Energy Efficient Routing for Multi-layer Design of Wireless Sensor Networks

Abstract

Wireless Sensor Network (WSN) has its applications from terrestrial to aquatic, from earth to cosmic, from agriculture to jungle and from home and offices to international borders. For it, tens to hundreds to thousands of nodes are deployed. So, the distance between the sources to destinations ranges from meters to kilometers that requires well-organized communication architecture with respect to software and hardware. Among constituents of communication architecture, routing is the most energy squeezing process. On focusing this issue, energy efficiency in the routing process is the ultimate goal of this dissertation that is achieved by setting some milestones. First target is to do a comprehensive analysis on energy aware routing in WSN and to present a detailed analysis on those factors that affect energy aware routing. This step gives backing for preparing the underpinning grounds for our further research. In a continuation of this work, clustered network architecture is analyzed and customized since it is extensively used by researchers for energy aware routing. However, more attention is required to ameliorate the energy consumption aspect of its cluster designing process. Hence, a novel design of clustered network architecture is introduced. The proposed design technique is innovative in its idea. Our novel layer-based hybrid algorithms for cluster head and cluster member selection come up to novel communication architecture. Since its substantial constituent is cluster designing, so it is named Multilayer Cluster Designing Algorithm (MCDA). The proposed design not only has an effect on lessening blind broadcasting, but also on decreasing the message exchange in a passionate way. It also encapsulates the beauty of efficient centralized decision making for cluster designing and energy-aware distributed cluster head selection and cluster member allocation processes. Comprehensive experimentations have been performed on the comparative analysis of MCDA with state-of-the-art centralized and distributed cluster designing approaches. Calculation of energy consumption at various aspects, number of designed clusters and number of packets broadcast during the cluster designing are the main performance parameters. It has been found that MCDA outperforms its competing algorithms with respect to the aforementioned parameters due to its multilayered synergistic mating approach. In the extension of this work, its constructive ramification; Extended MCDA is proposed to ameliorate the performance in network lifetime. Novel algorithms for time slot allocation, minimizing the cluster head competition candidates, and cluster member selection play underpinning roles to achieve the target. These incorporations in MCDA result in minimizing transmissions, suppressing unfavorable response of transmissions and near equal size and equal load clusters. Extensive simulations in NS2 are done to evaluate the performance of E-MCDA in energy consumption at various aspects, packets transmission, number of designed clusters, number of nodes per cluster and un-clustered nodes. It is found that the proposed idea optimistically outperforms the competition with MCDA and EADUC (Energy Aware Distributed Unequal Clustering). Next step is energy aware routing mechanism for WSNs is proposed. Constituents of this mechanism are analysis of MCDA for highlighting the parameters to be used for energy aware routing algorithm, Energy Efficient Cluster Head Rotation Technique and Energy Aware Routing Strategy for MCDA that is made up of inter-cluster and intra-cluster routing algorithms. Each constituent of this mechanism plays vital supportive role to other to achieve the target of energy awareness in the routing process. We conducted simulations in NS2 simulator to validate the performance of proposed scheme in comparison to TLPER (Threshold Based Load Balancing Protocol for Energy Aware Routing) and EADUC. Simulations results demonstrate the performance efficiency of proposed scheme in terms of various metrics compared to similar approaches published in the literature.