ER-CRLB Analysis for Indoor Multi-Antenna Localization System : (Invited Paper)

摘要

Using channel state information (CSI) values to achieve fine grained localization has received much attention in the recent years. However, the impacts of signal power, the non-line-of-sight, and the scale of the playing field have not been investigated yet. In this paper, we propose an extended recursive Cramér-Rao lower bound (ER-CRLB) method as a tool to analyze the performance of multi-antenna based wireless indoor localization systems. The Fisher information matrix of the channel state information based wideband wireless system is formulated, in which the unknown estimated vector is classified as two parts: the estimated vector and the auxiliary vector. The observations are the CSI values which indicate the attenuation and the propagation delay. We employ the ER-CRLB to analyze the impacts of geometry impact, NLOS identification, power control and prior target state information. The simulation results indicate that the power control scheme is more essential than related prior information.