Research On Indoor Fingerprint Location Algorithm Based On WiFi And PDR Technology

The integration of WiFi(Wireless Fidelity)technology and PDR(Dead reckoning for pedestrians)technology is a relatively new indoor positioning method.Compared to other methods,this method is more universal,easier to operate,and cheaper,and it can realize the actual function of reacting quickly and accurately locating,indoors.It has great potential and research value in locating.This paper deals with the complex and time-consuming problem of building an offline fingerprint library in an indoor fingerprint algorithm based on WiFi and PDR technology.The online phase matching positioning algorithms K nearest neighbor(KNN)and weighted K nearest neighbor(WKNN)have been proposed to improve the problem of long positioning time and low positioning accuracy.The research contents of this paper are as follows:First,in Chapter 3 of this paper,the fingerprint base is constructed at the offline stage,and the signal strength(RSS)is accurately acquired using the path loss model,and the signal strength display(RSSI)is performed.We propose an adaptive path loss model interpolation algorithm that processes information.Then,by utilizing the use of the interpolation method at some reference points(APs),a finger print that reduces the amount of work for acquiring the signal strength is constructed.Next,in Chapter 4,the offline fingerprint bank is divided into sub-fingerprint banks using the Affinity Propagation Cluster(APC)method.In the online stage,a two-step positioning method is adopted,which reduces the position area to a thick position in a certain area according to the size of the distance metric and saves time for the next position.After that,fine positioning was performed using a matching positioning algorithm,and by improving the K-nearest neighbor algorithm,similarity measurement was performed using the improved Manhattan distance instead of the conventional European distance,and the positioning accuracy was improved.By replacing the fixed K value with the dynamic K value,the environmental adaptability of the algorithm is improved and the positioning accuracy is improved.Fusion of two positioning technologies Positioning is compared using the Extended Kalman Algorithm(EKF)to improve stability.Finally,in Chapter 5,we will examine the improved algorithm here by simulating a positioning system in a virtual environment.Simulation results show that the interpolation algorithm using the adaptive path loss model is closer in accuracy to artificial finger prints and less complex than other algorithms.At the same time,the improved WKNN algorithm has higher position accuracy than other algorithms;the improvement of the WKNN algorithm is more stable in fusion technology,the positioning time is significantly reduced,and the purpose of this paper is achieved.Figure[58]Table[13]Parameter[72]...