论文部分内容阅读
SSDF(Spectrum Sensing Data Falsification)攻击是认知无线网络中对频谱感知性能危害最大的攻击方式之一。基于认知无线网络中信号频域的固有稀疏性,本文结合了压缩感知(CS)技术与平均一致(average consensus)算法,建立了可防御SSDF攻击的分布式宽带压缩频谱感知模型。本文建立了次用户的声望值指标,用以在分布式信息融合的过程中更加准确地排除潜在的恶意次用户影响。在感知阶段,各个CR节点对接收到的主用户信号进行压缩采样以减少对宽带信号采样的开销和复杂度,并做出本地频谱估计。在信息融合阶段,各CR节点的本地频谱估计结果以分布式的方式进行信息融合,排除潜在恶意次用户的影响,得到最终的频谱估计结果。仿真结果表明,本文提出的分布式频谱感知模型可以有效地抵御SSDF攻击,提高了频谱感知的性能。
SSDF (Spectrum Sensing Data Falsification) attack is one of the most damaging attacks on spectrum-aware performance in cognitive wireless networks. Based on the inherent sparsity of signal frequency domain in cognitive wireless networks, this paper combines the compressed sensing (CS) technique with the average consensus algorithm to establish a distributed broadband compressed spectrum sensing model that can defend against SSDF attacks. This article establishes the sub-user’s reputation value index, which can be used to more accurately exclude potential malicious sub-users in the process of distributed information fusion. During the sensing phase, each CR node samples the received primary user signal to reduce the overhead and complexity of sampling the wideband signal and make local spectrum estimation. In the information fusion phase, the local spectrum estimation results of each CR node are distributed in a way of information fusion to eliminate the influence of potentially malicious secondary users to obtain the final spectrum estimation result. Simulation results show that the proposed distributed spectrum sensing model can effectively resist SSDF attacks and improve the performance of spectrum sensing.