IEEE PerCom2024
Four Presentations at IEEE PerCom 2024, Including Main Conference (Short) and Workshops
Publications & Talks
International Conference · 2024
A Preliminary Study on Angle of Arrival Estimation by MUSIC Algorithm Using Backscatter Tags
Abstract
Various methods employing RF signals, such as Wi-Fi, BLE, and UWB, have been proposed for estimating the locations of people and objects. However, one of the major challenges impeding their widespread adoption is the time and effort required to manage these systems, especially in terms of recharging and replacing batteries. To address this issue, our study focuses on localization using backscatter tags, which boast ultra-low power consumption. These tags operate by backscattering surrounding radio signals, thereby creating a frequency shift in the backscattered signal. As a preliminary investigation into localization, we utilized the MUSIC algorithm, commonly employed for Angle-of-Arrival (AoA) estimation. We adapted this algorithm for use with our backscatter tags and assessed its AoA estimation performance. Our experimental results indicate that AoA estimation is achievable with an error of 10.8° in a 5m$\times$ 9m conference room.
Research Note
Indoor positioning of people and objects has potential applications in various fields, such as shopping malls, nursing homes, and offices. Possible use cases include tracking human movement and displaying the location of lost items. However, attaching small devices (tags) that actively transmit Wi-Fi or BLE signals to numerous objects poses challenges in terms of maintenance, such as charging and battery replacement. Therefore, this study aims to estimate the position of Backscatter tags by performing angle of arrival (AoA) estimation.
Backscatter is a communication technology that switches between reflecting and absorbing ambient signals from nearby Wi-Fi or Bluetooth devices. Since Backscatter does not require the tag to generate its own carrier signal, it enables ultra-low-power communication. Additionally, when Backscatter tags scatter ambient signals, they generate unique frequency shifts. By registering pairs of shift frequencies and their corresponding tag locations in a database, it is possible to identify objects based on their frequency shifts.
For angle of arrival estimation of Backscatter tags, it is necessary to extract only the shift frequency components from the received signals. To achieve this, this study applies Fast Fourier Transform (FFT) to separate the received signals in the frequency domain and extract only the shift frequency components. Then, Inverse Fast Fourier Transform (IFFT)is used to convert the extracted frequency spectrum back into a time-domain signal, after which the MUSIC (Multiple Signal Classification) algorithm is applied for angle of arrival estimation.
Published Papers
- 山口雄大, 内山彰, & 東野輝夫. (2022). Backscatter タグを用いた MUSIC 法による到来角推定の性能評価. 研究報告高度交通システムとスマートコミュニティ (ITS), 2022(28), 1-7.
- Yamaguchi, Y., Erdélyi, V., Uchiyama, A., & Higashino, T. (2024, March). A Preliminary Study on Angle of Arrival Estimation by MUSIC Algorithm Using Backscatter Tags. In 2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) (pp. 696-701). IEEE.
Related Lab Activities
