Research On Physical Layer Security And Optimization Based On Intelligent Reflecting Surface

With the rapid development of wireless communication technology,Internet of Things has greatly improved people's quality of life in an information society.Physical layer security is the cornerstone of the development of Internet of Things.How to ensure the security of wireless transmission of sensitive data under the premise of low-cost equipment,miniature,limited energy and computing capabilities,is very important This paper proposes a lightweight physical layer security solution for the Internet of Things based on the characteristics of massive terminals,limited power consumption and computing capabilities in the Internet of Things scene.Specifically,this paper first proposes a key generation protocol enhanced by intelligent reflecting surface,which can further increase channel fluctuations on the basis of traditional protocols and increase the key generation rate under static channel conditions.At the same time,this paper also designs a phase control strategy for the intelligent reflecting surface based on deep reinforcement learning for the key generation protocol,so that the intelligent reflecting surface can most effectively improve the channel fluctuations.In this paper,the use of channel sampling data to drive the intelligent reflecting surface phase control strategy design.The mutual information estimated by the K-nearest neighbor algorithm is used as the key generation rate and the return function of deep reinforcement learning.In order to maximize the key generation rate,the deep reinforcement learning algorithm obtains the optimal phase control strategy of the intelligent reflecting surface,which effectively avoids the complex theoretical derivation of key capacity and channel estimation.The simulation results show that the key generation rate of the proposed intelligent reflecting surface phase control strategy based on deep reinforcement learning is better than that of the traditional fixed phase shift strategy and random phase shift strategy.