Energy Efficient Hybrid Precoder for Cell-Free Wideband mmWave Massive MIMO Systems

In a cell-free (CF) wideband millimeter-wave (mmWave) massive multiple-input multiple-output (mMIMO) system, a small number of users are simultaneously served by a much larger number of distributed access points (APs). These APs employ hybrid precoders to achieve beamforming gain. However, the beam squint caused by the spatial-wideband effect deteriorates the system performance. Moreover, the application of numerous radio frequency (RF) chains increases the power consumption and reduces the global energy efficiency (GEE). The letter focuses on designing energy efficient wideband hybrid precoders using low-resolution phase shifters (PSs) for fully connected (FC) and subarray-based phase shifter (SPS) architectures. A beam squint compensation (BSC) algorithm is proposed based on an iterative heuristic Gram-Schmidt algorithm to design the wideband hybrid precoder with the maximal number of RF chains. Based on the above wideband hybrid precoders, the number of RF chains is optimized as a shortest-path problem and solved by a novel bidirectional greedy tree search algorithm. Simulation results demonstrate that the proposed wideband hybrid precoder design can improve the GEE performance.