Research On Physical Layer Security Technology In MIMO Full Duplex Two-way Communication Systems

The security of information is an important performance metric of communication systems.Physical layer security,which is based on information theory,realizes the security transmission of information by employing the random and time-varying characteristics of wireless channels.The multiple input multiple output(MIMO)technology not only has the advantage of increasing the transmission capacity of the wireless communications system,but also provides resources to realize the physical layer security transmission.The full-duplex technology can effectively improve the system's spectrum efficiency because nodes can transmit and receive signals on the same frequency at the same time.In addition,the two-way simultaneous transmission in fullduplex systems is beneficial to the security of information.This thesis studies the optimization of physical layer security scheme in a MIMO full-duplex two-way communication system with residual self-interference,where legitimate nodes send confidential information and artificial noise while receiving information.The main contents are as follows:1.For the scenario where the legitimate channel state information is known and the eavesdropping channel state information is only statistically known,the closed-form expression of the ergodic achievable secrecy sum rate and its lower bound are derived first.Aiming at maximizing the lower bound of the secrecy sum rate,the power allocation factors between the artificial noise and the information signal,and the power allocation matrices of the information signal are jointly optimized.An iterative method is adopted.In each iteration,the former is fixed,the latter is optimized,then the latter is fixed and the former is optimized.The difference of concave/convex(DC)programming and the genetic algorithm are used to optimize the information's power allocation matrices and the power allocation factors respectively.The simulation results prove that the theoretical derivation is correct,and the optimization algorithm can effectively improve the ergodic secrecy sum rate.2.For the scenario that the channel state information of legitimate channels and eavesdropping channels is known,the precoding matrices of information signals and artificial noise are optimized under the conditions that the obtained channel matrices are perfect or imperfect,respectively.When the channel state information is perfect,an optimization problem with the goal of maximizing secrecy sum rate is constructed.The problem is transformed into the form of the difference between two convex functions.Then DC programming is used to optimize the precoding matrices of information signals and artificial noise.For the scenario where the channel state information has random deviations,the worst-case criterion is used to model the channels,and the weighted minimum mean square error algorithm is proposed to design the robust precoding matrices of the information signals and artificial noise.The simulation proves that the optimization algorithm can effectively improve the secrecy sum rate of the system.