Mitigating primary user emulation attacks using analytical model

Abstract

Cognitive radio has been a widely studied area to increase the efficiency of the spectrum. Most of the present work focuses on sensing techniques but very less work has been done for security in Cognitive Radio Networks (CRNs). Primary User Emulation Attack (PUEA) is a very prominent Denial of Service (DoS) attack that degrades the performance of the network to a large extent. This paper deals with an analytical model which depends on Neyman–Pearson Composite Hypothesis Test (NPCHT) to determine whether a PUEA is present in CRN. Log-normal shadowing and Rayleigh fading have been taken for the signals received from the primary transmitters as well as the attackers. The movement of good secondary user has been assumed in vertical direction ranging from an angle of (–π/4) to (π/4) with respect to primary transmitter. The performance of the system model has been evaluated by plotting the Receiver Operating Characteristic (ROC) curve. Variation in miss detection and false alarm probabilities has been studied for different angles. Results demonstrate that the likelihood of successful PUEA increments with the increasing distance between the good secondary user and primary transmitter.