Intelligent power control for covert communication in cognitive Internet of things

Abstract

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.

Keywords

• cognitive IoT
• covert communication
• generative adversarial network
• power control