Cognitive radio is an intelligent and adaptive radio that improves the utilization of the spectrum by its opportunistic sharing. However, it is inherently vulnerable to primary user emulation and jamming attacks that degrade the spectrum utilization. In this paper, an algorithm for the detection of primary user emulation and jamming attacks in cognitive radio is proposed. The proposed algorithm is based on the sparse coding of the compressed received signal over a channel-dependent dictionary. More specifically, the convergence patterns in sparse coding according to such a dictionary are used to distinguish between a spectrum hole, a legitimate primary user, and an emulator or a jammer. The process of decision-making is carried out as a machine learning-based classification operation. Extensive numerical experiments show the effectiveness of the proposed algorithm in detecting the aforementioned attacks with high success rates. This is validated in terms of the confusion matrix quality metric. Besides, the proposed algorithm is shown to be superior to energy detection-based machine learning techniques in terms of receiver operating characteristics curves and the areas under these curves.

认知无线电是一种智能的自适应无线电，它通过机会共享提高了频谱的利用率。但是，它固有地容易受到主要用户仿真和干扰攻击的影响，从而降低频谱利用率。本文提出了一种在认知无线电中检测主要用户仿真和干扰攻击的算法。所提出的算法基于基于信道的字典上的压缩接收信号的稀疏编码。更具体地，根据这样的字典的稀疏编码中的会聚模式用于区分频谱空洞，合法的主要用户以及仿真器或干扰器。决策过程是基于机器学习的分类操作。大量的数值实验表明，该算法在检测上述攻击中具有很高的成功率。这根据混淆矩阵质量度量进行了验证。此外，在接收器的工作特性曲线和这些曲线下的面积方面，该算法被证明优于基于能量检测的机器学习技术。