Takamochi Kanda
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Takamochi Kanda has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Koji Yamamoto
Faculty of Information and Human Sciences, Kyoto Institute of Technology, Kyoto, Japan
2 shared publications - Hiroki Shimomura
Graduate School of Informatics, Kyoto University, Yoshida-Honmachi, Kyoto, Japan
1 shared publication - Hiromitsu Awano
Graduate School of Informatics, Kyoto University, Kyoto, Japan
1 shared publication - Sota Kondo
Graduate School of Informatics, Kyoto University, Kyoto, Japan
1 shared publication - Takashi Sato
Kyoto University
1 shared publication - Takayuki Nishio
School of Engineering, Institute of Science Tokyo, Tokyo, Japan
1 shared publication
Latest Publications
- Open Access
- 19 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 146008-146019
Beamforming Feedback-Based Respiration and Heart Rate Estimation Toward Firmware-Agnostic WiFi Sensing
WiFi-based vital sign monitoring has attracted growing attention for its potential applications in contactless healthcare. However, most existing techniques rely on channel state information (CSI), which typically requires custom firmware and specific chipsets. To address this issue, this study explores firmware-agnostic respiration and heart rate estimation using beamforming feedback (BFF), compressed representation of CSI. This eliminates the need for custom firmware or chipset support, enabling broader applicability using off-the-shelf devices. However, it is not trivial to apply CSI-based estimation techniques to BFF-based estimation because the information content and data structure of BFF differ from those of CSI. The proposed BFF-based estimation algorithm addresses this issue by adapting the CSI-based estimation techniques to work with BFF. The algorithm consists of four key components: subcarrier selection, data calibration, signal extraction, and respiration and heart rate estimation. The performance of the BFF-based estimation algorithm is experimentally validated in several indoor environments using commodity IEEE 802.11ac devices. Results show that respiration rate and heart rate can be estimated with average errors below 1 breaths/min and 10 beats/min, respectively. Furthermore, accuracy comparisons between BFF-based and CSI-based estimations are provided to investigate the impact of lossy compression from CSI to BFF, specifically singular value decomposition (SVD) calculation and quantization. Comparisons reveal that the accuracy degradation of the BFF-based estimation compared to CSI-based estimation is primarily caused by the quantization rather than the SVD calculation.
- 19 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 146008-146019
Open AccessBeamforming Feedback-Based Respiration and Heart Rate Estimation Toward Firmware-Agnostic WiFi Sensing
WiFi-based vital sign monitoring has attracted growing attention for its potential applications in contactless healthcare. However, most existing techniques rely on channel state information (CSI), which typically requires custom firmware and specific chipsets. To address this issue, this study explores firmware-agnostic respiration and heart rate estimation using beamforming feedback (BFF), compressed representation of CSI. This eliminates the need for custom firmware or chipset support, enabling broader applicability using off-the-shelf devices. However, it is not trivial to apply CSI-based estimation techniques to BFF-based estimation because the information content and data structure of BFF differ from those of CSI. The proposed BFF-based estimation algorithm addresses this issue by adapting the CSI-based estimation techniques to work with BFF. The algorithm consists of four key components: subcarrier selection, data calibration, signal extraction, and respiration and heart rate estimation. The performance of the BFF-based estimation algorithm is experimentally validated in several indoor environments using commodity IEEE 802.11ac devices. Results show that respiration rate and heart rate can be estimated with average errors below 1 breaths/min and 10 beats/min, respectively. Furthermore, accuracy comparisons between BFF-based and CSI-based estimations are provided to investigate the impact of lossy compression from CSI to BFF, specifically singular value decomposition (SVD) calculation and quantization. Comparisons reveal that the accuracy degradation of the BFF-based estimation compared to CSI-based estimation is primarily caused by the quantization rather than the SVD calculation.
- Open Access
- 1 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 58858-58868
ACK-Less Rate Adaptation Using Distributional Reinforcement Learning for Reliable IEEE 802.11bc Broadcast WLANs
As a step towards establishing reliable broadcast wireless local area networks (WLANs), this paper proposes acknowledgement (ACK)-less rate adaptation to alleviate reception failures at broadcast recipient stations (STAs) using distributional reinforcement learning (RL). The key point of this study is that the algorithms for learning the strategy of ACK-less rate adaptation are evaluated in terms of the broadcast performance, which is composed of the data rate of the broadcast access point (AP) and the reception success rate at the recipient STAs. ACK-less rate adaptation framework was realized using the received signal strength (RSS) of the uplink frames transmitted by the non-broadcast STAs to the non-broadcast APs, which correlated with the broadcast performance with a confounding effect from the deployment of the broadcast recipient STAs. However, this rate adaptation framework has the problem that it incurs the reception failures at a part of the broadcast recipient STAs, because deep Q-learning used in the previous framework cannot deal with the wide distribution of the broadcast performance. To address this challenge, this paper further discusses the rate adaptation using distributional RL, which approximates the entire distribution of the broadcast performance. The simulations confirmed the following: 1) Using the expected broadcast performance learned by deep Q-learning improved the performance in terms of the Pareto efficiency. 2) Learning the entire distribution of the broadcast performance enabled the broadcast AP to determine the tail of the distribution using risk measure, and to alleviate reception failures while implementing the rate adaptation in the same way as the method that learns only expected broadcast performance.
- 1 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 58858-58868
Open AccessACK-Less Rate Adaptation Using Distributional Reinforcement Learning for Reliable IEEE 802.11bc Broadcast WLANs
As a step towards establishing reliable broadcast wireless local area networks (WLANs), this paper proposes acknowledgement (ACK)-less rate adaptation to alleviate reception failures at broadcast recipient stations (STAs) using distributional reinforcement learning (RL). The key point of this study is that the algorithms for learning the strategy of ACK-less rate adaptation are evaluated in terms of the broadcast performance, which is composed of the data rate of the broadcast access point (AP) and the reception success rate at the recipient STAs. ACK-less rate adaptation framework was realized using the received signal strength (RSS) of the uplink frames transmitted by the non-broadcast STAs to the non-broadcast APs, which correlated with the broadcast performance with a confounding effect from the deployment of the broadcast recipient STAs. However, this rate adaptation framework has the problem that it incurs the reception failures at a part of the broadcast recipient STAs, because deep Q-learning used in the previous framework cannot deal with the wide distribution of the broadcast performance. To address this challenge, this paper further discusses the rate adaptation using distributional RL, which approximates the entire distribution of the broadcast performance. The simulations confirmed the following: 1) Using the expected broadcast performance learned by deep Q-learning improved the performance in terms of the Pareto efficiency. 2) Learning the entire distribution of the broadcast performance enabled the broadcast AP to determine the tail of the distribution using risk measure, and to alleviate reception failures while implementing the rate adaptation in the same way as the method that learns only expected broadcast performance.