A Collaborative Data-Driven Intelligence for Future Wireless Networks
Rashid Ali, Hyung Seok Kim · 2023
Future wireless networks, such as IEEE 802.11be (Wi-Fi 7), are vital to provide ubiquitous ultra-reliable and low latency communication services with massively connected devices to beyond 5th generation (B5G) communication technologies. Amalgamating Wi-Fi networks with B5G networks has attracted strong researcher interest over the past two decades, because over 70% of mobile data traffic is generated by Wi-Fi devices. However, Wi-Fi channel resource scarcity for B5G is becoming ever more critical. One current problem regarding channel resource allocation is channel collision handling due to increased user densities. Reinforcement learning (RL) algorithms have recently helped develop prominent behaviorist learning techniques for resource allocation in B5G networks. An agent optimizes its behavior in an RL-based algorithm based on reward and accumulated value. However, densely deployed Wi-Fi environments are distributed and dynamic, with frequent changes. Thus, relying on individual local estimations without considering the distributed data-driven intelligence leads to higher error variance. Therefore, this chapter proposes a collaborative data-driven intelligent channel resource allocation framework for B5G networks and suggests collaborating learning estimates of local RL models for faster learning convergence. Experimental results verify that the proposed approach optimizes Wi-Fi performance in terms of throughput by collaborative channel access parameter selection. This chapter also highlights six potential applications for the proposed framework.