Wearable technology has rapidly advanced, opening new possibilities for context-aware applications in fields such as healthcare and gait analysis, where distinguishing between indoor and outdoor environments is essential. This is often accomplished through technologies like GPS, Wi-Fi, cellular, and Bluetooth which, however, come with privacy concerns, high power consumption, and dependency on external infrastructure. To address these challenges, recent studies have preliminary exploited the ambient magnetic field, though comprehensive validation with real-life data is lacking. This paper seeks to validate machine learning techniques, i.e., random forest, extreme gradient boosting, and stacked long short-term memory networks, for indoor-outdoor discrimination using exclusively magnetometer data from the daily activities of 20 participants in four cities across three countries. The study investigated the most effective magnetometer placement (feet, lower back, and non-dominant wrist) and pre-processing techniques (e.g., features and window size). Reference data is obtained through a GPS-based algorithm coupled with a geographical database running on a smartphone. The extreme gradient boosting algorithm yielded the best results, with an accuracy of 0.91, an F1-score of 0.90, and an area under the ROC curve of 0.94. These findings confirm the feasibility of accurately estimating indoor/outdoor context information from magnetometer at the same update frequency of a common GPS, but with an important energy savings. The proposed model can be integrated into state-of-the-art gait analysis systems, being able to discriminate the location to avoid misinterpreting gait deviations in real-world settings, thus supporting continuous and ubiquitous gait monitoring. Datasets and algorithm implementations have been made publicly available.

Discriminating Between Indoor and Outdoor Environments During Daily Living Activities Using Local Magnetic Field Characteristics and Machine Learning Techniques / Marcianò, Vincenzo; Cereatti, Andrea; Bertuletti, Stefano; Bonci, Tecla; Alcock, Lisa; Gazit, Eran; Ireson, Neil; Bevilacqua, Antonio; Mazzà, Claudia; Ciravegna, Fabio; Del Din, Silvia; Hausdorff, Jeffrey M.; Ifrim, Georgiana; Caulfield, Brian; Caruso, Marco. - In: IEEE SENSORS JOURNAL. - ISSN 1530-437X. - 25:1(2025), pp. 1507-1515. [10.1109/jsen.2024.3490253]

Discriminating Between Indoor and Outdoor Environments During Daily Living Activities Using Local Magnetic Field Characteristics and Machine Learning Techniques

Cereatti, Andrea;Bertuletti, Stefano;Caruso, Marco
2025

Abstract

Wearable technology has rapidly advanced, opening new possibilities for context-aware applications in fields such as healthcare and gait analysis, where distinguishing between indoor and outdoor environments is essential. This is often accomplished through technologies like GPS, Wi-Fi, cellular, and Bluetooth which, however, come with privacy concerns, high power consumption, and dependency on external infrastructure. To address these challenges, recent studies have preliminary exploited the ambient magnetic field, though comprehensive validation with real-life data is lacking. This paper seeks to validate machine learning techniques, i.e., random forest, extreme gradient boosting, and stacked long short-term memory networks, for indoor-outdoor discrimination using exclusively magnetometer data from the daily activities of 20 participants in four cities across three countries. The study investigated the most effective magnetometer placement (feet, lower back, and non-dominant wrist) and pre-processing techniques (e.g., features and window size). Reference data is obtained through a GPS-based algorithm coupled with a geographical database running on a smartphone. The extreme gradient boosting algorithm yielded the best results, with an accuracy of 0.91, an F1-score of 0.90, and an area under the ROC curve of 0.94. These findings confirm the feasibility of accurately estimating indoor/outdoor context information from magnetometer at the same update frequency of a common GPS, but with an important energy savings. The proposed model can be integrated into state-of-the-art gait analysis systems, being able to discriminate the location to avoid misinterpreting gait deviations in real-world settings, thus supporting continuous and ubiquitous gait monitoring. Datasets and algorithm implementations have been made publicly available.
File in questo prodotto:
File Dimensione Formato  
caruso_2025_Discriminating_Between_Indoor_and_Outdoor_Environments_During_Daily.pdf

accesso aperto

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Creative commons
Dimensione 8.85 MB
Formato Adobe PDF
8.85 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2996406