Abstract

The study and the detection of possible network attacks are essential for wireless networks, in particular for mobile cognitive radio networks due to its characteristics such as the dynamic spectrum allocation and constant frequency hopping. The primary user emulation attack is one of the most significant attacks in cognitive radio, because it hazards the complete cognitive cycle. The techniques used for the detection of primary user emulation found in the literature are based on a fixed attacker location. However, in a mobile environment, the attacker usually has dynamic locations and this compromises the current applied security techniques and generates inefficient attack detection. Therefore, our work proposes a novel technique using cross-layer design for the detection of primary user emulation with mobility. This attack detection technique was tested with experiments using software-defined radio equipment and mobile phones at indoor scenarios with dynamic locations and with a mobile phone base station built up also with software-defined radio. The obtained results show that the combination of the three utilized techniques, energy detection, motion estimation, and application information analysis, are able to optimize the detection with around 100% of effectiveness for the primary user emulation attack with dynamic location. The proposed technique shows that the energy detection time is around 100 ms and for the processing time of the information analysis in the mobile phone is about 30 s. This result shows a practical and effective approach to detect primary emulation attacks. The proposed technique, to the best of the authors’ knowledge, has not been presented before in the literature with experiments neither with mobility conditions of the attacker as presented in our proposed work.

中文翻译：

---

使用USRP在移动认知无线电网络上检测动态位置主要用户仿真

摘要

由于其特征（例如动态频谱分配和恒定跳频），对可能的网络攻击的研究和检测对于无线网络至关重要，尤其是对于移动认知无线电网络而言。主要用户仿真攻击是认知无线电中最重要的攻击之一，因为它危害整个认知周期。文献中发现的用于检测主要用户仿真的技术均基于固定的攻击者位置。但是，在移动环境中，攻击者通常具有动态位置，这会损害当前应用的安全技术并产生无效的攻击检测。因此，我们的工作提出了一种使用跨层设计来检测具有移动性的主要用户仿真的新技术。该攻击检测技术已通过在室内场景中使用软件定义的无线电设备和移动电话进行动态实验，并在移动电话基站也安装了软件定义的无线电的情况下进行了测试。所获得的结果表明，将三种利用的技术（能量检测，运动估计和应用程序信息分析）相结合，能够对具有动态位置的主要用户仿真攻击以100％左右的有效性优化检测。所提出的技术表明能量检测时间约为100毫秒，而手机中信息分析的处理时间约为30 s。此结果显示了一种检测主要仿真攻击的实用有效方法。据作者所知，所提出的技术，