DEFENSE
FREQUENCY

The only platform that turns public data into real-time physical threat detection.

Explore the Science Request Investor Briefing
LIVE COINCIDENCE RANK
99.4
RED ALERT
p < 0.006 • Non-random
Identity Frequency 11 overlaps • 28 days
Spatial
≤ 100 m
Temporal
≤ 30 min
POISSON + GAUSSIAN DP
SCROLL TO BEGIN ↓
THE SYSTEMIC THREAT

Witness & VIP Intimidation Is Silencing Justice

In high-stakes cases, intimidation influences up to 50% of witness decisions. Modern tactics exploit public spaces and digital transparency, leaving only statistically improbable patterns of repeated proximity.

The Human & Institutional Cost

  • 30–50% of serious urban prosecutions compromised by witness intimidation
  • Cases collapse. Violent offenders walk free. Public trust in the rule of law erodes.
  • Victims and protected individuals face relocation, isolation, and trauma, or are silenced entirely.

Modern Coordinated Tactics

Orchestrated “chance” encounters, rotating surveillance teams, and psychological pressure designed to instill fear without leaving direct evidence. These patterns are only visible through rigorous analysis of public geospatial footprints.

This is 21st-century witness and VIP intimidation.

Traditional protections are reactive, resource-intensive, and often arrive too late.

Defense Frequency shifts the paradigm to proactive, mathematically rigorous detection.

THE PLATFORM

Identity Frequency.
Identity Coincidence.
Mathematical Certainty.

Defense Frequency is in a class of its own: the only privacy-first system that converts public geospatial data into actionable, real-time physical-world proximity intelligence for protecting witnesses and high-value individuals.

OUTPUT 1

Identity Frequency

The observed rate at which two pseudonymized digital identities occupy proximate real-world space, derived exclusively from their voluntary public digital footprints (geotagged posts, photos, location shares).

k = raw spatio-temporal overlaps per identity pair
OUTPUT 2

Identity Coincidence Rank

A mathematically rigorous score that quantifies how unlikely it is for observed proximity patterns to occur by chance, distinguishing random movement from coordinated activity.

High Coincidence Score Extremely low probability of randomness → strong signal of potential coordinated intimidation

Privacy & Legal Integrity by Design

  • Public data exclusively. No private device tracking, advertising IDs, or warrants required for baseline monitoring.
  • Irreversible pseudonymization. Source identifiers mapped to cryptographically random UUIDs on ingestion. De-pseudonymization requires judicial warrant + multi-admin approval.
  • Tunable (ε, δ)-differential privacy via Gaussian mechanism — Mathematical guarantee against re-identification while preserving signal for genuine high-k patterns.
SECURE BY DESIGN
Court-Admissible.
Constitutionally Sound.
Zero Private Data Liability.
SCIENTIFIC FOUNDATION

Rigorous Statistical Analysis for Real-World Patterns

Advanced probabilistic modeling that distinguishes random everyday encounters from coordinated, non-random patterns of proximity — built specifically for the sparse, real-world nature of public geospatial data.

Identity Frequency
Weekly overlaps detected
0 / week
Coincidence Rank
0
/ 100
Non-Random Detection
0%
Probability of coordinated pattern
1
Ingest Public Signals

Collect open-source location data and map it to protected identifiers.

2
Detect Proximity Events

Identify when different identities appear near each other in space and time.

3
Adjust for Normal Patterns

Account for expected movement based on location type, time, and conditions.

4
Score for Coordination

Calculate how unlikely the observed pattern is to occur by chance.

Why This Approach Excels for Judicial Protection
  • Purpose-built for sparse public data — no need for artificial “trial” counts that other models require
  • Reliable performance even in low-density or remote environments where many AI systems struggle
  • Produces clear, explainable scores that investigators, prosecutors, and courts can understand and defend
  • Transparent methodology with strong statistical grounding — designed for high-stakes legal use
VALIDATED PERFORMANCE
Significantly reduces false alerts compared to simpler methods while maintaining high sensitivity to genuine coordinated patterns — tested for real operational reliability.
REAL-WORLD IMPACT

Real Patterns. Real Protection.

Scenarios drawn from actual threat profiles handled by prosecutors and protection teams.

◉ Coordinated Intimidation
RANK 99.8 • 14+ overlaps • λ deviation 4.2×

Targeted individual reports repeated, patterned proximity from rotating groups across multiple locations. Platform identifies non-random clustering.

Outcome: Protection deployed. Witness testified successfully.
GANG-STALKING
◉ Multi-City Pattern
RANK 99.92 • 9 overlaps • λ deviation 5.1×

Informant experiences repeated encounters across jurisdictions. Platform shows synchronized movement at hotels and transit hubs with very low expected probability.

