面向认知物联网的隐蔽通信智能功率控制  

Intelligent power control for covert communication in cognitive Internet of things

作　　者：李赞[1] 廖晓闽 石嘉 肖培 Zan LI;Xiaomin LIAO;Jia SHI;Pei XIAO(State Key Laboratory of Integrated Services Networks,Xidian University,Xi’an 710071,China;School of Information and Communications,National University of Defense Technology,Xi’an 710106,China;Home of 5G Innovation Centre,University of Surrey,Guildford GU27XH,U.K.)

机构地区：[1]西安电子科技大学综合业务网理论及关键技术国家重点实验室,陕西西安710071 [2]国防科技大学信息通信学院,陕西西安710106 [3]萨里大学5G创新研发中心,萨里郡吉尔福德GU27XH

出　　处：《物联网学报》2020年第1期52-58,共7页Chinese Journal on Internet of Things

基　　金：国家自然科学基金重点项目(No.61631015),国家自然科学基金项目(No.61941105,No.61901327);国家杰出青年科学基金项目(No.61825104).

摘　　要：针对认知物联网的安全问题,提出了一种基于生成对抗网络的认知物联网隐蔽通信智能功率控制算法。首先将认知物联网隐蔽通信问题转化为认知物联网用户和窃听者之间的动态博弈问题,然后利用生成器模仿认知物联网用户,利用鉴别器模仿窃听者,两者分别采用3层神经网络构建,并通过二人零和博弈实现学习优化过程,最终达到纳什均衡,获得隐蔽功率控制方案。仿真结果表明,所提出的算法收敛速度快,不仅可以获得近似最优的隐蔽功率控制方案,而且在未来认知物联网中更具有实用性。In order to solve the security problem of cognitive Internet of things(IoT),an intelligent power control algorithm of covert communication in cognitive IoT based on generative adversarial network was proposed.Firstly,the covert communication optimization problem in the cognitive IoT was transformed into a dynamic game between the cognitive IoT user and the eavesdropper.Then,the generator imitated the cognitive IoT user,while the discriminator imitated the eavesdropper.The generator and the discriminator were constructed by the three-layer neural network respectively.Through the two-person zero-sum game,the learning optimization process was realized to achieve the Nash equilibrium,and finally the covert power control scheme was obtained.The simulation results show that the proposed algorithm can not only obtain near-optimal covert power control scheme with rapid convergence ability,but also be more practical in the future cognitive IoT.

关 键 词：认知物联网 隐蔽通信 生成对抗网络 功率控制 

分 类 号：TN929.5[电子电信—通信与信息系统]