Application of Selected Computational Intelligence Techniques to Improve QoS Parameters in WSNs

Abstract

A Wireless Sensor Network (WSN) is a collection of distributed sensor nodes which have limited sensing, processing and communicating capabilities. The main objective of any WSN is to efficiently sense and collect information, process it and finally relay it to the base station. Sensor nodes in WSN are battery powered and in most of cases, it is not possible to recharge them or change their batteries. Thus, to prolong the network lifetime, energy efficiency becomes the most important design criterion. In WSNs thousands of sensors are randomly deployed in a particular region. These networks are mostly deployed at places where human intervention is rarely feasible and thus there is remote possibility of uniform deployment density. Due to high density of nodes and resulting contentions in shared communication medium, there are communication failures. Besides these, broadcasting nature of sensor nodes leads to high probability of data redundancy, which in turn leads to scalability which is another important factor for system design. This thesis addresses these inherent constraints of WSNs by developing methods and algorithms with energy efficiency, scalability and optimal resource management as the main research goals. newlineIn both uniformly and randomly deployed WSNs, clustering is an effective method for improving energy efficiency, robustness and scalability. However, due to random node deployment, the clusters have unequal node density. The cluster heads of clusters with higher node density consume comparatively higher energy which leads to lower time span. Further, in cluster based WSN, inter cluster communication is to be structured either as one-hop or multi-hop communication mode. In case of one-hop communication, nodes distant from the sink consume higher energy. In case of multi-hop communication, there is heavy relay traffic load for the nodes adjacent to sink. Thus, both of these modes lead to an unavoidable problem of unbalanced energy consumption...