Few-shot learning in Wi-Fi-based indoor positioning

This paper explores the use of few-shot learning in Wi-Fi-based indoor positioning, utilizing convolutional neural networks (CNNs) combined with meta-learning techniques to enhance the accuracy and efficiency of positioning systems. The focus is on addressing the challenge of limited labeled data, a...

Full description

Saved in:
Bibliographic Details
Main Authors: Xie, Feng, Lam, Soi Hoi, Xie, Ming, Wang, Cheng
Other Authors: School of Mechanical and Aerospace Engineering
Format: Article
Language:English
Published: 2024
Subjects:
Online Access:https://hdl.handle.net/10356/181449
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-181449
record_format dspace
spelling sg-ntu-dr.10356-1814492024-12-07T16:49:06Z Few-shot learning in Wi-Fi-based indoor positioning Xie, Feng Lam, Soi Hoi Xie, Ming Wang, Cheng School of Mechanical and Aerospace Engineering Engineering Few-shot learning Indoor positioning This paper explores the use of few-shot learning in Wi-Fi-based indoor positioning, utilizing convolutional neural networks (CNNs) combined with meta-learning techniques to enhance the accuracy and efficiency of positioning systems. The focus is on addressing the challenge of limited labeled data, a prevalent issue in extensive indoor environments. The study explores various scenarios, comparing the performance of the base CNN and meta-learning models. The meta-learning approach involves few-shot learning tasks, such as three-way N-shot, five-way N-shot, etc., to enhance the model's ability to generalize from limited data. The experiments were conducted across various scenarios, evaluating the performance of the models with different numbers of samples per class (K) after filtering by cosine similarity (FCS) during both the stages of data preprocessing and meta-learning. The scenarios included both base classes and novel classes, with and without meta-learning. The results indicated that the base CNN model achieved varying accuracy levels depending on the scenario and the number of samples per class retained after FCS. Meta-learning performed acceptably in scenarios with fewer samples, which are the distinct datasets pertaining to novel classes. With 20 samples per class, the base CNN achieved an accuracy of 0.80 during the pre-training stage, while meta-learning (three-way one-shot) achieved an accuracy of 0.78 on a new small dataset with novel classes. Published version The APC was paid by Sanda University, Shanghai, China. (Key Funded Project of the Sanda University, funding number: 2023ZD01). 2024-12-02T06:57:14Z 2024-12-02T06:57:14Z 2024 Journal Article Xie, F., Lam, S. H., Xie, M. & Wang, C. (2024). Few-shot learning in Wi-Fi-based indoor positioning. Biomimetics, 9(9), 9090551-. https://dx.doi.org/10.3390/biomimetics9090551 2313-7673 https://hdl.handle.net/10356/181449 10.3390/biomimetics9090551 39329573 2-s2.0-85205088050 9 9 9090551 en Biomimetics © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering
Few-shot learning
Indoor positioning
spellingShingle Engineering
Few-shot learning
Indoor positioning
Xie, Feng
Lam, Soi Hoi
Xie, Ming
Wang, Cheng
Few-shot learning in Wi-Fi-based indoor positioning
description This paper explores the use of few-shot learning in Wi-Fi-based indoor positioning, utilizing convolutional neural networks (CNNs) combined with meta-learning techniques to enhance the accuracy and efficiency of positioning systems. The focus is on addressing the challenge of limited labeled data, a prevalent issue in extensive indoor environments. The study explores various scenarios, comparing the performance of the base CNN and meta-learning models. The meta-learning approach involves few-shot learning tasks, such as three-way N-shot, five-way N-shot, etc., to enhance the model's ability to generalize from limited data. The experiments were conducted across various scenarios, evaluating the performance of the models with different numbers of samples per class (K) after filtering by cosine similarity (FCS) during both the stages of data preprocessing and meta-learning. The scenarios included both base classes and novel classes, with and without meta-learning. The results indicated that the base CNN model achieved varying accuracy levels depending on the scenario and the number of samples per class retained after FCS. Meta-learning performed acceptably in scenarios with fewer samples, which are the distinct datasets pertaining to novel classes. With 20 samples per class, the base CNN achieved an accuracy of 0.80 during the pre-training stage, while meta-learning (three-way one-shot) achieved an accuracy of 0.78 on a new small dataset with novel classes.
author2 School of Mechanical and Aerospace Engineering
author_facet School of Mechanical and Aerospace Engineering
Xie, Feng
Lam, Soi Hoi
Xie, Ming
Wang, Cheng
format Article
author Xie, Feng
Lam, Soi Hoi
Xie, Ming
Wang, Cheng
author_sort Xie, Feng
title Few-shot learning in Wi-Fi-based indoor positioning
title_short Few-shot learning in Wi-Fi-based indoor positioning
title_full Few-shot learning in Wi-Fi-based indoor positioning
title_fullStr Few-shot learning in Wi-Fi-based indoor positioning
title_full_unstemmed Few-shot learning in Wi-Fi-based indoor positioning
title_sort few-shot learning in wi-fi-based indoor positioning
publishDate 2024
url https://hdl.handle.net/10356/181449
_version_ 1819113003265556480