Wi-Fi Fingerprint Database Refinement Method and Performance Analysis

Wi-Fi fingerprint database is a common method of indoor positioning, but its construction process is tedious. In order to solve the workload of constructing the fingerprint database, this paper studies a variety of interpolation methods to estimate the signal strength of unknown points, by using the signal strength of the known reference points. In this paper, the credibility function is proposed to classify the different signal strength restoring errors, the segmentation function is used to judge the error of the signal and compute the credibility of the interpolation algorithm in the refinement of the Wi-Fi fingerprint database. In order to study the feasibility of various interpolation algorithms, the corresponding experiments are carried out. The experimental data were collected on the 4th floor corridor of Beihang E Block, and the performance of the fingerprint database was analyzed by both credibility and positioning results. The experimental results show that using the appropriate interpolation algorithm can indeed greatly reduce the workload of constructing the fingerprint database under the two evaluation criteria of credibility and positioning results.