COMSATS University Islamabad

Abbottabad Campus
Department of Computer Science
Assignment # 02 & 03

Q # 1: Assume any type of network, along with it’s QoS issues and apply the
concept of traffic engineering in it. And then show how its overall performance has
affected?

Traffic Engineering:
Traffic engineering is a method of optimizing the performance of
a telecommunications network by dynamically analyzing, predicting and regulating the behavior
of data transmitted over that network. Traffic engineering is also known as tele-traffic
engineering and traffic management. The techniques of traffic engineering can be applied to
networks of all kinds, including the PSTN (public switched telephone network), LANs (local
area networks), WANs (wide area networks), cellular telephone networks, proprietary business
and the Internet.

Type of network that I choose is:

The type of network I choose is 4G network.

QoS issues in 4G network:

The user mobility, mobility management, integration and interoperability of diverse networks,
streaming multimedia-based services, and call admission control are some of the QoS issues for
future 4G technologies.

1. User Mobility:

User mobility in 4G eras users must get a convenient access to the services needed at any given
situation. In this context user mobility has become an important aspect in the design of 4G
wireless communication systems. The terminal mobility deals with mobility of users having a
single device. The session mobility deals with users in a public access network (PAN) having
multiple personal devices to provide a live session. The personal mobility concentrates on
provision of personalized operating environments for users along with user movements. In
personal mobility a service required by the user will be delivered instantaneously, irrespective of
the user’s location, device/device location, operator/provider domain, and type of network. So
unlike terminal mobility, session and personal mobility concentrate more on user movements
rather than terminal movements.

The disadvantages in the single-layer mobility approach that maintains strict layer modularity when
applied to the emerging wireless systems and argue that such an approach can only provide incomplete
mobility functionality. They propose the multi-layer approach to explore contributions from multiple
layers for extended mobility functionality and improved performance.

The scheme uses mobile agents and signaling protocols to facilitate significant optimal usage of
the communication channels. This improves the system usability and allows accessibility of the
required services anytime globally. This framework suffers from considerable overhead cost

2. Mobility management:

The 4G mobility management includes moving networks, seamless roaming and vertical
handoff. In 4G scenario users expect to be connected to the Internet from “anywhere” at
“anytime”, in fixed wireless locations or while on the move, provided that any available access
network can be accommodated. For doing so, mobile networks may be multihomed i.e. having
multiple points of attachment to the Internet. Moreover a user may have more than one mobile
device, say a mobile phone, a laptop and a personal digital assistant (PDA). Each of these
devices is likely to have multiple network interfaces that enable them to interconnect with each
other as well as with other networks. These devices moving with the user together are an
example of a small scale mobile network. The access networks deployed on public
transportations such as ships, trains, buses and aircrafts are examples of mobile networks at a
larger scale.

3. Integration and interoperability of diverse networks:

The 4G systems intend to facilitate the seamless integration and interoperation of a broad range
of existing systems like satellite broadband, 3G systems, wireless local loop, fixed wireless
access systems, WLAN, WiMAX, digital video broadcasting (DVB), and multimedia
broadcasting multicast service (MBMS). The intention of such open wireless architectures is to
provide unhampered connectivity, full broadband access, with global roaming, and global
Internet/data/voice. However, it is difficult to achieve perfect QoS and user controlled services
due to the diverse nature of the constituent access technologies in terms of varying bit rates,
bandwidth allocation, channel characteristics, fault tolerant levels and handoff management
mechanisms.

4. Streaming multimedia-based services:

The aim of the 4G wireless multimedia communications is to provide efficient transmission

 of streaming data for video applications such as telemedicine, multimedia video
conferencing, three dimensional (3D) virtual reality and virtual navigation. For such
applications the constraints like scarce system resources, high QoS, bandwidth requirements,
variation in delays and packet losses need to be overcame. The bursting and streaming video
services are the popular types of services in 4G systems. The memory requirement for
bursting is much larger than streaming whereas streaming should use less bandwidth as it is
possible to encode multiple streams within the same file which can be transmitted using the
available bandwidth. So a new streaming video application scheme is required to be
implemented for optimum use of available bandwidth using limited available memory. The
choice of an appropriate protocol is important in this context. The user datagram protocol
(UDP) and transport control protocol (TCP) are the two important transport layer protocols
for video streaming. However UDP suffers from acute congestion related problems, which
may be more conveniently handled by the TCP.

