CoMP-Assisted NOMA and Cooperative NOMA in Indoor VLC Cellular Systems

TL;DR

This paper explores CoMP-assisted NOMA and cooperative NOMA in indoor VLC systems, proposing power allocation schemes to optimize data rates and fairness amid inter-cell interference.

Contribution

It introduces two novel CoMP-assisted NOMA schemes with tailored power allocation frameworks for indoor VLC systems, addressing interference and QoS challenges.

Findings

Proposed CoMP-assisted NOMA schemes improve data rates.

Optimized power allocation enhances user fairness.

Heuristic solutions address non-convex optimization challenges.

Abstract

In this paper, we investigate the dynamic power allocation for a visible light communication (VLC) cellular system consisting of two coordinating attocells, each equipped with one access-point (AP). The coordinated multipoint (CoMP) between the two cells is introduced to assist users experiencing high inter-cell-interference (ICI). Specifically, the coordinated zero-forcing (ZF) precoding is used to cancel the ICI at the users located near the centers of the cells, whereas the joint transmission (JT) is employed to eliminate the ICI at the users located at the edge of both cells and to improve their receptions as well. Furthermore, two multiple access techniques are invoked within each cell, namely, non-orthogonal-multiple-access (NOMA) and cooperative non-orthogonal-multiple-access (C-NOMA). Hence, two multiple access techniques are proposed for the considered multi-user multi-cell system, namely, the CoMP-assisted NOMA scheme and the CoMP-assisted C-NOMA scheme. For each scheme, two power allocation frameworks are formulated each as an optimization problem, where the objective of the former is maximizing the network sum data rate while guaranteeing a certain quality-of-service (QoS) for each user, whereas the goal of the latter is to maximize the minimum data rate among all coexisting users. The formulated optimization problems are not convex, and hence, difficult to be solved directly unless using heuristic methods, which comes at the expense of high computational complexity.