Outcome: Federal task force coordinates protection. Network dismantled.
RICO / ORGANIZED CRIME
◉ Workplace Harassment
RANK 99.7 • 10 overlaps • λ deviation 3.8×

Whistleblower notices repeated “chance” meetings near office and cafes by former colleagues, including during low-probability conditions.

Outcome: Company security and law enforcement intervene. Proceedings protected.
WHISTLEBLOWER
◉ Judicial Protection
RANK 99.4 • 8 overlaps • λ deviation 3.1×

High-profile judicial officer reports unusual activity near chambers and residence. Platform detects synchronized movement across high-λ venues.

Outcome: Security detail augmented. Pattern disrupted before escalation.
VIP / JUDICIAL
◉ Relocated Victim
RANK 99.5 • 7 overlaps • λ deviation 4.7×

Domestic violence survivor relocates after restraining order. Abuser appears near new locations. Platform flags repeated proximity in low-traffic areas.

Outcome: Court grants enhanced protection. Abuser charged with stalking.
DOMESTIC VIOLENCE
THE ARCHITECT BEHIND THE MISSION

About the Architect

Catherine 'Cat' Allen, Architect of Defense Frequency

Catherine “Cat” Allen

Architect of Defense Frequency

Cat is a digitally diverse woman with a passion for developing socially responsible, sustainable, and ethical technology. A data privacy advocate with six years in cybersecurity and nearly a decade writing production code, she is on a mission to defend individual freedom and protect people from both each other and the machines.

She architected Defense Frequency to fill a critical gap: turning public geospatial signals into real-time, privacy-preserving physical threat detection for VIPs and witnesses. Built secure-by-design with cryptographic protections and a serious on-premises capable, optimized-for-real-time tech stack.

Defense Frequency is in a class of its own. Designed to protect good humans.

BUILT FOR SCALE & TRUST

Privacy-First. Secure-by-Design. On-Prem.

Designed to uphold constitutional rights and protect against abuse of authority. Public data only. No private surveillance. Built to defend individuals, not enable overreach.

TECHNICAL ARCHITECTURE
  • FastAPI + PostGIS for real-time spatial queries (<1ms on millions of records)
  • React + MapLibre GL interactive investigator dashboards with satellite overlays
  • Dockerized with hybrid public sources (OSM + Overture + weather + NORAD TLEs)
  • Exponential backoff + redundant data pipelines for maximum resilience
PRIVACY & SECURITY POSTURE
  • Public data only — zero private telemetry or device access
  • AES-256 at rest, TLS 1.3 in transit, comprehensive audit logging
  • Role-based access control with mandatory MFA
  • Containerized least-privilege deployment ready for classified or air-gapped environments
TECHNOLOGY EDGE
  • Secure-by-design architecture with immutable audit logs
  • Quantum-ready encryption layered on top of proven cryptographic primitives
  • Full on-premises deployment capability with air-gapped support
  • Built for maximum operational security and long-term resilience
FIELD-READY OPSEC
Purpose-built for high-consequence operations. Fully supports air-gapped on-premises deployments and secure management from hardened GrapheneOS devices — delivering real-time physical threat detection where maximum operational security is non-negotiable.
The era of reactive witness protection is over. The era of proactive, science-driven judicial defense begins now.

Defense Frequency is the only platform that turns publicly available geospatial data into court-admissible, privacy-preserving intelligence capable of detecting coordinated intimidation against witnesses and high-value protected individuals before cases collapse.

Ready to protect justice at scale?

Sign NDA → Receive full technical whitepaper
Ethical geospatial intelligence • Defensible methodology • Clear path to scale
COMING SOON

Defense Frequency: Target Edition

Empowering individuals with direct control. In supported high-stakes deployments, Target Edition will connect via GrapheneOS devices to deliver secure encrypted messaging and real-time threat dashboards — giving both operators and those being protected live visibility.

▸ GrapheneOS Connected
▸ Encrypted Messaging
▸ Real-time Dashboards

More details coming soon.

FOR DOD & DOJ DECISION MAKERS

Frequently Asked Questions

Defense Frequency supports on-premises and air-gapped deployments. It can be configured with containerized architectures and does not require external cloud connectivity.

The platform operates exclusively on publicly available data. It does not require access to private devices or networks and is built to respect constitutional protections while helping detect abuse of authority and insider threats.

Defense Frequency does not require access to private devices or networks. It is designed to analyze open geospatial signals to help identify coordinated physical-world patterns.

The system uses statistical modeling with environmental adjustments. Outputs are probabilistic and intended to support human analysis rather than replace it.
This page is for informational and marketing purposes only. Defense Frequency and its representatives disclaim all liability for any actions, decisions, or consequences arising from the use of information presented herein.