phat 19(18): e5

Research Article

Indoor Localization Services for Hearing Aids using Bluetooth Low Energy

Download841 downloads
  • @ARTICLE{10.4108/eai.15-5-2019.162799,
        author={Michał Kapiczynski and Sergi Rotger-Griful and Stefan Wagner},
        title={Indoor Localization Services for Hearing Aids using Bluetooth Low Energy},
        journal={EAI Endorsed Transactions on Pervasive Health and Technology},
        keywords={Indoor Localization Services, Bluetooth Low Energy, Hearing Impairment, Hearing Aids},
  • Michał Kapiczynski
    Sergi Rotger-Griful
    Stefan Wagner
    Year: 2019
    Indoor Localization Services for Hearing Aids using Bluetooth Low Energy
    DOI: 10.4108/eai.15-5-2019.162799
Michał Kapiczynski1, Sergi Rotger-Griful2, Stefan Wagner1,*
  • 1: Aarhus University, Finlandsgade 22, DK-8240, Aarhus N, Denmark
  • 2: Eriksholm Research Centre, Oticon A/S, Rørtangvej 20, DK-3070, Snekkersten, Denmark
*Contact email:


BACKGROUND: Hearing loss is a common disorder which is usually treated with hearing aids. The context of use, whether being outside, inside of large or small rooms, in a quiet or noisy environment, or talking with someone near or at a distance, often requires the hearing aid user to adjust the settings of the hearing aids manually in order to reach an optimal or even acceptable hearing experience. Knowing the context of a hearing aid user in the home setting, including the position within the home, could likely automate and optimize relevant hearing aid settings as needed. Several modern hearing aids are already equipped with Bluetooth Low Energy (BLE) antennas enabling their connectivity with smartphones and other compatible devices. This allows for the development of positioning and localization services using BLE.

OBJECTIVES: The aim of this study was to investigate the potential of using the BLE radio signal of hearing aids to provide indoor localization and positioning support when used in combination with fixed radio access points in a home setting.

METHODS: A research platform based on a single Oticon hearing aid and three embedded Raspberry Pi computers placed at three strategic locations in a home setting was developed. The Raspberry Pis used the two statistical learning methods K-Nearest Neighbors and Decision Trees for non-obtrusive detection, classification, and determination of the location of the hearing aid. The efficacy of the research platform was evaluated during two hearing experience relevant efficacy studies: 1) room classification and 2) proximity detection of hearing aid users to their conversation partners. RESULTS: Room classification results provided an accuracy of 88.79% and it was found feasible and reliable to differentiate whether the hearing aid user was within a comfortable conversation distance or not.

CONCLUSION: These results open up for a wide range of audiological applications in indoor environments for supporting new context-aware services for improving the hearing experience of the users.