Ensemble learning of deep learning-based receiver for 5G communication system implementation

• Main Author: Xu, Ziang
• Other Authors: Teh Kah Chan
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Ensemble deep learning
• Online Access: https://hdl.handle.net/10356/181868
• Institution: Nanyang Technological University

This project explores the application of ensemble learning techniques for implementing deep learning (DL) based receivers in the fifth generation (5G) communication systems. Deep learning has shown promising results in handling the complexity of 5G channels; however, the performance of individual DL models can be limited due to the inherent variability of wireless channels. Ensemble learning offers a solution by combining multiple DL models to improve overall receiver performance. This project investigates different ensemble learning approaches, such as bagging, boosting, and stacking, and explores their effectiveness in enhancing the accuracy and robustness of DL-based receivers. Experimental evaluations will be conducted using realistic 5G channel models and performance metrics, comparing the ensemble learning approach with individual models. The findings of this research contribute to the practical implementation of DL-based receivers in 5G systems, providing insights into the benefits of ensemble learning for improved system performance and reliability.