Reinforcement Learning Approach for Hybrid WiFi-VLC Networks

Abstract

The number of mobile devices in indoor environment has dramatically increased and the capacity of conventional RF wireless networks may not be enough to support the indoor traffic demand. Recently, Visible Light Communication (VLC) systems have emerged as a complementary unlicensed media. In this paper, we proposed a hybrid WiFi-VLC system where multiple VLC access points (AP) coexist with a WiFi AP. A number of indoor users share the hybrid WiFi-VLC system. All users employ WiFi for uplink whereas one access point (WiFi or VLC) is assigned for each user to maximize the overall capacity of the network. We propose a new reinforcement learning algorithm which can be implemented at the WiFi AP and results in the selection of an access point such that the total throughput is maximized. Numerical simulation results show that the proposed method improves the total system throughput significantly. Furthermore, the throughput achieved by the worst user in the proposed Q-Learning algorithm becomes higher than what would be received by the average user who used the conventional hybrid systems based on best connection.