Research On Physical Layer Security Based On MBM And Massive MIMO

With the development of wireless communication technologies,it is more difficult to secure communications.The physical layer security has gained considerable attention as a practicable communication security mechanism.The key idea behind this strategy is to exploit the randomness of wireless channels to ensure that the adversary cannot acquire any information.This security mechanism is based on information theory and signal processing.The physical layer security opens a new sight for security research.It is a valuable problem to research the physical layer security for emerging wireless communication technologies.Media-based modulation(MBM)is a novel transmission method.It exploits the rich scattering environment to generate a series of channels and embeds information in the variations of the transmission media,which means that it exploits the variety of channel realizations to transmit different information.The core idea of MBM is same as physical layer security technology.So it is worth researching the physical layer security based on MBM systems.Massive multiple input multiple output(MIMO)has recently emerged as a potential technology which provides higher spectral and power efficiency.Equipping tens or hundreds of antennas in a cell,it greatly improves the spectral efficiency and data rate.However,researchers pay little attention to the effect of the unlimited antennas to the physical layer security.In this dissertation,we mainly investigate the physical layer security for MBM and massive MIMO systems according to their characteristics.The contents of this dissertation are listed as follows.Before discussing the physical layer security for MBM systems,we analyze the basic prin-ciples,basic models and signal detection algorithm of MBM system.Different from tradition-al source-based modulation like phase shift keying(PSK)and quadrature amplitude modulation(QAM),MBM embeds information in the variations of transmission media and exploits the ran-domness and uniqueness of channels to transmit information.All the channel realizations form a unique constellation at receiver.We research the basic principles of the system.Then we present the systems models of single-input and multi-input and their transmission characteristics.We also propose a practicable receiving signal detection algorithm.Then we focus on the physical layer security for MBM systems.We present the method to improve system security in the presence of an eavesdropper.Because of the characteristics that MBM systems have a large number of channel realizations,we choose parts of these channels to improve the system security without the requirement of transmitting the channel information between transmitter and receiver.Based on different premises,we propose three schemes:optimal channel realization selection scheme,practical channel realization selection scheme and artificial noise(AN)based selection scheme.Through the detailed study and simulation of secrecy outage probability and security diversity for three channel selection schemes,we compare the performance of these schemes on system security.The simulation results also demonstrate that MBM systems can improve the system security through channel realization selection.We research the physical layer security for massive MIMO.After presenting the communica-tion eavesdropping model,we analyze the effect to security rate when the adversary uses different receiving technologies to decode the receive signals,and eavesdrop or attack the train phase or data phase.Through the analysis of security rate,secrecy outage probability and security diversi-ty,it demonstrates that massive MIMO combats passive eavesdropping because of the unlimited antennas and precoding at base station.So the adversary has to attack the train phase to destroy the system security.Thus we research the attack detection scheme for further study.After present-ing and analyzing the existing power detection scheme and pilot correlation detection scheme,we combine the two schemes to propose our simple detection scheme.The simulation results display that our scheme is robust and it can improve the detection performance effectively.