This study is aimed at the development of a mechanism for black hole attack mitigation in Wireless Sensor Network (WSN) using a modified Bacterial Foraging Optimization Algorithm (BFOA). A total of 200 randomly generated bacterial sensor nodes with a communication range of 20m were deployed in a 100mx100m network coverage area, consisting of four base stations. The radii 20m, 30m and 40m were chosen for the black hole region. The algorithm was implemented in MATLAB R2015b. In all tests carried out, the results obtained at 40m radius showed the effect of the black hole attack better than those at 20m and 30m. Successful packet delivery probabilities of 83.52%, 95.78%, 97.26% and 99.78% respectively were achieved at 40m radius for one, two, three and four base stations respectively. Significant reduction in false positive was observed when the base stations were increased. A negligible value of about 0.003% false positive was observed with four base stations using 40m radius of black hole region. Average delivery times of 31sec, 37sec, 43sec and 49sec were achieved at 40m radius for one, two, three and four base stations respectively. The times indicated that the routing complexity increased as the number of base stations increased. The performance of the modified BFOA based method showed packet delivery probability improvement of 5.48%, 9.67%, 0.18% and 1.01% over the standard BFOA based method as the base stations were increased from one to four respectively. As the base stations were increased to five, six, seven and eight, successful packet delivery probabilities of 99.39%, 99.69%, 99.79% and 99.82% respectively were achieved using 40m radius of black hole region. The trend observed for packet delivery showed that optimum efficiency is achieved with quadrant placement of base stations. The results obtained indicated that optimal placement of the base stations minimized the effect of black hole attack and ensured successful packet delivery.
TABLE OF CONTENTS
Cover Page i Title Page ii Declaration iii Certification iv Dedication v Acknowledgement vi Abstract vii Table of Contents viii List of Figures xii List of Tables xiii List of Abbreviations xiv CHAPTER ONE: INTRODUCTION
1.1 Background 1
1.2 Statement of Problem 4
1.3 Motivation 5
1.4 Aim and Objectives 5
1.5 Justification 6
1.6 Methodology 6
1.7 Dissertation Organization 7
CHAPTER TWO: LITERATURE REVIEW 2.1 Introduction 9 2.2 Review of Fundamental Concepts 9 2.2.1 Wireless sensor networks 9 2.2.2 Characteristics and constraints of wireless sensor network 10 2.2.3 Security goals 11 2.2.4 Types of attack in wireless sensor networks 12 2.2.5 Bacterial foraging optimization algorithm 14 2.2.6 Modified step size 19 2.2.7 Objective function formulation 20 2.2.8 Performance metrics 21 2.3 Review of Similar Works 24 CHAPTER THREE: MATERIALS AND METHODS 3.1 Introduction 31 3.2 Initialization of BFOA and Network Parameters 31 3.3 Standard Bacterial Foraging Optimization Algorithm 33 3.3.1 Chemotactic and swarming Step 33 3.3.2 Reproduction step 34 3.3.3 Elimination and dispersal step 34 3.4 Development of the Modified Bacterial Foraging Algorithm 35 3.5 Optimized Positioning of Base Station and Detection of Black hole 38 3.6 Performance Evaluation 39
3.6.1 Packet delivery success and failure 39 3.6.2 False positive 40 3.6.3 Convergence speed 40 3.7 Comparison of the Results obtained from the Modified BFOA with the Results of the Standard BFOA 40 3.8 Extended Eight Base Stations 40 CHAPTER FOUR: RESULTS AND DISCUSSION 4.1 Introduction 41 4.2 Packet Delivery Success Results of Modified BFOA 41 4.3 Packet Delivery Failure Results of Modified BFOA 43 4.4 False Positive Results of Modified BFOA 44 4.5 Convergence Speed Results of Modified BFOA 46 4.6 Comparison between the Results of the Modified BFOA and the Standard BFOA 47 4.7 Result of Extended Eight Base Stations 53 CHAPTER FIVE: CONCLUSION AND RECOMMENDATION 5.1 Conclusion 57 5.2 Limitations 58 5.3 Significant Contributions 58 5.4 Recommendations for Further Works 59 REFERENCES 60 APPENDIX A
MATLAB script for the modified BFOA 64
APPENDIX B1 MATLAB script for the objective function formulation 70 APPENDIX B2 MATLAB script for the cell-to-cell attractant 71 APPENDIX B3 MATLAB script for the extended base stations 72 APPENDIX C MATLAB script used to generate response 79 APPENDIX D MATLAB script used for result comparison
INTRODUCTION 1.1 Background
According to Annu and Chaudhary (2015), “Wireless Sensor Network (WSN) is an interconnection of a large number of nodes deployed for monitoring a system by means of measurement of its parameters”. Due to its wide range of applications in both the military and civilian domain, it is emerging as a prevailing technology for the future. It is used in industrial process control and monitoring, healthcare monitoring, environment and habitat monitoring, disaster management, structural monitoring and lots more (Arya & Raina, 2014). Wireless Sensor Networks are prone to security attacks due to severe constraints such as broadcast nature of transmission medium, limited battery power, small memory and susceptibility to physical capture because they are deployed in hostile and physically non-protected areas. As such, security is a major concern in WSN. There are many possible attacks on sensor networks such as Denial of Service (DoS) attack, selective forwarding, sybil attack, sink hole attack, black hole attack, hello flood attack and wormhole attack (Sharma & Thakur, 2014; Sharma & Ghose, 2010).
