Research On Security Issues For Cooperative Spectrum Sensing In Cognitive Radio Networks

With the rapid development of wireless communication technologies, the spectrum requirement increases quickly and the current static spectrum allocation policy already cannot meet the radio users’ demands. Meanwhile, the allocated spectrum is utilized efficiently, and the spectrum scarcity coexists with low utilization. The cognitive radio(CR) technology makes dynamic spectrum access possible. It has enabled the opportunity to transmit signals in several idle licensed bands without causing the harmful interference to primary users(PUs). Thus, the spectrum utilization and the satisfaction of radio users’ requirements can be improved.Spectrum sensing is a premise for the realization of cognitive radio networks(CRNs). For improving the accuracy of sensing, the cooperative spectrum sensing technology has been proposed, which uses spatial diversity to overcome the multi-path fading, shadowing and receiver uncertainty issues for the local spectrum sensing in CRNs. But the cooperative spectrum sensing is not able to solve the security issues in the spectrum sensing phase, either.The main purpose of this dissertation is to study the security issues for cooperative spectrum sensing in CRNs. The main work and contributions of this dissertation are as follows:Firstly, the current local spectrum sensing and cooperative spectrum sensing, including the main technical characteristics, are analyzed and discussed.Secondly, the security issues for cooperative spectrum sensing in the hostile environment are summarized and analyzed. The models of the primary user emulation(PUE) attack and the spectrum sensing data falsification(SSDF) attack are established. The influences on sensing results are presented, which demonstrate the severe performance degradation for a CR system, under these two kinds of attacks.Then, a secure sensing scheme based on Helper Nodes is proposed for identifying a mimic user and protecting from the SSDF attacks. These nodes playing the assistant role are deployed within the geographical area of the CRN, using the link signatures and cryptographic mechanisms to identify the activity of PU. The corresponding fusion algorithm based on reputation is also proposed. Simulation results demonstrate that the proposed scheme can efficiently defend against attacks in hostile environment.Finally, for the presence of PUE attacker in CRN, a fusion scheme based on the maximal ratio combination(MRC) is proposed. As a PUE attacker is launched in a CRN with cooperative sensing technique, each secondary user(SU) will receipt the signals which are emitted by the attackers while receiving the signals coming from the PUs. To maximize of the detection probability of PU, the optimal weighting scheme is proposed. The studying works of this dissertation show that the proposed scheme can be used for defend PUE attack in the hostile environment. The higher performance for eliminating the malicious interference on the detection capacity can be achieved while increasing the number of cooperative SUs, according to the simulation results.In summary, this dissertation studies the influence of the two main attacks, PUE and SSDF, on the cooperative spectrum sensing in CRNs. The feasible schemes for rightly treating the threats are also studied for the cooperative sensing with offering the solutions to facilitate the application of CR technology.