针对认知无线电网络中的频谱感知数据伪造(spectrum sensing data falsification，SSDF)攻击问题，提出一种基于哈里斯鹰优化(Harris hawks optimization，HHO)算法和支持向量机(support vector machine，SVM)的抗SSDF攻击协作频谱感知方法。首先从报告信息矩阵中提取用于区分次用户(secondary users，SU)类别的特征向量。其次通过HHO算法优化SVM内核参数，通过优化的SVM模型检测恶意SU，提高了在复杂感知环境中对SU分类的准确率。最后根据优化的SVM模型计算获得SU的可信度，并以可信度为权重融合感知数据，进一步加强系统的抗攻击性。仿真结果表明，所提方法能够对不同的SSDF攻击场景实现有效防御，相比现有的方法具有更好的频谱感知性能。
Cooperative spectrum sensing method based on HHO-SVM for resisting SSDF attacks
Aiming at the problem of spectrum sensing data falsification(SSDF)attacks in cognitive radio networks,a cooperative spectrum sensing method based on Harris hawks optimization(HHO)algorithm and support vector machine(SVM)for resisting SSDF attacks is proposed.Firstly,feature vectors are extracted from the report information matrix to distinguish the category of secondary users(SU).Secondly,the SVM kernel parameters are optimized with HHO algorithm,and the malicious SU are detected by the optimized SVM model,which improves the accuracy of SU classification in complex sensing environments.Finally,the reliability of SU are calculated according to the optimized SVM model,and the sensing data is fused with credibility as a weight to further strengthen the anti attack capability of the system.Simulation results show that the proposed method can effectively defend against different SSDF attack scenarios,and has better spectrum sensing performance compared with the existing methods.
spectrum sensingspectrum sensing data falsification(SSDF)attacksupport vector machine(SVM)weighted fusion
NETL
NSTL
万方数据
