Research And Implementation Of WiFi Indoor Location Algorithm Based On RSSI

With the rapid development of science and technology,the demand for indoor location service is more and more.In indoor location technology,WiFi positioning technology has been developing rapidly in recent years because of its simple implementation,strong expansibility and low cost.Therefore,this paper focuses on WiFi location technology.When WiFi is used to locate,there are many kinds of location characteristics that can be chosen.In this paper,we choose the received signal strength RSSI as the location feature,because it does not need special hardware devices to obtain,and is easy to apply and popularize.This paper mainly includes the following aspects:Firstly,the research on how to obtain the WiFi signal intensity of the mobile terminal equipment is carried out,and the RSSI change characteristics are analyzed in both static and dynamic environments.Secondly,the WiFi signal is easily disturbed by the environment during the transmission process,which makes the positioning accuracy inaccurate.In this paper,a weighted median Gauss filter is proposed to preprocess RSSI.Compared with the single filter method,the method has better effect.Then using the two positioning methods-based on RSSI ranging and RSSI based on distance measurement,also called the model method and the position fingerprint method.Three signal propagation path loss models and three location algorithms based on RSSI ranging are introduced,and the location experiment is carried out in laboratory based on the logarithmic distance loss model and the three edge location algorithm.In the location matching stage of location fingerprinting,we studied the K algorithm,the weighted K method and the Bayesian location algorithm.Based on these three matching location algorithms,we carried out the location experiments in the real environment,and analyzed the location data.Finally,in view of the above algorithms,there are many problems in the research of signal propagation path loss,such as the large dependence of the signal propagation path loss,the inaccurate location results,and the large time and labor consumption.This paper further studies the WiFi indoor location algorithm based on neural network,mainly based on BP neural network and RBF neural network.This method has relatively strong linear ability and self-organizing training ability.A RBF neural network indoor localization algorithm based on LM is proposed.In order to verify the localization performance of the algorithm,a location experiment is carried out in the library.The location data confirm the effectiveness and applicability of the proposed algorithm.