kimo
DefinitionKimo DefenseAviation & intelligence

Multilateration

MLAT

Definition

Multilateration (MLAT) locates an aircraft from the differences in the time its transponder signal arrives at several synchronized ground receivers. Four receivers give a 3D position, or three if altitude is known. Because the position is computed on the ground, it does not depend on the aircraft’s own navigation.

Updated 4 sources3 min read

Multilateration (MLAT) locates an aircraft by measuring the tiny differences in the time its transponder signal reaches several ground receivers with synchronized clocks. Four receivers are enough to solve a 3D position. Three are enough if the altitude is known from another source. Because the position is computed on the ground, MLAT is independent of what the aircraft says about itself.

§01How does multilateration work?

A multilateration system is a set of antennas plus a central processor. The processor computes the aircraft’s position from the time difference of arrival (TDOA) of the same signal at different antennas. Mathematically, each TDOA between two antennas puts the aircraft on a hyperboloid. With four antennas the intersection gives a 3D position. With three, a position can still be calculated if altitude is known from another source, such as the Mode C reply.1 When receivers are spread across a wide region rather than one airport, the technique is called wide area multilateration (WAM).

§02Example: why timing matters

Radio covers about 300 m per microsecond. If two receivers’ clocks disagree by just one microsecond, the range difference is off by about 300 m before any geometry is applied. That is why MLAT networks invest heavily in synchronization (GNSS-disciplined clocks or reference transmitters) and why positions far outside the polygon formed by the receivers degrade quickly.

§03How is MLAT different from ADS-B?

ADS-BMultilateration
Who computes the positionThe aircraft (from its own navigation)The ground network (from signal timing)
NeedsADS-B Out equipmentAny transmitting transponder, plus 3–4+ synchronized receivers
Affected by aircraft GNSS problemsYesNo (receivers still need good timing)
Typical usePrimary position sourceCoverage for non-ADS-B aircraft; independent cross-check

The OpenSky researchers describe WAM as independent of the aircraft, unlike ADS-B, though it still requires the aircraft’s cooperation by transmitting.2 That independence makes MLAT a natural cross-check on ADS-B positions, for example during GNSS interference.

§04Common misconceptions

  • “MLAT can see any aircraft.” Only aircraft whose transponders are on and transmitting, and only where enough receivers hear them.
  • “More receivers always means better positions.” Geometry matters as much as count. Research on crowdsourced networks found classic multilateration struggles with random, unplanned receiver layouts and proposed alternatives that improved coverage by up to 2.5×.3
  • “MLAT replaces radar.” Operational systems fuse it with radar and ADS-B. EUROCONTROL’s ARTAS tracker, for one, combines WAM with primary and secondary radar, Mode S and ADS-B.4

§05How Kimo uses MLAT

Kimo keeps the position source on every point of the tracks model, so the Airspace map can style MLAT and ADS-B differently and compare them where both exist. A persistent gap between an aircraft’s broadcast position and its MLAT position goes to the review queue. Feeds come in through the ADS-B and OpenSky connectors; see the ingestion guide.

  • ADS-B: self-reported positions that MLAT can verify.
  • Mode S: the transmissions MLAT receivers time.
  • Data fusion: merging MLAT, ADS-B and radar into one track.

Frequently asked questions

How many receivers does MLAT need?

Four receivers that hear the same transmission give a 3D position. Three can work if altitude is known from the transponder’s altitude reply.

Is MLAT affected by GNSS jamming?

The aircraft’s position is not taken from its own GNSS, so MLAT keeps working when the aircraft’s navigation is degraded. Ground receivers that use GNSS for timing need their own protection.

Why do some flight trackers label positions “MLAT”?

To show the position was computed by the receiver network rather than broadcast by the aircraft. MLAT is typically used for aircraft without ADS-B Out.

Sources

4 references
  1. Wide Area Multilateration: Report on EATMP TRS 131/04 (NLR-CR-2004-472) (opens in a new tab)
    National Aerospace Laboratory NLR for EUROCONTROL2005eurocontrol.int

    Principle of multilateration: TDOA, hyperboloids, four vs three antennas.

  2. Bringing up OpenSky: A Large-scale ADS-B Sensor Network for Research (opens in a new tab)
    Schäfer, Strohmeier, Lenders, Martinovic, Wilhelm (IPSN 2014)2014cs.ox.ac.uk

    WAM independent of the aircraft unlike ADS-B; TDOA with four or more sensors.

  3. A Localization Approach for Crowdsourced Air Traffic Communication Networks (opens in a new tab)
    Strohmeier, Martinovic, Lenders, arXiv2016arxiv.org

    Limits of multilateration with unplanned receiver geometry; coverage up to 2.5× with k-NN.

  4. Air traffic management surveillance tracker and server (ARTAS) (opens in a new tab)
    EUROCONTROLeurocontrol.int

    Multi-sensor tracking using PSR, SSR, Mode S, WAM and ADS-B.

External sources were accessed at the time of writing. Kimo product details, customers and figures in examples are illustrative unless a source is cited.

Used in

Where Multilateration shows up in practice

3 resources
GuideAdvanced
Defense

Ingest ADS-B feeds into Kimo

Connect a receiver network or OpenSky, normalize state vectors, and model flights and tracks.

Hugo Lefèvre
12 min read
Whitepaper
Defense

Airspace Awareness from Open Data

ADS-B, Mode-S and GNSS interference: what open aviation data can reveal, its limits, and how to fuse it responsibly.

Hugo Lefèvre
32 pages
ArticleEngineering
Defense

Airspace awareness from open ADS-B data

How ADS-B works, what open receiver networks can and cannot tell you, and how to fuse it into one air picture.

Hugo Lefèvre
12 min read

Airspace awareness from open data.

Fuse ADS-B, AIS and OSINT feeds on your own infrastructure. The demo runs entirely on simulated data.