Research On The Evaluation Method Of Abnormal Cooperative Nodes For Cognitive Networks

Along with the fast-growing of wireless technology and the growing of user's needs, the radio spectrum which is in short supply faces serious and nervous shortage problem.The appearance of Cognitive Wireless Networks (CWN) has provided the significant mentality to solve the lack of the radio spectrum and to improve the utilization of the existing spectrum resource, has become a hot spot of research in recent years.By sensing the surrounding environment dynamically and carrying on the study and analysis, and then it will make the appropriate planning and decision-making based on the current situation, in order to achieve"secondary utilization" of the spectrum resources.However, the communication environment of opening and sharing makes CWN more vulnerable to suffer security attacks, especially the interference and attack from malicious entities and selfish entities.This would undermine the credibility of the sensing data and affect the fairness of spectrum allocation.The traditional network encryption and decryption,authentication and other security technology can not solve the internal security threat effectively. The reputation mechanism as a supplement of the traditional security scheme, can effectively solve the problem of internal network security. So, this paper proposes a anomaly cooperative node evaluation mechanism for CWN by drawing the theory of the reputation theory as reference and combining own characteristics of CWN.This mechanism starts from the cognition process to counter the malicious entity and selfish entities.The mechanism can timely discovery malicious and selfish entities by evaluating each entity's behavior,in order to enhance substantive cooperation and provide the security performance of network.The main work and innovation of this paper are following:Firstly, we analyze the existing security threats in CWN, and tease out the research context of security mechanism in CWN.Then, we analyze the background and motivation of using reputation mechanism in CWN.(2) Based on the existing centralized network topology, we propose a CWN topology based on the partition of location according to the characteristics of CWN.This topology not only considers the influence of location for the cognitive user's sensing apability,but addresses the gapof the topology control.This topology is also to provide the basis for the proposing of the reputation evaluation mechanism.(3) The existing methods always study each protocol layer threat in isolation and lack the considerations for the whole cognitive process and not to combine the reputation with privoding cognitive ability.This paper establishes a reputation evaluation mechanism on the basis of the network topology based on the partition of location.This mechanism improves the reliability of cognitive ability and guarantees the fairness of spectrum access by evaluating the node's reputation statusto find abnormal entity, and to deal with by the reward and punishment mechanism.(4) At last,we verify the effectiveness and superiority of the proposed mechanism to assess and resist malicious entities and selfish entities through simulation.