Performance of gravitational search algorithm in wireless sensor network localization

Abstract

Along with the collected data, the location information of sensor nodes will also be required in many Wireless Sensor Network (WSN) applications. The process of determining the location information or position coordinates of sensor nodes after their deployment is defined as localization. This work is concerned with localizing sensor nodes using a swarm-based optimization algorithm called Gravitational Search Algorithm (GSA). Localization in WSN is defined as a non-linear optimization problem and is solved using GSA. A detailed performance analysis is carried out through simulation. One of the major problems affecting the range based localization system is flip ambiguity. In this work, we tried to eliminate flip ambiguity from a network of collinear anchor nodes in a refinement phase of localization.