FAMAC: A Federated Assisted Modified Actor-Critic Framework for Secured Energy Saving in 5G and Beyond Networks

TL;DR

This paper proposes a federated AI framework combining actor-critic algorithms to optimize energy consumption in 5G networks by intelligently switching base stations, achieving significant energy savings while maintaining QoS.

Contribution

It introduces a novel federated assisted actor-critic framework for energy-efficient base station management in ultra-dense 5G networks, outperforming existing solutions.

Findings

Achieves approximately 77% more energy savings than state-of-the-art methods.

Effectively balances energy efficiency with quality of service constraints.

Demonstrates the effectiveness of federated learning combined with actor-critic algorithms in network management.

Abstract

The constant surge in the traffic demand on cellular networks has led to continuous expansion in network capacity in order to accommodate existing and new service demands. This has given rise to ultra-dense base station deployment in 5G and beyond networks which leads to increased energy consumption in the network. Hence, these ultra-dense base station deployments must be operated in a way that the energy consumption of the network can be adapted to the spatio-temporal traffic demands on the network in order to minimize the overall energy consumption of the network. To achieve this goal, we leverage two artificial intelligence algorithms, federated learning and actor-critic algorithm, to develop a proactive and intelligent base station switching framework that can learn the operating policy of the small base station in an ultra-dense heterogeneous network (UDHN) that would result in maximum energy saving in the network while respecting the quality of service (QoS) constraints. The performance evaluation reveals that the proposed framework can achieve an energy saving that is about 77% more than that of the state-of-the-art solutions while respecting the QoS constraints of the network.