Research Of Modulation And Detection Technologies In Pixelated Visible Light Communication System

Pixelated visible light communication uses pixels to modulate signal,and a camera as a receiver.Owing to its easy implementation,it has been well applied in some special scenes and has been widely studied in recent years.In this thesis,the pixel modulation and demodulation technology,the boundary detection technology in pixelated visible light communication have been studied.A test system of pixelated visible light communication has also been set up.The main work is as follows:Firstly,based on the introduction of two common color spaces,a chroma modulation scheme based on YUV color space is proposed,and the modulation and demodulation schemes are given respectively for the case of 4 constellation points and 8 constellation points.By setting the test environment,the bit error rate of the proposed chroma modulation scheme was measured.It is found that the chroma modulation of the 4 constellation points has good anti-noise and anti-interference ability,and the chroma modulation of the 8 constellation points has a lower anti-interference ability than the 4 constellation points.Secondly,based on introduction of the classical Harris corner detection algorithm,an improved hierarchical corner detection algorithm is proposed and a corner point filter scheme is proposed.The improved hierarchical corner detection scheme performs the first corner detection on the thumbnail of the image,obtains the rough position of the boundary corner,andthen looks for a small neighborhood containing the boundary corner on the original.image to perform the second corner detection.By double corner detection,the proposed algorithm avoids finding the corners on all the pixels of the whole image,thus reducing the computational complexity of the algorithm.In addition,the proposed corner point filter scheme uses thumbnail and corner point morphological methods to improve the stability of boundary detection.Thirdly,based on the temporal and spatial correlation of the boundary corner points in the received image sequence,this thesis proposes a corner point tracking scheme which uses the trajectory prediction and the corner detection in the small area to correct the prediction error.This tracking solution maintains accurate positioning of the boundary corner points during communication.By using tracking rather than re-performing boundary detection,the effect of non corner points in the background object is reduced,the stability of the boundary location is improved,and the complexity of the boundary detection is reduced.Fourthly,using the chroma modulation and demodulation scheme,the improved hierarchical corner detection algorithm and the boundary corner tracking scheme proposed in this thesis,a pixelated visible light communication test system has been set up and the test results are given.