Physical Layer Security Transmission Scheme And Capacity Optimization Based On Artificial Interference

With the rapid development of computer technologies, the drawbacks of thetraditional encryption algorithm are increasingly shown. To further guarantee thesecurity of wireless communication, physical layer security was born as a new researcharea. Artificial interference technique is one of the methods to enhance the physicallayer security performance of wireless communication networks, which interferes theinformation signal received by the eavesdropper by sending artificial noise or specificwaveform to achieve secure communication in physical layer. Although the physicallayer security has made some progress, for widely used Gaussian Z channel andGaussian wiretap channel, there are few results on physical layer security transmissionschemes and secrecy capacity optimization base on artificial interference. This thesismainly studies this problem, and the main work includes:For Gaussian Z channel with single antenna, the security transmission scheme andthe corresponding secure degrees of freedom are investigated. First, the upper bounds ofthe secrecy capacity and the secure degrees of freedom for this channel are obtained viatheoretical analysis. Second, a security transmission scheme based on artificialinterference is proposed, which also proves that the upper bounds of the securitycapacity and the secure degrees of freedom are achievable. Then, the secure degrees offreedom for this channel are obtained. Last, the transmission scheme without artificialinterference and the corresponding secure degrees of freedom are analyzed andcompared with the proposed scheme. The results show that the security transmissionscheme in this thesis can improve the system secrecy capacity and the secure degrees offreedom.For the secure communication of multiple-input multiple-output (MIMO) Gaussianwiretap channel, two security transmission schemes based on artificial interference arepresented. In the first scheme, since there is no cooperation between the transmitter andthe helper, the helper implements artificial interference by using generalized singularvalue decomposition (GSVD) algorithm, and chooses the subchannels that can weakenthe eavesdropping capacity of the eavesdropper. Then, the helper sends the artificialnoise through these subchannels, improving the system secrecy capacity. Further, toeliminate the effect of the artificial noise on the information signal at the receiver, the second scheme utilizes the cooperation between the transmitter and the helper. In thisscheme, the helper designs the precoding matrix for the artificial noise by using spaceprojection for antenna configuration with arbitrary number, and adopts suboptimalweighting strategy based on the channel quality to reduce the computational complexity.Finally, the performance and computational complexity of the two schemes areanalyzed via theoretical analysis. The performance is also analyzed via simulation.Theoretical analysis and simulation results show that, compared with the transmissionschemes without artificial interference, the two transmission schemes based on artificialinterference can effectively improve the system secrecy capacity. The results also showthat the second scheme with the suboptimal weighting strategy is better than the equalweighting strategy, and is quite close to optimal weighting strategy.For the physical layer security transmission of energy constrained Gaussianwiretap channel, the security transmission scheme and capacity optimization for energyharvesting Gaussian wiretap channel based on save-then-transmit (ST) protocol arestudied. First, the capacity optimization in the scheme without artificial interference isinvestigated. Second, a security transmission scheme based on artificial interference isproposed, and the sufficient and necessary conditions for this scheme to improve thesecrecy capacity are analyzed. Also, an iterative optimization algorithm for the secrecycapacity in this scheme is proposed. Finally, a low complexity selection strategy forsingle helper is given. Simulation results show that, in the scheme without artificialinterference, the optimization algorithm performs better than the alogirhtm with halfenergy harvesting ratio. They also show that in the transmission scheme based onartificial interference, the iteration optimization algorithm can further improve thesystem secrecy capacity and has a fast convergence speed. Furthermore, when thetransmission scheme without artificial interference is failed, the transmission schemebased on artificial interference can still guarantee secure communication under certainconditions.