Analytical and Simulation Modeling to Analyze Reliability State of Wireless Sensor Networks

Abstract

Abstract. Wireless sensor networks have gained abundant interest due to their potential wide range of applications. Reliability of deployed wireless sensor network is defined by covered area of alive nodes and redundancy of data. Redundancy in data occurs because of overlapped sensed area. An initial reliable wireless sensor network switches to unreliable state because nodes perish in the field randomly. Consequently, the quality of data starts diminishing. It is imperative to know when the network will switch to unreliable state from reliable state so that proper action can be conducted in the field. Work of this paper analyzes and compares analytical and simulation modeling for reliability state of wireless sensor network. Multi-objective genetic algorithm based method is operated for analytical modeling which determines minimum number of nodes (randomly) that covers almost complete area while having required minimum overlapped area. Clustering algorithm, LEACH, is implemented in NS-2 for simulation modeling. Comparative results of analytical and simulation modeling are different because of their different nature but both highlights that reliability of wireless sensor network is salient.