频谱感知的多节点协作干扰与抗干扰研究

发布时间：2018-11-22 09:55

【摘要】：认知无线电通过动态频谱接入的方式实现频谱共享,被广泛视为一种提高频谱利用率、解决频谱资源短缺的有效方法。频谱感知是认知无线电实现的基石,其为动态频谱接入提供指导性信息。然而,由于无线通信环境的开放性与认知无线电“先感知,后接入”的特殊性,频谱感知过程面临着潜在的安全威胁。在噪声与信道衰落存在的感知环境中,相较于单节点频谱感知,多节点协作频谱感知的方式能有效改善频谱感知性能。可是,多节点协作频谱感知过程却易受到恶意用户的干扰,其通过某种干扰方式迷惑认知系统使其做出与频谱真实状态不符的判决结果,导致协作感知性能的下降。本论文在现有频谱感知干扰与抗干扰总结分析基础上,分别从多节点频谱感知过程中的干扰与抗干扰两方面展开了研究。研究了频谱感知的多节点协作干扰对频谱感知性能产生的影响,提出了一种多节点协作干扰的方法。所提出方法通过多节点协作,对目标认知系统的感知性能相关参数进行学习,据此选择相应的干扰策略实施协作攻击。具体分析了对感知性能相关参数进行学习所需的长度、影响学习准确率的因素,研究了具体的多节点协作干扰策略,最后通过计算机仿真验证了所提多节点协作干扰算法的有效性。对频谱感知多节点协作抗干扰方法进行了研究,提出了一种基于聚类的多节点协作频谱感知算法,以此减小恶意用户对认知系统造成的不利影响。该算法在未知恶意用户相关先验信息的条件下,利用聚类的方法识别被干扰的感知数据,并为每个认知用户分配一个身份标识值以减小聚类识别错误对认知系统感知性能造成的影响,最后通过仿真验证了所提算法的鲁棒性以及有效性。
[Abstract]:Cognitive radio realizes spectrum sharing by dynamic spectrum access, which is widely regarded as an effective method to improve spectrum efficiency and solve the shortage of spectrum resources. Spectrum sensing is the cornerstone of cognitive radio, which provides guiding information for dynamic spectrum access. However, due to the openness of wireless communication environment and the particularity of cognitive radio "first perception, then access", spectrum sensing is faced with potential security threats. In the presence of noise and channel fading, multi-node cooperative spectrum sensing can effectively improve spectrum sensing performance compared with single-node spectrum sensing. However, the multi-node cooperative spectrum sensing process is vulnerable to the interference of malicious users. It can confuse the cognitive system to make the decision results that are not in accordance with the true state of the spectrum, which leads to the degradation of cooperative sensing performance. Based on the summary and analysis of spectrum sensing interference and interference, this paper studies the interference and interference in the process of multi-node spectrum sensing. The effect of spectrum sensing multi-node cooperative interference on spectrum sensing performance is studied, and a multi-node cooperative interference method is proposed. The proposed method is based on the multi-node collaboration to learn the parameters related to the perceived performance of the target cognitive system, and then select the corresponding interference strategies to implement cooperative attacks. The length needed to learn the perceptual performance parameters and the factors affecting the learning accuracy are analyzed in detail, and the specific multi-node cooperative jamming strategy is studied. Finally, the effectiveness of the proposed multi-node cooperative interference algorithm is verified by computer simulation. A multi-node cooperative spectrum sensing algorithm based on clustering is proposed to reduce the adverse effects of malicious users on the cognitive system. Under the condition of unknown prior information of malicious user, the algorithm uses clustering method to identify the disturbed perceptual data. An identity value is assigned to each cognitive user to reduce the impact of clustering errors on cognitive system perception performance. Finally, the robustness and effectiveness of the proposed algorithm are verified by simulation.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925

【参考文献】