File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/WF-IoT48130.2020.9221271
- Scopus: eid_2-s2.0-85095600723
- WOS: WOS:000627822200090
Supplementary
- Citations:
- Appears in Collections:
Conference Paper: Accurate Stride Length Estimation via Fused Radio and Inertial Sensing
Title | Accurate Stride Length Estimation via Fused Radio and Inertial Sensing |
---|---|
Authors | |
Keywords | Stride length estimation inertial sensing WiFi sensing |
Issue Date | 2020 |
Citation | IEEE World Forum on Internet of Things, WF-IoT 2020 - Symposium Proceedings, 2020, article no. 9221271 How to Cite? |
Abstract | Stride length estimation has various applications, ranging from pedestrian tracking to individual healthcare. It is usually achieved by inertial sensing, which, however, suffers from large errors due to the noisy readings on the low-cost commodity sensors and unconstrained human walking. Different from prior methods that explore inertial sensors only, in this paper, we present a fused radio and inertial sensing design that estimates fine-grained stride length. Our approach incorporates recent advances in WiFi sensing that underpins walking distance estimation at centimeter accuracy from radio signals. We then present a novel step detection algorithm using inertial sensor readings, which not only counts steps but also reports the time information of every detected step. The proposed algorithm then fuses the time-annotated distance estimates and steps to derive the stride length. The evaluation on a large public dataset shows that our step counting algorithm yields an error of 3%. Furthermore, experiments on commodity hardware with eight users demonstrate an error of about 2 cm in stride length estimation. |
Persistent Identifier | http://hdl.handle.net/10722/303711 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wu, Chenshu | - |
dc.contributor.author | Liu, K. J.Ray | - |
dc.date.accessioned | 2021-09-15T08:25:52Z | - |
dc.date.available | 2021-09-15T08:25:52Z | - |
dc.date.issued | 2020 | - |
dc.identifier.citation | IEEE World Forum on Internet of Things, WF-IoT 2020 - Symposium Proceedings, 2020, article no. 9221271 | - |
dc.identifier.uri | http://hdl.handle.net/10722/303711 | - |
dc.description.abstract | Stride length estimation has various applications, ranging from pedestrian tracking to individual healthcare. It is usually achieved by inertial sensing, which, however, suffers from large errors due to the noisy readings on the low-cost commodity sensors and unconstrained human walking. Different from prior methods that explore inertial sensors only, in this paper, we present a fused radio and inertial sensing design that estimates fine-grained stride length. Our approach incorporates recent advances in WiFi sensing that underpins walking distance estimation at centimeter accuracy from radio signals. We then present a novel step detection algorithm using inertial sensor readings, which not only counts steps but also reports the time information of every detected step. The proposed algorithm then fuses the time-annotated distance estimates and steps to derive the stride length. The evaluation on a large public dataset shows that our step counting algorithm yields an error of 3%. Furthermore, experiments on commodity hardware with eight users demonstrate an error of about 2 cm in stride length estimation. | - |
dc.language | eng | - |
dc.relation.ispartof | IEEE World Forum on Internet of Things, WF-IoT 2020 - Symposium Proceedings | - |
dc.subject | Stride length estimation | - |
dc.subject | inertial sensing | - |
dc.subject | WiFi sensing | - |
dc.title | Accurate Stride Length Estimation via Fused Radio and Inertial Sensing | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/WF-IoT48130.2020.9221271 | - |
dc.identifier.scopus | eid_2-s2.0-85095600723 | - |
dc.identifier.spage | article no. 9221271 | - |
dc.identifier.epage | article no. 9221271 | - |
dc.identifier.isi | WOS:000627822200090 | - |