Visible Light Based Positioning System And Application

Visible Light Positioning(VLP)is a promising technology.Thanks to the widespread deployment of lighting and the linear propagation of visible light signals,VLP has the potential to overcome the reliability issues of radio positioning.This thesis studies the positioning problems in the indoor environment and wireless sensor network by using the Light Emitting Diode(LED),Photodiode(PD)and camera sensor.Both studies aim to take full advantage of the stability and reliability of visible light signals and explore the potential of VLP in various high-precision positioning requirements.Most of the existing positioning methods are based on radio signals.However,radio signals are subject to the influence of multipath propagation.For example,the complex layout in the indoor environment and electromagnetic interference in the same radio band inevitably affect the positioning accuracy.Although some researches based on the visible light signal can avoid such hassle,they cannot meet the requirements of simplicity and low cost.The positioning accuracy needs to be further improved.In particular,there are few efforts to achieve accurate orientation calculations while simultaneously achieving positioning.To solve this problem,this thesis designs a fast and accurate positioning scheme for movable objects in the indoor environment by comprehensively considering the requirements of high precision,low cost and easy-touse.Besides,this thesis also explores the application of visible light in neighbor identification and localization,providing a new solution to the problem of node localization.The main research contents and contributions of this thesis include the following aspects:(1)This thesis designs Polar Light,an indoor positioning system based on visible light,which focuses on the features of easy-to-use and mobility-support.Based on the rationales of light polarization and radiation propagation,the device orientation and position are calculated simultaneously.Polar Light can achieve a median device orientation error of 0.97°,a median 2D positioning error of 3.87 cm,and a median height error of 4.71 cm.The positioning delay is 985 ms.(2)This thesis designs Li Sensor,a node identification and localization system based on visible light.It designs an identity expression method and a multi-node identification algorithm.The network topology inference method based on neighbor identification is presented and verified.Li Sensor can achieve an average identification latency of 1.087 s,an average distance error of 0.37 cm and an average angle error of 1.67°.(3)This thesis implements the prototype of Polar Light and Li Sensor.Both of them use LEDs to send special light signals.PD and camera are utilized to receive light signals respectively.Combining with the Arduino development board,the positioning of mobile devices in the indoor environment and the identification and localization of neighbor nodes in the sensor network can be realized.