Research On Positioning Method Of Indoor Visible Light Based On Position Fingerprint

With the development of the information age,the popularity of handed devices,and changes in human behavior,the changes in demand scenarios for Location Based Services(LBS)have become more apparent.The most prominent part of these is the gradual transition of application scenarios from outdoor to indoor.The LBS in the indoor environment has also begun to attract the attention of major players.The traditional outdoor global satellite navigation systems(such as China's Beidou,etc.)cannot be applied to complex indoor spaces because satellite signals have low power and cannot penetrate thick walls and obstacles.Therefore,wireless indoor positioning technology as the core of LBS technology has been booming in recent years,many scholars and technology companies have begun to increase investment in indoor wireless positioning technology research.Visible Light Communication(VLC),which is a member of many indoor wireless communication technologies,has the functions of data transmission and illumination,strong confidentiality,does not occupy wireless channel resources,no electromagnetic interference,and is free of licenses.With such advantages,indoor VLC has gradually become a research hotspot for many scholars at home and abroad.Correspondingly,the indoor VLC positioning system based on LED also slowly entered the field of view of researchers.After analyzing the principle of several localization algorithms commonly used in indoor VLC systems,this paper focuses on the research of indoor VLC location technology based on location fingerprinting.Currently,there are relatively few studies on position fingerprinting in the field of visible light.Based on the full discussion of the existing Weighted K Nearest Neighbor(WKNN)algorithms,this paper presents a SWKNN(Simple WKNN)algorithm from the perspective of K-value,and dynamically reduces K by each time.The value improves the positioning accuracy.Secondly,based on the time complexity and the intrinsic characteristics of Received Signal Strength(RSS)high-dimensional signals,a WKNN positioning technique based on the bipartite K-means clustering algorithm is proposed.The light intensity of the position coordinates corresponds to the high-dimensional vector value of the RSS.After preserving the original database,preprocess the data,and then iteratively divides the K-means clustering algorithm to form multiple different clusters.The similarity of elements within the cluster is high,and the similarity between clusters is lower,thus reducing the classic WKNN algorithm.At the same time as complexity increases positioning accuracy.Its main work is as follows:(1)Introduce the research status and development of indoor visible light location technology;(2)Study the channel model of indoor visible light location system,including LOS channels,NLOS channels,and noise effects,and discuss several common indoor visible light location technologies.Principle,comparing its performance,the final choice of location fingerprinting as the focus of this study;(3)Focused on the WKNN positioning model for indoor visible light positioning system,discuss the advantages and deficiencies,and give A SWKNN algorithm is proposed to improve the accuracy of the WKNN algorithm by periodically changing the value of K,and a simulation comparison is made based on the LOS channel;(4)Proposed the positioning scheme of this paper: based on the two-point K-means clustering and WKNN positioning algorithm,and given the simulation model,simulation parameters,discuss and verify the appropriate size of the two K values,and finally give the program and the classics proposed in this paper.The comparison of the positioning accuracy and algorithm time complexity of the WKNN algorithm validates the feasibility and effectiveness of the proposed algorithm.