Edge to Cloud Service Placement Based on Reinforcement Learning in 6G Networks

Telecommunications are undergoing a transformation due to the introduction of 5 G and the upcoming 6G technologies, which promise previously unheardof speeds, extremely low latency, and widespread access. The resource management of edge, fog, and cloud infrastructures is facing new difficulties as a result of these improvements. In order to optimize performance, cost-efficiency, and energy consumption, this paper investigates how machine learning (ML) might be used to improve resource allocation to services across this continuum, specifically in the context of B5G and 6G networks. In this work, we propose a reinforcement learning based algorithm using Q-learning for determining the best locations of services across the 6 G edge to cloud continuum. The aim is to ensure better QoS to the users and efficient use of network resources. To evaluate the performance of the proposed approach we compare it with a traditional state of the art algorithm in the context of content placement in CCNs, taking into consideration network related metrics such overall network traffic and delay. The experimental analysis shows that the proposed algorithm performs better than the traditional state of the art algorithm in most cases and reveals some insights on how network topological configurations affect the performance.