PNT-resilience framework cross-walk

June 20, 2026 · View on GitHub

Aligned to, not certified. This maps Kshana's simulated capabilities to the published PNT-resilience frameworks. It is a simulation-derived self-assessment aligned to the DHS/CISA Resilient PNT Conformance Framework (RPCF) v2.0, the RethinkPNT/Firesmith Resist-Detect-Respond-Recover (RDRR) model, and Yang Yuanxi's resilient-PNT criteria. It is not a certification, accreditation, or compliance statement, and it carries no endorsement from DHS, IEEE, or any authority. Every row is labelled with its honest verification status (VALIDATED against an external oracle, or MODELLED from first principles with tests); the resilience-scoring layer itself is MODELLED.

The scoring engine (src/resilience/) consumes these capabilities and emits per-dimension sub-scores, never a single phantom number. See src/verification.rs for the machine-checked status invariants.

DHS RPCF v2.0 technique categories (the seven; "OLVIDMR" is our mnemonic)

CategoryKshana capability used as the sub-score driverModuleStatus
ObfuscateDeclared technique × source quality (no direct behavioural driver)resilience::scoreMODELLED
LimitDeclared technique × source qualityresilience::scoreMODELLED
VerifyImpairment-detector AUC under the scenario (spoof/jam/meacon monitors)impairment_eval, resilience::scoreMODELLED
IsolateDeclared technique × source qualityresilience::scoreMODELLED
DiversifyIndependent-group count → inverse-Simpson effective diversityresilience::diversityMODELLED
MitigateAvailability under denialfom, resilience::scoreMODELLED
RecoverHoldover coast × bounded-degradation gatefom, holdover, resilience::timelineMODELLED

RethinkPNT / Firesmith RDRR functions

FunctionDriverModuleStatus
ResistMean of Obfuscate/Limit/Isolate/Diversify sub-scoresresilience::scoreMODELLED
DetectVerify sub-score (detector AUC)resilience::scoreMODELLED
RespondMitigate sub-score (availability)resilience::scoreMODELLED
RecoverHoldover × bounded gate; resilience-timeline KPIs (detect/react/recover/loss)resilience::timelineMODELLED

Yang Yuanxi criteria (the subset the timing/detection FoMs speak to)

CriterionDriverModuleStatus
AvailabilityAvailability under denialresilience::scoreMODELLED
ReliabilityFilter integrity fractionfom, resilience::scoreMODELLED
ContinuityHoldover × bounded gateresilience::timelineMODELLED
AccuracyTiming/quality proxy — position-domain accuracy is NOT modelled (fom::positioning_performance returns an explicit gap)resilience::scoreMODELLED (timing-domain only)

What this cross-walk does not claim

  • No certified RPCF Level. The assigned Level is a tentative, simulation-derived reading with an explicit bounded-degradation gate, not a conferred maturity.
  • No position-domain accuracy. Kshana's figures of merit are timing-domain and detection metrics; position CEP/SEP/HPL require a multi-axis nav solution Kshana does not yet provide.
  • No field validation. The reference panel and threat ensemble are synthetic, parameter-grounded reductions (resilience::panel), stated as a first-class limitation.

The companion study (resilience::study) exists precisely to show that collapsing these dimensions into one composite score or one Level produces a rating whose architecture ranking is unstable under defensible weighting and threat choices — so this cross-walk is a measurement layer to be read per-dimension, not a single grade to certify against.