Spectrum occupancy-based PUEA detection using SVM-PSO in cognitive networks. (20th October 2020)

• Record Type: Journal Article
• Authors: Rajagopala, M.V.; Lingareddy, S.C.
• Is Part Of: International journal of communication networks and distributed systems. Volume 26:Number 1(2021)
• Page Start: 30
• Page End: 49
• Publication Date: 2020-10-20
• Subjects: primary user emulation -- PUE -- attack -- spectrum occupancy -- kernel -- bias -- fast energy detector; Computer networks -- Periodicals; Telecommunication systems -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; 004.6
• Languages: English
• ISSNs: 1754-3916

Abstract

Cognitive radio (CR) is an evolutionary technology for next-generation wireless networks in order to solve the spectrum scarcity problem and efficiently utilise the limited spectrum resources. The potential security threat in cognitive radio networks is the primary user emulation (PUE) attack, which causes failure to dynamic access within the network. In our proposed method, we detect the attack by three phases. In the first phase, the occupancy and status of the spectrum are calculated. The spectrum occupancy is identified by using support vector machine (SVM) method. The optimal selection of kernel and bias values are done by particle swarm optimisation (PSO) in the second phase. By multiple thresholds based fast energy detector, the PUE attack is detected in the final phase. The proposed method is evaluated in the MATLAB platform and the results are analysed to prove the classification accuracy of the system.