Professional Documents
Culture Documents
Data Communication &fieldbus Systems
Data Communication &fieldbus Systems
Conducted for
OMV Aktiengesellschaft
Dates
03 - 07 December 2012
By
David Olson
Senior Consultant
INTRODUCTION
This programme takes an in-depth look at the divisive field of data communications and Fieldbus systems. The increasing trend towards digital communications within the industrial plant environment has brought in its wake a myriad of problems concerned with interconnectivity and differing standards offered by a wide variety of competing vendors. For the user, it has become progressively more difficult to differentiate between the proffered options.
WHO SHOULD ATTEND? Professionals involved in designing, selecting, sizing, specifying, installing, testing, operating and maintaining data communication and Fieldbus systems. Any professional needing to get to grips with the ever expanding and complex field of data communications and Fieldbus systems: Automation Engineers Chemical Engineers Consulting Engineers Electrical Engineers Electricians Industrial IT specialists Installation and maintenance technicians Instrumentation I t t ti and dC Control t lE Engineers i Maintenance Engineers
Operations Engineers Process Engineers Process Operators Production professionals f Project professionals System Integrators Other professionals who want a better understanding of f the th subject bj t matter tt
INTRODUCTION
INTRODUCTION o This workshop is designed to provide engineers and technicians with an overview of modern g communication standards - starting g at the basic RS232 standard right g through g to digital Profibus/Foundation Fieldbus and beyond. PERSONAL IMPACT On successful O f l completion l ti of f this thi workshop k h d delegates l t will ill b be able bl t to:
Appreciate the basic principles of data communications Apply modbus in a practical manner and make use of Understand the essential differences between asynchronous troubleshooting techniques and synchronous y transmission Differentiate between modbus and modbus p plus Apply basic faultfinding techniques covering rs232 up to rs485 Gain an insight into the essential differences between asibus Understand the importance of the iso osi model and devicenet Understand the basis of cyclic redundancy Make an informed appraisal of the differences between Recognise the different aspects of cable selection profibus and foundation fieldbus Apply different shielding technologies Understand the basics of ethernet Understand the application of fibre optics Appreciate the application of ethernet in the industrial Appreciate the value of hart in both calibration, data collection environment and diagnostics Understand tcp/ip and its associated protocols Recognise the wide range of fieldbus options Differentiate between net and sub-net masking Distinguish the functionality f of f ports and sockets
INTRODUCTION
ORGANISATIONAL IMPACT o Following the training and development experience provided by this workshop, participants will g equipped q pp with new skills and knowledge g that will enable them to return to their organizations evaluate the impact of new and existing technology in respect of your plant's communication strategy and apply a rationalized communication philosophy based on the best available practices. o By leveraging these skills skills, your enterprise can expect an improvement in overall productivity through the reduction in downtimes resultant from inappropriate data communications strategy and through the ability to apply, install, and troubleshoot a modern digital communication system. TRAINING METHODOLOGY o Designed for both novice and experienced engineers and technicians, this workshop starts at fundamental principles and progresses through basic coding and formatting systems, error detection, and protocols through to advanced Fieldbus and device networking concepts. o Throughout the workshop, participants will learn through active participation using exercises, questionnaires, and practical troubleshooting using protocol analysis and covering: RS 232, RS 485, Modbus, Networking, TCP/IP.
INTRODUCTION
COMPETENCIES EMPHASIZED
o o o o
Understanding of the major technologies used in modern data communication and fieldbus systems Select and specify different types of communication strategy according to the specific requirement i t and d application li ti The experience of exchanging ideas, problems and solutions with other delegates from different backgrounds and processes A renewed or refreshed ability y to make an informed decision regarding g g their communication requirements according to their process, plant or operation
SPEAKER PROFILE
Mr. David C. Olson is a senior consultant with GLOMACS, specializing in IT management, t N Networking, t ki IT security, it T Telecommunications l i ti and d wireless i l systems. David has over 28 years of professional experience in IT systems management t and d engineering, i i with ith responsibilities ibiliti f for all ll phases h of f system t desgn and development, delivery of services and network architecture. Prior to joining GLOMACS, David worked with the U.S. department of defense and as a senior systems engineeer and project manager with f Digital Equipment Corporation. Mr. Olson holds degrees in Electrical engineering and Informations systems management, and currently lectures in a E.U. sponsored technology program in Spain. David has presented over three hundred seminars to government clients and private concerns in North America, Europe, Asia and the Arabian Gulf.
TABLE OF CONTENTS
DAY ONE Pg. 10 I t d ti to Introduction t data d t communications i ti Overview Modern Instrumentation and Control y Systems "Smart" Instrumentation Basic principles Data communications standards RS-232 interface standard Faultfinding RS-422 interface standard RS-485 interface standard Interface Converters
DAY TWO ISO OSI model d l Communications media Cabling Shielding Grounding Fibre optics Error detection Checksum Cyclic Redundancy LAN Standards Topologies Protocols CSMA / CD (Ethernet) Token Ring Token Bus
pg.100
DAY FOUR Pg. 259 CANBus /DeviceNet Frame format Arbitration Fragmentation Error reporting DeviceNet media layer Network configuration Profibus FMS DP physical layer PA physical layer Foundation Fieldbus Comparison between Profibus PA and FF Publisher/subscriber model Physical layer Frame structure Link Li k active ti scheduler h d l Function block Device descriptions Safety and Reliability
Day One
Modern Instrumentation and Control Systems "Smart" Smart Instrumentation Basic principles Data communications standards RS-232 RS 232 interface standard Faultfinding RS-422 interface standard RS-485 RS 48 interface i f standard d d Interface Converters
10
SCADASYSTEMSCOMEIN MANYSHAPESANDSIZES
Thereisaremarkablevarietyinthesize,scopeandcomplexityofSCADAsystems.
Becausethereissuchawidescaleofapplications,thereisasimilarlyhugevarietyofcomponents andsystemsthatareusedintheindustry.
K-069
11
SUMMARY FEATURES
SCADA and Industrial Computing applications make use of technology that is in many ways the same as business Information Technology, with some very important differences. The main distinguishing features of this class of service are:
A key design feature is the systematic use of redundancy in areas of systems activity that are either particularly mission-critical or particularly fragile.
Redundancy measures often complicate the operation and administration of systems (e.g. managing duplicate, inconsistent data and designing fail-safe hand-over transfer systems).
Typically, process-related computer systems are required to coordinate process-wide operations with careful attention to time. Time sensitivity is a two-fold issue:
It is important that a common measure of absolute time be distributed, or synchronized across the physical system.
Relative time coordination is more critical, where the relative timing of complex sequences must be carefully managed.
12
SUMMARYFEATURES,cont.
Generallylessdemandingcapacityconcerns
Thisgeneralizationisrapidlybecominglessvalidasincreasinglysophisticatedvalue-added servicesaremixedwithraw,real-timedatastreamsinindustrialapplications.
Generallycautiousapproachtonewtechnology
Thescaleofindustrialapplicationstendsnottochallengetheperformanceenvelopeasmuch asbusinessapplications.
Asaresult,thiskindofperformance-hungryapproachtosystemsdesignisfarlessimportant thanassured,reliable,resilientandmaintainablesystems.
Forindustrialcomputingapplications,thesearefarmoreimportantdesignobjectives--the conceptofbleedingedgetechnologyhasnoplaceincoreindustrialapplications.
13
PRIMARYANDSECONDARYAUTOMATION
Theautomationandcontrolresourcesthataretypicallyusedfallintotwobroadcategories:
PrimaryControl
Atthislevelprocessvariablesaremeasuredandoperationalset-pointsaredefined,typically usingclosed-loopcontrolsystems,andtypicallyoverphysicallysmalldimensions(localcontrol).
K-016
14
PROTOTYPESCADASYSTEM
HMI
MTU COMPUTER SYSTEM
COMMUNICATIONS EQUIPMENT INTERFACES
TELEPHONE MODEM
WIRELESS LINK
LINE DRIVER LINE DRIVER
WIRED LINKS
REMOTE TERMINAL UNIT
TELEPHONE MODEM
K-007
15
The Programmable Logic Controller (PLC) is a class of device that evolved from traditional relay-based local control systems. Classical features and applications are:
A control-oriented operating regime that is based on deriving a closed-loop local control function based on Boolean logical conditions applied to input conditions -- a relationship that is commonly expressed as a relay ladder diagram.
Applications are typically stand-alone, where the PLC functions as an autonomous controller of some local activity, such as burner control or as a waste water pump control.
PLCs have been enhanced in several ways, making them much more capable and easily used:
Addition of human interface features, supporting graphical displays, event logging and trend reporting.
Support for personal computer (PC) software development/testing/downloading, making application development a greatly simplified task.
Networked connectivity linking multiple PLCs, Human-Machine Interfaces (HMIs) and PCs -- creating capability for supporting greatly enhanced system scope and capacity.
16
The Distributed Control System (DCS) is an integrated hardware/software system concept that is designed to efficiently perform data acquisition and control functions.
The original concept, and the class of application from which current generation DCS products have evolved was:
A digital programmable system optimized for performing closed loop control -- and replacing earlier analog controllers for this class of application.
A system concept using digital microprocessor technology, which adds reliability, functionality and convenience features largely unavailable with earlier analog technology.
Major deficiencies that preexisting analog control technology had was stability problems due to sensitivity to the environment (voltage, temperature, mechanical) and difficulty in easily implementing non-linear control functions.
a distributed architecture, with local control modules, human interface consoles, (typically) proprietary backbone network.
support a reasonably user-friendly programming language and applications development capability, such that end-user modifications to system software become feasible.
17
OPERATOR CONSOLES
18
THEROLEOFMINICOMPUTERS
Formanyyears,minicomputerswerethecomputingplatformofpreferencetodesignersof SCADAandrelatedsystems.Reasonsforthispreferencewere:
K-012
19
THEPERSONALCOMPUTERAS ANINDUSTRIALCOMPUTINGRESOURCE
Thepersonalcomputerhasbecomeauniversalcomputingresource--onethathascompelling advantagesinmostallcomputerapplications:
Itistheglobalcomputingplatformofchoice,makingitthecommondenominatorinanyindustrial applicationthatrequirescomputerfunctionality.
Theprice/performancefeaturesareunbeatable,eveninruggedizedpackages.
MostdevelopmentsinPCandrelatedtechnologiesareorientedtowardbusinessapplications (andthereforetypicallynotideallysuitedtoindustrialapplications).
Itisnotaninherentlyfault-tolerantorruggedarchitecture,makingimprovementsinthese areaspiecemealandthereforelackinginoverallsystemintegrity.
Regardlessofargumentsproandcon,thePCwillundoubtedlyhaveanevermoreimportantrole inindustrialcomputingapplications.
Data Communications & Fieldbus Systems
K-013
20
THEUBIQUITOUSPERSONALCOMPUTER
Mostorganizations--indeedthewholeworld--haveeffectivelystandardizedonIntel-compatible personalcomputers(runningavariantofMicrosoft'sWindows--orincreasingly,LINUX).
RecentcorporateexperienceintheUnitedStatesshowsthatdesktopcomputersystemsare replaced(onaverage)everythreeyears,andnotebookcomputerseverytwoyears.
D-329
21
RUGGEDIZEDPCPACKAGING
NumerousvendorshavedevelopedruggedizedPCpackagesfordiverseapplications--designedto helptheratherfragileconventional"officeoriented"PCtosurviveintheenvironmentallyharsh worldofindustrialapplications. FactorsthatthreatenconventionalPC'sinharshenvironmentsinclude: Heat Vibration Humidity Dustanddirt Causticatmosphere Powersystemstabilityandreliability Someconfigurationissuesthatare importantinthiscontextare: Choiceofoperatingsystem; typicallyaWindowsvariantor Linux. Choiceofmassstorage technology;typicallyarugged harddriveorflashdrive.
RUGGEDIZEDPCPACKAGINGEXAMPLE (www.acuraembedded.com)
22
The roles of SCADA, PLC and DCS technology have changed with the progressive development of technology:
TRADITIONAL ROLES
Replacement for discrete analog local closed loop control systems Digital technology offers flexibility in development and modification Discrete logical closed loop control applications traditionally served by relay systems Digital technology benefits Open loop data collection and control functions supported by human operator Typically involves off-site network
EXPANDING ROLES
DCS
Increased sophistication of the human interface and modular features encroach on SCADA Many features similar to PLC
PLC
PLC systems frequently serve as high-end RTUs in SCADA app's PLC features and extensions encroach on role of DCS
SCADA
SCADA systems integrate with DCS and PLC systems Low end SCADA applications lose out to DCS and PLC
23
ENTERPRISE,CONTROL,SENSOR ANDDEVICENETWORKS
HMI/MMI
SCADA NETWORK (ControlApplications)
HMI/MMI
RTU/PLC
RTU/DCS
24
THESHIFTFROMACOMPUTER-CENTRIC TONETWORK-CENTRICPARADIGM
USER NETWORK
USER
COMPUTER
COMPUTER
NETWORK
Twokeyobservationsapplyhere:
Thenetworkhasbecomethestrategicheartofourcomputingenvironment.
totheuserthanpreviously. Thecomputerismuchcloser(andthereforemuchlessthreatening)
25
DTE
COMMUNICATION FACILITIES
DCE
MODEM
DTE: DCE: -
Data Terminal Equipment (DTE) consists of whatever apparatus physically terminates the circuit (dumb terminal, personal computer, mainframe, etc.).
Data Communications Equipment (DCE) consists of equipment which interfaces the DTE to the communications facilities (DCEs are typically analog modems or digital line drivers).
Note that packaging decisions may blur the distinction between these two functions (e.g. an internal modem mounted inside to the PC cabinet).
AA
D-003
26
HALF-DUPLEX OPERATION provides for communication in either direction, but not simultaneously. The reversal of direction may be sufficiently fast so as to create the illusion of full duplex operation.
Operates like a street capable of working in either direction, but alternating, perhaps as controlled by a flag-person.
FULL-DUPLEX OPERATION provides for simultaneous transmission in both directions. Thus, neither direction of communication is forced to wait for activity to complete in the reverse direction, as is often the case with half-duplex.
Operates like a typical city street, or divided highway, with no interaction between opposing traffic (that is, there is never any requirement to wait for, or be slowed down by opposing traffic)
D-007
27
SYNCHRONIZATION
There are three dimensions to the synchronization, or timing problem:
1. BIT SYNCHRONIZATION: Bit synchronization is accomplished with electronic circuitry (in either DCE's or DTE's). Bit synchronization is dependent on some prior knowledge of digital bit patterns, and predictability of transitions between binary 1 and 0 values.
1 0 1 1 0 0 1 1
1 0
IN ORDER FOR USEFUL CLOCK RECOVERY TO FUNCTION, THERE MUST BE GUARANTEES OF SUFFICIENTLY FREQUENT TRANSITIONS (between 1 and 0) IN THE DATA
2. CHARACTER SYNCHRONIZATION: Character synchronization involves assuring all bits from one character (or data octet) remain properly associated with the byte or octet to which they belong.
THE BYTE (Octet) BOUNDARIES MUST BE RESPECTED IN THE PROCESS OF RECREATING MESSAGES. ACCIDENTAL REALIGNMENT OF THESE BOUNDARIES WILL RENDER DATA USELESS.
0
1
3. MESSAGE SYNCHRONIZATION: Message synchronization involves unambiguously marking (and observing) the beginning and end of a message, which will typically involve multiple segments or packets.
PACKET/FRAME NUMBER 2
PACKET, OR MESSAGE FRAME SEQUENCE MUST BE RESPECTED IN THE PROCESS OF RE-ASSEMBLING FRAMES. VARIATIONS IN NETWORK TRANSIT DELAYS CAN RESULT IN OUT-OF-SEQUENCE FRAME ARRIVALS.
D-064
28
STOP PULSE (The unconditional MARK or ONE value of this pulse assures that the next START pulse -- which is always a MARK-to-SPACE transition -- will be recognizable)
7 DATA BITS (Most asynch. systems operate with 7 data bits) THE 8th. BIT IS MOST COMMONLY USED AS A PARITY BIT (Parity checking is a simple form of error detection)
INCREASING TIME
D-001
29
SYNCHRONOUSSYSTEMCHARACTERISTICS
TYPICALSYNCHRONOUSMESSAGEFORMATCONSISTSOF:
DTE DTE
SYN SYN STX X X X ETX BCC
DCE
MODEM
DCE
MODEM
DIRECTIONOFDATAFLOW
D-002
30
INTERPRETATION EXAMPLE:
010
111
011 4 @ p q b r A B R S
T
100
101 110
0000 0 0 1 2 3
4 D
SP
P Q a
0001 1
2 3
! " #
C
0010 0011
0100
c
d
s
t
4
0
1
U e u
6
8 7 6
4 5 6
7 BEL 8 ETB CAN
$ % & ' ( )
5 6 7
8
0
E
F
5
V G H W
X f
4
v g
h
0111 BS HT 9
A
w x EM
9
Bit Positions:
1001
1010
*
B VT ESC
: + ; ,
>
J
K
[
C
k FF D E
CR
{
FS
GS
NOTES:
L
\ _
SO RS
| <
M
.
1111 F SI
US
DEL
D-004
31
SERIALvs.PARALLELCOMMUNICATION
"O" "K" 1 1
PARALLELDATASEQUENCES:
CircuitNo.2
CircuitNo.1
1 1
1
CircuitNo.4
1 0
1
CircuitNo.3
"O"CHARACTER:8
0
CircuitNo.5
0 0
CircuitNo.6
"K"CHARACTER:-
0
1
CircuitNo.7
1
CircuitNo.8
SERIALDATASEQUENCE:
BITSEQUENCEREPRESENTING ASCIICHARACTER"O" 1 1 1 1 0 0 1 1
BITSEQUENCEREPRESENTING ASCIICHARACTER"K"
STARTPULSE (TRANSITION):
STOP PULSE:
STARTPULSE (TRANSITION)
STOP PULSE:
D-081
32
ANALOG
TRADITIONAL LEASED, DEDICATED, PRIVATE LINE SERVICES (Require analog modems) GLOBAL, UBIQUITOUS PUBLIC SWITCHED TELEPHONE NETWORK (PSTN) SERVICE (Commonly referred to as "POTS")
DIGITAL
DEDICATED
CIRCUIT SWITCHED
BASIC RATE ISDN SUPPORTS CIRCUIT-SWITCHED 64 KBPS SERVICE; TYPICALLY CONFIGURED WITH 2 CHANNELS CAPABLE OF FUNCTIONING AT 128 KBPS.
PACKET SWITCHED
NOT FEASIBLE
OBSOLESCENT X.25 PACKET SWITCHING SERVICE IS STILL USED; FAST PACKET TECHNOLOGY (Frame Relay and Cell Relay/ATM) HAS LARGELY REPLACED X.25
D-024
33
ANALOG SIGNALS ARE INFINITELY AND CONTINUOUSLY VARIABLE (Meaning the signals are very complex, and it is very difficult to efficiently characterize their behavior so that electronic circuits can interpret their meaning)
DIGITAL SIGNALS ARE DISCRETE IN BOTH TIME AND AMPLITUDE (Meaning they change only at predictable times, and vary between a small number of predictable values)
D-005
34
FUNDAMENTAL OF COMMUNICATIONS
Fourier Series Approximation Nyquist Theorem Shannons Theorem Modulation and Demodulation
35
36
g1(t)=sin(2f t)
g2(t)=1/3sin(2( 3f )t)
37
FOURIER TRANSFORM
Jean B. Fourier found that any periodic function can be expressed as an infinite sum of sine function.
38
FOURIER APPROXIMATIONS
39
Transmission speed can be measured in bits per second(bps). Technically, transmission is rated in baud, the number of changes in the signal per second that the hardware generates. Using RS-232 standard to communicate, bit rate rate = baud rate. In general, bit rate rate = N * baud rate, where N is the number of signals in a string.
40
Sender sends the bit string, by b1 b2 bn. The transmitter alternately analyzes each string and transmits a signal component uniquely determined by the bit values. Once the component is sent, the transmitter gets another bit string and repeats this process. The different signal components make up the actual transmitted signal. The frequency with which the components change is the baud rate. At the receiving end, the process is reversed. The receiver alternately samples the incoming signal and generates a bit string.
Data Communications & Fieldbus Systems
41
Consequently, the bit rate depends on two things: the frequency with which a component can change (baud rate) and n, the number of bits in the string. That is why the formula:(signal may have up to 2n different amplitudes) bit rate = n * baud rate
42
Dr. Harry Nyquist (1920) Nyquist showed that if is the maximum frequency the medium can transmit, the receiver can completely reconstruct by sampling it 2 times per second on a perfectly noiseless channel. In other words, the receiver can reconstruct the signal by sampling it at intervals of 1/(2) second. For example, if the max frequency is 4000 Hz, the receiver needs to sample the signal 8000 times per second or using 2 as the baud rate. Bit rate = 2 * n.
43
The formula Bit rate = 2 * n seems to imply that there is no upper bound for the data rate given the maximum frequency. Unfortunately, this is not true for two reasons.
Amplitude limitations Noise
44
First, if we used amplitude to represent data bits, each time we separate the amplitude into smaller ranges to represent more data bits, the receiver must be more sophisticated (and more expensive) to be able to detect smaller differences. If the differences become too small, we eventually exceed the ability of a device to detect them (or a medium to react to them).
45
46
S/N is known as the signal-to-noise ratio in decibels. Because S is usually much larger than N, the ratio is often scaled down logarithmically and the unit is measured in bels and 1 dB = 0.1 bel. You could expect somewhere around 1000-1 S/N on a typical telephone circuit.
47
SHANNONS THEOREM
Bit rate = Bandwidth * log 2 (1+S/N) bps. According to this result, a bit rate around 35,000 bps is an upper limit for conventional modems. Remember telephone specs!
48
According to this result, a bit rate around 35,000 bps is an upper limit for conventional modems. How about 56k modems?
49
TELEPHONESYSTEMSCHEMATICVIEW
TELEPHONE EXCHANGE
TELEPHONE EXCHANGE
LOCALLOOP
INTER-EXCHANGENETWORK
LOCALLOOP
LOCALLOOPFEATURES:
D-060
50
0 1 1 1
1
0 1
1 0 1 1 0 0 1 1
D-023
51
With digital communication technology, there is a general relationship between the maximum feasible distance of communication and maximum feasible data rate -- for any given technology and/or implementation.
In the box illustration, the trade-off between maximum data rate and distance is shown to follow the classic hyperbola curve.
Different technologies will support different hyperbola curves, but for any given (level of sophistication of) technology, it will always be possible to double the data rate so long as one is willing to half the distance capability. This characteristic is common to all digital format signals (electrical or optical), and answers questions like:
How far can I stretch an EIA-232-C interface cable ? What is the maximum feasible data rate for twisted pair (UTP) cable ? How fast can line drivers operate ? Why can't LAN systems be stretched to greater physical dimensions ?
HYPERBOLIC CURVES HAVE THE CHARACTERISTIC OF FOLLOWING THE LINE OF CONSTANT PRODUCT OF THE TWO AXES;
TRANSMISSION DISTANCE
D-265
52
COMPARISON:ANALOGAMPLIFICATIONvs. DIGITALREGENERATION
Digitalsignalsaretypicallyviewedas rectangularpulses,assuggestedbythe 'sourcesignal'inbothsketches. Signalsinevitablyweakenandbecome distortedoverthedistanceof communication,assuggestedbythe 'weakenedsignal'view. Amplifierssimplyincreasethesignal strengthwithoutreshapingorrepairing thedistortion.
SOURCE SIGNAL
WEAKENED SIGNAL
AMPLIFIED SIGNAL
SOURCE SIGNAL
WEAKENED SIGNAL
REGENERATED SIGNAL
53
DIGITALTELEPHONENETWORKOVERVIEW
CODEC
TIMEDIVISION MULTIPLEXER SYSTEMS
MODEM
CODEC
TIMEDIVISION MULTIPLEXER SYSTEMS
CODEC
CODEC
MODEM
TERMINAL SYSTEMS
SIGNAL CONVERSION
MULTIPLEX SYSTEM
SIGNAL CONVERSION
TERMINAL SYSTEMS
TERMINALSYSTEMS :
Terminalsystemcomponentsareanyuser-selectedterminaldevices (voice,video,graphics,etc.).
Inmostcountries,terminalsystemsareprivatelyowned,althoughhistoricallythetelephonecompanies havejealouslyguardedtheirrightofownership.
continued ....
D-246
54
DTE
Link multiplexers may function as frequency division multiplexers, time division multiplexers or statistical time division multiplexers. The common concept is that the multiplexer-to-multiplexer segment is adapted to carry the aggregate data volume for the various services being connected.
Multiplexing is often integrally packaged with other functions, such as circuit or packet switching capability, modems, data compression, etc.
A
DCE
MODEM
A A A
REMOTE MULTIPLEXER
D-008
55
ENCODINGDATAFORTRANSMISSION --THEROLEOFTHEDCE
Thus,inordertocommunicateovergreaterdistances,thesesignalsneedtobeconvertedinsome way--intosignalsthataremorerobust,andconformtothefeaturesofthefacilitiesavailable.
Thatis,foranalogcommunicationsfacilities,ananalogsignalisrequiredandfordigital communicationsfacilities,adigitalsignalisrequired.
DTE
MODEM
DCE
DTE
LINEDRIVER
LINEDRIVER
D-224
56
BITRATEvs.BAUDRATE
Whereonebitofinformationis representedbyoneelectrical signalelement--orsymbol,thebit rateandBaudratehavethesame numericalvalue.
1
1
0
ONEBITPERBAUDCASE
DATASEQUENCE
DIGITALSIGNAL
ANALOGSIGNAL
1 0
INCREASINGTIME
INCREASINGTIME
DATASEQUENCE
1
DIGITALSIGNAL
11
10
01
INCREASINGTIME
00
ANALOGSIGNAL
11 10 01
00
INCREASINGTIME
D-240
57
The AT&T organization was at one time the leader in developing modem technology. Some examples of "Bell-compatible" modems are:
DATA RATE TYPE 103: (1) TYPE 201C: TYPE 202S: TYPE 208A: TYPE 209: TYPE 212A: 2400 bps 1200 bps 4800 bps 9600 bps 1200 bps 300 bps DUPLEX Full Half Half Full/Half (2) Full/Half (2) Full TIMING Asynch. Synch. Synch. Synch. Synch. Synch/Asynch.
LINE TYPE
Dial (2W)
Dial (2W)
Dial (2W)
Leased (4W)
Leased (4W)
Dial (2W)
With the advent of a liberalized telephone interconnect policy (allowing connection of customer owned equipment) in the early 1980's, competitive specialist organizations passed AT&T by in this market. These are largely based on ITU (ex CCITT) "V-Series" modem standards.
NOTES: (1) Type 103 modems were acoustically coupled, thus requiring no metallic wiring connection. (2) Types 208 and 209 modems operate half duplex on 2-wire connections, full duplex on 4-wires.
D-179
58
Most significant, current developments in modem technology are coordinated through the ITU-T (formerly CCITT -- see Note 1 below). All ITU-T modem standards carry a "V" prefix, and are commonly referred to as the "V-Series" standards. Some examples of ITU-T modems are:
DATA RATE V.22 1200 bps 2400 bps 4800 bps 9600 bps 9600 bps 14,400 bps 14,400 bps 28,800 bps V.22 bis V.27 V.29 V.32 V.32 bis V.33 V.34 DUPLEX Full Full Full/Half (2) Full/Half (2) Full Full Full/Half (2) Full TIMING Asynch. Synch/Asynch Synch. Synch. Synch/Asynch. Synch/Asynch. Synch. Synch/Asynch.
LINE TYPE
Dial (2W)
Dial (2W)
Leased (2W/4W)
Leased (2W/4W)
Dial (2W)
Dial (2W)
Leased (2W/4W)
Dial (2W)
NOTE: (1) The ITU-T functioned for many years as the CCITT -- only being renamed in 1994. Thus, many documents and references still use this earlier name.
(2) Type V.27 and V.29 modems operate half duplex on 2-wire connections, full duplex on 4-wires
D-180
59
SCHEMATICVIEW-2-WIREAND4-WIREDATA
2-wire(e.g.dialaccess)dataconfigurationsshareacommonpairoflocalloopcable. Thissharedaccesslocalloopmustservebothdirectionsofinformationflow.
DTE DCE
DTE
DCE
TELEPHONE EXCHANGE
TELEPHONE EXCHANGE
2-WIRELOCALLOOP
INTER-EXCHANGENETWORK
2-WIRELOCALLOOP
4-wire(i.e.dedicatedaccess)configurationshaveadedicatedlocalloopcablepairforeach directionofinformationflow(thusnoconflictorcompromiseiscreatedbyopposingtraffic).
Thus,a"dividedhighway"configurationprevailsallthewayfromtheDCEtotheDCE.
DTE DCE
DTE
DCE
TELEPHONE EXCHANGE
TELEPHONE EXCHANGE
4-WIRELOCALLOOP
INTER-EXCHANGENETWORK
Data Communications & Fieldbus Systems
4-WIRELOCALLOOP
D-281
60
LINEDRIVERS...DIGITALDCE's
DTE
LINEDRIVERFEATURES:
LINEDRIVERS
DCE DCE
D-210
61
LOADINGCOILSONCOPPERCABLE
NORMAL TELEPHONE CHANNEL BANDWIDTH PROFILE
AMPLITUDE
INCREASINGFREQUENCY
LOADINGCOILS
TOCUSTOMERPREMISES
TELEPHONECENTRAL OFFICEEXCHANGE
LOADINGCOILSARETYPICALLYINSTALLEDEVERY6000ft.(FOLLOWING THEFIRSTPAIRAT3000ft.)--ONLOOPLENGTHSEXCEEDING3miles.
AMPLITUDE
D-211
62
Link protocols are required to create a structured, reliable environment for data to be exchanged. Elements of this role are:
Provide an environment where integrity of data is assured (integrity focuses on the requirement that data should be delivered to the appropriate destination -- intact).
Provide for accuracy in the data delivered (assurance of error detection and correction, so that other, higher layer functions can safely assume data and control information is correct).
In combination, these integrity and accuracy concerns result in a formal process of custody transfer regarding data. Additional responsibilities include:
Flow control -- responding to congestion in the network, or temporary jeopardy of buffer overflow.
Management of contention -- without which an ethernet LAN or a multipoint line could not function.
D-175
63
Link layer protocols provide an organized, structured data delivery capability, validating received data, thus confirming the correct delivery destination and completeness of delivery (integrity verification) and correct message content (accuracy verification). TELETYPE PROTOCOL (Asynchronous Systems)
Teletype protocol (TTY) is unstructured, in that there are no embedded guarantees of data integrity or accuracy.
Teletype protocol was developed for the electro-mechanical teletypewriter systems used 60 years ago -- and presumes no higher level of equipment "intelligence" than these early devices provided. BINARY SYNCHRONOUS PROTOCOL (Synchronous Systems)
Binary Synchronous (BISYNC) is an inherently half-duplex protocol which was the first widely used structured protocol -- offering generally appropriate accuracy and integrity control. In most cases, BISYNC delivers poor facilities utilization efficiency.
HDLC, patterned after IBM'S Synchronous Data Link Control (SDLC) protocol, is a link service that meets most current application needs for accuracy, message integrity and efficiency. HDLC is inherently insensitive to code formats and message content, making it a convenient, general purpose link layer support protocol.
D-103
64
TELETYPE(TTY)PROTOCOL
Teletypeprotocolsystemstendtobesatisfactoryinapplicationswhereahumanisinvolvedinthe process(i.e.eitherhuman-to-humanmessaging,orhumaninteractiveterminalservice).
Whenmachine-to-machinecommunicationisrequired(withinherent"zerotolerance"oferrors), basicunaugmentedTTYprotocolisunacceptable.
AnumberofauxiliaryApplication-layerprotocols(suchasXMODEM,KERMIT,BLAST,MNPand V.42)havebeenusedtocompensateforsomeoftheselimitations.
D-104
65
HALF-DUPLEXvs.FULLDUPLEX PROTOCOLOPERATION
HalfDuplexProtocol:
Halfduplexissimpletoimplement,simpletooperateandveryeffectiveinsimpleapplications.
Inefficiencyoccursthroughonlytransmittinginonedirectionatatime--andwaitingtimedelaysare requiredtoreversethedirectionoftransmissiontwiceforeachblockofdatasent.
FullDuplexProtocol:
Messagereferences,suchasrequestsforretransmission,aremadeusingthesenumbers.The numberofdifferentnumbersavailablefornumberingthemessagesiscalledthewindowsize.
Fullduplexprotocolssupportefficientutilizationofexpensivetelecommunicationsfacilities.
Complexityofimplementationslowedearlyadoption,butthisisanon-issueincurrentpractice.
D-068
66
BINARYSYNCHRONOUS MESSAGEFEATURES
REPEATED'SYNCHRONOUSIDLE' CHARACTERSARESENTFIRST TOPERMITBITANDCHARACTER SYNCHRONIZATIONTOBE ESTABLISHED.
S Y N S T X
S Y N
S Y N
E T X
B C C
B C C
VARIABLELENGTHDATAFIELD.TYPICALBLOCKLENGTHSMAYBE200TO2000CHARACTERS.
D-065
67
BINARYSYNCHRONOUS TRANSPARENTMODEFEATURES
THENETMESSAGECONTENTIS NOWBRACKETEDBYTWO CONTROLCHARACTERSATBOTH ENDS;DLE/STXandDLE/ETX.
S Y N S T X
S Y N
S Y N
D L E
D L E
D L E
D L E
D L E
D L E
E T X
B C C
B C C
ThereceivingDTEhastheresponsibilityofremovingallinsertedDLEsinthedatafield,thus restoringittoitsoriginalunmodifiedformat.
D-066
68
STATION "B"
INCREASING TIME
NO RESPONSE EXPECTED
D-069
69
HIGHLEVELDATALINKCONTROL FRAMESTRUCTURE
FLAG ADDRESS CONTROL
CRCERROR CHECKFIELD
FLAG
VARIABLELENGTHDATAFIELD
DATAFIELD:TheDATAfieldisvariablelength.Itistypicallyrequiredthatthisfieldcontainan integernumberofoctets--with2,000octetsbeingatypicallength.
D-067
70
HDLCBITSTUFFINGFORTRANSPARENCY
ORIGINALDATA (Incl.controlfields)
0
-----FLAG-----
1 1 0 0 1 1 1 0 1 1 1 1 0 0 1 1 1 1 1 0 0 0 1 1 1 1 1
-----FLAG-----
Thereceivingstationmustsubsequentlylocateandremoveanystuffedbits--asimpletask,since every"0"bitwhichfollowsfive"1"bitsisunconditionallyastuffedbit,andmustbediscarded.
D-221
71
HDLCCONTROLFIELDCOMPONENTS
The8-bitControlFieldtakesone ofthreeforms--orformats:
InformationFormat;whichis alwaysusedwhendataisactually beingsent, SupervisoryFormat,whichis usedtoassertnormalcontrolover thelink,and UnnumberedFormat, which performsextraordinarycontrol tasks,suchasresettingthelink.
--------BITPOSITIONSINCONTROLFIELD--------
1
I-FRAME (InformationFormat)
SENDSEQ. NUMBER
P/F
RCV.SEQ. NUMBER
S-FRAME (SupervisoryFormat)
SUPV. BITS
P/F
RCV.SEQ. NUMBER
U-FRAME (UnnumberedFormat)
MODIFIER FCN.BITS
P/F
MODIFIER FCN.BITS
3-BIT(Modulo8) CIRCULARCOUNTER
D-222
72
HDLCCONTROLFUNCTIONS,cont.
InformationFormatControlFieldContents:
Twosequencenumbersarecarried--SendandReceivenumbers
Thetwo-bitfieldofsupervisorydataisencodedasfollows:
continued ....
Data Communications & Fieldbus Systems
D-223
73
HDLCCONTROLFUNCTIONS,cont.
SupervisoryFormatControlFieldContents,cont.
HDLCwasderivedfromtheSynchronousDataLinkControl,orSDLClinkprotocoldevelopedby IBM.
TherearecountlessotherHDLCimplementations,bothproprietaryandpublicdomain.
D-282
74
MESSAGESEQUENCENUMBER CIRCULATIONEXAMPLE
MESSAGEINTRANSITEASTBOUND
4
STATION"A"
DATA SOURCE: OUTPUT MESSAGE BUFFER
STATION"B"
INCOMING(Receive)MESSAGE SEQUENCENUMBERSARECOPIED OVERTOTHEOUTGOING(Send) SEQUENCENUMBERFIELDFOR TRANSMISSIONWITHTHENEXT MESSAGE.
5 4
2
MESSAGEINTRANSITWESTBOUND
D-177
75
RELATIONSHIP:OPTIMUMPACKETSIZE DEPENDSONERRORRATEEXPECTATION
Foreverynetworkenvironment,there willalwaysbeanoptimummessage unit,orpacketsize,largelydetermined bytheerrorrateencounteredinthat environment.
THROUGHPUTEFFICIENCY
HIGHERROR RATECASE
INCREASING(NET)PACKETSIZE
T-196
76
EIA-232-CFEATURES/FUNCTIONS/DEFICIENCIES
TheTelecommunicationsIndustriesAssociation(formerly,theElectronicIndustriesAssociation) hasdefinedeffectivelyallequipmentinterfacestandardsinNorthAmerica.
Earlierversionsweremoredatarate-restricted;EIA-232-CandDrevisionsarelimitedto20kbps. --inlesssophisticatedtimes,thiswasnotaseriouslimitationbutitcurrentlyisakeyconstraint.
Dependingondatarate,electromagneticenvironmentandcabletype,this50-footlimitationmay bereasonablystretchedtoafewhundredfeet--perhapsasfaras500.
D-171
77
THE FORMAL EIA-232-C INTERFACE ACTUALLY OCCURS AT THIS POINT (at the DCE connection rather than DTE)
TELEPHONE CABLE
MODEM
DTE DCE
THE CABLE END TERMINATING AT THE DCE IS A MALE GENDER CONNECTOR, ... THEREFORE, THE CHASSIS CONNECTOR ON THE DCE IS FEMALE GENDER
AA
D-026
78
EIA-232-CSIGNALANDCONTROL VOLTAGECONVENTIONS
1 OFF
VOLTAGESCALEVERTICALLY
0 1 1
ON OFF OFF ON
0
ON
1 OFF OFF
+25
+3 0 -3
TIMESCALEHORIZONTALLY
BIPOLARNONRETURNTOZERO (BNRZ)FORMAT
-25
D-029
79
5. CLEAR TO SEND
4. REQUEST TO SEND 3. RECEIVED DATA 2. TRANSMITTED DATA 1. PROTECTIVE GROUND 7. GROUND & COMMON RETURN
11. UNASSIGNED
25. UNASSIGNED
D-022
80
ABBREVIATION
SUBSET EQUIPPED: 4 9 15 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 SDCD SCTS STXD TXCLK SRXD RXCLK SRTS DTR SQD RI RATE XTCLK
Frame Ground Transmitted Data Received Data Request to Send Clear to Send Data Set Ready Signal Ground Data Carrier Detect Positive Test Voltage Negative Test Voltage Unassigned Secondary Carrier Detect Secondary Clear to Send Secondary Transmitted Data Transmitter Clock Secondary Received Data Receiver Clock Unassigned Secondary Request to Send Data Terminal Ready Signal Quality Detect Ring Indicator Data Rate Selector External Clock Unassigned
D-074
81
9. RING INDICATOR
8. CLEAR TO SEND
7. REQUEST TO SEND
D-053
82
6. DCE READY
5. CLEAR TO SEND
4. REQUEST TO SEND 3. RECEIVED DATA 2. TRANSMITTED DATA 7. GROUND & COMMON RETURN
11. UNASSIGNED
1. SHIELD GROUND
D-038
83
UNBALANCED ELECTRICAL CIRCUIT -SUCH AS EIA-232-C. THE GROUND RETURN PATH IS SHARED BY ALL ELECTRICAL SIGNALS.
INPUT
OUTPUT
UNBALANCED ELECTRICAL CIRCUIT -SUCH AS EIA-423. GROUND RETURN PATH IS ISOLATED FROM OTHER FUNCTIONS, BUT ELECTRICAL BALANCE IS LOST THROUGH THE GROUND CONNECTION.
