Alternating Optimization Based Low Complexity Hybrid Precoding in Millimeter Wave MIMO Systems

Abstract

Hybrid precoding is an effective solution to reduce the cost of millimeter wave systems. In recent years, joint optimization has begun to attract the attention of researchers due to its excellent performance. In this letter, we adopt the alternating optimization method to design a low complexity hybrid precoding. In particular, the spectral efficiency optimization problems are simplified to sub-optimization problems under certain constraints. For each sub-optimization problem, we alternatively optimize the columns of the analog precoding matrix and combining matrix by using the projected gradient method. Then, the digital domain matrices are derived from the equivalent channel. Simulation results and complexity analyses show that the performance of the proposed hybrid precoding based on alternating optimization is close to the unconstrained optimal precoder and the algorithm has lower complexity.