COMPUTERIZED SYSTEM FOR MONITORING

WIRELESS NETWORK PERFORMANCE

ABSTRACT

Computer networks need protocols to govern all transmission and presentation processes. The
Transmission Control Protocol (TCP) is one of the most important protocols that has the
compatibility to work with all types of computer networks, overcoming all architectural and
operating system differences. Nowadays, networks depend on the TCP protocol to control data flow
between all types of connected computers, whether it is client or server, over any type of media
whether it is wired or wireless networks, for all network topologies. A simulation of a university
campus network has been conducted to determine TCP protocol features, those features are taken
into consideration as one of the most important network parameters. In all digital networks, the data
transmission is not a continuous transmission – instead, it is a discreet transmission, presenting itself
as packets. These packets transfer and propagate within the network between computers, and network
nodes using the TCP protocol depending on the address, which is embedded in its header. TCP has a
great influence on the network speed. The network simulator OPNET provides an easy way of
campus design, predicting and estimating the performance of networks in a university campus
environment.
In this research, wired connections reach all computer network users at fixed points to maintain
higher Mbps and ensure reliable communications between all the campus network nodes, as well
as to increase the overall network performance taking into account the future expansions for the
university campus network design.

INTRODUCTION AND BACKGROUND OF THE STUDIES

Today’s world is entirely dependent on information. For the ease of sharing and communication of
information, data networks play a vital role by providing many services. Therefore the whole world is
connected and empowered using various kinds of networks based on different technologies and scales.
These networks handle day-to-day information, news, multimedia as well as sensitive data such as
banking details, financial and critical details to companies and firms of the government and private
sector. Therefore data networks should be dependable for their users even at situations where there is a
heavy load on the network. The ideal data network should be available, reliable, maintainable and
durable consistently. Unfortunately, failures are inevitable. The best solution to minimize the cost of
failures is to always learn, understand and detect risks and failures as early as possible. For this
monitoring of networks has become one of the vital practices for network managers.

There are many open source (eg: Nagios) and commercial products (eg: OPmanager, Orion) which
provide network monitoring facilities. These applications provide network engineers and administrators
with many features and tools to identify and examine network infrastructure (Perera, 2014 pp 2). Many of
the monitoring tools available in the industry provide specific monitoring tasks which are limited for one
feature. Most monitoring tools provide interfaces to display network status, or server status. If
configurations of devices require alterations a separate tool or suit is required for configurations and
device management purposes. Majority of the network monitoring tools are standalone system which
does not enable network engineers and administrators to remotely monitor or examine network events and
statistics.

Commercial monitoring tools are very expensive and the cost adds up during the network
designing and implementation. However there is a selection of open source network monitoring tools.
They also provide similar functionality and features as the commercial tools. The current monitoring
systems completely monitor the data network as a whole. However monitoring of physical server details
and resources are limited or not provided.Systems and network administrators however require tools that
are affordable to ease the management of computer networks (Martin). This research culminated into the
development of a Network Monitoring Tool for network administrators. The tool has a light foot print and
does not require the use of a database. It instead monitors network traffic as well as available network
adapters on a networked computer. T The proposed system tends to develop an application that can
monitor and manage a network over LAN and wireless connection. The system development will come to
reality through the use ofvisual C# and Mysql database.

STATEMENT OF THE PROBLEM

There are rules and regulations which users agree to abide by, but if they choose to ignore these,
they are able to do so. This also makes the network more vulnerable to viruses and advanced hackers of
the 21stcentaury. Inappropriate traffic can slow the network down, or even bring it to a complete
shutdown, causing frustration to legitimate users of the network. Illegal traffic, such as pirated movies
and music, can get both the college and the individual users into serious litigation. To an extent network
abuse is inevitable. In most organizations and especially academic institutions.
AIM AND OBJECTIVES
This project aimed at implementing an application in monitoring and managing a network. The
objectives of the research are stated below.

1 To maintain the effectiveness of the system when compared with any other way of monitoring
network.
2 To develop a system that monitor and give details of the host.
3 To develop a system that views attitudes of network users and administrators

SCOPE OF THE STUDY

As network management has a wider scope in which all the sense cannot be covered in this project work
as it will only focus on managing and monitoring of Local area network. So as not to mismanage the vital
information stored on the local server.
LITERATURE REVIEW

REVIEW OF PAST WORK

There are many systems for remote monitoring and control designed as commercial products or
experimental research platforms. Internet Based Monitoring Internet monitoring is one of the common
approaches for remote monitoring. Many researchers have worked in field of Internet based remote
monitoring.

