An investigation into performance of MIMO wireless communication system under semi-blind channel estimation

Abstract

A new approach to Semi–Blind Channel Estimation (SBCE) technique for Multiple Input Multiple Output (MIMO) wireless communication system is proposed in this paper. It combines the least squares (LS) and Minimum Mean Square Error (MMSE) Training–Based Channel Estimation (TBCE) scheme with whitening rotation–based SBCE scheme. In this approach, the singular value decomposition has to be applied only once, thereby reducing the computational complexity. The channel correlation matrix required by MMSE technique can be obtained based on latest channel conditions. The performance of this scheme is compared with whitening rotation–based SBCE under various parameters such as modulation schemes, Signal–to–Noise Ratio (SNR) and frame length. The performance of the new approach is found to be better in terms of Bit Error Rate (BER) for high SNR, higher modulation order and smaller frame lengths. With low SNR, lower modulation order and larger frame lengths, there is negligible reduction in performance.