Adaptive Quality of Service and Trust Based Lightweight Secure Routing Algorithm for Dense Wireless Sensor Networks

Abstract

Wireless Sensor Networks (WSNs) are group of wireless devices that are deployed in an adhoc manner and are generally left unattended. The main advantages of WSNs are that they are simple to use, allow the use of inexpensive sensor nodes, and have good scalability. WSNs are useful in object tracking, periodic monitoring, and event detection applications. However, the inherent characteristics of the WSNs, such as limited resources and low computation, make them vulnerable to various types of security attacks. Therefore, security mechanisms are needed to secure the network and protect against various security attacks. Conventional security mechanisms, such as cryptography (encryption/decryption) and authentication based systems, are generally used to ensure the security of traditional networks. However, due to the resource constrained nature of WSNs, conventional security mechanisms can be too resourceheavy to allow the reliable and lightweight operation of a WSN. Therefore, providing security, while maintaining Quality of Service (QoS) and energy efficiency, represents an important research challenge in the design of WSNs. In this thesis, we critically investigate the problem of security provisioning in WSNs. We identify challenges, limitations, and requirements for implementing security with QoS and energy efficiency for dense WSNs. We find that the security constraints for WSNs have not been well discussed in the literature. Also, the simultaneous optimization of energy, QoS, and security has not gained much attention. We develop two novel algorithms that address the above issues in WSNs and optimize energy, QoS, and security using a metaheuristic technique known as Ant Colony Optimization (ACO). These algorithms are called Dynamic Trust-aware Secure Routing (DTSR) and Lightweight Secure Routing (LSR). DTSR improves the connectivity and improves the tradeoff between coverage and lifetime for dense WSNs. Furthermore, LSR provides an improved method for the detection and isolation of a compromised node by using direct and indirect trust calculations for dense WSNs. We show through analytical and simulation results that our presented algorithms can outperform existing techniques in terms of network lifetime, average routing delay, and packet delivery ratio. Also, we perform an analysis of network lifetime over varying network sizes to find a good range of nodes for the effcient performance of the algorithm. Furthermore, we present a runtime analysis of the algorithms to understand the simulation time in the MATLAB environment based on changing network size.