Enhancing Uplink Communication with Multi-User Detection in NOMA Through Deep Neural Networks

Abstract

In the realm of wireless communication, Multi-user Detection (MUD) techniques have become crucial for ensuring efficient and reliable transmission in complex network scenarios. Particularly in the uplink channel of Non-Orthogonal Multiple Access (NOMA) systems, effective MUD approaches are essential to overcome interference and enhance overall performance. One innovative solution that has shown promise in addressing these challenges is the utilisation of Deep Neural Networks (DNNs) for MUD in grant-free NOMA uplink communications. By leveraging the power of artificial intelligence and machine learning, DNNs can effectively distinguish and decode signals from multiple users sharing the same frequency band in NOMA networks. By training deep learning algorithms on large datasets of multi-user signals, DNNs can learn complex patterns and correlations, enabling them to accurately separate and detect individual user signals in the presence of interference. This capability improves communication’s overall reliability and efficiency in NOMA systems and opens up opportunities for enhancing spectral efficiency and capacity.