Research On Indoor Positioning Signal Filtering And Positioning Algorithm Based On Wi-Fi Fingerprint

At present,location information is more and more important to people's lives,such as location information needed when sharing location with others,route navigation needed when traveling,indoor navigation needed when shopping in supermarkets,etc.Outdoor or indoor positioning technology.At the same time,among all the location information requirements,the location accuracy of outdoor location information is sufficient for daily use,but the location accuracy of indoor location information is not satisfactory,and the location is difficult.Therefore,the research related to indoor positioning is becoming more and more important.At present,the positioning technologies of indoor location information mainly include ultrasonic,ultra-wideband,RFID,infrared and so on.However,due to factors such as cost,network scale,and ubiquity,the mentioned positioning technologies cannot meet the needs of ubiquitous and popular computing applications.The Wi-Fi positioning technology has the advantages of low cost and strong applicability,which has become the focus of current scholars' research.However,the accuracy of the positioning method used in this technology is limited.Therefore,in the research of indoor location information,this paper uses Wi-Fi technology as the basis to conduct research on indoor positioning methods to improve the accuracy of Wi-Fi positioning.The main research work of this paper is as follows:1.After introducing the network principles and characteristics of Wi-Fi,filtering algorithms and other related technologies,some problems in current wireless network positioning are verified experimentally in Chapter 3,such as: Wi-Fi signal propagation is subject to multipath effects,absorption Influenced by factors such as effect,distance consumption,etc.,the distribution of RSSI at the same distance in different directions is different,and may even produce signal fluctuations such as sudden drop or sudden increase,and in the histogram statistics,RSSI is mostly skewness distribution.2.For the RSSI distribution problem described above,it is difficult for common filtering algorithms to effectively eliminate the gross errors caused by skewed distribution.Therefore,this paper proposes an improved RSSI filtering algorithm that can deal with skewed distribution noise.The algorithm is based on the Boxplot digital signal processing method,and then uses the Kalman Filter algorithm to update the RSSI,and finally obtains the RSSI average value to represent the RSSI of a single AP at the fingerprint point.The experimental results show that the filtering algorithm is more stable and reliable than the commonly used moving average filtering algorithm,and it can detect and eliminate the gross error of RSSI observation with skewed distribution characteristics,making the signal strength more stable,and finally improving the indoor Wi-Fi fingerprint.Accuracy and reliability of positioning technology.3.After filtering through the above algorithm and establishing an offline fingerprint database,an improved WKNN positioning algorithm is proposed to solve the problems of low positioning accuracy and large variance of positioning accuracy of traditional positioning methods.This algorithm is different from the traditional WKNN algorithm.The Sigmoid function is introduced as the weight distribution function,the speed of the target to be located is substituted into the weight distribution function,and the weights of the two positioning results before and after are allocated to improve the positioning accuracy.At the same time,after the positioning algorithm based on the sigmoid function,the Taylor series expansion method is introduced to further improve the accuracy of the positioning point.The experimental results show that,compared with the WKNN algorithm,the average error of the localization algorithm of the Sigmoid function is reduced by 22%,and the standard deviation is reduced by 16.1%;the average error of the localization algorithm combining the Sigmoid function and the Taylor series is reduced by 35%;and in the motion trajectory description,the trajectory curve described by the positioning algorithm of the sigmoid function is closer to the real route.To sum up,this paper has achieved good results in this research direction.At the same time,the proposed algorithm is simple and fast,and has wide application significance.