Turning reactions into recognizable patterns.
A sensor array's raw response is a tangle of curves. A fingerprint is what survives processing: a stable, comparable pattern a model can learn and a protocol can reproduce.
What we research
Signal processing, sensor fusion, baseline normalization, and feature design — the step between raw electronics and machine-readable smell. The fingerprint is the unit everything downstream depends on, so it has to be stable across runs before any claim rests on it.
Technical terms: signal processing, feature extraction, sensor fusion, baseline normalization, time-series representation.
Status
The feature shapes and data contracts the fingerprint depends on are fixed. First hardware fingerprints — stable, repeatable, and comparable across runs — are the goal of the bench experiments now starting.
- →Machine smell — the array the fingerprint comes from.
- →On-device AI — the model that reads it.