One of the most severe attacks is the black hole attack, which drops the entire packet. It is a type of routing attack whereby an intruder captures and reprograms a set of sensor nodes in a network so that they do not transmit the generated or received data packets to their original destinations.Black hole attack prevention techniques proposed in literature either use neighbourhood interactions and message overhearing (Karakehayov, 2005; Roy et al., 2008), or secret sharing and path diversity (Ketel et al., 2005; Lou & Kwon, 2006).
It is assumed that based on neighbourhood message interactions and overhearing, a sensor node close to a black hole node can monitor and report the black hole node. Though, when many sensor
nodes within the same location are compromised, they can conspire and make the technique
ineffective. A much better technique is that based on path diversity and secret sharing, though it is
still not very effective (Arya & Raina, 2014; Dighe & Vaidya, 2013; Misra et al., 2011; Sheela &
Mahadevan, 2012). In this technique, messages are transformed into multiple shares by secret
sharing schemes and then the shares are delivered via multiple independent paths to the
destination.Figure 1.1 shows a sensor node, which transmits data to the base station using four node
disjoint paths. The figure indicates that none of the packets passing through the four paths reaches
the base station due to the deliberate placement of the black hole region, despite the path diversity.
This indicates that routing based on multiple paths can perform badly in the presence of black hole
attacks (Dighe & Vaidya, 2013; Misra et al., 2011).
Successful packet delivery to the destination, which is the base station, without loss of data in a
WSN, is more essential than the requirement of prevention of data being captured by an adversary
(Arya & Raina, 2014; Misra et al., 2011). As such, this research developed a mechanism for
successful packet delivery using multiple base stations in the presence of black hole nodes in a
WSN. Modified bacterial foraging optimization technique was used to determine the optimal
location to deploy the base stations in this work and in the mitigation of the black hole nodes.
Figure 1.1: Data Delivery and Black Hole Region in a WSN with One Base Station (Misra et al.,
BFOA proposed by Passino (2002) based on the social foraging behaviour of a bacteria Escherichia Coli is a new member in the coveted domain of swarm intelligence that has drawn the attention of researchers in different fields of knowledge with regards to its biological motivation and graceful structure (Das et al., 2009). It has been applied for solving practical engineering problems like optimal control, harmonic estimation, machine learning, channel equalization and others (Annu & Chaudhary, 2015; Latiff et al., 2007). The algorithm is based on imitating the searching behaviour of the bacteria in terms of positioning, handling and food ingestion. It has advantages such as high computational efficiency, high precision, fast convergence and global optimization (Kandasamy et al., 2014; Li et al., 2014). This research is aimed at the development of a black hole mitigation (detection and elimination) based model using an intelligent bio-inspired optimization algorithm (BFOA). A lot of research have been conducted on the area of malicious beacons detection using swarm intelligence techniques such as: Intelligent Water Drop (IWD) (Qureshi et al., 2011), Ant Colony Optimization (ACO) (Iftikhar & Fraz, 2013), Particle Swarm Optimization (Khan & Iftikhar, 2013) and Bacterial Foraging Optimization (Arya & Raina, 2014). In this research, a modified BFOA was employed for the black hole mitigation and optimal deployment of base stations. This is due to its robustness, fast convergence and good optimization results.
One of the shortcomings associated with the standard BFOA is the constant step size which results in unnecessary chemotactic steps as the bacteria is close to the optimal values (Dasgupta et al., 2009). With a large step size, the accuracy of the final solution is reduced because the bacteria are not able to reach the optimum location. Likewise, with a small step size, the convergence speed is affected because it will reach the optimum location but at a slow speed (Nasir et al., 2015). However, since the BFOA is a meta-heuristic optimization technique, which is regeneration dependent, any constant parameter at the initial stage may have significant influence on the
performance of the algorithm at the later stages. This led to the development of an improved BFOA with regards to the step size by some researchers (Dasgupta et al., 2009: Supriyono & Tokhi, 2011; Nasir et al., 2015;Niu et al., 2015). The nature of their improvements was based on introduction of step size variation through adaptation of mathematical formulation and fuzzy logic approach by establishing a relationship between bacteria step size and nutrient value. However, linear value iteration based behaviour was introduced and discussed in this work to address the shortcoming of the standard BFOA. 1.2 Statement of Problem
Most routing protocols used in Wireless Sensor Networks do not consider message security due to resource constraints (Ngai et al., 2006), which create opportunities for attackers. Among such attacks against wireless sensor network, a black hole attack has the ability to undermine the effectiveness of the network by partitioning the network to prevent useful information from reaching the base stations. Since successful delivery of data to the base station is more essential than prevention of data captured by an adversary because it can be countered with cryptographic techniques, several techniques based on secret sharing and multi-path routing were proposed in literature to overcome black hole attacks in the network but seems not to be very effective (Ketel et al., 2005; Lou & Kwon, 2006). Thus, this research developed an algorithm to mitigate black hole attack in wireless sensor network based on modified BFO using optimized multiple base stations. Also the number of base stations was increased from four (4) to eight (8) to determine the increasing trend of packet delivery as the number of base stations is increased beyond four and to investigate the effectiveness of the modified BFOA as the complexity of wireless sensor network environment increases.
