QoE based Joint Adaptive User Grouping and Power Allocation in Indoor VLC-NOMA Network

Abstract The growing demand for high data rate and the increasing scarcity of spectrum resources have attracted non-orthogonal multiple access (NOMA) into visible light communication (VLC) network. For improving user’s quality of experience (QoE) in VLC-NOMA network, a joint optimization problem of user grouping and power allocation is formulated to maximize the sum mean opinion scores (MOS) of users in the VLC-NOMA network. Then, we decompose it into an inner power allocation and an outer user grouping subproblems. For the inner power allocation subproblem, we propose a power adjustment-based power allocation (PAPA) algorithm to guaranteeing the QoE and successive interference cancellation (SIC). For the outer user grouping subproblem, a hierarchical agglomerative clustering-based (HAC) user grouping algorithm is proposed that users can adaptively select NOMA groups and transmission modes. Then, we propose a joint adaptive user grouping and power allocation (JAUGPA) algorithm to solve user grouping and power allocation iteratively. The simulation results show that the proposed algorithm can increase the sum MOS and guarantee the fairness of users in the VLC-NOMA network.