Low‐complexity near‐optimal signal detection for uplink large‐scale MIMO systems

The minimum mean square error (MMSE) signal detection algorithm is near‐optimal for uplink multi‐user large‐scale multiple‐input–multiple‐output (MIMO) systems, but involves matrix inversion with high complexity. It is firstly proved that the MMSE filtering matrix for large‐scale MIMO is symmetric positive definite, based on which a low‐complexity near‐optimal signal detection algorithm by exploiting the Richardson method to avoid the matrix inversion is proposed. The complexity can be reduced from 𝕆( K 3 ) to 𝕆( K 2 ), where K is the number of users. The convergence proof of the proposed algorithm is also provided. Simulation results show that the proposed signal detection algorithm converges fast, and achieves the near‐optimal performance of the classical MMSE algorithm.