Cognitive radio network is one of the proposed solutions to handle the scarcity of spectrum in the wireless communications systems. Spectrum sensing is the key step in the cognitive radio implementation and any problem in this functionality leads to the severe degradation in network performance. In this paper, a smart attacker is introduced which during the sensing period time, senses the spectrum and whenever primary user leaves, enters to channel. Then by emulating the behavior of the primary user, the attacker tries to reduce the spectral efficiency. We model the primary user and smart primary user emulation attacker traffics by a time dependent Markov process. By calculating the performance of the cognitive radio network and throughput of secondary user’s network in the presence of the smart primary user emulation attacker, the impact of the traffic parameters over such performances are evaluated. Also launching primary user emulation attack in conventional and smart attacking methods are compared. The simulation results demonstrate that in the case of smart attacking, the throughput of the secondary user’s network degrades considerably and if the appropriate attack traffic parameters are chosen, throughput might get close to zero.
 

Keywords: Cognitive radio networks, Spectrum sensing, Network security, Smart primary user emulation attacker, Network throughput.

Full-Text [PDF 1243 kb]   (687 Downloads)