频谱感知技术毕业论文

《频谱感知技术毕业论文》由会员分享，可在线阅读，更多相关《频谱感知技术毕业论文（69页珍藏版）》请在毕设资料网上搜索。

1、 I 中文摘要中文摘要 认知无线电技术（Cognitive Radio）是一种新颖的通信技术，它 能有效地缓和现有固定频谱管理方式导致的频谱稀缺问题，被认为是 下一代最具潜力的通信技术之一。频谱感知作为认知无线电的关键技 术之一，它能检测周围空间的频谱空洞，为认知无线用户提供可利用 的频谱；监视授权用户的出现，以防止对授权用户干扰。 现阶段对频谱感知技术的研究主要集中在提高频谱检测准确度 上，而对频谱检测面临的安全问题的研究甚少。本文在详细分析现有 单节点频谱检测技术，深入研究多节点协作感知的网络框架、数据融 合技术和信号检测理论的基础上，重点研究了频谱检测过程中的数据 混淆攻击（SSDF）问

2、题，提出了基于信誉的权重序贯对数似然比算法 （AWSPRT）。在 AWSPRT 算法中，我们主要有两个贡献点：首先 提出了基于检测准确度得信誉更新算法。它能有效反映认知用户检测 信息的可靠性，其次根据授权用户占用信道时间的特点，提出了恶意 用户剔除方法。仿真中，我们利用 matlab 对 AWSPRT 算法和现有 4 种算法进行了分析比较，结果显示，我们算法是在 SSDF 攻击下检测 准确度最高，最稳定的算法。 关键词关键词：认知无线电;频谱感知;序贯对数似然比检测;数据混淆攻击。 邵阳学院毕业设计 II Abstract Cognitive radio technology is a rev

3、olutionary new communication technology that promises to solve the spectrum shortage problem led by existing fixed spectrum management, and has been considered as one of the most promising communication technologies in the next generation. Spectrum sensing which is one of the key technologies in cog

4、nitive radio can detect the spectrum hole to provide secondary users with available spectrums, and monitor the appearance of primary users to avoid interference. Most of current spectrum sensing studies mainly focus on the detection ratio improvement, and pay little attention to the security issue i

5、n spectrum sensing In this paper, we analyze the single-node-based spectrum sensing technology, and investigate the network framework, the data fusion technology, and signal detection theory in Multi-node cooperative spectrum sensing technologies. Based on the above research, we focus on the spectru

6、m sensing data falsification (SSDF) attack problem, and propose an new algorithm named reputation-based Weight Sequential Log-likelihood Ratio Test (AWSPRT). In this paper, we have two contributions: (1)first, we propose a detection ratio-based reputation update algorithm which can effectively reflect the reliability of cognitive users sensing information; (2)second, we analyze the characteristic of the time when primary users appear or not, and propose a Malicious User detection algorithm. E