BLEAR: Practical Wireless Earphone Tracking under BLE protocol

Ge, Linfei1,2; Xie, Wentao1,2; Zhang, Jin1; Zhang, Qian2

10.1109/PERCOM59722.2024.10494424

2024

IEEE International Conference on Pervasive Computing and Communications (PerCom)

2474-2503

2024 IEEE INTERNATIONAL CONFERENCE ON PERVASIVE COMPUTING AND COMMUNICATIONS, PERCOM

MAR 11-15, 2024

null,Biarritz,FRANCE

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

Motion tracking is an important aspect of human-computer interaction (HCI) and recent research focuses on motion tracking using earphones' embedded acoustic sensors. However, these solutions can only be deployed on wired earphones, while most of the commercial earphones are wireless ones. This limitation arises because wireless earphones utilize the Bluetooth Low Energy (BLE) protocol for handling audio data, which blocks the usage of existing acoustic sensing solutions. Firstly, the low sampling rate of BLE prevents the system from processing high-frequency ultrasounds. However, the sensing signal for earphones must be ultrasonic to prevent disturbance to the user. Secondly, BLE employs an audio compression process that is applied with different compression rates with different bandwidths. This will break the structure of wideband signals usually used for acoustic sensing. To overcome these challenges, we present BLEAR, the first earphone-tracking system compatible with the BLE audio recording protocol. To let BLE earphones receive ultrasounds, BLEAR utilizes a specially designed bandwidth conversion scheme that uses a mask signal to trigger a non-linear effect that converts high-frequency components to low-frequency ones, thereby overcoming the low audio sampling rate restriction of BLE. Additionally, by strategically designing beacon signals to align with BLE's subband compression pattern, BLEAR mitigates the influence of audio compression and achieves accurate wireless earphone tracking. We implement a wireless earphone prototype for BLEAR and conduct extensive experiments involving 8 subjects to demonstrate its feasibility. The experimental results show that BLEAR achieves a mean distance tracking error of 3.37 cm, an angle tracking error of 5.3 degrees, and an accuracy of 97.14% in recognizing 7 common user activities. This work not only introduces a BLE-compatible earphone tracking solution but also establishes a foundation for broader BLE device tracking applications.

Human-computer interaction earphone motion tracking BLE

第一 ; 通讯

英语

CPCI-S

RGC["CERG 16204820","16206122","AoE/E-601/22-R","R8015","3030_006"]

Computer Science ; Telecommunications

Computer Science, Artificial Intelligence ; Computer Science, Information Systems ; Computer Science, Theory & Methods ; Telecommunications

WOS:001219488200003

Web of Science

1.Southern Univ Sci & Technol, CSE Dept, RITAS, Shenzhen, Peoples R China
2.Hong Kong Univ Sci & Technol, CSE Dept, Hong Kong, Peoples R China

Ge, Linfei,Xie, Wentao,Zhang, Jin,et al. BLEAR: Practical Wireless Earphone Tracking under BLE protocol[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2024.