Channel Estimation Method for LTE Downlink MIMO Transmission

Channel estimation often makes use of time-domain windowing to reduce noise and thus improve the accuracy of estimate. However, it is not easy to find a window size which removes as much noise as possible while preserving most of the signal energy. In this paper, we propose a metric, called estimated mean squared error (EMSE), which can be used in the search of optimal window size for Long Term Evolution (LTE) downlink multiple-input multiple-output (MIMO) transmission. Simulation results show that the proposed method improves the mean squared error for all signal-to-noise ratio (SNR) less than 20 dB. In the case of SNR higher than 20 dB, no windowing should be applied to the original channel estimates.