INPUT
OUTPUT
BALANCED ELECTRICAL CIRCUIT -SUCH AS EIA-422. BOTH CONDUCTORS ARE ELECTRICALLY EQUAL (NEITHER IS DESIGNATED AS "GROUND" CONNECTION). THE RESULTING SYMMETRY CREATES SUPERIOR NOISE/INTERFERENCE IMMUNITY. CABLE CONNECTION CIRCUIT RECEIVER CIRCUIT
D-079
84
4. SEND DATA
3. UNUSED 2. SIGNALING RATE INDICATOR
1. SHIELD GROUND
21. UNUSED
22. SEND DATA
23. SEND TIMING 24. RECEIVE DATA 25. REQUEST TO SEND
D-025
85
6. DCE READY
5. CLEAR TO SEND
4. REQUEST TO SEND 3. RECEIVED DATA 2. TRANSMITTED DATA 1. SHIELD GROUND 7. GROUND & COMMON RETURN
NOTE: DUPLICATED TRANSMIT SIGNAL ELEMENT TIMING ASSIGNMENTS PROVIDE FOR CLOCK SOURCE TIMING TO BE PROVIDED BY EITHER DTE OR DCE
D-039
86
10,000
DISTANCE IN FEET
1000
6 8
6 8
100
6 8
2 2 4
10
100 bps
AAAAAAAAAAAAAAAAAAAAAAA A AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA A A A A A A A A A A A A A A A A A A A A A A A AAAAAAAAAAAAAAAAAAAAAAA A A AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA AA A
2 4 6 8 2 4 6 8 2
1kbps
10 kbps
100 kbps
1 Mbps
10 Mbps
D-078
87
The RS-485 interface standard has some strong similarities to the RS-422 interface: It uses the same balanced electrical configuration It uses essentially the same signaling voltage conventions
The key concept difference is the RS-422 is a point-to-point service (operating on a two-pair, or 4-wire connection), while the RS-485 supports multi-drop, or bus configurations (operating over a two-wire, or single pair connection) -- see sketch.
RS-485 operates up to 56 kbps (compared with RS-422 operation to 10 mbps under some circumstances).
This multi-drop bus format configuration lends itself to network connection of field instruments -suitable to applications where the simplicity of the configuration is desirable, and high performance is not required.
BUS-CONNECTED DEVICE
BUS-CONNECTED DEVICE
BUS-CONNECTED DEVICE
K-030
88
?
PIN NO. 2 3 7 PIN NO. 2 3 7
A DIRECT, SIMPLE CONNECTION IS INOPERABLE. TRANSMIT AND RECEIVE DATA SIGNALS (ON PINS 2 and 3 RESPECTIVELY) ARE IN DIRECT CONFLICT. OTHER INTERFACE FUNCTIONS (LIKE READY STATUS AND RTS/CTS) WOULD BE SIMILARLY MISHANDLED.
D-040
89
1 2 3 4 5 6 7 8 20
1 2 3 4 5 6 7 8 20
1 2 3 4 5 6 7 8 15 17 20 24
NOTE: (1) DETAILS OF SPECIFIC SYSTEMS, AND INTERCONNECTION SUBTLETIES PREVENT DEFINITION OF A UNIVERSALLY "CORRECT" MODEM ELIMINATOR WIRING CONFIGURATION. THESE EXAMPLES ARE INTENDED AS ILLUSTRATIONS OF HOW SUCH CONFIGURATIONS MIGHT BE ESTABLISHED.
(2) THIS EQUIPMENT CONNECTION PROBLEM IS GENERALLY ASSOCIATED WITH THE NEED TO DIRECTLY CONNECT DTE-to-DTE; IT IS AN IDENTICAL PROBLEM IN SITUATIONS WHERE DIRECT DCE-to-DCE CONNECTIONS ARE REQUIRED.
D-075
90
CROSSOVERCABLES
Itisalwaysasourceofconfusiontohavetodealwiththeinevitableasymmetryof electricallycabledconnections.
Transmitting,orsendingdevicesarealwaysrequiredatoneendofawiredlink,anda receivingcircuitisrequiredattheotherend--afundamentallyunavoidablyasymmetric featureofwiredconnections (seetheillustrationintheaccompanyingbox). Thetwotypesofdevicesconnected withtheRS-232aremodemsand terminalequipment(Data CommunicationsEquipment,or DCEsandDataTerminal Equipment,orDTEs)--the terminal,orDTEviewissuggested intheillustrationhere. A'straightthrough'cablewillclearly notworkinthiscase.
continued ....
Data Communications & Fieldbus Systems
PIN NO. 2 3 7
PIN NO.
TERMINALSTRANSMIT ONPIN2
2 3 7
TERMINALSRECEIVE ONPIN3
D-735
91
CROSSOVERCABLES,cont.
Inthecaseof10/100BASE-TEthernetconnectivity,theproblemisidentical--exceptthe terminologyisdifferent.
continued ....
Data Communications & Fieldbus Systems
D-736
92
CROSSOVERCABLES,cont.
ThestandardEthernet100/10Base-T connectorconfigurationisillustratedin theboxtotheright. ThedesignersoftheUniversalSerial Bus(USB)solvedthisproblemquite elegantlybyspecifyingtwodifferent connectors,completelyavoiding conflictbetweendeviceroles.
PINNUMBER 12345678
COMPUTER-ENDCONNECTOR
PERIPHERAL-ENDCONNECTOR
D-737
93
WIRELESSSENSORNETWORKS
Achievinganeffective,economicalconnectiontosensorsand/orcontrollersisanon-goingproblem forindustrialnetworkdesigners.
ZigBeeisactuallyasetofnetworking,securityandapplicationsoftwarestandardsthatsitsatopthe 802.15.4low-datawirelessstandardapprovedbytheIEEE.
K-077
94
The industrial computing industry has used several analog instrument interfaces -- or analog inputs (AI) -- for the past number of years. A partial list of signal types/interfaces that are used is:
4 - 20 mA current loop signal (overwhelmingly the most commonly used analog interface signal type)
0 - 5 Volt signal (an interface type that is generally difficult to be successful with, due to the need to maintain a high impedance interface -- to minimize voltage loss in associated cabling -- which in turn increases vulnerability to noise induction)
frequency and pulse rate signal (often referred to as accumulator signals, commonly associated with flow rate measurement)
resistance value (commonly associated with thermocouple-based temperature measurement -- which is typically limited by power dissipation limits in the thermocouple to millivolt-level signals, which are very vulnerable to noise problems)
It is a significant engineering challenge to design data acquisition systems so that ambient electrical noise is controlled, so as to preserve the usefulness of the signals being measured (typically accomplished through use of shielded wiring and low impedance circuits), while respecting the need for intrinsic safety in many applications.
K-023
95
Analog instruments have historically supported a standard 4-20 mA current loop interface to associated control equipment -- over a dedicated 2-wire shielded cable pair.
This current loop prevailed in the industry for many years, in spite of several distinct disadvantages:
The low levels of current required are subject to interference distortion problems that limit the accuracy and confidence of measured results.
The dedicated (typically shielded) cable pair required to support each instrument interface results in a large, bulky cable plant, which is expensive to install and difficult to maintain --especially in hazardous environments.
Cable terminations are also bulky, resulting in the cable termination cabinets and racking occupying typically more control room space than the functional systems they serve.
The unsophisticated characteristics of the current loop interface (e.g. no opportunity for carrying test or calibration signals over the dedicated cable pair without disturbing the measured results), making maintenance and calibration procedures very disruptive.
The traditional current loop interface, and the associated hub-and-spoke cabling configuration is not capable of supporting sophisticated microprocessor-based instruments, without adding greatly to the bulk and expense of the cable plant.
K-014
96
INTELLIGENT INSTRUMENTS
Intelligent, or "smart" instruments can be defined in several ways, but their key differentiating features are:
They support a sophisticated interface to associated control systems which enable a bus-cabled wiring system.
They are capable of performing self-calibration procedures and test functions non-intrusively (without disrupting the process being monitored).
The concept of intelligent instrumentation led to the industry initiatives to develop a fieldbus standard, enabling the bus-connection of numerous devices in an industrial network, in many ways the equivalent of an office local area network.
The fieldbus initiative has been plagued by the proliferation of numerous fieldbusses, all essentially incompatible -- leading to considerable cynicism as to whether a real inter-operable, industry consensus standard will ever emerge.
The so called Foundation Fieldbus shows promise as the possible solution to this difficulty, but it will take time to determine if it will yield the desired result.
K-015
97
INSTRUMENTS
InstrumentationisasubjectquitedistinctfromSCADA,andincludesitsownuniquebodyof technicalknowledgeandconventionalpractices.
Instrumentsalwaysmeasuresomesecondaryelectricaleffectthatisconfiguredinsome waytobedependentontheprimaryphenomenonbeingmeasured.
K-055
98
FLOWMEASUREMENTCONCEPTS
Twoofmanyfluidflowmeasurementschemesareillustratedhere:
Orifice(PressureDifferentialMeter)
PRESSURE DIFFERENTIAL METER PRESSURE SENSORS
FLOW DIRECTION
ORIFICE
DISPLACEMENT METER
FLOW DIRECTION
IMPELLER
K-056
99
Day Two
100
LAYEREDARCHITECTURESYSTEMCONCEPT
SYSTEM "A"
APPLICATIONLAYER PRESENTATIONLAYER SESSIONLAYER TRANSPORTLAYER NETWORKLAYER LINKLAYER PHYSICALLAYER
END USER
END USER
APPLICATIONLAYER
SYSTEM "B"
TERMINOLOGY/CONCEPTDISCUSSION:
Theyarenotabletocommunicatedirectly,so theydelegatethetasktounderlyingsystem(s).
MODEM
Thus,theactualinformationflowfollowsa"U" shapedpathasshown.
D-247
101
PRESENTATION LAYER
SESSION LAYER
Network layer functions apply to the task of managing resources by controlling switching and routing functions involving potentially many different links, so as to create an end-to-end coherent network. A common addressing scheme is essential at this layer, enabling network level flow control, congestion control, and common routing procedures. Examples are X.25 packet layer protocol and the IP portion of the TCP/IP protocols.
TRANSPORT LAYER
NETWORK LAYER
Link layer functions apply to the task of organizing data bits into a structured format (and in the process attending to error control, flow control, link management and data integrity). Examples of representative link layer protocols are BISYNC, HDLC, CSMA/CD, etc.
PHYSICAL LAYER
Physical layer functions apply to electrical bit level characteristics of the physical interface. Examples of representative standards are EIA-449, EIA-232-C and EIA-232-D.
D-303
102
The application layer manages communication between application processes. Examples of application layer functions are SMTP, FTP and Telnet functions in TCP/IP.
PRESENTATION LAYER
SESSION LAYER
The presentation layer defines the structure of data -- effectively ensuring that data syntax requirements are taken care of. Presentation layer functions provide for code conversion (e.g. ASCII to EBCDIC) and graphical user interface (GUI) support.
TRANSPORT LAYER
NETWORK LAYER
Session layer functions apply to the task of initiating, managing and terminating sessions, and support different message sizes and flow mechanisms (FDX, HDX), and provide for recovery in event of lower layer failure. An example of session (along with transport and network) functions is NETBIOS.
PHYSICAL LAYER
The transport layer sends data to the network layer in segments sized to suit the network's maximum transmission unit, or MTU limit. The transport layer is principally responsible for end-to-end flow control, and message packetization/segmentation and reassembly.
D-304
103
OPEN SYSTEMS
The issue of OPEN SYSTEMS speaks to the desire to work with information systems that are:
DOWNSIZING
OPEN SYSTEMS
capable of interoperation supportive of application and data portability scalable across a range of platform sizes, vendors and configurations
CLIENT/SERVER PARADIGM
Key concept: CLIENT/SERVER empowers the desktop systems and along with that, superior:
CLIENT/SERVER SYSTEMS
The primary thrust of new information systems development effort focuses on the area in this diagram that is common to all three concepts.
performance, to support the GUI economy, using the available PCs data management features of RDBMS
That is, appropriate Information Systems design strategies typically embrace ALL THREE concepts.
D-335
104
COMPARISONOFISSUES: PROPRIETARYvs.OPENSCADATECHNOLOGIES
PROPRIETARY Systemstendtobesimpler GOODNEWS FEATURES Systemstendtobeefficient "Securitybyobscurity" SCADAoperationsismore autonomous Implementationsareveryinflexible BADNEWS FEATURES Staffhaslimitedscopeinmodifying details/features Uncompetitive--singlesource prices(especiallyafter-market).
OPEN
Multi-vendorenvironmentresultsin competitiveprices
Commoditymanufacturingresults inbettercomponentcosts
Increasedsecurityrisks
K-084
105
2. The number of twists per foot will be a factor in determining a cable's suitability for LAN service (Conventional UTP has 1 twist per 18 inches, high performance cables have 1 twist every 6 inches or less).
3. The type and dimensions of insulation will determine cable capacitance; commonly low capacitance cable is specified in demanding applications, such as high speed LAN installations.
The purpose of manufacturing cable with a helical twist to the individual cable pairs is to minimize electromagnetic energy transfer to and from the cable pairs (this energy transfer is the principal cause of crosstalk, and other performance-limiting interference).
Shielded Twisted Pair (STP) cable is a related technology, and although technically superior to UTP, there is minimal interest in using STP due to cost, physical bulk and installation difficulty.
D-132
106
SUMMARY--TWISTEDPAIRCABLE LIMITATIONSWITHFASTETHERNET
BANDWIDTH
100MHz
300MHz
300MHz
1Gbps 1Gbps
100m 100m
350MHz
750MHz
D-732
107
INSTALLATIONPRACTICESTHATMAYIMPAIR PERFORMANCEOFCATEGORY5UTP
Thisissuewillbecomemoreacuteasorganizationsmoveawayfromtraditional10Mbps Ethernet,to100MbpsandhigherrateEthernetsand/orATM.
Someinstallationpracticesthatarecapableofcausingbelow-standardperformanceare: Excessivelytightcablebendingradius
D-733
108
(a) Contains the desired electromagnetic energy, or signals (i.e. limits radiation of signals -- which can cause interference and security threats).
(b) Excludes undesired electromagnetic energy, or signals from entering the cable and conflicting with desired signals (i.e. limits incoming interference signals).
Coaxial cable is specified in terms of impedance; a characteristic determined by the relative diameters of the conductors, and measured in Ohms (e.g. RG-58 cable is 50 Ohms). It is always critical to respect the specified cable impedance in coaxial cable applications.
D-131
109
FIBEROPTICCABLECHARACTERISTICS
THEREAREMANYDIFFERENT PACKAGINGTECHNIQUESUSED WITHFIBEROPTICCABLE --SOMEAREMUCHMORE COMPLEXTHANTHEONE ILLUSTRATED
PACKING MEDIUM CORE FIBERS (SILICA)
SILICONE COATING
BUFFER JACKET
Thereareprincipallytwotypesoffiberopticcable (andanumberofvariationswithineachcategory):
MultimodeFiber:
D-164
110
FIBEROPTICCABLECROSS-SECTIONS
Byfar,themajorshareofthebulkinfiberoptic cablesisassociatedwiththefeaturesofthe cablethatprovidestrengthandtoughness. Thesketchesattherightshowthreedifferent cablefabricationconcepts--eachwithexactly thesameopticalfibercapacity. Ineachcase,therearejustfouropticalfibers (atthe12,3,6and9o'clockpositions) adjacenttotheverycentercorewire. Thelargest,armoredcableis64.7mm (2.55 inches)indiameter;themid-sizedcableis36.8 mm(1.45inches)--andthesmallest,minimally sheathedcableis21mm(0.83inches) diameter. Electricalpowerisfeddownthecablefrom shorestations,usingacopperfoilsheath wrappedaroundthesteelwires (blackouter ring,lowerdiagram)-andseawaterasthe returnpath.
D-768
111
To understand the pulse dispersion effect, consider two light rays traveling along an optical fiber -Indicated here as the gray dashed arrow representing a direct, axial light ray, and the black, solid arrows which reflect off the fiber sidewalls -- representing a less optimally aligned light path:
These two light pulses start off at the same instant, but follow paths of different lengths, and thus arrive at different times.
When they recombine at the optical detector, the resulting sum of all input pulse components has become distorted by being broadened as a result of pulse dispersion; in reality there are a vast number of light ray components, resulting in a more uniformly dispersed light pulse than this simple example would suggest (as indicated by the smooth dotted line curve).
Note that both the distance traveled (cable length), and pulse duration (data rate) affect the amount of pulse dispersion encountered. Pulse dispersion thus limits the product of data rate and cable length -and fiber cables are specified in terms of megabit-per-second-meters.
D-243
112
113
FIBER SPECTRUM
114
FREESPACEOPTICS
FreeSpaceOptics(otherwiseknownasopticalwireless)isanopen-airlasertechnologycapable ofsupportingeffectivehigh-capacityshort-rangecommunicationsapplications.
Severalconfigurationsarefeasible,includingpoint-to-point,point-to-multipoint,andmeshed topologies.
Operatingdataratesofuptoabout10Gbpsarefeasible--althoughmorecommonlysystems operateatmegabitrates.
Operatingdistancesofuptoabout4km.arepractical,dependingonatmosphericconditions.
continued ....
VISIONSAFETYFACTORS
FSOsystemsoperateintheinfraredspectrum andgeneratenoenergyinthevisiblelight spectrumwherehumanvisionismostvulnerable. Powerlevelsareverylow--typicalFSOlasers operateatabout10mW(comparewithlaser pointersthattypicallyoperateatabout5mW.)
TYPICALTERMINALDEVICE
Copyright
W-163
(40508)
115
FREESPACEOPTICS,cont.
AdvantagesofFSOTechnology
Usesunlicensedelectromagneticspectrum;noregulatorydelaysorfeesareaconcern.
Equipment/infrastructurecostisverylowcomparedwithmicrowaveorfiberandelapsedtimeto operateisverybrief--typicallyinstallationandtestingtimeisafewhours.
W-164
(30330)
116
Microwave radio signals are really like any radio communication signals; two key features differentiate microwave radio from lower frequency radio applications:
Since microwave systems operate at a very high frequency, the distance the signal can travel during one cycle of the signal frequency (the wavelength) is very short -- literally a "micro-wave". This very high frequency of operation makes it possible to carry more end-user signal information (such as voice channels or data signals) than lower frequency systems.
The propagation characteristics of microwave radio are such that only line-of-sight communication is feasible (contrast with shortwave radio, which can propagate beyond the horizon).
Microwave systems typically communicate with carrier frequencies in the range from about 2,000,000,000 Hz (2 GHz) to around 15,000,000,000 Hz (15 GHz). Frequencies above about 10 GHz are adversely affected by rain and fog in the path.
Physical separation distances between microwave radio stations are typically limited by the curvature of the earth, and the need to maintain sufficient vertical clearance at the middle of the path (typical maximum feasible path lengths are about 40 miles, or 65 km.)
DESIRED, DIRECT SIGNAL PATH WHERE SMOOTH EARTH OR WATER SURFACES ARE LOCATED AT MID-PATH, TROUBLESOME REFLECTIONS OCCUR, CAUSING INTERFERENCE
D-244
117
Geosynchronous satellite orbits (which permit a satellite's position in space to appear stationary from earth) require that satellites orbit eastward over the equator (in the direction of earth's rotation) at an altitude exactly consistent with a 24 hour period.
There are a limited number of practical "parking spots" in space; satellites cannot be positioned any closer than about 2 degrees of longitude separation -- assuming moderate cost earth-based antennas. Actually, this 2 degree separation is a time separation -- all of these satellites share an identical orbit -- separated by about 8 minutes and 780 miles.
The area of coverage of a satellite is called its footprint, the shape of which is custom designed for each nation's geographic shape (see illustration).
International satellites have been typically operated by the Intelsat organization (by international agreement) -- although competitive international services now exist. Communication within one country (or regional service) is managed by a domestic authority.
D-245
118
GEOSYNCHRONOUS COMMUNICATION SATELLITES ARE ALWAYS POSITIONED OVER THE EARTH'S EQUATOR
NOTE: Earth diameter and satellite separation distance are scaled approximately correctly.
Earth-to-satellite distance = 35,888 km, or 22,300 miles, which results in an earth-to-satellite propagation time delay of 125 milliseconds. Radio signals propagate at the speed of light (300,000 km, or 186,000 miles per second).
Minimum round-trip time delay (the time it takes for a signal to travel from a point on earth up to the satellite, down to another point on earth, and then back again via the satellite to the point of origin) is 4 x 125 = 500 milliseconds.
Actual round-trip time delays are greater than the above estimate for three reasons:
1. Earth stations are rarely located on the equator. Latitudes other than zero increase the path length (note that the dotted line to a northern point is longer than the dashed line to the equator). 2. A similar longitude offset increases path length (The two earth stations and the satellite will never be all three on exactly the same longitude). 3. Signal processing delays in electronic systems (the two earth stations and the space platform). Thus a typical round-trip time delay budget is 650 to 750 msec.
D-183
119
Traditionally, radio communication systems have been engineered to maximize the range of effective coverage from any transmitter site -- resulting in poor use of radio frequency spectrum.
Cellular radio technology (referred to as "mobile" service in many parts of the world) was the first break with this tradition, where the radius of cell coverage is designed to be deliberately limited.
Cellular systems require a greater number of transmitter sites, and some kind of interconnection -but have the key advantage of a remarkably superior efficiency of spectrum use, due to the fact that frequencies can be reused without interference in cells that are separated from one another (thus greatly increasing total system capacity).
Capacity can be further increased by subdividing cells (to a limit of down to a few hundred meters radius).
A maximum power, maximum height transmitter will command a coverage radius of perhaps 50 km (30 miles) -but there will be no other service able to use the same frequency in the same area. Multiple low power transmitters will cover the same service area, but it will now be possible to re-use frequencies several times in a given coverage area.
D-262
120
WIRELESSDATANETWORKING TECHNOLOGY-OVERVIEW
Wirelessdatanetworkingapplicationsfitintotwobroadcategories:
LowSpeed,WideAreaTechnologies
Anumberofproductsareonthemarketthatenableintegratedaccesstowideareadatanetworksfor mobile/cellularorportableandtrunkedradiousers.
Operatingspeedistypicallymeasuredintensofkilobitspersecond(practicallylimitedtoabout 50kbps),andfreedomofphysicalmovementisamajorbusinesspriority.
Applicationsincludecourierservices,lawenforcementandvehicledispatch/fleetmanagement.
HighSpeedLocalTechnologies
Therangeofoperationofthisclassofequipmentistypicallylimitedtoadistancesintheorderofone hundredmeters--andisverydependentoninteriorbuildingstructures.
continued ....
D-606
121
WIRELESSDATANETWORKING TECHNOLOGY-OVERVIEW,cont.
Tremendousprogressincostreduction,compactpackagingandimprovedbatterylifehascreatedmany opportunitiesthatwerepreviouslynotfeasible.
Asamplerofthemanyapplicationswhichmakegooduseofthetechnologyinclude:
Industrialapplications-Complexityinmanagingawiredinfrastructureandchallengesincalibrating instrumentationarguestronglyforwirelessnetworkingsolutions.
Healthcare-aroamingpen-basedorpalm-topcomputercanbeusedtoenterpatientdataand/or accessexistingdatarecordsfromthebedside.
Extendingofficenetworkfunctions-Thin-clienttechnologyenablesfullnetworkedWindows functionalitywiththe10'sofkilobitdataratessupportedby2Gpublicwirelesspacketnetworks.
D-607
122
THEAPPROXIMATERELATIONSHIPBETWEEN SOMEWIRELESSTECHNOLOGIES
WWAN (Cellular)
INCREASINGRANGE
WMAN
WLAN
WPAN
0.01
0.1 1 10 INCREASINGDATARATE(Mbps)
100
1000
W-332
123
MOBILEDATACOMMUNICATIONSYSTEMS
SAFETY
TAXI
MAIN SERVER
COMM'N CONTROL
DCE
DCE
TX/RX
TRANSIT
INDUSTRIAL
D-295
124
MOBILEDATACOMMUNICATIONSYSTEMS,cont.
DatatransmissionratesoverwirelessinfrastructureareconstrainedbygenerallylimitedRF bandwidth--andthetypicallyhigherrorrateoverthistypeofchannel.
Wheredataapplicationsaredeliveredoveravoice-orientedframework,themaximumfeasible dataratesaretypicallylessthan20kbps--commonlyinthe10to12kilobitrange.
D-296
125
CELLSIZEISHIGHLYVARIABLE
Microcells:Coverageareastypicalofmobile/cellulartelephonesystemsinverydenseurban environments.
Macrocells:Coverageareastypicaloflessdense,suburbanmobile/cellulartelephonesystems.
Ratherthanbeingdeterminedbycellsize,inter-cellmobilityisdeterminedbythefeaturesofthe systemarchitectureandfeaturesoftheapplicationsbeingserved.
Copyright CANOVA SYSTEMS CORP.
1)
126
MULTIPATH PROPAGATION
Whenever more than one comparable strength propagation path exists between a transmitter and receiver (which is typically the case) the two (or typically more) signals will arrive at the receiver with differing phase values.
The destructive interference effect of these multiple signals arriving over different paths is multipath propagation, and results in multipath fading varying over time.
Radio paths between stationary devices will demonstrate multipath fading effects that are described by the Rayleigh statistical model -- so called Rayleigh fading.
Radio paths between devices in motion (one end of the path or both) will similarly exhibit Rayleigh fading statistics -- but in this case variable in both space and time.
It is not possible to engineer a particular path -- highly variable conditions prevail as a mobile unit moves around its service area.
Omnidirectional antennas are unable to discriminate between the preferred and secondary path signals -- and thus offer no relief for the multipath problem.
Signal reflections off major buildings and other structures in urban environments typically create serious multipath fading problems.
W-073
(00105)
127
Multipath fading phenomena are a greater problem with narrow bandwidth signals, since the sharp nulls which reduce signal strength below usable levels are very phase specific, and the sharpest, most troublesome nulls will only occur in a very narrow bandwidth.
Planning radio signal strengths and propagation effectiveness in an urban environment is challenged by the unpredictability of multipath-induced signal fades, variability of terrain which distorts median signal strength, physical obstructions (man-made and natural), and frequently, exposure to high noise levels caused by automobiles, power infrastructure, signs, etc.
Observable nulls in received signal strength are spaced in the order of 1 wavelength apart. Occurrence of multipath fading events is highly variable; greater regularity is observable when a small number of signals are present.
05)
128
The depth of multipath fades (and associated error consequences) is affected by a number of factors, including propagation effects and carrier frequency of operation.
Perhaps the most critical issue is the occupied bandwidth of the signal being viewed, since extremely deep multipath fades are only observed with very narrow bandwidth carrier signals.
The accompanying time-domain view of multipath signal strength variations is overlaid with a sequence of four packets, where the horizontal scale is implicitly assumed to be time-calibrated. It is evident from this sketch that some packets will be hopelessly corrupted by the noise burst that accompanies deep fading events (probably packets A and D in this example). Since effective performance in marginal coverage areas will always be a key operating priority, a signal processing and error mitigation strategy that is effective through these fading events is key to the success of data transmitted over mobile wireless channels.
0
AAAAAAAAAAAAAAAAAAA AAAAAAAAAA AAAAAAAAAA AAAAAA AAAAAAAAAAAA AAAAAAA AAAAAAAAA AAAAAAAAA AAA AAAAAAAAAA AAAAAAAAAA AAAAAAAAAA
129
The most capital-intensive portion of a traditional telephone system is the local loop infrastructure -- the subscriber lines.
The numerous configurations and equipment concepts that can make telephone and enhanced services available without traditional copper (or fiber optic) cable connections are referred to as subscriber radio or wireless subscriber systems. Three major application areas are developing for this class of equipment:
In underdeveloped countries, the cost of conventional cabled technology is cost-prohibitive, and impedes other development efforts -- wireless offers an instant telephone solution, and is very cost-effective even in sparse communities.
Many developed countries are moving toward deregulation of local access telephone service. Competitors to the traditional telephone companies are anxious to offer state-of-the-art services without the incredible cost and delay in getting to market with a cabled product.
These emerging technologies support enhanced service possibilities parallel to conventional wired technology -- enabling traditional, established telephone companies to position themselves for a more competitive market.
130
WIRELESSSUBSCRIBER TECHNOLOGY,cont.
Theneedtoconserveradiofrequency spectrumdictatesthattheconceptof cellularfrequencyreusemustprevail (minimumradiusofcoverageper transmitter,manylowpowercellsites abletoreuseradiospectrumthroughout anurbanarea...). Thetrunkdistributionsysteminan alternate,competitivecarriermodelof servicecouldbeservedbymicrowave radiolinksorfiberopticcablesystems Inurbanapplications,thesubscriber radiolinksareveryshort--measuredin afewhundredsofmeters. Inunderdevelopedcountries,the subscriberlinksmaybemuchlonger, sometimesuptotensofkilometers.
3
D-648
(10831)
131
THEAPPROXIMATERELATIONSHIPBETWEEN SOMEWIRELESSTECHNOLOGIES
WWAN (Cellular)
INCREASINGRANGE
WMAN
WLAN
WPAN
0.01
0.1 1 10 INCREASINGDATARATE(Mbps)
100
1000
132
ULTRAWIDEBAND(UWB)RADIO
Anemergingclassofshort-range,veryhighdataratewirelessservicehascometobeknown asultrawideband,orUWBradio--consistentwiththeIEEE802.15.3aspecifications.
3)
133
IEEE802.16--WIMAXFEATURESUMMARY
TheIEEE802.16standarddefinesawirelessmetropolitan-area(WirelessMAN)networking technology--awireless"lastmile"solution.
Suitableforfixedlinkserviceand"nomadic"users(i.e.notadynamicallymobileservice)
ThetechnologyisIntendedasacommunity-wideservice,suitedtodeliveringtelephony, broadbandInternetandcable-televisionstyleentertainmentprogramming.
Oneapplicationthatisattractingalotofinterestistheuseofthisnewtechnologytoprovide connectivityto802.11hotspots,creatingacommunity-widewirelessLANnetwork.
W-266
(31211)
134
TX
3 2 1 3 2 1
RX
1
2
2
1
2
1
3
RX
3
3
SOURCE DEVICES SHARE RESPONSIBILITY FOR TIME SLOT COORDINATION (VIA A MULTIPLE ACCESS CONTROL SYSTEM)
135
Multiplexing is a fundamental feature of many communications system, involving the ability to share a communications channel of some type by a number of simultaneous users.
Generally, the term 'multiplexing' is used to describe a channel sharing environment where all parties share a small number of common physical locations -- as in a point-to-point service.
A very closely related term is 'multiple access' which applies to a multiplexed channel environment where the user population is disbursed over a broader service area, as in the case of population of mobile radio users.
Thus, the various channel sharing techniques used to equitably and efficiently share radio spectrum amongst a population of mobile users are referred to as multiple access techniques.
There are four such sharing mechanisms that are important to wireless services, based on sharing in the space dimension, in the time dimension, in the frequency dimension and in a code dimension:
The structuring of cellular wireless systems with deliberately defined and managed physical dimensions to each cell is essentially what SDMA is all about.
SDMA makes it possible for multiple users to occupy the same frequencies at the same time in a common service area by deliberately restricting each to a different subset of that service area.
136
Multiplexing in the frequency domain is perhaps the oldest, most familiar channel sharing mechanism.
With FDMA, individual users are assigned different, non-interfering frequencies of operation.
The familiar spectrum management scheme used in broadcast services is FDMA, where each broadcaster operates in a coordinated, protected frequency segment of the available spectrum.
Time division multiplexing techniques yield efficient, effective channel sharing -- strictly in context of digital communication.
The essential concept of TDM/TDMA systems is that the entire channel is assigned to each user sequentially, so that operation is characterized by intermittent, high rate bursts of data being sent.
TDM/TDMA systems are highly flexible, making it possible to dynamically assign channel resources on an 'as required' basis; busy users receiving a larger share of shared capacity than less demanding users.
137
The newest, and therefore least familiar channel sharing technology is CDM/CDMA.
With CDMA, each individual user's data stream is modulated with a unique code -- the uniqueness of which is the basis of channel sharing.
The ability of spread spectrum systems to support multiple simultaneous, co-channel users is keyed to the concept of each user using a unique chipping code that is orthogonal (see accompanying box) to that used by all other simultaneous users.
ORTHOGONALITY
Orthogonality is a mathematical concept, in which two sets of data values are independent of each other. The three axes of three-dimensional space (x, y and z) are orthogonal; the projection of each axis on the other two is unconditionally zero.
Expressed in matrix algebra terminology, two sets of data are orthogonal to each other if their dot product is zero.
and
138
Each of the medium-sharing, multiple access mechanisms are imperfectly managed, such that guard space is required between each of the concurrent users. This guard space becomes, for each multiple access mechanism:
With SDMA, overlapping radio coverage forces a physical separation between user groups.
Unique frequencies are the basis of FDMA operation -- and in this case individual users must be assigned frequencies sufficiently separated so as to minimize adjacent channel interference.
With TDMA, absolute control over timing of individual devices' data bursts is impossible, and a guard-time (gap) between individual transmissions is necessary to manage interference.
The selection of the code space used for CDMA is a highly mathematical, complex process -- but the mathematical concept of code distance is the analog here.
139
Frequency Division Multiple Access (FDMA) involves a number of channel management schemes -- all of which separate users in the frequency domain (i.e. assign individual users unique, non-interfering frequencies)
Most FDMA systems and frequency allocations are designed for full duplex, telephony-style (two open, continuously available channels) operation where the transmit and receive signals are assigned different frequencies -- a configuration known as frequency division duplex, or FDD.
As in the example below (based on the 900 MHz GSM frequency plan) each pair of frequencies is typically offset by a uniform value -- here 45 MHz.
FREQUENCIES AND CHANNEL ALLOCATIONS SHOWN ARE CONSISTENT WITH 900 MHz GSM 124 FDD CHANNELS ARE ASSIGNED IN TWO 25 MHz. FREQUENCY BANDS -- WHICH ARE SEPARATED BY A 20 MHz GAP. EACH SIDE OF THE FDD CHANNEL NOMINALLY OCCUPIES 200 kHz, THE FDD CHANNEL-PAIRS ARE SEPARATED BY 45 MHz.
TRANSMIT-FROM-MOBILE
1 2 3 4 123 124
TRANSMIT-TO-MOBILE
20 MHz SEPARATION 890.2 MHz 915 MHz
1 2 3 4 123 124
935.2 MHz
960 MHz
140
Time Division Multiple Access (TDMA) provide for time-coordinating a given digital communications medium so that multiple users are able to share this common resource.
TDMA offers a much higher level of flexibility than FDMA, in that time slots allocated to individual users are not necessarily all of the same duration, or repetition rate, nor are they necessarily managed in a highly structured way (e.g Aloha is the ultimately unstructured access method).
Time Division Duplex (TDD) is the term used for use of the same RF channel for both transmit and receive functions (see illustration below) -- the essential feature is a manageable, predictable time separation between the transmit and receive functions.
TIME-SLOT ALLOCATIONS SHOWN ARE ONLY TYPICAL OF TDMA OPERATION (THE DIFFERENT TDMA STANDARDS [e.g GSM, IS-54, IS-136, etc.] GENERALLY USE DIFFERENT NUMBERS OF TIME SLOTS) IN THIS EXAMPLE, WHICH IS CONSISTENT WITH THE DECT CORDLESS PHONE SYSTEM, 12 TIME SLOTS ARE ALLOCATED FOR ONE DIRECTION OF COMMUNICATION, FOLLOWED BY 12 IN THE REVERSE DIRECTION -- WITH A FRAMING PATTERN THAT REPEATS EVERY 10 ms, resulting in 417 s INDIVIDUAL TIME SLOTS. BY SEPARATING THE TRANSMIT AND RECEIVE FUNCTIONS IN TIME, WITH THIS TDD FEATURE, IT IS POSSIBLE TO SHARE RF CIRCUITRY BETWEEN A DEVICE'S RECEIVER AND TRANSMITTER.
TRANSMIT-FROM-MOBILE
1 2 3 11 12
1
TRANSMIT-TO-MOBILE
2 3 11 12
141
SPREAD SPECTRUM
Spread spectrum technology was developed as a U.S. military initiative, the key elements of which were declassified in the 1980's.
A number of features of the technology that were attractive to military communications are similarly attractive in civilian applications -- see accompanying box.
Spread spectrum technology deliberately modulates a transmitter's spectrum so as to occupy considerably more bandwidth than would be required using conventional modulation techniques.
On the surface of it, this tactic would seem to wasteful of spectrum -- but spread spectrum is arguably more spectrum-efficient than more conventional channel sharing technologies, like FDMA and TDMA.
A key element of spread spectrum technology, as adapted to civilian applications is the ability to efficiently share spectrum with multiple simultaneous users -each of which uses a unique SPREAD SPECTRUM FEATURES frequency spreading code NET CIVILIAN BENEFIT INITIAL MILITARY OBJECTIVE sequence that ensures each user's signals are recoverable.
Jam-proof communication
Interference immunity
The resulting channel sharing scheme is Code Division Multiple Access, or CDMA.
Superb security
142
With DSSS, a high rate pseudo-random binary code (called the chipping code) is Exclusively OR'd against the data stream -- resulting in a similarly high rate binary sequence that is directly modulated as a wide band carrier signal.
Chipping code sequences are in fact not random -- they are carefully chosen sequences that demonstrate a high degree of orthogonality with all other simultaneously used chipping codes.
In civilian applications, chipping code rates are typically between 10 and 100 times the data rate (in military applications the number is more typically 10,000), resulting in a channel spreading factor (bandwidth increase) of a similar number.
With FHSS, the available spectrum is divided into a number of discrete channels -- much as is done with FDMA. Transmitters and receivers operate in a carefully orchestrated sequence, pseudo-randomly selecting first one and then another channel on which to communicate a burst of data.
With slow hopping FHSS, the carrier stays on frequency long enough to transmit several bits of user data (slow FHSS is a standard option with GSM systems) -- and with fast hopping FHSS systems, the carrier frequency changes several times per user data bit interval.
143
Direct Sequence Spread Spectrum (DSSS) systems perform an exclusive-OR between the user's data sequence and a spreading code chipping sequence, as illustrated in the accompanying box.
The degree of frequency spreading is the spreading factor -- which is the ratio of the chipping sequence rate to the user data rate (in this simple example the chipping sequence is the 7-bit data pattern 1001010, so the spreading factor is 7:1).
The modulation spectrum is spread, or widened by the spreading factor; that is, with all other system features unchanged (e.g. modulation and filtering strategy) the spread signal will be wider than the single channel spectrum by this factor.
USER DATA
A
0 11 01 01 1 00 10 10
CHIPPING SEQUENCE
1 00 10 10 1 00 10 10
0 0 1 1
0 1 0 1
0 1 1 0
144
This and following pages illustrate the concepts of a DSSS CDMA signal processing sequence:
1 Data Encoding Sequence: (See 'Data Encoding Sequence' illustration next page)
Prior to commencing this transmit sequence, the CDMA system will have assigned the two distinct and orthogonal 6-bit spreading codes.
These two codes (101100 and 110110) can be shown to be orthogonal as follows: Consider the logical 1 value to have an arithmetic +1 value and the logical 0 to have an arithmetic -1 value -- then bit-by-bit, the product of these two codes is +1 -1 -1 +1 -1 +1 = 0
The two transmitted data sequences are obtained by applying the Exclusive OR function to the the source data and the spreading code -- and are sent at the rate of the spreading code sequence.
145
1 1 0 1 0 0 1 1 1 0 0 1 1 0 0 1 0 1 1 1 1
0 0 0 0 0 1 0 0 1 1 0
1 1 0
SOURCE DATA
USER #1
TRANSMITTED DATA
0 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 1 0 0
0 1 1 1 1 0 0 1 0 1 0 0 1
1 1 0
SOURCE DATA
USER #2
TRANSMITTED DATA
Notes: (1) 6-bit spreading codes used here are shorter than typically used -- but are orthogonal. (2) Transmitted data is Exclusive-OR of Source Data and Spreading Code
W-119
(01028)
146
0 1 1 0 1 1 0 1 1
1 0
1 1
0 1
0 0
0 0
1 0
0 1
0 0
USER #1 DATA
USER #2 DATA
+1 0 -1
Notes: (1) CDMA systems are very dependent on careful, accurate control of the signal strength recovered from all simultaneously received signals. (2) This is because the "Combined Received Signal" shown here is an additive value and it is vital that all contributing signals contribute a like amount of signal from each received symbol.
8)
147
3 Data Decoding Sequence: (See 'Data Decoding Sequence' illustration next page)
The data recovery sequence with CDMA is critically dependent on precise amplitude and phase coordination of incoming multiple access users, so that their signals can be simply combined (or added) in the receiver.
The resulting signal is a multistate sequence -- shown here as a three-state because of additively combining two 2-state signals.
The data decoding sequence involves calculating the bit-by-bit product of the incoming signal and each of the in-service spreading codes -- here just two.
It is convenient to assign an algebraic value to the binary spreading codes (i.e. view them as having values +1 and -1 compared with the more common 1 and 0 valuation).
The bit-by-bit products from this process are accumulated, or integrated to yield a soft decision process.
Note that the decision is not necessarily assisted by all bit intervals in the spreading code (i.e. some bit intervals do not contribute an increased signal accumulation, or integral), but over the source data bit interval (and therefore numerous spreading code symbol intervals) a usable binary detection process results.
