Enhancing wireless communications using Machine Learning

18/03/2025 - Guy Anthony NAMA NIAM, Doctorant

Today, wireless communication networks face numerous challenges, such as ensuring quality of service (QoS) in terms of throughput, latency, and quality of experience (QoE), while also guaranteeing reliability and massive access for users. The proliferation of connected wireless devices generates a huge volume of data, particularly with the surge in popularity of multimedia services (streaming, video conferencing, etc.). It is therefore essential to design next-generation networks (NGN) to meet this ever-growing demand. Given the increasing complexity of systems and the exponential growth in the amount of data exchanged over networks, intelligent and autonomous services and applications should be deployed on the user side, leveraging both established and emerging technologies, particularly artificial intelligence (AI) and machine learning (ML). For instance, integrating deep reinforcement learning (DRL) can support scalable and diverse services and applications. Indeed, network intelligence could be implemented directly on programmable devices, enabling faster responses to network events without relying on time-consuming centralized exchanges. Furthermore, NGNs should be self-learning, self-reconfigurable, self-optimized, self-healing and self-organizing, ensuring dynamic and resilient network management.