Machine Learning Inspired Hybrid Precoding for Wideband Millimeter-Wave Massive MIMO Systems

Millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) has already been considered as a promising solution to meet the requirement of the higher data rate for the future Internet of Things (IoTs). Hybrid precoding is an effective solution for the mmWave massive MIMO systems to significantly decrease the number of radio frequency (RF) chains without an apparent sumrate loss. However, the current literature on hybrid precoding considers either the high-resolution (HR) phase shifters (PSs) with high power consumption or the impractical narrowband mmWave channel model. To this end, in this paper, we investigate an energy-efficient hybrid precoding scheme using one-bit PSs for practical frequency-selective wideband mmWave massive MIMO systems. Specifically, we provide the energy consumption analysis to reveal the fact that the energy consumed by the one-bit PSs is much lower than that by the HR-PSs, and the array gain loss incurred by using one-bit PSs is minimal. Moreover, motivated by the cross-entropy optimization (CEO) algorithm evolved for machine learning, we propose the CEO-based hybrid precoding scheme to maximize the achievable sum-rate of the considered system. In the CEO-based hybrid precoding, we update the probability distributions of the elements in the hybrid precoder to minimize the cross-entropy between the two probability distributions so that we can generate the final solution close to the optimal one. Furthermore, we extend the proposed CEO-based hybrid precoding scheme from the case with one-bit PSs to the general case with HR-PSs to show that our solution can also be applied in other scenarios. The performance evaluation demonstrates that our proposed scheme can obtain near-optimal sum-rate and considerably higher energy efficiency than some existing solutions.