Association rule mining for detection of colluding SSDF attack in Cognitive Radio Networks

We use collaborative sensing in a cognitive radio network for a more accurate sensing decision. Individual radio nodes sends its local sensing report to the fusion center (FC), which combines the local sensing reports using some fusion rule thereby arriving at a final sensing decision. However, a malicious node may launch the Spectrum Sensing Data Falsification (SSDF) attack in which the local sensing report is falsified before it reaches the fusion center (FC). This is done so as to adversely influence the final sensing decision. The attack can be more severe if the malicious nodes collude. In this paper, we propose a detection scheme based on Apriori algorithm which detects colluding SSDF attack. We support the validity of the proposed scheme through extensive simulation results.