Research On Physical Layer Security Technology With Active Eavesdropper

Physical layer security technology is potential to provide "perfect secrecy" independent on the computational complexity of devices and provides a new solution for wireless secure communications.At present,many researchers have proposed insightful ideas in passive wiretapping scenario,but a variety of active wiretapping methods have emerged and created much serious damage to secure transmission.The current research lacks the analysis and optimization of the physical layer security transmission performance under the active eavesdropping scenario,and the game analysis between the active eavesdropper and the legitimate nodes.Motivated by this,the thesis explores artificial-noise-aided communication scheme in the presence of an active eavesdropper.It reveals the influence of interaction between legitimate nodes and illegal nodes on secrecy performance.Main contributions of this thesis are summarized as follows.(1)This thesis focuses on confidential transmission under full-duplex active eavesdropping scenario and establishes the artificial noise-assisted Multi-Input Single-Output Single-Eavesdropper(MISOSE)downlink transmission model.The closed-form ergodic secrecy rate is also derived based on Jason inequality.Then the optimal power allocation between information signals and the artificial noise is derived.The result reveals the relationship between the system parameters and the optimal transmission strategy of the source node,and indicates the upper limit of the system reachable security rate.Moreover,it reduces the system design complexity,and guides the legitimate node to determine the optimal power transmission strategy directly to achieve the optimal system secure transmission scheme.(2)Aiming at the security-oriented joint power distribution problem between full-duplex eavesdroppers and legal nodes,the game model between legal nodes and eavesdropping nodes is established,and the Nash equilibrium solution that satisfies both states of equilibrium is deduced.This result reveals the intrinsic relationship between the active eavesdropper and the legitimate node.In addition,the result indicates the minimum power allocation scheme of the legal node and the optimal interference power transmission scheme of the eavesdropping node,and gives the actual safe transmission rate of the system in the balanced state.Comparing the security performance of active eavesdropping and passive eavesdropping,we analyze the effect of eavesdropper's behavior on security performance.(3)In the scenario of stochastic collaborative eavesdropping,a Multi-Input Single-Output Single-Eavesdropper Multi-Jammer(MISOSEMJ)downlink transmission model with artificial noise is established and the closed-form ergodic secrecy rate is derived based on Jason inequality.The game theory is used to analyze the power allocation scheme of the source node and the cooperative node selection scheme.The existence of Nash equilibrium in the game model is proved and an iterative algorithm is proposed to obtain the equilibrium solution quickly.The result reveals the influence of eavesdropping on the security performance of the system,and gives insights on resisting the active eavesdropper who can select cooperative jammer.