Computational Intelligence Algorithms For Energy Optimization Problems In Wireless Sensor Networks

Abstract

Recent advancements in hardware and wireless technology enabled the development of low cost and energy-constrained tiny devices known as sensors that communicate with each other at short distances through a wireless link. The collaborative settings of these tiny devices form a Wireless Sensor Network (WSN). In the recent past, it has gained tremendous interest among researchers and industrial communities due to its wide spectrum of applications in the real world. One of the important issue is coverage of the given set of targets under specified con- nectivity constraint. The other main issue is the interference of signals in the wireless media. This results in message drop and requires message retransmission which in turn affects the en- ergy efficiency of the network. Energy conservation is the most critical problem in WSN to extend stability or lifetime of the network. Many artificial intelligence methods are proposed in the literature to solve these problems in the wireless sensor network. The first objective of the thesis work is to deploy an optimal number of sensor nodes with k-coverage and m-connectivity constraints in an area of interest. The problem of ensuring all the targets are covered by at least k number of sensor nodes and all the sensor nodes have at least m connectivity with other sensors nodes is termed as k-coverage and m-connectivity problem in WSN. Many meta-heuristic algo- rithms have been proposed to solve different problems like clustering and localization in WSN. In this work, a novel meta-heuristic based differential evolution algorithm to solve k-coverage and m-connectivity problem in WSN is proposed. The second objective of the thesis is inter- ference minimization in wireless sensor network. Therefore biogeography based optimization and multi-attribute decision making techniques are proposed for sensor placement which min- imizes the interference of sensors by preserving connectivity and coverage constraints. The third objective of the thesis is to propose an energy efficient clustering technique using artificial intelligence methods. Therefore a hybrid of game theory and fuzzy logic based hierarchical clustering algorithms are proposed to increase stability of the network. Also, an interference aware clustering technique is proposed using TOPSIS to extend stability of the network. Sim- ulations are carried out to check validity of the proposed methods and compared with other methods.