Deep learning based receiver for downlink NOMA system

• Main Author: Lim, Wei Kang
• Other Authors: Teh Kah Chan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Deep learning
• Online Access: https://hdl.handle.net/10356/176804
• Institution: Nanyang Technological University

In the technologically advanced age that we live in, the transmission of data in fifth Generation(5G) has been critical in serving different masses across diverse demographics. Specifically in the downlink stream of data, due to the massive amount of people using the internet, utilizing the most optimal channel access method is crucial in the assurance of massive connectivity while ensuring high throughput. Orthogonal Frequency Division Multiplexing Access is currently mainly utilized in the transmission of data in 5G, where signals are separated to different sub-carriers where there will be no interference in the timeslot. In contrast, Non-Orthogonal Multiple Access (NOMA) will allow subcarriers to be shared between users regardless of whether they have good or bad channel conditions while being more spectrum efficient, able to serve a large number of users simultaneously. Hence, sparking the avenue for research in evaluation on the performance of using a NOMA based system in downlink for the receiver. In this report, literature review is done on the critical factors relating to this project, the proposed methodology, work completed and schedule for what will be accomplished moving forward till the end of this project.