A channel estimation algorithm for large-scale MIMO system using block sparsity adaptive matching pursuit

A new adaptive channel estimation algorithm for large-scale multiple input multiple output (MIMO) systems is proposed by combining the block sparsity adaptive matching pursuit (BSAMP) technique with adaptive beamforming. Firstly, the structure model based on continuous constant is optimised randomly, and it is used alternately with the basic denoising optimisation scheme to find the sparse characteristic channel. Then the sparse matrix is optimised based on adaptive beamforming to enhance the channel sparsity. Furthermore, based on BSAMP technology, using the joint sparsity of large-scale MIMO system subchannels, we set threshold and find the maximum backward difference position to select the atoms of the support set quickly and preliminarily. At the same time, the energy dispersion caused by the non-orthogonality of the observation matrix is considered to improve the estimation performance of the algorithm. Finally, the atoms are filtered by regularisation to improve the stability of the algorithm. Simulation results show that the algorithm can recover large-scale MIMO channel information with unknown sparsity quickly and accurately, and the average running time is only 0.12 s.