Research on Fast Location of WKNN Indoor WiFi Fingerprint Based on Double Search

In order to reduce the influence of Wi-Fi signal fluctuation on the positioning accuracy, this paper improves the Wi-Fi fingerprint positioning algorithm by variational mode decomposition (VMD), and uses VMD to smooth the noise while maintaining the original characteristics of the signal. In order t...

Full description

Saved in:
Bibliographic Details
Published in2024 20th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) pp. 1 - 6
Main Authors Mao, Yongyi, Wang, Xiuhong, Wu, Zhiwei, Hui, Yani
Format Conference Proceeding
LanguageEnglish
Published IEEE 27.07.2024
Subjects
Accuracy
Fingerprint recognition
Indoor positioning
Knowledge discovery
Matched filters
Noise
Parallel processing
Real-time systems
Search problems
Two-way search
Wireless fidelity
WKNN
Online AccessGet full text

Cover

More Information
Summary:In order to reduce the influence of Wi-Fi signal fluctuation on the positioning accuracy, this paper improves the Wi-Fi fingerprint positioning algorithm by variational mode decomposition (VMD), and uses VMD to smooth the noise while maintaining the original characteristics of the signal. In order to solve the problem of low real-time positioning of WKNN algorithm when dealing with a large number of fingerprint data, this paper constructs a guide database and a target database, optimizes their data structures by using ball-tree, and realizes a two-way fast search mechanism by using multi-thread parallel processing. On the premise of ensuring the matching accuracy, the matching number of fingerprint points in the positioning process is effectively reduced and the positioning efficiency is accelerated. The data collected by esp8266 Wi-Fi signal collector designed by ourselves in real environment shows that the positioning accuracy and speed of this method are improved by about 20% and 40% respectively compared with traditional methods.
DOI:10.1109/ICNC-FSKD64080.2024.10702327