TOA estimation in dense multipath environment for mobile device using audible LFM signal

This paper presents a threshold-based time of arrival (TOA) estimation method for audible linear frequency modulation (LFM) signal, which focus on eliminating global error caused by multipath fading or interference. Unlike conventional methods that try to directly detect first arriving path corresponding to the actual TOA, the proposed algorithm estimates TOA in two steps thus reducing early detection caused by noise and increasing TOA estimation accuracy. It utilizes adaptive thresholds based on matched filter, a jump-back and search-forward algorithm for the detection of first path. The relation between signal to noise ratio and optimal threshold are investigated via simulations. The proposed method yields efficient direct path detection and low complexity implementation for mobile device, performance of TOA estimation is examined by simulations and filed measurement.