Zhiyong (2003) emerged network worms and attacks have distributive characteristic, which can spread
globally in a very short time. Security management crossing network to co-defense network-wide attacks
and improve efficiency of security administration is urgently needed. This paper proposes a hierarchical
distributed network security management system (HD-NSMS), which can centrally manage security
across networks. First describes the system in macrostructure and microstructure; then discusses three key
problems when building HD-NSMS: device model, alert mechanism and emergency response
mechanism; at last, describes the implementation of HD-NSMS. The paper is valuable for implementing
NSMS in that it derives from a practical network security management system (NSMS).

Rafiullah, SarmadUllah, Rifaqat and Muhammad (2003,pp..1-12Alkar and Buhur, (2005) implemented
Internet based wireless flexible solution where home appliances are connected to slave node. The slave
nodes communicate with master node through RF and master node has serial RS232 link with PC server.
The nodes are based on PIC 16F877 µc. PC server is formed of a user interface component, the database
and the web server components. An Internet page has been setup running on a Web server. The user
interface and the Internet front end are connected to a backend data base server. The control of devices is
established and their), Large organizations always require fast and efficient network monitoring system
which reports to the network administrator as soon as a network problem arises. This paper presents an
effective and automatic network monitoring system that continuously monitor all the network switches
and inform the administrator by email or SMS when any of the network switch goes down. This system
also point out problem location in the network topology and its effect on the rest of the network. Such
network monitoring system uses smart interaction of Request Tracker (RT) and Nagios software in Linux
environment.

This scenario is common place in academic institutions because of the aforementioned reasons. This was
the inertia for this research and project development. Improved techniques for building a better network
monitoring system were devised and investigated. A new network monitoring tool was built from the
ground up using new technologies. Some of the new techniques devised were integrated into the new
network monitoring tool to improve usability and usefulness of a network monitoring tool from the
perspective of network users and systems administrators.

According to Barry (2010, P:1-12), Network infrastructure is growing rapidly and as a result, network
traffic is expanding beyond the point where traditional approaches can still be used to efficiently detect
anomalies. To aid in overcoming this issue, modern approaches to tra ffic monitoring need to be
investigated. One particularly good way of representing large amounts of network traffic is with the aid of
data visualization. InetVis, an interactive tool for visualization, was developed in 2005 by J.P van Riel
under the security and networks research group at Rhodes University. It displays traffic in a 3
dimensional cube which uses source and destination addresses as well as port numbers for each
dimension. Users have the ability to navigate through this 3D space through the use of a simple interface.

Internet Based Monitoring

Internet monitoring is one of the common approaches for remote monitoring. Many researchers have
worked in the field of Internet based remote monitoring. They developed home gateway system for
interconnecting home network consisting of IEEE 1394 AV network and X10 power line home
automation network with Internet. This provided remote access functions from Internet for digital AV
appliances like Digital Video Camera, Digital VCR connected to IEEE 1394 network and home
appliances like TV, desk lamp, electric fan connected to X10 controller. They developed Java based home
automation system via World Wide Web. The home appliances were controlled from ports of embedded
system board connected to PC based server at home (Dariusz, et al 2016). They implemented Internet
based wireless flexible solution where home appliances are connected to slave node. The slave nodes
communicate with master node through RF and master node has serial RS232 link with PC server. The
nodes are based on PIC 16F877 μc. PC server is formed of a user interface component, the database and
the web server components. An Internet page has been setup running on a Web server. The user interface
and the Internet front end are connected to a backend data base server. The control of devices is
established and their condition is monitored through the Internet. They used X10 controller interfaced
through serial port to PC server for control of devices. The Common Gateway Interface (CGI) is used to
interface between the browser and the X10 protocol via http connection. The server executes CGI
programs in order to satisfy a particular request from the browser, which expresses its request using the
http.
They developed model of web services based email extension for remote monitoring of embedded
systems which integrates web services into emails. It uses a general purpose email messaging framework
to connect devices and manipulators. This low cost model fits for systems with low connection
bandwidth, small data transportation volume and non real- time control, e.g., monitoring of home
appliances and remote meter-reading (Englander, 2013). They developed a system for controlling home
electrical appliances over the Internet by using Bluetooth wireless technology to provide a link from the
appliance to the Internet and Wireless Application Protocol (WAP) to provide a data link between the
Internet and a mobile phone. However, technical details relating controller are not revealed.

