TUHH Open Research
 

POSTER: Discrete wavelets as zero-parameter front-end for near-sensor cardiac-interval estimation from seismocardiography

Start Page
364
End Page
366
Citation
22nd Annual International Conference on Distributed Computing in Smart Systems and the Internet of Things, DCOSS-IoT 2026
Contribution to Conference
22nd Annual International Conference on Distributed Computing in Smart Systems and the Internet of Things, DCOSS-IoT 2026  
Publisher DOI
10.1109/DCOSS-IoT69657.2026.00066
Scopus ID
2-s2.0-105049715894
Publisher
IEEE
ISBN of container
979-8-3315-4670-0
979-8-3315-4671-7
We present the Discrete Wavelet Transform (DWT) as a zero-parameter front-end for near-sensor cardiac-interval estimation from Seismocardiography (SCG), preserving timing-critical structure under embedded constraints. Across 10 wavelet families, we benchmark reconstruction, peak-timing fidelity, and energy on an nRF52840 SoC, showing that energy scales with filter length, while signal fidelity depends on wavelet family. This yields an energy-fidelity Pareto frontier for principled wavelet selection, with ongoing work extending this representation into neural network pipelines for cardiac interval estimation with morphology-dependent inductive bias.
Subjects
cardiac interval estimation
DWT-NN frontend
Seismocardiography (SCG)
wavelet benchmarking
DDC Class
600: Technology