W-127
(01028)
148
1
1 1 0 1 1 0 1
0
1
1
0
1
1
0
1
0
0
1
1
0
1
1
0
1
1
Note: Each user signal is integrated (accumulated) over the source data bit interval using the following calculation and the numeric calibration shown on the left margin above: -1 x (combined receiver signal) x (user spreading code)
W-121
(01028)
149
WIRELESSMODEMS
Variouspacketradiotechnologies(e.g.GPRSand1xRTT)andfrequencybands(e.g.850/1900 or900/1800MHz.).
Externalpowerrequirements--theytypicallyuseaDCpowerfeed(suchasisreadilyavailable forvehicle-mountedandSCADAapplications).
ManyofferanintegratedGPSreceiveroption,asisoftenrequiredformobileapplications.
DirectlyconnectedtelemetryportsforsimpleSCADAordatacollectionapplications.
W-311
150
WIRELESSMODEMS-SAMPLER
W-312
151
GROUNDPOTENTIALRISE
continued ....
ELECTRICALSUBSTATIONEXAMPLE
GROUNDINGGRIDSYSTEM, USUALLYSUB-SURFACE
RESIDUALEFFECTIVE GROUNDIMPEDANCE
K-035
152
GROUNDPOTENTIALRISE,cont.
Lightningcurrentsarenaturallylimitedinduration;inthecaseofanelectricalfailure,circuit breakerswillblowandeliminatethecurrentflow.
However,briefly,thesecurrentscaneasilypeakat10,000Amperesormore,allofwhichwill flowintothegroundingstructuresprovidedforthispurposeandthusintotheearth.
continued ....
Data Communications & Fieldbus Systems
K-036
153
GROUNDPOTENTIALRISE,cont.
Someonetalkingonthistelephonewouldexperiencetheabove-described10,000Voltpotential differencebetweentheirfeetonthegroundandhandsonthetelephone!
Inadditionto"off-site"issues,thereareotherhazardstopersonnelandequipmentthatare causedbynon-uniformityinthestationgroundingsystem:
Personnel"TouchHazards"--Similarly,thegroundunderone'sfootandapieceofstation steelinone'shandcanhaveenoughpotentialdifferencetobelethal.
K-037
154
ERRORSANDTHEIREFFECTONOPERATIONS
Detectederrorsmustbedealtwithreliablyandeffectivelysoastoavoidtheseundesirable consequencesandresultinglossofinformationintegrity.
Undetectederrorsmustbeeffectivelyeliminated--althoughtherewillalwaysbeamathematical probabilityoftheirresidualpresenceinreceiveddata.
D-279
(41102)
155
Parity calculations involve defining the value of a "parity bit" such that the original data segment plus parity bit will always have an even (or odd when using odd parity) number of 'one' bits in the now larger data segment. Even parity is more common than odd parity, and is illustrated here:
THE
0 0 1 0 1 0 1 1
QU I C K
0 0 0 0 0 1 0
1 0 0 0 1 0 1 1 0 0 1 0 0 1
B R OW N
0 0 0 0 0 1 0
FOX
0 1 0 0 0 0 1 0 1 0 0 1 0 1
1 1 1 1 0 0 1 1 1 1 0 1 0 1 0 1 1 1 0 0 1 0 0 0 0 0 1 0
. . .
0 1 1 0 0 0 1
1 1 1 1 0 0 1
0 0 0 1 0 0 1
1 0 1 0 0 0 1
1 0 1 0 1 0 1
1 1 0 0 0 0 1
1 1 0 1 0 0 1
0 0 0 1 1 0 1
0 0 0 0 0 1 0
0 1 1 1 0 1 0
0 1 1 1 0 1 0
0 1 1 1 0 1 0
1 0 1 0 0 1 0
0 1 1 1 0 1 1 0 1 0 1 1 1 0 1 1 1 1 1 0 0 0
VERTICAL PARITY BITS ARE APPENDED TO EACH CHARACTER, TYPICALLY BY ADDING AN 8th. BIT TO A 7-BIT ASCII CHARACTER. A CHARACTER-SEQUENTIAL VIEW OF DATA IS DESCRIBED AS HORIZONTAL, OR LONGITUDINAL, OR BLOCK SEQUENCE.
HORIZONTAL PARITY IS IMPLEMENTED AS A BCC (BLOCK CHECK CHARACTER) APPENDED TO THE END OF A BLOCK OF DATA. DATA BLOCKS ARE TYPICALLY HUNDREDS OF CHARACTERS IN LENGTH.
D-034
(60113)
156
Cyclic Redundancy error detection calculations involve doing a long division calculation on the data.
The remainder which results from this calculation becomes the block check data which -- is transmitted along with the desired information.
The receiving system repeats the identical long division calculation, and checks for consistency between the received CRC check data and the results of its own calculation.
The mathematical basis of CRC error detection is derived using polynomial algebra, where all binary sequences are regarded as polynomials for purposes of analysis. Example:
11010001 10011 110001100000 10011 ----10111 10011 ----01001 00000 ----10010 10011 ----00010 00000 ----00100 00000 ----01000 00000 ----10000 10011 ----0011
THE QUOTIENT IS THE INTEGER PORTION OF THE RESULT -- THE VALUE OF WHICH IS OF NO PARTICULAR SIGNIFICANCE HERE.
THE DEFINITION OF AN APPROPRIATE DIVISOR POLYNOMIAL IS A BLACK ART; IT WILL ALWAYS HAVE ONE MORE BIT THAN THE REMAINDER. USING POLYNOMIAL ALGEBRA NOTATION THIS 5-bit VALUE WOULD BE EXPRESSED AS: X+X+1
4
THE INFORMATION FIELD TO BE TRANSMITTED WILL BE MUCH LONGER THAN THIS 12 BIT EXAMPLE (REAL-WORLD DATA FIELDS IN HDLC FRAMES ARE COMMONLY 10,000 bits LONG)
THE REMAINDER, OR FRACTION PART OF THE RESULT, BECOMES THE BLOCK CHECK SEQUENCE (EQUIVALENT TO PARITY BITS) USED AT THE RECEIVING SITE TO VALIDATE THE DATA.
D-106
(61209)
157
CYCLICREDUNDANCYCHECK(CRC) ERRORDETECTION,cont.
SomecomparativeCRCpolynomialvalues (usedindifferentapplications)are:
4
Preceding(hypothetical)example:
X+X+1
16152
IBM'sEBCDICBISYNC:
X+X+X+1
16125
HDLC,SDLCandITU-T:
X+X+X+1
322623221612111087542
Ethernet:
X+X+X+X+X+X+X+X+X+X+X+X+X+X+1
Cyclicredundancyerrordetectiontechniqueshaveseveralkeyadvantages:
CRCsystemsaresimplyimplemented (typicallyinhardwareontheI/Ocard).
CRCerrordetectionhasbecometheindustryde-factostandarderrordetectiontechnique.
D-107
(30421)
158
There are two basic error correction techniques (and several variations of each). These are Automatic Repeat Request (ARQ) and Forward Error Correction (FEC).
Automatic Repeat Request requires that the receiving device validate incoming data to determine if there are any errors in the data:
If the received data appears to be correct, the receiving station will respond with a positive acknowledgement message -- sometimes just an ACK character.
If the data are found to be incorrect, the response will be a negative acknowledgement (possibly just a NAK character) rejecting the data as received.
The sending station thus becomes responsible for retransmitting the flawed data -- potentially more than once -- until it is received correctly.
Forward Error Correction systems operate without any feedback, or referral to the sending station.
Data are sent with additional redundancy, such that the receiving station has sufficient information about the message content to enable it to not only locate errors in data, but to make corrections.
Since FEC systems are limited as to the number of bits that can be reliably corrected, seriously degraded data channels will result in too many errors for dependable handling, delivering an unacceptable residual error rate.
D-112
(61209)
159
COMPARISON;FECvs.ARQ
AutomaticRepeatRequest:
Automaticrepeatrequestsystemspassalldatathroughauniformtestofaccuracy;thusno questionabledataiseveraccepted.
Thisisafavorablecompromisewithbusinesssystems,wherewewouldrathersufferathroughput and/ordelaypenaltythantrytodealwitherroreddata.
ForwardErrorCorrection:
Forwarderrorcorrectionsystemsmaintainaconstantpaceofdatatransmissionthatisinsensitiveto prevailingchannelconditions.
Sinceseriouslydegradedchannelconditionswillresultinanunusualfrequencyoferroroccurrences, systemperformancewilldeteriorate,duetoexcessiveuncorrectableerrorsinthedata.
FECsystemsthusdeliveraconstantdatathroughputrate,butofferuncontrolledqualityofdelivered data--anundesirablecharacteristicforbusinesssystems.
Forthisreason,FECistypicallydeliveredinconjunctionwithARQ--ahybridapproach.
D-113
(30421)
160
Automatic Repeat Request (ARQ) error recovery techniques are basic to business data communication systems.
Half-Duplex Protocols like BISYNC invoke a simple process, whereby ACK (acknowledge) or NAK (negative acknowledge, or reject) messages are anticipated in response to a transmitted message. This process is known as a Stop and Wait ARQ.
There is no ambiguity to this ACK/NAK reference, because there is never more than one message being handled at any given time.
Go Back-N ARQ provides for the receiving station to pass advice back to the sending station identifying the message number of the last correctly received message.
The sending station is then responsible for retransmitting all message segments that were sent subsequent to that specific message segment.
Selective ARQ provides for the receiving station to save in buffer all message segments -specifically including those received subsequent to a flawed data segment.
Thus, the sending station only needs to selectively re-send message segments identified by the receiving station as flawed.
D-108
(61209)
161
2 3
WAIT
WAIT
WAIT
WAIT
SENDING STATION
WAIT 1 2
AC K
AC K
AK
K AC K
AC K
AC K
AC
AC
AC
AC
AC
NA
AC
WAIT 1
AC K
AC K
NA K AC K
AC K
AC K
AC K
AC K
AC K
AC K
AC K
AC
AC
AC
AC
AC
AC
NA
AC
WAIT
A AT D D
D
A AT D
AC
AC
AC
AC NA
K A AT
A AT
A AT
WAIT WAIT
WAIT
WAIT
WAIT
ERROR
RECEIVING STATION
WAIT
SENDING STATION
TA DA
2 3 1 2 3
T DA A TA DA A T DA
WAIT
WAIT
ERROR
ERROR
RECEIVING STATION
WAIT
SENDING STATION
ERROR
ERROR
RECEIVING STATION
D-109
(30311)
162
Following is an illustration of a Hamming Code implementation. The Hamming Code is useful for illustration -- in practice, more powerful codes (e.g. the Reed-Solomon codes) would be used:
1 2 3 4 5 6 7 8 9 10 11
11 ACTUAL DATA BITS ARE MAPPED INTO 16 BEFORE BEING TRANSMITTED, ADDING 4 CHECK BITS (SHOWN BELOW IN GRAY NUMERALS)
0 0
1 2 3 4 5 6 7 8
0
9
0
10
1
11
1
12
0
13
1
14
1
15
DATA AS TRANSMITTED NOW HAS 4 EXTRA BITS -- BEING REDUNDANT INFORMATION TO BE USED TO FIND AND CORRECT ERRORS IN TRANSMISSION.
The four parity check bits (Bits 1, 2, 4 and 8) are defined using even parity calculations, involving a subset of the full sequence:
Bit No. 1 is calculated using Bits 3, 5, 7, 9, 11, 13 and 15 (i.e. all odd numbered bits -- or all members of every second group of one bit).
Bit No. 2 is calculated using Bits 3, 6, 7, 10, 11, 14 and 15 (i.e. all members of every second group of two bits).
Bit No. 4 is calculated using Bits 5, 6, 7, 12, 13, 14 and 15 (i.e. all members of every second group of four bits).
Bit No. 8 is calculated using Bits 9, 10, 11, 12, 13, 14 and 15 (i.e. all members of every second group of eight bits).
D-110
(61208)
163
1
15
-- TO --
0 0 1
4 5 6 7 8 9 10
1
11
1
12
0
13
1
14
1
15
ERROR
Assume there is an error in transmission, such that one bit is incorrect. Assume this is bit No. 9. To perform the Hamming error correction calculation, we start by checking to see if the parity calculations (involving bit positions 1, 2, 4 and 8) are still apparently correct.
There are four tests to perform -- designed to verify the correctness of each of these parity bits:
Bit Bit Bit Bit No. No. No. No. 1 2 4 8 was was was was defined defined defined defined with with with with
bits 3, 5, 7, 9, 11, 13 and 15; This set no longer has even parity and this test fails. bits 3, 6, 7, 10, 11, 14 and 15; This set still has even parity and this test passes. bits 5, 6, 7 , 12, 13, 14 and 15; This set still has even parity and this test passes. bits 9, 10, 11, 12, 13, 14 and 15; This set no longer has even parity and the test fails.
We enter these test results into a 4-bit register, using a "0" value for each test that passes, and a "1" for each failure. This resulting 1-0-0-1 bit pattern is the binary value of 9, which identifies the bit position we arbitrarily flawed. Thus, we have the necessary information to correct this single bit error condition.
1
1
0
2
0
4
1
8
9
Data Communications & Fieldbus Systems
D-111
(61208)
164
FORWARDERRORCORRECTION APPLICATIONEXAMPLE
ROUTERCONNECTEDLANSYSTEMSVIASATELLITELINK:
GEOSYNCHRONOUSSATELLITE
64kbpsDATALINKTOTHE APPLICATIONPROCESS
ROUTER
F E C
TX
TX F
E
RX
RX C
ROUTER
THEPRIMARYERRORCONTROL MECHANISMISFECBETWEENCODEC'S WHICHAREPARTOFTHE SATELLITEEARTHSTATIONS. SECONDARY,ORBACKSTOPERROR CONTROLISPROVIDEDBYARQLINK PROTOCOLSSUPPORTEDBYTHE DATATERMINALEQUIPMENT
D-035
(41102)
165
LANCONFIGURATIONCONCEPTS
BUSARCHITECTURE LANSYSTEM RINGARCHITECTURE LANSYSTEM
D-009
(99999)
166
RELATIONSHIP--IEEELANSTANDARDS TOOSILAYEREDARCHITECTUREDEFINITIONS
IEEELANSTANDARDS
OSISTANDARDS
HIGHER LAYERS
IEEE802.1--SYSTEMMANAGEMENT
IEEE802.1--OVERVIEW
NETWORK LAYER
DATALINK LAYER
802.15MAC
802.15 PHYSICAL BLUETOOTH PAN
802.16MAC
802.16 PHYSICAL WIRELESS MAN
PHYSICAL LAYER
NOTE:'MAC'REFERSTOTHEMEDIUMACCESSCONTROLFUNCTION,WHICHIS REGARDEDASASUB-LAYERINCONTEXTOFIEEELANSTANDARDS.
D-115
(99999)
167
SIXTECHNICALFEATURES OFLOCALAREANETWORKS
1.MEDIATYPE
Characteristicsofcommunicationsmediumaffectsystemcapacity,quality, costandeaseofuse.
OptionsincludeUTP,STP,coaxialcable,fiberopticcableandwireless.
2.ACCESSCONTROL
Accesscontrolmethods(protocols)affectsystemperformance,dimension constraints,costandcomplexity.
Optionsincludecontentionaccessmethods,suchasethernet'sCSMA/CD andcontrolledaccessmethods,suchastokenpassingmechanisms.
3.TOPOLOGY
4.HUBTYPE
Thereisascaleofsophisticationinhuboptions,rangingfromeffectively passivefirst-generationhubs,throughhighlyintelligentswitchinghubs.
5.SWITCHINGTYPE
Therewasatimeinthe1980'swhencircuitswitchingtechnologywasthought tobeanoptionofinterestbutpacketswitchingistheonlyoptionatpresent.
6.SIGNALTYPE
D-010
(99999)
168
Each user transmits whenever it has data to send, without any attempt at first determining system status (busy/idle), or developing any physical layer view of transmission success or failure. Relies on upper layer functions to validate message transmission success. Works well with lightly loaded systems; theoretical maximum capacity about 18% of channel capacity. Variant: SLOTTED ALOHA operation provides system-wide clocking to mark time slots, ensuring either 100% collision or 100% missed collisions (i.e. no partial message overlaps). CONCEPT: collisions result from two stations transmitting at the same (or overlapping) times, resulting in garbled transmissions.
CSMA operation provides for a system status check prior to initiating transmission. Thus, collisions will only occur as a result of effectively simultaneous decisions to transmit. Relies on higher layer functions to validate message transmission success, as with ALOHA. Once transmission is initiated, all transmissions run to completion (i.e. no detection of collisions) means no ability to abort failed transmissions.
D-086
(60506)
169
Concept is similar to CSMA, except CSMA/CD adds ability to monitor the progress and continuing success of data transmission, and take remedial action when collisions occur. Whenever collisions are detected (by each of the two contending stations), they respond by first sending a jamming signal to ensure no attempt is made to decode the garbled message content; Secondly, each station starts a random timer, which defines the waiting time before repeating the "listen-before-you-transmit" CSMA logical process. Assuming contending stations select different random timer intervals, two stations should not encounter subsequent conflict. In the event of multiple unsuccessful transmission attempts (due to collisions), the interval within which the random timer is set is forced to grow with each failure. It actually doubles with every repeated collision event -- for up to 10 occurrences, and tolerates a further 6 collisions prior to giving up.
CSMA/CA is similar to CSMA/CD, except a pre-specified scheduling sequence is invoked every time the channel goes idle, preventing simultaneous attempts at transmission. CSMA/CA is really a hybrid contention-controlled mode of operation due to this scheduling mechanism.
D-087
(60506)
170
LAN A
LAN B
Extend the physical dimensions, or range beyond the limits of a single LAN, or Expand the traffic carrying capacity beyond that of a single LAN, or
Increase the manageability and/or apparent reliability by segmenting the system into more manageable pieces. Equipment options include: Repeaters Bridges Routers Gateways
D-145
(50813)
171
Connectivity between LAN's is achieved with four different types of devices (and some hybrid solutions which share features of these four): REPEATERS
Repeaters provide a means of extending a single, contiguous LAN to distances that cannot be supported without such aid.
Repeaters simply regenerate the electrical (or optical, as the case may be) signals on the LAN; they do not interpret any portion of the message, and assert no discretion over forwarding data (ie. all packets are unconditionally forwarded).
Repeater connected LAN's remain a single logical LAN (e.g. limitations on the number of addresses supported on a LAN are not overcome by a repeater). BRIDGES
Bridges provide a way of transparently internetworking two logically distinct networks so as to yield the perception of having a single, contiguous wider area, higher capacity LAN than can be achieved with repeaters.
D-134
(60113)
172
Bridges are ideally suited as a means of interconnecting a small number of departmental LAN's (perhaps up to 3 or 4), particularly where the community of users are uniformly trustworthy (Bridges provide no administrative or security partitioning capability).
Most bridges are self learning, in that they construct an address table which associates the address of each device observed on each LAN with that LAN. They then use this table to make forwarding decisions -- such that only traffic addressed to a device on the alternate LAN passes through the bridge. ROUTERS
Routers are suited to interconnecting a larger number of departmental LAN's than bridges. Routers are more complex devices than bridges, and more easily integrated into an automated network management system -- and can be configured to yield effective security partitions in a network (referred to as "firewalls" in this mode of use).
Routers perform a more demanding role than bridges, and the resulting processing burden makes routing a more computationally intensive task than bridging (accordingly, routers are typically custom, multiprocessor hardware solutions whereas bridge functions can be supported by a conventional PC).
D-135
(50813)
173
LANINTERCONNECTION ALTERNATIVES,cont.
ROUTERS,cont.
Gatewaysareusedtointerconnectdissimilar (andotherwiseincompatible)networks,providing:
Gatewaysareoftenapplication-specific(suchase-mailgatewaysthatmanagee-mailservices,or e-commercegatewaysthatarbitratee-commercetrafficforaserverfarm).
D-384
(21106)
174
WAJAX Forks Pty. 6784 Alberhill Boulevard Tiddlybank, New South Wales Australia
The first-step postal sorting process has to be able to identify which of a possible 200 (approximately) destination countries has been specified.
Susan Henderson
Flat addresses are superficially simpler -- but they require unreasonable detailed knowledge of all possible destinations when there is a large population of users.
With a flat address system, every sorting clerk would have to know the location of every possible recipient -- some 6,000,000,000 possibilities!
Bridges work with this address model, requiring each bridge to maintain exhaustive address tables containing location information for all possible destinations.
This works well for small networks, but is not a reasonable approach for complex configurations.
D-442
(60607)
175
When an organization initially deploys LAN services, the obvious first configuration is a free-standing (non-connected) system.
As traffic volumes and/or the physical dimensions of the service area increase, a strategy for defining a connectivity-based infrastructure becomes necessary. Directly Connected LAN's Where a small number of LAN's (probably departmental in scale at this point) are in use, direct connectivity is a reasonable strategy -typically using bridges as the preferred entry-level connectivity devices. Once more than two DLANs (Departmental LANs) are involved, it becomes inevitable that there will be through-traffic adding potential congestion problems to intermediate networks -- much like vehicle traffic adds congestion to an urban district where there is no through highway available. The maximum number of LANs suitable to this configuration is small (perhaps 4 or 5) and will depend on traffic, types of applications and physical dimensions of the site.
DLAN #1
DLAN #2
DLAN #3
DLAN #4
D-582
(61207)
176
With a backbone architecture, a purpose built network segment is added, with the dedicated mission of providing connectivity services to the other networks. Backbone capacity typically exceeds that of the individual networks it serves (perhaps the most common configuration is a 100 Mbps FDDI backbone system serving a cluster of 10 Mbps ethernet departmental LANs). The physical dimensions of the backbone system are highly variable, ranging from a few feet (collapsed backbone limited to an equipment rack) to tens of miles/kilometers (MAN-structured backbone). Backbone networks should have no directly connected workstations or servers (all such application-oriented equipment should be on one of the DLAN segments). Possible backbone-connected resources would include routers and/or firewalls to external networks and network management automation resources.
DLAN #3
DLAN #1
DLAN #2
BACKBONE LAN REACHES TO ALL DLAN NETWORKS (Connections are by local bridges/routers)
DLAN #4
D-583
(61207)
177
The Fiber Distributed Data Interface (FDDI) was the first recognized high speed LAN (HSLAN) and metropolitan area network (MAN) technology -- developed in 1990 and accepted by ANSI as a standard. Key features of FDDI are:
100 Mbps dual ring, token passing fiber optic technology Fault-tolerant, capable of sustaining full operational throughput through any arbitrary single-point failure (device failure of cable break). Compatible with IEEE 802-Series LAN standards at 802.2 LLC sub-layer Operates on multi-mode and single-mode fiber -- or copper twisted pair (UTP) cable. Maximum ring circumference allowed = 200 km. Maximum number of attachments = 1000 FDDI LOWER LAYER STRUCTURE Supports two classes of stations: dual-attachment and single-attachment (Dual-attachment configurations have full fault-tolerant features). Applications: campus backbones, data center bus systems, metropolitan area networks and high-speed desktop applications. FFOL (FDDI Follow On LAN) is a planned upward compatible extension projected to operate at 2.4 Gbps.
D-213
(61024)
178
FRAME CHECK
END DELIMITER
FRAME STATUS
FRAME FEATURES:
Preamble: Consists of alternating 1-0 bits for clock synchronization characters -a maximum of 16 octets Marks the start of the frame, much like the flag field with HDLC Designates the frame type being carried (ULP identification) and designates token/message status Either a 48 bit Universally Administered address field, or a 16 bit Locally Administered address (not typically used) Like Destination Address, either a Universally or Locally Administered address Information Payload
IEEE/ISO compatible LLC frames SNAP-encoded general ULP data (such as IP datagram) IEEE/ISO MAC layer frame encapsulated Frame Check: End Delimiter: Frame Status:
Destination Address:
Source Address:
D-667
(11206)
179
DAS
DAS
Devices which populate the FDDI Primary Ring will always be Dual Attachment Stations (or DAS) -supporting full access to both fiber rings. As will be seen later, a non-redundant class of termination known as a Single Attachment Station (SAS) has a useful role in making non-critical connections to the core ring shown here.
The PRIMARY RING operates at 100 Mbps, and always carries active message traffic
The SECONDARY RING operates at 100 Mbps also, and functions in one of two modes:
it may be designated as a hot-standby back-up to the primary ring, idling while the system operates normally -- probably for about 99.999 percent of the time (typical) it may carry preemptable message traffic, which gets dumped when called upon to back-up the primary ring (increasingly widely used)
D-374
(11206)
180
DAS
DAS
DAS
DAS
DAS
DAS
The FDDI dual ring system wraps primary ring traffic onto the secondary ring -- at the nearest surviving dual attachment station (DAS) connection point, thus maintaining operation through any single failure event.
D-375
(11206)
181
VIRTUALLOCALAREANETWORKSCONCEPTILLUSTRATION
IllustratedbelowaretwovirtualLANsystemsconfiguredonathree-hubphysicalLAN.
Thekeyconceptisthattherenolongerneedstobeanassociationbetweenthephysicalnetwork configurationandthefunctionalconfiguration.
ROUTER TOPLANT
K-094
(41103)
182
The first temptation is to suspect the physical cable system is out of capacity -- which is typically not the problem. However, a related problem may be a factor:
Violation of cable length and repeater placement rules (with either Ethernet or Token Ring) can result in excessive numbers of failed transmission attempts, with too much time unproductively consumed with retransmissions.
Server performance is the most common system bottleneck: hard disk access time, RAM capacity and configuration, NIC throughput limitations, server cpu type and clock speed are all performance-related features.
To a lesser extent, workstation performance will be a factor, with the two most significant being workstation cpu type, RAM capacity and clock speed and NIC characteristics.
System diagnostic software and/or test equipment can be used to identify these choke points -- or bottleneck conditions.
Conversely -- people responsible for making LAN systems upgrade decisions should be cautious about spending sums of money to upgrade components of the system that may not be the performance limiting assets.
D-232
(10420)
183
Day Three
184
PROTOTYPESCADASYSTEM
HMI
MTU COMPUTER SYSTEM
COMMUNICATIONS EQUIPMENT INTERFACES
TELEPHONE MODEM
WIRELESS LINK
LINE DRIVER LINE DRIVER
WIRED LINKS
REMOTE TERMINAL UNIT
TELEPHONE MODEM
K-047
(30818)
185
PROTOTYPESCADASYSTEM
HMI
MTU COMPUTER SYSTEM
COMMUNICATIONS EQUIPMENT INTERFACES
TELEPHONE MODEM
WIRELESS LINK
LINE DRIVER LINE DRIVER
WIRED LINKS
REMOTE TERMINAL UNIT
TELEPHONE MODEM
K-049
(30818)
186
ELEMENTSOFBASICSCADASYSTEMS
ThethreebasicelementsofSCADAtechnologyare:
MasterTerminalUnit(MTU)
TheMTUhastwokeyfunctions:
Itcontrolsallremotedatacollectionfunctionsanddataflowsthroughthesystem.
continued ....
Data Communications & Fieldbus Systems
K-050
(40308)
187
ELEMENTSOFBASICSCADASYSTEMS,cont.
RemoteTerminalUnits(RTUs)
RTUsactasagentsfortheMTUinfieldlocationswheretheSCADAsystemhasaneedto collectdataand/ordistributecontrolsignals.
TherearetypicallymanyRTUsperSCADAsystem--sometimeshundreds. ThetwokeyfunctionsoftheRTUare:
Tointerfacewiththenetworkandsupportthenecessarycommunicationprotocolsand functions--enablingthetransferofdataandcommandstoandfromtheMTU
continued ....
Data Communications & Fieldbus Systems
K-051
(40308)
188
ELEMENTSOFBASICSCADASYSTEMS,cont.
AConnectingNetworkInfrastructure
ThenetworkinfrastructurethatconnectstheMTUwiththeassociatedgroupofRTUsisthe mostvariableanddifficult-to-summarizepartofSCADAsystems:
Thescaleofdistanceinvolvedcanvaryfromafewfeet/meterstohalfwayaroundthe world.
K-052
(30818)
189
SELECTBEFOREOPERATEvs. OPERATEWITHACKNOWLEDGE
TheoperationofSCADAcontrolproceduresmaybesetupinaboutthreedifferentways:
SelectBeforeOperate(Sometimescalled"ConfirmBeforeOperate")
Highvalue,criticaloperations(typicallythosewhoseincorrectoperationcouldresultinfacilities damageorfinancialloss)aretypicallyhandledwiththistwo-stepprocedure.
TheSCADAoperatorsetsupacontrolcommand,andessentiallyarmstheRTU,readytoactivate thedesignatedcommand.
TheRTUthensendsaconfirmingmessagebacktotheoperator,effectivelyaskingforconfirmation thatthisactioniswhatisreallyintended.
DirectOperateWithAcknowledgement
Thisprocedureprovidesasimplerframework,withadesignatedcontrolprocedurebeing implementedattheRTU,noquestionsasked.Anacknowledgmentoftheactionfollows.
DirectOperateWithoutAcknowledgement
Veryuncommon--thisproceduresimplyperformsthedesignatedactionwithnoconfirmation.
K-087
(41031)
190
OBJECT-ORIENTEDPROGRAMMINGCONCEPTS
Thisinheritancemodelisopen-ended;AinheritsfromB,BinheritsfromC,CinheritsfromD...
Object-orientedprogramsaresignificantlylessproceduralthanconventionalprogramming structures,althoughultimatelythereisasequentialorproceduralcomponenttotheirexecution.
K-064
(40306)
191
ANOBJECT-ORIENTEDVIEWOF INDUSTRIALDATASOURCESANDPROCESSES
Traditionally,representingthedifferentelementsofdatathatareaccessiblefrom,oraredelivered toremoteequipmentismassive,tediousandfraughtwithhumanerror.
Accommodationofnewrequirementsthatemergeafterthesystemsarecommissioned,suchas newHMIfeaturesanddisplayscreensornewprocesscontrolmodels.
K-065
(40307)
192
MTUPLATFORMREDUNDANCY
Thiswilltypicallyinvolvetransferringpowerand/orcommunicationscircuitsandinitializingsystem software(whichmeansdataassociatedwithworkinprogressislost).
HotStandby
Thisisaclassoffault-tolerancethatprovidesapowered,operationalreplacementsystemready andwaitingforservice.
continued ....
Data Communications & Fieldbus Systems
K-066
(40307)
193
MTUPLATFORMREDUNDANCY,cont.
FullDynamicRedundancy
Thetechnologysupportsstandard"middleware"developmenttechnologieslikeTuxedoandCorba --andisstronglyorientedtowardwebserviceapplications.
Inmostcases,itistoocostlyasasolutionforSCADAapplications,butitwoulddeliver extraordinaryavailabilityifusedasanMTUhostmachine.
Thereareanumberoflessambitiousfault-tolerantequipmentconfigurationsbasedonUnixand MicrosoftWindows2000technologies--butnonewitha"nosinglepointoffailurearchitecture".
K-067
(40308)
194
SERVERDISKARRAYSYSTEMS
Fault-tolerantdiskarraysenableon-linereplacementoffaileddiskhardware,andseamless reconstructionofdata--transparenttotheuser
Dataisbrokenintopiecesandwritteninduplicatetoatleasttwoofseveraldisks,usinga processcalled'striping'
Themostsuccessfuloftheseschemesisthefive-levelRedundantArraysofInexpensiveDisk (RAID)techmnology,whichhasgainedacceptancethroughMIcrosoft'ssupport:
RAID-1:Usesmirroreddiskstoprovideone-on-onedataredundancy;recoveryrequires disruptingservice.
RAID-2:Usesamoreefficientdatarepresentation,requiringlessthan100%duplicationof datavolume;notcommonlyused.
RAID-5:Usesdistributeddataguarding,eliminatingthededicatedguarddrive--resultinginall driveshavingequalstatus,andsupportingmultiplesimultaneousreadsandwrites.
D-333
(99999)
195
RTUBLOCKDIAGRAM
TWO-WAYEXTERNAL COMMUNICATION TERMINALSTRIPFOR CONNECTINGINPUTS
CPU
OUTUTSIGNAL INTERFACE
MEMORY
TERMINALSTRIPFOR CONNECTINGOUTPUTS
K-060
(31011)
196
CONTROLSAREIMPLEMENTEDASEITHER CLOSEDLOOPOROPENLOOPSYSTEMS
OpenLoopSystems
Anopenloopprocessisaone-wayprocess--inputcontrolsignalspassthroughalogical sequence,orprocessandresultinmodifyingthebehavioroftheplantorprocess.
continued ....
OPEN-LOOPCONTROL
INITIAL INPUT
+ -
ACTUATING SIGNAL
PLANTOR PROCESS
OPERATOR CONTROLS
HUMAN OPERATOR
SENSOR INPUTS
K-041
197
CONTROLSAREIMPLEMENTEDASEITHER CLOSEDLOOPOROPENLOOPSYSTEMS,cont.
OpenLoopSystems,cont.
Agoodexampleofanopenloopcontrolsystemisthefuelcontrolmechanismusedinan automobile.
Thedesiredinputisdefinedbythepositionoftheacceleratorpedal;theactuatoristhethrottle valveandtheprocessoutputisamodifiedengineoperatingspeed.
Humanobservationswilltypicallyresultinanassessmentofthesuitabilityofagivenengine speedandtheoperatorwillchangetheacceleratorpedalpositionaccordingly.
SCADAsystemsalmostalwaysoperateasopenloopcontrolprocesses--withtheprocess controloperatorperformingafunctionanalogoustotheautomobiledriverabove.
Thelackofanyfeedbackmechanismforallowingoutputconditionstoautomaticallyinfluence inputsignalsputssomeconstraintsontheprocess:
Thesystemresponse/operatingdynamicsmustbecompatiblewithhumanreactiontimes. Humanvigilanceisrequiredtoensuresafe,continuedoperation.
continued ....
K-042
198
CONTROLSAREIMPLEMENTEDASEITHER CLOSEDLOOPOROPENLOOPSYSTEMS,cont.
ClosedLoopSystems
Closedloopcontrolsystemshavethepotentialforbeingunstableinsomeorall-operating conditions--andmustbecarefullydesignedtooperatereliablyunderallconditions.
continued ....
CLOSED-LOOPCONTROL
DESIRED INPUT
+
ACTUATING SIGNAL
PLANTOR PROCESS
-
SENSOR
K-043
(30615)
199
CONTROLSAREIMPLEMENTEDASEITHER CLOSEDLOOPOROPENLOOPSYSTEMS,cont.
ClosedLoopSystems,cont.
Examplesofclosedloopcontrolsystemsinclude: Anautomobile"cruise-control"systemfor regulatingthrottlepositionbasedonthespeedof thevehicle. Ahouseholdfurnaceorairconditionerthatuses asimpleon-offswitchingthermostattoactivate therequiredtemperaturecontrolsystem. JamesWatt'sfly-ballgovernor,(see accompanyingillustration)whichwasoneofthe firstclosedloopcontrolsystems. Ineachoftheseexamples,adesiredinputvalue (setpoint)ismodifiedbyinformationreturnedviaa feedbacklink,enablingtheinputinstructionstothe controlprocesstobemodifiedoradjustedsoasto moreexactlyconformtothedesiredinputvalue,or setpoint.
JAMESWATT'S FLY-BALLGOVERNOR
K-044
200
REAL-TIME OPERATIONS
Industrial computer systems are typically required to service their "clients" -- robots, instruments, device actuators, controllers, etc. -- in real-time, meaning that results/outputs must be assured of being available while the input conditions are still valid.
There are numerous alternative definitions of real-time computing, but all interpretations of the concept emphasize a guarantee of timeliness of results, where the meaning of "timeliness" depends on application-specific details. Examples of real-time computing applications include:
a servo control sub-system in an aircraft fly-by-wire system that must respond quickly enough to modify flight attitude to assure timely reaction to changing flight circumstances
a factory production system that involves a robot arm picking something off a conveyor belt -in which delayed response from the arm would mean the object had moved beyond recovery reach
In both cases, the computational result must be complete in a time-frame that will assure timely accomplishment of the overall system's operating objectives.
The achievement of effective real-time computing is more than a computer hardware/software problem, since it includes data recovery delays, network transit delays, output delays and actuator operation delays.
K-028
201
Real-Time Operating Systems (RTOS) have distinctly different features and requirements, compared with typical business applications.
Summary features that are necessary to support time-critical, event-driven industrial computing applications include:
multi-thread operation - alternatively known as reentrancy, is the ability to support multiple concurrent processes with a single memory-resident version of key software modules.
preemptible tasking - capability of designating time-critical processes as top priority, and therefore eligible to be scheduled for immediate execution.
interrupt driven (asynchronous scheduling) - capable of responding to external events that trigger hardware-level interrupts (to normal, otherwise sequential processing)
controlled, short intervals with interrupts disabled - processes which normally operate with interrupts enabled must disable interrupts while context switching activities are under way (and while certain discrete processes operate) -- but must assure a rigorous bound on the maximum time interval that will be encountered with interrupts disabled.
The unique data handling processes associated with real time applications are frequently well served by a two-stage architecture, where the front-end process is supported by a RTOS (and designed to be lean and simple), feeding a conventional scheduled environment operated on a backend process.
K-029
202
RELAYBASICS&CONCEPTS
Therewasatimewhenindustrialcontrolsystemswereentirelybasedontheuseof electromagneticrelaystosenseelectricaloperatingconditionsandactivatecontrolfunctions.
Theessentialfeatureofelectromagneticrelaysisthewayinwhichthemagneticfieldthatresults fromacontrolledcurrentflowactuatingaswitchedelectricalcontactforadifferentcircuit. Inthediagramhere,withnocurrentflowinginthe controlcircuit,therestingpositionoftherelay contactsisdown,makingcontactbetweenthe bottompairofcontactpoints. Therestingpositionofarelaywithnocurrent flowinginitsmagneticcoilisthe"Normal"position, socontactsthatareclosedinthisconditionare "NormallyClosed"orN/C--andcontactsthatare openinthisconditionare"NormallyOpen",orN/O. Whentheswitchisclosed,andcurrentflowsinthe relaycoil,thesecontactconditionsreverse--but the"Normal"definitionremainsasbefore.
BASICRELAYFEATURES
SWITCH
CONTROL CIRCUIT
BATTERY
N/O N/C
K-039
(30616)
203
INTHEBEGINNINGPROCESSESWERE CONTROLLEDWITHRELAYS
Thefirstautomationsolutionsforprocesscontrolandrelatedapplicationswerebasedon electro-magneticrelays--many,manyofthem.
Forthisreason,PLCsinheritedthesamelogicalsystemforrepresentingtheiroperationasthe technologytheyreplaced.
Thus,the"ladderdiagram"livesonasadesignconceptthatisimbeddedintheprogramming toolsthatareusedtodevelopmodernPLCinstructions,orprograms.
K-040
204
LADDERDIAGRAMS
Thefirstprocessandindustrialcontrolautomationsolutionswerebasedon electro-magneticrelays--typicallymanyofthem.
Forthisreason,PLCsinheritedthesamelogicalsystemforrepresentingtheiroperationas therelay-basedtechnologytheyreplaced.
Thus,the"ladderdiagram"livesonasadesignconceptthatisimbeddedinthe programmingtoolsthatareusedtodevelopmodernPLCinstructions,orprograms.
K-053
(30906)
205
LADDERDIAGRAMCONCEPTS
ElementsofasimpleLadderDiagramareillustrated intheboxontheright. SW1andSW2areswitches;SW1isnormallyopen (NO)andSW2isnormallyclosed(NC). Thebatterypositiverailistheleft-sideverticalline; thebatterynegativerailistheright-sideverticalline. Sequentiallogic (whichisremarkablysimplehere) flowstop-to-bottom--eitherlooping,orterminating onthe"END"command. Eachlogicalinputandoutputisassignedamemory addressvaluesothatoperationscanreferencethe TRUEorFALSEconditionofeachelement. ElementsareconsideredtobeFALSEwhenintheir normal,orrestingstate,andTRUEwhenactivated. Thus,NOinputsareFALSEwhenopen;NCinputs areTRUEwhenopen. Outputsareenergizedonlywhenallinputconditions areTRUE.
VERYSIMPLERELAYCIRCUITAND CORRESPONDINGLADDERDIAGRAM
SW1
SW2
BATTERY
COIL
INPUTS SW1
SW2
OUTPUT COIL1
END
K-059
(31011)
206
FUNCTIONBLOCKINSTRUCTIONS
Tasksofthissortcouldbeprogrammedwithcombinationsofstandard,simpleinstructions -butthefunctionblockwasintroducedinordertomakethesemorecomplextaskseasier.
Justaswithanarithmeticrunginaladderdiagram,afunctionblockrunghasthreeparts:
Inputconditionsconsistingofcombinationsofexamineinstructions.Therewilltypically beseveralinputconditionsthatare"examined",orusedtodrivethefunctionblock.