1.3 Motivation Due to the distributed nature and deployment environment of WSNs, they are vulnerable to numerous security threats that can adversely affect their performance. Therefore to ensure the effectiveness and efficient functionality of WSNs, security is a major concern. The WSN mechanisms cannot at present ensure that an attack would not be launched and existing security mechanisms are inadequate. As such,there is need to develop new efficient and more reliable methods to protect the network from such attacks. 1.4 Aim and Objectives The aim of this research is to develop an intelligent optimization based wireless sensor model capable of ensuring a successful packet delivery in the presence of black hole attack. In order to achieve this aim, the following objectives were set:
1) Replication of the standard BFOA and development of the modified BFOA for the optimization of the multiple base station locations and detection of the black hole using MATLAB programming language.
2) Evaluation of the performance of the developed algorithm through simulation and comparison with existing algorithm to determine its effectiveness and accuracy in ensuring successful delivery of packets. Packet delivery success and failure, false positive and convergence speed were used as performance metrics.
3) Investigation of the effect of increasing the number of base stations from four (4) to eight (8) in order to find the trend of packet delivery as the number of base stations is increased beyond four (4).
1.5 Justification Security attacks such as black hole poses great challenges to wireless sensor networks by hindering delivery of message to the destination. Since the sensor nodes that are responsible for sensing and forwarding the data to the base station for analysis and compilation can be compromised by attacks (such as: black hole, sink hole, hello flood, sybil, etc). There is need to optimally place the base stations to reduce the effect of these attacks. This has led to optimal placement of multiple base stations using robust intelligent meta-heuristic technique, which is capable of reducing the effect of attacks on WSNs. But the results of the mitigation technique can be improved to enhance accuracy and searching speed of the algorithm. Also, the number of base stations is a critical factor of the sensor network architecture that significantly affects the network performance. Therefore, the need to determine the optimal number of base stations for a particular coverage area. 1.6 Methodology The methodology adopted in this research is as follows:
1) Initialization of BFOA parameters and the network parameters.
a) Number of Bacteria
b) Problem dimension
c) Chemotaxis step
d) Swarm limit
e) Reproduction step
f) Elimination probability
2) Replication of the standard BFOA using MATLAB programming language for:
a) Chemotaxis and swarming step
b) Reproduction step
c) Elimination and dispersal step
3) Development of the modified BFOA using MATLAB programming language for:
a) Adaptive step size
b) Steps a) to c) of item 2) are implemented
4) Deployment of the randomly generated sensor nodes using the standard and modified BFOA for the four base stations.
5) Application of model developed in 3) using scenario of 4) to detect the black hole nodes in the network.
6) Performance evaluation of the developed model using Packet Delivery Success and Failure, False Positive and Convergence Speed as performance metrics.
7) Comparison of the developed algorithm with the Standard algorithm.
8) Increasing the base station topology to eight and deployment of the randomly generated nodes.
9) Application of model developed in 3) using scenario of 8) to detect the black hole nodes in the network and evaluating the delivery performance.
1.7 Dissertation Organization
The general introduction has been presented in Chapter One. The rest of the chapters are structured as follows: Firstly, a detailed review of related literature and relevant fundamental concepts about wireless sensor networks, its constraints and characteristics, security goals, types of attack in wireless sensor networks, black hole attack and standard BFOA are carried out in Chapter Two. Secondly, an indepth approach and relevant mathematical models describing the development of
the modified bacterial foraging optimization algorithm are presented in Chapter Three. Thirdly, the analysis, performance and discussion of the result are shown in Chapter Four. Finally, conclusion and recommendations of further work makes up the Chapter Five. The list of cited references and MATLAB codes in the appendices are provided at the end of this dissertation.
GET THE COMPLETE PROJECT»
Do you need help? Talk to us right now: (+234) 08060082010, 08107932631, 08157509410 (Call/WhatsApp). Email: firstname.lastname@example.org