Localization scheme for large scale Wireless Sensor Networks

Localization of Wireless Sensor Networks (WSNs) is an issue for location-based services. This paper proposes a new Wireless Sensor Network Localization Scheme based on Self-Organizing Maps (SOM) to deal with this issue. Our scheme utilizes only connectivity information and information from some heard anchors in the network to determine the location of nodes. With anisotropic topologies or topologies having obstacles, localization schemes tend to get higher error. Our scheme not only solves well the isotropic networks, but also minimizes the error in anisotropic networks and networks having obstacles by filtering the information from heard anchors. From our intensive simulations, the results show that the proposed scheme achieves very good accuracy in most cases.