Localization in Ad-Hoc networks

Localization is an inevitable challenge when dealing with wireless sensor nodes and a problem that has been studied for many years. Nodes can be equipped with a Global Positioning System (GPS), but this is a costly solution in terms of volume, money and power consumption. The algorithm presented here in relies on range measurements between pairs of nodes and a priori coordinates of sparsely located anchor nodes. In this paper, we discuss a robust and distributed RSSI-based estimation algorithm for wireless sensor network. A neural network is trained with precise localized nodes coordinate training set and localizes any unknown node physical location.