Robust Single-Carrier Frequency-Domain Equalization for Broadband MIMO Systems With Imperfect Channel Estimation

Single-carrier frequency-domain equalization (SC-FDE) with multi-input multi-output (MIMO) has been recognized as an alternative technology to orthogonal frequency-division multiplexing with MIMO for broadband wireless communication systems because of its single carrier transmission advantages. Conventional SC-FDE MIMO schemes are designed under the assumption of perfect channel estimation. In this paper, we investigate the robust SC-FDE MIMO design for systems with imperfect channel estimation. Based on a statistical model for channel estimation, the optimal equalization coefficients for the robust SC-FDE MIMO schemes with both parallel interference cancellation and successive interference cancellation (SIC) are derived with the objective of minimizing the sum of the multiple data streams' mean square errors (sum-MSE). We propose an optimal ordering algorithm in the sense of minimum sum-MSE for the SC-FDE MIMO scheme with SIC, and further propose a low complexity sub-optimal ordering algorithm. The hit-error-rate (BER) performance in the uncoded case is analyzed and a tight BER approximation is derived. Numerical results show that the proposed robust SC-FDE MIMO schemes achieve 1.SdB similar to 4dB gains in E-b/N-o over the conventional non-robust schemes.