Jamming Attack Identification in Cognitive Radio Networks: An Adversarial Behavior Perspective

Cognitive Radio Networks (CRNs) enable opportunistic spectrum utilization, but their performance is compromised by adversarial users that exploit vulnerabilities in spectrum sensing and management. Two major adversary categories are selfish and malicious users. Selfish users misreport sensing results or occupy channels for personal gain, often due to malfunctioning sensors or spectrum hoarding strategies. Malicious users, by contrast, deliberately disrupt CRNs through targeted attacks, including Primary User Emulation Attacks (PUEA), Spectrum Sensing Data Falsification (SSDF), and jamming. These attacks degrade detection probability, increase interference with licensed users, and reduce network throughput. Although prior research has extensively examined jamming detection, the classification and distinction between selfish and malicious adversaries remain insufficiently addressed. This work analyzes adversarial behavior in CRNs, focusing on differentiating selfish misuse from intentional malicious disruption. By characterizing attack patterns, evaluating their impact on sensing performance, and identifying classification gaps, this study provides a foundation for developing robust adversary detection frameworks. The ultimate objective is to strengthen CRN resilience against heterogeneous threats and ensure reliable spectrum access.