TECHNICAL REPORT ON

THE PROBLEMS OF BANDWIDTH ALLOCATION IN COMMUNICATION SYSTEM

BY

OJO PRECIOUS OLUWASEYE

(Matric No. 13/ENG02/018)

SUBMITTED TO THE

DEPARTMENT OF COMPUTER & ELECTRICAL ENGINEERING

COLLEGE OF ENGINEERING, AFE BABALOLA UNIVERSITY,

ADO-EKITI, NIGERIA

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF

BACHELOR OF ENGINEERING DEGREE (B.ENG) IN COMPUTER ENGINEERING

NOVEMBER 2016
 CERTIFICATION

This is to certify that write-up was put together and presented by OJO PRECIOUS
OLUWASEYE with matriculation number 13/ENG02/018, Department of Computer
Engineering, College of Engineering, Afe Babalola University, Ado-Ekiti, Ekiti State, Nigeria.
 DEDICATION

I dedicate this first to The Almighty God for giving me the strength, wisdom and grace to write
this report. To my Parents: Mr. & Mrs. Ojo for their support & encouragement.
 ACKNOWLEDGEMENT
I give thanks to God Almighty for his guidance over my life throughout the period of my
research in putting together this report. I also give thanks to my Parents: Mr. & Mrs. Ojo. May
God in His infinite mercies grant all your heart desires.
 ABSTRACT

Different broadband services require different amounts of bandwidth and have different
priorities. For example, a connection for visual communications will in general require more
bandwidth than one for data communications, and a voice connection will in general be of higher
priority than either a data or a video connection. In response to these varied demands, the
network designer may choose to assign different amounts of bandwidth to different types of
traffic. The motivation for such an approach stems from the desire to support different kinds of
multimedia services with a reasonable level of performance and without letting the demand from
any one type shut-out other types of services. The challenge for the designer is to come up with
techniques that are able to balance the needs of the various applications with the need of the
system to accommodate as many connections as possible. This task of providing guaranteed
quality of service with high bandwidth utilization while servicing the largest possible number of
connections can be achieved through a combination of intelligent admission control, bandwidth
reservation and statistical multiplexing.
 TABLE OF CONTENT

Title PageI

Certification...II

DedicationIII

Acknowledgment. IV

Abstract..... V

Table of content........VI

CHAPTER ONE: DEFINITION OF BANDWIDTH AND BANDWIDTH

ALLOCATION.7

CHAPTER TWO: BANDWIDTH ALLOCATION ISSUES .8

CHAPTER THREE: SOLUTIONS TO SOLVING THE PROBLEM.9

REFERENCES12
CHAPTER ONE

BANDWIDTH

Bandwidth allocation is the process of assigning radio frequencies to different applications. The
radio spectrum is a finite resource creating the need for an effective allocation process. In the
United States, the Federal Communications Commission or FCC has the responsibility allocating
the portions of the spectrum, or bands, to various industries. For example, the FCC shifted the
location of television on the spectrum, to open up more space for mobile applications. Different
bands of spectrum are able to transmit more data than others, and some bands of the spectrum
transmit clearer than others. These bands are of critical importance for companies that intend to
operate a business involving wireless communications.

Bandwidth is like lanes on a highway. It allows for more traffic to go through at once, while still
retaining high speeds. A lone car can speed down a single lane road, in the same way a lone
network request can function with low bandwidth. However, often times many requests must go
through a network simultaneously and a low bandwidth connection will severely slow speeds. A
high bandwidth connection can service many requests and users without sacrificing precious
speed.

Bandwidth limitations
The exponential increase in mobile data traffic during the decades of the 1990s and 2000s has led
to the massive deployment of wireless systems. As a consequence, the limited available RF
spectrum is subject to an aggressive spatial reuse and co-channel interference has become a major
capacity limiting factor.
Therefore, there have been many independent warnings of a looming "RF spectrum crisis"as the
mobile data demands continue to increase while the network spectral efficiency saturates despite
newly introduced standards and great technological advancements in the field.
It is estimated that by 2017, more than 11 exabytes of data traffic will have to be transferred
through mobile networks every month.
CHAPTER TWO
Common Causes of Bandwidth Issues
Bandwidth Issues can almost always be traced to one or two specific activities. These activities
almost always have two characteristics: large amounts of data, and extended duration. Common
activities causing bandwidth problems are:
Watching videos from Internet (YouTube)
Large file transfers between computers (greater than 100 megabytes in size)
Constant stream of data (surveillance footage from security cameras)
Downloading files from internet
All of the above can contribute greatly to bandwidth issues in a network, and should be done only
when there is only light network traffic. Large file transfers or data streams within a network
should be placed on a separate network, in order to avoid bottlenecking other users.

