Study On Cognitive Radio Security

Cognitive Radio (CR) can sense its environment and alter its communication parameters so as to dynamically access white spectrum without interfering incumbent users. It's a smart spectrum sharing technology to effectively alleviate the spectrum shortage problem and improve the utilization of radio resources.Real-Time spectrum sensing and intelligence are two special functions of CR. They bring about new security problems which traditional wireless networks never meet. During spectrum sensing, cognitive users exit channel once sensing incumbents'signals. An adversary transmits signals emulating those of incumbent signals to prevent normal CR users accessing, even interfere with incumbents'communication. This malicious behavior can severely reduce network performance. Besides, at the mention of intelligence, Cognitive Radio makes decisions by spectrum sensing and learning. Malicious users tamper certain parameters to prevent CR reaching to Self-Adaptation, and finally mislead decisions by influencing memory bank of the network.In this paper we propose a scheme to defend against primary user emulation (PUE) attacks after in-depth study on Cognitive Radio security. The scheme based Hash function can distinguish adversary by detecting signals. Besides, we propose a multi-objective programming model, called MOP by analyzing Objective Function Attacks, and construct fitness value function for sub-objectives with Particle Swarm Optimization(PSO) algorithm.Our simulation results indicate that the MOP model can defend against Objective Function Attacks effectively in the context of optimal network configuration.