Multi-IRS-aided Millimeter-wave Massive MIMO with Energy-Efficient Hybrid Precoding Schemes

Intelligent reflecting surfaces (IRS)-aided multiuser millimeter-wave (mmWave) massive multiple input multiple output (mMIMO) systems have recently attracted a great interest as an attractive paradigm for the future 6G wireless communications. The main benefits of these systems include achieving a transmission in a cost-effective, energy-efficient and intelligent manner. In this paper, multi-IRS-aided systems are introduced as a promising technique to assist the adaptive cross entropy (ACE)-based hybrid precoding, which is a machine learning (ML) inspired energy-efficient hybrid precoding scheme that requires a small number of low-cost and energy-efficient switches and inverters. Extensive computer simulations have been conducted to evaluate the achievable sum-rate performance and the energy efficiency (EE) of the proposed multi-IRS-assisted mMIMO systems with ACE-based hybrid precoding architecture. For the sake of fair comparisons, the conventional precoding schemes are also evaluated. The simulation results demonstrate the remarkable advantages of the proposed scheme compared with the conventional state-of-the-art schemes. It is also shown that the EE can be further improved as the number of cascaded IRSs and their elements increase. This makes the proposed system a promising candidate for 6G wireless communications.