Performance analysis of data fusion for ground target tracking using GSM networks
Miao Zhang, Stefan Knedlik, Otmar Loffeld · 2009
cations (GSM) networks can provide position informa-tion. However, a GSM positioning system based on cur-rent specifications faces many difficulties to yield accu-rate position estimate for ground target tracking. Data fusion solutions which integrate two or more types of measurements have been proposed to provide estima-tion with better accuracy, reliability and coverage. The performance of Mobile Station (MS) positioning has been broadly discussed, but the performance analysis of ground target tracking, i.e., dynamic position estima-tion, is rare. This paper focuses on the performance analysis for a data fusion approach of integrating two different types of measurements from GSM networks us-ing posterior Cramér-Rao lower bound (PCRLB). The simulation results verify that the data fusion approach greatly improves the estimation accuracy and is more robust to noisy measurements as compared to the ap-proaches using single kind of measurements.