· Decision · Claim-safe

How to choose a signal filter: GSRF vs EMA vs Kalman

Restates locked Evidence / workbench numbers only. Not universal performance.

Most “which filter?” articles end in a single champion. Locked GSRF characterization does the opposite: job first, then tool.

One table chooser

JobPreferLocked reason
Mid-band ring / chatter calmGSRF Practical~+35% osc31; ~−70% sweet-band power on phase6.2 packs
Peak chase toward a normalGSRF Practical~38% peak-dev mean; 49/52 wins
Minimum MAD tracking of rawEMA / KalmanGSRF 0/52 MAD wins when spring on
Delayed plant as command feedbackNot balanced GSRF+59% / +97% / +106% MAD worse at 3/6/10 min
Fixed-FA residual event detectionEMA residual / detectorRecall ~0.44 GSRF vs ~0.83 EMA
Probabilistic state + noise modelKalman (or similar)Different objective entirely
Permanent level shiftsAdaptive x* GSRF or EMAFrozen x* ~9.7 → adaptive ~1.5 error probe

Questions that decide

  1. Is the costly failure ring around a known normal, or lagging the truth?
  2. Is your “noise” mid-band (period ~50–120 samples) or slow drift you must track?
  3. Will the filtered stream drive a delayed closed loop?
  4. Is your KPI MAD-to-raw, peak excursion, or false-alarm rate?

Trap: “just replace EMA”

Dropping balanced GSRF into an EMA slot without checking delay, FA, and MAD will look worse on the metrics EMA was hired for. Read When GSRF lost first.

Verdict

If ring/spikes around a normal dominate → try GSRF (/try). If tracking/delay/detection dominate → EMA/Kalman/detector. Full table: /compare.

Sources: Evidence · compare · excellence hunt v2 · characterization v4 delay · probes v5 alarm

Questions people actually ask

How do I choose GSRF vs EMA vs Kalman?

If costly failure is ring/spike chase around a normal → try GSRF. If lagging the truth, delayed command, or fixed-FA residual → EMA/Kalman-class tools. Probabilistic state fusion → Kalman.

Is GSRF a drop-in EMA replacement?

No. Different identity (log-space soft thermostat). Evaluate on your KPIs and read the safety cage.

Where is the decision table?

On the Compare page and this note — always subordinate to Evidence numbers.

Related research