Low Complexity ICI Mitigation for MIMO-OFDM in Time-Varying Channels

Abstract

Time-varying channels destroy the orthogonality among subcarriers in orthogonal frequency division multiplexing (OFDM) systems, and introduce intercarrier interference (ICI). Lots of efforts have been devoted to mitigate ICI in OFDM systems, with different frame structures and channel models, but the computational complexity of the methods is usually very high. In multiple-input multiple-output (MIMO) systems, the complexity is even higher. In this paper, a low-complexity ICI mitigation method is proposed for MIMO-OFDM systems under the assumption of linear time-varying channels. It reduces the complexity of ICI compensation from $ {O(K^{3}(N^{3}+MN^{2}+MN)+NK\log (K))}$ to $ {O(K(N^{3}+2MN^{2}+2MN+2M^{2}+N\log (K)))}$ , where ${K}$ is the number of subcarriers, ${M}$ the number of transmitters, and ${N}$ the number of receivers. It requires channel estimation based on the linear time-varying channel model, and no transmission overhead is needed. The proposed algorithm applies to all OFDM systems as long as linear time-varying channel estimation is applicable. Time-domain synchronous-OFDM naturally suits for the proposed ICI mitigation algorithm because its receiver is able to easily estimate linear time-varying channels. Simulation with QPSK and 16QAM modulation demonstrates the performance of the proposed method, in comparison with no ICI mitigation case. It shows that 2 dB signal to noise gain is achieved when the uncoded bit error rate is $ {10^{-3}}$ and the normalized Doppler frequency is 0.1.