The MMSE MIMO Detector Under a New LMMSE Channel Estimation Error Analysis Method

The MIMO detector should output the log-likelihood ratio (LLR) of each coded bit to the channel decoder in order to achieve the desired bit error rate (BER) performance in a channel coded multiple-input multiple-output (MIMO) system, whereas channel estimation error which will result in performance degradation in the practical MIMO system. This paper proposes to divide the linear minimum mean square error(LMMSE) into two parts: white noise error and Wiener filter error (the error caused by the ideal channel after Wiener filtering). Based on the Wiener filter coefficients, noise, and channel response, we can derive channel estimation errors and incorporate them into the MMSE detector to improve system performance. The simulation results show that the proposed new method can improve system performance with negligible complexity increase.