Multiobjective Approach to Localization in Wireless Sensor Networks

Abstract

Wireless sensor network localization is a complex problem that can be solved using different types of methods and algorithms. Nowadays, it is a popular research topic. What becomes obvious is that there are several criteria which are essential when we consider wireless sensor networks. Our objective is to determine accurate estimates of nodes location under the constraints for hardware cost, energy consumption and computation capabilities. In this paper the application of stochastic optimization for performing localization of nodes is discussed. We describe two phase scheme that uses a combination of the trilateration method, along with the simulated annealing optimization algorithm. We investigate two variants of our technique, i.e., centralized and distributed. The attention is paid to the convergence of our algorithm for different network topologies and trade-off between its efficiency and localization accuracy.