Theoretical Research On The Physical Layer Security Of Downlink NOMA Systems

In order to cope with the high-speed growth of mobile data traffic and massive device connectivity,the 5th generation mobile communication(5G)system is required to provide higher data transmission rates,more device connectivity capabilities and more than 100 times energy efficiency improvement for the network.However,because of the natural factors such as the fading characteristics of wireless channels,multipath propagation,and inter-user interference,the design of high-rate transmission schemes in wireless environments becomes more challenging.In recent years,Non-orthogonal Multiple Access(NOMA)technology has received extensive attention for its superior spectrum utilization and capacity and is generally regarded as one of the most promising technology in 5G systems.Due to the broadcast characteristics of the transmission channel,its security problems become more prominent.Physical layer security utilizes the randomness and time-varying characteristics of the wireless channel from the perspective of information theory to achieve secure communication.It overcomes the major shortcomings of traditional security technologies and has been considered as an effective means to achieve secure communications.This thesis model and analyzes the physical layer security performance of the downlink NOMA system,and utilizes the transmit antenna selection technology to improve the security performance of the system.The specific researches can be summarized as follows:1.Model and analyze the physical layer security performance of the downlink multiple-input single-output NOMA system.First of all,the performance of the singleantenna NOMA system as a benchmark is analyzed,and then the security performance under different antenna selection strategies are analyzed.The closed-form analytical expressions of the secrecy outage probability(SOP)are derived.The results show that the secrecy diversity orders(SDO)of the system under this model is zero,and a dynamic power allocation scheme is proposed to make the SDO non-zero.Finally,the correctness of the theoretical analysis is verified by simulation.2.Although the traditional transmit antenna selection strategy enhances the physical layer security performance of a single user,the improvement of the overall system is negligible.Therefore,this thesis proposes a Max-Min antenna selection scheme that considers a multiple input multiple output NOMA system with multiple eavesdropping nodes.Based on whether the eavesdropping nodes are independent or not,the security performance of the eavesdropping node in collusion and non-collusion situations is analyzed,a closed-form analytic expression of their exact and asymptotic SOPs are derived.In addition,through the asymptotic analysis under the dynamic power allocation scheme,the SDO under the Max-Min strategy is obtained.The result shows that the SDO is related to not only the power allocation parameters,the number of transmit antennas,but also the number of destination antennas,the channel parameters and the order in which legitimate users are sorted.