| At present,visible light communication technology(VLC)mainly uses LED for lighting and communication at the same time,which has great research value and application prospect in future communication fields such as 6G,so it has been favored by researchers continuously.Visible light communication uses light with wavelength from 380nm to 780nm in visible band and communication frequency from 400THz to 800THz as the information carrier,which has the advantages of high speed data transmission,large communication capacity,rich spectrum resources,safety and green.In recent years,with the increasing demand for indoor accurate positioning,indoor positioning technology based on VLC has been widely concerned.The traditional indoor visible light positioning technology has problems such as low positioning accuracy and complex system implementation,researchers have proposed a variety of improvement schemes,among which the indoor visible light positioning technology based on intelligent algorithm can effectively improve the indoor positioning accuracy,and has become one of the research hotspots.At the same time,in the communication system,modulation technology has been an inevitable focus of research.Compared with the traditional RF communication field,visible light is incoherent and adopts intensity modulation/direct detection(IM/DD)mode,so the modulation technology suitable for visible light communication is greatly limited.In the visible asymmetrically clipped optical OFDM(ACO-OFDM)system,Hermitic symmetry is used to eliminate complex numbers,and the negative numbers are eliminated by the pair zero limiting.Therefore,in theory,when the ACO-OFDM system is used in the visible light communication system,most digital modulation modes can be used without considering whether it has complex value or negative value.Therefore,it is possible for the visible light communication system to adopt various modulation modes for signal transmission.In this case,the automatic modulation style classification AMC is particularly important.And artificial intelligence algorithm can provide easy to implement the classification decision process,which makes the modulation mode recognition based on artificial intelligence algorithm has been widely concerned.Aiming at indoor visible light positioning technology and modulation technology,this paper uses artificial intelligence algorithms,including genetic algorithm(GA),particle swarm optimization(PSO),supported vector machine(SVM),K-nearest neighbor(KNN)and artificial neural network(ANN),to propose a new indoor visible light positioning system based on artificial intelligence algorithm and three kinds of indoor visible light ACO-OFDM system modulation classification schemes based on artificial intelligence algorithm.The specific research content is as follows:Firstly,the thesis analyzes the related theories of the positioning technology and modulation technology in indoor visible light communication system and typical artificial intelligence algorithm.The channel model of visible light communication system is mainly established,and some common indoor positioning and modulation techniques of visible light are briefly introduced,and then five typical intelligent algorithms are introduced,including GA,PSO,SVM,KNN and ANN.Secondly,a new indoor visible light location system for point classification based on artificial intelligence algorithm is proposed.Artificial intelligence algorithm is used in different processes of indoor visible light RSSI location,including height estimation scheme based on PSO-SVR regression algorithm,point type classification scheme based on GA-SVM algorithm and multi-path effect reduction scheme based on ANN neural network.It greatly improves the observability of the positioning process,makes it more optimizable and improvable,and effectively extends the traditional RSSI trilateral positioning from two-dimensional to three-dimensional.The receiving terminal can directly use the model trained by artificial intelligence algorithm,which reduces the computing power requirement on the hardware equipment of receiving terminal.Finally,a complete system is built in the simulation experiment to verify the effectiveness of the proposed scheme flow,which can reach centimeter-level positioning.Finally,aiming at M-PAM(Pulse Amplitude Modulation),M-PSK(Phase Shift Keying),M-QAM(Quadrature Amplitude Modulation)and other modulation modes in indoor visible ACO-OFDM system,an artificial intelligence algorithm modulation classification scheme based on signal high-order cumulant characteristic parameters is proposed.In this paper,the channel model of indoor visible ACO-OFDM system with first-order reflection is constructed.The theoretical research and simulation of M-PAM,M-PSK,M-QAM modulation are carried out.Based on the characteristic value of signal high-order cumulants,each order cumulants of modulation style are calculated according to the formula.Finally,the fourth order cumulants and the sixth order cumulants are selected as the characteristic values of signals,so as to carry out effective classification.This paper considers three artificial intelligence algorithm schemes,including PSO-SVM,ANN and KNN algorithms.Finally,the simulation results show that the proposed artificial intelligence algorithm modulation classification scheme can effectively realize the inter-class and intra-class recognition of M-PAM,M-PSK and M-QAM modulation modes in the visible light ACO-OFDM system under the condition of unknown signal-to-noise ratio. |
