Cognitive radio for enhancing spectral efficiency over wireless communications

• المؤلف الرئيسي: Xu, Xuanzhi
• مؤلفون آخرون: Li Kwok Hung
• التنسيق: Thesis-Master by Coursework
• اللغة: English
• منشور في: Nanyang Technological University 2024
• الموضوعات: Engineering; Cognitive radio; Spectrum sensing; Energy detection; Machine learning; Support vector machine; Multilayer perceptron; Naive Bayes
• الوصول للمادة أونلاين: https://hdl.handle.net/10356/179101

In today's wireless networks, cognitive radio (CR) stands out as a crucial technology for efficiently managing limited spectrum resources. By adapting intelligently to changing environments, CR ensures optimal spectrum usage. This dissertation focuses on improving spectrum efficiency in CR networks through cooperative spectrum sensing (CSS). By addressing the challenges involved, the study aims to enhance the feasibility of CSS-based CR networks, evaluating key metrics like detection reliability and transmission throughput. Beginning with an exploration of CR's fundamental concepts and the critical role of spectrum sensing, the dissertation conducts a comprehensive review of existing literature and spectrum sensing methodologies, primarily focusing on energy detection methods. Moreover, the dissertation extends its analysis to spectrum sensing strategies within CR networks. By considering the standalone sensing and cooperative sensing, the research explores algorithms to maximize spectrum efficiency while ensuring the protection of primary users' signals. Through systematic experimentation and analysis, the study evaluates the performance of various spectrum sensing techniques, considering factors such as variable parameters and the number of sensing samples. Besides, By applying the machine learning (ML) model in CSS, the detection accuracy and efficiency of CR systems have been enhanced. After simulation on different ML techniques, the support vector machine (SVM) presents a better training model compared with other ML techniques such as multilayer perceptron (MLP) and Naive Bayes (NB), especially at low false alarm probability. In conclusion, this research contributes to advancing the field of cognitive radio by providing insights into effective spectrum sensing strategies. The findings offer valuable implications for improving spectrum efficiency in wireless communication networks.