A Variable Length Genetic Algorithm approach to Optimize Data Collection using Mobile Sink in Wireless Sensor Networks

Abstract

Data collection using mobile sink (MS) in wireless sensor network (WSN) is an efficient approach to collect data from sensor nodes (SNs). There are several advantages of using a MS like, it helps to alleviate the sink hole problem, balances energy consumption of the network, enhances network lifetime etc. The MS has to go through the WSN and collect data from SNs by stopping at some predefined locations from where SNs are in its vicinity. Finding the number of such locations is a fundamental problem in the path design of a MS. In this paper, we present a novel variable length gentic algorithm (VLGA) to find a solution to the problem. The algorithm finds the optimal number and location of sojourn points which are used to design the tour of the MS. The algorithm is simulated extensively and compared with the TSP algorithm. The results obtained shows better performance of VLGA over TSP algorithm in terms of path length and data collection time of the MS.