← Back to the chapter · Previous detail
Technical detail · Sensor Fusion Architectures
Metadata at each fusion level
At data level, combine measurements before feature extraction, usually after calibration and time alignment. Feature-level outputs need their schema, window, and sensor coverage. State-level outputs carry estimates, tracks, covariance, and confidence from local filters or trackers. Decision-level outputs carry alarms, classes, votes, or abstentions when lower-level evidence is unavailable.
| Level | Example | Required Metadata | Failure Mode |
|---|---|---|---|
| Data | Two temperature probes fused after calibration and time alignment. | Units, calibration id, event time, freshness, raw value, and accepted/rejected flag. | Combining measurements that describe different times or units. |
| Feature | IMU and magnetometer features fused for motion-state classification. | Feature schema, window, normalisation, sensor coverage, and feature version. | Comparing features from different windows or preprocessing versions. |
| State | Track-to-track fusion from two local Kalman filters. | State vector, covariance, process model, time step, and correlation assumptions. | Overconfident fusion because shared evidence is counted twice. |
| Decision | Gateway alarm and cloud classifier vote on the same incident. | Label definition, confidence, abstain rule, detector version, and voting policy. | Voting over incompatible labels or suppressing low-confidence disagreement. |