Analysis of Non-Recurrent Algorithms for Detecting 4-QAM Signal in a MIMO System with Different Numbers of Antennas

The paper investigates non-recurrent algorithms for obtaining soft solutions for a four-position quadrature amplitude modulation (4-QAM) signal in conditions of estimating communication channel parameters with different accuracy in a communication system with multiple transmitting and receiving antennas (MIMO). The communication channel was modeled using the Jakes model, which takes into account Doppler spreading and Rayleigh fading of the signal. The additive noise was assumed to be white Gaussian. The channel matrix takes into account the distortions received by the signal in the direct conversion receiver path, such as amplitude and phase imbalance between signal quadratures, phase noise, frequency shift remaining after the demodulation procedure, and DC components. The noise immunity of the proposed regularizing procedures was compared with the known algorithms Zero Forsing, minimum mean square error (MSE) and a procedure based on the least squares method (LSM) for a MIMO system with a number of antennas of 8, 16, 32 and 64. The computational experiment showed that that a regularizing algorithm for finding soft solutions can provide an energy gain in noise immunity with inaccurately known parameters of the communication channel relative to other signal detection procedures considered in this work.