Physical Layer Security Techniques For User Random Mobile Wireless Systems With Intelligent Reflecting Surface

With millions of industrial devices embedded in the industrial internet of things(IIo T),wireless communication security is an important issue for wireless control in real-time industrial wireless information physical system,traditional password-based security mechanism because the restrictions based on the calculation of the safety performance and high complexity,cannot meet the needs of industrial wireless security.Industrial real-time control requires low-complexity and light-weight security measures to reduce the processing delay of communication,improve communication efficiency,and achieve ultra-reliable,high-security and low-delay communication.Physical layer security technology mainly uses the physical characteristics of wireless channels(such as fading,noise and interference)to achieve secure transmission of secure messages.It does not rely on the upper layer key,and can achieve the requirements of low delay and high security.Faced with the transmission requirements of high security,high speed,large capacity and low delay in wireless networks,we need to find more and better methods to achieve low-complexity and high-security physical layer secure transmission methods.Intelligent reflecting surface(IRS)can realize intelligent and controllable wireless channel propagation environment.The parameters of the electromagnetic field,such as amplitude,phase,polarization and frequency,can be dynamically regulated by IRS through the control signal applied to the adjustable elements of the electromagnetic unit,thus realizing intelligent reconfiguration in the wireless environment.IRS assisted wireless network can intelligently adjust the uncontrolled electromagnetic wave of wireless transmission channel,in order to realize certain customizable electromagnetic response,improve the desired signal power or suppress co-channel interference,in order to realize the maximization of the spectrum and energy efficiency and the maximum rate,by reconstructing the legitimate correspondents,potential eavesdroppers,channel,can maximize the physical layer security rate etc.This dissertation studies physical layer security enhancement based on random mobile users and intelligent reflective surfaces.One of the major characteristics of wireless networks is mobility,but the physical layer security research rarely considers the influence of user mobility on the security performance of physical features.At present,the research of our research group shows that the random mobile users in steady motion state have higher physical layer security capacity characteristics than the static uniformly distributed users.However,the existing research results are limited to the single-antenna system.The main contributions and innovations of this dissertation are summarized as follows:1.Proposed a random mobile user physical layer security model based on artificial noise-assisted multi-antenna transmission.Aiming at the characteristics that random mobile users may improve the security capacity,a physical layer security model of random mobile users based on artificial noise assisted multi-antenna transmission is proposed.This dissertation studies the influence of mobile users with random waypoints on the physical layer security of wireless networks in MISO networks.By comparing the security performance of the mobile user and the average static scene,the simulation results show that the random waypoint mobile user can obtain better security performance than the average static user.The influence of transmit power allocation,path loss,antenna number and target secrecy rate on the safety is analyzed.2.Proposed a cooperative jamming(CJ)security scheme for random mobile users based on edge computing and intelligent node selection.The scenario of a multi-antenna system with random mobile users aided by artificial noise and collaborative jamming schemes is investigated.The traversal secrecy capacity of random mobile users with cooperative jamming(CJ)schemes is derived,and the security energy efficiency of CJ-free and CJ schemes is compared.Test validation is performed in industrial plants to verify the physical layer security of mobile users in real industrial wireless networks.In addition,intelligent selection of CJ nodes using edge computing nodes maximizes the traversal secrecy capacity as well as the security energy efficiency.3.Proposed a physical layer scheme to resist eavesdropping attacks in industrial wireless network(IWN)based on intelligent reflecting surface.Aiming at industrial wireless networks,this dissertation studied a physical layer scheme to resist eavesdropping attack in IWN network based on Intelligent Reflecting Surface(IRS).In industrial scenarios,IRS has the advantages of low power consumption,low environmental radiation,and improved channel conditions for legitimate partners.The optimization of the IRS phase shift user to maximize the average rate of confidentiality,considering the SISO systems and MISO systems in the IRS auxiliary legitimate users,in view of the channel information unknown statistical information known,improve the legal user can reach rate of secrecy,optimization of beam forming and IRS phase shift matrix,for the IWN provides a low cost and high security transmission solution.In addition,the proposed scheme can also work well without knowing the exact channel information of eavesdropper Eve,and is more suitable for practical application scenarios.4.Proposed a joint cooperative jamming and intelligent reflecting surface reflection coefficient optimization scheme.This dissertation proposes a physical layer security model of cooperative interference combined with IRS under edge computing.A joint cooperative jamming and IRS reflection coefficient optimization scheme is proposed in the IRS assisted security cooperation system.Deep reinforcement learning(DRL)technology is integrated into the optimization design of IRS-assisted wireless communication system in the Internet of Things to solve the large dimension optimization problem.An edge device is provided to speed up task processing and real-time control of the IRS unit.In order to overcome the difficulty of solving the non-convex optimization problem in dynamic environment,the non-convex optimization problem of maximizing secrecy energy efficiency is modeled as a DRL process.For the cooperative interference model,DDPG learning method is used to realize the optimal cooperative interference strategy.The algorithm uses edge computing devices as agents to study the environment and learns approximate solutions based on rewards.The agent constructs the combinatorial optimization strategy,builds the global model,and finds the associated subproblems,which is better than the benchmark scheme.Different from previous algorithms that use alternating optimization strategies,the proposed algorithm takes the transmit beamforming matrix and the phase shift as the output of the DRL algorithm.