Fusion of Wi-Fi and WSN Using Enhanced-SIR Particle Filter for Hybrid Location Estimation

It is extremely important for pedestrian navigation and location-based services (LBS) that users of portable devices can quickly and accurately locate themselves wherever they might be. In this paper, we propose a novel hybrid approach for indoor and outdoor localization in urban area based on Enhanced Sequential Importance Resampling (Enhanced-SIR) Particle Filter (PF). The approach implements by gathering and processing signal strength information from Wi-Fi and other signals of opportunity, such as wireless sensor network (WSN). It combines WSN (2.4 GHz Zigbee) with Wi-Fi for hybrid location estimation, in which, Wi-Fi signal is used for the entire area, and WSN signal serves as the enhanced signal to assist the resampling of particles. The method of Enhanced SIR can not only solve the problem of degeneracy, but also improve the sampling accuracy of particles. Experimental results demonstrate that the proposed hybrid approach can provide more accurate location estimates in indoor and outdoor areas, and costs equivalent runtime, compared to conventional SIR PF.