Compressive Channel Estimation in Space Domain for Massive MIMO Systems

In the existing compressive sensing (CS) theory, the accurate reconstruction of an unknown signal lies in the awareness of its sparsifying dictionary. For the estimation of channel parameters in massive multiple-input-multiple-output (MIMO) systems, however, it is impractical to set a fixed sparsifying Fourier dictionary prior due to the continuity of the angles of departure (AoD) for different wireless paths. To address this continuous optimization problem, the particle swarm optimization (PSO) algorithm was applied in this paper. Furthermore, combined with the orthogonal matching pursuit (OMP) algorithm based on the CS strategy, we develop a novel PSO-OMP channel estimation algorithm for massive MIMO system. Simulation results under different conditions demonstrate that the proposed method achieves much higher accuracy and yields superior performance in terms of bit-error-rate (BER).