Theactualfunctionblock,whichwilloftendependonanumberoflocations-or variables--thatwillaffectitsaction.
Outputs,whichbehavelikerelaycontacts,arethemeansbywhichthefunctionblock interactswithotherladderrungsinthesystem.
continued ....
Data Communications & Fieldbus Systems
K-057
(31011)
207
FUNCTIONBLOCKINSTRUCTIONS,cont.
Likeinputs,therearecommonly severaloutputs: Numericaloutputsarecommonlyused byotherrungs(functionblocks)as inputstotheirarithmeticinstructions. Relaycoil(bit-level,orbinary)outputs canbe"examined",orreadbyexamine instructionsinotherfunctions. Therehasbeenacleartrendtoward PLCmanufacturersconvertingtiming, counting,andarithmeticinstructionsto themoreuser-friendlyfunctionblock format. Programmingdocumentationavailable foranyparticularPLCwillexplainthe programmingofthevariousavailable functionblockinstructionsindetail.
FUNCTIONBLOCKUSEDINA SIMPLELADDERDIAGRAM
INPUTS SW1
SW2
OUTPUT COIL1
SW3
OUTPUT COIL2
END
K-058
(31011)
208
The Programmable Logic Controller originated in the late 1960's as a discrete electronic logic system intended to replace hardwired relay control systems.
Relay systems have been designed using a so-called ladder diagram, or ladder logic -- effectively a Boolean logic scheme where relay features (such as make-before-break and latching operation) are central to the logical operation of the system.
With PLCs, these relay features are modeled and embedded in the programmable characteristics of the controller.
Although the programming interface to PLCs typically focuses on the logical relay-based equivalent, the PLC is really a computer with computer-like components, including: Central Processing Unit (CPU) Memory Unit Input Interfaces Output Interfaces
The PLC has been adapted to countless applications, many well beyond the original scope intended by the original concept -- but the core application remains local control of complex machines based on Boolean logic conditions (such as tank levels, motor or pump status, pressure conditions, position of movable parts, etc.)
K-024
(11206)
209
PLCBLOCKDIAGRAM
TERMINALSTRIPFOR CONNECTINGINPUTS
CPU
OUTUTSIGNAL INTERFACE
MEMORY
TERMINALSTRIPFOR CONNECTINGOUTPUTS
K-054
210
The PLC has proven to be a remarkable efficiency tool to industry, offering the following key advantages over precursor electro-mechanical hardware solutions:
Superior flexibility to accommodate various applications Superior reliability (maintenance is radically simplified) Superior economy (cost to install and operate/maintain) Investment is recyclable (a software change changes the entire system) Development effort is portable (one software development effort can serve many installations)
Other features that have no direct comparison to hardware relay logic solutions includes:
PLCs can be applied to applications where critical response times must be guaranteed PLCs can support networked communication with peers and/or master control systems PLCs can support graphical, user-friendly human interface sub-systems PLCs can support a wide range of sophisticated instruments, creating a systems control capability that is incomparable to precursor technology.
The range of PLC product size, sophistication, capacity and cost is remarkable; PLCs can be cost-effectively used in applications where only a handful of relays might otherwise be used, all the way to big, complex, networked multi-million dollar process control applications.
With this range of applications and scale of capacity available, the understanding the technology and capabilities of PLC systems is a technical specialty in its own right.
K-026
(11206)
211
The Programmable Logic Controller has been adapted to many industrial automation applications -- but most are applied to factory automation (FA) systems, and most are stand-alone configurations operating independently of other control systems. A sampler of PLC applications includes: Injection molding Warehouse automation Pump (water, chemicals, food products) control Rolling mills Industrial mixing and blending applications Production line testing Bottling and filling applications Motor controls Paper machine controllers Speed control (such as elevators, escalators and conveyor systems) Cutting, drilling, welding and stamping applications Remote Terminal Unit (RTU) interfaces to SCADA systems Petrochemical distillation/cracking/blending/mixing applications Flow control systems (aggregates, liquids, discrete items)
K-025
212
PROGRAMMABLELOGICCONTROLLERS --PARTIALVENDORLIST
AsamplerofthebetterknownPLCvendorsincludes:
ABB AllenBradley(RockwellAutomation) Aromat ControlMicrosystems Eaton/Cutler-Hammer GEFanuc(GeneralElectric) Hitachi IDEC Keyence Mitsubishi Omron SchneiderElectric(Modicon) Seprol Siemens Toshiba
(www.abb.com) (www.rockwell.com) (www.aromat.com/pgh.htm) (www.controlmicrosystems.com) (www.cutler-hammer.eaton.com) (www.gefanuceur.com) (www.hitachi-ds.com) (www.idec.com) (www.keyence.com) (www.global.mitsubishielectric.com) (www.omron.com) (www.schneider-electric.com) (www.seprol.com) (www.sea.siemens.com/controls/) (www.tic.toshiba.com)
K-027
(30615)
213
The technical feasibility of using microprocessor-based intelligence as an integral part of field instruments opens many doors of opportunity for enhancing the operational effectiveness of industrial networked control systems.
With this opportunity comes the inevitable requirement that a different means of communicating with this class of instrumentation is necessary -- more consistent with the functions of business local area networks. From this need and opportunity the fieldbus concept was born.
Unfortunately, like many good ideas, the fieldbus concept has not lived up to industry expectations because of a lack of standards and industry consensus about the myriad of technical details that must be coordinated in actually making it all happen. This problem is especially acute in the industrial context because:
The many vendors involved have a vested interest in promoting their proprietary solutions, which lock customers into their complete systems solution.
The industry does not have a history of creating systems from plug-and-play modular components -- systems integration is a complex, specialists domain.
Industrial practice and preferred vendors vary from country-to-country, creating regional centers of influence.
K-017
214
There are at least 10 alternative fieldbus technologies making claim to a "standard" industrial network solution. Of these, the following are most credible:
Profibus DP/PA
Network technology similar to IEEE 802.4 Token Bus Standardized as a German National Standard (DIN 19245, Part 3/4), and is a European "interim fieldbus" standard CENELEC 50170. Siemens and other German industry suppliers have been principal advocates. Primary/initial application is networking PLC equipment Profibus DP is a 12 Mbps technology, Profibus PA operates at 31.35 kbps.
LONworks
Features a non-deterministic network technology (uses CSMA/CD access technique like Ethernet) LONworks is not a formal standard technology, although it is an openly published technical specification and coordinated by ASHRAE of BACnet. A major application area is with control applications in intelligent buildings; (HVAC, fire-safety, security, elevator controls, etc.), although many applications in general fieldbus use (petrochemicals, water control, food processing and energy management) have been successfully developed. LONworks networks operate at 1.25 Mbps full duplex over various media.
K-018
(11206)
215
Works with a twisted pair or fiber optic bus architecture network WorldFIP is a European standard technology (originally developed and sponsored in France) WorldFIP technology has been very influential in the creation of Foundation Fieldbus standards -- being the source technology most closely adopted. FIP is a French National Standard, and is one of four European CENELEC 50170 "interim fieldbus" standards. FIP is based on a 31.25 kbps, 1 Mbps, 2.5 Mbps or 6 Mbps peer-to-peer network infrastructure.
The IEC/ISA SP50 Fieldbus standardization initiative is an attempt to pull together much of the divergent architecture of the various "fieldbusses". Part of what makes this initiative challenging is the vested interest groups; several national interest groups (notably Profibus and WorldFIP) and groups of equipment manufacturers, none of which have much motivation to simplify the user's situation. The design features of the SP50 initiative are intended to be flexible, with a view to accommodating as many of the other technology features as possible. The SP50 Fieldbus supports a 31.25 kbps service, as well as 1, 2.6 and 5 Mbps mode. of operation.
K-019
216
Foundation Fieldbus is the result of an industry-sponsored initiative to unify the diverse segments of the fieldbus industry. The Foundation Fieldbus technology development and standards are coordinated with the IEC/ISA SP50 initiative, although there is no commitment to synchronizing the two standards tracks. There are two categories of communication services defined (as with several of the other fieldbusses): The H1 Instrument Bus - oriented toward servicing field instruments operates at 31.25 kbps. 1.9 km maximum cable length The H2 Systems Network Bus - oriented toward supporting networked connectivity between PLCs, DCS systems, human interface subsystems, etc. operates at 1.0 or 2.5 Mbps. 750 m maximum cable length
Intrinsically safe operation is supported in both bus configurations. Both configurations operate over (nominally shielded ) twisted pair cable media. Both configurations support bus-powered instruments, although the bus powered feature limits the number of devices that can be connected per fieldbus segment (and restricts the H2 service to 1.0 Mbps). The standard is much more than a network technology; it includes a so-called User Layer set of features that facilitate coordination of measurement units and point descriptions (based on a procedure and data representation method referred to as function blocks).
K-020
(11206)
217
CONTROLLER
K-021
218
COMPARISONANDUSAGEOFCOMMON INDUSTRIALNETWORKS
NETWORK ARCNet ASI CANOpen UTP Coax,Fiber UTP UTP,Fiber UTP UTP,Fiber Device Device Control/Device Control/Device UTP,Fiber, Powerline UTP UTP,Fiber ControlNet DeviceNet Ethernet Foundation Fieldbus Control Interbus LonWorks Modbus ProfibusDP/PA Enterprise Device Control Device Device FlatorUTP Enterprise UTP,Coax, Fiber 10Mbps 167Kbps 1Mbps 5Mbps 500Kbps 10Gbps 100Mbps 500Kbps 1.25Mbps 38.4Kbps 12Mbps LEVEL MEDIA MAXRATE MAXDIST. 10Km 100m 1Km 30Km 500m 2Km 1.9Km 400m 2Km 500m 24Km MAXNODES 255 31 127 99 64 1024 240 256 32000 250 127
USAUSAGE
INT'LUSAGE
9%
12%
2%
9%
3%
12%
15%
14%
22%
17%
73%
50%
3%
7%
7%
7%
4%
6%
29%
22%
8%
26%
K-045
(30616)
219
Modbus is a nominally proprietary communications architecture/protocol initially developed as part of the Modicon brand of PLCs and attached equipment.
Modbus is a widely used industrial communications environment that specifies both: an equipment configuration (either a bus or star topology) a communications protocol ( a master-slave polling protocol)
The master-slave architecture of Modbus supports only one master device per bus segment, consistent with a single (non-redundant) PLC interacting with a number of slave devices.
Applications which require interaction with higher-level systems (such as Human Machine Interfaces [HMIs] or possibly SCADA systems) are often configured redundantly -- requiring duplicate Modbus networks to support redundant operations.
A newer version of Modbus (Modbus Plus) is a more sophisticated architecture/protocol, but is not nearly as widely accepted as an industry standard.
Modbus Plus is a token-passing, distributed control network system -- not remarkably different from the common token passing LAN systems used in business communications applications.
A key feature of this approach is it offers peer-to-peer capability, in distinct contrast to the master/slave communication in basic Modbus.
K-031
220
The two common Modbus network configurations are bus and star topology -- and several variations within each. Typical implementations are suggested in the accompanying sketches.
MASTER DEVICE MODEM CONFIGURATION SHOWN SUGGESTS TELEPHONE CONNECTIVITY. 4-WAY 4-WIRE BRIDGES WOULD SUPPORT A LOCAL HARD-WIRED CONFIGURATION MODEM SLAVE DEVICE
BUS TOPOLOGY
MODEM
SLAVE DEVICE
STAR TOPOLOGY
MASTER DEVICE DIGITAL BRIDGE IS AN ACTIVE ELECTRONIC RS-232 (typically) BROADCAST OR REPEATER UNIT DIGITAL BRIDGE
SLAVE DEVICE
SLAVE DEVICE
SLAVE DEVICE
K-032
(11206)
221
The Modbus protocol is a master-slave polling regime, where simple polling instructions are issued to addressed destination devices, and a corresponding response is received.
Typical Modbus master station polling message format is illustrated in the accompanying box.
The key to flexibility with the Modbus protocol is the approximately 21 standard function codes (plus 200-odd reserved/special use codes).
The polling message example suggested in the accompanying box consists of five fields:
Slave Address (05) -- allows up to 247 slave addresses per bus segment. Function Code (03) -designates the functional purpose of this message (see code summary next page) Read at Start Register (00F1) designates the register number at which data recovery is to begin
0002
E3
K-033
222
Number of registers to Read (0002) -- specifies the number of data register contents to be recovered in this transaction
Error Check Code (E3) -- data integrity calculation used to validate received data in preceding message. A sampler of Modbus function codes is given in the accompanying box.
Data is always sent as an asynchronous character, and will be encoded either as: ASCII representations of HEX characters, or Binary codes (8-bit bytes, or octets)
POLL PROGRAM COMPLETE FETCH EVENT COUNTER POLL PROGRAM COMPLETE FORCE MULTIPLE COILS PRESET MULTIPLE REGISTERS REPORT SLAVE ID RESET COMMUNICATIONS LINK READ GENERAL REFERENCE WRITE GENERAL REFERENCE
K-034
(11206)
223
ETHERNET/IP
Ethernet/IPisanemergingstandardspecificationforopenarchitectureindustrial networkingthatisconfusinginitsfirstencounter:
The"IP"partofthespecificationdoesnotdenoteInternetProtocolasonemight anticipate,butratherIndustrialProtocol--Thus,itisEthernetIndustrialProtocol.
Havingsaidthat,Ethernet/IPisbasedonaTCPoverInternetProtocolframework.
CIPprovidesasetofreal-timeI/O-andpeer-to-peermessagingservicesthatsupportsthe MTUinitscommunicationwithboththeRTUandwiththeHMI.
K-061
(40308)
224
CIPLAYEREDARCHITECTUREVIEW
VARIOUSDEVICEPROFILES(SENSORS,CONTROLS,etc.) CIPAPPLICATIONLAYER&APPLICATIONOBJECTLIBRARY
CIPDATAMANAGEMENTSERVICES(ExplicitMessages&I/OMessages) CIPMESSAGEROUTING&CONNECTIONMANAGEMENT ENCAPSULATION CONTROLNET TRANSPORT DEVICENET TRANSPORT CAN CSMA/NBA CTDMA PHYSICAL PHYSICAL TCP INTERNETPROTOCOL UDP
IEEE802.3ETHERNET-LLC&MACLAYERS
IEEE802.3ETHERNET-PHYSICALLAYER
CONTROLANDINFORMATION PROTOCOL
Data Communications & Fieldbus Systems
K-062
(40306)
225
DEVICENET'SCSMA/NBAPROTOCOL
DeviceNetisanindustry-standard,openspecificationdevice-levelnetworkthathasbeendesigned withahighpriorityonreal-timecontrolapplications.
Akeyfeatureofthisclassofapplicationistheneedforahighlypredictable,ordeterministicnetwork deliverysubsystem.
ThisdifferencemakestheCSMA/NBAaccessprotocolmuchmoredeterministicandthussuitable totime-criticalapplications.
K-063
(40306)
226
MODBUS PROTOCOL
Defines a message structure that controllers will recognize and use Independent to the type of networks (RS-232, RS-422/485, Ethernet, etc) Describes the process a controller uses to : o request access to another device, o respond to requests from the other devices, o detects and reports errors. Master-slave arrangement Establishes a common format for the layout and contents of message fields
227
MODBUS TRANSACTION
Data exchange between devices is called transaction Only the master can initiate transactions (query) The slave respond by supplying the requested data to the master, or by taking the action requested in the query. The master can address individual slaves, or can initiate a broadcast message to all slaves. Maximum 246 slaves can be addressed (1-247) Slaves return a message (called a response) to queries that are addressed to them individually. Responses are not returned to broadcast queries from the master.
228
MESSAGE FORMAT
START o starting delimiter of a message ADDRESS o address of destination devices o address 0 is used for broadcast FUNCTION code of masters command function
DATA o additional information which the slave must use to take the action defined by the function code LRC/CRC CHECK message frame check
END
229
FUNCTION CODES
Each code represents the masters command to the slaves Categories : o Data query (coil/register) o Status/register change command o Programming command o Diagnostics o Device information query Some function codes are not supported by all devices
230
ERROR CHECKING
Error checking method : o Parity check (even/odd) for each character (optional) o Frame check for entire message (CRC/LRC) Both the character check and message frame check (CRC/LRC) are generated in the master device and applied to the message contents before transmission The slave device checks each character and the entire message frame during receipt and generates CRC/LRC If the LRC/CRC generated by master device is not the same as LRC/CRC generated by slave device, the message contains some errors
231
232
MODBUSPLUS
Local area network system designed for industrial control applications. Each network supports up to 32 addressable node devices (64 if repeater is used) 1 Mbps data transfer rate Token passing derivative that uses the MODBUS messaging structure Provides host level peer-to-peer communication for the network devices. Applications include transferring of process control and supervisory messages.
233
Modbus messages sent on Modbus Plus networks are embedded into the Logical Link Control (LLC) level frame.
Modbus message fields consist of 8bit bytes, similar to those used with RTU framing.
The Slave Address field is converted to a Modbus Plus routing path by the sending device.
The CRC field is not sent in the Modbus message, because it would be redundant to the CRC check performed at the Highlevel Data Link Control (HDLC) level.
The application software (e.g., MSTR blocks in controllers, or Modcom III in hosts) handles the framing of the message into a network packet.
234
235
MODBUSTCP
A variant of the MODBUS family of simple, vendor-neutral communication protocols intended for supervision and control of automation equipment.
Covers the use of MODBUS messaging in an Intranet or Internet environment using the TCP/IP protocols.
Use mostly for Ethernet attachment of PLCs, I/O modules, and gateways to other simple field buses or I/O networks.
236
MODBUSTCP FRAME
The connection-oriented TCP protocol is used rather than the datagram-oriented UDP to keep control of an individual transaction
The MODBUS checksum is not used as the Ethernet TCP/IP link layer checksum mechanisms are relied upon to guarantee data integrity
237
A local area network designed to support remote programming for factory-floor applications. o Connects a maximum of 64 devices per link (although 15 or fewer nodes are recommended per link) o PLC-5 and SLC 5/04 programmable controllers o Color graphics systems o Personal computers o Host computers o Numerical controls o Programmable RS-232-C/RS-422 devices
Cabling(1770-CD shielded twin-axial): o trunk-line - 10,000 ft max o drop-lines - 100 ft max floating master token passing arrangement o 57.6k bit/s data rate
238
DH PLUS NETWORK
239
DH-485
DH-485 is a local area network (LAN) designed for factory- floor applications.
o
Connects up to 32 devices, including SLC 500 and MicroLogix 1000 programmable controllers, color graphics systems, and personal computers (with RSLinx software)
The DH-485 link together with auxiliary RS-232-C (DF1 protocol) links make up the DH-485 network.
A single PC connected to network can be used to program all the SLC 500 controllers (using SLC500 programming software) o Trunkline length 4000 ft max o 19.2k bit/s max data transfer rate
240
DH-485 NETWORK
241
o o o o
Interconnects smart transmitter in a two-wire network 1200 bps data rate Shielded Twisted Pair (2 wire loop) Support multivariable transmitter with the use of HART splitter Can be used as either conventional transmitter or smart transmitter
o o o
242
HART LAYERS
243
HART BASICS
HART (Highway Addressable Remote Transducer) was developed by Fisher-Rosemount to retrofit 4-to-20mA current loop transducers with digital data communication.
HART modulates the 4-20mA current with a low-level frequency-shift-keyed (FSK) sinewave signal, without affecting the average analogue signal.
HART uses low frequencies (1200Hz and 2200 Hz) to deal with poor cabling, its rate is 1200 Baud - but sufficient.
HART uses Bell 202 modem technology, ADSL technology was not available in 1989, at the time HART was designed
244
data
data
checksum
247
248
HART COMMANDS
The HART command set provides uniform and consistent communication for all field devices
Host applications may implement any of the necessary commands for a particular application.
UNIVERSAL : o Understood by all devices o Provide access to information useful in normal operations (e.g., read primary variable and units).
COMMON PRACTICE Provide functions implemented by many, but not necessarily all, HART communication devices. DEVICE SPECIFIC Represent functions that are unique to each field device. Access setup, calibration, and construction information Available from device manufacturers.
o o o
249
HART COMMANDS
250
HART - Commands
Universal commands (mandatory): identification, primary measured variable and unit (floating point format) loop current value (%) = same info as current loop read current and up to four predefined process variables write short polling address sensor serial number Instrument manufacturer, model, tag, serial number, descriptor, range limits, Common practice (optional)
time constants, range, EEPROM control, diagnostics,
Total: 44 standard commands, plus user-defined commands Transducer-specific (user-defined) calibration data, trimming,
251
HART - IMPORTANCE
Practically all 4-20mA devices come equipped with HART today About 40 Million devices are sold per year.
more info:
http://www.hartcomm.org/
http://www.thehartbook.com/default.asp
252
Very simple sensor bus for building automation, combining power and data on the same wires, transmitting mostly binary signals
D0 = sensor 1 one connection D1 = sensor 2 D2 = actuator 1 AS-Interface Slave IC 1 module enclosure D3 = actuator 2 P0 Watchdog energy
253
ASI
Mechanically coded flat cable - two wires for data and power
insulation piercing connectors simple & safe protection class up to IP67, even after disconnecting
o o o
o o o
directly connected slaves sensors, actuators valve terminals electrical modules etc.
vampire-connector
254
ASI
o o
Master-slave principle o up to 31 slaves on one line o cycle time < 5 ms Each slave can have up to 4 digital inputs + 4 digital outputs additional 4 parameter bits / slave Max. 248 digital Inputs and Outputs
master
To Slave1
To Slave 2
ToSlave 31
To Slave 1
master calls
Slave 1 Slave 2 Slave31 Slave1
slave response
255
ASI TOPOGRAPHY
o o o o no terminating resistor necessary free tree structure of network protection class up to IP67 unshielded 2-wire cable
o data and power on one cable o extension: 100 m (300 m with extender)
star
controller
line
controller
branch lines
controller
tree
Master
controller
Master
Master
Master
Slave Slave Slave Slave Slave Slave
Slave
Slave
Slave Slave Slave Slave Slave
Slave
Slave
Slave
256
INTERBUS-S
Discrete Manufacturing bus
+ + + + + + + +
Standard in CENELEC 1700 products, 270 manufacturers, 375.000 applications Good experience in field wiring (intelligent wiring bar) Easy to engineer Easy to program (IEC 61131) Far extension (400m-13 km) Good response time Conformance test
- Market centered on manufacturing - Limited number of variables (4096 bits) - Ring structure sensitive to disruptions - Sensitive to misplacement - Clumsy and slow message service - Medium user community - Few and costly tools - Strong ties to Phoenix Contact
257
INTERBUS-S TOPOLOGY
BA BA
5-wire
258
Day Four
259
CAN OVERVIEW
CAN is an important embedded protocol Primarily automotive, but used in many other places CAN specifies: Physical layer Protocol layer Message filtering layer (with add-on protocols)
o o o
Note
o o
How message prioritization achieved How small nodes can be kept from overloading with received messages
260
The development of CAN began when more and more electronic devices were implemented into modern motor vehicles. Examples of such devices include engine management systems, active suspension, ABS, gear control, lighting control, air conditioning, airbags and central locking. All this means more safety and more comfort for the driver and of course a reduction of fuel consumption and exhaust emissions.
To improve the behavior of the vehicle even further, it was necessary for the different control systems (and their sensors) to exchange information. This was usually done by discrete interconnection of the different systems (i.e. point to point wiring). The requirement for information exchange has then grown to such an extent that a cable network with a length of up to several miles and many connectors was required. This produced growing problems concerning material cost, production time and reliability.
261
BEFORE CAN
262
WITH CAN
The solution to this problem was the connection of the control systems via a serial bus system. This bus had to fulfill some special requirements due to its usage in a vehicle. With the use of CAN, point-to-point wiring is replaced by one serial bus connecting all control systems. This is accomplished by adding some CAN-specific hardware to each control unit that provides the "rules" or the protocol for transmitting and receiving information via the bus.
263
CAN is a broadcast type of bus. o This means that all nodes can "hear" all transmissions. There is no way to send a message to just a specific node; all nodes will invariably pick up all traffic. The CAN hardware, however, provides local filtering so that each node may react only on the interesting messages.
264
BASIC CONFIGURATION
265
The physical layer uses differential transmission on a twisted pair wire. The bus uses Non-Return To Zero (NRZ) with bit-stuffing.
The nodes are connected to the bus in a wired-and fashion: if just one node is driving the bus to a logical 0, then the whole bus is in that state regardless of the number of nodes transmitting a logical 1.
Max. transfer rate of 1000 kilobits per second at a maximum bus length of 40 meters or 130 feet when using a twisted wire pair which is the most common bus medium used for CAN.
Message length is short with a maximum of 8 data bytes per message and there is a low latency between transmission request and start of transmission. The messages are protected by a CRC type checksum
266
The bus access is handled via the advanced serial communications protocol Carrier Sense Multiple Access/Collision Detection with Non-Destructive Arbitration. This means that collision of messages is avoided by bitwise arbitration without loss of time.
There is no explicit address in the messages, instead, each message carries a numeric value which controls its priority on the bus, and may also serve as an identification of the contents of the message.
An elaborate error handling scheme that results in retransmitted messages when they are not properly received.
There are effective means for isolating faults and removing faulty nodes from the bus.
267
268
269
Only if all nodes transmit recessive bits (ones), the Bus is in the recessive state.
If any one node transmits a dominant bit (zero), the bus is in the dominant state.
oT is Transmitter, R is receiver. Note nodes can therefore check the line while transmitting. This is important particularly during arbitration.
270
CSMA/CD NDA Carrier Sense Multiple Access/Collision avoidance by Non Destructive arbitration
271
272
o o o o
Specifies how small packets of data may be transported from point A to point B using a shared communications medium. It (quite naturally) contains nothing on topics such as flow control transportation of data larger than can fit in a 8-byte message node addresses establishment of communication, etc.
273
274
The messages uses a clever scheme of bit-wise arbitration to control access to the bus, and each message is tagged with a priority.
The CAN standard also defines an elaborate scheme for error handling and confinement.
CAN may implemented using different physical layers, and there are also a number of different connector types in use.
275
Summary: "Hello everyone, here's some data labeled X, hope you like it!" The Data Frame is the most common message type. It comprises the following major parts (a
few details are omitted for the sake of brevity):
The Arbitration Field, which determines the priority of the message when two or more nodes are contending for the bus. The Arbitration Field contains: o For CAN 2.0A, an 11-bit Identifier and one bit, the RTR bit, which is dominant for data frames. o For CAN 2.0B, a 29-bit Identifier (which also contains two recessive bits: SRR and IDE) and the RTR bit. the Data Field, which contains zero to eight bytes of data.
o o o
the CRC Field, which contains a 15-bit checksum calculated on most parts of the message. This checksum is used for error detection.
an Acknowledgement Slot; any CAN controller that has been able to correctly receive the message sends an Acknowledgement bit at the end of each message. The transmitter checks for the presence of the Acknowledge bit and retransmits the message if no acknowledge was detected.
276
Note 1: It is worth noting that the presence of an Acknowledgement Bit on the bus does not mean that any of the intended addressees has received the message. The only thing we know is that one or more nodes on the bus has received it correctly Note 2: The Identifier in the Arbitration Field is not, despite of its name, necessarily identifying the contents of the message. CAN 2.0A (standard CAN 11-bit ID) Data Frame.
277
Summary: "Hello everyone, can somebody please produce the data labeled X?"
o o
The Remote Frame is just like the Data Frame, with two important differences: It is explicitly marked as a Remote Frame (the RTR bit in the Arbitration Field is recessive), and there is no Data Field.
The intended purpose of the Remote Frame is to solicit the transmission of the corresponding Data Frame. If, say, node A transmits a Remote Frame with the Arbitration Field set to 234, then node B, if properly initialized, might respond with a Data Frame with the Arbitration Field also set to 234.
Remote Frames can be used to implement a type of request-response type of bus traffic management. In practice, however, the Remote Frame is little used. It is also worth noting that the CAN standard does not prescribe the behaviour outlined here. Most CAN controllers can be programmed either to automatically respond to a Remote Frame, or to notify the local CPU instead.
278
There's one catch with the Remote Frame: the Data Length Code must be set to the length of the expected response message. Otherwise the arbitration will not work.
Sometimes it is claimed that the node responding to the Remote Frame is starting its transmission as soon as the identifier is recognized, thereby "filling up" the empty Remote Frame. This is not the case. A Remote Frame (2.0A type):
279
280
Summary: "I'm a very busy little 82526 device, could you please wait for a moment?"
The Overload Frame is mentioned here just for completeness. It is very similar to the Error Frame with regard to the format and it is transmitted by a node that becomes too busy. The Overload Frame is not used very often, as today's CAN controllers are clever enough not to use it. In fact, the only controller that will generate Overload Frames is the now obsolete 82526
281
282
283
STANDARDIZATION
Vehicle bus system applications can be separated in three different categories according to their real-time capabilities. Class A for a low speed bus with bit rates up to 10 kbps, e.g for body control applications, Class B for a low speed bus with bit rates from 10 kbps to 125 kbps, e.g. for dashboard and diagnostics, Class C for a high speed bus with bit rates from 125 kbps to 1 Mbps for real time applications like engine management, Gearbox, ABS etc.
For the use of CAN in vehicles two standards have been defined for the bus interface: CAN High Speed according to ISO-IS 11898 for bit rates between 125 kbps and 1 Mbps CAN Low Speed according to ISO-IS 11519-2 for bit rates up to 125 kbps
284
These are the bus levels according to ISO-IS 11898. A recessive bit is represented by both CAN bus lines driven to a level of about 2.5 V so that the differential voltage between CAN_H and CAN_L is around 0 V. A dominant bit is represented by CAN_H going to about 3.5 V and CAN_L going to about 1.5 V. This results in a differential voltage for a dominant bit of about 2V.
285
Cheap CAN controller CPU could get overrun with messages even if it didnt need them.
286
Hardware message filters sort & filter messages without interrupting CPU
287
288
CATERPILLAR EXAMPLE
289
CATERPILLAR EXAMPLE
290
291
PROFIBUS
Fieldbus is the generic name of a family of industrial network systems for real-time distributed control. Human Machine Interface (HMI) at the top, Programmable Logic Controllers (PLCs) in the middle, and the Fieldbus at the bottom.
Area Controller
Factory Level
Bus Cycle Time < 1000 ms
Cell Level
Bus Cycle Time < 100 ms
Profibus-FMS
Field Level
Bus Cycle Time < 10 ms
Profibus-PA Transmitter
Drive
I/O
Valves
Field Device
292
PROFIBUS
PROcess FieldBUS o DIN 19245 o FMS (Fieldbus Message Specification) o DP (Distributed Peripheral) o PA (Process Automation)
32 stations per segment 9.6, 19.2, 93.75 kbps for segment length less than 1200 m
293
Vendor-independent ensured by EN50170 and EN50254 International Standard Open field bus standard for a wide range of applications in manufacturing and process
automation
Allows communication between devices of different manufacturers without any special interface adjustments
Can be used for both high-speed time critical applications and complex communication Continuing further technical developments prepared for the future Widely accepted in European industrial community
294
Token passing procedure o Ensures that the bus access right (token) is assigned to each master within a precisely defined timeframe o The token is passed around the logical token ring once to all masters within a maximum token rotation time o Used only for communication between complex stations (masters)
Master-slave procedure o Permits the master (active station) which currently owns the token to access the assigned slaves (passive stations) o Enables the master to send messages or retrieve them from the slaves ini the following configurations :
Pure master-slave system Pure master-master system (token-passing) A combination of the two
295
296
PRIORITY MECHANISM
Two priority classes: high and low o TTR: target rotation time o TRR: real rotation time o TTH: token holding time, TTH =TTR TRR
At least one high priority message per token reception regardless of TTH Additional high priority messages only if TTH is still positive
Low priority messages after high priority message transmission if TTH is still positive
297
PRIORITY MECHANISM
298
299
COMMUNICATION PROFILES
Defines how users transmit their data serially via the common transmission medium
Decentralized Periphery (DP) o The most frequently used o Optimized for speed, efficiency, and low connection cost o Designed especially for communication between automation systems and distributed peripherals (PLC and field devices) o Suitable as a replacement for conventional, parallel signal transmission
Fieldbus Message Specification (FMS) o Universal communication profile for demanding communication tasks o Offers many sophisticated application functions for communication between intelligent devices (PLCs and PCs) o Will be less frequently used in the future with the usage of TCP/IP in cell level network
300
H1 MACROCYCLE
301
Application layer o program invocation management o variable access o event management o context management (connection establishment) o virtual fieldbus device object o object dictionary management
302
DESCRIPTION
INPUT TRANSDUCER BLOCKS DESCRIPTION TEMP Temperature Transducer INPUT FUNCTION BLOCKS AI DI MAI MDI PUL Analog Input Discrete Input Multiple Analog Input Multiple Discrete Input Pulse Input
Data Communications & Fieldbus Systems
DESCRIPTION
303
DESCRIPTION
ENHANCED PID It has all the standard features plus : bumpless or hard transfer from a manual mode to an automatic mode and bias.
APID
ADVANCED PID It has all the standard features plus: bumpless or hard transfer from a manual mode to an automatic mode, bias. Adaptive gain, PI sampling, deadband for error, special treatment for error, ISA or parallel algorithm,..
Arithmetic - 9 predefined equation types Splitter - split ranging and sequencing. Signal Characterizer - 20 points curve Integrator Analog Alarm Input Selector - 4 inputs Setpoint Ramp Generator Timer
Data Communications & Fieldbus Systems
304
Density Constant FLIP-FLOP and EDGE TRIGGER Modbus Control Slave Modbus Supervision Slave Modbus Control Master Modbus Control Slave Advanced Equations
DESCRIPTION
305
COMMUNICATION PROTOCOL DP
PROFIBUS uses a single, open communication protocol (PROFIBUS DP, Decentralized Periphery) for all applications
The protocol uses the Master-Slave model: One device (master) controls one or more other devices (slaves). The protocol uses the Token Passing model: The token is transmitted across the network; the station in possession of the token controls the access to the network.
Token
DP Slave 1
DP Slave 3
Slave 3
.....
Slave 3
306
COMMUNICATION PROTOCOL DP
PROFIBUS DP exists in three versions:
DP-V0: Overall command structure, cyclic data exchange DP-V1: Extension by acyclic data exchange et al. DP-V2: Further extension by time stamp, clock synchronization et al.
Functional Levels
plus extensions: Clock Synchronization & Time Stamps HART on PROFIBUS Up/Download (Segmentation) Redundancy DP-V1 Acyclic Data Exchange between PC or PLC and Slave Devices
Time
307
Device Features
plus extensions: Integration within Engineering: EDD and FDT Portable PLC Software Function Blocks (IEC 61131-3) Fail-Safe Communication (PROFIsafe) Alarms
PROFIBUS DP (RS-485)
32 devices max. (incl. controller) on one segment. Devices must be daisy chained; no spur lines. Segment must be terminated (T). Baud rate depends on segment length. Repeater are possible, 9 max. per segment. Use of recommended grounding methods.
Devices
Controller Segment
308
One main cable (trunk) and spur lines Maximum length of spurs depends on number of spurs T-connectors with or without short-circuit protection With optional overvoltage protection
309
PROFIBUS DP
Link/ Coupler
PROFIBUS PA
Junction Box
Devices
310
PROFIBUS DP
Link/ Coupler
Junction Box
PROFIBUS PA
Junction Box
Trunk
Spur lines
Devices
311
PROFIBUS PA
T
Junction Box Spur lines Junction Box Trunk
Junction Box
Devices
312
TRANSMISSION TECH
....
PROFIsafe, I&M, iPar-Server, Time Stamp, Redundancy,
Wireless
313
TRANSMISSION TECH
Wired transmission
RS 485; MBP RS 485-IS; MBP-IS (Intrinsically Safe)
RS - 485 PROFIBUS DP Baud rate Devices/segment (max.) Devices/segment (typic.) Cable length max. Spur line length max. 1200 9.6 ... 12.000 kBit/s 32 MBP PROFIBUS PA 31.25 kBit/sec 32 14 ... 20 1900 m 120 m
MBP- IS PROFIBUS PA
31.25 kBit/sec
32
4 ... 6
1000 m
60 m
314
TRANSMISSION TECH
Wired transmission
MBP transmission technology
Fieldbus standard IEC 61158-2 for MBP transmission technology
Up to 32 nodes in one segment Data transmission rate 31.25 Kbit/s Per field device: Min. working voltage 9 V DC Min. current consumption 10 mA Transmission of digital communication signal in zero-mean Manchester II coding (MBP) through 9 mA amplitude Signal transmission and remote power supply using twisted-pair cable Fieldbus cable type A Connection of field devices via stubs (spur) to a main cable (trunk) for trouble-free disconnection of devices without affecting other nodes Max. total length of main cable, including all stubs, is 1900 m
315
TRANSMISSION TECH
Optical transmission
Various types of fiberoptic cables are supported. Typical topology structures are star and ring, linear structures are also possible. The implementation of a fiberoptic cable network involves the use of electrooptical converters.
Fiber type Multi-mode glass fiber Single-mode glass fiber Plastic fiber HCS fiber Core diameter [m] 62,5 / 125 9 / 125 980 / 1000 200 / 230 Transmission range 2 - 3 km > 15 km Up to 100 m Approx. 500 m
316
TRANSMISSION TECH
Wireless transmission
PROFIBUS & PROFINET International do not specify its own wireless solution.
Numerous wireless solutions are available for PROFIBUS from different vendors using gateways to translate the electric signals into a wireless signal.
See the product guide on the PROFIBUS website and search for wireless. Various types of fiberoptic cables are supported.
317
ASSET MANAGEMENT
Unlike conventional communication systems, PROFIBUS allows a detailed view into field devices.
Controller Controller
Devices
Conventional system Very limited system view, device details are invisible.
PROFIBUS
318
ASSET MANAGEMENT
Example from a chemical plant: Level in reactor 2B gets out of spec
Controller
319
DIAGNOSTICS
Before profile 3.02 was introduced
all diagnosis messages have been provided to all users. >> Difficult to manage by the operators
PROFIBUS PA
Device A
Device B
Device C
320
DIAGNOSTICS
With Profile 3.02
Diagnosis messages are mapped to categories already by the manufacturer, categories comply with NAMUR NE 107. Plant operator gets categorized information. Maintenance department gets full information.
Maintenance department Plant operator
Maintenance Failure Required
PROFIBUS PA
Device A
Diag 1
321
In hazardous environments, fieldbus systems must comply with two IEC standards:
IEC 60079: Explosive atmospheres IEC 61158-2: Fieldbus/Physical layer specification
Zone 0, 1 and 2 define areas of a plant, where explosive substances may exist in the air and an electrical spark could trigger an explosion The respective PROFIBUS solution limits the energy going to the bus and the devices to eliminate the danger of generating a sparc
The Intrinsically Safe (IS) version of the MBP physical layer (MBP-IS) complies with this approach.
322
MBP PROFIBUS PA Baud rate Voltage Current Devices/segment (max.) Devices/segment (typic.) Cable length max. Spur line length max. 31.25 kBit/sec 24 ... 30 V 1000 mA 32 14 ... 20 1900 m 120 m
Note: RS485 is also available in an IS-version, which runs at lower power levels with a special coupler and a special wiring. .
323
The FISCO (Fieldbus Intrinsically Safe Concept) provides easy and fast design of PROFIBUS PA installations in hazardous areas.
Only one power source permitted. All other components are drains. Maximum overall cable length 1000 m Maximum spur line length 60 m Power supply, coupler and field devices must be FISCO certified.
324
I.S. does not satisfy completely the needs regarding to cable length and number of devices compared to applications outside of hazardous areas. The High-Power Trunk Concept solves this limitation for use in hazardous areas.
325
PROFIBUS PA
intern
The trunk is installed with increased protection in zone 1 to allow increased supply current for more field devices. The field devices are connected using Ex i ignition protection.
General Purpose / Safe Area
PROFIBUS DP
Zone 2 Class I, Div. 2
PROFIBUS PA
Field Barriers
Terminator
T
Trunk: Ex e Spur: Ex i
Field Devices
Zone 1 Class I, Div. 1/2 Zone 0 Class I, Div. 1
326
intern
PROFIBUS PA
PROFIBUS PA
Coupler
With Short-circuit Protection
Terminator
T
Trunk: Ex nA Non sparking Spur: Ex ic/nL Energy limited
Field Devices
Zone 2 Class I, Div. 2
327
PROFIsafe
Objective: Reduce the risk to an acceptable level.
Risk
Risk reduction measures
Acceptable risk
Zero risk
328
PROFIsafe
PROFIsafe is standardized in IEC 61784-3-3 and complies with SIL 3 according to IEC 61508.
SIL: A performance criteria of a Safety Instrumented System (SIS) which describes, among other things, the Probability of Failure on Demand (PFD). SIL covers four levels SIL 1 to SIL 4.
