Localization and Tracking in Wireless MIMO Systems

Abstract

Due to the demands for location-based services, localization and tracking of a mobile terminal (MT) have attracted much attention recently. When performing localization and tracking in multiple-input multiple-output (MIMO) systems, the space-time processing capability which MIMO can provide enhances the overall performance. In this thesis, several new approaches are proposed with the aim to improve the accuracy of the location estimate. The space-time processing techniques together with the prior channel state information (CSI) are used to enhance the accuracy of location systems through achieving more accurate parameter estimates. We first propose a precoder design strategy to enhance the estimation of angle-of-arrival (AoA) and location. An optimal precoder which achieves the new error bound of the MUSIC (MUltiple SIgnal Classification) algorithm, as well as a more implementable precoder which leverages on the feedback CSI estimated at the receiver and performs close to the optimal precoder, are both developed. We next study the impact of signal pre-processing on the time-of-arrival (ToA) estimation through transmit beamforming and transmit diversity. We demonstrate that the accuracy of the ToA estimator is enhanced with the availability of CSI at the transmitter (CSIT) and the number of antennas. A maximum a posterior (MAP) based channel estimation algorithm is also proposed to jointly estimate the temporal and spatial domain channel parameters, which leverages on prior statistical information of the channel parameters used in the extended Saleh-Valenzuela (SV) model. Finally, we propose and study a novel MT tracking method based on the space-time processing capability of MIMO systems. The proposed algorithm consists of three steps: motion-dependent parameters estimation based on the space-time correlation of the received signal, extend Kalman filter (EKF) based tracking method to estimate the location, and high-resolution AoA (HR-AoA) based performance enhancement to reduce the accumulative error. The performances of all the proposed algorithms are studied through simulations.