Research On VLC Physical Layer Security Technology Based On Artificial Noise Assistance

With the rapid development of the Internet,the number of users has increased rapidly leading to difficulty to guarantee users' security.Traditional network security mainly uses high complexity upper-layer encryption technology.In recent years,the calculating ability of mobile devices has been continuously enhanced,so the upper-layer encryption method based on the amount of calculations has flaws that can be broken.The physical layer security technology has low computational complexity and can theoretically completely protect user information security.Physical layer security technology provides the first line of defense against eavesdropping and will become an important solution to communication security.Visible Light Communications(VLC)can simultaneously satisfies communication and illumination in the indoor environment,which is an important scenario for future green communications.Because light cannot transmit through walls,light has natural security indoors.However,when the eavesdropper is located within the coverage area of the LED,the eavesdropper can eavesdrop on the direct signal of light.In addition,the light signal might be reflected on walls or ceilings so eavesdroppers can eavesdrop the reflected light signal.Therefore,this paper focuses on the physical layer security in VLC system.In wireless communications,delay is an important indicator of Quality of Service(QoS).Most of the existing physical layer security researches only focus on the physical layer without considering the statistical delay QoS requirements from the link layer perspective.The effective capacity reflects the throughput performance of the network under the statistical delay QoS constraints.Therefore,this paper studies the secure effective capacity which ensures statistical delay QoS requirements and secure transmission of user information.This paper mainly analyzes the security performance of the VLC system from two aspects: uplink and downlink VLC scenarios.In order to achieve secure transmission of the VLC physical layer,this paper introduce a scheme that eavesdroppers are interfered by artificial noise.In addition,this paper also propose algorithms to suppress the effect of artificial noise on users.Finally,the performances of secrecy rate,secure effective capacity,and Signal Noise Ratio(SNR)are analyzed in this paper.The work and contributions of this paper are summarized as follows:(1)In the uplink scenario of VLC,the signal sent by the user might be reflected indoors and eavesdroppers could eavesdrop the reflected optical signal.In order to improve thesecurity of the VLC system,an artificial noise aided scheme is proposed.This paper has designed a special receiver,whose transceiver works in full-duplex mode.While its Photo-Diode(PD)receiver receives signals from the user,its Light Emitting Diode(LED)transmitter sends artificial noise to interfere with eavesdroppers.As the received signal of the full-duplex transceiver would be affected by reflected artificial noise,this paper first introduce the Independent Component Analysis(ICA)method in the VLC system.At the full-duplex transceiver's receiving end,the self-interference of artificial noise is eliminated by the ICA algorithm so signals sent by the user are almost separated.But the eavesdropper would receive user's reflected signals and strong artificial noise.Therefore,the safety of user's data can be guaranteed.In this paper,as the estimated signal of the user's signal is obtained through the ICA algorithm,this paper can derive the lower limit of the secrecy rate.In addition,this paper compare the distribution of secrecy rate with and without artificial noise.The SNR distribution of the eavesdropper and the full-duplex transceiver's receiving end are also analyzed.The results show that the secrecy performance of the artificial noise aided full-duplex scheme is far superior to that without artificial noise.(2)As for the downlink VLC,most of the VLC physical layer security researches focus on the physical layer and do not consider the QoS guarantee in wireless communication.In order to guarantee the statistical delay QoS requirements,effective capacity based on physical layer security technology is studied.The effective capacity reflects the throughput performance of the system under statistical delay QoS guarantee.This paper assumes that the user's location is determined and the eavesdropper's location obey uniform distribution,so the user's secrecy rate is different when the eavesdropper is located in different locations.In this scenario,this paper calculates approximate secure effective capacity to guarantee the statistical delay QoS.With the goal of secure effective capacity,Particle Swarm optimization(PSO)algorithm is used to find a suitable artificial noise precoding matrix and a signal precoding matrix.Moreover,this paper analyzes the suitable power ratio of artificial noise and useful signals.The secure effective capacity of the Zero Forcing(ZF)algorithm and the PSO algorithm with artificial noise are compared with secure effective capacity without artificial noise.Simulation results show that adding artificial noise can improve the secrecy performance of the system and the overall performance of the PSO algorithm is the best.