The Research Of Defensing Against Primary User Emulation Attack In Cognitive Radio Networks

The purpose of the development of cognitive radio technology to improve spectrum resource utilization,to solve the shortage of spectrum resources increasingly appears the status quo.But due to the special features and functions of cognitive radio,make it not only has security problems of general wireless communication,but also brings new security threats.Primary User Emulation(PUE)Attack is one of the security threats,it through emulates the characteristics of primary user signal to pretend to be the primary user,and to achieve spectrum precedence of other cognitive users reaching priority or useing a frequency bandwidth alone,PUE attcsk will make other cognitive users can't normally access idle spectrum,the normal cognitive wireless network communication have much damage.The PUE attack will seriously interfere with the spectrum sensing process and significantly reduce the available spectrum of cognitive users.In this paper,we will discuss and research the PUE attack in the cognitive radio network.In the situation of the location of primary user is fixed,in this paper,a method which based on the BP neural network is proposed to defense the primary user emulation attack.Individual cognitive users used the received the primary user's signal strength and their own local two-dimensional coordinates as BP neural network's input,and it will directly output the detection results of the primary user.Then cognitive users will sent the detection results to the data fusion center of cognitive radio network,according to the linear fusion rule,the fusion center makes a final judgment of all individual cognitive users's detection results,thus finally reach the purpose of identifying the primary user identity.When primary user is moving,the method of defensing the primary user emulation attack is based on the method which uses the BP neural network when the loacation of primary user is fixed.Specific improvement step is that we will divide the time that the primary user signal begins to appear until the end divided into k moments,in every moments,we can assume the position of the primary user is fixed.So we can use the method which based on BP neural network to detect the PUE attacsk in every moments,and we also can get the judgements which the fusion center gathers cognitive users's detection results in every moments,finally,we will reach the purpose of identifying the primary user identity.In the procedure of the cognitive users measure the signal strength of primary user,we use the Kalman filter algorithm to smooth the received signal strength(RSS)value.The proposed method to against the primary user emulation attack is more accurate than the traditional method,not only solved the primary user emulation attack when the situation of the primary user is static,but also puts forward the corresponding methods when the primary user is moving.