Summary
International Conference on Emerging Technologies for Communications
2021
Session Number:D1
Session:
Number:D1-3
Coordinate interpolation of Indoor Neural Network Localization by Particle Filter
Kaishin Hori, Satoru Aikawa, Shinichiro Yamamoto, Yuta Sakai ,
pp.-
Publication Date:2021/12/1
Online ISSN:2188-5079
DOI:10.34385/proc.68.D1-3
PDF download (431.8KB)
Summary:
In this study, we attempted to interpolate the coordinates output by the fingerprinting method for indoor location estimation using wireless LAN. Using Wi-Fi-based RSSI fingerprinting, we can know our indoor location. The fingerprinting method requires the RSSI measurement by the operator in advance. That has a problem the user can only estimate the location at the pre-measured coordinates. It is known that this problem can be solved by using a particle filter. By using a particle filter, we can estimate not only the pre-measured coordinates but also the inter- coordinates among them. Not only that, we can improve the accuracy of the position estimation based on the temporal dependency of the user assuming a pedestrian. In this paper, the result of experimental validation shows the proposed method improves the accuracy. We evaluated and enhanced the accuracy of position estimation using MSE with a particle filter having more particles than before. In addition, we did the accuracy of position estimation when using CNN with a particle filter.