A Time Window Positioning Method Based on Traditional Weighted k-Nearest Neighbor Technology

Abstract

The invention discloses a time window positioning method based on the traditional weighted K nearest neighbor technology, which belongs to the time window positioning method in a WiFi fingerprint positioning system. In this method, the signal strength data is first calculated according to the traditional weighted K-nearest neighbor technology, and if the same position is judged N times, the historical data is directly used for position calculation. Compared with the traditional weighted K-nearest neighbor technology, the present invention avoids the comparison with all fingerprint point data in each positioning process, and can effectively improve the calculation speed. At the same time, the traditional weighted K-nearest neighbor technology does not fully consider the inertial information of the user's movement, but the present invention tends to select the position of the user's previous N times of the same positioning, which improves the positioning accuracy.

Description

technical field [0001] The invention relates to a time window positioning method based on traditional weighted K nearest neighbor technology, in particular to a time window positioning method in a WiFi fingerprint positioning system, and belongs to the technical field of pattern matching in wireless fingerprint positioning. Background technique [0002] Traditional wireless fingerprint positioning technology generally uses pattern matching technology in the online positioning stage to find the fingerprint data that is closest to the input data. Common matching technologies include nearest neighbor technology (NN, Nearest Neighbor), K nearest neighbor technology (KNN, Kth Nearest Neighbor), weighted K nearest neighbor technology (WKNN, Weight Kth NearestNeighbour). Among them, WKNN technology is widely used in positioning and matching at the current stage due to its simple algorithm and good positioning effect. [0003] However, the traditional WKNN technology needs to compa...

AI Technical Summary

Problems solved by technology

[0003] However, the traditional WKNN technology needs to compare all the fingerprint point information in the system for each positioning request, and calculate the fingerprint data segment closest to the input data according to specific matching rules

This method is inefficient in calculation and ignores the continuity of the location of the user to be located, resulting in slow calculation, and the positioning accuracy needs to be improved

Tang Yang et al. (Tang Yang, Bai Yong, Ma Yue, Lan Zhangli, "Research on the Application of WiFi-based Fingerprint Matching Algorithm in Indoor Positioning", Computer Science, pp.73-75, 2016) proposed a method based on The way of fingerprint clusters greatly reduces the time required for positioning. This scheme sets up characteristic fingerprint points and reduces the number of searches through hierarchical search. The disadvantage is that it depends heavily on the distribution status of access points and does not take into account the historical location. information, positioning accuracy needs to be improved

Images