Deep Reinforcement Learning for spectrum coexistence between future 6G networks and incumbent systems

In this dissertation, a study was conducted on the use of Machine Learning techniques to address the problem of spectrum coexistence between 5G New Radio and incumbent systems in the context of future 6G networks. After an introduction to the coexistence challenges in next-generation wireless environments, the focus was placed on the application of Deep Reinforcement Learning for interference-aware radio resource management. A learning agent was developed to control the behaviour of the cellular system — through dynamic user scheduling — while assuming that incumbent systems remain unchanged. The proposed method was evaluated in simulated coexistence scenarios, analysing standard performance metrics to assess its effectiveness.