Research On Indoor Positioning Based On Visible Light

At present,satellite-based outdoor positioning technology is quite mature,but satellite signals cannot be received indoors,so satellite signal-based positioning technology is not suitable for indoor positioning system's research and development.Therefore,a positioning technology that can complete indoor high-precision positioning without relying on satellite signals has received extensive attention and research.For indoor positioning technology,a large amount of research has been carried out at home and abroad,and the significant results have been achieved.The positioning accuracy and technical methods based on WiFi,Zigbee,RFID and UWB are very mature,but each of them has its shortcomings,so that they are not suitable for actual production.LED-based visible light positioning technology is a research hot spot at present.It has fast signal transmission speed,high precision,low price of LED,convenient layout and installation,and low cost.It can effectively solve the various problems and troubles encountered by the current indoor positioning technology.Therefore,the focus of the research is LED-based visible light positioning technology and further improved the positioning accuracy of the technology.This dissertation is based on visible light communication technology,the software algorithm of indoor positioning technology is researched to improve the positioning accuracy based on LED visible light positioning technology.This dissertation is mainly divided into the following four research sections.First,analyze the actual indoor environment and establish a channel model for visible light communication technology.Secondly,based on the establishment of the channel model and combining the physical characteristics of the photoelectric sensor.We focus on the analysis of the various positioning algorithms currently available in the visible light communication system,and propose two-dimensional positioning and three-dimensional positioning algorithms based on RSS and other algorithms for deep fusion to improve positioning accuracy.Thirdly,a weighted fusion method based on RSS and centroid is proposed for two-dimensional positioning.The two-dimensional positioning algorithm is optimized to optimize the simulation of noise interference,obstacle interference and multipath interference.Finally,aiming at 3D positioning,a deep fusion algorithm based on RSS and least squares position estimation is proposed to optimize the 3D positioning algorithm.(1)The traditional visible light communication channel model is based on the Lambert model theory.The Lambert model is an ideal environment.The signal receiving end and the signal transmitting end at the position to be measured are relatively parallel and have a certain distance.Therefore,the system model is not suitable for positioning measurement of abnormal posture.In reality,the posture of the signal receiving device at the position to be measured is uncertain,and there may be vertical displacement or tilt.In this regard,this paper takes such errors into consideration,and establishes a new visible light communication channel model that is more in line with the actual situation.(2)According to the analysis of the first part,the characteristics of the indoor environment,the characteristics of the visible light communication channel and the positioning analysis method based on RSS are determined,the interference factors affecting the positioning are determined,and the influence of these factors on the positioning accuracy is analyzed.The analysis results show that the main factors that interfere with the positioning accuracy include: noise signal interference,obstacle obstruction,multipath interference,etc.Combining the characteristics of these factors,it is finally clear that the positioning algorithm is optimized by different methods of eliminating errors.For the noise signal existing in the environment,this paper uses the Kalman filter method to eliminate the noise signal in the received signal as much as possible,so that the positioning accuracy is improved by about 23% In order to reduce the positioning accuracy caused by obstacles,this paper proposes a solution to restore and correct the missing channel state information.The adaptive correction method is used to eliminate errors,so that the positioning performance and accuracy of the visible light positioning system are generally improved.According to the test results,the average value of the positioning error has been increased by 0.09 m.Aiming at the multi-path interference problem in the positioning environment,a nonlinear estimation method is used to eliminate the error,which improves the positioning accuracy by 30% compared with the traditional linear least square estimation method.(3)This part is mainly based on the current research technology to conduct a three-dimensional positioning accuracy optimization algorithm for visible light.On the basis of the second part of the research,this part is based on the RSS distance measurement method to analyze and demonstrate the three-dimensional visible light positioning method.Based on the RSS distance measurement method to analyze and demonstrate the three-dimensional visible light positioning method.It is determined to optimize the design of the positioning algorithm that uses RSS positioning and least squares for deep fusion to solve the above factors on the positioning accuracy and obtain accurate three-dimensional positioning targets.The specific idea of the design in this part is: use Taylor series expansion to find the actual coordinate value of the position to be measured by multiple iterations of the positioning algorithm and the small square method to obtain higher positioning accuracy and better positioning performance.