Sensor‑Level CO₂ Analytics & Benchmarking (C2D)
Purpose of the Algorithm
This algorithm analyzes CO₂ behavior at the individual sensor level to detect abnormal pollution patterns, sensor drift, and relative deviations between sensors operating in similar environments.
It is designed to support both environmental insight and sensor quality assurance.
What the Algorithm Does
Per‑sensor CO₂ statistics
- mean, median
- standard deviation
- volatility index
- min / max
- P01 / P99 spike detection
Relative benchmarking
- Compares each sensor’s CO₂ levels against the global median
- Identifies sensors reporting consistently higher or lower values
Anomaly detection
- Detects CO₂ spikes and extreme values
- Flags sensors with excessive variance
Quality vs environment signal separation
- Helps distinguish real environmental pollution from sensor malfunction
Value in a Compute‑to‑Data Workflow
- Enables dirty‑farm vs broken‑sensor differentiation
- Produces compact, monetizable analytics outputs
- Safe for confidential agricultural or industrial datasets
Industry Value
Supports:
- Pollution hotspot detection
- Sensor calibration audits
- Environmental certification
- Advanced benchmarking products