Research On The Methods Of Indoor Wireless Location Based On WiFi

The requirements for information of people grow day by day due to an era of information ? large data and cloud computing nowadays.The position information is particularly important,and the service based on position information has changed our life all the time.It has showed powerful development in the fields of medical assistance,disaster relief,security supervision,social network,smart home,personalized information transfer,commodity circulation and so all.It is the development direction of information service in the future.With the rapid development of wireless network technology and the applications of smart mobile terminal in a wide range of areas,the existing positioning technology,especially indoor wireless location technology,meets the demand for pervasive computing so that to enter our life step by step in terms of use costs?operability?convenience,etc.High-accuracy indoor wireless location technology is the key and prerequisite for the realization of location information service.This paper aims to investigate indoor wireless location technology based wireless-fidelity(WiFi)and received signal strength(RSS).The following is the main content of the work:1?Firstly,we summarize the classification? methods as well as algorithms of current indoor wireless location technology,and the advantages and disadvantages of each are analyzed so that to select corresponding positioning method.The experiments are carried out to verify the influence of key parameters on indoor wireless location.2?Further,the basic principles of indoor wireless signal's propagation model and BP neural network are studied in detail.The model parameters of BP neural network are determined.The process for indoor wireless location method is presented based on signal propagation loss model fitting by BP neural network.The positioning accuracy is estimated by simulation.3?In the end of the paper,the main work concentrates on three aspects:1)study the basic principles of location fingerprinting and machine learning theory(including K-means clustering and support vector machines.2)determine the kernel function and parameters and improve the multi-class of support vector machine.3)design the flow of indoor wireless location method based on the above-mentioned principles,including pre-treated of AP signal?clustering block?combined classifier?training data?sub-regional determination and user location estimated by positioning algorithm.