EN
Talk to AIR NEXUS
MenuClose
Talk to AIR NEXUS
Data status: IllustrativeIllustrative system view

Airspace Intelligence Infrastructure

From Signals to Understanding.

AIR NEXUS turns fragmented airspace observations into trusted operational understanding.

Explore the PlatformTalk to AIR NEXUS

Evidence levelE1Architecture Defined

  1. Sense
  2. Understand
  3. Verify
  4. Protect
  5. Decide
  6. Remember
  7. Improve

Data status: Synthetic demonstration

Pick any object. Every panel follows.

A synthetic demonstration of how the platform is designed to hold one object across RF, PCL and EO sources. The scenario is fixed and drawn for illustration; it shows no real place, sensor or result.

Scenario time 01:04

Panoramic EO/IR strip

Tags mark the objects the EO channel sees. Select a tag or a row.

Situation view

RF gives a direction only. PCL gives range ellipses. A wider ellipse means a wider confidence band.

Observation table

TimeTrackClassSourcesConfidenceEvidenceFreshness
00:08Small drone 78%RFEOHighCorroboratedFresh
00:16Aircraft 93%PCLRFHighCorroboratedFresh
00:28Bird-like 54%EOMediumSingle sourceFresh
00:40UnknownPCLEOLowUnresolvedFresh
00:56Aircraft 81%PCLEOHighCorroboratedFresh

Selected object

A zoomed, grey, thermal-looking crop of a small multirotor aircraft against a hazy sky. It shows no real object.

T1 Small drone 78%

Track
T1 · Small drone 78%
Sources
RFEO
Confidence band
High
Evidence
Corroborated
Known then
00:16
Freshness
Fresh
Synthetic demonstration image. It shows no real place.

Sense

One airspace. Many signals.

Every family of sources sees part of the picture and none sees all of it. A source that reports nothing has not reported that the air is empty.

Six families of sources

  • Passive RF sensing

    Adds
    Notices objects that emit radio energy, without transmitting anything itself.
    Cannot prove alone
    Says nothing about an object that stays silent.

  • Passive coherent location

    Adds
    Reads existing broadcast signals as objects reflect them, to estimate range and motion.
    Cannot prove alone
    Cannot say what an object is, and depends on the broadcasts available.

  • Electro-optical and infrared

    Adds
    Gives a visual look that helps classify what another source found.
    Cannot prove alone
    Needs a line of sight and workable light or weather.

  • Cooperative and external aviation data

    Adds
    Brings in what aircraft and aviation services report about themselves.
    Cannot prove alone
    Misses anything that does not report, and a report can be wrong.

  • Atmospheric context

    Adds
    Explains why another source may be weaker at a given moment.
    Cannot prove alone
    Does not detect objects at all.

  • System telemetry

    Adds
    Tells the picture how healthy and how fresh each source is.
    Cannot prove alone
    Says nothing about the airspace itself.

Acoustic sensing belongs to the same family and is described in text only.

Unknown is not zero. When a source sends no report, the picture shows unknown, never clear.

  • Source A
  • Source B
  • Source C
Contribution by source: Data table
Time (relative)Source ASource BSource C
11060
21280
313102
412124
511137
6101310
791413
8101316
Contribution by source CH02Data status: Synthetic dataEvidence level: E0 · ConceptQuestion answered: Which sources contributed over time?Three stacked bands show how much each of three example sources added at each time step. The upper band grows as the third source comes in.Limitation: Sources are generic and the amounts are invented for illustration.
Source presence: Data table
ItemValue
Source AContributes
Source BContext only
Source CNot yet available
Source presence RD03Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: Which sources contribute to this state?Three source markers connect to one state and carry a text status each. A source being present does not mean it is trusted; the label says what it contributes.Limitation: Sources are generic and their positions are not geographic.

Understand

From observation to understanding.

