5G Tiny-ML AI-based IoT eNose system for Hazardous Odour Detection and Classification

Rafael Fayos-Jordan; Mohammad Alselek; Mohamed Khadmaoui-Bichouna; Jaume Segura-Garcia; Jose M. Alcaraz-Calero

doi:10.1109/JSEN.2025.3567576

5G Tiny-ML AI-based IoT eNose system for Hazardous Odour Detection and Classification

Rafael Fayos-Jordan
, Mohammad Alselek
, Mohamed Khadmaoui-Bichouna
, Jaume Segura-Garcia
, Jose M. Alcaraz-Calero

University of the West of Scotland

Research output: Contribution to journal › Article › peer-review

72 Downloads (Pure)

Abstract

More than 2 million people have died in the world in 2019 exposed to hazard substances. In this context, it is of paramount importance to deliver effective systems to help minimizing such number of dead. The usage of advanced technologies, such as artificial intelligence (AI) and the Internet of Things (IoT), plays a critical role in the detection of hazardous substances within Industry 4.0. The combination of these technologies enhance the efficiency and accuracy of monitoring harmful materials, improving safety standards and operational processes in industrial environments. AI can be used to analyse vast amounts of data for identifying patterns and predicting potential hazards, while IoT connects various devices and sensors to ensure real-time tracking and prompt responses to risks. This technological synergy is essential for modern industries aiming to create safer and more automated systems. In this work, we propose a 5G AI-IoT e-Nose system for the real-time detection and classification of 5 hazardous odours. The proposed AI model is very lightweight and it is affordable for our IoT micro controller unit. The system has been validated in laboratory conditions, but has the advantage and potential impact to be effective in any scenario.

Original language	English
Pages (from-to)	25439-25449
Number of pages	11
Journal	IEEE Sensors Journal
Volume	25
Issue number	13
Early online date	12 May 2025
DOIs	https://doi.org/10.1109/JSEN.2025.3567576
Publication status	Published - 1 Jul 2025

Bibliographical note

Funding

This work was supported in part by United Kingdom Research and Innovation (UKRI) Engineering and Physical Sciences Research Council (EPSRC) National Edge Artificial Intelligence (AI) Hub for Real Data: Edge Intelligence for Cyberdisturbances and Data Quality under Grant EP/Y028813/1, in part by the European Union (EU) Horizon Programming Platform for Intelligent Collaborative Deployments Over Heterogeneous Edge-Internet of Things (IoT) Environments (P2CODE) Project under Grant HORIZON-CL4-2022-DATA-01-03/101093069, in part by the Ministry of Science and Innovation through the Project Agriculture 6.0 under Grant PID2021-126823OB-I00 and Grant TED2021-131040B-C33, and in part by the Generalitat Valenciana under Grant CIBEST/2023/101.

Keywords

Sensors
Artificial intelligence
Electronic noses
Sensor arrays
Industries
Intelligent sensors
Chemicals
Hazards
Compounds
Chlorine
Internet of Things (IoT)
hazardous substances
classification
5G
artificial intelligence (AI)

Access to Document

10.1109/JSEN.2025.3567576

AAM_5G_Tiny-ML_AI-based_IoT_eNose_system_for_Hazardous_Odour_Detection_and_Classification
Copyright © 2025 IEEE. This accepted manuscript version is made available under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).
Accepted author manuscript, 975 KBLicence: CC BY 4.0

Cite this

@article{977e1288aaa642bd8827c98bef3a0f37,

title = "5G Tiny-ML AI-based IoT eNose system for Hazardous Odour Detection and Classification",

abstract = "More than 2 million people have died in the world in 2019 exposed to hazard substances. In this context, it is of paramount importance to deliver effective systems to help minimizing such number of dead. The usage of advanced technologies, such as artificial intelligence (AI) and the Internet of Things (IoT), plays a critical role in the detection of hazardous substances within Industry 4.0. The combination of these technologies enhance the efficiency and accuracy of monitoring harmful materials, improving safety standards and operational processes in industrial environments. AI can be used to analyse vast amounts of data for identifying patterns and predicting potential hazards, while IoT connects various devices and sensors to ensure real-time tracking and prompt responses to risks. This technological synergy is essential for modern industries aiming to create safer and more automated systems. In this work, we propose a 5G AI-IoT e-Nose system for the real-time detection and classification of 5 hazardous odours. The proposed AI model is very lightweight and it is affordable for our IoT micro controller unit. The system has been validated in laboratory conditions, but has the advantage and potential impact to be effective in any scenario.",

