Location estimation in multi-carrier systems using extended Kalman based interacting multiple model and data fusion

Abstract

Researchers have shown that extended Kalman based interacting multiple model (EK-IMM) can significantly mitigate location estimation errors. In this study, we propose an EK-IMM smoother generalized for multi carrier systems. In our system, we use COST 231 Walfisch — Ikegami path loss model instead of street microcell path loss model in order to improve the applicability of the method. By implementing the new path loss model, we do not need to know distances from base stations to building corners, hence a database to hold building coordinates is not required. Via simulation results, we have shown that by implementing the new EK-IMM smoother in multi-carrier systems, location estimation accuracy can be improved by approximately 60%, making use of 48 subcarriers.