ABSTRACT
Mobile ad-hoc networks are multi hop wireless networks comprising of a collection of wireless mobile nodes configured to communicate among each other without the aid of an existing infrastructure. They are free to move randomly and organize themselves arbitrarily. The network’s wireless topology may change rapidly or unpredictably. As a result, the link becomes unreliable leading to poor communication.This research work presents the prediction of link reliability in a wireless Mobile Ad-hoc Network (MANET) with path loss effect. The link reliability was investigated using the Weibull distribution function due to its flexibility, simplicity and reliability and was simulated using the developed MATLAB 2007Rb Graphic User Interface (GUI). The free space and two ray ground path loss models were used to study the path loss effects on the link reliability. The results obtained were validated by comparing the performance of the model with existing results. It was noticed that the developed model performed better by 7% for coverage area, 2% for range, 0.3% for number of mobile nodes at the point where attenuation factors were at maximum point. This proves the authenticity of the developed model. The network reliability was 98.70% when the bandwidth was 3000 MHz and propagation parameters were δ = 1.0 and γ = 1.0 respectively. In the case of link reliability and signal to interference plus noise ratio (SINR), 99.99% of link reliability was attained which occurred when the SINR was 0.9 and propagation parameters were δ = 1.0 and γ = 1.0 respectively. The effect of SINR and bandwidth on the link reliability and the development of a user dependent Matlab GUI for the link reliability prediction are significant contributions of this research.
TABLE OF CONTENTS
TITLE PAGE ii
DECLARATION iii
CERTIFICATION iv
DEDICATION v
ACKNOWLEDGEMENT vi
ABSTRACT ix
LIST OF FIGURES xiii
LIST OF TABLES xvi
LIST OF ABBREVIATIONS xvii
CHAPTER ONE: INTRODUCTION
1.1 Background 1
1.2 Statement of Problem 2
1.3 Aim and Objectives 3
1.4 Motivation and Justification 3
1.5 Methodology 3
1.6 Dissertation Organization 4
CHAPTER TWO:LITERATURE REVIEW
2.1 Introduction 5
2.2 Review of Fundamental Concepts 5
2.2.1 Wireless Networks 5
2.2.1.1 Characteristics of Mobile Ad-hoc Networks (MANETs) 11
2.2.2 Reliability of Telecommunications Systems 12
2.2.3 Reliability Distributions 13
2.2.3.1 Statistical Distributions 14
2.2.4 Reliability System Modeling Techniques 18
2.2.4.1 The Link Reliability Model 20
2.2.4.2 Reliability of Series and Parallel connections 23
2.2.4.3 Link Reliability 24
2.2.5 Path Loss 26
xi
2.3 Review of Similar Works 34
CHAPTER THREE:MATERIALS AND METHODS
3.1 Introduction 43
3.2 Development of Signal Propagation Model 43
3.3 Development of Network Reliability Model 44
3.4 Development of Node Mobility Model 45
3.5 Development of Node Reliability Model 47
3.6 Development of Link Reliability Model 48
3.7 Implementation of the Link Reliability Algorithm 51
3.8 The Input Parameters 52
CHAPTER FOUR:RESULTS AND DISCUSSIONS
4.1 Introduction 55
4.2 Results 55
4.2.1 The Effect of Coverage Area on the Link Reliability 55
4.2.2 The Effect of Range on the MANET Link Reliability 56
4.2.3 The Effect of Communication Duration on the MANET Link Reliability 57
4.2.4 The Effect of Number of Nodes on the Link Reliability 58
4.3 Validation 59
4.3.1 Performance evaluation using link reliability and coverage area (m2) 61
4.3.2 Performance Evaluation using Link Reliability and Range(m) 63
4.3.3 Performance Evaluation using Link Reliability and Time(s) 66
4.3.4 Performance Evaluation using Link Reliability and Number of Nodes 70
4.3.5 Effect of Bandwidth on Link Reliability 73
4.3.6 Effect of SINR on Link Reliability 75
CHAPTER FIVE: CONCLUSION AND RECOMMENDATION
5.1 Introduction 77
5.2 Conclusion 77
5.3 Significant Contributions 78
5.4 Limitations 78
xii
5.5 Recommendations for Further Work 79
REFERENCES 80
Appendix A
Reliability Parameters 84
Appendix B
The MATLAB GUI Main Function 86
Appendix C
The Link Reliability Main Function 97
xiii
CHAPTER ONE
INTRODUCTION
1.1 Background
In recent past, networks have been developing at a faster pace by connecting a large number of devices from servers to micro- devices embedded in objects (Wazwaz, 2005). These devices provide services that can support a variety of applications like: environment monitoring, medical services, military applications, disaster recovery (fire, flood, earth quake etc.,), law enforcement, etc. There are many instances where the network supporting such applications are mobile wireless networks, whose necessary infrastructure support is installed either in a permanent or ad hoc manner (Wazwaz, 2005). For instance, MANET serves in situations where setup of the infra- structure based network may not be feasible or desirable because of the coverage limitations, network failures, congestion, etc. Telecommunications system reliability is of great interest and major importance to developers and customers. The ultimate goal is in improved service quality, throughput, reduced cost, delay,etc (Erickson et al.1990). The special feature of MANET such as self-organizing ability and high mobility of Mobile Nodes (MN) gives a dynamic topology. But strong linking of each MN within the coverage area distinguished it from other types of networks. New scenarios are observed where users can benefit from anywhere and at any time as a result of growth experienced in the area of MANET (Haboub & Ouzzif, 2012). The node failure in a MANET may be due to variety of reasons, for example, low transmission range, out of coverage area, atmospheric effects, physical obstacles, and limited battery life time. In other words, MN time-to-failure follows some statistical distribution. Similarly, the links between the nodes areas summed to have failed when either the distance between the nodes are beyond the specified transmission range of each node, or it may be due to mobility, interference, highly
