AN ACCELERATED ALGORITHM FOR LARGE-SCALE MIMO SIGNAL DETECTION IN 5G SYSTEM

In multiple input multiple output (MIMO) system, because the MIMO system supports a large number of antennas, it is difficult and almost impossible for traditional detection algorithms to balance the system performance loss caused by the high complexity of the algorithm. Under the premise of ensuring better performance requirements, in order to reduce complexity as much as possible, we propose a block-modification successive over-relaxation (MBSOR) detection algorithm based on the block matrix and the correction factor. The algorithm blocked the filter matrix to obtain more channel information from the diagonal matrix and introduced a correction factor to the iterative matrix to further accelerate the convergence speed. Through the simulation of this algorithm and the traditional successive over-relaxation (SOR) algorithm, it is shown that the MBSOR algorithm is closer to the optimal solution than the SOR algorithm while reducing the number of repetitions.