Indoor Wifi positioning data preprocessing and trajectory reconstruction method

Abstract

The invention discloses an indoor Wifi positioning data preprocessing and trajectory reconstruction method, comprising the following steps: A, defining an indoor space POI; B, extracting correspondingpositioning data according to mac, and forming a mac behavior trajectory, Strj = { p1..., pi, pj,...pn }; C, analyzing the signal loss and an AP point, and judging the invalid trajectory; D, intersecting trajectory points in the Strj trajectory with the indoor POI, ranking all points in the POI, and constructing a temporary trajectory sequence Strjtmp = { p1,..., pi,...pn }; E, merging the trajectory points in the same POI, namely only keeping the earliest and latest trajectory points in a continuous time period in the POI, reconstructing a trajectory Strjnew = { p1,..., pi,...pn }; F, mapping the trajectory data to a two-dimensional plane based on which the drift points and the jump points in the trajectory are deleted; and G, implementing step E again, and merging the trajectory pointsin the same POI, thus realizing trajectory compression, and completing the extraction of a new trajectory. According to the invention, indoor space characteristics and indoor trajectory characteristics are met, the result is reliable, and the processing efficiency is high.

Description

technical field [0001] The invention relates to the technical field of indoor Wifi positioning data preprocessing and trajectory reconstruction, in particular to a method for indoor Wifi positioning data preprocessing and trajectory reconstruction. Background technique [0002] In recent years, with the development of mobile communication and wireless positioning, important progress has been made in obtaining the spatial position of individuals inside buildings, including Wifi, Bluetooth, Zigbee and RFID and other positioning methods, and the accuracy has reached sub-meter level. By analyzing indoor positioning data, we can understand and grasp the activities of indoor crowds, such as the distribution, density and trajectory of crowds, which is helpful for public disaster protection, public facilities optimization and commercial services. Positioning data preprocessing refers to the need to deal with many issues such as noise filtering, data compression, and stop point ident...

AI Technical Summary

Problems solved by technology

Traditional positioning data preprocessing methods are not necessarily applicable. For example, median or mean filtering cannot handle continuous noise points, and particle filtering cannot be applied to trajectory data processing of multi-floor structures in indoor spaces; commonly used trajectory clustering to extract stay points The method is also not suitable for indoor narrow locations, let alone give effective semantic information to the extracted stay points, and cannot achieve the fundamental purpose of extracting stay points

Images