Research On Physical Layer Secrecy Coding Based On Wireless Channel Characteristics

The physical layer security is an information security guarantee method for wireless communication, and it exploits the diversity of wireless channel, the reciprocity and uniqueness of the channel characteristic to prevent the information leaking at the bottom of the communication link. As one of the key technologies of physical layer security, secrecy coding combining the error correcting properties of channel coding and t he wireless channel characteristics, and then differencing the receiving performance of the legitimate receiver and the third party eavesdropper. So it can guarantee the secure and reliable communication between the legitimate transmitter and receiver However, there are still the following problems: 1) The channel characteristics are not fully used, which restricts the secrecy gain of secrecy coding; 2) Lack of the secrecy coding and transmission joint design, which reduce the secrecy coding efficiency and the power efficiency of the system. Physical layer secrecy coding based on wireless channel characteristics is undertaken in this work with the support of the National Science Fund.Aiming at above problems, this dissertation first put forward two kinds of secrecy coding method which strong coupling the time-domain characteristics of the legitimate channel, and they were Physical Layer Secrecy Coding Based on the Fading Channel Amplitude Difference and Physical Layer Secrecy Coding Method Based on Random Phase Perturbation, respectively. Then we studied the joint design method of secrecy coding and the physical layer security transmission technology to improve the efficiency and applicability of secrecy coding. The main research contents are as follows:1. We propose a physical layer secrecy coding based on the fading channel amplitude difference. Through the proof that, in low-density parity-check coding, when the channel energy is identical, the smaller the channel fading amplitude of information nodes is, the lower the bit error rate becomes, we makes the confidential bits in the positions where the legitimate channel amplitudes are minimum and hides the confidential bits in the random sequence, thus turns the decoding condition better for legitimate receiver. Combined with this selection,the scrambling is used to spread the error to the whole code word. The simulation results show that our method improves the bit error rate of the eavesdropper, and the security gap of our metho d is reduced by 5dB compared with scrambling.2. We propose a physical layer secrecy coding method based on random phase perturbation. By using the product of the fading amplitude of legitimate channel and the amplitude information of phase perturbation weights, we determine the boundary of the phase random perturbations of the transmitted symbols. Then the phase perturbation weights are determined by the randomly generated amplitude information under the average power constraint, and randomly generated phase information in phase perturbation boundary. Finally, each transmitted symbols is precoding with the phase perturbation weights, so that it ensures a stable anti-noise performance for the legitimate receiver, while prevent the eavesdropper correctly deco ding the confidential information by the phase perturbation. The simulation results show that, under the same average transmission power and channel noise power, the bit error rate of the eavesdropper is close to 0.5.3. We study a joint design on secrecy coding and transmission based on the equivalent channel model. First, a wireless physical layer equivalent channel model is set up to analyses the key parameters affecting the performance of secrecy coding and artificial noise transmission method. And then we quantitatively describe the inner relationship between the weighted vector and the noise vector of the artificial noise method and the characteristics of equivalent channel, as well as the relationship between the error rate threshold of the secrecy co ding and the characteristics of equivalent channel. In this way, a close relationship between secrecy coding and artificial noise is established. Finally, a joint optimization algorithm is designed in order to maximize the confidential information rate. The simulation results show that the confidential information rate can reach to 0.43, which improve the secrecy coding efficiency, and the security margin is reduced by 3dB compared with the cascade connection of channel coding and secrecy transmission when 0.35 sR ?, which improve the power efficiency of the system.