Joint Channel Tracking and Symbol Detection for MIMO-OFDM Mobile Communications

In this paper, a new joint channel tracking and symbol detection algorithm is proposed for MIMO-OFDM systems over the time-varying frequency-selective fading channel. The iterative detection/decoding and channel estimation are combined to improve system performance. With the aid of the candidate list from list sphere decoder (LSD), the channel frequency response of each subcarrier is estimated by the recursive expectation-maximization (EM) algorithm which uses sequential processing over all the significant symbol combinations in the limited candidate list. Moreover, with the prior knowledge of the second-order statistics of the fading channel in time domain, the time-varying channel impulse responses are tracked by the Kalman predictor based on the estimated channel frequency response. Simulation results demonstrate that the proposed algorithm can track the time- varying channel effectively and offer substantial performance gains over the conventional soft decision-directed Kalman filtering techniques.