A non-parametric change-detection gate based on dynamic template matching reduces HAR computational load by over 67% with 97-98% sensitivity on two public datasets while requiring only brief device calibration.
On-Sensor Online Learning and Classification Under 8 KB Memory,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SP 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
An Algorithm for On-Sensor Agnostic Detection of Changes in Human Activity for Ultra-Low-Power Applications
A non-parametric change-detection gate based on dynamic template matching reduces HAR computational load by over 67% with 97-98% sensitivity on two public datasets while requiring only brief device calibration.