A Nested Two Stage Game-based Trust Management In Cognitive Radio Network

With the continuous development of the wireless network technology,the cognitive radio network(CRN)has been widely regarded as one of the most effective ways to solve the problem of spectrum shortage.Cognitive Radio(CR)can adaptively change its transmission parameters by interacting with its wireless environment,which is an intelligent radio.CRNs use idle spectrum resources to communicate to improve the spectrum utilization,mitigate the problem of wireless spectrum resources.And it can be adaptive to perceive and learn the environment,and optimize the re-certification of radio networks.However,the dynamics of the spectrum and users' cognitive characteristics has also created new security threats such as conspiracy attacks,malicious occupation channel attacks,on-off attacks,and data usurpation in the perception phase.The traditional security policy,cryptography,and access control technology can not solve the above security threats,especially the data security attacks caused by the secondary users in the perception phase,the transmission phase and the learning process.So far,one of the most effective ways to solve the soft-security threats of cognitive networks is trust management.Therefore,it is necessary to propose an effective trust management scheme for CRNs.The superiority of trust management scheme directly determines the fairness of the spectrum allocation phase and the security of the system,and the robustness of various malicious attacks in the presence of malicious users.In order to solve the cognitive wireless network adaptability,robustness and security issues,we find that the secondary users in the system can express the information and transmit data in the process of system cognition,which should be the standard of system evaluation user reputation value through the research on the related progress of trust management in CRNs.In addition,there is a competition among the second users,i.e.,game.To conclude,it is feasible to apply the nested game theory and marginal utility theory to propose a trust management scheme which can effectively resist malicious attacks based on the above research status and needs.This thesis aims to study the trust management scheme under the central architecture of CRN,and analyze the problems and challenges of trust management scheme in the CR environment.In order to solve the existing problems,the author has made a comprehensive analysis of and research on the trust management plan request.The analysis enumerates the features that the reputation value function requires.It is considered that the identification of the reputation value function should be based on the cognitive cycle.Hence the reputation evaluation function is determined based on this cognition,and the game theory is used to optimize the analysis to realize the fairness and security of spectrum allocation.This thesis has made the following achievements in the study of trust management program.1.This thesis focuses on the whole cognitive cycle.The trust management evaluation process is divided into the perception stage and transmission action phase.This thesis evaluates its historical behavior and current direct transmission behavior in two stages,and integrates the evaluation data in the data center.This makes the trust-based evaluation of cognitive cycles more objective and accurate and provide a more reliable standard for spectrum allocation.2.This thesis applies the time window mechanism,marginal utility theory,and diminishing principle based on the marginal utility.Users can participate in the spectrum competition according to the good perception behaviors when they first enter in the system so as to participate in the spectrum competition.But if users want to acquire a higher reputation value,they must keep the long-time accumulation of good behaviors so that the reputation value function can realize slow rise and quick fall to resist malicious users ' On-Off attacks.3.This thesis introduces the nested game theory to optimize the trust management scheme.The decision of secondary users at different times affects the constant change of the reputation value.In order to obtain the spectrum,the secondary users continue to game with each other and learn good behaviors around to make the most favorable decision for themselves,and this can also maximize the efficiency of the system decision.So it can achieve the maximum fair spectrum distribution under the program,resist malicious users and remove malicious users.4.Finally,the simulation and analysis of the proposed scheme of trust management shows that it has strong adaptability,and the reputation value function has the characteristics of slow rise and quick fall,which is fair to the spectrum distribution.The program has a greater resistance to On-Off attacks,which can improve the robustness and security of CRNs.