Eight stages carry a raw observation to a state a reviewer can check. Step through them, then move the time control to compare what was known then with what is known now.

Time 36 of 36: 40 observations arrived, 38 attached to the track, 2 unknown.

Legend

  • Passive RF: direction only, distance unknown
  • PCL: a coarser position, and later
  • Electro-optical: a fine position, with gaps in haze
  • Unknown: fits no track
  • Track, as known then
  • Confidence band
  • Gate
  • Known now: hindsight, labelled as such
  • Sensor site (abstract)

  1. Three sources report on their own clocks. Each has its own delay, noise and gaps.

  2. Around each predicted position, a statistical gate marks which observations could belong to the track.

  3. The nearest observation inside the gate is attached to the track. Nothing outside it is forced in.

  4. A filter turns the attached observations into one smooth state: position and velocity.

  5. The band around the track is its uncertainty. It widens when sources go quiet and narrows when they agree.

  6. An observation that fits no track stays grey and unassigned. It is not promoted to a threat or thrown away as noise.

  7. Move the time control back. The track shows only what had arrived by that moment.

  8. Later data refine earlier positions. The dashed line is that corrected track, labelled as hindsight and never passed off as the original.

Fusion Theatre SX-03Data status: Synthetic dataA synthetic scenario on an abstract plane. It shows how the method works, not how any real system performed.

Two views of the same idea

How relationships form: Data table
StageValue
1Observations
2Alignment
3Candidate link
4Supported state
5Evidence kept
How relationships form CH09Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How do relationships form?Five steps, from separate observations to a supported state, show how a candidate relationship is formed, checked and then kept with its evidence.Limitation: A schematic of the logic, not of any particular system.
  • Confidence
  • Uncertainty band
Confidence and its drivers: Data table
Time (relative)ConfidenceUncertainty band
14026 - 54
24228 - 56
34533 - 57
46052 - 68
56356 - 70
66660 - 72
76862 - 74
85240 - 64
95038 - 62
105444 - 64
115849 - 67
126052 - 68
Confidence and its drivers CH04Data status: Synthetic dataEvidence level: E0 · ConceptQuestion answered: How did confidence evolveA line with a shaded band shows confidence rising when a new source joins and falling when one drops out. The band narrows as more sources agree.Limitation: The band is a drawn example of uncertainty, not a computed interval.

Remember

Live state. History. Context. Evidence.

The Airspace Digital Twin is designed to hold five layers of one picture together, so a reviewer can ask what is there, what was known, in what setting, how sure we are and on what evidence.

Layers of one picture

  • The unified state of each track, with its unknowns kept visible.

  • Every earlier state stays stored, so any moment can be reopened.

  • Terrain, weather and protected areas that change how a signal should be read.

  • How sure the system is, and which source quality drives it.

  • The method, record and limit behind each claim.

Data status: Illustrative

Step through a replay

Replay: Data table
ItemValue
Observed1, 4
Inferred2
Unknown3
Replay RD08Data status: SimulationEvidence level: E0 · ConceptQuestion answered: What was known at each moment?A replay line separates what was observed from what was inferred and what was unknown. Stepping forward or back reveals only what was known at that step.Limitation: A simulated sequence. It never shows predicted paths or knowledge gained afterwards.

Step 1 of 4Step 1

Explore the Digital Twin

One intelligence layer. Multiple sources.

Four bands carry data from the first observation to the system that uses it. Open a band to see what it holds.

Four bands of the architecture

  1. ObserveMany sources enter through adapters that record where each observation came from and how fresh it is.
    • Source adapters
    • Clock and freshness
    • Source health
  2. UnderstandObservations are correlated, fused and given a confidence that can be inspected.
    • Correlation
    • Multi-sensor fusion
    • Classification in context
  3. PreserveState, history and evidence are kept together, so any past moment can be reviewed.
    • Unified state
    • History and replay
    • Provenance chain
  4. DeliverGoverned interfaces hand the picture to the systems that need it, with its limits attached.
    • Role-based views
    • Interfaces that state what is unknown
    • Audit trail
Data flow: Data table
StageValue
1Sources
2Model
3Policy
4Interface
5Consumer
6Audit
Data flow CH13Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How does data flow through the architecture?A pipeline of six stages, from sources to audit, shows how data moves through the model, the policy and the interface to the consumer, with a record kept at the end.Limitation: A conceptual flow. It does not describe a deployed system.

