Communications that survive the section's worst day
For signals branches, dismounted sections and command posts operating against a peer adversary that captures radios, jams bands and direction-finds emitters into fires.
- Users
- Corps of Signals, infantry sections, CPs
- Threats
- Capture, jamming, DF, traffic analysis
- Products
- Sigilcore, Driftwell, Saelvor, Watch
What happens to a section under EW
One radio, a net-wide re-key
The key is on the device, so a captured radio forces a re-key measured in days and run under contact. Units delay reporting the loss.
Primary, fallback, silence
Jamming forces the fallback and then forces silence. Failure is cliff-edged at exactly the moment the net matters.
The plan in the metadata
Traffic rises before every operation and the command post is the loudest node. Order of battle falls out without a word decrypted.
Remove the target, not just the weakness
- Capture becomes one revocation entry. Sigilcore keeps no long-term key on any radio; a captured set is inert and its loss is reported, not hidden.
- Jamming degrades gracefully. Driftwell sprays shards over every bearer the section carries; above threshold, a jammed band changes nothing.
- Tempo disappears from the envelope. Saelvor emits at a constant rate, so the orders group and H-hour are not visible in volume.
- The signals officer sees custody and exposure. Irythra Watch shows who can talk, who is compromised and where emission exposure is rising.
Bench
Run the air-gapped prototype and test vectors with the evaluating team; agree the threat model.
Lab against EW
Saelvor on issued ruggedised COTS against a real EW team: correlation under live collection, emission within the agreed DF budget.
Field with a regiment
Operators complete tasks without workarounds; the isolated-node grace policy is argued out with doctrine, not defaulted.
The costs for this mission, stated plainly
A radio genuinely cut off from its formation cannot originate traffic, and the bounded grace window that mitigates this is inherited by a captured radio. Constant-rate cover costs ≈4.8× bandwidth at full cover, painful on a dismounted battery over 72 hours. Quorum assembly is fine at message cadence and wrong at packet cadence. These are exactly the questions a user trial should answer.
What this mission uses
The binding constraint is a signals regiment willing to run a trial.
Not capital. If you can open that door, or want the architecture note and the runnable prototype to take apart yourself, we would like to hear from you.