PFD: A value that indicates the probability of a system failing to respond to an actual demand. PFD is also referred to as safety unavailability.
Safety Integrity Level (SIL) SIL 1 SIL 2 SIL 3 SIL 4 Probability of failure on demand (PFD) per year >= 10 -2 to <10 -1 >= 10 -3 to <10 -2 >= 10 >= 10
-4 -5
10 000 to 1000
329
PROFIsafe
Standard Controller
SF
7 2 1
SF
77 22 11
SF
SF
7 2 1
7 2 1
7 2 1
Black Channel: Not safety-related components such as ASICs, links, cables etc.
SF
PROFIsafe (Safety Function, Safety Layer) Part of the safety-related communication system, located above layer 7 Safety Layers checks addressing, signature, fault tolerance time etc.
Safety-related components (I/Os, controller, control systems) These are not part of PROFIsafe! Not safety-related functions, e.g. diagnosis
Data Communications & Fieldbus Systems
330
WORLDFIP
331
Centralized polling o Bus arbitrator o Producer-consumer model o Bus arbitrator transmitting an ID frame for a certain pair of producer and consumers o Elementary cycle with a period equal to that of most frequent message o Aperiodic variable and message exchange after periodic data transmission o Filling for excess time
332
333
334
IEC 61158
IEC 61158 Fieldbus standard for use in industrial control systems (IEC/SC/65C)
o o o o o o o o Type 1: IEC Fieldbus Technical Specifications Type 2: Control Net Type 3: Profibus Type 4: Pnet Type 5: HSE of the Fieldbus Foundation Type 6: SwiftNet Type 7: WorldFip Type 8: Interbus
335
ISO 15745
ISO 15745 Industrial automation systems and integration - Open systems application integration frameworks (TC 184 SC5)
Part 2: ISO 11898 based control systems DeviceNet, CANKingdom, CANOpen Part 3: IEC 61158-based control systems P-Net, Profibus, WorldFIP, ControlNet, Interbus
336
FOUNDATION FIELDBUS
The Fieldbus market is dominated by Foundation Fieldbus and PROFIBUS. Both technologies use the same physical layer but are not interchangeable.
As a general guide, applications which are controlled and monitored by PLCs tend towards PROFIBUS, and applications which are controlled and monitored by a DCS (digital/distributed control system) tend towards FOUNDATION Fieldbus.
Foundation Fieldbus is an all-digital, serial, two-way communications system that serves as the base-level network in a plant or factory automation environment. Developed and administered by the Fieldbus Foundation.
337
338
o o
In 4-20mA circuits, the power supply, field device and input card are connected in series. In a Fieldbus system, the power supply, field devices and input card are connected in parallel.
fieldbus
Input card
339
CONTROL ROOM
Field wiring hub with spur shortcircuit protection
FIELD
Safe area
340
CURRENT STATE
Two related implementations of FOUNDATION fieldbus have been introduced to meet different needs within the process automation environment.
o H1 is a bi-directional communications protocol used for communications among field devices and to the control system. It works at 31.25 kbit/s and generally connects to field devices. It provides communication and power over standard twisted-pair wiring. H1 is currently the most common implementation.
o HSE (High-speed Ethernet) works at 100 Mbit/s and generally connects input/output subsystems, host systems, linking devices, gateways, and field devices using standard Ethernet cabling. It doesn't currently provide power over the cable.
341
CURRENT STATE
342
Installed base worldwide (published by Fieldbus Foundation, 2008): o >10,000 systems o >700,000 field instruments
Diverse adoption: o 9 of the top 10 oil & gas companies o 24 of the top 25 pharmaceutical manufacturers o 23 of the top 25 chemical producers o 15 of the top 20 pulp & paper companies o 10 of the top 20 food & beverage companies
ARC Advisory Group: powerful value proposition for fieldbus in the hybrid industries-namely, food & beverage and pharmaceutical, particularly when it comes to easing the process of validation and regulatory compliance and the integration of batch procedural operations with continuous processes. The function block structure of Foundation technology lends itself to these processes very well.
343
Industry
Mining and Metals 3% Education 3% Pulp and Paper 4%
N A 43%
O ther 8%
AP 13%
MEA 7%
R EU 25%
344
SUPPORTING TECHNOLOGY
Fieldbus Components: o Bus Terminal o EtherCAT o Fieldbus Box o Lightbus o PC Fieldbus Cards o Switches o And more
345
346
Economic savings o Different incentives at each stage of a facility life cycle: Capital/Construction Commissioning Operations, maintenance o Reduced infrastructure
347
J-Box
Fieldbus devices
J-Box
348
Replace
Cash Flow
Time
Analog fieldbus
349
350
TERMINATION COUNT
Field
FT-100
Junction Box
+ 4 pr 1 2 3 4 5 6 7 8 9 10 11 12 1 pr
Marsh Cab.
4 pr 1 2 3 4 5 6 7 8 9 10 11 12 1 pr
I.S. Barrier
+ + + + 1 + 2 + 3 + 4 + -
I/O Assembly
4 pr
1 pr
FCV-100 + -
16 4
Totals:
TT-100 + -
+ + + + + + -
75 Analog 34 Fieldbus
8 24 15 20 5
PT-101 + -
351
Number of devices per network segment o 15-20 mA per transmitter o 20-25 mA per valve
Bus Bandwidth o Balance cycle time against number of devices o Function of host control system
Virtual Communication Resources o Required for communication between devices and host
352
Cash Flow
353
IS Interface
12.87mA
I/O card
PV = 392.8mb
Leakage + noise
Fieldbus
PV = 392.8mb
IS Interface
PV = 392.8mb
H1 I/O card
PV = 392.8mb
PV = 392.8mb
PV = 392.8mb
354
6%
C a lib ra tio n s h ift Z eroOff
6%
4%
P lu g g ed lin es
F a iled
355
Estimated Project Savings o Materials and Field Devices Some increase o Installation Labor Savings up to 50% o Commissioning:
Significant savings up to 60%
Overall Capital Savings of 2530% However savings will only be realized if the project is planned as a fieldbus project from the start.
356
FF DEVICE COSTS
Costs below are a per device cost o Traditional field instruments $890 o Ff devices w/ single-pair homerun $672 o Ff devices w/ multi-pair homerun $652 o Ff devices w/ connectors $567
357
Advantages
o Reduces installation costs, wiring, less termination and fewer screwdriver turns. o Reduces hardware requirements and Capital Expenditures. o Reduces Operating expenses through plant efficiencies, better asset management, and less maintenance.
Limitations
o
o o
Fieldbus systems are more complex, so users need to be more extensively trained or more highly qualified The price of Fieldbus components is higher Fieldbus test devices are more complex compared to a (high-spec) multimeter that can be used to read and simulate analog 4-20 mA signals One or more Fieldbus standards may predominate in future and others may become obsolete. This increases the investment risk when implementing Fieldbus.
Data Communications & Fieldbus Systems
358
VENDORS
359
PROTOCOL ARCHITECTURE
360
MESSAGE FORMAT
361
362
Manages access to the fieldbus through a deterministic centralized bus scheduler (Link Active Scheduler - LAS)
363
The message in the data buffer is broadcast to all devices on the fieldbus when the LAS issues the compel data (CD) to the publisher The subscribers listen to the message broadcast
364
The message in the queue buffer is transmitted on the fieldbus when the LAS issues the pass token message to device x The message can be sent to a single destination or to multiple destinations (multicast)
365
A clear, and unambiguous, structured text description that precisely describes field device data to host systems.
366
DEVICE DESCRIPTIONS
Attributes like coding, name, engineering unit, write protection, how to display etc.
The arrangement of the parameters in a menu structure, names of menus and submenus. Information about the relation of parameters to others. Information about help texts and help procedures.
Information about necessary operating interactions (e.g. calibration), also called methods. Information about visualization tools (i.e.: charts and graphs)
367
The international standard for developing Electronic Device Descriptions (EDDs) according to the IEC 61804-2.
The Fieldbus Foundation FF-900 Device Description Language Specification is fully conformant to the IEC 61804-2 EDDL Standard.
368
EDDL HISTORY
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
2004
2005
2006
2007
2004
369
Function Blocks (FB) for Process Control Part 1: Overview of system aspects Part 2: Specification of FB concept and Electronic Device Description Language (EDDL)
o o
Covers DDL used in the FOUNDATION fieldbus, HART and Profibus communication protocol Became an International Standard in March 2004
370
DEVICE DESRIPTIONS
What is EDDL?
o EDD is an extended description of the data, user interface and communication in a device which is used by PC applications and handhelds for engineering, commissioning, monitoring, operation and diagnostics.
o EDDs are written in a language called Electronic Device Description Language (EDDL).
o EDDs are operating system (OS) and communication protocol independent: EDDs are not executable code, do not affect OS configuration Adding a new EDD will not disrupt existing applications EDDs support HART, PROFIBUS, FOUNDATION fieldbus & other protocols
371
EDDL BENEFITS
Established millions of devices worldwide System Independent (including OS) A single EDD for all hosts Robust Revision Control Testing and Registration International Standard Uniformity Supported by major suppliers Backward compatible Preserves investment Easy to implement
372
Joint Fieldbus Foundation, PNO and HCF project to specify visualization and data storage management extensions
373
development tools
testing tools
Proven technology: EDD has been in use for over 10 years Easy to use: Developers and end users like it Open and interoperable
374
EDDL COMBINES
Protocol documentation Product Specification
Attributes for Display Coding Engineering Units Menu Structure Menus, Submenu Names Parameter Relationships Help Texts Help Procedures Methods for operating interactions like calibration IEC 61804-2 Standard System Independent (including OS)
375
o Extension to the IEC 61804-2 International Standard o Operating System & Communication Protocol Independent o No Changes to Existing Device Descriptions o Protects Investment in DD-based Devices, Tools & Training o Single Technology for Suppliers and End Users o Reduces Device Development Time & Development Cost o Improves Asset Management Capability
376
EDDL EXTENSIONS
Charting
New CHART construct to define display characteristics New SOURCE construct enables multiples curves on a CHART New AXIS construct
Graphing
New GRAPH construct to define display characteristics New WAVEFORM construct enables multiple curves on a GRAPH. New AXIS construct
New FILE construct describes parameters that will be stored New LIST construct is used with FILE to access specific parameters
377
GRAPHS
A GRAPH is used to present one or more waveforms. Waveforms may be device data, calculated data or persisted data.
Radar Calibration
GRAPH WAVEFORM
AXIS
FILE/LIST
Data Communications & Fieldbus Systems
378
CHARTS
A CHART is used to present the real-time (continuous) device values
Valve Step Response
SOURCE
AXIS
379
Visualization Built-ins MenuDisplay for enabling Wizard-like interface using enhanced Menus
o ListInsert inserts an element into a list o ListDeleteElementAt deletes an element from a list
Improved User Interface (UI) DD Developer can describe screen layout Enhanced MENU construct with screen layout attributes (e.g. dialog boxes)
380
IMAGE
Language codes permit different images for each language supported in EDDL
381
SCADA SECURITY
SCADA insecurity may have contributed to the end of the Cold War
382
The Most Monumental Non-Nuclear Explosion and Fire Ever Seen From Space."
Thomas C. Reed, Ronald Regans Secretary of the Air Force, described in his book At The Abyss (Ballantine, 2004, ISBN 0-89141-821-0) how the United States arranged for the Soviets to receive intentionally flawed process control software for use in conjunction with the USSR's natural gas pipelines, pipelines which were to generate t critically iti ll needed d dh hard d currency f for th the USSR USSR.
Reed stated that "The pipeline software that was to run the pumps, turbines, and values was programmed to go haywire haywire, after a decent interval interval, to reset pump speeds and valve settings to produce pressures far beyond those acceptable to pipeline joints and welds."
The result? A three-kiloton blast in a remote area of Siberia in 1982, which, only by some miracle, apparently didn't result in any deaths. (For context, the Halifax Fire Museum lists the massive 1917 Mont Blanc ship explosion in the Halifax Harbor at a force of 2.9 kilotons.)
383
A forensic summary of the investigation, prepared in the Defense Department, said the bureau found "multiple casings of sites" nationwide. Routed through telecommunications switches in the Gulf, Indonesia and Pakistan, , the visitors studied emergency g y telephone p systems, y , electrical generation and transmission, water storage and distribution, nuclear power plants and gas facilities. Some Some of the probes suggested planning for a conventional attack attack, U U.S. S officials said. But others homed in on a class of digital devices that allow remote control of services such as fire dispatch and of equipment such as pipelines More information about those devices -- and how to program pipelines. them -- turned up on al Qaeda computers seized this year, according to law enforcement and national security officials.
384
In 2000, in Maroochy Shire, Queensland, Vitek Boden released millions of liters of untreated sewage using a wireless laptop, apparently taking revenge against former employers. He was arrested, convicted and jailed arrested jailed.* *
385
Paul Blomgren [] measures control system vulnerabilities. Last year, his company assessed a large southwestern utility that serves about four million customers. Our people drove to a remote substation," he recalled. "Without leaving g their vehicle, , they y noticed a wireless network antenna. They yp plugged gg in their wireless LAN cards, fired up their notebook computers, and connected to the system within five minutes because it wasn't using passwords [ passwords. [] ] Within 15 minutes minutes, they mapped every piece of equipment in the operational control network. Within 20 minutes, they were talking to the business network and had pulled off several business reports.
386
Starting around 14:14, FE [FirstEnergy] control room operators lost the alarm function that provided audible and visual indications when a significant piece of equipment changed from an acceptable to problematic status. Analysis of the alarm problem performed by FE after the blackout suggests that the alarm processor essentially ti ll stalled t ll d while hil processing i an alarm l event. t With th the software ft unable to complete that alarm event and move to the next one, the alarm processor buffer filled and eventually overflowed. After 14:14, the FE control computer displays did not receive any further alarms alarms, nor were any alarms being printed or posted on the EMSs alarm logging facilities. FE operators relied heavily on the alarm processor for situational awareness, since they did not have any other large large-scale scale visualization tool such as a dynamic map board. The operators would have been only partially handicapped without the alarm processor, had they known it had failed. However, by not knowing that they were operating without an alarm processor processor, the operators did not recognize system conditions were changing and were not receptive to information received later from MISO and neighboring systems. The operators were unaware that in this situation they needed to manually, and more closely, monitor and interpret the SCADA information they were receiving.
Data Communications & Fieldbus Systems
387
BLACKOUT
388
SCADA systems are often physically distributed over large areas, making physical security a challenge. Simple vandalism is a real/well known risk:
-- [] vandals shot out approximately 80 individual insulators on the BPA Cougar-Thurston 115 000 volt 115,000 lt t transmission i i li line causing i it t to go out t of f service i at t th that t ti time. Th The vandalism d li occurred near Cougar Dam, which is approximately 25 miles east of Eugene. BPA crews replaced the damaged insulators at an estimated cost of $6,000. Even though no electrical service to EWEB and Lane Electric Cooperative p customers was disrupted p by y the vandalism, , Eugene Water and Electric had to purchase additional power to serve its customers during the 13 hours that it took to repair the damaged line.
-- A A Washington man who admitted to tampering with more than 20 high high-voltage voltage transmission towers in four Western states said yesterday he was trying to point out the power system's vulnerabilities. "I intended to loosen the bolts and by doing so illustrate the vulnerabilities of these towers," Poulin told the judge. Poulin said in a telephone interview before his arrest that he considered his actions necessary to point out that he was able to damage the towers despite being "62 years old, overweight, arthritic, diabetic, half-blind and a cancer patient living on a minimum of 12 medication pills a day.
389
The present state of security for SCADA is not commensurate with the threat or potential consequences consequences. The industry has generated a large base of relatively insecure systems, with chronic and pervasive vulnerabilities that have been observed during security assessments. Arbitrary applications of technology, informal security, and the fluid vulnerability environment lead to unacceptable risk. [] Security for SCADA is typically five to ten years behind typical information technology (IT) systems because of its historically isolated stovepipe organization.
390
Traditionally network and security people have focused virtually all our attention on the enterprise side of the network, ignoring the parallel hidden half of the network associated with process control systems and embedded systems systems.
Process control systems y and embedded systems y use different p protocols, different jargon, and no one ever really mentioned them. They were out of sight and out of mind, and handled by hardware guys.
391
In the old days, process control systems used proprietary protocols and ran with serial communications (e.g., modems) or on physically separated (air gapped) private dedicated networks, but thats no longer always the case.
These days, process control systems often run using MODBUS/TCP on the enterprise LAN and over the Internet; process control traffic may be commingled i l d with ith web b pages, email, il P2P t traffic, ffi V VoIP IP t traffic, ffi etc. t
392
Most SCADA systems were originally built before and often separate from other corporate networks. As a result, IT managers typically operate on the assumption p that these systems y cannot be accessed through g corporate p networks or from remote access points. Unfortunately, this belief is usually fallacious.
393
While enterprise network security is undeniably important, unlike enterprise network security security, SCADA compromises can have real world life safety impacts impacts.
Enterprise p network security y breach: financial consequences, q , customer p privacy y is compromised, systems need to be rebuilt, spam gets sent, etc., but life goes on.
SCADA security breach? Property can be destroyed and people can be hurt or killed.
394
SIMPLE PROTOCOLS
Because SCADA devices with embedded controllers tend to have limited computational power, and were historically connected via low speed serial lines, SCADA protocols tend to be quite simple, with little or no protection against i t spoofing, fi replay l attacks, tt k or a variety i t of fd denial i l of f service i attacks. tt k
In In a demonstration at a recent security conference conference, [Jeff Dagle, Dagle a PNNL (Pacific Northwest National Laboratory) EE] hacked into his testbed system and tripped an electrical breaker. The breaker then signaled the SCADA software that it had opened opened. But the SCADA controller did not respond because it had not instructed the breaker to open. It was a classic denial-of-service attack. "We were demonstrating a weakness at the protocol t ll level l it itself," lf " said id Dagle. D l
395
Industrial plants, and the instrumentation they include, tend to be long life cycle projects ten, fifteen or twenty year project lives are by no means uncommon. As a result, the devices that may be deployed as part of that construction may be virtual antiques by the time the facility is finally decommissioned, and theres no p provision for refreshing g those devices the way you might upgrade out of date PCs in some office.
"A ti i "Anti-virus software ft d doesn't 't work k on these th SCADA systems," t " said id R Robert b t Childs, information security analyst at the Public Service Company of New Mexico, who spoke at NetSec about the challenges in working with SCADA vendors to get them to comply with the new rules. "Many of these systems are based on old Intel 8088 processors, and security options are limited to us.
396
SCADA devices are often controlled from central monitoring stations i (MTU (MTUs, or master terminal i l units). i ) Hi Historically i ll those h were Unix-based systems, but many contemporary MTUs are now Microsoft Windows based.
397
-- the device may use an OS and application that was burned to ROM, and which is not rewritable (upgrade = replacing ROMs)
-- the device may be physically sealed and not upgradeable, or be located in a difficult location, or have no removable media
--- the vendor may no longer be in business, or may not be producing upgrades, or the vendor may not be allowing upgrades
398
CERTIFYING PATCHES
Health care IT professionals say medical device makers prohibit them from changing the systems and even from running anti-virus software in some cases. These Th IT administrators d i i t t say manufacturers f t often ft are slow l t to supply software patch updates and routinely claim the Food and Drug Administration (FDA) requires approval of patch-base changes. However the FDA says it has no such rules
399
Because of the need for p positive access and control, there is a trend toward simple, known, and shared passwords. Users like to avoid situations such as: Do you know the password to turn off the nuclear reactor before it melts down? I forgot mine today today
But theres hope: people in the SCADA community are beginning to talk about strong authorization systems
400
Theres also the sheer issue of managing passwords for thousands of devices passwords will tend to be common across devices as a practical matter (this is much like SNMP community strings)
And of course those passwords arent changed very often (if at all), even when staff transitions occur or years have gone by
401
These days, few of us would be willing to send our passwords over plain text transmissions paths (as we would when using telnet), yet plain text transmissions are still very common in the SCADA world.
In the realtime world, encryption overhead and jitter may be the crucial problems to overcome
402
In many cases, SCADA traffic will be on just one port such as 502/tcp (e.g., Modbus/TCP). This is both good and bad.
The use of a single port (or just a couple of ports) makes it easy to track that traffic, traffic or to poke a hole in firewalls to allow that traffic to pass pass, but it also makes it easy for the bad guys to scan for connected devices, and it makes it impossible to do port-based selective filtering.
403
Despite tantalising accounts of terrorist interest in targeting SCADA networks and other critical infrastructure, there actually appears to be little interest among the hacker community in developing tools and exploits against PLC or industrial protocols such as Modbus/TCP or Ethernet/IP. Unlike IT products, tools for automatically "hacking " PLCs, remote IO devices, robots, or Ethernetbased b d sensors are not t readily dil available. il bl
Bedroom hackers with little or no knowledge of automation systems are, in reality, unlikely to cause deliberate harm.
404
Day Five
405
Ethernet was the first LAN technology to be developed -- the pioneering research work being done in the early 1970's. It was developed by Xerox at a time when support technology was not cost-effective, nor was there a significant market. The microprocessor developments of the late 1970's changed both issues, making the technology viable and creating a market via the personal computer.
To further develop and promote ethernet technology, Xerox partnered with Intel and DEC, creating a strong corporate credibility that has helped make ethernet the most popular LAN technology.
The IEEE adopted ethernet as the basis of the IEEE 802.3 standard (although with subtle differences from the version promoted by DEC/INTEL/XEROX -- the so-called "DIX" ethernet).
Ethernet is a bus architecture system (which implies a common, or party-line broadcast medium) which is shared using a contention access control technique known as carrier sense multiple access with collision detection (CSMA/CD).
D-116
406
Whenever there is a collision involving two or more stations, they will each independently determine that there has been a problem, stop transmitting, and reattempt the transmission later.
The retry waiting time is randomly selected (from within a specified time interval), so as to minimize the risk of a second collision event.
However, when the system is busy, there is a significant probability of a second collision with some other station.
In order to separate (in time) stations that have collided (and to smooth out traffic patterns that peak at some moment -- presumably caused by some external event), the protocol provides for a station that encounters more than one collision to double the size of the window within which it sets its random timer for each subsequent retry interval.
This process makes it increasingly likely that a station that has had a run of bad luck (in getting its message transmitted) will encounter increasingly bad luck (i.e. It becomes a first in - last out queuing process).
This doubling of the time window continues for 10 events (so that the random time selection window can grow to 2**10 = 1024 times its original value). If the collisions continue beyond 10 events, the time window stays unchanged up to 16 collisions, at which time an error is reported to the LLC layer, typically requiring some kind of manual intervention. This timer control scheme is called the truncated exponential backoff algorithm.
As a consequence of this mechanism, it is not possible to guarantee a bound on ethernet access time delay -- such as is often required for real-time applications.
D-276
407
TWO CONDUCTOR PRESSURE CONTACT/CONNECTION (Vampire Tap) LOCATED AT MIN. 2.5 m INTERVALS (Cable is marked every 2.5 m)
COAXIAL CABLE: Comprised of a center core conductor, cylindrical insulator and outer cylindrical shield conductor surrounding the assembly. Typically, an outer insulating jacket is provided around the composite cable. Ethernet uses a 50 Ohm cable(similar to RG 213), colored yellow if it is PVC insulated; orange if it is plenum-safe).
D-082
408
UPPER LEVEL FUNCTIONS LOGICAL LINK CONTROL MEDIUM ACCESS CONTROL PHYSICAL (SIGNALING INTERFACE) OSI PHYSICAL LAYER
AAA A AA AA AA A AAAA A AA AAA
BNC-Type twist-lock connector is used. Each workstation has a "T" connection, thus providing for an onward connection to other users.
COAXIAL CABLE: Comprised of a center core conductor, cylindrical insulator and outer cylindrical shield conductor surrounding the assembly. Thinwire coax is RG58 R/U, 50 Ohm cable, approximately 1/4 inch diameter.
A AA AA AA A
D-084
409
ETHERNETUTPWIRINGCONCEPTS
UNSHIELDEDTWISTED-PAIR(UTP) HOUSECABLINGISALWAYS"HOMERUN"
PATCHCORD
WALLPLATEAND RJ-45CONNECTOR
D-021
410
ETHERNETRJ-45PIN CONNECTIONCONVENTIONS
StandardCableConfigurations: Ethernet(10and100Base-T)
1 + 2 3 + 4 5 6 7 8
EthernetTransmitPair EthernetReceivePair
PINNUMBER 12345678
D-424
411
INDUSTRIALETHERNETCONNECTORS
Thenormal,office-orientedRJ-45Ethernetconnectorisa sourceofconcernwhenadaptedtoindustrial applications. Thisismainlyduetothelight-duty,fragileconnections thataremadewhentheconnectorsarecrimpedontothe cables,andtheminimalconductingsurfaceareawhen theyaremated. Industrialenvironmentalissuesincludeexposuretodust, temperatureextremes,excessivehumiditylevels, electromagneticinterference,andvibration--allofwhich cansignificantlyjeopardizereliability. Inthiscontext,standardRJ-45plugsandjacksmay corrode,wear,clogwithdebris,andeventuallyfail. Failingtotakeappropriatecarewiththeseseemingly trivialitemscanleadtoseriousunreliabilityproblemsand resultinghighmaintenancecosts.
continued ....
RJ45CONTACTPROBLEMS
K-074
412
INDUSTRIALETHERNETCONNECTORS,cont.
OneapproachistofittheconventionalRJ-45plugsandjackswithanouterhousingthatwill protectthemfromdirt,humidityandvibrationexposure--seeexamplesshowninboxbelow.
Variationsonthisideatypicallyincludescrewandbayonet-stylesheaths,orjacketsthatare designedtoprotecttheencasedRJ-45plugandjackassembly. Alternativestothese"hardened" RJ-45Ethernetconnectordesigns isacompletelydifferentconnector approach. SomehaveusedDB-9shell connectors(commonlyusedon PCserialports)--whichcanbe solder-connectedandscrew anchored. Terminatingindividualconductors withscrewlugsandbarrierstrips isanotherpossibility--avery reliableandrobustsolution,but labor-intensivetoinstall.
Siemon'sIndustrialMaxconnectors(www.siemon.com)
AMPNetconnectEthersealconnectors(www.ampnetconnect.com)
K-075
413
Cable segment maximum length: 500 meters Maximum number of attachments per segment: 100 taps Minimum tap separation: 2.5 meters Maximum number of devices per network: 1024 (limitation common to all configurations) THIN-WIRE ETHERNET (10-BASE-2) Cable segment maximum length: 200 meters (some 189 meters) Maximum number of attachments per segment: 30 attachments Minimum connection-point separation: 1 meter TWISTED-PAIR ETHERNET (10-BASE-T)
Star-wired cable configuration; most vendors limit cable lengths to 100 meters Each wire run supports only one user Multi-port repeater hubs aggregate multiple user connections FIBER OPTIC BASED ETHERNET (10-BASE-F)
Fiber cable configurations are nominally suited to "backbone" roles, linking star couplers Maximum fiber cable segment/system length with passive star couplers: 2,200 meters Maximum fiber cable system (end-to-end) length with active collision detectors: 4,400 meters
D-085
414
Ethernet based on fiber optic cable segments will deploy in one of three configurations; point-to-point, passive star coupled and active star coupled. Point-to-Point
Point-to-point use of fiber optic technology is simple, involving a copper-to-fiber repeater connection at each end (of a two-fiber cable span).
This class of service is typically configured using components compatible with passive star coupling (see below). Thus, this connection is called a passive fiber optic segment (PFOS). Passive Star Coupled
The passive star coupler provides an optically coupled connection (light path) between all of its cable ports -- really an optical hub.
Passive-connected ethernet components (medium attachment units and couplers) use a different data format and a different collision detection method than comparable active components -- and also different from conventional metallic cabled systems. Passive optical segments (PFOS) may be up to 500 meters length.
D-426
415
Correct functioning of the collision detection logic in this passive configuration is based on a data reformatting procedure somewhat like that used with token ring, whereby an 'illegal' manchester data pattern is generated, thus eliminating the long synchronization preamble in metallic ethernet data formats. Active Star Coupled
The active star coupled configuration provides for complete optical signal regeneration at the coupler, or hub point -- thus significantly increasing the number of ports that can be feasibly supported, and the practical maximum span distance. Maximum active fiber optic segment length is 2,000 meters
Active hubs support collision detection at the hub itself (with the SQE collision signal encoded in the modified data stream) -- this being the principal feature enabling the now 4,000 meter edge-to-edge optical backbone segment.
D-427
416
CABLECONNECTIVITYBETWEENBUILDINGS
FIBEROPTIC CABLELINK
FIELDTERMINAL OH H
FIELDTERMINAL
K-093
417
60 %
40 %
10
D-241
418
ETHERNETFRAMELENGTHvs.PHYSICALLENGTH
Toappreciatetherelationshipbetween message(frame)lengthandoverallphysical dimensionsoftheLAN,considerthefollowing:
Step1:Assumethechannelisidle,and"A"is readytotransmit,anditdoes.Theframefrom "A"propagatesatfinitespeedtotheextremity ofthecablesystem,takingatimewhichis dependentoncablelength. Step2:Assume"B"nowmakesthe (unfortunate)decisiontostarttransmitting momentarilybeforetheframefrom"A"arrives, andbeforeitisinapositiontoknow.thatthere isothertrafficonthechannel.
W/S
1 A
W/S
W/S
ONE-WAYTRANSITTIME= tseconds
2 A
W/S
W/S
ONE-WAYTRANSITTIME= tseconds
3 A
RETURNTRANSITTIME= tseconds
W/S
Step3:"B"learnsimmediatelythattherehasbeenacollision--but"A"hasnoknowledgeofthe conflictuntilsufficienttimeelapsesfortworound-tripframetransitsofthenetwork.
Therefore,theminimumlengthmessagemustoccupythechannelfor2xtseconds.Inpractice, theIEEE802.3standardspecifiesa64byteminimumframelengthforthisreason.
D-270
419
DISTANCECONSTRAINTSWITHETHERNET
TherearetwounrelatedconstraintsoncablelengthinEthernetsystems:
CollisionDiameter
ItiscriticalthatdevicesthatshareacommonEthernetsegmentarecapableofeffectively detectingcollisionsthattheyhavecontributedto.
Eachoftheseveral802.3-seriesofEthernetstandardspecificationsprescribeamaximum deviceseparationwithinwhichdistancethiscapabilityisassured--thecollisiondiameter.
Drivedistanceistypicallyalessrestrictiveconstraintthancollisiondiameter,anddependingon cabletypeselectedcanbeaslargeasmanykilometers.
Thus,fullduplexfiber-connectedEthernetsystemsarefeasiblyabletooperateovermetropolitan areadimensions--makingEthernetintoacredibleMANtechnology.
D-743
420
DRIVEDISTANCE
Drivedistanceisthemaximumdistanceasignalisabletotraveloveraspecifiedmedium, andstillbeusableatthereceivingend.
Drivedistanceconstraintsarealwaysmuchlesslimitingwiththeuseoffiberopticmedia, becauseofthesuperiortransmissioncharacteristicsofthemedium.
D-501
421
Full duplex LAN technology eliminates all problems with access control (especially significant with ethernet, where the diameter of the collision domain strongly interacts with performance).
Full duplex configurations are by definition point-to-point (with one device having unchallenged access to one communication path).
Application of the concept is typically limited to backbone (hub-to-hub) connectivity, but may be reasonably extended to include high volume end systems, such as enterprise servers.
D-619
422
SSAP
1 BYTE
CONTROL FIELD
1 or 2 BYTES
42 to 1497 BYTES
SOURCE ADDRESS
2 or 6 BYTES
LENGTH COUNTER
2 BYTES
4 BYTES
14 or 22 BYTES
4 BYTES
PREAMBLE: Consists of a 7 byte string of alternating 1's and 0's. Purpose is to establish bit-level synchronization at all receiving stations. START FRAME DELIMITER: Consists of a 10101011 bit pattern (continuation of the preamble pattern, marked with "11" at the end) ; Purpose is to mark the start of the frame. ADDRESS FIELDS: Message destination and source address specification. There are two forms; 2 byte form is "locally administered", 6 byte form is "universally administered" (no 2 devices are manufactured with the same universal address).
DATA FIELD: Contains data content provided by LLC layer; combined with PAD FIELD, composite must be no shorter than 46 bytes, ensuring complete message is never less than 64 byte length.
PAD FIELD: Optional field provides filler characters to ensure compliance with message length requirements.
DSAP and SSAP: Destination and Source Service Access Point indicators (pointers or links to upper layer protocols -- destination and source respectively).
D-143
423
Ethernet users are moving rapidly to replace a default 10 BASE-T architecture with 100 BASE-T -or at least implement new systems with a simple upgrade path (switch selectable NIC's and Category 5 UTP).
100 BASE-T technology operates over the same UTP/fiber optic media as 10 BASE-T, and is capable of supporting the conventional 100 meters span required for typical workstation access.
This new technology is interoperable with 10 BASE-T, and supports the same shared cable collision-arbitrated mode of operation. Similarities to 10 BASE-T CSMA/CD access protocol IEEE Standard 802.3 and broadly supported by vendor community 100 meter lobe star topology (hub based) Cable media: Category 3, 4 or 5 and fiber optic Differences from 10 BASE-T
10-fold increase in transmission speed Does not support obsolescent bus topology cable plant Operating diameter of collision domain max. = 210 meters (c.f. 500 meters for 10 BASE-T)
D-451
424
10 BASE-T CONNECTION
SWITCHING HUB
SWITCHING HUB
SWITCHING HUB
D-452
425
MEDIA ACCESS CONTROL (MAC) PHYSICAL LAYER PROTOCOL (PHY) PHYSICAL MEDIUM DEPENDENT (PMD)
Fast Ethernet (100 Base-T) borrows the lower layer protocols directly from the ANSI FDDI specifications, and the upper layers of the traditional IEEE 802-series standards
D-659
426
AUTOSENSING
Animportantmanagementchallengewithanysystematicupgradeoftechnologyisthelogisticalissue ofmaintainingoperationalcontinuitythroughtheprocess.
TheIEEEstandardspecifiesanoptionalspeedsensingfeature,properlycalledNwayautosensing, whichallowsboth10BASE-Tand100BASE-Ttechnologytocoexistonthesamenetwork.
Autosensingdevicesareabletosensewhetherornotthereare10BASE-TdevicesontheLAN,orif thenetworkis100%100BASE-Tequipped.
Thisparallel10/100dualspeedfeaturecreatesabigoperationalbonusbyenablingaseamless upgrade/transitionpathtothemoredesirable100BASE-Tmodeofoperation.
Earlyversionsofthisautosensingtechnologyhadaproblemwithmarginalcableconfigurations,ill advisedlyshiftingto100Mbpswhenconnectedbyinferiorcableinfrastructure.
D-500
427
GIGABITETHERNET
TheapplicationofgigabitEthernetisprincipallydeployedasabackbonetechnology,typicallywith 100MbpsEthernetusedforaccesstoenddevices.
ThephysicallayerstandardadoptedforgigabitEthernetistakenfrompre-existingFibreChannel standards.
AlthoughcontentioncontrolproceduresarepartofthegigabitEthernetstandard,mostoftheactual applicationsoperatefullduplex,point-to-pointbetweenswitchinghubs.
ThisisachievedthroughaschemetomodifytheEthernetframetoextenditslengthasittransits gigabitEthernetsegments,soastoensurethatcollisiondetectioncapabilityissustained.
(20422)
428
RELATIONSHIP:OSIANDICALAYERED PROTOCOLVIEWS
TRADITIONAL OSI7-LAYER MODEL
APPLICATION
HTML JAVASCRIPT
PRESENTATION
APPLICATION LAYER
HTTP
SESSION
TRANSPORT
HOST-TO-HOST LAYER
NETWORK
INTERNET PROTOCOL
INTERNET PROTOCOL
LLCSUB-LAYER DATALINK NETWORK ACCESS LAYER(NAL) PHYSICAL MACSUB-LAYER PHYSICAL LAYER (PHY)
IEEELANSTANDARDS PROVIDEFOR SPLITTINGTHEDATA LINKLAYERINTOTWO PARTS: THELOGICALLINK CONTROL SUB-LAYER(LLC) THEMEDIAACCESS CONTROL SUB-LAYER(MAC)
D-250
429
CONNECTIONLESSvs. CONNECTION-ORIENTEDSERVICES
CONNECTION-ORIENTEDSERVICES
Involveacall(session)set-upprocedure,acontinuingdialogandformalclosureprocedures.
SimilartoX.25packetswitchingcallcontrolprocedures--andmuchlikeHDLCcontrolsequences.
Errorrecoveryistheresponsibilityoftheseconnectionservices--atwhateverlayertheyoperate.
CONNECTIONLESSSERVICES
Oftencalledunreliableservice,messageunitsarecalled datagrams. Messagesaresentwithoutpriornotice--andwithoutany prioragreementfromthereceivingstationtoparticipate. Becausethereisnoformalcustodytransfer understanding,thereceiverisnotaccountablefor acknowledgingreceiptofmessagesdeliveredthisway. Analogoustofirstclassletterservices--thepostoffice makesa"bestefforts"attemptatdelivery,butneither failurenorsuccessareadvisedtothesender.
Networktechnologiestowhich thisconceptapplies:
TCP/IP--TCPisconnection- oriented;IPisconnectionless.
LANsystems-typicallylow-level LANprotocolsareconnectionless.
D-220
430
FTP
SMTP
ELECTRONIC MAIL
DNS
RTP
NAME SERVER
HOST-TO-HOST LAYER
INTERNET PROTOCOL
REGIONAL NETWORKS
ENTERPRISE WANs
ETHERNET LANs
PPP
INTERNET INFRASTRUCTURE
INTRANET INFRASTRUCTURE
D-185
431
RELATIONSHIPBETWEENTCP/IPLAYERS
HOSTSYSTEM
APPLICATION
MSG HDR
HOSTSYSTEM
MESSAGE
APPLICATION
HOST-TO-HOST
TCP HDR
SEGMENT
HOST-TO-HOST
ROUTER
INTERNET
IP HDR
ROUTER
IP HDR
DATAGRAM
INTERNET
DATAGRAM
INTERNET
IP HDR
DATAGRAM
INTERNET
NET.ACCESS
PKT HDR
PACKET
NET.ACCESS
PKT HDR
PACKET
NET.ACCESS
PKT HDR
PACKET
NET.ACCESS
FRAME HDR
FRAME
FRAME TLR
FRAME HDR
FRAME
FRAME TLR
FRAME HDR
FRAME
FRAME TLR
NETWORK "A"
NETWORK "B"
NETWORK "C"
T-084
432
INTERNET PROTOCOL
The Internet Protocol (IP) header (as illustrated in the box) is always found on the front of a datagram -- a uniform, and uniformly interpreted part of an otherwise highly variable environment.
32 BIT WORDS
1 2
3 4 5 6 VERSION IHL
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
TYPE OF SERVICE IDENTIFICATION FLAGS
TOTAL LENGTH
FRAGMENT OFFSET
TIME TO LIVE
PROTOCOL
HEADER CHECKSUM
SOURCE (IP) ADDRESS
The IP header is never shorter than 5 words (20 octets) in length, but options fields allow it to be as large as 15 words long. The Internet Header Length (IHL) field specifies the actual header length.
PADDING
7 8
The Version number field specifies the version of IP to which the header format conforms -here Version 4 is assumed (although Version 6 is designated as the near-term replacement).
The Type of Service field specifies quality of service (QoS) requirements; four bits in this field specify routing priority based on minimization of delay, maximization of throughput, optimization of reliability or minimization of cost (but not all routers are able to support multiple routing decision criteria).
T-355
433
Total Length specifies the total datagram length -- measured in octets. In practice, message length is constrained by the associated network technology, which will define the maximum transmission unit (MTU) -- the largest message payload that a given network technology can support. Fragmentation Issues
IP supports flexible interconnection of networks with differing MTU limitations -- resulting in a need to accommodate message size conversion at the point of interconnection (routers) -- a process described as fragmentation.
The Identification field carries a unique number -- effectively a sequence number -- that makes it possible for the receiving system to properly associate all fragments of the original message, making reassembly feasible. The Flags field carries two single bit flags used as follows:
The DF (don't fragment) bit is used to prevent fragmentation of a datagram The MF (more fragments following) bit is used to indicate that this is not the last fragment in a series.
The Fragment Offset is a pointer (calibrated in 8 octet units) indicating the position that a given fragment occupied in the original datagram.
T-356
434
The Time To Live (TTL) field is used to prevent the endless circulation of datagrams that are undeliverable because of routing problems. The TTL is a counter which is initialized by the source station (at typically 128), and decremented by one unit by each router that it transits along the way. If the TTL counter value ever reaches zero, the datagram is judged to be undeliverable and is destroyed.
