Wireless Localization Method Based On AHP-WKNN And Amendatory AKF

The demands for high precision positioning and navigation in indoor scenario are increasing along with the popularization of smartphones and Internet of things nowadays.The WiFi-based indoor positioning can fully make use of the widely deployed Access Points to achieve high positioning performance.With the increasing popularity of WLAN,indoor positioning technology based on WiFi has become a hot spot.In this context,indoor localization based on WiFi including offline stage and online stage is discussed in this paper.The main work is as follows:(1)In the online phase,the signals from all the access points are measured at the test point.Amendatory AKF is applied for noise reduction of the measured signal.Based on the existing AKF,a comparison and an amendment are introduced in each iteration.Simulation results show that the amendatory AKF filter has higher accuracy.(2)In the online phase,the RSS signal denoised by amendatory AKF is compared with the fingerprints in the fingerprint database and then the position of the test point is determined by AHP-WKNN.The traditional WKNN uses Euclidean distance to assign weights to the k selected fingerprints.The AHP is designed as a new weight assignment scheme in this paper.The simulation results show it effectively enlarges the influence of the fingerprint position gap has on the weight and improves the accuracy of the existing WKNNs.In conclusion,this paper provides theoretical and practical support for noise reduction of RSS signal of test points and assignment scheme of weights of WKNN in the indoor positioning systems based on WLAN.