See the architecture

Protect

Built for real environments.

Six kinds of environment, with the protected ones first. Each panel says what AIR NEXUS is designed to add and where that stops.

Six environments

  • Critical infrastructure
    What it is designed to add
    A reviewable picture of the airspace around sites that must keep running.
    Where it stops
    A description of the intended contribution. No site result is claimed.
  • Low-altitude airspace
    What it is designed to add
    Awareness of low-altitude airspace, brought together from several modest sources.
    Where it stops
    Low-altitude detection depends on terrain, clutter and the sources available. Nothing is promised without them.
  • Airports and aviation
    What it is designed to add
    A shared picture of aircraft, UAS, wildlife and weather around active operations.
    Where it stops
    It supports human review. It does not replace air traffic control or any certified system.
  • Ports and coastal areas
    What it is designed to add
    Awareness over harbours and shorelines, where low flight, weather and traffic meet.
    Where it stops
    Coastal conditions change what each source can see. The panel claims no coverage.
  • City and regional airspace
    What it is designed to add
    One picture across many sites, built up from local ones.
    Where it stops
    Scale adds sources and gaps together, and the picture states both.
  • Wildlife and aviation safety
    What it is designed to add
    Context on birds and animals near the runway, for the people who manage that risk.
    Where it stops
    A planning aid. It predicts no strike and removes no risk.

  • Critical infrastructure

    What it is designed to add
    A reviewable picture of the airspace around sites that must keep running.
    Where it stops
    A description of the intended contribution. No site result is claimed.

  • Low-altitude airspace

    What it is designed to add
    Awareness of low-altitude airspace, brought together from several modest sources.
    Where it stops
    Low-altitude detection depends on terrain, clutter and the sources available. Nothing is promised without them.

  • Airports and aviation

    What it is designed to add
    A shared picture of aircraft, UAS, wildlife and weather around active operations.
    Where it stops
    It supports human review. It does not replace air traffic control or any certified system.

  • Ports and coastal areas

    What it is designed to add
    Awareness over harbours and shorelines, where low flight, weather and traffic meet.
    Where it stops
    Coastal conditions change what each source can see. The panel claims no coverage.

  • City and regional airspace

    What it is designed to add
    One picture across many sites, built up from local ones.
    Where it stops
    Scale adds sources and gaps together, and the picture states both.

  • Wildlife and aviation safety

    What it is designed to add
    Context on birds and animals near the runway, for the people who manage that risk.
    Where it stops
    A planning aid. It predicts no strike and removes no risk.

Data status: Illustrative Abstract marks for illustration, not coverage

Fifteen projects. One architecture.

The portfolio is grouped into five families. Choose a family to see how its projects connect to the core platform and which evidence level each one has reached.

The platform that holds the picture together: the core, the fusion engine and the data interface.

Projects in this family

  • Airspace Intelligence Core PlatformE1 Architecture Defined
  • Multi-Sensor Fusion EngineE0 Concept
  • Airspace Data & Intelligence APIE0 Concept
All fifteen projects by family

Core Intelligence

The platform that holds the picture together: the core, the fusion engine and the data interface.

  • Airspace Intelligence Core PlatformE1 Architecture Defined
  • Multi-Sensor Fusion EngineE0 Concept
  • Airspace Data & Intelligence APIE0 Concept

Sensing & Perception

The ways the architecture observes: passive RF, passive coherent location, electro-optical sensing and the distributed network.