The Protocol field specifies the transport layer protocol, which defines the format of the header which follows the IP header -- typically TCP (a short sampler of protocols and their associated code numbers is provided in the accompanying box).
The Header Checksum is defined by a simple error check calculation, and is used to verify that the IP header contents are correct. The 32-bit Source and Destination IP Addresses define the logical end-to-end source and destination for a given datagram. It is common to represent these address values using a decimal value for each octet individually, separated by decimal points. The resulting "dotted decimal" address representation will be in the form (for example) of 204.174.16.1
PROTOCOL DESIGNATIONS (NUMBERS) ARE ADMINISTERED BY THE INTERNET ASSIGNED NUMBERS AUTHORITY (IANA)
DESCRIPTION INTERNET CONTROL MESSAGE PROTOCOL GATEWAY-TO-GATEWAY PROTOCOL TRANSMISSION CONTROL PROTOCOL EXTERIOR GATEWAY PROTOCOL INTERIOR GATEWAY PROTOCOL (ANY) USER DATAGRAM PROTOCOL ISO TRANSPORT PROTOCOL CLASS 4
T-357
435
A number of options are supported by IP, four of which are of general interest:
The Strict Source Routing option provides a complete list of the IP addresses of routers that the datagram is required to visit along the way (this completely prespecifying and controlling the route the datagram is to follow).
The Loose Source Routing is similar, but only incompletely specifies the route to be taken.
The Record Route option provides for logging the IP addresses of routers visited along the way (effectively the inverse of source routing).
The Timestamp option provides for capturing time-of-day clock information at routers visited (for purposes of network time synchronization).
IP header options are formatted as entries in a list, allowing more than one option to be supported in a given datagram (subject to not exceeding the maximum allowable IP header size of 15 words).
Frequent use of IP header options will degrade router-based network performance, because of the considerably greater computational burden which results from forcing an interpretation of header options at every routing point along the way.
T-358
436
DOTTEDDECIMALADDRESSREPRESENTATION
The32bitbinaryformatofanIPaddressisnotaveryusefulwayofrepresentingaddressesfor humanuseorrecognition.
Considerthefollowingexample,ahypothetical32bitIPaddress:
10000000 00001011 00000101 00011001
T-071
437
THERE ARE THREE PRINCIPAL NETWORK ADDRESS FORMS IN USE ON THE INTERNET, ILLUSTRATED BELOW (additional, less common forms are given on the next page) CLASS A NETWORK ADDRESSES, Featuring: 7 bits of NETWORK ID (Sufficient to identify 126 major networks) 24 bits of HOST ID (Sufficient to identify 16,777,214 hosts on each network)
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
NETWORK ID
HOST ID
14 bits of NETWORK ID (Sufficient to identify 16,382 intermediate sized networks) 16 bits of HOST ID (Sufficient to identify 65,534 hosts on each network)
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
10
NETWORK ID
HOST ID
CLASS "B" ADDRESSES START WITH VALUES FROM 128 through 191
CLASS C NETWORK ADDRESSES, Featuring: 21 bits of NETWORK ID (Sufficient to identify 2,097,150 small networks) 8 bits of HOST ID (Sufficient to identify 254 hosts on each network)
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
11 0
NETWORK ID
HOST ID
CLASS "C" ADDRESSES START WITH VALUES FROM 192 through 223
T-069
438
INTERNETADDRESSINGCONVENTIONS,cont.
TherearetwoadditionalnetworkaddressformsprovidedforintheInternetstructure:
Multicastaddressing,whichprovidesforakindofselectivebroadcastmessagetoallmembersof somepredefinedgroup.
Featuresandfeasibilityofmulticastingaredependentonthetechnologyinuse(Ethernet,for example,supportsmulticastoperation).
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
1 1 1 0
MULTICASTADDRESS
Thefifthclassofaddressing(illustratedbelow)isreservedfounspecifiedfutureuse.
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
1 1 1 1 0
HOSTID
T-070
439
CLASSLESSINTERDOMAINROUTING
ClasslessInterdomainRouting(CIDR)wasdevelopedtodealwiththeseproblems.
T-657
440
CLASSLESSADDRESSHIERARCHY
IANA
ARIN
LACNIC
APNIC
RIPE NCC
AFRINIC
RegionalInternetRegistries (Possiblyallocatedbits3-6)
NIR
NationalInternetRegistries (Possiblyallocatedbits7-12)
LIR
LIR
LocalInternetRegistries (Possiblyallocatedbits13-17)
ISP
ISP
ISP
InternetServiceProviders (Possiblyallocatedbits18-22)
END USER
END USER
END USER
END USER
END USER
End-UserOrganizations (Possiblyallocatedbits23-31)
1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
Class"C"Format: 1 1 0
T-658
441
Internet Control Message Protocol (ICMP) is used for a range of applications, primarily in the category of error reporting services.
The connectionless, "best efforts" delivery philosophy of IP is made more effective with feed back information that can be used to modify behavior at the sending site. ICMP provides that feedback mechanism with the following message types:
Destination Unreachable - Datagram was undeliverable because of addressing or administrative prohibition (including 'DF' bit restriction) Time Exceeded - Time-to-live or fragment reassembly timers expired
Source Quench - A flow control feature designed to prevent data loss at network congestion points.
Redirect - A routing advisory message (router-to-host) instructing a better starting router choice. ICMP also supports two administrative applications: Timestamp Request/Reply - A host-to-host time coordination service.
T-359
442
The echo request/reply ICMP messages comprise a transaction pair that is used to verify the reachability (Is it a valid IP address, and is the network connectivity intact ?) and operational status (Is it turned on ?) of a specified destination router or host.
The echo request/reply is a useful first level network diagnostic tool for verifying feasibility of a network connection, and for roughly estimating network performance.
PING application-level programs provide a common human interface to the echo request/reply service.
Many different versions of PING have been developed, but they typically allow someone to initiate an echo transmission with a simple keyboard command like
ping destination_address
In its simplest form, PING will respond by confirming success in reaching the specified destination (Failure is indicated by the absence of a reply), and displaying the network round-trip transit time delay that was encountered.
Many implementations of PING will permit initiation of multiple such transactions, and present statistically summarized transit time delay results
T-039
443
TRACEROUTE
Tracerouterecordseachofthehopsalongtheroutefromthelocalhosttothedesignatedremote host,andthereforeprovidesconsiderablymoreinformationthanping.
TraceroutesendsaseriesofUDPdatagrams,alladdressedtothedesignatedremotehost--with theTime-To-Live(TTL)valueinitiallysetto1--andincrementedforeachiteration.
Withthefirstdatagram,thefirstrouterinthepathdecrementstheTTLtozeroand,returnsanICMP TIME-EXCEEDEDtypemessagetotheoriginatinghost,andintheprocess,revealsitsIPaddress.
ThescreendisplayfromTracerouteshowstheresponsefromeachrouteronanewline,andthe round-tripdelay(s)thatwereobservedfromtypicallythreetestsperformedperrouter.
T-621
444
ADDRESSESMAKE CONNECTIVITYHAPPEN
Bridges,RoutersandSwitchesarethedeviceswe usetoachievenetworkconnectivity.
BridgesworkwiththeDataLinkLayeraddresses-- hereassumedtobethe48-bitMACaddresses.
UPPERLAYERS
UPPERLAYERS
RoutersworkwithNetworkLayeraddresses--here assumedtobethe32-bitIPaddresses.
TRANSPORT
TCP
3
NETWORK
LAYER
ADDRESSESUSEDBYTHENETWORKLINKLAYER (LAYER3)ARESYSTEM-WIDE(GLOBAL).
LLC MAC
DATALINK
INTHECASEOFIP,THEYARETHE32-BITIP ADDRESSES.
PHYSICAL
PHYSICAL
ADDRESSESUSEDBYTHEDATALINKLAYER(LAYER2) ARELOCAL.
INTHECASEOFLANs,THEYARETHE48-BITMAC ADDRESSES
T-680
445
The Address Resolution Protocol (ARP) is a procedure used to determine the physical (hardware) address corresponding to a given IP address -- a capability that is a critical link in the process of actually delivering a datagram to the destination host over the physical network.
ARP is very effective in resolving addresses on broadcast-oriented networks (such as shared media LAN's like ethernet) but would be disastrous on wide area networks like X.25 or frame relay.
The router creates and maintains an ARP table based on the results of ARP transactions (avoiding the need to repetitively ARP for the same data).
The ARP sequence of events unfolds as follows: 1. Incoming datagrams are accepted by the router. 2. The router broadcasts over the local network with an ARP request, asking who out there recognizes the destination address. 3. The device that owns the specified IP address will generate an ARP reply message carrying its local hardware address -- sufficient to deliver the datagram to the correct device.
3 1
ROUTER
T-046
446
HOST #1
NETWORK #2
C
ROUTER
INTERNET
RO A UT ER
HOST #3
NETWORK #1
B
ROUTER HOST #2
NETWORK #3
D
ROUTER
NETWORK #4
ARP TABLE
HOST COMPUTER SYSTEM
T-228
447
IPDELIVERYPROCESSREVIEW
SOURCE NETWORK
SOURCE HOST ENTRY ROUTER
TRANSIT NETWORK
EXIT ROUTER
DESTINATION NETWORK
DESTINATION HOST
TheSourceHostIPprocessinitiatestheIP datagram,specifyingthesourceanddestination IPaddress. SincetheDestinationHost'sIPaddressidentifies adifferentnetworkfromthesource,theSource HostwillconsultitsHostRoutingTable,and identifytheIPaddressoftheEntryRouter. UsingtheEntryRouterIPaddress,the correspondingMACaddresswillbediscovered withtheARPprocess,andtheIPdatagramwill beencapsulatedinaMACframeusingthisMAC destinationaddress.
T-642
448
Point-to-Point Protocol (PPP) was developed as a replacement for SLIP, with the intent that it would address the limitations which frustrate SLIP usage
PPP is a fully functional (and therefore radically more complex than SLIP) implementation capable of:
administrative coordination of addresses supporting multiple network protocols error detection and correction
PPP is structurally an HDLC link protocol implementation, with additional features to enable identification of upper layer protocols (ULP's) and higher level link control processes PPP was developed to support the TCP/IP protocol suite in a more generally functional way than SLIP; the result has been a highly versatile point-to-point services protocol which is becoming increasingly important in enterprise-wide WAN connectivity (See box)
Most existing router-connected wide area networks are served by proprietary point-to-point protocols, forcing users into a single vendor network environment
Considerable progress has been made in using PPP as a open-systems solution to this problem
In addition to native IP support, PPP has been extended to support IPX, Appletalk, DECnet and OSI CLNP network layer protocols -- thus supporting general multi-protocol WAN environments
T-122
449
Point-to-Point Protocol (PPP) is a data link protocol designed to carry IP datagrams over serial lines, taking a much more sophisticated approach to the problem than was done with SLIP.
PPP is uniquely able to operate over either character-at-a-time (asynchronous protocol) links and bit-oriented (synchronous) link systems.
Various extensions to PPP have been developed, supporting data compression, inverse multiplexing (operating over multiple parallel paths concurrently) and authentication procedures.
PPP is based on the long-standing HDLC (High Level Data Link Control) protocol frame structure, widely used in X.25 and other synchronous wide area networks.
ADDRESS
11111111
CONTROL PROTOCOL
00000011 16 BITS
INFORMATION (PAYLOAD)
FCS
FLAG
16 BITS
01111110
NOTES:
(1) ADDRESS FIELD IS UNIFORMLY SET TO HEX FF (2) CONTROL FIELD IS HEX 03 (Unnumbered connectionless frame)
(3) THREE PROTOCOL TYPES ARE COMMON: LINK CONTROL PROTOCOL (FIELD SET TO HEX C021) NETWORK CONTROL PROTOCOL (FIELD SET TO HEX 8021) INTERNET PROTOCOL (FIELD SET TO HEX 0021)
T-474
450
The protocol field specifies the upper layer protocol being serviced, equivalent to the Services Access Point (SAP) codes used with the IEEE 803.2 LLC standard.
The three protocol field uses common with IP applications provide for link initiation (Link Control Protocol), link control, such as initializing upper layer protocols (Network Control Protocol), and Internet Protocol itself.
In IP applications, the IP-specific version of the Network Control Protocol (NCP) is called Internet Protocol Control Protocol (IPCP).
In addition to these IP-related protocols, PPP supports a wide variety of other network layer protocols, like Novell IPX, XNS, DECNET and Appletalk.
Because the control logic of HDLC (in synchronous link applications) is critically dependent on correctly recognizing the delimiting flags, a special bit-stuffing process is used to break up happenstance message data that coincidentally replicates the flag pattern (HEX 7E).
The problem of possible flag ambiguity is handled in asynchronous implementations of PPP by inserting a HEX 7D (ASCII: }) in front of all happenstance flag data patterns, or any octet of value less than HEX 20 (ASCII control codes).
Operationally, the LCP serves to establish a link connection (and maintains responsibility for link maintenance and termination); IPCP then takes responsibility for IP module initialization, address coordination and compression functions, if any.
T-475
451
WINDOW
URGENT POINTER
PADDING
Unlike the IP case, TCP options (although provided for) are of no practical business interest.
The Source and Destination Port Number fields specify the upper level protocol (ULP) which is being served by TCP. Sample Port Number codes are presented in the box on the next page.
Sequence Numbers are used to indicate the progress of transmitting message segments, where the sequence number identifies the byte position of the first byte in a given segment. At the receiving site, logical consistency of incoming sequence numbers is used to verify that no segments have gone missing.
T-360
452
PORT CODE ASSIGNMENTS ARE SET BY THE INTERNET ASSIGNED NUMBERS AUTHORITY (IANA) - AND ARE CALLED "WELL KNOWN PORTS"
DESCRIPTION
REMOTE JOB ENTRY FILE TRANSFER PROTOCOL - DATA CONNECTION FILE TRANSFER PROTOCOL - CONTROL CONNECTION TELNET INTERACTIVE TERMINAL PROTOCOL SIMPLE MAIL TRANSFER PROTOCOL GOPHER PROTOCOL HTTP (WORLD-WIDE WEB PROTOCOL) Z39.50 (WIDE AREA INFORMATION SERVICE) PROTOCOL
The Window field defines an allowable number of bytes that the receiving TCP handler is willing to accept -- thus TCP asserts flow control in a very proactive way. No data are ever sent that are not previously agreed to.
The TCP Checksum field is similar in objectives and calculation method to that in the IP header. A major difference is the TCP header includes the segment payload in the calculation -- and regrettably, portions of the IP header (IP addresses, IP protocol code and Segment Length).
By involving the IP header in TCP activity, the desirable separation of layered system functions is lost, making it impossible to make changes to IP without simultaneously changing TCP.
T-361
453
TCP operates with a positive acknowledgment -- meaning that successful delivery of data is explicitly confirmed by the receiving station.
There is no response when delivery of a TCP segment fails, for whatever reason.
The sending TCP system is forced to make its own decision to retransmit a message segment if no positive acknowledgment is received in a reasonable time.
The complicating consideration is that the duration of an appropriate waiting interval is dependent on the type of network being served; LANs typically require waiting times in the order of 10's of milliseconds, WANs (especially satellite-served networks) typically require waiting times in the order of seconds.
The Karn Jacobson Algorithm specifies a moving average calculation, based on observed round-trip acknowledgement time delays, as a way of setting this retransmission timer.
Getting the process started with a new network connection typically involves some inefficiency in defining the retransmission time interval (with impaired response times being evident) -- but the algorithm very quickly converges to a viable time interval.
The net effect is that TCP/IP is uniquely capable of delivering effective, efficient results over a remarkable diversity of networking technology.
T-362
454
SCADAPROJECTS
PlanningandIdentificationofRequirements
PreparationofFunctionalSpecifications
continued ....
Data Communications & Fieldbus Systems
K-0
455
SCADAPROJECTS,cont.
PreparationofFunctionalSpecifications,cont.
Thisstageincludesdefinitionsofhowthenewprojectwillintegratewithexistingsystemsand providesaplanforoperationsandmaintenance(O&M)throughthelifecycleoftheproject.
DetailedDesign
Detailatthispointshouldbesufficientforimplementation/fabricationtoproceedeitherin-houseor contractedout.
Stakeholders,executivesanddesignersshouldagreeandsign-offonstrategicmilestonesinthe detaileddesignusingasystematic,periodicdesignreviewprocess.
continued ....
Data Communications & Fieldbus Systems
456
SCADAPROJECTS,cont.
Implementation
LargeSCADAprojectsaretypicallycontractedouttoaturn-keysystemsvendor,whowillbe responsiblefororchestratingresourcessufficienttomeetprojectschedulecommitments.
AcceptanceandProjectCompletion
K-0
457
PROJECTISSUESTHATAREUNIQUETOSCADA
SCADAprojectshaveuniquenessesthatrequirespecialattentionintheprocurementprocess:
TheOrganization'sVulnerabilitytoNon-Performance
SCADAsystemsalwaysservethecorebusinessoftheenterprise--itmanagesthe"goosethatlays thegoldeneggs",andisthereforetheepitomeofa"missioncritical"technicalsystem.
Failuretoperformwiththisserviceisnotanoption,soaprofessional,highlystructuredprocess mustbefollowedfromconceptthroughtofinalacceptance.
ValidationofPartsofTheSolutionisDifficult
continued ....
458
PROJECTISSUESTHATAREUNIQUETOSCADA,cont.
SCADASystemsHaveaLargeCustomComponent
NotwoSCADAsystemsarethesame--withtheconsequencethateverysystemisacustom design,tosomeextent.
Projectswithalargecustomcomponentmustbemanagedwithacarefullystructuredprogramthat measuresandvalidatesprogressthroughsometimes-elaboratetestprocedures.
SCADASystemsHaveaLongLifetime
Comparedwithothercomputer-based,electronicsystems,SCADAsystemstendtobeoperate throughamuchmoreextendedtimeframe.
459
ALTERNATIVESINPROCURINGSCADASOLUTIONS
Likewithanyprocurementprocess,thereareanumberofalternateapproachestoprocuring SCADAsolutions.
Considerationsthatwilldeterminetheapproachthatbestsuitsanyparticularapplicationinclude:
Theleveloffamiliarityandskillsavailablein-house--i.e.howcapableisthehostorganizationto doallorpartoftheworkthemselves?
Havemuchbetterfamiliaritywithcorporateneedsandcultureissues--anin-housedesigntends to"fit"better
Bemorelimitedintheirdesignfreedom--theydothingsthewaytheyhavealwaysdonethem-- forbetterorforworse.
460
IMPLEMENTATIONAND COMMISSIONINGISSUES
SCADAsystemsoperatewithlive,mission-criticalinfrastructure.Howdoyoutest"Emergency Shutdown"controlsequenceswithoutactuallydoinganemergencyshutdown?
461
PIONEER
PIONEERS GET BURIED AT THE SIDE OF THE ROAD
LATE ADOPTER
LAGGARD
BUY INTO MATURE TECHNOLOGY; THUS: PRODUCT COSTS ARE STABILIZED AND/OR REDUCED
OPPORTUNITY HERE IS TO LEAD WITH CAUTION BE A CAREFUL SHOPPER FOR MANY BUSINESS SITUATIONS, BEING AN EARLY FOLLOWER IS OPTIMAL STRATEGY (manageable mixture of risk vs reward)
PRESENTS EXCITING OPPORTUNITY TO BE A BREAKTHROUGH LEADER REQUIRES VERY CAREFUL, HANDS-ON MANAGEMENT APPROACH
TECHNOLOGY IS STANDARDIZED
THERE IS STILL A RISK OF ACTING BEFORE THE TECHNOLOGY STABILIZES -- BUT WAITING LONGER IS ALSO RISKY
RISKS OF FAILURE ARE MINIMIZED -- BUT RISK OF BEING LEFT BEHIND IS HIGH
COSTS ARE NOT MUCH MORE ATTRACTIVE THAN EARLIER OPTIONS BUT THERE ARE MAJOR BUSINESS PENALTIES FROM BEING NONPROGRESSIVE
OTHER COSTS NOW DEFINE THE BUSINESS CASE (Personnel efficiency, opportunity costs ... )
462
DECREASING COST, EFFORT AND RISK - ALSO DUE TO MATURITY OF THE TECHNOLOGY
The consequence of moving too late is principally lost opportunity (to deliver business effectiveness to the organization).
The cross-over point is sometimes called the 'ah-ha' point -- because of the inspiration that it is now time to move is not altogether obvious!
463
Appendices
Glossary of Terms Suggestions For Further Reading
464
GLOSSARY OF TERMINOLOGY
A-to-D Analog to digital conversion. The conversion process whereby analog signals are converted into a digital format, or encoded into a digital, or numeric representation. Inverse of D/A process. ACCESS METHOD In context of LAN systems, a means of managing access from individual stations to the shared cable medium. Examples of common LAN access control techniques are Token Passing and Carrier Sense Multiple Access with Collision Detection (CSMA/CD). ACK A positive acknowledgement of message (typically a block or frame message segment) receipt -- indicating acceptance of the message segment. An ACK is typically signaled with a special control character (with half-duplex link protocols) or a pre-defined bit-flag (with full-duplex link protocols). The alternative to an ACK is a NAK, or negative acknowledgement. ADAPTIVE DIFFERENTIAL PULSE CODE MODULATION (ADPCM) ADPCM is a voice signal encoding technique which delivers digitally coded toll-quality signals at 32 kbps. -- compared with 64 kbps. using conventional PCM. Several ADPCM standards exist, including the most widely supported ITU-T specification. ADAPTIVE ROUTING A network routing capability that selects the most appropriate route available at the moment of connection demand. Some data network adaptive routing procedures continuously evaluate this routing configuration, and redefines the route in mid-session. ADDRESS RESOLUTION The process whereby an internet (IP) address is interpreted (typically by a router/gateway device) and a physical address value deduced. See "address resolution protocol". ADDRESS RESOLUTION PROTOCOL (ARP) The ARP is a capability required on many TCP/IP systems which enable the gateway router to associate an incoming IP address with a local piece of hardware -- such as an Ethernet workstation. AMERICAN NATIONAL STANDARDS INSTITUTE (ANSI) ANSI is a New York, NY based organization that both defines standards applicable to USA interests, and represents the USA on a number of international forums, such as the ISO. ANSI defines standards over a wide range of technologies; in context of computer communications they are currently playing a major role is participating in the process of defining standards for high speed WAN and MAN technologies.
AMERICAN STANDARD CODE FOR INFORMATION INTERCHANGE The ASCII code is a 7-bit code used to represent 128 data and control characters. Extended ASCII is an 8-bit version of the code with 128 normal ASCII code values, and 128 additional codes accessed using the "Alt" keyboard key. Several international adaptations of ASCII have been developed to enable representation of "non-English" characters, such as accented vowels. The original ASCII is sometimes referred to as USASCII to differentiate it from these derivative versions. ASCII was defined by the American National Standards Institute, or ANSI.
AMPLITUDE MODULATION (AM) Amplitude modulation is one of three modulation techniques (See definition of "modulation"). With AM, the baseband signal information is encoded, or impressed on the carrier signal by modifying its amplitude to instantaneously respond to variations in the baseband signal.
ANALOG LOOPBACK A test procedure used to effectively short-circuit the remote end of an analog data line to verify the end-to-end connection. The actual loopback is not actually a wired short-circuit, but rather a logical echo of the outgoing test signal.
ANALOG SIGNAL An analog signal models or imitates the behavior of some (generally naturally occurring) process, such as human speech. The analog signal literally can be regarded as analogous to the process being modeled, and represents that process by means of a continuously varying physical quantity such as an electrical voltage. In such a a case, the voltage will change in the same proportions, and at the same rate as the speech itself. Thus the electrical signal is an analog of the naturally occurring, continuously variable (speech) process.
ANALOG TRANSMISSION Analog signals are continuously variable signals, and there is information conveyed in the continuum of signal values. Contrast with digital signals which demonstrate only signal values, and convey information only by their values at these discrete levels. Analog transmission facilities and equipment thus are responsible for faithfully reproducing the infinitely variable continuum of signal states. The complexity of analog signals makes it fundamentally impossible to develop transmission equipment that understands and interprets the signal values (as is done with digital systems), thus making it necessary to amplify and manipulate the signals with a minimum of distortion, but always (because equipment is imperfect) in an imperfect way.
465
APPLICATION PROGRAM Application programs in context of computer systems are the programs developed (often by the end-user organization) that deal with the organizations reason for owing the computer in the first place. Application programs include payroll, accounting, text processing, spread sheet systems. In contrast to application programs, supervisory programs are typically provided by the computer vendor, and include operating systems (such as MAC-OS and the various Windows variants). APPLICATION PROGRAM INTERFACE (API) The software interface between an application program and the computer and network operating systems which support it. The API effectively specifies the technical envelope within which a programmer must operate. On Unix-based systems, the API is referred to as a "socket". ARPANET ARPANET was the first operational packet switching network, from which evolved both X.25 packet switching standards (mid 1970s) and the TCP/IP internetworking protocols (early 1980s). ARPANETs research/development role has become largely unnecessary due to OSI standards development, and its operational role has been replaced by the DDN (Defense Data Network). ARQ Automatic Repeat reQuest. ARQ systems operate by providing for retransmission of data whenever errors are encountered (which would occur in response to a "NAK" message sent back to the source). Thus, with ARQ, the sending station remains accountable (and thus must maintain all data in a buffer), ready for a possible need to retransmit it, until the receiving station accepts responsibility with a positive "ACK" message. See "Stop-and-Wait ARQ", "Go-Back-N ARQ" and "Selective ARQ". ASCII See definition of "American Standard Code for Information Interchange". ASYNCHRONOUS TRANSFER MODE (ATM) ATM has emerged as the preferred switching technology for implementing cell relay fast packet switching. ATM is currently viewed as the most appropriate way of delivering functional, user controllable interconnection between high speed local access and wide area network facilities. ATM is scalable -operating at up to multi-Gigabit rates. ATMs main uniqueness is its flexibility -- it is capable of supporting the unique requirements of voice, video, graphical and conventional data applications. ATM is currently the favored switching technology with both wide area network planners (telephone companies), metropolitan area network designers and is feasible (but not popular) as a high speed local area network (HSLAN) technology -- a remarkable range of applications. ASYNCHRONOUS TRANSMISSION Asynchronous transmission utilizes a character-by-character data synchronization method in which each information character is individually handled, by the use of start and stop bits. The time interval between each character is not controlled. See definition of Synchronous transmission for a comparison. Asynchronous transmission is commonly called "start-stop" transmission.
ATTENUATION Attenuation is the loss or decrease in analog signal strength (as measured by voltage, current or power) caused by some component of a communications system -- typically the medium itself. Attenuation is usually measured and expressed in decibels.
AUDIO FREQUENCIES Audio frequencies are those components of speech, music, etc. (sound) that are audible to a person with normal hearing, typically taken to be in the range from about 30 to 20,000 hertz.
AUTHENTICATION Authentication deals with the question of confidence that a messages claimed originator was in fact the authentic originator of the message. Commonly, public key encryption technology is used to assure electronic message authentication. The problem is similar to verifying that a traditional documents signature is authentic -- and indeed, digital signatures are one tool used in network-oriented authentication procedures.
AUTOMATIC NUMBER IDENTIFICATION (ANI) The ability for the telephone number of the calling station to be made available to the called location. There are numerous applications of ANI; message accounting, integration with business computer applications and personal caller identification to name a few.
AUTOMATIC ROUTE SELECTION (ARS) ARS is a business telephone system feature used to select the most cost-effective outgoing service or route for each telephone connection.
BACKBONE NETWORK A high speed network system that links other typically lower speed networks together. The most common usage is associated with a connectivity solution for a number of local area networks, such as is encountered in a typical campus environment. Metropolitan area networks (MANs) may be configured as backbone networks.
BALANCED CIRCUIT A balanced communication circuit or interface is configured so that the two conductor required to support the electrical connection are identically configured. The two conductors (normally wires) will therefore be of the same type, have equal electrical characteristics and neither will be directly connected to ground or earth.
BAND A frequency band is a range of analog frequencies that are normally associated with the same application, such as the AM radio broadcast band, which occupies the range of frequencies from about 550 khz to 1600 khz.
466
BANDWIDTH The range of frequencies present in a signal (signal bandwidth), or the range of frequencies supported by a communication system (system or equipment bandwidth). Bandwidth is a measure of the frequency difference between the highest and lowest frequencies in this range. The usable telephone channel bandwidth is nominally 3 khz; the highest frequency on a standard telephone channel is approximately 3,300 hz; the lowest 300 hz. BASEBAND SIGNAL The native format of a signal as derived from the source apparatus, before any modulation or multiplexing procedures are effected. Thus, in telephony, the baseband signal is the speech signal itself; in data applications the baseband signal is the digital voltage square wave waveform. BAUD The baud rate of a signal is the rate at which recognizable changes of coded signal status (signal elements) are transmitted. In applications where one bit of information is encoded as one recognizable signal element, or symbol, the baud rate is the same numerical value as the bit rate. In most modems, data is encoded with multiple bits per signal element, and thus the bit rate is almost always greater than the baud rate. It is a widely mistaken assumption that bit rate and baud rate are the same; thus it is common to refer to the baud rate of a modem in an attempt to describe its data transmission capability. BINARY CODE A binary code is simply the representation of information in the simplest of units, being combinations of the binary pair of digits (the "1" and "0" values) . The executable form of a computer program is in binary code; effectively unintelligible to humans but necessary for computer execution. BIND Berkeley Internet Name Domain - a specific implementation of the Domain Name Server function on Berkeley Unix systems. BIT BIT is a contraction of "BInary digiT," the smallest unit of information in a binary system. A bit of information provides the ability to differentiate between a "1" or a "0" condition. BIT RATE The rate at which bits of information are transmitted, usually expressed in bits per second (bps), and multiples thereof: kbps, mbps, gbps, tbps, etc. Compare with the definition of baud. BLOCK One complete message segment, or data unit on a data communications link. A block is structured by pre-agreement to convey control information such as source/destination addresses, message number, retransmission requests ("NAK" codes), and error checking features. The message units known as blocks tend to be associated with half duplex systems; the comparable unit of information in full duplex systems is typically known as a frame.
BRIDGE A bridge is one of the simplest LAN connection devices based on examination of physical, or MAC-layer addresses only. Bridges do not examine network layer addresses,or make any changes to the contents of a message -- in contrast to the role of routers. Bridges must encounter the same technology (e.g. Ethernet) on all segments, and are ideally suited to a relatively small number of locally connected LAN segments.
BROADBAND (1) Communication facilities which have analog bandwidth values greater than that of a voice-grade telephone channels, and are thus able to support data transmission rates in excess of those feasible on telephone channels. (2) Generally high capacity data communications systems. An example is "Broadband ISDN", which is a fully digital (and thus has no literal "bandwidth" features at all) system offering very high data transmission rate capability.
BROADCAST A message delivery concept where all members of a community receive exactly the same message. Examples are television broadcasting and Ethernet LANs.
BROUTER A hybrid bridge-router. Features of brouters vary amongst vendors, but they typically offer routing services in some circumstances (e.g. where they recognize the network protocol in use) and act as bridges in others (e.g. where they dont recognize the network protocol in use).
BUFFER A buffer is a place to store information temporarily while some (usually) related activity can be performed. Buffers are used to save data while error checking and retransmission can be performed, they are necessary where interfaces occur between data transmission systems that operate at different data rates, or operate with different protocols. Buffers will generally introduce a time delay, and require careful management so as to avoid overflow (and consequent loss of data).
BUS A bus is a physical medium where there is no significance to the position or connection point of devices. A computer backplane is a bus (one can plug accessories and auxiliary equipment into any available "slot") and a coaxial cable ethernet LAN is a bus system (one can connect at most any point along the cable). Bus systems are always operated concurrently -- that is to say, signals are delivered to all points on the bus at effectively the same time.
BYTE In most contexts, a byte is an 8-bit unit of information, which is more exactly, and more technically referred to as an "octet". A byte typically represents one character of text information. Since computers process data in parallel format, the byte becomes the smallest unit of information in most data manipulation activities. Since some earlier computers used other than multiples of 8 bit words sizes, there have been situations where the term byte was used to refer to other than 8 bits of information. For example, the old DEC PDP-8 computer was a 12-bit memory system; the characters were
467
encoded into 6 bits of data -- thus a byte was 6 bits length. BYTE MULTIPLEXING Byte multiplexing, or byte interleaved multiplexing is a form of time-division multiplexing in which whole bytes of data are captured from each sub-channel, and sent as a unit. Bytes of data from different sub-channels follow each other in sequential time slots. Alternative mechanisms are bit interleaving, and message interleaving; byte interleaved multiplexing is the most common. CABLE A cable is an assembly of one or more conductors (or optical fibers) within a bundle (which typically includes an enveloping protective sheath). Cables are typically constructed so as to allow the use of conductors separately or in groups. The term is often used to suggest the coaxial cable that is the key element in cable television systems -- but is not an accurate generalization. CARRIER A carrier signal is a simple continuous frequency signal created for the purpose of being modulated by another information carrying signal. CARRIER FREQUENCY The carrier frequency is the operating frequency of the carrier signal used in the modulation process. In radio communication, the carrier frequency corresponds to the frequency to which one tunes the radio to receive a desired program or message. CATV Community Antenna Television, or CATV refers to the cable television service are offered in most communities. Unlike the over-the-air broadcast signals used in commercial television and radio, CATV signals are confined to a coaxial cable distribution networks. Additionally, microwave and satellite links, may be used as part of these distribution networks. CCIR See "ITU-R", which is the current identity for the former CCIR. CCITT See "ITU-T", which is the current identity for the former CCITT. CCTV Closed Circuit Television, or CCTV systems use coaxial cable or other "closed circuit" (ie. non-broadcast) media to carry television signals. Examples of CCTV applications include CATV (cable television entertainment services) and security video systems. CELL RELAY Cell relay is a packetized networking technology based on the use of small, fixed length packets called cells. The routing logic and packet control functions are implemented largely in hardware, making it possible to handle the call switching process at very high speeds -- much more quickly than software-driven technology (as is used in conventional X.25 packet switching and frame relay systems). Asynchronous Transfer Mode (ATM) switching implementations are the currently favored technology for delivering cell
relay services -- for both WAN, MAN and high speed LAN applications.
CENTRAL OFFICE The telephone central office, or exchange is the place where local access lines (local loops), and inter-exchange communications lines are terminated, and where switching equipment (which serves to interconnect these communication facilities) is located.
CENTREX A business telephone service where the features and functions of on-premises private branch exchange (PBX), or "switchboard" equipment are delivered from the telephone companys central office. Centrex tends to be favored by large, multi-office corporations and government departments.
CEPT Council of European Postal and Telecommunications administrations. CEPT is a body which assumes responsibility for the special interests of European telecommunications providers by making policy and standards recommendations for telecommunications practice within Europe. CEPTs role has been largely superseded by the European Telecommunications Standards Institute (ETSI), which includes in its membership the growing number of non-government European telecommunications providers.
CHANNEL A channel is a path for communication of information, which may be derived from a communication circuit or transmission facility. Multiplexing techniques make it possible for multiple channels to be created on a single communication facility such as a microwave radio system. A channel can be thought of as having the same relationship to the communication facility as a painted lane on a highway has to the highway itself. In some usages, the term channel implies a unidirectional communication capability.
CHARACTER Characters may be human-readable alphanumeric characters, such as the letters of the alphabet, numeric and punctuation symbols. In context of computer systems, the term "character" is often taken to mean the 7 or 8 bit ASCII (or other code) representations of data. These "characters" include the full set of alphanumeric characters described above, but also include a number of "control characters", which invoke system control functions. This latter category includes, for example, the Carriage Return Character (ASCII 0DH) and the Escape Character (ASCII 1BH).
CHROMINANCE The color part of a video signal, which pertains to the hue and saturation (but not the brightness) of the signal.
CIRCUIT A communication circuit is a physical telecommunication facility which provides a way of transmitting information between two or more points. An electronic circuit consists of an assembly of components connected together to form some signal processing or other electronic function.
468
CIRCUIT - FOUR WIRE A 4-wire communication path uses four wires (two pairs) to support fully separated information flow in each of two directions. That is, an independent, dedicated pair of wires is used for each direction. This term had its origins in an era when all communication facilities were derived from metallic wire; the term is now used to describe the functional equivalent of 4-wires, which might be actually supported by glass fiber optic cable, or perhaps microwave radio technology. CIRCUIT - TWO WIRE A 2-wire communication path uses two wires (ie. one pair) to support communication services. A 2-wire facility can support bidirectional communication if the cable is short enough so as to not require any signal amplification (which is an inherently one-way process) and does not include any multiplexing or other signal processing functions (which are also unidirectional activities). Otherwise, a 2-wire service will be unidirectional only. CIRCUIT SWITCHING Circuit switching systems provide an impermanent connection between two (or more) locations, and provide the system users with exclusive use of a continuously open channel, over which information may be exchanged. Circuit switching costs are proportional to holding time, and are unrelated to the volume of information transmitted. Compare with Packet Switching. CLIENT The software supporting user-oriented computing functions in a client/server computing environment. Client software typically (at least) concerns itself with managing the human interface; for this reason client systems are commonly associated with the graphical user interface (GUI). The term client is often (somewhat casually) associated with the hardware computing system which supports the client as defined here. CLIENT/SERVER PARADIGM The Client/Server computing model has been accepted as the de facto standard of the industry because it supports enriched human interfaces (e.g. the graphical user interface, or GUI), and is amenable to distributed processing (consistent with downsized hardware computer architectures), among other things. Client/server systems tend to be intensive users of network resources. CLOCK A precisely timed, repetitive digital pulse signal sequence which is used to control a data transmission and reception procedures. CLOCK RECOVERY The process of deriving a local clock signal at a receiving location by extracting the embedded timing information from the signal received on a synchronous channel.
COAXIAL CABLE A communications cable which consists of central conductor surrounded by an concentric, or co-axial outer conductor. The two conductors are separated from each other by a cylindrical shaped insulator. Coaxial cable is technically very much superior to (twisted) paired telephone cable.
CODEC A codec, or Coder-Decoder is a device that converts analog signals to digital, and digital signals to analog. Codecs are used to convert analog signals such as speech to a digital format as required for transmission over digital media -- and the companion function at the receiving end back to the original analog format.
COLOR BURST A few cycles (typically 8 to 10) of the 3.58 Mhz. color subcarrier which occur during the "back porch" interval in the luminance video signal. The color burst provides frequency and phase reference values for the receiver color oscillator.
COLOR SUBCARRIER In NTSC television, the 3.58 MHz signal which is modulated with color information, and superimposed (amplitude modulated) onto the luminance signal defines the color subcarrier. The amplitude of the modulated subcarrier is proportional to color saturation, and phase is proportional to hue.
COMMON CHANNEL INTEROFFICE SIGNALING (CCIS) CCIS is a telephone signaling technique in which call control signaling data is transmitted over a separate communication system from that used to support the message traffic. Typical CCIS systems use packet communication methods. Current interest focuses on Signaling System Number Seven (SS#7) -- which is a CCIS service being deployed with ISDN.
COMMON GATEWAY INTERFACE (CGI) The Common Gateway Interface, or CGI technology creates an interface between web servers and other information systems, enabling access to databases and legacy computing systems, for example. Without CGI (or a similar function), web servers are limited to presenting fixed format, prepackaged information. With CGI (or one of the proprietary extensions) it is possible to construct fully interactive client/server business applications.
COMMON MANAGEMENT INFORMATION PROTOCOL CMIP is the OSI standard network management protocol. It is less widely supported and more complex to implement than the TCP/IP equivalent, Simple Network Management Protocol.
COMMUNICATIONS SATELLITE Communications satellites are earth orbiting satellites designed to function as telecommunications microwave radio relay devices. Typical communications satellites are positioned in geosynchronous orbits, which are 22,300 miles (35,888 km) above the equator, and traveling eastwardly, so that they appear from earth to be stationary in space -- thus their orbital position is referred to as a "geostationary orbit".
