Sequential Load Balancing for Link Aggregation Enabled Heterogeneous LiFi WiFi Network

Abstract

Light fidelity (LiFi) is one of the promising communication technology for 6G internet of everything (IoE), however, it requires a line-of-sight (LoS) link; in contrast, WiFi can support moderate data rates even in the absence of LoS. As the electromagnetic spectrums of LiFi and WiFi do not overlap, these technologies can be used for concurrent communication, thus, resulting into a link aggregation (LA) enabled heterogeneous LiFi WiFi networks (HLWN). However, for the most efficient utilization of a LA enabled HLWN, proper load balancing is essential. Therefore, in this paper, we propose a novel sequential load balancing method with reinforcement learning (RL)-based access point (AP) assignment followed by optimum resource allocation for LA enabled HLWN. It is observed that the proposed method outperforms the baseline received signal strength (RSS) scheme by around 37% and 56% in terms of average data rate and user satisfaction, respectively. Furthermore, the proposed method performance closely matches to the exhaustive search with the added advantage of reasonably low complexity. Additionally, the robustness of the proposed method is proven by considering two different user’s mobility models in this work.

Keywords

• Heterogeneous LiFi WiFi network (HLWN)
• link aggregation (LA)
• LiFi
• resource allocation (RA)
• reinforcement learning (RL)