Research On Power Distribution And Receiver Antenna Selection Algorithms In Secure Spatial Modulation Systems

One of the key challenges for future mobile communication research is how to achieve an attractive compromise between the spectrum efficiency and energy efficiency of wireless networks.Among them,the multi-antenna technology constitutes the key technology of modern wireless communication.It obtains better error performance and higher data rate at the expense of system complexity and cost.Among many multi-antenna transmission principles that use transmitters,receivers,or both,spatial modulation(SM)can use antenna index as the spatial dimension in addition to traditional amplitude and phase modulation to transmit information,which is a promising multi-antenna technology.Compared with single input single output system,it can provide higher data rate with very low system complexity,and has lower hardware cost and complexity compared with other multi-antenna systems.However,due to the openness and broadcast nature of radio transmission,confidential information is vulnerable to eavesdropping by illegal users.Therefore,it is crucial to ensure the secure transmission of SM systems.This thesis focuses on the physical layer security technology of SM system.Our main research contents and innovations are as follows:(1)In SM system,artificial noise is the key technology to ensure its safe transmission.For the power-constrained transmitter,how to reasonably allocate the power of useful signal and artificial noise has an important impact on improving the safety of SM system.Firstly,the secrecy rate of the SM system is derived,but there is no closed-form expression for the secrecy rate,the closed-form solution of the power allocation factor cannot be obtained.If the optimal power allocation factor is solved through the exhaustive search method,the complexity will be high.In order to reduce the complexity,a power allocation method based on deep learning is proposed.The simulation results show that the power allocation scheme based on deep learning is better than the existing power allocation method in terms of secrecy rate,and its implementation complexity is lower than that of the exhaustive search method where the secrecy rate performance of the former is close to that of the latter.(2)For the antenna selection problem in precoding-aided spatial modulation systems,in order to improve secrecy rate performance,the maximizing secrecy rate antenna selection algorithm is firs analyzed,but its complexity is high.Subsequently,the approximate expression of secrecy rate is derived,and then the method of maximizing the secrecy rate under low SNR and maximizing the secrecy rate under high SNR are proposed,and these two methods obtain the secrecy rate close to the optimal method under its conditions,respectively.In addition,according to the special form of the spatial modulation symbol,the method to maximize the secrecy rate at all SNRs is proposed.Simulation results show that the proposed three methods not only significantly reduce the complexity of antenna selection,but also obtain the secrecy rate performance close to the optimal solution.