Bandwidth Allocation in Wireless Networks for Multiresolution VBR Video Traffic

Supporting real-time VBR video along with voice and data over bandwidth-constraint networks
continues to be formidable problem. The difficulty arises because VBR video is unpredictably
bursty and because it requires performance guarantees from the network. While resource
reservation schemes work best for CBR traffic, there is no consensus on which strategy should be
used for VBR traffic. On one hand, since real-time VBR traffic is delay sensitive, a resource
reservation scheme seems to be the right choice, on the other hand, because VBR video is bursty,
if resources are reserved according to peak rates, the network may be under- utilized if the peak-
to-average rate ratios are high. These two opposing characteristics have resulted in a common
belief that it is unlikely that performance guarantees can be provided to such bursty sources with
very high network utilization. This is the problem that is addressed in this work, that is, can
performance guarantees be provided to VBR video without significantly under-utilizing the
bandwidth and can this be done in conjunction with minimizing the maximum blocking probability
for voice and data connections?
CHAPTER THREE
SOLUTIONS TO THE PROBLEM
Our solution to this problem consists of three parts:
(1) use a multi-resolution joint source-channel video codec,
(2) provide connection lifetime reservation for real-time video connections with optimum
bandwidth utilization, and
(3) partition the available bandwidth in a manner that ensures that the maximum blocking
probability for voice and data traffic is minimized.

From a connection's perspective, we advocate the use of a multi-resolution subband video codec
(or a segmented region- based video codec) . We suggest reserving the peak bandwidth for the
primary subband (or region) while letting the secondary and tertiary subbands (or regions) compete
dynamically for bandwidth. A potential problem with this approach is that most of the time the
actual amount used by the primary subband is far below the amount reserved for the peak. We
concentrate on this aspect and develop techniques to completely use all of the reserved bandwidth.

From the system's perspective, we develop a simple yet effective algorithm that partitions the
available bandwidth in a manner that minimizes the maximum blocking probability for voice and
data connections while providing guaranteed QoS to VBR video connections. It should be noted,
that even when the distribution of the different traffic types is given, finding the optimal
partitioning of the bandwidth is a very difficult task, and for the general case can be modeled by
an NP-complete graph coloring problem. The intractability of finding the optimum is present
already in the simplest situation when the traffic consists of voice connections only and the
statistics of the offered traffic are completely known. However, the problem becomes even more
difficult when the wireless network is carrying integrated non-homogeneous traffic, a situation
occurring naturally in the case of wireless multimedia networks. In this case estimating the
blocking probability of connections and its application in resource allocation strategies are further
complicated for two fundamental reasons:
Although there are methods for computing blocking probabilities for integrated systems under
specific statistical assumptions (e.g. multirate Poisson models), there are no simple closed
formulas that can easily be applied to optimizing resource allocation.

It is realistic to expect that traditional statistical assumptions will not describe the traffic load
precisely. Therefore, it is injudicious to make concrete assumptions based on any advance
knowledge regarding the detailed statistical properties of traffic in a wireless multimedia network.
This calls for a bandwidth management methodology that works under incomplete information
and does not critically depend on specific statistical assumptions.
We propose a solution for the allocating transmission resources among different traffic types under
incompletely known conditions. This when combined with our proposal for using multiresolution
video and bandwidth reservation with intra-frame statistical multiplexing, our solution has the
following main properties:

It provides guaranteed QoS for on-going real-time visual communication sessions. This guarantee
does not come at the expense of wasting bandwidth since all of the reserved bandwidth is used up
through intelligent statistical multiplexing.

It is robust and insensitive to statistical assumptions, as it depends only on the average rates of the
aggregated flow of traffic types, but not on detailed statistics of the traffic mix and of the arrival
process. From practical viewpoint, this insensitivity is highly advantageous, since the detailed
statistical information is typically unavailable or uncertain.

The resulting allocation is based on minimizing a bound on the blocking probabilities that is proven
to be asymptotically optimal. The optimality is also important as it signifies that for large systems
it is sufficient to know aggregated flow rates, as the detailed knowledge of the traffic mix would
not significantly contribute to achieving smaller loss.
It is counter-intuitive in the sense that the allocation is different from the naive solution where the
bandwidth is allocated proportionally to the load. If only the per traffic type average load is known,
one could easily think that the best we can do is to allocate the channels in proportion to the load.
It is surprising, as we prove and demonstrate, that by allocating the bandwidth in a different way
we can obtain better results for each traffic type. This is why we call the algorithm Smart Allocate,
since it is better for every traffic type than the apparently fairer load-proportional allocation.
REFERENCES
1. P. Bahl and I. Chlamtac, Strategies for Transmission of Compressed Video Over Error-Prone
Radio Channels, Proceedings of the 3rd International Workshop on Mobile Multimedia
Communications, Princeton, New Jersey (Sept. 1996)
2. K. W. Ross, Multiservice Loss Models for Broadband Telecommunication Networks, Springer,
1995