Improved coded massive MIMO OFDM detection using LLRs derived from complex ratio distributions

Abstract

In this paper, a novel receiver is proposed for coded massive Multiple-Input-Multiple-Output systems using Orthogonal Frequency Division Multiplexing (MIMO-OFDM). The receiver utilizes log-likelihood ratios (LLR) derived from complex ratio distributions to model the noise probability density function (PDF) at the output of a zero-forcing detector. These LLRs are subsequently used to improve the performance of the decoding of Low Density Parity Check (LDPC) codes. The Neumann large matrix approximation is employed to simplify the matrix inversion in deriving the PDF. To verify the new findings, Monte Carlo simulations are used to demonstrate the optimality of the fitting between the derived PDF the experimentally obtained histogram of the noise. Further simulations results over time-flat frequency selective multi-path fading channels demonstrated improved performance over equivalent systems using the Gaussian approximation for the PDF of the noise. A significant gain of 1 dB was observed at bit error rate of 10-4 which corresponds to a reduction of approximately 30 receive antenna elements.