ECSS Communications — Cessation of Emission (space-systems/ecss/e50-cessation-of-emission)
Use when the task is the single obligation of ECSS-E-ST-50C clause 5.3.3 — that the space-to-ground emission can be stopped by command — and the question is whether a particular inhibit architecture actually delivers it.
Domain quick reference
- The obligation is on the capability, not on a box. It is met by a path from a ground request to a carrier that is off, and the path is only as good as its weakest link: the receiver that takes the request, the decoder that acts on it, and the switch that removes drive from the power amplifier.
- The failure the requirement exists for is a transmitter that will not stop. An inhibit routed through the emitting chain — through the same unit, the same power rail, the same software task that is misbehaving — is not a second opinion on that failure. It is the same opinion.
- Safe mode is where cessation is most often wanted and least often available. A path that is configured out of existence when the platform drops to its survival configuration has no value in the case that matters, even though it tests perfectly in nominal mode.
- Reliance on flight software is a real limitation but not a disqualification. A software-decoded inhibit still works for the large majority of causes; it is recorded as a limitation so a reviewer can weigh it, and it is never allowed to be the only path.
- The quantity to hold is a budget, not a box latency. Ground reaction, uplink propagation and on-board execution all sit between the decision and the carrier going quiet, and only their sum can be compared with a regulatory or coordination deadline.
- One path is not a capability. Cessation that a single failure removes is cessation that is unavailable in exactly the compound cases that drive the requirement, so the time to quote is the time left after the fastest path has been taken away.
Workflow
- List every emitter the spacecraft can radiate from before looking at any path, since the obligation covers all of its transmissions and a single transmitter nobody can silence defeats it. Then validate each inhibit path as a record: an identifier, an execution latency, and whether it runs through the emitting chain, survives safe mode, and needs flight software. Reject a malformed record rather than assuming a default.
- Name the impairments on each path, then split them: running through the emitting chain or vanishing in safe mode removes the path from the credible set; software dependence is recorded as a limitation.
- Build the end-to-end budget for every credible path — ground reaction plus uplink delay plus on-board execution — and sort them.
- Take the second-fastest credible budget as the time that survives one failure. With fewer than two credible paths there is no time to quote, and that is itself the finding.
- Compare both the nominal and the degraded budget with the deadline using an explicit tolerance, so a budget that lands exactly on the bound reads the same on every host.
- Return the verdict with the findings that produced it, so a reviewer can see whether the gap is speed, independence, or a path that never existed. Grade the design met only where each emitter can be brought off the air by telecommand at any point in the mission; a transmitter that can be silenced in some configurations and not others is not one the obligation lets through.
Obligations
| Item | Step |
|---|---|
| ECSS-E-ST-50C Rev.2 5.3.3a | 6 |
Pitfalls
- Counting an inhibit that is decoded by the transmitter's own controller as an independent path. It shares the failure it is meant to answer, and an architecture built of two such paths has no cessation capability at all.
- Quoting the fastest path's latency as the cessation time. That number is only true while nothing has failed, which is not the condition the requirement is written for.
- Budgeting the on-board execution alone. Uplink propagation on a distant mission can exceed every on-board contribution combined, and a budget that omits it will pass on paper and miss in flight.
- Treating safe-mode unavailability as a configuration detail. The survival configuration is the one a stuck emission is most likely to coincide with, so a path lost there is lost when it is needed.
- Using a bare strict comparison against the deadline. Budgets are sums of floats, and a design that lands exactly on its bound must not be read as compliant on one host and non-compliant on another.
- Reading software dependence as a pass or a fail. It is neither: it is a limitation to record, and the architecture is judged on whether a non-software path exists beside it.
Behavior contract (gate 3)
The path validation, impairment naming, credible-path filtering, end-to-end budget, single-failure response time, tolerance-based deadline comparison and final verdict are exercised by the gate 3 contract test: scripts/test_e50_cessation_of_emission.py against scripts/e50_cessation_of_emission_logic.py (stdlib unittest, offline). Run: python3 scripts/test_e50_cessation_of_emission.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
