Security Threat Analysis in 5G Cognitive Radio Networks: A Deep Learning Ensemble Approach

Abstract

Spectrum constraints are a significant problem in the field of communication networks. A cutting-edge wireless communication technology called cognitive radio allows devices to maximize spectrum consumption and adjust to their surroundings dynamically. Despite its promise, cognitive radio technology has several security vulnerabilities that endanger the network. Cognitive radio security is crucial to accomplish dynamic spectrum access. We can ensure that cognitive radio technology is deployed and operated securely by being aware of and responding to these security concerns. In cognitive radio, artificial intelligence is crucial in identifying malevolent users. This work employs an ensemble of long-term, and short-term, GRU approach to distinguish fraudulent users from authorized users. The suggested method for detection was implemented on data sets containing several parameters, including SNR and modulation scheme energy. The proposed algorithm shows compelling evidence of outperforming the state-of-the-art algorithms in detection.