Compressive Sensing Based Sparsity Adaptive Channel Estimation Algorithm for Massive MIMO Systems

The generalized block sparsity adaptive matching pursuit (GBSAMP) approach is suggested for Massive Multi-Input Multi-Output (MIMO) systems with unexpected channel routes. The endorsed set atoms are quickly and initially chosen using the combined sparsity of the massive MIMO system sub-channels by imposing the threshold and identifying the highest backward differential location, as well as the energy dispersion caused by a non-orthogonality observation matrix is taken into consideration to enhance the algorithm's estimation accuracy. The algorithm's stability is enhanced by a secondary screening of atoms using regularisation. Simulations demonstrate that the technique is capable of rapidly and accurately recovering large amounts of MIMO channel information with unpredictable sparsity.