Secure cooperative spectrum sensing and access against intelligent malicious behaviors.

Wei Wang,Lin Chen,Kang G. Shin,Lingjie Duan

DOI: https://doi.org/10.1109/INFOCOM.2014.6848059

2014-01-01

Abstract:Sensing falsification is a key security problem in cooperative spectrum sensing for cognitive radio networks. Most previous approaches assume that malicious users only cheat in their sensing reports following a predefined rule. However, some malicious users usually act intelligently to strategically adjust their malicious behavior according to their objectives and the network's defense schemes. The existing schemes cannot resist the malicious behaviors of intelligent malicious users (IMUs) without long-term collection of information on their reputation. In this paper, we construct a moral hazard principal-agent framework and design an incentive compatible mechanism to thwart the malicious behaviors of rational and irrational IMUs. We find that neither spectrum sensing nor spectrum access alone can prevent the malicious behavior without any information on users' reputation. According to the analysis of malicious behavior resistance methods, we propose a joint spectrum sensing and access mechanism to optimally prevent the IMUs from sensing falsification. Our evaluation results show that the proposed mechanism achieves almost the same performance as the ideal case with perfect sensing.