GSM-SMS Based Monitoring

With the wide spread use of cellular networks, this approach is also popular when small amount of data
is to be transferred through the network. Researchers using this approach especially in medical field
have carried out extensive work. They describe a remote monitoring system based on SMS of GSM. The
system includes two parts which are the monitoring center and the remote monitoring station. The
monitoring center consists of a computer and a TC35 GSM communication module. The computer and
TC35 are connected by RS232. The remote monitoring station includes a TC35 GSM communication
module, a MSP430F149 MCU, a display unit, various sensors, data gathering and processing unit. They
developed a tele-monitoring system, based on short message service (SMS), to remotely monitor the
long-term mobility levels of elderly people in their natural environment. Mobility is measured by an
accelerometer-based portable unit, worn by each monitored subject. The portable unit houses the
Analog Devices ADuC812S microcontroller board, Falcon A2D-1 GSM modem, and a battery-based
power supply (Jabhera, et al 2011). Two integrated accelerometers are connected to the portable unit
through the analog inputs of the microcontroller. Mobility level summaries are transmitted hourly, as an
SMS message, directly from the portable unit to a remote server for long-term analysis. Each subject’s
mobility levels are monitored using custom-designed mobility alert software, and the appropriate
medical personnel are alerted by SMS if the subject’s mobility levels decrease. They proposed a system
for early diagnosis of hypertension and other chronic diseases. The proposed design consists of three
main parts: a wrist Blood Pressure (BP) measurement unit, a server unit and a terminal unit. Blood
Pressure is detected using data acquired by sensors intelligently using DSP microchip. The data is then
transmitted to the remote server unit located at Community Healthcare Centers/Points (CHC/P) by using
Short Messaging Service (SMS), and notification information is sent to the terminal unit to inform users
if patient’s BP is abnormal.
 RESEARCH METHODOLOGY

Basic theory used to check the availability is using ICMP messages (PING) to check the availability of
host. But there are few ways to do this in C# language.
Using the above method to check the availability and if the host is available, some information can be
gathered about it using SNMP messages. Basically SNMP GET request was used. Using SNMP GET it
fetch the values of certain OIDs (Object Identifiers). Each entity in the network and the system has its
own OID.

ANALYSIS OF DATA AND THE EXISTING SYSTEM

Cacti (cacti.net, 2015) is a network monitoring system designed for drawing time-series graphs of
performance data on a monitored network. It typically draws a different graph for each monitored data
source. A graph can be drawn from multiple data sources, but requires a suitable template created using
the cacti format. The configuration system is entirely web based and there is no provided method for
performing bulk configuration. The additional effort required to update the network model in Cacti often
discourages the user from monitoring everything. Cacti also does not include an event detection system or
a notification system and therefore is usually used to supplement another network monitoring system by
providing historical graphs. These provide more visibility and insight as to why an event may have
triggered in another monitoring system.

PROBLEM OF THE EXISTING SYSTEM

It is very typical in every industry that as its client base continuous to grow, the industry faces increased
number of problems in monitoring their operations. In network monitoring aspect the problem may
include:

i. Poor network strength

ii. Network hacking
iii. No signal when the network is down etc.
3.4 DESCRIPTION OF THE PROPOSED SYSTEM
SNMP (Simple network management protocol) is the industry standard for data collection in a network
monitoring system. SNMP is commonly configured in a centralized architecture, where a single SNMP
collector collects data from every device on the network.

The subject of monitoring by the system will be the network devices (ie. routers and switches).

The system will be maintained on a single machine, on which will be running:

 Netmon software - to monitor critical IT infrastructure components, including system metrics,

network protocols, applications, services, servers, and network infrastructure
 SNMPTRAPD - an SNMP application that receives and logs SNMP TRAP andINFORM
messages.
 SNMPTT - an SNMP trap handler for use with the NET-SNMP/UCD- SNMP snmptrapd
program. Received traps are translated into friendly messages using variable substitution. Output
can by to MySQL (Linux/Windows) or any other ODBC database.
 Database Server.

3.5 ADVANTAGES OF THE PROPOSED SYSTEM

The proposed application for network management and monitoring system is intended to detect numerous
attacks. Since it is not possible to design a network management and monitoring system for every type of
attack, the system be will designfor denial of service attack, Web server attack and buffer overflow attack.
Some of it advantages are:
i. to design low cost intelligent embedded system based remote monitoring system .

ii. to provide flexibility for remote monitoring.

iii. to incorporate alternative mechanism for communication when messaging fails.

iv. to implement a simple embedded system as a proof of concept.