| With the rapid development of various novel technologies and applications,the location-based services have been increasing fast relying on accurate positioning technology.Typical outdoor positioning systems such as global positioning system(GPS)and Beidou navigation satellite system(BDS)meet great challenges in indoor environment,while the signal is easy to be interfered by obstacles or difficult to cover the user's area.Due to the serious indoor positioning error,the outdoor positioning technology cannot be applied in indoor positioning scene directly.Compared with the traditional indoor positioning methods based on infrared,ultrasonic,ultra wideband,wireless local area network(Wi Fi),the visible light communication(VLC)based indoor positioning technology has the advantages of green environmental protection,rich spectrum resources,high communication capacity,safety and low cost,which is widely considered as one of the main solutions for indoor precise positioning in the future.In this thesis,the key technologies of indoor positioning technology based on visible light communication are studied by theoretical analysis and numerical simulation.Firstly,the principle of positioning technology based on visible light simultaneous interpreting is briefly introduced,and the advantages and disadvantages of different traditional location schemes are analyzed.The application of machine learning technology in indoor precise positioning is also introduced emphatically.In view of the shortcomings of traditional fingerprint location methods,such as long time and high computational complexity,this thesis proposes an improved visible fingerprint indoor location method based on k-means method.The improved indoor location method uses the signal strength information from different indoor light sources received by the location terminal to build features,physical coordinates as labels,and the improved k-means method to build fingerprint database.The position procedure will be roughly determined by triangle location firstly,then the fingerprint matching method is used to obtain accurately location.The simulation results show that the improved k-means method can improve the computing efficiency of mobile user terminals and reduce the energy consumption of mobile devices on the premise of ensuring the positioning accuracy compared with the traditional fingerprint based method.The optimal performance can be obtained by changing the number of fingerprint databases and adjusting the size of K.The simulation results show that when k is 4,the positioning accuracy and positioning time reach a balance. |