5. Call admission control (CAC):

Efficient call admission control algorithm is required for 4G wireless networks due to the diverse
QoS requirements for multimedia applications and the presence of different wireless access
technologies. The call admission control algorithm must be able to handle vertical handoff as
discussed in section 3.2. It must be able to accommodate different types of users and applications
with different QoS requirements. The system utilization and QoS performance can be improved
for multimedia applications by adjusting the bandwidth allocation depending on the state of the
network and users’ QoS requirements. At both call-level and packet-level, QoS needs to be
considered to design call admission control algorithms so that not only the call dropping and call
blocking probabilities, but also the packet delay and packet dropping probabilities can be
maintained at the target level.

Traffic Engineering in Fourth Generation Wireless Network:

Extension to the Multiprotocol Label Switching (MPLS) traffic engineering in Fourth Generation
(4G) composite networks. In a 4G composite network users should be able to select among
various radio access technologies that can individually or collectively satisfy the users’ capacity
and quality of service requirements. The extension provides the ability to diverge traffic flows
away from a path that cannot satisfy the QoS requirements of a flow. When the QoS
requirements of a traffic flow exceed a predefined threshold, the extension will distribute the
flow among multiple paths based on the cost of the traffic distribution and the effectiveness of
Label Switching Routers (LSPs).

The development effort toward the integration of various wireless access technologies that will
form the foundation of Fourth Generation (4G) systems is called “Beyond 3G Systems” or the
composite radio concept. The main theme of the composite radio concept is the synergy of the
services of the various service providers with different access technologies in order to provide
more efficient services to the end-users than the individual service providers could do
individually. Fourth Generation (4G) wireless systems will offer personalized services for the
end-users by making use of a broad range of access technologies and terminal devices. Many of
these devices will be invisible and embedded in the users’ environment. One of the most
challenging design goals of 4G systems is the efficient collaboration between existing
heterogeneous access technologies and communications protocols in addition to the development
of new radio interfaces and technologies.

The procedure to switch between wireless networks is called handoff. A new handoff protocol
that can provide QoS in a composite network environment. We consider cases when the mobile
terminal cannot be granted the desired QoS for certain applications by the current service
provider due to the lack of network resources. We assume end-to-end IP communications
between mobile hosts. Our handoff mechanism, called Flow Distribution Protocol (FDP)
incorporates and extends both the RSVP and the MPLS. The QoS may be provided by more than
one link between the end users. When the traffic characteristics delay, jitter, packet loss) of a
newly requested or existing packet flow approach the capacities of the initial or current path, the
mobile terminal selects one or more new wireless connection to handoff the new or existing
traffic flow to other traffic flows that can provide the requested QoS. FDP identifies multiple,
loop-free, end-to-end paths between mobile terminals jointly satisfying the requested QoS. The
new paths’ selection is based on cost estimates of allocating or reallocating network resources to
accommodate the handoff and the effectiveness of the intermediate network nodes in reducing
the number of discarded packets, delay, and jitter in previous handoffs.

MPLS has emerged as an enabling technology for new public networks due to its support for
traffic engineering. MPLS does not replace IP routing, but will work together with existing
routing technologies to provide very high-speed data forwarding between Label-Switch Routers
(LSRs). A fixed format label is inserted in front of each data packet on entry into the MPLS
network. At each hop across the network, the packet is routed based on the value of the incoming
label and sent out on an outgoing interface with a new label value. The path that data traverses
through a network is defined by the transition in label values, as the label is swapped at each
LSR. Since the mapping between labels is constant at each LSR, the path is determined by the
initial label value. Such a path is called a Label Switched Path (LSP). At the entry to the
network, called ingress, each packet is examined to determine which LSP it should use based on
the destination address, the quality of service requirements, and the current state of the network.
The group of similar packets forwarded in the same way is known as a Forwarding Equivalence
Class (FEC). One or more FEC may be mapped to a single LSP.

