[1]尹拓凯*,岳文静,陈 志.面向拜占庭攻击的认知用户分类[J].计算机技术与发展,2023,33(04):102-107.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 015]
　YIN Tuo-kai*,YUE Wen-jing,CHEN Zhi.Cognitive User Classification for Byzantine Attack[J].,2023,33(04):102-107.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 015]

点击复制

面向拜占庭攻击的认知用户分类()

分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年04期

页码:

102-107

栏目:

网络空间安全

出版日期:

2023-04-10

文章信息/Info

Title:

Cognitive User Classification for Byzantine Attack

文章编号:

1673-629X(2023)04-0102-06

作者:

尹拓凯^1* ; 岳文静¹ ; 陈 志²

1. 南京邮电大学 通信与信息工程学院,江苏 南京 210000;

2. 南京邮电大学 计算机学院,江苏 南京 210000

Author(s):

YIN Tuo-kai1* ; YUE Wen-jing1 ; CHEN Zhi2

1. School of Communication & Information Engineering,Nanjing University of Posts & elecommunications,　Nanjing 210000,China;

2. School of Computer,Nanjing University of Posts & Telecommunications,Nanjing 210000,China

关键词:

认知无线电; 图神经网络; 拜占庭攻击; 恶意用户; 无线通信

Keywords:

cognitive radio; graph neural network; Byzantine attack; malicious user; wireless communication

分类号:

TP309;TN925

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 04. 015

摘要:

万物互联的蓬勃发展使得承载业务的无线资源日益短缺,认知无线电技术是具有广大前景的技术。 而作为实现认知无线电的基础环节的频谱感知却面临特有的安全威胁,其中拜占庭攻击伪造频谱感知数据,利用认知网络的开放性和合作机制,恶意篡改上传数据,致使融合中心做出错误判断,造成频谱资源损失。 因此,针对拜占庭攻击下的安全问题,该文提出了基于图神经网络的异常度检测,依据图神经网络得到数据异常程度,结合全局判决与邻居用户冲突程度建立异常度模型。 首先,介绍了认知网络频谱感知的相关原理;接着,展示了认知网络的模型,并对拜占庭攻击做了简单的概括。 研究结果表明,该算法在拜占庭攻击中取得优异的识别性能,不仅降低了正常用户的误判概率,同时提高了恶意用户的检测概率。

Abstract:

The vigorous development of the Internet of things makes the wireless resources carrying services increasingly scarce. Cognitiveradio technology is a technology with broad prospects. Spectrum sensing,as the basic link of cognitive radio, faces a unique securitythreat. Byzantine forges spectrum sensing data and maliciously tampers with the uploaded data by using the openness and cooperationmechanism of cognitive network,resulting in the wrong judgment of fusion center FC and the loss of spectrum resources. Therefore,aiming at the security problem under Byzantine attack,we propose a reputation detection based on graph neural network,obtain the degreeof data abnormality according to graph neural network,and establish a reputation model combined with the degree of conflict betweenglobal judgment and neighbor users. Firstly,we introduce the relevant principles?

of spectrum sensing in cognitive network,then show themodel of cognitive network,and briefly summarize the Byzantine attack. The results show that the proposed algorithm achieves excellentrecognition performance in Byzantine attack,which not only reduces the false positive rate,but also improves the real rate.

常用功能