文档介绍：南京邮电大学
硕士学位论文
认知无线电网络中的频谱感知方法研究
姓名:姚继明
申请学位级别:硕士
专业:信号与信息处理
指导教师:赵君喜
2011-03
南京邮电大学硕士研究生学位论文摘要
摘要
认知无线电技术是在现有频率资源分配不合理、频率资源利用率不高的背景下提出的,
大量的研究表明该技术可以有效缓解频率资源紧张现状和提高频率资源利用率。频谱感知技
术是认知无线电的核心技术之一,也是整个感知过程中最先利用的技术。
本文首先介绍了单个用户感知的常用的几种算法,并对这几种算法的优缺点分别进行了
分析和比较。针对单个用户的频谱感知容易受到阴影、衰落以及无线信道的时刻变化等噪声
因素的影响这一情况,协作频谱感知被提出用来对抗这些不利因素。在协作感知频谱的过程
中,来自于不同感知用户的信息被结合在一起从而做出主用户存在与否的判决,所以融合算
法的选择对协作感知的性能起着关键作用。本文分决策融合和数据融合两个方面对协作感知
中融合算法进行了讨论,并对这些算法的性能进行了分析和比较。
以上算法都是基于固定样本数检测的方法,实际判决过程中非固定样本数检测方法更能
节省信道资源,同时考虑到在实际的通信系统中会存在着一些恶意用户对系统的检测造成干
扰。本文在此背景下给出了一种改进的基于声望值的加权序贯检测的感知方法。该方法是非
固定样本数的检测,不仅能够利用恶意用户发送的数据,而且可以实时对恶意用户的名单进
行更新。仿真结果表明,该方法可以减少融合中心判决所需样本数,提高系统检测性能,也
即有效地减少了恶意用户的影响。

关键词:协作频谱感知,决策融合,数据融合,序贯检测
I
南京邮电大学硕士研究生学位论文 ABSTRACT
ABSTRACT
Because of many spectrum segments have been allocated to the licensed spectrum users and
the spectrum utilization of the existing systems is little, cognitive radio technology is proposed as a
solution to this question. A lot of research shows that the technology can effectively relieve the
scarcity of spectrum and improve the utilization of frequency resource. Spectrum sensing is a key
technology in cognitive radio systems, and also the first step of the whole cognition process.
This paper first introduces normal algorithms of individual user's sensing. The advantages and
disadvantages of these algorithms are analyzed pared. Because of a single user's spectrum
sensing is affected by shadowing, fading and time-varying natures of wireless channels, cooperative
spectrum sensing schemes have been researched bat negative factors. During process of
cooperative spectrum sensing, information from different CR users bined to make a decision
on the presence or absence of the primary user, so choice of fusion algorithm is important to
collaborative spectrum sensing. This paper studies the collaborative spectrum sensing and divides
the collaborative spectrum sensing algorithms into decision f