Hybrid Precoding Design based on Alternating Optimization in mmWave MIMO Systems

Millimeter waves (mmWave) offer significant potential for wireless telecommunications due to their expansive bandwidth. Practical constraints limit the ability to equip all antennas with radio frequency (RF) chains, leading to the exploration of hybrid beamforming architectures. This approach faces challenges due to phase shifter limitations and non-convex optimization problems. This paper introduces a versatile algorithm suited to various hybrid beamforming architectures, utilizing the Phase Coherence Method (PCM) to derive closed-form solutions for the baseband and RF matrices. Our findings demonstrate the effectiveness of alternating optimization, achieving high performance with minimal iterations. Simulations indicate that a hybrid beamforming structure can mimic a fully digital beamformer when the number of RF chains is double the number of data streams, regardless of the number of antenna elements. Comparative analyses with benchmark algorithms further underscore the superiority of the proposed algorithm.