基于ESD测试的无线电网络抗SSDF攻击协作频谱感知方案

针对认知无线电网络中协作频谱感知容易遭受数据伪造攻击的问题，提出一种基于检验统计和极端学生化偏差检验法的协作频谱感知方案。首先，将差分进化算法与加权增益合并软决策融合方法相结合，形成一种高效的节点决策融合机制。然后，在协作感知中，根据节点的软决策数据，利用检验统计消除故障认知节点。最后，利用提出的改进型ESD检验法消除恶意认知节点，从而形成全局决策。仿真结果表明，该方案在协作感知中能够有效过滤SSDF攻击数据，具有较低的误检测率。

Cooperative spectrum sensing scheme against SSDF attack based on ESD test in radio network

Aiming at the problem that the cooperative spectrum sensing vulnerable to falsify data attacks in cognitive radio network, a cooperative spectrum sensing scheme with against SSDF attack based on test statistic and extreme studentized deviate test was proposed. Firstly, the differential evolution algorithm and weighted gain combining soft decision scheme were combined to form an efficient node decision fusion mechanism. Then, the fault cognitive radio nodes were eliminated by test statistic according to soft statistics data. Finally, the malicious cognitive radio nodes were eliminated by improved ESD test, so as to make a global decision. The simulation results show that the proposed scheme can effectively filter SSDF attack data in cooperative sensing, and it has lower error detection rate.