Research On Antenna Selection Technology Based On Support Vector Machine In Indoor Spatial Modulation VLC System

With the further development of technologies such as big data,cloud computing and the internet of things,people today are in an era of information explosion.Digital and intelligent new family lifestyles are constantly being proposed,and the demand for high-speed communication and spectrum resources is also rising sharply.The spectrum resources of traditional wireless communication have gradually been unable to meet human needs,so it is urgent to develop new spectrum to solve the contradiction between the demand for super-capacity communication and the spectrum crisis.Visible light communication(VLC)has become an effective supplement to the wireless communication system based on radio frequency(RF)link by virtue of its rich free spectrum resources,relatively high network security and excellent anti-electromagnetic interference capability.In the VLC system,the spatial modulation(SM)technology introduces the modulation range by using the position index of the Light Emitting Diode(LED)located at the transmitter as the spatial domain,and each time slot will activate one LED to send a message,avoiding inter-channel interference(ICI)and synchronization problems in the Multiple-Input Multiple-Output(MIMO)technology.Since LED need to be selected to transmit information in the indoor SM-VLC system,how to quickly select the ideal LED is one of the key issues that need to be solved.Due to the superiority of machine learning technology in dealing with complex problems,this paper will adopt the support vector machine(SVM)technology in machine learning to solve the problem of LED selection.The main research contents of this paper are as follows:The research background and significance of the antenna selection technology based on machine learning in the indoor VLC system are explained,the research status of this research direction at home and abroad is summarized,and the research questions to be carried out in this paper will be given.The characteristics of the MIMO-VLC system,including its composition architecture and communication transmission method,are deeply discussed.At the same time,some spatial modulation technologies in the indoor VLC system are introduced: Space Shift Keying(SSK)modulation,Generalized Space Shift Keying(GSSK)modulation,SM and Generalized Spatial Modulation(GSM).The basic principles and system models of these four modulation schemes are analyzed in detail.Finally,the related concepts of machine learning technology and physical layer security are introduced.Based on the Euclidean Distance Antenna Selection(EDAS)algorithm,this paper proposes a smart antenna selection algorithm to improve the overall performance of the indoor GSSK-VLC and GSM-VLC systems.By modeling the LED selection in indoor SM-VLC system as a multi-classification problem,an optimization problem is constructed by utilizing kernel SVM.After the optimal parameters of the learning system are obtained,the LED selection procedure can be accomplished efficiently for any given user's channel state information(CSI).Simulation results complexity analysis show that,compared with traditional LED selection algorithms,the proposed algorithm can achieve an ideal bit error ratio(BER)performance while having considerably lower complexity,which proves the effectiveness of the algorithm.Additionally,to improve the physical layer security performance of the indoor GSSK-VLC system,an SVM aided secure LED selection algorithm is proposed.Simulations and complexity analysis show that,compared with traditional antenna selection algorithms,the proposed SVM aided LED selection algorithm can acquire nearly the same secrecy performance,and the complexity of antenna selection is greatly reduced.