keywords = "Sensors, Artificial intelligence, Electronic noses, Sensor arrays, Industries, Intelligent sensors, Chemicals, Hazards, Compounds, Chlorine, Internet of Things (IoT), hazardous substances, classification, 5G, artificial intelligence (AI)",

author = "Rafael Fayos-Jordan and Mohammad Alselek and Mohamed Khadmaoui-Bichouna and Jaume Segura-Garcia and Alcaraz-Calero, \{Jose M.\}",

note = "Copyright {\textcopyright} 2025 IEEE. This accepted manuscript version is made available under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).",

year = "2025",

month = jul,

day = "1",

doi = "10.1109/JSEN.2025.3567576",

language = "English",

volume = "25",

pages = "25439--25449",

journal = "IEEE Sensors Journal",

issn = "2379-9153",

publisher = "IEEE",

number = "13",

}

TY - JOUR

T1 - 5G Tiny-ML AI-based IoT eNose system for Hazardous Odour Detection and Classification

AU - Fayos-Jordan, Rafael

AU - Alselek, Mohammad

AU - Khadmaoui-Bichouna, Mohamed

AU - Segura-Garcia, Jaume

AU - Alcaraz-Calero, Jose M.

PY - 2025/7/1

Y1 - 2025/7/1

N2 - More than 2 million people have died in the world in 2019 exposed to hazard substances. In this context, it is of paramount importance to deliver effective systems to help minimizing such number of dead. The usage of advanced technologies, such as artificial intelligence (AI) and the Internet of Things (IoT), plays a critical role in the detection of hazardous substances within Industry 4.0. The combination of these technologies enhance the efficiency and accuracy of monitoring harmful materials, improving safety standards and operational processes in industrial environments. AI can be used to analyse vast amounts of data for identifying patterns and predicting potential hazards, while IoT connects various devices and sensors to ensure real-time tracking and prompt responses to risks. This technological synergy is essential for modern industries aiming to create safer and more automated systems. In this work, we propose a 5G AI-IoT e-Nose system for the real-time detection and classification of 5 hazardous odours. The proposed AI model is very lightweight and it is affordable for our IoT micro controller unit. The system has been validated in laboratory conditions, but has the advantage and potential impact to be effective in any scenario.

AB - More than 2 million people have died in the world in 2019 exposed to hazard substances. In this context, it is of paramount importance to deliver effective systems to help minimizing such number of dead. The usage of advanced technologies, such as artificial intelligence (AI) and the Internet of Things (IoT), plays a critical role in the detection of hazardous substances within Industry 4.0. The combination of these technologies enhance the efficiency and accuracy of monitoring harmful materials, improving safety standards and operational processes in industrial environments. AI can be used to analyse vast amounts of data for identifying patterns and predicting potential hazards, while IoT connects various devices and sensors to ensure real-time tracking and prompt responses to risks. This technological synergy is essential for modern industries aiming to create safer and more automated systems. In this work, we propose a 5G AI-IoT e-Nose system for the real-time detection and classification of 5 hazardous odours. The proposed AI model is very lightweight and it is affordable for our IoT micro controller unit. The system has been validated in laboratory conditions, but has the advantage and potential impact to be effective in any scenario.

KW - Sensors

KW - Artificial intelligence

KW - Electronic noses

KW - Sensor arrays

KW - Industries

KW - Intelligent sensors

KW - Chemicals

KW - Hazards

KW - Compounds

KW - Chlorine

KW - Internet of Things (IoT)

KW - hazardous substances

KW - classification

KW - 5G

KW - artificial intelligence (AI)

UR - https://ieeexplore.ieee.org/document/11002305/

UR - https://www.scopus.com/pages/publications/105005192920

U2 - 10.1109/JSEN.2025.3567576

DO - 10.1109/JSEN.2025.3567576

M3 - Article

SN - 2379-9153

VL - 25

SP - 25439

EP - 25449

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

IS - 13

ER -

5G Tiny-ML AI-based IoT eNose system for Hazardous Odour Detection and Classification

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this