Advanced MDS based localization algorithm for location based services in Wireless Sensor Network

Abstract

Wireless Sensor Networks (WSNs) are increasingly being used for various information collection in the future. For many applications, a large number of sensors in an active wireless network are more effective and preferable to a few expensive wireless sensors. Location discovery of sensor nodes after they have been deployed is known as a challenge, especially for moving sensors. Algorithms based on radio signal strength indication are considered as the lowest cost method to solve localization problem. But in practice, the accuracy of signal strength indication algorithm varies with the average distances between wireless nodes - the nearer, the better. This paper describes a new integrated dynamic localization algorithm, based on Received Signal Strength Indication (RSSI) and Multidimensional Scaling (MDS) algorithm. In this concept, signals with RSSI values among beacon nodes could help to generate a real time 2D map of the network by self iterations. Unknown nodes use the map to determine their locations in this region. A software simulator is created to test the performance of the algorithm, and practical experiments also show the development of localization accuracy and efficiency.