Precoder Design for Time-Varying mmWave MIMO Detection Based on Expectation Propagation

This paper proposes a novel precoder design for time-varying millimeter-wave (mmWave) multiple-input multiple-output (MIMO) signal detection based on expectation propagation (EP). In mmWave wireless communications, millisecond-scale fluctuations in the surrounding environment cause significant changes in the wireless channel, and the channel aging over time causes a mismatch between the beam and actual channel. The proposed precoder attempts to maximize detection performance by adjusting information propagated across iterations when employing the EP-based iterative algorithm to suppress interference components due to channel aging at the receiver. Simulation results show that the precoders that impose heterogeneity in quality among streams of MIMO spatial multiplexing minimize the bit error rate (BER). We conjecture that this phenomenon is due to the improved detection accuracy in the early iterations caused by the heterogeneity among streams, and discuss appropriate precoder design criteria by quantitatively evaluating the performance improvement.