Joint User Association and Resource Allocation for Smart-Grid Powered Wireless Networks Under Constrained Carbon Emission

As small base stations (SBSs) are densely deployed to meet the explosive communication demands, the increases in energy consumption and carbon emissions call for the green and sustainable design of next-generation cellular networks. In this paper, we jointly optimize user association and resource allocation for a wireless network powered by a combined smart grid (SG) and renewable energy. We reduce the system transaction cost under constrained carbon emission while ensuring users’ quality-of-service (QoS). Generalizing the stochastic dual-subgradient method, an online resource allocation (ORA) approach is proposed to decide on beamforming and energy trading of SBSs online. A Multi-Attribute Decision Making (MADM)-based user association scheme is developed by leveraging the historical transaction cost of the SBSs derived by the ORA, energy harvesting rates, and distances between the SBSs and users through user mobility prediction. The feasibility and asymptotic optimality of the ORA are analyzed. Simulations show our algorithm cuts down the average system cost by 62.3% compared to the benchmarks.