  • Distributed Passive Radar NetworkE0 Concept
  • Passive RF Sensing NodeE1 Architecture Defined
  • Passive Coherent Location (PCL) RadarE1 Architecture Defined
  • Electro-Optical AI Detection & ClassificationE0 Concept

Protected Environments

Where the architecture is applied first: critical infrastructure, drone intrusion and airports.

  • Critical Infrastructure Airspace ProtectionE0 Concept
  • Drone Detection & Airspace Intrusion IntelligenceE0 Concept
  • Airport Airspace IntelligenceE0 Concept

Domain & Environmental

Specialist views of the surroundings: birds and wildlife, ports and coasts, and the local atmosphere.

  • Bird & Wildlife Radar IntelligenceE0 Concept
  • Port & Coastal Air-Surface IntelligenceE0 Concept
  • Hyperlocal Atmospheric IntelligenceE0 Concept

Networked Airspace

Scale across places: city and regional low-altitude networks, and the near-ground network.

  • Near-Ground Airspace Intelligence NetworkE0 Concept
  • City / Regional Low-Altitude Airspace NetworkE0 Concept

Data status: Illustrative An abstract map for illustration, not a system layout

Verify

Built on evidence.

Each project states the evidence level it has actually reached, and the levels above it are shown as not yet demonstrated.

The evidence ladder

  1. E0 · ConceptAn idea and a stated purpose, with nothing yet built to test it.Reached
  2. E1 · Architecture DefinedA written architecture that others can review.Reached
  3. E2 · Prototype ValidatedA working prototype checked against stated criteria.Not yet demonstrated
  4. E3 · Integrated ValidationSeveral parts checked together as one system.Not yet demonstrated
  5. E4 · Controlled Field ValidationChecked in the field, under controlled conditions.Not yet demonstrated
  6. E5 · Qualified / OperationalQualified, and operating under independent review.Not yet demonstrated
The curved lattice reflector of a radar antenna, painted orange, against an overcast sky.
A radar antenna shown as illustrative context. It is not AIR NEXUS hardware.Photo: FBuHL09, CC BY-SA 4.0, via Wikimedia Commons

A higher visual finish does not imply a higher evidence level.

Real hardware looks convincing in a photograph. It tells you nothing about what has been demonstrated.

How a claim is kept honest

  • Claims written to the level
  • Limits stated beside results
  • Data status on every chart
  • Unknown never shown as clear

Launch register

  • P01
  • P02
  • P03
  • P04
  • P05
  • P06
  • P07
  • P08
  • P09
  • P10
  • P11
  • P12
  • P13
  • P14
  • P15

Projects P01 to P03: architecture defined

Projects P04 to P15: concept

Maturity by stage: Data table
StageValue
E0 ConceptReached
E1 Architecture DefinedReached
E2 Prototype ValidatedNot yet reached
E3 Integrated ValidationNot yet reached
E4 Controlled Field ValidationNot yet reached
E5 Qualified / OperationalNot yet reached
Maturity by stage CH06Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How did maturity progress?A six-step ladder from E0 to E5 marks the stages an item has reached and the stages still ahead. The example stops at E1; later steps stay open.Limitation: The ladder is a reading aid. A step is reached only when its evidence exists.
Evidence attachment: Data table
ItemValue
StateEvidence
StateMethod
StateLimitation
Evidence attachment RD07Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: What evidence stands behind a state?A state is linked to its evidence, its method and its limitation by three short connectors. The link appears once and stays; it implies no proof by itself.Limitation: The linked items are examples. Real links point to documented records.

From site to region.

The architecture is designed to grow one stage at a time, from a single facility to a region, without changing how evidence is stated.

Five stages of scale

  1. Facility

    One site, with its own sources and its own local picture.

  2. Campus

    Several sources around one campus share a single picture.

  3. Multi-site

    Separate sites exchange what they know, and each keeps its own record.

