摘要
In this paper,we study the covert performance of the downlink low earth orbit(LEO)satellite communication,where the unmanned aerial vehicle(UAV)is employed as a cooperative jammer.To maximize the covert rate of the LEO satellite transmission,a multi-objective problem is formulated to jointly optimize the UAV’s jamming power and trajectory.For practical consideration,we assume that the UAV can only have partial environmental information,and can’t know the detection threshold and exact location of the eavesdropper on the ground.To solve the multiobjective problem,we propose the data-driven generative adversarial network(DD-GAN)based method to optimize the power and trajectory of the UAV,in which the sample data is collected by using genetic algorithm(GA).Simulation results show that the jamming solution of UAV generated by DD-GAN can achieve an effective trade-off between covert rate and probability of detection errors when only limited prior information is obtained.
基金
supported in part by the National Natural Science Foundation for Distinguished Young Scholar 61825104
in part by the National Natural Science Foundation of China under Grant 62201582
in part by the National Nature Science Foundation of China under Grants 62101450
in part by the Key R&D Plan of Shaan Xi Province Grants 2023YBGY037
in part by National Key R&D Program of China(2022YFC3301300)
in part by the Natural Science Basic Research Program of Shaanxi under Grant 2022JQ-632
in part by Innovative Cultivation Project of School of Information and Communication of National University of Defense Technology under Grant YJKT-ZD-2202。
