Fuzzy extended Kalman filter for dynamic mobile localization in urban area using wireless network

The problem of accurate mobile positioning in cellular network is very challenging and still subject to intensive research, especially given the uncertainty pervading the signal strength measurements. This paper advocates the use of fuzzy based reasoning in conjunction with Kalman filtering like approach in order to enhance the localization accuracy. The methodology uses TEMS Investigation software to retrieve network information including signal strength and cell-identities of various base transmitter stations (BTS). The distances from the mobile station (MS) to each BTS are therefore generated using Walfish-Ikigami radio propagation model. The performances of the established hybrid estimator −fuzzy extended Kalman filter (FEKF)- are compared with extended Kalman filter approach and fuzzy-control based approach. Both simulation and real-time testing results demonstrate the feasibility and superiority of the FEKF localization based approach.