  4. City or corridor

    Sites along a city or a corridor are read as one connected area.

  5. Region

    Many areas are brought together, and the gaps between them are shown as unknown.

Data status: Illustrative An abstract sequence for illustration, not a deployed network

Two views of scale

From site to region: Data table
StageValue
Site8
Local5
District3
Region1
From site to region CH18Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How does it scale from site to region?Four nested levels, site, local, district and region, show how detail is abstracted step by step as the view widens, with fewer marks at each larger level.Limitation: Levels are abstract. No real site, boundary or coverage is drawn.
  • Normal
  • Degraded
  • Unavailable
System health over time: Data table
Time (relative)Component 1Component 2Component 3
1NormalNormalNormal
2NormalNormalNormal
3NormalNormalNormal
4NormalNormalUnavailable
5NormalNormalUnavailable
6NormalDegradedNormal
7NormalDegradedNormal
8DegradedNormalNormal
9DegradedNormalDegraded
10NormalNormalNormal
11NormalNormalNormal
12NormalNormalNormal
System health over time CH07Data status: Synthetic dataEvidence level: E0 · ConceptQuestion answered: How has system health changed?Three bands, one for each of three example components, show normal, degraded and unavailable periods as different fills and patterns, so state never depends on colour alone.Limitation: Periods are invented and say nothing about any real component.
Distributed nodes: Data table
ItemValue
Node7
Relationship8
Moving marker1
Distributed nodes RD09Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How are distributed nodes related?Generalised nodes and a few relationships form a schematic network, with one marker moving slowly along a path. The layout is symbolic and carries no real topology.Limitation: Schematic only. Positions, distances and counts are not real.

Decide

Intelligence where your systems need it.

A governed interface is designed to hand the picture to other systems, with its time, its sources and what is still unknown attached to every object.

Fields of one object

observed_at
When the observation was made, not when it arrived.
sources
Which families of sources contributed.
position
Where the object was, with its stated uncertainty.
confidence
How far the picture can be trusted, kept apart from the evidence level.
state
Known or unknown. Unknown is never sent as clear.
provenance
A reference to the chain of steps that produced it.
An example object
{
  "object": "example-object",
  "observed_at": "example-time",
  "sources": [
    "passive_rf",
    "pcl"
  ],
  "position": {
    "value": "illustrative",
    "uncertainty": "stated"
  },
  "confidence": "stated",
  "state": "unknown",
  "provenance": "ref:example-chain"
}

A fictional object with invented values, shown to explain the fields. It comes from no real system.

How delivery stays governed

Data freshness: Data table
ItemUpdates
Under 1 min62
1 to 5 min28
Over 5 min10
Data freshness CH08Data status: Synthetic dataEvidence level: E0 · ConceptQuestion answered: How fresh is the data?Bars group example updates by age (under a minute, one to five minutes and over five minutes). Most updates are fresh, with a thin tail of older ones.Limitation: Ages are invented. No latency figure is implied for any real system.

Improve

Engineering the next layer of airspace understanding.

Short notes on method and limits. Each one says what it does not establish.

Latest notes

  • ExplainerDraft

    Unknown Is Not Zero: Reading Airspace Without False Reassurance

    Why an empty display must never be read as an empty sky, and how a state can say what it does not know.

  • Method noteDraft

    Known Then, Known Now: Why a Replay Must Not Rewrite the Past

    A replay is useful only if it shows what was known at the time, not what we learned afterwards.

  • Architecture noteDraft

    Passive Radar as One Layer Among Many

    Passive coherent location earns its place by joining other sources, not by standing alone.

Discuss your airspace environment.

Tell us about the place, the question and the systems around it. We will answer where the architecture applies and where it does not.

Talk to AIR NEXUS

Start with a topic

Please keep sensitive operational details out of your first message. We will agree a suitable channel first.

We say what is known, and we say what is not yet.