Anti-interference Analysis And Optimization Of Indoor Visible MIMO Communication System

Visible light communication is a new type of optical communication technology.It was mainly used to transmit information through brightness changes of LED(Light-Emitting Diode)lamps.It could be used for both illumination and communication,which had the advantages of high speed,low power consumption,anti-electromagnetic interference,etc.in recent years,visible light communication technology was developing rapidly.It had become one of the research hotspots in the field of wireless communication.Indoor LED MIMO systems have been widely studied due to the widespread use of LED arrays in indoor lighting systems.MIMO technology can be used in visible light communication systems to increase transmission bandwidth and increase transmission rate.However,There are some serious problems in VLC MIMO systems,such as severe inter-symbol interference,high channel correlation,and complex channel noise,which restricts the development of VLC MIMO systems.Aiming at these problems,this paper analyzes and optimizes anti-interference and noise reduction technology in VLC MIMO systems.The system model,channel link model and inter-symbol interference model have been established firstly.Then the channel capacity of VLC MIMO system has been analyzed,and the distribution of optical power and ISI power in the receiving plane has been simulated.Based on it,an optimized pyramid-type diversity reception model is proposed.The parameter-optimized diversity receiver can effectively improve the channel capacity of the system.Finally,based on the optimized diversity receiver,a BP neural network model is proposed to process communication signal,which can achieve better bit error rate performance than the conventional scheme.The specific work is as follows:(1)Firstly,a relatively complete indoor VLC MIMO system model included the direct and diffuse channel link models had been established.The relationship between the channel capacity and the diversity dimension of the VLC MIMO system had been analyzed.The distribution of optical power and ISI power in the receiving plane had been simulated.Simulation results showed that inter-symbol interference and channel correlation were the key factors that affected the performance of VLC MIMO systems.(2)Secondly,based on the established VLC MIMO system,an optimized pyramid angle diversity reception scheme with 4 transmitters and 4 receivers was proposed.Furthermore,an analytical framework is proposed to analyze the performance of pyramid angle diversity receiver,where different propagation scenarios such asline-of-sight and no-line-of-sight are considered.To improve system performance based on the proposed scheme,the diversity parameters had been optimized.The various inclination angle and azimuth angle effects to the channel capacity and BER were discussed,and the different separation distance between LEDs and PDs in 4x4 structure was also analyzed.Numerical simulation results showed that the maximum channel capacity could be obtained when the horizontal rotation angle was 45°.For transmission systems with a transmission rate of 100 Mbps the system could achieve the maximum average channel capacity,when the inclination angle was 45° and the LED spacing was2.6 m,.By optimizing the simulation parameters,the channel diversity gain could be effectively enhanced,the inter-symbol interference could be reduced,and the overall performance of the reception could be improved.(3)Simultaneously,based on the optimized pyramid-type diversity receiving structure,a signal processing scheme of BP neural network was proposed,which included multi-channel signal combining scheme and multi-channel signal detection scheme.Multi-channel signal combining scheme referred to combine multiple received signals of a pyramid diversity receiver with the BP neural network,which could improve the diversity gain of the system.The multi-channel signal detection scheme was aimed at MIMO multiplexing technology.Specifically,it referred that to replace the traditional signal detection technology with the BP neural network,which could remove inter-symbol interference while reducing noise interference.