Enhanced Primary User Emulation Attack Inference in Cognitive Radio Networks Using Machine Learning Algorithm

Abstract

Cognitive Radio (CR) is a competent technique devised to smart sense its surroundings and address the spectrum scarcity issues in wireless communication networks. The Primary User Emulation Attack (PUEA) is one of the most serious security threats affecting the performance of CR networks. In this paper, machine learning (ML) principles have been applied to detect PUEA with superior decision-making ability. To distinguish the attacking nodes, Reinforced Learning (RL) and Extreme Machine Learning (EML-RL) algorithms are proposed to be based on Reinforced Learning (EML). Various dynamic parameters like estimation error, attack detection efficiency, attack estimation rate, and learning rate have been examined with the Network Simulator 2 (NS2) tool.