Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

Darwin Quezada-Gaibor; Joaquín Torres-Sospedra; Jari Nurmi; Yevgeni Koucheryavy; Joaquín Huerta

doi:10.1109/ICL-GNSS54081.2022.9797021

Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

Darwin Quezada-Gaibor, Joaquín Torres-Sospedra, Jari Nurmi, Yevgeni Koucheryavy, Joaquín Huerta

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

2 Citations (Scopus)

31 Downloads (Pure)

Abstract

Machine learning models have become an essential tool in current indoor positioning solutions, given their high capa-bilities to extract meaningful information from the environment. Convolutional neural networks (CNNs) are one of the most used neural networks (NNs) due to that they are capable of learning complex patterns from the input data. Another model used in indoor positioning solutions is the Extreme Learning Machine (ELM), which provides an acceptable generalization performance as well as a fast speed of learning. In this paper, we offer a lightweight combination of CNN and ELM, which provides a quick and accurate classification of building and floor, suitable for power and resource-constrained devices. As a result, the proposed model is 58% faster than the benchmark, with a slight improvement in the classification accuracy (by less than 1 %).

Original language	English
Title of host publication	2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings
Editors	Jari Nurmi, Elena-Simona Lohan, Joaquin Torres Sospedra, Heidi Kuusniemi, Aleksandr Ometov
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781665405751
ISBN (Print)	9781665405768
DOIs	https://doi.org/10.1109/ICL-GNSS54081.2022.9797021
Publication status	Published - 2022
Publication type	A4 Article in conference proceedings
Event	International Conference on Localization and GNSS - Tampere, Finland Duration: 7 Jun 2022 → 9 Jun 2022

Publication series

Name	International Conference on Localization and GNSS
ISSN (Print)	2325-0747
ISSN (Electronic)	2325-0771

Conference

Conference	International Conference on Localization and GNSS
Country/Territory	Finland
City	Tampere
Period	7/06/22 → 9/06/22

Keywords

deep learning
extreme learning machine
Indoor Localisation
Wi-Fi fingerprinting

Publication forum classification

Publication forum level 1

ASJC Scopus subject areas

Computer Networks and Communications
Aerospace Engineering

Access to Document

10.1109/ICL-GNSS54081.2022.9797021

Lightweight Hybrid CNN-ELM Model
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Accepted author manuscript, 227 KBLicence: Unspecified

https://urn.fi/URN:NBN:fi:tuni-202212139135Licence: Unspecified

Supplementary material "Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification"
Quezada Gaibor, D. (Creator), Torres-Sospedra, J. (Contributor), Nurmi, J. (Contributor), Koucheryavy, Y. (Contributor) & Huerta, J. (Contributor), Zenodo, 11 Mar 2022
DOI: 10.5281/zenodo.6347465
Dataset

Cite this

Quezada-Gaibor, D., Torres-Sospedra, J., Nurmi, J., Koucheryavy, Y., & Huerta, J. (2022). Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification. In J. Nurmi, E.-S. Lohan, J. T. Sospedra, H. Kuusniemi, & A. Ometov (Eds.), 2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings (International Conference on Localization and GNSS). IEEE. https://doi.org/10.1109/ICL-GNSS54081.2022.9797021

Quezada-Gaibor, Darwin ; Torres-Sospedra, Joaquín ; Nurmi, Jari et al. / Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification. 2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings. editor / Jari Nurmi ; Elena-Simona Lohan ; Joaquin Torres Sospedra ; Heidi Kuusniemi ; Aleksandr Ometov. IEEE, 2022. (International Conference on Localization and GNSS).

@inproceedings{c0db1d5cf0e24e2eafa068f72074cba9,

title = "Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification",

abstract = "Machine learning models have become an essential tool in current indoor positioning solutions, given their high capa-bilities to extract meaningful information from the environment. Convolutional neural networks (CNNs) are one of the most used neural networks (NNs) due to that they are capable of learning complex patterns from the input data. Another model used in indoor positioning solutions is the Extreme Learning Machine (ELM), which provides an acceptable generalization performance as well as a fast speed of learning. In this paper, we offer a lightweight combination of CNN and ELM, which provides a quick and accurate classification of building and floor, suitable for power and resource-constrained devices. As a result, the proposed model is 58% faster than the benchmark, with a slight improvement in the classification accuracy (by less than 1 %).",

keywords = "deep learning, extreme learning machine, Indoor Localisation, Wi-Fi fingerprinting",

author = "Darwin Quezada-Gaibor and Joaqu{\'i}n Torres-Sospedra and Jari Nurmi and Yevgeni Koucheryavy and Joaqu{\'i}n Huerta",

note = "Funding Information: Corresponding Author: D. Quezada Gaibor ([email protected]) The authors gratefully acknowledge funding from European Union{\textquoteright}s Horizon 2020 Research and Innovation programme under the Marie Sk{\l}odowska Curie grant agreements No. 813278 (A-WEAR: A network for dynamic wearable applications with privacy constraints, http://www.a-wear.eu/) and No. 101023072 (ORIENTATE: Low-cost Reliable Indoor Positioning in Smart Factories, http://orientate.dsi.uminho.pt). Publisher Copyright: {\textcopyright} 2022 IEEE. JUFOID=72237; International Conference on Localization and GNSS ; Conference date: 07-06-2022 Through 09-06-2022",

