Iterative Massive MIMO Demodulation Method with Non-Gaussian Approximation

Nowadays, the problem of synthesizing demodulation algorithms with good noise immunity and acceptable computational complexity is a particularly pressing issue. The application of a non-Gaussian approximation of the prior distribution of the estimated parameters for demodulation in Massive MIMO systems using Newton’s method is proposed. The demodulation problem is presented in the form of a problem of solving a system of nonlinear equations. A study of the new algorithm was conducted under the conditions of a different number of antennas and different orders of QAM modulation. The results of modeling a new demodulation algorithm using Newton’s method are presented, as well as a comparison with the well-known MMSE algorithm, Monte Carlo, and K-best methods, which confirm the effectiveness of the proposed approach of non-Gaussian approximation for demodulation in Massive MIMO systems.