An Efficient Channel Estimation Scheme for mmWave Massive MIMO Systems

Due to the massive deployment of antennas, the computational complexity and channel dimension increase enormously and the computations are tedious. To solve this problem, we propose low complexity precoding methods to effectively estimate the channel using compressive sensing and hybrid precoding techniques. In compressive sensing technique, based on the sparsity of the beamspace channel matrix, we choose several beams to collect the main energy of the channel to approximate the large-dimensional channel matrix. In the hybrid precoding method, an iterative precoding algorithm is analyzed for analog precoding. The algorithm can utilize the approximate orthogonality between different paths, and the asymptotically optimal transmitting precoding matrix and the corresponding combining moment of the corresponding receiving end can be obtained through iterative processing. In the digital part, two linear precoding methods, MRT and BD are considered and their performance is compared. Simulation results show that the proposed reduced-complexity precoding methods can significantly reduce the mmWave Massive MIMO system complexity, improve the system sum rates, and provide a reference for hardware and software design of subsequent research.