Research On Physical Layer Security Transmission Methods In UAV-assisted Wireless Networks

Due to the broadcast characteristics of wireless communication links,it is easy to cause communication content leakage and lead to data leakage.Physical Layer Security(PLS)technology is widely regarded as a promising method for secure wireless communication by exploiting the physical characteristics of wireless channels.Compared with the security technology based on traditional cryptography,the physical layer security technology uses the random characteristics of the channel to improve the security performance,and does not need to use complex algorithms to effectively solve the problem of information leakage in wireless communication,so that in the process of information transmission,it can not only ensure the secure communication of the information receiver and transmitter,but also reduce the eavesdropping effect of the eavesdropping.However,with the improvement of eavesdrop technology,if the ground receiver and transmitter only use the physical layer security technology,the communication transmission efficiency will be reduced,and other communication equipment can be used to combat the eavesdrop of the eavesdrop,which can further improve the security of the communication between users.Unmanned Aerial Vehicle(UAV)can effectively assist wireless network communication due to its high flexibility and low cost.The use of UAVs to assist physical layer security transmission can significantly improve the problems of low communication efficiency and security performance caused by the receiver and transmitter in the fight against eavesdropping.However,the UAV battery capacity is limited,and the UAV can only rely on the onboard energy to maintain its flight when communicating.The flight trajectory of the UAV and its transmission power affect the energy consumption of the system.How to manage the network resources under the limited airborne energy to improve the secure transmission efficiency in the process of UAV-assisted wireless communication and the secure energy efficiency of the system are the key problems that need to be solved.The resource management scheme based on traditional optimization methods has high computational complexity and is difficult to adapt to the complex and changeable environment.The explosive growth of data variables makes it difficult for traditional methods to solve the problem of secure transmission.To solve this problem,this thesis focuses on the physical layer security transmission method in UAV-assisted wireless networks based on Deep Reinforcement Learning.The main research contents of this thesis are as follows:(1)Aiming at the problem of secrecy rate in Device-to-Device(D2D)scenario,a UAVassisted secure communication system in D2 D scenario was studied.Due to the broadcast characteristics of wireless channels,users are easily eavesdropped by eavesdropper during data transmission,resulting in data leakage.In order to ensure the security of D2 D communication,this thesis proposes a method to use mobile UAV to transmit jamming signals against ground eavesdropper to improve the security of communication for the UAV-assisted D2 D scenario.In order to ensure the continuity of communication,firstly,the connection outage probability model is established,and the closed-form expression for the connection outage probability of D2 D users is derived,where the secrecy rate is defined as the difference between the transmission rate and the eavesdropping rate.Then,by jointly optimizing the UAV flight trajectory and transmit power,the optimization problem is established to maximize the secrecy rate.To solve this problem,a joint optimization algorithm for UAV trajectory and jamming power based on Twin Delayed Deep Deterministic Policy Gradient(TD3)was proposed.In this algorithm,the UAV is regarded as a single agent.After the random parameters are initialized,the UAV observes the state,selects the corresponding action according to the initial strategy,adds noise,and selects a smaller target value in the twin network for the next network update to reduce the cumulative error of the algorithm.The simulation results show that using TD3 algorithm to optimize the UAV trajectory and jamming power can effectively improve the secrecy rate of the system.(2)Aiming at the security and energy efficiency problem of unmanned aerial vehicle(UAV)assisted eavesdropper in fuzzy coordinate scenario,the security transmission problem of UAV assisted wireless network communication in air-to-ground communication scenario and the energy consumption problem of communication system are studied.In the air-to-ground communication scenario considered in this thesis,the mobile UAV acts as the receiver and transmitter.When the ground user sends information to the mobile UAV,the UAV simultaneously sends jamming signals to counter the ground eavesdrop with fuzzy coordinates to ensure the security performance of the system.By jointly optimizing the flight trajectory of the UAV,the transmit power of the ground user and the transmit power of the UAV,an optimization problem is established to maximize the security and energy efficiency.To solve the problem,this thesis proposes a joint optimization algorithm of ground user’s transmit power,UAV trajectory and transmit power based on a dual-delay deep deterministic policy gradient to ensure the maximum system secure energy efficiency during the communication process.The performance of TD3 algorithm and Deep Deterministic Policy Gradient(DDPG)algorithm is also compared in solving this problem.The simulation results show that the TD3 algorithm outperforms the DDPG algorithm and can significantly improve the secure transmission performance of the communication system.