MDS-LM for Wireless Sensor Networks Localization

Recent advancements in micro electro mechanical system and wireless communication have made it possible to use wireless sensor nodes for many real world applications. Wireless sensor nodes are widely been used now a days in many tracking and monitoring problems such as vehicle tracking, environmental and health monitoring. It is one of the most challenging problems to find the geographic locations of nodes given that proximity information only between the nodes. There are large number of localization algorithms, each having different approach to find the final location. Multidimensional Scaling (MDS) is one of that technique which works in more robust fashion for noise sparse networks, even with less number of anchor nodes. The proposed work addresses the problem of location discovery of nodes based on MDS and Levenberg-Maurquardt (LM) method. In this paper we used MDS to find the initial estimated position of nodes and proposed Levenberg-Maurquardt as a refinement algorithm for final position. Our simulation results show that the proposed scheme is very competitive compared to other common schemes.