Assured Spectrum — Contested EW Deinterleaving
Bounded-coherence RF pulse deinterleaving on PDW streams — median V=1.0 vs CDIF ~0.03 on synthetic contested cells. Parallel EW path in Assured Sensing.
- 1.0
- Median V-measure (DEINT)
- 0.03
- CDIF baseline V-measure
The scenario
Set the picture
An ES/RWR front-end in a contested EM environment receives interleaved pulse descriptor words from dozens of emitters — friendly, hostile, and deceptive. Before threat-library correlation, the pulses must be separated into per-emitter trains.
Classical CDIF-family deinterleavers collapse when pulse density rises or when the environment does not match the tuning assumptions.
What it costs today
CDIF and variants require oracle emitter counts or heavy per-environment tuning.
ML deinterleavers need labeled PDW corpora and retraining when the battlespace shifts.
What changes with SolvNum
Assured Spectrum applies bounded-coherence rules to cluster PDW reports into per-stream trains without fixed K or learned weights.
Median V-measure 1.0 on synthetic contested PDW cells vs CDIF ~0.03. Deterministic, auditable assignment rule — same certification story as MoridainDefense. Complements the time-series Assured Sensing chain; does not replace BSE or MoridainDefense.
Measurable outcome
What we'll claim — and how it survives review
Each line below maps to a captured number in the demo section. Every number is reproducible from the SolvNum validation suite.
- DEINT synthetic benchmark: median V-measure 1.0 vs CDIF ~0.03 on dense/contested cells.
- No fixed emitter-count oracle required.
- Deterministic rule set — no ML training-data provenance chain.
- External Turing Challenge / operational PDW corpus: ROADMAP (G-038).
The demo
What was tested. How. What the script printed.
Synthetic contested PDW generator: interleaved trains at varying density. Compared Assured Spectrum DEINT vs CDIF baseline. Measured V-measure per cell.
Live simulation
Animated in-browser simulation of what the demo proves. The numbers underneath are the captured demo output.
Float64 + ML classifier — arbitration in progress
round 0 · 100.0% disagreeSolvNum — set union of integer tier tags
round 0 · 0.0% disagreeEach cell is one node's classification of one track (color = tier 0–3). Float baseline starts disagreeing and converges over 2 consensus rounds. SolvNum is identical-by-construction at round 0 — the band field is an integer; aggregation is set union, not consensus.
Composes with
Where this POC sits in the substrate
Every POC reinforces — and is reinforced by — others. Click through to see how each piece locks into the larger picture.
Evidence pointers
Where the claims live in the repo
These are the files a reviewer should run, read, or grep to re-derive every number on this page.
- moridain/business/deint/deint_product.md
- assured-spectrum portfolio technical.mdx
- solv/17-defense-validation-gaps.md (G-038)
Previous · POC 17
SolvFilter — Scalable Multi-Target EKF
Next · POC 19
PrecisionEdge — Tactical Link Compression
Want to see this in your environment?
Brief us on a program where this POC matters.
ITAR-aware. Air-gapped delivery available. Every claim above traces back to a script in the public repo.