Spectrum Allocation and Power Control in Full Duplex Dense Heterogeneous 5G Networks

Our Project focuses on improving 5g networks by using smart techniques in crowded and diverse environments. We aim to make data transmission faster and more reliable while minimizing interference. By using Non-Orthogonal Multiple Access (NOMA) and Deep Reinforcement Learning (DRL), we allow devices to share the network efficiently. NOMA assigns different power levels, like varying voices in a conversation. DRL teaches the network to manage resources wisely, adapting to changes. Our algorithms allocate spectrum and control device power, like a traffic manager for data. Simulations show our approach boosts speed, reduces confusion, and ensures fair resource sharing. In the future, these innovations could lead to faster downloads, smoother connections, and a smarter wireless experience. Our project advances the journey towards a more connected and efficient 5G world.

Keywords: Non-Orthogonal Multiple Access, Deep Reinforcement Learning, Data Transmission, Interference, Resource