Analysis Of Physical Laver Security Performance Based On Random Geometry

With the widespread use of mobile Internet and the increasing use of eavesdropping methods,the.security requirements for communication networks are increasing.The physical layer,as the first natural barrier to communication security,provides a reliable guarantee for secure communications.The physical layer security based on random geometry mainly studies the link transmission of wireless channels with noise and fading.Link transmission success probability often provides limited information.R.K.Ganti first proposed the concept of Meta-distribution in 2010.It provides more detailed network connectivity information than the average probability of success.Because it can not only provide the average probability of network transmission,but also solve the problem of how many users in the network can reach the corresponding probability of success.Therefore,this thesis aims at analyzing and improving the security performance of the physical layer,and uses Meta-distribution to carry out related research on the network reliability of physical layer security.The main work and achievements are as follows:(1)Deriving the SIR(Signal to Interference Ratio)of the physical layer security Meta distribution:Using the Meta distribution to study the physical layer security problem in the D2D communication scenario.A network model in which a base station and a legitimate user form a Poisson bipolar network,an eavesdropper adopts a Poisson distribution characteristic,and a nearby eavesdropping mode is selected.Based on the stochastic geometry theory,the Meta distribution characteristics of SIR are derived to analyze the connection reliability and safety reliability.The analysis results show that the stability of link connection success changes rapidly,that is,as the probability of successful connection increases,the number of users in the network that can meet this requirement will decrease rapidly;the security stability changes slowly,that is,with the success of security.As the probability increases,the number of users who can meet this requirement decreases slowly,and the number of users decreases rapidly only when high security and stability are required.(2)Deriving the distribution characteristics of successful connection probability in physical layer security considering interference correlation:Consider the impact of interference correlation on network security performance when the distance between eavesdropper and legal person is diff erent in the scenario of cellular network communication.Both the base station and the legitimate user in the cellular network are Poisson point processes,and the eavesdropping network model in which the user is located at a certain distance from the legitimate user for eavesdropping.Deriving the probability distribution characteristics of the network when the link connection success and the secrecy success occur simultaneously considering the interference correlation.The analysis results show that the correlation of interference will lead to lower probability of success than independent cases,and under different security constraints,not only the probability of successful connection of the network will decrease,but also the number of links that reach this probability will decrease.(3)The SINR(Signal to Interference and Noise Ratio)Meta distribution and the safe transmission rate under artificial noise Interference are derived.The problem of adding artificial noise to the network security performance improvement under interference-related conditions is studied.In the cellular network,the base station and the legitimate users in the cellular network are Poisson distribution,and the network model in which the user is located at a fixed location for eavesdropping is eavesdropped.The performance of the SINR and the secret transmission rate after adding artificial noise in the interference-related situation are derived.Adding interference correlation considerations can provide a more accurate performance analysis of the effects of artificial noise,and use Meta distribution to provide a deeper amount of information on the network.The analysis results show that the addition of artificial noise will have a certain improvement effect on the safe transmission rate,but there will also be a decrease in the probability of successful connection of the corresponding link.The above results can provide theoretical guidance and evaluation criteria for physical layer security network performance,and provide some reference for actual network operation.