Research On Indoor Visible Light Positioning Technology

In order to solve the poor performance of traditional satellite positioning in indoor,various indoor positioning technologies have come out one after another.Visible light indoor positioning using visible light communication is one of them.Compared with RF that is the traditional positioning technology,visible light positioning has many advantages such as no spectrum license,no electromagnetic radiation and lighting compatible positioning,which can be used in special occasions including mines,nuclear power plants and shopping malls.For this reason,this paper studies the system characteristics,positioning algorithms and detection methods of visible light positioning.The specific contents include the following aspects:1.By analogy with visible light non-imaging communication,the basic principle of visible light imaging communication,transmission characteristics of optical communication signals,and gain characteristics of optical communication channels are systematically introduced.By equating the system characteristics of imaging communication with transmitter radiation characteristics and receiver concentration characteristics of non-imaging communication and the dispersion characteristics of imaging pixels,the signal and channel model for visible light imaging communication and positioning is established.2.The K-means algorithm and KNN algorithm were reconstructed on the Hadoop cluster in parallel.The scene analysis method based on distributed computing of MapReduce was proposed,and the method was used to test the acceleration ratio and the scaling ratio.The off-line fingerprint database is clustered into an index subspace by a parallel K-means algorithm,and the parallel KNN algorithm is used to calculate and match the fingerprint distance in the index subspace,which solves the single-machine computation when calculating the fingerprint distance in the scene analysis method.3.Combining visual ranging with visible light imaging communication,a new imaging location method based on binocular ranging is proposed,which solves the problem that traditional wireless ranging cannot be applied to imaging communication and positioning,and reduces the complexity of implementing visible light positioning on imaging terminals such as tablets and mobile phones.4.The speckle signal of the visible light imaging communication light source was detected and analyzed by filtering denoising,gamma enhancement,graying,local threshold segmentation and Hough transform.The illumination,communication and positioning compatible design was optimized using the shutter shutter effect of the camera and the alternate nullness of the ITF code,and the detection method of the optical stripe and the coding and decoding scheme of the optimized signal were given.5.A visible light imaging platform for positioning is built,which transplanted u-boot,Linux kernel,minimum system of no desktop and touch screen dependent library tslib,graphical interface library Qt for development,and computer vision library OpenCV on the ARM processor as the imaging positioning receiver.A related GUI program was written to test the positioning system's imaging communication and imaging positioning functions.It was found that the positioning error of the positioning system in this paper is mainly concentrated in the range of 3cm to 11 cm when the receiver is indoors at different heights.