469
COMPANDOR The word "compandor" is a hybrid term for a device that performs a compression function on outgoing signals (which reduces the dynamic range, or amplitude scale of an analog signal (in order that the full range of signal strength variation may be represented at a higher average volume level), and a companion expansion function performed on incoming signals (that have been compressed at the other end of the communication channel). The purpose of compandor operation is to reduce the vulnerability of weak signals to noise distortion on the communication link. The Dolby tape recording process is an example of a compandor system. COMPONENT DIGITAL SIGNAL A digital video format in which separate video signals are used to represent the luminance (Y), and color difference signals (R-Y and B-Y), or alternatively, in RGB format. COMPOSITE VIDEO A mixed signal comprised of the a luminance (monochrome) component and a chrominance (color) component, along with synchronizing signals (such as synchronization pulses, blanking pulses and color burst). Standard over-the-air television broadcast signals are composite. CONCENTRATOR A concentrator is a device which connects a number of circuits, or terminal connections to a single higher speed, shared communication circuit. Concentrators are functionally very similar to multiplexers -- except that concentrators are equipped without a companion concentrator at the host end of a link, since the host system itself performs the companion role. Since their operation is integrally linked to host system operation (contrast with multiplexer devices, which are transparent to host systems), they are generally proprietary products capable of working on only one equipment vendors networks. CONDITIONING Conditioning is an optional procedure applied to leased line analog communication circuits, with the intent to control the quality of transmission. Conditioning procedures require custom analysis and mitigation of system impairments on an end-to-end basis -- and are therefore not applicable to dial services. Conditioning is not a common service in current practice, since most leased line services are provided with digital technology. Where used, there are two types of conditioning: C and D conditioning. CONTROL CHARACTER Control characters are character codes (which along with alphanumeric characters comprise the character code sets like ASCII and EBCDIC) whose occurrence in a data sequence will result in the initiation, modification, or interruption of operational data flow. An example is the ASCII "Escape" character. CROSS TALK Cross talk is the unintended transfer of energy from one communications channel to another. It is most commonly encountered as the low volume occurrence of one voice telephone conversation on top of another -- caused by both connections sharing wires in a common cable bundle, and simultaneously,
CSMA/CA Carrier Sense Multiple Access with Collision Detection is the contention access protocol used with ethernet LAN systems. The "CSMA" part of the protocol provides for a station first listening for activity on the cable system before commencing transmission that would be destructive to the current use application. The "CD" function provides for continuing to listen for activity from other stations on the cable system after transmission is initiated -- to detect collisions which result from the transmission attempt. In the event of a collision, both affected stations will be required to back off and attempt the communication later.
CUG Closed User Group. A GUG is an administratively defined virtual network service, such that members of a GUG are able to communicate freely amongst themselves, but are isolated from the larger public community that shares the common communications infrastructure.
CUSTOMER PREMISES EQUIPMENT (CPE) CPE is telecommunications terminal equipment connected to the end of a telecommunications network termination link. The common telephone is CPE -on a customers premises, and connected to the public telephone network. A well defined demarcation point will establish the boundary of responsibility between CPE and the network service provider. In the U.S.A., clear policies have been defined prohibiting ownership of CPE by the telecommunications network service company.
D-to-A Digital to analog conversation. The conversion process whereby digital signals are converted into an analog format, or decoded from a digital, or numeric representation. The inverse of the A-to-D process.
DATA Data is the encoded representation of information, which may be in the form of numbers, character-oriented texts, tabulated facts, instructions, executable programs, etc. Data is a plural term (datum is the formal singular term) -however common usage ignores this distinction, using the "data" term for both singular and plural.
DATA COMMUNICATIONS EQUIPMENT (DCE) In the context of a data communications, it is the equipment which is interposed between the Data Terminal Equipment (DTE) and the communication circuit. The purpose of the DCE is to translate signals presented by the DTE into a format suitable for passing across the communication link. Thus, the DCE must be selected to operate within the constraints of the medium (based on analog/digital and 2-wire/4-wire options, for example).
DATA DICTIONARY A document or database file that specifies the precise content characteristics of data elements, which are in turn are individual items of information, like perhaps a telephone number.
470
DATA ENCRYPTION STANDARD (DES) The public domain data encryption methodology developed by the US National Institute of Standards and Technology (NIST, formerly the National Bureau of Standards). DES systems all operate using the same 56 bit key encryption/decryption algorithm -- and maintain the security of data through a private key code known only to the two communicating entities. DATA SET The data set is a largely disused telephone industry term for the analog modem. This usage continues with respect to circuit designations on the EIA-232-C interface standard. DATA TERMINAL EQUIPMENT (DTE) In the context of a data communications, it is the data processing equipment which is attached to, and terminates the network connection, and which is the entity being served by the communication facility and the associated data communications equipment, or DCE. DATAGRAM The term for the message unit which is supported by the Internet Protocol, or IP in TCP/IP networks. Although technically distinct, a datagram is somewhat like a packet. (The term datagram implies an uncoordinated, unannounced communication; packet implies a pre-established logical connection prior to sending data.) DBS (DIRECT BROADCAST SATELLITE) Direct broadcast satellites a geosynchronous orbit satellites designed to deliver a sufficiently high powered signal so as to be recovered by a small receiving dish antenna. By radically reducing the cost of the receiving earth terminal in this manner, DBS technology is competitive alternative for delivery of entertainment broadcast services. DCE See definition of "Data Communication Equipment". DECIBEL The decibel is formally the tenth part of a "Bel" -- named in honor of Alexander Graham Bell. In practice, the "Bel" unit is never used. The decibel is a unit (abbreviated "dB") for measuring and describing the relative strength of two signals -- and is typically based on units of power or voltage. The decibel is defined as ten times the logarithm (to the base 10) of the ratio of the power of two signals. Common practice often substitutes one of these two signals with an arbitrary reference level, making the decibel into an absolute measure of signal strength. Where so used, the reference level must always be indicated, such as a 1 watt power reference -in which case the unit would become "dBW". DEMODULATION Demodulation is the process of recovering the encoded information or data from a received, modulated carrier wave. Demodulation is the inverse of the inverse of modulation.
DIGITAL LOOPBACK Digital loopack capability is provided by data communications equipment (DCE) -- such as modems and DSUs -- which involves taking an incoming data signal and sending it back out over the return data path to the origination location. The purpose is to test a communications circuit, including DCE. See "Analog Loopback".
DIGITAL MODULATION Digital modulation involves a process of encoding (and subsequent demodulation, or decoding) of binary digital data onto an analog carrier of some type. Whether using an audio or RF carrier, the device that performs this function will be a modem.
DISTRIBUTED COMPUTING ENVIRONMENT (DCE) The DCE is an open standards-oriented collection of software technologies (primarily related to data base management and human interface aspects of client/server systems) supported by the Open Systems Foundation, or OSF. A key feature of the DCE is that it is intended to enable functional client/server interoperability between heterogeneous hardware and software elements of a computing environment.
DISTRIBUTED MANAGEMENT ENVIRONMENT (DME) DME is a collection of network management specifications and standards which act as a kind of management umbrella to the DCE, or Distributed Computing Environment -- both of which are defined and supported by the Open Software Foundation, or OSF.
DOMAIN NAME SYSTEM (DNS) A human-friendly site naming system used on the Internet to avoid the need to remember numeric address values. The DNS uses a syntax where strings of characters are separated by decimal points or periods, as in "ee.mit.edu".
DOTTED DECIMAL ADDRESS A numeric TCP/IP (Internet) address which uses the syntax of breaking the 32 bit binary IP address into four bytes, representing each byte as a decimal value, and separating each byte segment with a decimal point. A hypothetical example could be "128.34.222.197".
DPCM Differential pulse code modulation, or DPCM is a variant of pulse code modulation (PCM) in which successive signal sample values are encoded on the basis of differences between these sample values -- instead of their absolute value, as is done with simple PCM. See definition of "PCM".
DS-0 Individual voice channels are encoded using industry standard pulse code modulation (PCM), with a resulting 64 kbps. data sequence, and are known as DS-0 channels. This 64 kbps. data stream doesnt necessarily carry encoded voice information, but is regarded as voice channel-equivalent. North American telephone networks aggregate 24 DS-0 channels into one DS-1 (or T-1) service at 1,544,000 bps. using time division multiplex techniques. ITU-T standards used in Europe aggregate 30 DS-0 channels into what is widely known as "E-1", or First Order Multiplex channels at 2,048,000 bps.
471
DS-1 Digital Signal - 1 is a generic term for the "T-1" high speed data communications service available commercially in North America and Japan. It is a 1,544,000 bps point-to-point data service. The "T-1" designation has its origins as a vendor-specific (AT&T) product name -- the DS-1 equivalent is seen to be a more generic reference. Similarly, DS-3 and T-3 are equivalent. The ITU-T equivalent to DS-1 used in Europe is properly referred to as "First Order Multiplex", but is very widely referred to as "E-1", parallel to the "T-1" designation. DSAP Destination Service Access Point -- the upper layer protocol specification for the network layer protocol which is to receive incoming packets. DSI Digital speech interpolation is a digital implementation of time assigned speech interpolation (TASI). See definition of "TASI". DTE See definition of "Data Terminal Equipment". DUPLEX TRANSMISSION Two-way transmission capability or process. Usage varies; to be unambiguous, one should specify full duplex (simultaneous bidirectional communication), or half duplex (sequential bidirectional communication) in both directions. Less formal usage may associate full duplex with "duplex", and differentiate only half duplex configurations. EBCDIC Extended Binary Coded Decimal Interchange Code: an IBM proprietary 8-bit alphanumeric information code. EBCDIC is comparable to, but incompatible with ASCII code. ECHO SUPPRESSOR Used to control the subjective impairment caused by echoes on telephone communication by reducing the amplitude of echoed signals. The echo suppressor senses which speech path (in a 4-wire, or dual speech path configuration) is active at any particular moment and functions by reducing the amplitude in the direction of communication opposite to that with the observed speech activity. Since human speech is fundamentally a half-duplex process, and because the echo suppressor is capable of quickly sensing normal reversals in direction of speech activity, this reduction in signal amplitude will only affect the unwanted echo signal. Echo suppressors are of limited effectiveness where long time delay (such as encountered on satellite-derived communication channels, and long international terrestrial channels) impairs switching responsiveness. Suppressor switching times of up to 100 milliseconds will (along with other factors) limit line turn around times on half-duplex dial telephone data connections. EDI See definition of "Electronic Data Interchange".
EFT Electronic Funds Transfer involves the electronic communication of financial assets, either between trading partners directly, or more commonly, between one organization and a bank.
EIA INTERFACE The Electronic Industries Association has specified the physical, electrical, functional and procedural aspects of interconnection of equipment (DTE and DCE) in internationally recognized standards. The most commonly encountered EIA interface standard is the EIA-232-C, formerly known as the RS-232-C.
ELECTROMAGNETIC INTERFERENCE (EMI) EMI is interference in the form of electromagnetic energy radiated by electrical power systems (such as power transmission lines, power transformers, industrial motors), natural phenomena (such as lightning strikes) and electronic apparatus (such as radio transmitters, radar equipment, cathode ray tube display devices). Vulnerability is very much affected by the kind of communications cable systems used (unshielded twisted pair cable, or UTP, shielded twisted pair cable, or STP, coaxial cable or fiber optic media) , and in extreme cases, buy shielding measures taken to protect terminal equipment as well.
ELECTRONIC DATA INTERCHANGE The computer-to-computer exchange of standard business documentation in machine processable form. EDI is/was an early form of e-business and has largely been overtaken by this class of solution.
ELECTRONIC MAIL The electronic communication of generally free-formatted information over a computer network. The largest e-mail system in the world is supported by the global Internet, comprising hundreds of millions of users world-wide.
ELECTRONIC MAILBOX A computer file system managed by a network provider where incoming e-mail for a given recipient is lodged, pending their convenience in taking delivery -very analogous to a physical postal mail box.
ENCRYPTION Encryption technology provides for systematically and thoroughly encoding a communicated message so that any unintended recipient will be functionally incapable of correctly interpreting its content. There are several widely used encryption systems, many of which are developed in the U.S.A. and are restricted to use in that country. Military use of encryption is generally regarded to be considerably more sophisticated and effective than the systems available for civilian applications.
ERROR CONTROL Error control is a term used to describe any of a number of techniques for controlling errors on a data communication link. Error control always requires some capability of detecting the presence of errors. Upon detection, there are basically two procedural options: (1) invoke a
472
retransmission request (See definition of "ARQ"), or (2) perform additional calculations on the composite message received, and make a determination of error location(s) without further retransmission (See definition of "FEC"). ERROR CORRECTING CODE Data protected by error correcting codes is transmitted with additional bits of data -- error correcting code fields. Error correcting codes define these additional data elements in such a way that it is possible to validate received data by checking the error correcting code bits, since the sending and receiving stations both follow the same identical error checking calculation procedure. ERROR DETECTING CODE A means for encoding data signals, such that they conform to specific rules of construction -- so that departures from those rules at the receiving location can be automatically detected. Such codes require additional, or redundant data to be transmitted than are necessary to merely convey the desired information. ERROR RATE Data error rates are calculated and expressed as the ratio of errored data bits to the total number transmitted. Error rates are typically expressed in terms of numbers like "1 in 1,000,000) -- meaning that for every million data bits transmitted, one bit is statistically likely to be errored, or incorrect. ETHERNET Ethernet LAN technology was developed initially by Xerox, who subsequently partnered with DEC and Intel to further its market acceptance. The IEEE 802.3 standard version is slightly different from ethernet. Ethernet is a concurrent (messages are delivered to all stations at one time), bus oriented contention LAN technology. The original ethernet was a 1 mbps. design; all currently supported versions are 10 mbps. (with the exception of developing 100 mbps. "pseudo-ethernet" versions. Early implementations were all based on "Fat Ethernet" (10-Base-5) cable designs; development of "Thin Ethernet" (10-Base-2) and "10-Base-T" telephone cable versions have changed the cable environment of typical installations. ETSI European Telecommunications Standards Institute, an independent and broadly based standards organization which has assumed the carrier coordination role previously held by CEPT. EVEN PARITY CHECK (ODD PARITY CHECK) Even and Odd parity checking are examples of error detection codes which are used to validate data at a receiving location. Parity provides for constructing data sequences (typically character-sized segments) in such a way that there will be a predictably even (or odd) number of "1" bits in the sequence. EXCHANGE, PRIVATE BRANCH (PBX) A private business telephone switching system (located on users premises) which is connected to the public telephone network via trunk circuits. In time past it was operated by a switchboard attendant. More recent usage of the term PBX presupposes that the equipment will be fully automatic. In an
era when this automatic role could not be assumed, the term "PABX", with the "A" used to indicate "automatic". The acronym "PBX" is the most commonly used form in current practice.
EXTENDED SUPERFRAME (ESF) ESF is a DS-1/T-1 enhancement which makes all 192 data bits in each of the 24 repeating frame patterns available for customer data. With ESF in place, each DS-0 subchannel on the T-1 or DS-1 service can deliver a "clear channel" 64 kbps. connection. This is in contrast to the precursor version of T-1 known as T-1/D-3. With the T-1/D-3 configuration, each DS-0 subchannel is required to sacrifice occasional "robbed signaling bits", limiting effective clear channel use to 56 kbps. ESF is an enhancement to T-1 network systems to make them capable of supporting ISDN services.
FACSIMILE Facsimile is a system for transmitting information in graphical form, typically over the public switched telephone network. The image is raster scanned at medium resolution, resulting scanning information communicated by data modem, and reconstructed by the receiving equipment -- often on thermally sensitive paper.
FAST PACKET SWITCHING Fast packet is a general term used to describe packet communications technology based on a more dynamic packet switching methodology than that permitted by the X.25 international standard. The two principal technologies included in this category are "Frame Relay" and "Cell Relay". Implementations of cell relay are commonly associated with Asynchronous Transfer Mode (ATM) -- a currently important technical development.
FIBER DISTRIBUTED DATA INTERFACE (FDDI) One of two leading high speed standard networking technologies (Alternative is DQDB, see definition of "Distributed Queue, Dual Bus") suitable for either MAN applications (shared by a number of users, and physically reaching a public metropolitan area user community) or HSLAN (high speed LAN applications), where the physical scale of the network is usually limited to a single, contiguous piece of real estate (eg. a university campus or military base), and provides a backbone service to typically a large number of lower capacity LAN systems. FDDI operates at 100 Mbps as a dual token ring system with a total circumference of up 200km. Extensions to FDDI include CDDI, and FDDI-II which extends the data rate capability into the Gigabit per second range. FDDI enjoys the benefit of being an ANSI standard technology, but is not included in the almost universally embraced IEEE/ISO LAN standards -- thus its long term future compared with other high-speed LAN/MAN technologies is limited.
473
FIBER OPTIC SYSTEMS Fiber optic systems communicate with (generally infrared) light pulses transmitted through extremely thin, highly pure glass fibers that trap, or contain the light energy so that it follows the cable around bends and emerges from the far end with sufficient energy as to drive an opto-electric detector device. Short distance applications sometimes us a plastic "fiber" optic cable medium. In addition their incredible distance and data rate capabilities (commercially available systems operate over many kilometers without repeaters at data rates of many Gbps) -- there is considerable benefit in using fiber optic technology where immunity to electromagnetic interference (EMI), where industrial hazard voltages are present and/or where security issues are of concern. FIELD One half of an interlaced television picture, or frame. In 525 line television frame systems (North American NTSC), the field consists of 262.5 lines and is repeated at a 59.94 Hz. rate. FILE TRANSFER PROTOCOL (FTP) The original, and most commonly used means of moving files over TCP/IP networks (and therefore the Internet). See anonymous FTP. FINGER Finger is a client/server configured application process for TCP/IP internetworked systems that make it possible to enquire as to the availability of remote host systems and/or individuals. By invoking the "finger" command, a message is send over the TCP/IP network to the specified host system, which returns statistical data regarding current usage or personal information of individual users. FLOW CONTROL In data communication systems or networks which make use of buffer storage, the possibility of congestion preventing normal data flows would, if ignored, cause a risk of buffer overflow. Thus, information systems connected to such systems must be capable of responding to signals requiring temporary (typically momentary) cessation of transmission. There are two types of flow control: (1) control via the interface ready signals (typically DSR), and (2) through the use of flow control characters, such as the ASCII XOFF and XON control characters. FORWARD ERROR CORRECTION FEC systems operate by providing sufficient redundant information accompanying each original message transmission that it is possible to not only identify the presence of errors on data, but to also locate them -- thus enabling a correction decision to be made without any reverse transmissions. Thus the name. Since FEC systems operate without any feedback to the point of origin, there is no definitive custody transfer action, as there is with ARQ (See definition of "ARQ") systems. Consequently, FEC systems for business applications are typically configured with an accompanying backstop ARQ capability -- or more correctly, in a hybrid ARQ/FEC configuration.
FRAME The complete representation of a single video image, which is reproduced at the field refresh rate (29.97 Hz. in NTSC equipment).
FRAME One complete message segment, or data unit on a data communications link. A frame is structured by preagreement to convey control information such as source/destination addresses, message number, retransmission requests ("NAK" codes), and error checking features. The message units known as frames tend to be associated with full duplex systems; the comparable unit of information in half duplex systems is typically known as a block.
FRAME RELAY Frame relay is a simplified, expedited equivalent of conventional X.25 packet network switching technology -- such that the delivered service has a higher throughput capability -- and is thus more suitable to LAN connectivity applications. Frame relays greatest simplification (compared to X.25 packet switching) is that it dispenses with link-by-link data validation, making it possible to commence forwarding a packet onto its outbound route before the trailing portion has arrived (in contrast to the store-validate-forward process inherent in packet switching). Thus the term "relay" rather than "switching".
FREQUENCY The rate at which signal values alternate. Thus, frequency is the measure of how frequently analog signal values repeat their cyclic pattern of amplitude variation. Frequency was once measured in terms of "cycles per second", which was a useful, descriptive term. Current practice describes frequency in the Hertz unit. Variations are the Kilohertz, Megahertz, Gigahertz, etc.
FREQUENCY DIVISION MULTIPLE ACCESS (FDMA) FDMA is a technique for frequency, or bandwidth sharing an analog common communication channel -- typically a geographically dispersed channel, such as a satellite service. Devices at different geographical locations can feasibly share such a channel if they are disciplined to coordinate their individual transmissions so as to not transmit in exactly the same frequency bandwidth segment as any other user. FDMA is in some ways like a distributed frequency division multiplexer system. Compare with TDMA.
FREQUENCY DIVISION MULTIPLEX (FDM) Frequency-division multiplexer or frequency division multiplexing. FDM is a process by which different analog signal sources can co-exist on a communication circuit or facility by allocating each signal a unique fraction of the composite system bandwidth.
FREQUENCY MODULATION (FM) Frequency modulation is one of three modulation techniques (See definition of "modulation"). With FM, the baseband signal information is encoded, or impressed on the carrier signal by modifying its frequency to instantaneously respond to variations in the baseband signal.
474
FREQUENTLY ASKED QUESTIONS (FAQ) Frequently Asked Question -- or FAQ -- documents are commonly created and distributed on the Internet pertaining to thousands of different topics. FAQ documents are available dealing with most of the topics of the USENET Newsgroups, and many other technical and non-technical topics of interest to the Internet community. FAQ documents are typically organized in question and answer format, and almost always are designed to give the reader an entry level exposure to the title subject. A comprehensive FAQ archive site is maintained at MIT, at an anonymous FTP server designated as "rtfm.mit.edu" in the "pub" directory. FULL DUPLEX (FDX) Full Duplex refers to a data flow sequences, or communications systems which operate (or permit operation) in both directions at the same time. Compare with "Half Duplex". GO-BACK-N ARQ Go-Back-N ARQ systems provide for a sequence numbered message stream (such as supported by HDLC) to identify a message segment, or frame that is found to be flawed by transmission errors, and provide for retransmission of the flawed or errored frame, and all message frames that have been transmitted since the errored frame. Thus, the sending station simply "backs up" to the point where the error occurred and starts over. Thus it would be more literal to refer to the procedure as "Go Back To N", rather than "Go Back N". Contrast with "Selective ARQ", which provides for only retransmitting the errored frame. See definition of "ARQ". GOPHER Largely obsolete: Gopher is a menu-based hierarchical information search and recovery tool widely used on the Internet. It was developed at the University of Minnesota, where many gopher search sequences still originate. Gopher is a client/server structured environment that operates with either a text or graphical presentation. GROUND RETURN CIRCUIT A ground return circuit is an electrical connection connecting two points, which is generally a metallic connection (although it may be an earth conduction path) that (1) provides an electrically common, neutral connection medium, and (2) creates an electrical return path for signals passed between two systems -- such as with the pin no. 7 connection on the EIA-232-C interface. HALF DUPLEX (HDX) Half duplex systems (and circuits) demonstrate bi-directional communication capability -- but (due to communication equipment or protocol limitations) are only able to communicate in one direction at a time. They are thus sequential bi-directional systems or circuits. Human speech is an example of a half duplex process. HAMMING CODE The Hamming code is an early, simple, and relatively ineffective forward error correcting code that uses a unique parity calculation to detect and isolate transmission errors in data.
HDTV High Definition Television; a developmental digital television broadcasting technology designed to reproduce many more lines of resolution than are available with existing analog television broadcasting. The U.S. definition is based on 1080 lines, 1920 horizontal samples (pixels) per line distributed as an MPEG-2 compressed digital signal.
HERTZ The "Hertz" is the universally used unit of frequency, equivalent to the now disused "cycles per second" unit. Larger units include kilohertz (abbreviated khz, and meaning thousands of hertz), megahertz (abbreviated Mhz, and meaning millions of hertz) and gigahertz (abbreviated Ghz, and meaning billions of hertz).
HIGH LEVEL DATA LINK CONTROL (HDLC) HDLC is the most widely used synchronous, serial bit oriented link protocol. HDLC is an ISO standard patterned after IBMs proprietary Synchronous Data Link Protocol, or SDLC. The U.S. ANSI equivalent of HDLC is called ADCCP Advanced Data Communications Control Procedure.
HIGH SPEED LAN (HSLAN) The HSLAN term is used to describe the application of high speed technology to a (campus) backbone service. HSLAN and MAN systems share the same technology -- see definition of "Metropolitan Area Network" for discussion of technology alternatives.
HOME PAGE A home page is a starting point for exploring a World Wide Web site. The default file name for the home page is index.htm.
HORIZONTAL PARITY ERROR DETECTION Horizontal parity calculations are based on defining a block check character (BCC), the individual bits of which are defined on the basis of maintaining an even (or odd) number of "one" bits throughout all of the bits in that position in all characters of a data block. Thus, for example, the third bit of the BCC will be defined so that it, along with all of the data bits in the third bit position in all characters in the block will collectively demonstrate even (or odd) parity.
HOT STANDBY Hot standby equipment is equipment redundantly provisioned so that it is powered up and instantly ready to assume an operational role -- in the event that an operating companion unit should fail. Hot standby equipment configurations are generally simpler to implement than full concurrent redundant operations, in which normal system activity is shared by two productive devices -- and in which the operational role is assumed by the survivor when one fails. The idle equipment in hot standby configurations is never productive with respect to the role for which it is standing by -although it may perform some low priority tasks which are subject to being preempted.
475
HUB A hub is a device which is located at some central point in a network, and provides a means of terminating a number of incoming connections. In LAN systems, the hub is typically located in a telephone wiring closet, and is equipped with a direct wire connection to each device on the LAN. Hubs are available in varying degrees of sophistication, some capable of supporting a number of different LAN technologies and centralized network management systems. HYPERTEXT Hypertext documents are specially formatted text which has embedded codes that highlight portions of the text, and provide a logical linkage to other files or information resources. By viewing hypertext documents with appropriate software, it is possible to point to a piece of highlighted Hypertext text, click the mouse, and establish the linkage to the embedded reference information. Thus, hypertext makes it possible to move between documents by keying on internal content of these documents, rather than using an external menu system. HYPERTEXT MARKUP LANGUAGE (HTML) HTML is an industry standard syntax for marking up, or annotating text so as to make it into Hypertext. See definition of Hypertext. Extensions/alternatives to HTML include SGML (Standard General Markup Language) , XML (Extensible Markup Language) and WML (Wireless Markup Language). HYPERTEXT TRANSFER PROTOCOL (HTTP) HTTP is the specified protocol for exchanging hypertext encoded documents (and graphics, audio and video information) on the World Wide Web. IEEE See definition of Institute of Electrical and Electronics Engineers. IEEE 802.3 IEEE 802.3 is the most widely used LAN technology, being the formalized, standard derived from the original ethernet. Ethernet as defined by the Xerox/DEC/Intel consortium is subtly different from the derived IEEE 802.3 version. See "Ethernet" definition. IEEE 802.4 The IEEE 802.4 is a specification for a token bus LAN technology developed at General Motors as a delivery system for industrial automation applications. The system characteristics are not easily adapted to typical office applications -- making the IEEE 802.4 largely irrelevant to non-industrial uses. IEEE 802.5 See definition of "Token Ring" LAN technology. IEEE 802.6 The IEEE standard specification for Metropolitan Area Networks (MAN) technology. The IEEE 802.6 standard is based on the Distributed Queue Dual Bus (DQDB) MAN technology. DQDB is essentially a dead technology.
IN-BAND SIGNALING In-band signaling is an analog telephone network signaling scheme whereby essential network signaling information is conveyed by means of tone coded signals that have frequency values such that they can be carried within the analog bandwidth normally used for voice transmission. (Compare with definition of Common Channel Interoffice Signaling).
INSTITUTE OF ELECTRICAL AND ELECTRONICS ENGINEERS (IEEE) The IEEE is the worlds largest professional organization, which over its history has established a reputation for publishing a remarkable array of highly regarded journals and other periodicals that define the state-of-the art in most dimensions of electrical and electronics technology. In recent time they have become involved in the computer communications standards setting process -- most notably with respect to the 802-series of LAN standards.
INTEGRATED SERVICES DIGITAL NETWORK (ISDN) ISDN is a standards-based technology keyed to the delivery of digital-format signals over public access networks. This is in contrast to existing telephone systems which are based on analog signals and an analog interface at the demarcation point between telephone company and customer premises equipment (CPE).
INTERFACE An interface is the formal point of interconnection -- or boundary -- between two pieces of equipment. The functions of each signals which are exchanged, their electrical characteristics and the physical features of connectors are generally the key components of interface standards. The standard EIA-232-C interface is an example.
INTERNATIONAL ORGANIZATION FOR STANDARDIZATION (ISO) The ISO (often incorrectly referred to as the International Standards Organization) is an international agency that develops and endorses standards -- for a wide variety of technologies. The ISO had a major role in creating a unified world view of information technology through the Open Systems Interconnection Model (OSI) -- and related protocols.
INTERNATIONAL TELECOMMUNICATIONS UNION (ITU) The ITU is an agency of the United Nations, which was established to provide global standardization of communications procedures and practices -including coordination of telephony practices and radio technology usage. Key operating branches of the ITU are the ITU-T and the ITU-R. (Refer to these individual definitions).
INTERNET The term Internet has two key interpretations: (1) When used with an upper case "I", Internet refers to the world-wide community of interconnected TCP/IP networks -- all of which necessarily support the Internet Protocol, or "IP". (2) When used with a lower case "I", internet refers to the generic interconnection of networks -- more properly referred to as an "internetwork" to avoid any confusion with (1).
476
INTERNET ENGINEERING TASK FORCE (IETF) The IETF is a largely volunteer group responsible for guiding the development of TCP/IP and Internet technical standards and practices. INTERNET PROTOCOL (IP) IP is the basic data delivery service and Network Layer protocol for TCP/IP networks. IP-level addressing and routing logic makes it possible to deliver message traffic (datagrams) through large, complex networks like the Internet. INTERNET SERVICE PROVIDER (ISP) A commercial enterprise that provides Internet access in typically a limited geographic service area -- for a modest fee. Some ISPs offer national service -- and a wide range of quality, competency, features and cost of service is evident. INTRANET There are differing views of just what an Intranet is. All would agree that it involves adopting the internet computing architecture and technology for company-private use. Some define it is an infrastructure, and thus is synonymous with a private TCP/IP network environment. Others argue that it is synonymous with web technology, and thus describes a private web service -- an application definition. In practice it is both; it would be unthinkable to create a private intranet without using both TCP/IP and web technology. IP SWITCHING There are two main categories of LAN switching technology deployed in support of Virtual LAN service -- switches based on adapting bridging technology (Layer 2 Switches) and switches based on adapting routing technology (Layer 3 switches). Of the Layer 3 switch alternatives, IP Switching is by far the most popular. It takes advantage of the near-universal use of Internet Protocol in business networks -- and makes switching decisions based on IP network and/or sub-network addresses. See "Tag Switching" for a contrasting Layer 2 switch discussion. ISO See definition of "International Organization for Standardization". ITU-R The International Telecommunications Union - Radio Committee (formerly the Consultative Committee on International Radio, or CCIR) is responsible for coordinating the usage of radio communication technology around the world. This includes communication satellite services, microwave radio services and broadcasting services. The ITU-R is one of the two key operating branches of the International Telecommunications Union (ITU) -- the other main branch is the ITU-T. ITU-T The International Telecommunications Union - Telephony Committee (formerly the Consultative Committee on International Telephony and Telegraphy or CCITT) is responsible for coordinating the development of technology and standards in matters relating to telephony and telegraphy -- from which data communications technology has evolved. The ITU-T/CCITT has historically been most influential in Europe, although their role is becoming
increasingly accepted globally. The ITU-T is one of the two key operating branches of the International Telecommunications Union (ITU) -- the other main branch is the ITU-R.
JAVA Java is a Sun MicroSystems development that adds executable software to the information resources captured in a web transaction. Thus, in addition to acquiring text and graphics information when web browsing, Java enables access to executable software which can animate a web presentation or be customized to suit specific business applications, such as data entry. One strong benefit of Java is its platform-independence; it runs without modification on any recent web browser on any machine. LAN See definition of "Local Area Network".
LEASED CIRCUIT Telephone leased (or "dedicated", "private" or "full period") lines are telephone company provisioned on a 24 hour a day, 7 days a week basis. Leased lines bypass all switching equipment. Leased lines are unsuitable for occasional use (too costly), or for applications requiring variable connection points (not conveniently reconfigurable). They are suitable for high capacity, continuous applications -- like perhaps headquarters to branch office data connections.
LOADING COIL Loading coils are coils of wire (inductance) added in series with long metallic local loop telephone cables for the purpose of minimizing amplitude distortion. Loading coils embedded in the telephone cable plant complicate deployment of high capacity digital technologies like ADSL.
LOCAL ACCESS AND TRANSPORT AREA (LATA) Specific to U.S.A. regulatory policy, the LATAs are the geographic areas within which the regulated telephone companies operate. Competitive long-distance carriers are not allowed to offer service between points within one LATA; only between LATAs. The physical size of LATAs vary from the scope of a single city (e.g. Washington D.C.) in densely populated areas of the country to entire states (e.g. North Dakota and New Mexico).
LOCAL AREA NETWORK A LAN is a somewhat loosely defined network which covers a relatively small geographic area, such as a single office building, an industrial campus, a university campus, or a military base. LAN technology typically supports high data rates (many Mbps), relatively low error rates, and all resources are owned and administered by the user organization.
LOCAL LOOP The telephone network connection to individual customer premises is called the local loop. It is a copper twisted pair cable connection, the length of which is determined by the distance separating customer premises from telephone exchange sites. The universality of the copper wire local loop
477
infrastructure is being challenged by wireless and cable television-based alternative technologies. LOOP The term "loop" is an alternate term for "local loop". See definition of "local loop". LOW EARTH ORBIT SATELLITES Low earth orbit satellite technology differs from classical geosynchronous satellite communications systems in the literally low altitude orbit of operation. Low orbit implies a satellite that appears to move across the sky -- thus requiring (1) a number of satellites serving the same application, and (2) non-directional antenna systems (almost always). LEO technology is the basis of a number of mobile telephone and data access networks, such as the financially challenged Iridium project. LUMINANCE The monochrome, or brightness part of a video signal. Whenever a monochrome video signal is encountered, it consists entirely of luminance information. MAIL GATEWAY A host computer system that is connected to two or more different networks, and serves to pass e-mail between them. Because there is a difference in technology supported or a need for security partitioning between the mail systems being gatewayed (otherwise a gateway would not be required), there are protocol conversion and/or proxy functions required at a gateway. MAILING LIST A distribution list to which bulk e-mail may be delivered. Mailing lists are used as a way of supporting listserv newsgroups, or discussion groups. MAIN DISTRIBUTION FRAME (MDF) The MDF is a rack system in telephone exchanges where all incoming cable connections are physically terminated. From the MDF, connections are made to telephone switching apparatus, or other cable connection termination point by means of cable "jumpers". MAN See definition of "Metropolitan Area Network". MARK SIGNAL The "mark" signal is the opposite, or complementary signal to a "space" signal. See the definition of "space signal" for a definition of both. MARK-TO-SPACE TRANSITION A mark-to-space transition in data (i.e. the transition from the "1" to the "0" signal state) is particularly critical as a timing reference point in asynchronous transmission systems -- an most references to this topic are in context of achieving bit level and character synchronization in asynchronous systems.
MEAN TIME BETWEEN FAILURE (MTBF) Applied to either systems or components of systems, MTBF specifies the mean (or average) length of time the device or equipment will be expected to operate between failures. Usually expressed in hours, such as "expected MTBF = 9,000 hrs." -- implying an expected failure rate of about once per year of 24 hour operation.
MEAN TIME TO REPAIR (MTTR) MTTR specifies the expected mean or average time required to restore a failed system to operational status -- measured from the point in time when a failure occurred.
MEDIUM The communications resource used to actually carry information; It may be a copper cable, a fiber optic cable, a microwave radio signal or a satellite communications system.
MEDIUM ACCESS CONTROL (MAC) The MAC corresponds to the firmware-based logic on a LAN Network Interface Card (NIC) that supports the LAN access protocol (e.g. ethernets CSMA/CD). The OSI standard specification of Layer 2 (Data Link Layer) is split into two components in the IEEE LAN standards; the upper Logical Link Control (LLC) sub-layer, which is identical for all IEEE-standard LANs, and the lower Medium Access Control (MAC) sub-layer, which is different for each LAN technology.
MESSAGE SWITCHING Message switching systems (obsolete) were computer-based applications that accepted incoming text messages, store them temporarily (in (typically some kind of rotating memory) until the appropriate outgoing circuit was available, and then retransmitted, or forwarded the message on to the addressed destination. These systems replaced torn tape message centers (WW-II era) -- and have in turn been replaced by electronic mail systems.
METROPOLITAN AREA NETWORK (MAN) A Metropolitan Area Network is a network service that has the data rate and distance features that make it suitable for internetworking LAN systems (or other high speed applications) across the dimensions typical of a metropolitan area -- a dimension of perhaps 100 km. in diameter. MAN technology is seen to require sufficient data rate capability so as to offer virtual private network services to a number of unrelated user organizations -- on a common MAN backbone. Data rates of Gigabits per second and higher are currently regarded as necessary for this role. MAN technology, when deployed as a private campus backbone service, is generally referred to as a High Speed LAN (HSLAN) service. Most MAN solutions are either ATM or high-speed switched Ethernet based.
MICROWAVE SIGNALS Radio signals are classified as microwave signals if their wavelength is less than 10 cm. -- corresponding to a frequency of 890 MHz. or greater. Commercial microwave systems operate in the frequency range between about 2,000 MHz (2 GHz.) and 15,000 MHz. (15 GHz.), although emerging wireless systems (such as LMDS) will operate at much higher frequencies.
478
MODEM The word "modem" is a contraction of the words "modulator" and "demodulator", recognizing the two functions that the device performs. The modulator function of the modem applies to the encoding of digital information presented to it by the attached DTE, such that the physical characteristics and limitations of the communication channel can be accommodated, and the data can be successfully conveyed across the channel -- carried on an audible analog "carrier signal". The demodulation function is simply the inverse process. Modems thus are used to pass digital signals across analog circuits. MODULATION Modulation is a process whereby one signals variations (typically a source information signal like voice, and referred to as the baseband signal) are modulated onto, or imposed upon a second signal, called the carrier signal. Carrier signals are synthetically generated, and their characteristics are selected on the basis of their ability to carry the desired information across the communication gap of interest. A typical example is a radio transmitters carrier signal. See definitions of amplitude modulation, frequency modulation and phase modulation. MODULATION, PULSE CODE (PCM) See definition of "Pulse Code Modulation". MODULATOR A modulator is a device which takes baseband signals (such as voice or digital data), and uses them to modify the characteristics of a synthetic carrier signal for purposes of distance communication, or to share high capacity analog communication facilities (frequency division multiplexing). MULTIPLEXER Multiplexers are devices which make it possible for more than one signal to be simultaneously carried on a single (typically high capacity) common circuit. In data communication applications, multiplexers are typically either "time division multiplexers" or "statistical time division multiplexers". MULTIPOINT LINE Multipoint, or multidrop lines or circuits provide for the interconnection of multiple access points on a single, common data circuit -- a daisy chain configuration. MULTIPROCESSING Multiprocessing involves the simultaneous application of more than one processor (in either a multi-processor or distributed processing system) to a single job or task. Multiprocessing is commonly handled with parallel or array processors. MULTIPROGRAMMING A multiprogramming computer operating system enables the (apparent) simultaneous execution of more than one task by time-slicing its sequence of operating instructions. Multiprogramming systems are able to demonstrate apparent simultaneous execution of tasks by rapidly switching context between applications, executing a few instructions for one task and then moving on and executing a few instructions for another.
MULTISTATION ACCESS UNIT (MAU) The MAU is the hub device specified for Token Ring LAN systems. Originally, the non-intelligent IBM 8220 MAU was the only option -- requiring specific connector types and mounting arrangements. More recently, numerous vendors offer a wide range of physical configurations, connector types and intelligence in MAU products. MULTITASKING See definition of "Multiprogramming".
MULTITHREADING Multithreading is the simultaneous processing of multiple service requests by an application program -- typically a server process. Multitasking/Multiprogramming are supported by the operating system while Multithreading is supported by the application process (although multithreading typically requires a multiprogramming operating system). Multithreading is associated with reentrancy.
NAK A negative acknowledgement of message (typically a block of frame message segment) receipt -- indicating a request to retransmit the message segment. A NAK is typically signaled with a special control character (with half-duplex link protocols) or a pre-defined bit-flag (with full-duplex link protocols). The alternative to a NAK is an ACK, or positive message acknowledgement.
NATIONAL INSTITUTE OF STANDARDS AND TECHNOLOGY (NIST) Formerly, the National Bureau of Standards (NBS), NIST has historically been responsible for maintaining USA physical standards, such as length and time. They have been responsible for defining standards for data encryption technology (such as the public domain Data Encryption Standard or DES) -in cooperation with the National Security Agency (NSA).
NETBIOS NETBIOS (Network Basic Input/Output System) was the essential core of the "IBM LAN Program", which was the first widely used LAN operating system. NETBIOS is still supported by Microsoft networking products, although it attracts little continuing interest.
NETSCAPE Netscape is a graphical browser client tool widely used for access to the World Wide Web. Netscape was inspired by the success of Mosaic (and developed by the original author of Mosaic) -- the first commonly used package of this type. Netscape was the dominant web browser for a time, but lost this claim to Microsofts Internet Explorer.
NETWORK An interconnected communication facility which provides services to a number of service points, or users.
