A New Hybrid Wireless Sensor Network Localization System

Abstract

Wireless sensor networks are used to monitor the environment and to report the occurrence of events. The geographical location of the sensed event is usually important to the application. Hence, dynamically determining the physical location of every sensor node in space is crucial. In this paper, we present a new hybrid localization system (ALS) developed based on three existing localization algorithms: ad-hoc positioning system (APS), multidimensional scaling (MDS), and semidefinite programming (SDP). We consider five network properties that affect localization performance and use machine learning to obtain parameter values of ALS. Simulation shows that the new method achieves more accurate position estimation than the individual algorithms across broad network conditions