A Coalitional Game-based Trust Management For Distributed Cognitive Radio Networks

In order to solve the increasingly shortage of spectrum resources, cognitive radio networks allows secondary users to lease licensed spectrum without affecting primary users. The cognitive characteristics of cognitive radio makes untrusted secondary users can cause harmful interference to the primary users easily. While, it can also cause internal competition among secondary users, which leads normal users can hardly access the available spectrum. Obviously, these concepts will introduce trust problems into the network both between the primary users and secondary users and between the secondary users and secondary users. Most of the traditional trust solutions only use secondary users’past trust to estimate their future behaviors. However, cognitive radio’s intelligent perception and learning characters will make the prediction not easy. In addition, for distributed cognitive radio networks, since there is no perfect trusted third party, competition for resources between secondary users will become increasingly fierce. Therefore, it’s time to design an efficient and reasonable trust management mechanism to reduce the malicious competition among the SUs in distributed cognitive radio networks and to achieve fair allocation of spectrum resources.Based on the adaptive features of the cognitive radio networks, this thesis chooses to take advantage of the basic principles of coalitional formation games to model the problem of resource completion in cognitive radio networks. The probability to get the available spectrum is modeled as the utility value of the secondary users. Besides the trust degree in the traditional trust management mechanism, the participation degree and the need degree of the secondary users are added to user’s utility function to encourage secondary users’participation of the network activities and improve the fairness of spectrum allocation. Meanwhile, this thesis also proposes the concept of spectrum attribution vector, taking account of the different properties of different spectrum. It allows spectrum to be allocate according to its unique attribution, by which the utilization rate of the free spectrum can be improved greatly. Simulation and analysis show that, this new trust management mechanism can improve the spectrum utilization while boosting network throughput, as well as improving system security, data fusion accuracy, and the fairness of spectrum allocation.