Analysis Of The Physical Layer Security Performance Of Large-scale NOMA Networks

Non orthogonal multiple access(NOMA)is one of the key technologies of the fifth generation mobile communication system(5G),which can greatly improve the spectrum efficiency.At the same time,the information security of users in NOMA system is also threatened by eavesdroppers.Therefore,this dissertation is devoted to research on enhancing the security performance of NOMA system through transmit antenna selection(TAS)and artificial noise(AN),in order to provide certain help for the design of secure NOMA system.The main contents of this dissertation are as follows:1.This dissertation analyzes the physical layer security performance of the multi-user multiple-input multiple-output(MIMO)-NOMA system in a scenario with randomly distributed multi-antenna eavesdropper,and proposes a performance improvement plan.First,this dissertation not only proposes a channel training scheme combined with TAS,but also proposes a corresponding dynamic uplink transmit power allocation scheme.Then,in order to allocate the power allocation coefficient of each user,this dissertation proposes a dynamic downlink transmission power allocation scheme to make the communication quality of different users more balanced.Finally,the closed form of the effective throughput of each legitimate user under imperfect channel state information(CSI)is obtained by using stochastic geometry theory.The simulation results proved the accuracy of the derived formula and the effectiveness of the proposed scheme.2.TAS can not reduce the signal-to-interference-plus-noise ratio(SINR)of the eavesdropper decoding the user's confidential information signal,so it is not suitable to continue to use TAS when the eavesdropper is close to the base station,while AN can interfere with the eavesdropper,thus reducing the SINR of the eavesdropper decoding the user's signal.Therefore,this dissertation analyzes the physical layer security performance of the multi-user multiple-input single-output(MISO)-NOMA system using AN in the scenario of randomly distributed single-antenna eavesdroppers,and proposes a performance improvement plan.First,this dissertation proposes a secrecy beamforming(SBF)scheme suitable for multi-user MISO systems.Compared with the traditional AN scheme(this scheme adds AN to each legal user separately),this scheme can reduce the interference to legitimateusers caused by AN leaking into the main channel due to channel estimation error.Then,this dissertation combines the SBF scheme with a channel training scheme based on the minimum mean square error,and proposes two power allocation schemes to solve the power allocation problem.Power allocation scheme II aims to make the communication quality of different users more balanced.Finally,the closed form of the effective throughput of each legitimate user under imperfect channel state information(CSI)is obtained by using stochastic geometry theory.The simulation results proved the accuracy of the derived formula and the effectiveness of the proposed scheme.3.The power allocation scheme of the previous two models can actually be further optimized to further improve the physical layer security performance of the NOMA system.Therefore,this dissertation optimizes the multi-user MISO-NOMA system and the multi-user MISO system in order to maximize the sum of the ergodic secrecy channel capacity of all legal users.Among them,this dissertation also considers and compares the use of traditional AN scheme and multi-user SBF scheme to construct AN.In addition,the channel training scheme will bring channel estimation error.Therefore,in the optimization process,this dissertation add relevant constraints to improve the probability that strong users can completely eliminate the interference of weak users' signals in the case of imperfect CSI.In addition,in order to narrow the gap between the effective throughput of different users,this dissertation also proposes an adaptive constraint scheme.The simulation results prove the effectiveness of the proposed scheme.