Is something there?
Find a candidate animal call or radio emission in the surrounding noise. Missed and false detections must be measured under the conditions of use.
AI & research
From an observation
to understanding.
Monitoring supplies repeated observations. Detection, classification, localisation and re-identification ask different questions of them—in both acoustics and radio.
What kind of signal?
Estimate a species from its call, or investigate the characteristics and class of an RF signal. A class label is distinct from an individual identity.
Where did it originate?
Combine observations from different positions to estimate a source location. Time-difference-of-arrival methods depend on shared reception, timing calibration, geometry and propagation; report uncertainty.
Have we encountered it before?
Compare separate observations to investigate whether they came from the same animal or transmitter. Receiver, environment and session effects must not be mistaken for identity.
Bioacoustic research / XinYenYana
From species
to individuals.
Xeroth’s XYY ecological model research learns from animal sounds.
Its direction spans call detection, species classification and the investigation of individual-animal re-ID. Localisation research can add spatial context to repeated observations.
Changes in microphone, habitat, day and background noise can mislead a model. Verified examples and independent evaluation are essential before treating a match as an individual animal.
Explore XYY research ↗RF research
Locate emissions.
Study repeat transmitters.
The RF node explores coordinated discovery and source localisation. The separate RF re-ID node captures selected waveforms for fingerprint research, retaining acquisition context needed to assess repeat encounters.
A device fingerprint must remain useful across different receivers, locations, days and power cycles—and distinguish unfamiliar transmitters. It does not identify a person or prove who operates a radio.
Evidence before confidenceOriginal signals.
Original signals.
Known uncertainty.
Independent checks.
Acoustic and RF re-ID remain research capabilities. No trained RF identity model, verified field accuracy or universal localisation performance is claimed.
Begin with a landscape
Explore a connected landscape.
See how acoustic and RF sensing can work together in a staged system.