Area partitioning and positioning method and system based on space partitioning

Abstract

The invention provides an area partitioning and positioning method based on space partitioning. The method comprises the following steps that 1, after a user designates the subarea number, various optimal initial cluster center positions are determined through distance calculation, the within-class distance, the between-class distance and the reference point spacing limiting condition are introduced by integrating a K-means algorithm, a Fisher criterion and a self-organizing iterative analysis algorithm, and by means of multiple times of the cluster merging and splitting process, the area partitioning target is completed finally; and 2, subarea positioning is achieved by applying a machine learning technology in a semi-supervised mode. The invention further provides an area partitioning and positioning system based on space partitioning.

Description

technical field [0001] The present invention relates to the technical field of indoor positioning, in particular to a region division and positioning method and system based on space partitioning. Background technique [0002] The general clustering algorithm can reasonably divide the location fingerprint database according to the received signal strength characteristics. But strictly speaking, these clustering methods only play the role of database filtering by dividing the location fingerprints. During the user positioning process, the corresponding sub-location fingerprint databases can be selected for calculation, which saves the time required for the algorithm to traverse the database. It optimizes the system performance, but it cannot really realize the indoor area positioning function. This is precisely because when the location fingerprint map is divided by signal strength features, it does not take into account whether the actual reference point locations are also ...

AI Technical Summary

Problems solved by technology

But strictly speaking, these clustering methods only play the role of database filtering by dividing the location fingerprints. During the user positioning process, the corresponding sub-location fingerprint databases can be selected for calculation, which saves the time required for the algorithm to traverse the database. It optimizes the system performance, but it cannot really realize the indoor area positioning function

This is precisely because when the location fingerprint map is divided by signal strength features, it does not take into account whether the actual reference point locations are also gathered together in physical space, so it is inevitable that there will be some reference points due to similar RSS similarity metrics (such as Europe distance) are divided into one category, however in the indoor area these reference point locations are far apart

For this kind of singularity problem, some literatures adopt different processing methods, such as directly deleting the reference points at the corresponding positions, directly ignoring these singular points, or artificially changing the category of these singular points. However, these processing methods may lead to The change of the clustering center affects the overall performance of the system, and it is difficult to solve the problem fundamentally

However, the categories in traditional clustering algorithms are mostly generated according to the algorithm, which is difficult to adjust properly and lacks flexibility.

Images