year = "2022",

doi = "10.1109/ICL-GNSS54081.2022.9797021",

language = "English",

isbn = "9781665405768",

series = "International Conference on Localization and GNSS",

publisher = "IEEE",

editor = "Jari Nurmi and Elena-Simona Lohan and Sospedra, {Joaquin Torres} and Heidi Kuusniemi and Aleksandr Ometov",

booktitle = "2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings",

address = "United States",

}

Quezada-Gaibor, D, Torres-Sospedra, J, Nurmi, J, Koucheryavy, Y & Huerta, J 2022, Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification. in J Nurmi, E-S Lohan, JT Sospedra, H Kuusniemi & A Ometov (eds), 2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings. International Conference on Localization and GNSS, IEEE, International Conference on Localization and GNSS, Tampere, Finland, 7/06/22. https://doi.org/10.1109/ICL-GNSS54081.2022.9797021

Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification. / Quezada-Gaibor, Darwin; Torres-Sospedra, Joaquín; Nurmi, Jari et al.
2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings. ed. / Jari Nurmi; Elena-Simona Lohan; Joaquin Torres Sospedra; Heidi Kuusniemi; Aleksandr Ometov. IEEE, 2022. (International Conference on Localization and GNSS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

AU - Quezada-Gaibor, Darwin

AU - Torres-Sospedra, Joaquín

AU - Nurmi, Jari

AU - Koucheryavy, Yevgeni

AU - Huerta, Joaquín

N1 - Funding Information: Corresponding Author: D. Quezada Gaibor ([email protected]) The authors gratefully acknowledge funding from European Union’s Horizon 2020 Research and Innovation programme under the Marie Skłodowska Curie grant agreements No. 813278 (A-WEAR: A network for dynamic wearable applications with privacy constraints, http://www.a-wear.eu/) and No. 101023072 (ORIENTATE: Low-cost Reliable Indoor Positioning in Smart Factories, http://orientate.dsi.uminho.pt). Publisher Copyright: © 2022 IEEE. JUFOID=72237

PY - 2022

Y1 - 2022

N2 - Machine learning models have become an essential tool in current indoor positioning solutions, given their high capa-bilities to extract meaningful information from the environment. Convolutional neural networks (CNNs) are one of the most used neural networks (NNs) due to that they are capable of learning complex patterns from the input data. Another model used in indoor positioning solutions is the Extreme Learning Machine (ELM), which provides an acceptable generalization performance as well as a fast speed of learning. In this paper, we offer a lightweight combination of CNN and ELM, which provides a quick and accurate classification of building and floor, suitable for power and resource-constrained devices. As a result, the proposed model is 58% faster than the benchmark, with a slight improvement in the classification accuracy (by less than 1 %).

AB - Machine learning models have become an essential tool in current indoor positioning solutions, given their high capa-bilities to extract meaningful information from the environment. Convolutional neural networks (CNNs) are one of the most used neural networks (NNs) due to that they are capable of learning complex patterns from the input data. Another model used in indoor positioning solutions is the Extreme Learning Machine (ELM), which provides an acceptable generalization performance as well as a fast speed of learning. In this paper, we offer a lightweight combination of CNN and ELM, which provides a quick and accurate classification of building and floor, suitable for power and resource-constrained devices. As a result, the proposed model is 58% faster than the benchmark, with a slight improvement in the classification accuracy (by less than 1 %).

KW - deep learning

KW - extreme learning machine

KW - Indoor Localisation

KW - Wi-Fi fingerprinting

U2 - 10.1109/ICL-GNSS54081.2022.9797021

DO - 10.1109/ICL-GNSS54081.2022.9797021

M3 - Conference contribution

AN - SCOPUS:85134559155

SN - 9781665405768

T3 - International Conference on Localization and GNSS

BT - 2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings

A2 - Nurmi, Jari

A2 - Lohan, Elena-Simona

A2 - Sospedra, Joaquin Torres

A2 - Kuusniemi, Heidi

A2 - Ometov, Aleksandr

PB - IEEE

T2 - International Conference on Localization and GNSS

Y2 - 7 June 2022 through 9 June 2022

ER -

Quezada-Gaibor D, Torres-Sospedra J, Nurmi J, Koucheryavy Y, Huerta J. Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification. In Nurmi J, Lohan ES, Sospedra JT, Kuusniemi H, Ometov A, editors, 2022 International Conference on Localization and GNSS, ICL-GNSS 2022 - Proceedings. IEEE. 2022. (International Conference on Localization and GNSS). doi: 10.1109/ICL-GNSS54081.2022.9797021

Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification

Abstract

Publication series

Conference

Keywords

Publication forum classification

ASJC Scopus subject areas

Access to Document

Fingerprint

Datasets

Supplementary material "Lightweight Hybrid CNN-ELM Model for Multi-building and Multi-floor Classification"

Cite this