基于Dempster-Shafer证据理论与抗频谱感知数据篡改攻击的协作式频谱检测算法

Cooperative Spectrum Sensing Algorithm against Spectrum Sensing Data Falsification Based on Dempster-Shafer Evidence Theory

针对恶意用户伪造认知用户或者控制已有认知用户、频谱感知数据篡改(SSDF)、使现有协作式频谱检测算法的性能大幅下降问题,提出了一种抗SSDF攻击干扰的协作式频谱检测算法。将SSDF攻击干扰分为Ⅰ类和Ⅱ类,通过Dempster-Shafer(D-S)证据理论设置用户权重和判决门限,区分Ⅰ类恶意用户、Ⅱ类恶意用户和可信用户,只使用可信的频谱检测数据进行信息融合,以判决频谱占用的情况、提高检测性能。理论分析与仿真结果表明,该算法能够有效对抗Ⅰ类和Ⅱ类SSDF的攻击干扰,其检测性能明显优于现有D-S证据融合频谱检测算法。

The malicious users forge and get command of the secondary users,which causes the spectrum sensing data falsification（SSDF） to decrease the performance of existing cooperative spectrum sensing algorithm.A cooperative spectrum sensing algorithm against SSDF is proposed.The SSDF attack is divided into Class Ⅰ and Ⅱ.User weight and decision threshold are set to distinguish the malicious users of Class Ⅰ and Ⅱ malicious users,and trusted secondary users by using Dempster-Shafer（D-S） evidence theory.Only trusted spectrum detection data can be used in information fusion to judge the spectrum usage and improve the detection performance.Theoretical analysis and simulated results show that the proposed algorithm can effectively confront SSDF attack interference,and its judgment performance is superior to that of existing D-S evidence fusion spectrum sensing algorithm.