NETWORK FILE SYSTEM (NFS) NFS was originally a proprietary data base access technology developed by Sun Microsystems, capable of operating over TCP/IP networks. It is still proprietary, but Sun has made NFS specifications publicly available, and it is now regarded as an "open" technology. A key part of the acceptance of NFS
479
has been the convenience of interprocess communication (e.g. client to server) enabled by the associated Remote Procedure Call (RPC) feature. NETWORK MANAGEMENT CENTER (NMC) A central facility equipped with network monitoring equipment capable of detecting network failures and aberrations, analyzing their significance and dispatching resources to deal with the problem. Typical NMC capabilities include collecting operating statistics, maintenance scheduling and configuration control. NIST See "National Institute of Standards and Technology". NODE A node is a junction point in a network, where communication links converge. The term has also come to imply a point of intelligence in a network, particularly in the context of packet switching. NOISE In context of communications systems, noise is any unwanted signal that disturbs or impairs the ability to correctly interpret desired incoming signals. Noise may be caused by external disturbances (e.g. fluorescent lighting) or communications equipment itself. The concept of noise and noise impairment relate to the analog view of signals and signal interpretation. NTSC National Television Standards Committee, which was the original television technology specification. NTSC is the television standard used in North America, and carries an embedded dependency on a 60 Hz. electrical power infrastructure. OCTET An 8 bit unit of information, equivalent to the widely recognized "byte". The octet term is favored in international usage (such as standards specifications) and other formal applications -- nominally because of the historically ambiguous byte (which hasnt always referred to 8 bits -- sometimes 6 bits), and pronunciation difficulties with differentiation "bit" and "byte" in the French language. OFF HOOK The condition of a telephone which occurs when the handset is removed from its cradle, thus engaging the hook switch. Off hook status thus means that a circuit switched telephone connection is either in use, or in process of being activated. An off hook telephone condition will render the line "busy" to any in-calling attempts. ON HOOK The condition of a telephone which occurs when the handset is placed on its cradle, thus disengaging the hook switch. On hook status thus means that a circuit switched telephone connection is available for use, or otherwise idle. An on hook telephone condition will render the line available for in-calling attempts.
ON LINE TRANSACTION PROCESSING (OLTP) On Line Transaction Processing is a general term for transaction-oriented business computing systems that offer on-line, or real-time access from remote users. OLTP systems are typically configured as client/server systems.
OPEN SYSTEMS INTERCONNECTION (OSI) The OSI model and related standards were developed by the ISO (in cooperation with other, regional standards groups, like ANSI and ETSI. The OSI model was developed with the objective of defining an international, open (ie. non-proprietary) set of standards for computer communications applications that will permit interoperability between different vendor systems. Any hope of actually building networks in conformance to the OSI standards was pretty much killed by the complexity of the final specifications and the concurrent success of TCP/IP and the Internet.
OPTICAL FIBER Optical fiber is a thin filament of transparent material; typically glass or plastic. Optical fiber is comprised of a very thin core (which actually carries light pulses over the length of the fiber), and a more bulky surrounding material known as cladding. See "Fiber Optic Cable" definition.
OUT-OF-BAND SIGNALING A method of telephone network signaling which communicates the necessary signaling and control information over a network connection other than the message channel. Common Channel Interoffice Signaling (CCIS) and Signaling System Number 7 (SS7) are examples of out-of band signaling technologies.
PACKET A (typically) brief message segment which includes both user data and message addressing and control information which is routed, or switched as a composite message. The data and related control information are arranged in a standard, prescribed format, thus enabling packets from other users to intermingle on common network facilities.
PACKET SWITCHING Packet switching systems support the transmission of data by means of specifically formatted and addressed packets, resulting in a communication capability that occupies a transmission facility for the duration of transmission of the packet only. The communication facility, or network is thus available to deliver packets many different users on a shared basis. Note: Large messages will be subdivided in to a number of discrete packets for transmission across the network, and reassembled at receiving end of the link -- by means of a packet assembly/disassembly function. Packet switching service costs are proportional to the volume of information transmitted, and are unrelated to holding time. Compare with Circuit Switching.
PACKET SWITCHING NETWORK A packet switching network is designed to carry packet-formatted data. Within the packet network, all users traffic is presented in an identical format (packetized), differentiated only by a standardized header -- analogous to the address on an envelope. Packet routing decisions are made at a number of
480
switching, or node sites on the basis of header contents. PAL Phase Alternate Lines, referring to the phase alternation used in the color subcarrier, with the purpose of creating more stable hue reproduction. PAL is the television standard used throughout most of western Europe (except France), and in most countries that use a 50 Hz. electrical infrastructure. PAL is the most widely used analog television technology standard. PARALLEL TRANSMISSION Parallel transmission involves the provisioning of multiple data communication paths so that the data bits pertaining to a unit of data (typically a byte, or data character) can be passed across different circuits. Parallel systems are typically 8 bits (or some integer multiple of 8) wide, enabling all data bits of one byte to be sent in one clock time unit. Parallel transmission is faster than comparable serial alternatives, but is seriously troubled by timing control and signal cross-talk problems if the link length exceeds a few meters. Compare with Serial Transmission. PARITY Parity checking involves the addition of (redundant) non-information bits to data (typically, but not always the seven or eight-bit character representations), in such a way as to force the number of one bits in the group to be either always an even (or always an odd) number. Thus parity implementations are described as being either even parity or odd parity. Parity implementations permit detection of single error occurrences in the data which is "covered" by the parity calculation. (See "Horizontal Parity" and "Vertical Parity". PBX Private branch exchange. A telephone exchange on the users premises with access to the public network. (See definition under "Exchange"). PHASE MODULATION Phase modulation is one of three modulation techniques (See definition of "modulation"). With PM, the baseband signal information is encoded, or impressed on the carrier signal by modifying its phase value to instantaneously respond to variations in the baseband signal. PING The "Packet Internet Groper", or PING application program is a utility program available on TCP/IP-enabled systems. PING performs the simple task of repetitively transmitting datagrams to a designated network address, and recovering the replies (using the Internet Message Protocol, or ICMP). PING verifies the validity of the chosen address, availability of a network connection to the specified destination, and the operational status of that system. POINT OF PRESENCE (POP) The point (in a country, state/province or region) where a network service provider offers access to their network. It is typically a customers problem (expense) to provide the connection between the nearest POP and their business location.
POINT TO POINT PROTOCOL Point-to-Point Protocol -- a link control protocol capable of encapsulating any of a number of network layer protocols; most importantly IP. Numerous variations and/or feature extensions have been added to the original PPP specifications, enabling PPP to be used in a wide variety of applications, such as VPNs.
POINT-TO-POINT A communication circuit that establishes a connection between only two locations (in contrast to a multi-point circuit).
POLLING Polling systems operate as a way of controlling the shared access to multipoint communication lines. The central site system undertakes to be a "master of ceremonies", directing each station to speak in a prearranged order. Individual remote stations that have no knowledge of other stations needs to communicate are inhibited from transmitting at the same time -- and thus interfering with one another.
PORT A logical (software) or physical (hardware) point of interaction with a system. Identification of appropriate ports to use for accessing systems (generally in context of several alternative ports being available) is often necessary; see DSAP and SSAP definitions.
PROPAGATION TIME DELAY Propagation time delay is the time required for a signal to travel from one point in a system circuit to another -- that is, the time required to traverse the network. Commonly, it is the round-trip time delay that affects system performance (eg. the waiting time between message transmission and receipt of acknowledgement). Open air radio signals (such as satellite communication signals) travel at (very close to) the speed of light; constrained signals operating within cables (metallic or optical) are commonly estimated at one half of the speed of light. It takes light approximately 5 milliseconds to transit 1,000 miles -- or 3 milliseconds to transit 1,000 km. Propagation delay is partly comprised of the transit time limitations of the speed of light, and partly due to signal processing delays in equipment along the way.
PROTOCOL A formally defined procedure which describes the format, and procedural sequencing of messages between two communicating entities. required to initiate and maintain communication. The term "Protocol" has become more generalized in recent practice. At one time it referred to what is now know as "Link layer protocol", which deals explicitly with the structuring (blocking or framing) of binary data sequences, their validity checking (error control) and addressing. The more general (current) usage is extended to all seven layers of the OSI standard architecture definition -- in which "protocol entities" exchange information in "protocol data units".
481
PT&T Postal, telephone and telegraph organization which acts as a nations public telephone service provider. The term PT&T is not used in context of investor-owned telephone companies -- PT&Ts are government-controlled monopoly organizations which acts as an arm of national telecommunication policy. The trend to global deregulation has very much reduced the number of countries where this class of telephone operation exists. PUBLIC SWITCHED TELEPHONE NETWORK (PSTN) The PSTN is the ubiquitous, global public telephone system that provides general connectivity to the approximately 1 billion telephone subscribers worldwide. The integration of the traditional wired telephone infrastructure with rapidly growing wireless services is greatly expanding the role of the PSTN. PULSE A pulse is a short change in the electrical or optical signal state of communication -- or electronic -- system. Pulse values are typically encoded to represent binary data, and are allowed to exist for very specific periods of time, consistent with some uniform, and mutually accepted measure of time. (See definition of "Clock"). PULSE-CODE MODULATION (PCM) PCM is a method of representing analog signals (such as speech) by taking periodic samples of the time-varying analog signal (the industry standard sampling rate is 8000 samples per second), and converting each sample value into a binary code (the industry standard is an eight-bit code, sufficient to represent 256 distinct sample values). PCM is not a true modulation process (See definition of "Modulation"). PUSH TECHNOLOGY In contrast to typical information recovery or retrieval systems which require the user to explicitly seek, or "pull" information to their workspace, push technology automates the outward dissemination of information. This is done by defining a user interest profile in some way, and providing an information delivery mechanism. RAID Redundant Arrays of Inexpensive Disks, or RAID systems use sophisticated disk controllers and associated software to create highly reliable aggregated, fault-tolerant disk systems from readily available, inexpensive disk hardware. There are five levels of RAID sophistication ranging from simple redundant mirroring through hot-swappable dynamically redundant systems. REDUCED INSTRUCTION SET COMPUTER (RISC) A computer which operates a simplified instruction set (compared with a traditional "Complex Instruction Set Computer", which supports a large number of often lengthy and complex built-in instructions as required by high level languages) which is specially designed to operate efficiently on hardware designed or this purpose. RISC systems therefore perform the majority of simple repetitive computing tasks very efficiently, and combine these simplified procedures into software routines when required to perform more complex functions.
REDUNDANCY Redundancy involves the concept of duplication, or (typically) underutilized capacity. There are two different contexts: (1) Information redundancy, which involves data representations (codes) that are less efficient than they might theoretically be. This is comparable to redundancy in language -- a typical sentence is still comprehensible when a few characters are incorrectly presented -- as with spelling errors. (2) Equipment redundancy involves provisioning of spare components in a critical system, so that failure of one will not terminate service. Dual automobile headlamps are an example -- one only could provide sufficient illumination, but its failure would be too serious to be an acceptable risk.
REPEATER, REGENERATIVE (1) Repeaters are used in telephone network connections to link high capacity, long distance systems together. The distance limitations of specific technologies (fiber optic cables, microwave radio systems, etc.) are overcome by cascading multiple maximum length spans of these technologies with repeater interconnections. (2) Repeaters are used in much this same role in context of local area networks as a way of extending the physical rang of LAN segments, or to interface different cable media types on a single LAN.
RESIDUAL BIT ERROR RATE, UNDETECTED ERROR RATE The residual error rates (See definition of "error rate") is the rate at which errors are able to escape error detection and remain undetected after whatever means of error detection has been utilized -- and thus residual in data that is thought to be correct.
RESPONSE TIME Response time is the time that a system requires to respond to a given input. This is issue is commonly associated with enquiry-response transactions on multipoint communication systems. When an enquiry (or other transaction initiative) is keyed into a terminal by an operator there is generally an expected reply from the controlling computer system. There is a short window of time following this initiative within which the terminal user perceives favorable responsiveness from the remote host system. Beyond this critical, and subjective time window, the user becomes quickly frustrated with process delays. Components of response time delay which must be individually addressed in system planning are: (1) time required to transmit enquiry data inbound to the computer; (2) time required to process the service request at the computer; and (3) time required to transmit the response data back to the terminal. The personal computer revolution has changed typical user response time expectations from a few seconds (perhaps as much as 5 or 8 seconds in time past) to sub-second times.
RFC Request for Comment -- a series of documents which specify TCP/IP and numerous related protocol standards, and provide a forum for developing consensus relative to emerging technical developments for TCP/IP and the Internet -such as the Next Generation versions of IP and TCP. The Internet Engineering Task Force (IETF) maintains a convenient web site for accessing the latest RFC and draft documents (www.ietf.org).
482
RGB Red, Green, Blue -- the three primary colors of light represented in television systems. Once color signals are fully decoded in a television receiver, they are presented to the color display in RGB format. ROUTER A router is a data network connection device that makes decisions about passing data between connected networks based on network layer addresses (OSI Layer 3). Routers are configured to deal with specific network layer protocols, and thus must be configured to cope with all protocols used on connected networks. Routers are required to coordinate "reachability" information with other routers with specific route exploration protocols. See contrasting definition of "Bridge". ROUTING INFORMATION PROTOCOL (RIP) A router-to-router coordination, or route exploration protocol used on small TCP/IP networks, or used internally on small TCP/IP sub-networks. RS-232-C The TIA interface standard commonly known as the RS-232-C is actually more correctly specified as the TIA-232-C. It is a 20 kbps. maximum data rate, unbalanced electrical interface suitable for connecting DCEs to DTEs (e.g. modems to computers). A "D" version of the interface has been defined but is not yet widely supported. SECAM Sequential Couleur Memoire, a television standard introduced by France, with the widely accepted (but unofficial) purpose of being incompatible with the rest of the world -- nominally to protect the French manufacturing industry. A variant of SECAM was adopted by the Eastern Bloc countries to discourage reception of Western European PAL signals off-air. SELECTIVE ARQ Selective ARQ systems provide for the explicit identification of an errored frame of data, and for the retransmission of only that frame. Contrast with "Go-Back-N ARQ", which provides for simply backing up to the point of error occurrence, and starting over a that point. HDLC link protocol supports this technique as an option. SERIAL LINE INTERNET PROTOCOL (SLIP) SLIP is a simple protocol for extending access to IP networks over point-to-point connections, such as dial telephone links. SLIP is regarded as obsolete, having been replaced by PPP as the IP access protocol of preference. SERIAL TRANSMISSION Serial transmission systems provide for transmitting all of the data bits in the information stream across a single communication channel -- in bit-sequential format. Thus, it requires 8 clock time units to transmit a single 8-bit byte of information. Effectively all implementations of data communication systems which operate beyond a few meters are based on serial transmission. Compare with Parallel Transmission.
SERVER The software supporting the host-oriented computing functions in a client/server computing environment. Server software typically concerns itself with managing shared data resources; for this reason database management systems are often a key part the server. The term server is often (somewhat casually) associated with the hardware computing system which supports the server as defined here.
SHIELDED TWISTED PAIR (STP) STP cable is configured in much the same format as unshielded twisted pair (See definition for comparison), except that a braided exterior cylindrical shield is constructed surrounding each individual pair of wires on the bundle. STP cable has technically superior performance (compared with equivalent UTP) , but the additional cost and bulk make it unpopular compared to UTP. The addition of the shield makes it necessary to package fewer pairs per cable bundle (thus occupying significantly more duct space) -- and the helical cable twist makes the exterior jacket irregular -- and difficult to manipulate through cable ducts.
SIGNAL A signal consists of electromagnetic impulses which propagate across a transmission channel, and are encoded, or modulated in such a way as to convey information. Signals can be broadly classified as being either analog or digital in character.
SIGNAL-to-NOISE RATIO (SNR) The ratio (typically expressed in decibels), of the strength of a signal to the strength of the noise energy present on the channel. A good quality analog telephone channel would demonstrate a 1000:1 (30 decibel) signal to noise ratio.
SIGNALING Signaling is the process by which network connection requests (such as dialed telephone numbers) and control signals (such as "hang up") are communicated across a network. Signaling can be either "in-band" (carried in the same bandwidth as message traffic) or "out-of-band" (carried over a separate channel or system).
SIMPLE NETWORK MANAGEMENT PROTOCOL (SNMP) SNMP is the network management coordination protocol used on TCP/IP-based networks. SNMP supports an effective and widely used set of network management services, and is very popular as a basis for managing enterprise networks.
SIMPLEX Simplex communication systems have the capability of carrying information in only one direction. An example is a pocket pager system -- where there is never any capability of transmitting a signal from the pager back to the base station to confirm receipt of a message.
483
SINGLE CHANNEL PER CARRIER (SCPC) SCPC configured equipment is used with satellite communications equipment that serves limited capacity or small earth stations. Typically, VSAT earth terminals operate in SCPC mode. Technically, SCPC means that only one message channel (such as a voice conversation or a data session) is carried on each satellite radio carrier. High capacity systems, on the other hand, multiplex and modulate a number of messages onto a single carrier simultaneously. SLIP See "Serial Line Internet Protocol". SONET Synchronous Optical Network (Known in Europe as the Synchronous Digital Hierarchy, or SDH) is an international standard for high speed WAN (i.e. long distance telephone company networks) applications. SONET is an open-ended specification, in that higher speed data rates will be defined as the need/capability emerges. Current SONET systems are defined up to the OC-768 rate (approximately 40 Gbps), typical implementations operate at the OC-192 (10 Gbps) rate. SPACE The "space" condition on a data communication circuit (opposite and complementary to a "mark" condition) is equivalent to a logical binary 0 signal state. The mark is equivalent to a logical binary 1. The terms mark and space are borrowed from the days of Morse code telegraphy, and the earliest automatic code reading equipment, which left marks or spaces on a strip of paper tape corresponding to the presence or absence of Morse code pulse. The terms mark and space are much more commonly associated with asynchronous than synchronous transmission. SPACE CHARACTER In most character codes, the requirement to leave a physical space between printing characters is handled through the use of a "space" character. This is the character code transmitted when the keyboard space bar is struck. SSAP Source Service Access Point -- the upper layer protocol specification for the network layer protocol from which data are derived to construct outgoing packets. STOP AND WAIT ARQ A process of transmitting data and waiting for a response from receiving station before proceeding with transmitting more data -- or retransmitting previously sent data -- as required. Stop-and-wait systems necessarily operate in a half-duplex mode. Binary synchronous link protocol and XMODEM are examples of stop-and-wait protocols. STORE AND FORWARD A message forwarding mechanism where an incoming message is fully buffered at a network node switching point (router), data validity confirmed, the possibility of retransmission contemplated (because of errors), and a message-by-message dynamic routing analysis performed. The message is then forwarded to the next way-point on route to the destination. This procedure
is time consuming, but is suitable to wide area networks where data rates are low-to-moderate and errors in transmitted data are frequent enough to warrant the overhead. The first store and forward systems were message switching systems (replacing torn-tape message centers); latterly, X.25 packet switching are described with this term.
SUPERSERVER Superservers are high-end multiprocessor computer systems designed to function as server platforms for demanding applications. Superservers are typically RISC architecture systems with optimized I/O capability and are frequently equipped with RAID mass storage systems.
SYNCHRONOUS Synchronous systems operate on a common, or uniform time base, creating a constant and uniform measure of time for purposes of data transmission and recovery.
SYNCHRONOUS DIGITAL HIERARCHY (SDH) The SDH is the European designation for SONET -- the Synchronous Optical Network -- standard for high speed WAN network multiplexing. See definition of "SONET".
SYNCHRONOUS TRANSMISSION A method of transmitting data in which both the sending equipment and receiving equipment are driven by a common clocking system -- typically from the transmitter end of the link. Furthermore, the clocking system is continuously operated, and provides for a uniform measure of time at both ends of the link -- so that data pulses can be accurately decoded.
SYSTEMS NETWORK ARCHITECTURE (SNA) SNA has been IBMs principal computer communications architecture, initially defined in 1974. It is significant as the industrys first structured, layered architecture definition, and as such, served as the model for developing the now-universal view of layered computer communications architectures. The proprietary characteristics of SNA very much limit is usefulness in current time.
TAG SWITCHING There are two main categories of LAN switching technology deployed in support of Virtual LAN service -- switches based on adapting bridging technology (Layer 2 Switches) and switches based on adapting routing technology (Layer 3 switches). Of the Layer 2 switch alternatives, Tag Switching is the most popular. It involves appending a tag (additional message field) to the data frame that provides a numeric identification of the virtual LAN to which the origination station is associated. See "IP Switching" for Layer 3 switch discussion. The IEEE 802.1p specification moved this technology from its initially proprietary category.
484
TARIFF Tariffs are published, standard rates for communications services or equipment. Regulated telephone companies have no flexibility in pricing their services once their regulator has approved a given tariff package. The term is applicable to other industries as well, such as the case of freight tariffs used by railways. This term is less widely used now that the industry works with competitive pricing models driven by deregulation. TCP/IP Transmission Control Protocol/Internet Protocol. TCP/IP is really two distinct protocols; TCP operating at layer 4 of the OSI reference model, and IP operating at Layer 3. In combination with many other less widely discussed protocols, TCP/IP has come to be used to describe the whole family of networking protocols, which collectively enable interoperation between dissimilar host systems and between dissimilar networks - a heterogeneous interoperability capability that is unique in the industry. TELECOMMUNICATIONS INDUSTRIES ASSOCIATION (TIA) The TIA (Formerly, the Electronic Industries Association, or EIA) is a Washington, DC based organization that has provided a leadership role in defining technology standards for the US community. TELEPRINTER An obsolete asynchronous terminal device that consists of a keyboard and integrated low-speed printer. An alternate term is teletypewriter. TELNET The Teletype Network, or Telnet is an interactive terminal service specific to TCP/IP network applications. Telnet (very similar to "rlogin") allows a remote network access user to operate with the same privileges and functions as a locally connected terminal user. The poor response time and primitive human interface features of Telnet make it unfavored for user-accessible applications. TERMINAL In general, a terminal is a point in a communication system where information can enter or leave the system. In common practice, a terminal is a device (typically operated by humans) which transmits and/or receives information over a communication channel. Although computers connected at the edge of a network are technically terminals, common usage restricts the term to non-intelligent devices. TIA See definition of "Telecommunication Industries Association". TIME ASSIGNED SPEECH INTERPOLATION (TASI) Time assignment speech interpolation, or TASI systems take advantage of the fact that human speech on any given telephone circuit in any given direction will never average more than 50% busy. In fact, due to pauses in conversation, it is more like 40% busy. It is often effective to equip devices at both ends of a telephone network to dynamically allocate the bursts of voice message activity to a pool of available one-way circuits -such that the otherwise inactive, and non-productive half of each conversation path can be utilized. TASI systems can improve utilization of
ocean cables, for example, by up to 80%. New technology TASI systems are also called Digital Speech Interpolation, or DSI systems.
TIME DIVISION MULTIPLE ACCESS (TDMA) TDMA is a technique for time sharing a common communication channel -typically a geographically dispersed channel, such as a satellite service. Devices at different geographical locations can feasibly share such as channel if they are disciplined to coordinate their short bursts of information transmission with all other users -- on a time shared, or time division basis. TDMA is in some ways like a distributed time division multiplexer system. Compare with FDMA.
TIME DIVISION MULTIPLEX (TDM) Time Division Multiplexers collect data from a number of typically low speed data sources, and interleave the incoming data streams onto a single, higher speed communication link to a companion multiplexer device, where the interleaved data is allocated back to a number of low speed distribution ports -- commonly ports on a computer system. If the allocation of time slots on the common high speed inter-multiplexer channel is fixed, the device is regarded as a simple, or traditional TDM. If the allocation is dynamically altered based on statistically variable demand from terminal devices -- such as with human interactive terminals -- it is said to be a Statistical Time Division Multiplexer -- or "statmux".
TOKEN The token used with token passing LAN systems is a special message that circulates over the same path as message traffic. The purpose of the token is to grant authority to the station that holds the token to transmit. The system design is configured to assure that there is only one token present on the system at any time (thus insuring there will be no contention for access) and that it is passed quickly around the community of attached devices (thus insuring that no station waits an inappropriately long time for a chance to transmit). The complexity of managing token behavior in the absence of any centralized authority makes token passing LAN systems more complex than ethernet.
TOKEN BUS Token bus LAN systems are the most complex of the common LAN technologies, due to the challenge of maintaining an addressing regime for passing the token to all attached devices. Because a bus is a broadcast medium, all devices receive all messages, and address differentiation is the only way to identify the intended destination for messages and tokens. The addressing technique used for token distribution is known as a "logical ring". The IEEE 802.4 is an example of token bus technology.
TOKEN RING The standard token ring LAN technology was developed by IBM in the mid-1980s, and accepted by the IEEE as a standard (IEEE 802.5). It is the second most commonly used LAN technology, after Ethernet, with perhaps 5% market share.
485
TRACEROUTE Traceroute (a common utility program) provides a listing of the routers traversed in establishing a connection over a TCP/IP internet. TRANSMISSION CONTROL PROTOCOL (TCP) TCP is the principal session control, so called "host-to-host" protocol with TCP/IP systems. TCP provides connection control and positive delivery acknowledgements that are not available with the underlying IP. TRANSPONDER An electronic subsystem (typically associated with communication satellites) which includes a combined receiver-transmitter function. Transponders receive, frequency-shift, amplify and retransmit users signals, operating on the signals as an analog processor. TRIVIAL FILE TRANSFER PROTOCOL (TFTP) Trivial File Transfer Protocol is a TCP/IP application process that offers a more rudimentary file transfer capability than the more common FTP. An advantage of TFTP is that it doesnt require a multitasking operating system for support, as does FTP, and can be implemented in firmware. TRUNK Trunks are telephone network connections which interconnect switching equipment. (Compare with "local loops" which extend connectivity to switches from non-switching equipment, such as telephones). TTY Abbreviation for "Teletype". Teletype has been historically a proprietary trade name (AT&T), but has evolved to refer to mechanical teleprinters of all manufacture, and the means of communication that they supported. TTY communication can be regarded as a primitive link protocol, associated with asynchronous timing, ASCII or Baudot coded data, and character-mode terminal systems. The serial port of a PC is a TTY-capable interface. TWO-WIRE CIRCUIT A circuit formed by a pair of conductors, or wires, which are insulated from each other, and typically twisted together. The two metallic conductors create and support the simplest possible communication mechanism, which typically supports transmission of information in either direction (without any directional preference). Telephone local loop connections to the public switched network are always connected as a two-wire service (ie. using one pair of wires). See definition of "Unshielded Twisted Pair". UHF Ultra high frequency. VHF is a term used to describe radio frequencies in the range 300 MHz to 3 GHz. UHF frequencies are characterized as "line-of-sight" services (not capable of use significantly beyond the horizon), and are widely used in personal communication (cellular), television , low-capacity point-to-point and radar services.
UNIFORM RESOURCE LOCATOR (URL) The URL concept provides for a common syntax for specifying the applicable access protocol, the host system address, path on the specified host system, and file specification for information resources on TCP/IP internets. The URL format is most commonly encountered in context of specifying Hypertext Transport Protocol (HTTP) information resources on the World Wide Web.
UNINTERRUPTABLE POWER SUPPLY (UPS) UPS systems provide for power continuity in event of failure of primary power. This is typically accomplished using batteries and power inverter circuits to generate the required primary power replacement. Where long duration backup capability is required, some alternate source of power is required, such as a diesel generator.
UNSHIELDED TWISTED PAIR (UTP) UTP is the term used to describe any telephone-type cable, usually in context of house cabling, or building wiring systems. UTP is comprised of a pair of insulated wires, twisted together into a helix. UTP cable pairs are typically packaged into cable bundles consisting of multiple pairs contained within a common exterior jacket. There a number of (most vendors agree on 5) different classes of UTP, differentiated on the basis of wire gauge, insulation type and diameter, and the number of twists per foot in the cable construction. These different "levels", or classes of cable are designated as being suitable for voice only, low or high speed data, ISDN applications, etc. See definition of "Two-Wire Circuit".
USENET USENET (Users Network) is the interconnection of a collection of UNIX systems -- all of which run the "netnews" software, which supports a news exchange service between users of the network. It has evolved into a remarkable means of exchanging information about any of hundreds of special interest subjects.
USER DATAGRAM PROTOCOL (UDP) A connectionless protocol which operates parallel to TCP in TCP/IP systems -and provides support for very simple application processes (such as PING and TFTP), more sophisticated processes which provide their own session control (such as Sun Microsystems NFS) or time-critical services like those supported by Real Time Protocol.
UUCP UUCP, or the "UNIX-to-UNIX Copy Program" is a common application program used for file transfer in the UNIX environment. There is a close association between electronic mail and file transfer (UUCP performs both tasks), and as a result, UUCP has come to be associated with mail transfer in UNIX-based systems.
VERONICA Obsolete: Veronica is a gopher-initiated internet search tool which locates files (or file directories) based on keyword specification.
486
VERTICAL PARITY DETECTION Vertical parity calculations are based on defining a single parity bit which accompanies each character that is transmitted -- such that the composite group of bits (Typically 7 character data bits plus 1 parity bit) will have a predictable even (or odd) number of "one" bits. (See definitions of "Parity" and "Horizontal Parity"). VERY SMALL APERTURE TERMINAL (VSAT) VSAT communications satellite earth terminals are suited to relatively low capacity remote locations -- typically used to deliver a small number of (perhaps up to 4) channels for low-to-moderate data rate applications (up to about 64 kbps. per channel) or voice applications. VHF Very high frequency. VHF is a term used to describe radio frequencies in the range 30 MHz to 300 MHz. VHF frequencies are characterized as "line-of-sight" services (not capable of use significantly beyond the horizon), and are widely used in personal communication (mobile) and television services. VIRTUAL CIRCUIT A term used to describe the addressable connection that exists across a packet switched network. Virtual circuits behave in some ways like real physical circuits, but they are "virtual" in that the only link across such a network is sustained through a packet delivery capability. Permanent virtual circuits are configured and available for service 24 hours a day continuously -- and offer no flexibility regards different connection options. Switched virtual circuits require user initiation of each session, and offer flexibility with regard to connection options. VOICE FREQUENCY Any (audible) frequency within that portion of the audio frequency spectrum which is consistent with the transmission of telephone quality speech. See "Voice Grade Channel"). VOICE GRADE CHANNEL A telephone-grade communication channel which is designed for, and is suitable for transmission of speech, modem-interfaced digital data, and facsimile information. In most countries, frequency bandwidth, or range of about 300 to 3000 Hz. characterizes the voice grade channel. WAN See definition of "Wide Area Network". WIDE AREA NETWORK (WAN) A WAN has traditionally been viewed as a low speed network service -- a view no longer accurate. WAN services are properly considered to be network services that extend beyond the immediately local environment (as supported by LAN systems) -- and except for the largest scale users, will be based on facilities leased or rented from a public network service provider, or common carrier. Some definitions of the WAN include metropolitan area network, or MAN access facilities -- in which case a LAN-to-LAN internetwork connection would be a LAN-WAN-LAN configuration. Other definitions differentiate, such that in inter-city LAN-to-LAN service would be seen as a
LAN-MAN-WAN-MAN-LAN configuration.
WIRELESS LAN Wireless LAN systems use radio or infrared technology to avoid or minimize the need for hardwired cable systems. High cost, limited capacity and security concerns initially limited widespread acceptance of wireless LAN technology. Standardization efforts (IEEE 802.11 and Hiperlan) are changing this, promising greatly increased acceptance.
WORD Historically, a word has been defined a six characters (typically comprising five characters plus one space). More typically, a word is considered to be a group of bits (typically comprising an integer number of octets), which is treated as a unit and is capable of being stored or communicated as a unit of information within a computer. A common word size is 32 bits.
WORKSTATION Workstations are high-end, powerful microcomputer systems. These kinds of equipment are often associated with graphics-intensive technical design applications and are also frequently deployed as high-end server platforms. Workstations are often configured with RISC processors and are often supported by the Unix operating systems.
WORLD WIDE WEB (WWW) The WWW is the collection of Internet-based resources that support the Hypertext Transport Protocol (HTTP), making it possible to exchange documents that are encoded with the Hypertext Markup Language (HTML). Using a client software package (such as Internet Explorer), the WWW becomes a simple, convenient vehicle for locating information resources.
X.25 X.25 is an obsolete ITU-T standard that defines the characteristics of first-generation packet switched networks. The X.25 standards are historically important because they inspired the more current frame relay and ATM network standards.
XEROX NETWORK SYSTEM (XNS) XNS was the first functional network operating system designed for LAN systems, and established many of the features and functions that have come to be associated with this class of system.
487
George Abe Residential Broadband Indianapolis IN Cisco Press 1997 (ISBN 1578700205)
1.2 1.3
Uyless Black Emerging Communications Technologies (Second Edition Upper Saddle River NJ Prentice Hall 1997 (ISBN 0137428340) 4.3 4.4
Martin P. Clark ATM Networks - Principles and Use Chichester UK John Wiley & Sons 1996 (ISBN 0471967017)
Ranier Hndel Manfred Huber and Stefan Schrder ATM Networks .. Concepts Protocols Applications Reading Mass. Addison Wesley 1998 (ISBN 0201178176) 4.5 4.6 4.7
David E. McDysan and Darren L. Spohn ATM Theory and Application New York NY McGraw Hill 1998 (ISBN 0070453462)
Stamatios V. Kartalopoulos Understanding SONET/SDH and ATM Piscataway NJ IEEE Press 1999 (ISBN 0780347455)
Timothy Kwok ATM: The New Paradigm for Internet Intranet and Residential Broadband Services & Applications Upper Saddle River NJ Prentice Hall 1998 (ISBN 0131072447) 4.8
2.2
Byeong Gi Lee Minho Kang and Jonghee Lee Broadband Telecommunications Technology (Second Edition) Norwood MA Artech House Inc. 1996 (ISBN 0890068666) 4.9
Mark A. Sportack Frank C. Pappas Emil Rensing et al HighPerformance Networking Indianapolis IN Howard W. Sams & Company (Division of Macmillan Inc.) 1997 (ISBN 157521 1874)
4.10 William Stallings ISDN and Broadband ISDN With Frame Relay and ATM (Fourth Edition) Upper Saddle River NJ Prentice Hall 1998 (ISBN 0139737448)
488
5.13 Radia Perlman Interconnections: Bridges Routers Switches and Internetworking Protocols Reading Mass Addison-Wesley 1999 (ISBN 0201634461)
5.14 Michael Santifaller TCP/IP and ONC/NFS - Interworking in a UNIX Environment Addison Wesley 1995 (ISBN 0201422751)
5.15 William Stallings High-Speed Networks; TCP/IP and ATM Design Principles Upper Saddle River NJ Prentice Hall 1998 (ISBN 0135259657)
5.4
5.16 W. Richard Stevens TCP/IP Illustrated Volume 1 Reading Mass. Addison Wesley 1994 (ISBN 0201633469)
5.17 Stephen A. Thomas HTTP Essentials; Protocols for Secure Scalable Web Sites New Y ork NY John Wiley & Sons 2001 (ISBN 0471398233)
5.5
5.18 Kevin Washburn and Jim Evans TCP/IP; Running a Successful Network (Second Edition) Harlow UK Addison Wesley Longman Ltd. 1996 (ISBN 0201877112)
5.6
5.19 Robert Wright IP Routing Primer Indianapolis IN Cisco Press 1998 (ISBN 1578701082)
Jennifer Niederst and Richard Koman Web Design in a Nutshell: A Desktop Quick Reference Sebastopol CA OReilly & Associates Inc. 1999 (ISBN 1565925157)
5.10 Christian Huitema Routing in the Internet (Second Edition) Upper Saddle River NJ Prentice Hall 2000 (ISBN 0130226475) 5.11 Craig Hunt TCP/IP Network Administration (Second Edition) Sebastopol CA OReilly & Associates Inc. 1998 (ISBN 1565923227) 5.12 Donald C. Lee Enhanced IP Services for Cisco Networks Indianapolis IN Cisco Press 2000 (ISBN 157870247X)
Christian Huitema IPv6: The New INternet Protocol Upper Saddle River NJ Prentice Hall 1998 (ISBN 0138505055)
Christopher Y. Metz IP Switching Protocols and Architectures New York NY McGraw Hill 1998 (ISBN 0070419531)
Ivan Pepelnjak and Jim Guichard MPLS and VPN Architectures Indianapolis IN Cisco Press 2001 (ISBN 1587050021)
Stephen A. Thomas IPng and the TCP/IP Protocols New Y ork NY John Wiley & Sons 1996 (ISBN 0471 130885)
489
9.5
Paul E. Renaud Introduction to Client/Server Systems: A Practical Guide for Systems Professionals New Y ork NY John Wiley & Sons 1996 (ISBN 0471133337)
Amjad Umar Object-Oriented Client/Server Internet Environments Upper Saddle River NY Prentice Hall 1997 (ISBN 0133755444)
8.3
8.4
10.1 Ira Brodsky Wireless Computing: A Managers Guide to Wireless Networking New Y ork NY V an Nostrand Reinhold 1997 (ISBN 0442019122)
10.2 Editors Official Wireless Application Protocol New Y ork NY John Wiley & Sons 1999 (ISBN 0471327557)
8.5
10.3 Vijay K. Garg and Joseph E. Wilkes Principles and Applications of GSM Upper Saddle River NY Prentice Hall 1999 (ISBN 0139491244)
8.6
10.4 Jim Geier Wireless LANs; Implementing Interoperable Networks New York NY Macmillan T echnical Publishing 1999 (ISBN 1578700817)
8.7
10.5 Lawrence Harte Steve Prokup and Richard Levine Cellular and PCS; The Big Picture New Y ork NY McGraw Hill 1997 (ISBN 0070269440)
10.6 Nathan J. Muller Mobile Telecommunications Factbook New Y ork NY McGraw Hill 1998 (ISBN 0070444617)
10.7 Jochen Schiller Mobile Communications Harlow UK Addison Wesley Longman Ltd. 2000 (ISBN 0201398362)
10.8 James D. Solomon Mobile IP; The Internet Unplugged Upper Saddle River NY Prentice Hall 1998 (ISBN 0138562466)
10.9 Mark S. Taylor William Waung and Mohsen Banan Internetwork Mobility; The CDPD Approach Upper Saddle River NY Prentice Hall 1997 (ISBN 0132096935)
10.10 Ellen Kayata Wesel Wireless Multimedia Communications; Networking Video Voice and Data Reading Mass. Addison Wesley 1997 (ISBN 0201633949)
490
12.8 Deborah Russell and G.T. Gangemi Sr. Computer Security Basics Sebastopol CA OReilly & Associates Inc. 1992 (ISBN 0937175714)
13.1 Paul W. Abrahams and Bruce A. Larson Unix for the Impatient (Second Edition) Reading Mass Addison-Wesley 1996 (ISBN 0201823764)
13.2 Michael Hammer and James Champy Reengineering The Corporation; A Manifesto for Business Revolution New Y ork NY Harper Collins Publishers 1993 (ISBN 088730687X)
13.3 Allan Leinwand and Karen Fang Conroy Network Management - A Practical Perspective (Second Edition) Reading Mass. Addison Wesley 1996 (ISBN 0201609991)
13.4 Kornel Terplan Communication Networks Management (Second Edition) Upper Saddle River NJ Prentice Hall 1991 (ISBN 0131564498)
13.5 Donald A. Waterman A Guide to Expert Systems Reading Mass. Addison Wesley 1986 (ISBN 0201083132)
13.6 Chuck Hutchinson Joel Kleinman and Dean Straw (Editors) ARRL Handbook for Radio Amateurs 2001 Newington CT The American Radio Relay League 2001 (ISBN 0872591867)
14.1 Tim Berners-Lee Weaving The Web; The Original Design and Ultimate Destiny of The World Wide Web New Y ork NY . Harper Collins Publishers 2000 (ISBN 006251587X)
14.2 George Gilder Telecosm: How Infinite Bandwidth Will Revolutionize Our World New Y ork NY . Free Press Division of Simon & Schuster 2000 (ISBN 0684609303)
14.3 Robert W. Lucky Silicon Dreams: Information Man and Machine New York NY St. Martins Press 1991 (ISBN 031205517X)
14.4 Cliff Stoll The Cuckoos Egg; Tracking a Spy Through the Maze of Computer Espionage New Y ork NY . Pocket Books Division of Simon & Schuster Inc. 1989 (ISBN 074341 1463)
491
IEEE associate membership is available to everyone and involves an annual fee of approximately $100.00 depending on small variations due to state/country of residence. Membership includes a subscription to IEEE Spectrum magazine. Inquiries regarding membership other publications and the IEEEs extensive and highly recommended list of books tutorial and collected reprint items direct communication to IEEE Service Center 445 Hoes Lane Piscataway NJ 08854-4150 USA (T elephone: 908-981-0060). (www.ieee.org)
ACM associate membership is available to everyone and involves an annual fee of $95 and includes a subscription to Communications of the ACM. Inquiries should be directed to ACM 1515 Broadway New Y ork NY 100365701 USA. (T elephone: 800-342-6626). (www.acm.org)
492