Hybrid Energy and Spectrum Efficient Wireless Network Design for 5G/6G And Wi-Fi 7/8 Applications

The migration of wireless communications from 5G to the sixth generation (6G) of cellular communications and next-generation wireless local area networks (Wi-Fi 7/8) has emerged to support ultra-high data rate, massive connectivity, and ultra-low latency. Nonetheless, these needs exacerbate both energy and spectral resource challenges, especially in ultra-dense and heterogeneous networks. The available techniques have focused dominantly on energy efficiency or spectral efficiency, making them less effective in multi-radio access technology (multi-RAT) networks. In this paper, a hybrid framework that simultaneously leverages energy and spectral efficiency for converged 5G/6G communications and Wi-Fi 7/8 networks based on the integration of resource control based on artificial intelligence (AI), cognitive spectrum management, and reconfigurable intelligent surfaces (RIS) is proposed. A model of a multi-RAT system assisted by reconfigurable intelligent surfaces has been established, and an optimisation problem that maximises energy efficiency while satisfying quality-of-service constraints has been formulated. The challenge of the resulting mathematical model being a non-convex optimisation problem has been resolved using a controller designed based on deep reinforcement learning to dynamically adjust both the power allocation to all the radios involved and the reconfigurable intelligent surfaces' phases. Simulation results have confirmed that the proposed scheme can obtain energy efficiency improvement of up to 35% compared to the typical scheme, and simultaneously, the spectral efficiency gain has also significantly improved when operating at a high frequency. The results have confirmed that integrated energy and spectral resource optimisation could provide a new solution for a sustainable 6G and Wi-Fi 8 network.