New handoff protocol (FDP) in 4G composite radio networks that can provide QoS for mobile
terminals in multi-access wireless environments. The protocol is an extension of the
Multiprotocol Label Switching traffic engineering for wireless networks. We assume that user
equipment’s have a choice to select services provided by overlapping radio networks and short-
distance access technologies. FDP can identify multiple, loop-free, least cost paths between
mobile terminals and services that collectively satisfy the requested QoS while terminals move
across heterogeneous networks. FDP allows mobile terminals to receive the required QoS instead
of reducing their transmission rates in times of congestion and receiving insufficient QoS as in
other protocols. FDP can also learn over time to remember the performance of network nodes
along the paths in previous handoffs. Similar protocols start over again without remembering
previous successful, least cost handoffs, therefore these protocols will involve ineffective
network nodes again in path discovery, even though it has been statistically proven that they
cannot provide the desired QoS in an efficient way.

Reference:

https://www.riverpublishers.com/journal/journal_articles/RP_Journal_1550-4646_531.pdf

https://cseweb.ucsd.edu/classes/wi13/cse222A-a/lectures/222A-wi13-l13.pdf

Q#2 Discuss in detail how quality of service will play it’s role in Internet?

Role of QoS in internet :

QoS, or Quality of Service, is a set of mechanisms that ensures the performance of critical
applications, guaranteeing enough bandwidth for their correct operation. Its concept is based on
the idea that access should not be treated in an equal way, allowing prioritization of specific
traffic. Over the Internet, QoS is typically measured in terms of bandwidth, loss, delay, jitter, and
availability

By using it, professionals can change the way network packets are routed to the internet or other
networks, allowing existing resources to be used more efficiently. With QoS, packets are marked
to distinguish service types, and routers are configured to create separate virtual queues for each
application, according to their priority.

 Ensures that mission-critical applications always have the necessary resources for their
operation;

 Enables administrators to better manage internet resources;

 Reduction of costs, through the efficient use of the internet resource, postponing
investments in the expansion of links;

 Minimizes the impact of concurrent non-work-related traffic;

 Improves user experience, ensuring better response times, according to the priority of
 access.

Since different applications (e.g. telephone, e-mail, surveillance video, etc.) may be using the
same IP network, there is a need to control how network resources are shared to fulfil the
requirements of each service. One solution is to let network routers and switches operate
differently on different kinds of services (voice, data, and video) as traffic passes through the
network. By using Quality of Service (QoS), different network applications can co-exist on the
same network without consuming each other’s bandwidth.
The term, Quality of Service, refers to a number of technologies such as Differentiated Service
Code Point (DSCP), which can identify the type of data in a data packet and so divide the
packets into traffic classes that can be prioritized for forwarding. The main benefits of a QoS-
aware network include the ability to prioritize traffic to allow critical flows to be served before
flows with lesser priority, and greater reliability in a network by controlling the amount of
bandwidth an application may use and thus controlling bandwidth competition between
applications. Pan–tilt–zoom (PTZ) traffic, which is often regarded as critical and requires low
latency, is a typical case where QoS can be used to guarantee fast responses to movement
requests. However, the prerequisite for the use of QoS within a video network is that all
switches, routers and network video products must support QoS.
Quality of Service (QoS) is a basic functionality that the Internet needs to be able to cope with
when the applications increased in number and varied in use. Internet QoS issues concerning the
Internet involves many different areas: end user QoS, ISP network QoS, backbone QoS.
QoS management is used to control traffic within the QoS enabled ISP network. It administer
users profile, the load status of each ingress/egress node and applications flow. Based on these
information QoS management may cluster (group) users into different service level, assign the
application flow different priorities.
In order to work with QoS management, the ISP network has some kind of "awareness"
functions like
• User awareness,
• Traffic awareness, and
• Application awareness
User awareness means that the ISP network should cluster different user groups. A business user
may, for example, be provided with a high level of service, regardless of ISP traffic load.
Traffic awareness is based on the analysis of IP flow. A IP flow can be checked by its IP header -
IP address, its TCPIUDP header - port number and SYNIFIN bit. QoS enabled servers on
UNI/SNI side may than make intelligent decision which kind service the IP flow can observe:
with low delay or no loss of data.
Application awareness is the way to classify applications. A HTTP application may involve real-
time or non-real-time applications. The traffic awareness cannot differentiate these applications.
A higher layer knowledge is necessary here to optimize the usage of network resource.
Reference:
https://link.springer.com/content/pdf/10.1007/978-0-387-35522-1_27.pdf

https://www.comparitech.com/net-admin/what-is-qos/

https://ostec.blog/en/perimeter/qos-and-its-benefits