A Novel Method of Combining Decision Making and Optimization for LiFi Resource Allocation

Light fidelity (LiFi) is a promising wireless communication technology which can be integrated into existing lighting infrastructure, fulfilling illumination and communication requirements simultaneously. Due to the relatively small coverage area of LiFi with a single access point (AP), a number of LiFi APs are usually needed to provide coverage. However, joint access point selection and resource allocation then become challenging. This issue is mainly tackled using optimization, iteration, and decision making methods. However, the current literature fail to balance the complexity and optimality well. In this paper, a novel approach that combines decision making and optimization is proposed to provide a better trade-off between optimality and complexity. Specifically, a distributed optimization is integrated into the decision making process of fuzzy logic, which is adopted to classify the user based on their channel quality and throughput requirements. Simulation results show that the mixed method can reach the near-optimal achievable throughput, reducing runtime by more than 3 orders of magnitude compared with the global optimization benchmark. Besides, the proposed method is capable of saving up to 72% runtime compared with iteration based benchmark.