2
dynamic topology and/or congestion of wireless links. These characteristics make modeling and reliability analysis of such networks a challenging and formidable task. In general, any system can be modeled using random graphs or probabilistic graphs with each user/terminal termed as a node and the links between them as edges. Based on the number of communicating nodes, the network reliability for infrastructure based networks are defined as Two Terminal(TTR), All-Terminal (ATR), K-Terminal (KTR) and All operational Terminal (AoTR) Reliabilities (Padmavathy & Chaturvedi, 2013). The probability that a specified pair of nodes (the source, s and the sink, r) with known success/failure probabilities of its elements, remain connected by a path created by the operating nodes and links is defined as TTR. The probability that every node be able to communicate with every other node in the network is ATR. Similarly, the probability that at least k designated nodes can communicate with each other is KTR. The probability that all the operational nodes can communicate with each other is termed as AoTR.
1.2 Statement of Problem
Transmitting high data rates is a big problem due to the link quality which diminishes rapidly as the number of hops increase (Paz, 2010). Since real time applications are delay sensitive, they require a good reliability and quality of service analysis. Despite the extensive work done in the prediction and analysis of the reliability of ad-hoc networks, the networks are still unreliable. This necessitates the need to explore more efficient means of predicting the link reliability in a wireless mobile ad hoc network due to path loss effects by using Weibull distribution. The bandwidth, Signal to Interference plus Noise Ratio (SINR), coverage area, transmission range, number of nodes and communication duration will be used as parameters of evaluation.
3
1.3 Aim and Objectives
This work is aimed at predicting the link reliability in a multi hop wireless mobile ad-hoc network due to path loss effect using Weibull distribution. The objectives are as follows:
1. Development of a Matlab Graphic User Interface (GUI) based algorithm for prediction and simulation.
2. Carrying out a wide variety of simulations using the Matlab GUI based program in order to demonstrate the effect of changes in network characteristics on link reliability.
3. Evaluation and comparison of the link reliability of the network with an existing research work.
1.4Motivation and Justification
The unpredictable dynamic nature of MANETS leads to poor link reliability of the network. Also, the environmental conditions which cause link path loss do not only affect the effective and efficient operation of the network but the reliability of the links as well. Finding the solution to these problems is the motivating factor of this research work. Though many researchers have already worked on these problems to improve link reliability, there is still room for improvement which justifies this research work as is evident in the contribution achieved.
1.5 Methodology
The methodology adopted in carrying out this research is as follows:
1. Development of :
i. Signal propagation model based on the combined Free Space (FS) and Two Rays to
Ground (TRG) path-loss model;
4
ii. Network reliability model; iii. Node mobility model based on Random Way Point Mobility (RWPM); iv. Node reliability model using Weibull distribution; v. Link reliability model for TTR based on the FS-TRG model 2. Implementation of the proposed algorithm using GUI in Matlab 3. Simulating various scenarios of the network using the Matlab GUI based script 4. Validationof the results of the developed model using Padmavathy &Chaturvedi, (2013).
1.6 Dissertation Organization
The general introduction has been presented in chapter one. The rest of the chapters are as follows: Detailed review of similar research works and relevant fundamental concepts about pathloss, link reliability, Weibull distribution and RWPM is carried out in chapter two. In depth approach and relevant mathematical models describing the individual link reliability matrix are presented in chapter three. The result, its analysis,and discussion are presented in chapter four. Conclusion and recommendation are discussed in chapter five. Quoted references and appendices are also provided at the end of this dissertation.
5
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
