Separate Estimation of Constant Components, Channel Matrix, Amplitude and Phase Imbalance of the QAM Signal in a System with MIMO

The article discusses an algorithm for separate estimation of distortions of a quadrature amplitude modulation (QAM) signal in a direct conversion receiver, as well as a channel matrix for communication systems with multiple transmitting and receiving antennas (MIMO). The estimation procedure is carried out using the test sequence and is based on the method of statistical averaging over time. The test sequence is the type of information signal. This approach allows estimating unknown parameters of the channel and signal, avoiding matrix inversion. It leads to a reduction in computational complexity in contrast to joint estimation procedures. There are advantages of this algorithm especially for systems with a large number N of transmitting and receiving antennas. The estimation algorithm is synthesized for conditions of a priori uncertainty about the channel statistics and distributions of additive and phase noise. In addition, the algorithm can work using information symbols after the detection procedure. A computational experiment was carried out for a system with MIMO, N = 32 and uncorrelated Rayleigh fading for different lengths of the test sequence. It is shown that an increase of the sample size of the processed signal leads to an increase of the accuracy of the estimated parameters and noise immunity.