Design of deep learning based receiver for downlink NOMA system

• المؤلف الرئيسي: Chen, Shuheng
• مؤلفون آخرون: Teh Kah Chan
• التنسيق: Thesis-Master by Coursework
• اللغة: English
• منشور في: Nanyang Technological University 2024
• الموضوعات: Computer and Information Science
• الوصول للمادة أونلاين: https://hdl.handle.net/10356/173684
• المؤسسة: Nanyang Technological University

Non-Orthogonal Multiple Access (NOMA) is a key technique that enables fifthgeneration mobile communication systems to function effectively. NOMA has received high attention in wireless communication. NOMA serves users simultaneously while enhancing frequency and spectral efficiency. The main NOMA detection approach is successive interference cancellation (SIC), which decodes signals at each user equipment (UE). However, SIC has issues with error propagation and power order constraints. Deep learning (DL) makes the system computationally simpler and mitigates the problems of error propagation encountered with traditional SIC schemes. In this dissertation, we explore the DL-based method using a convolutional neural network (CNN) and long short-term memory (LSTM) model for the downlink of the NOMA system and compare the results with the conventional SIC method. Results demonstrate the effectiveness and superior detection performance of the deep learning method.