Joint Channel Estimation and Signal Detection Based on MAP Criterion in MIMO-OFDM System With Phase Noise

The channel estimation and signal detection are key issues in the Multi-Input Multi-Output and Orthogonal Frequency Division Multiplexing (MIMO-OFDM) system. However, there exist severe impacts of phase noise (PN) on the estimations with the application of higher frequency in 5th Generation New Radio (5G-NR). In this paper, the joint channel estimation and signal detection (JCESD) method based on the maximum a posteriori (MAP) criterion under the assumption of Wiener process for PN is proposed in MIMO-OFDM system, which is called as the JCESD-PN-MAP method. Firstly, the MAP criterion is derived based on Bayesian theories and the structures of matrices in MAP criterion are analyzed to simplify the optimizations. Secondly, the optimizations for PN of receiving and transmitting antennas are translated into solving a tridiagonal linear equation and a sparse linear equation, respectively, which are optimized by the Gaussian elimination (GE) method with low computation complexity. To further reduce the computational complexity, the latter is solved by the alternating direction method of multipliers (ADMM) method. Thirdly, the optimizations for channel responses and transmitted signals are translated into solving two block diagonal linear equations, which are solved by calculating the inverse matrix with low computation complexity. The numerical results and complexity analysis confirm the effectiveness of our proposed method in terms of the accuracy and computation complexity.