Why MLAT Tracking Is Less Accurate Than ADS-B

What MLAT and ADS-B Actually Do

Aircraft tracking has gotten complicated with all the competing data sources and abbreviations flying around. As someone who has spent an unreasonable number of dinner parties staring at FlightAware on my phone while everyone else talked like normal people, I learned everything there is to know about why position dots jump sideways for no obvious reason. Today, I will share it all with you.

But what is ADS-B? In essence, it’s the aircraft telling you where it is. But it’s much more than that. The onboard GPS calculates a precise position, and the transponder broadcasts that position as a digital packet roughly twice per second. ADS-B stands for Automatic Dependent Surveillance-Broadcast. Anyone running a receiver can pick it up — no interpretation, no back-calculation, no guesswork involved.

MLAT is the opposite approach. Multilateration means you’re figuring out where the aircraft is by listening to echoes. The plane squawks a Mode C or Mode S transponder signal. A network of ground stations catches that same signal at slightly different moments. Those tiny arrival-time gaps get fed into math that back-calculates the transmission’s origin. Think of it like pinpointing a sound by timing when it reaches each ear. Clever. Genuinely useful. And — as we’ll get into — inherently fuzzier than just asking the plane where it is.

Why MLAT Position Data Is Inherently Fuzzier

Probably should have opened with this section, honestly. Because the whole MLAT accuracy problem comes down to one thing: time.

The system lives or dies on something called Time Difference of Arrival — TDOA. If a signal hits Receiver A four microseconds before Receiver B, that gap becomes a distance differential. Stack that against data from Receivers C and D, and you’ve got a position estimate. You need at least three receivers for 2D positioning. Four for reliable 3D positioning that actually includes altitude. And every single receiver needs its clock synced with the others to within nanoseconds.

That’s where the wheels come off. Real-world clocks drift. Network latency introduces jitter. A timing error of one microsecond alone translates to roughly 300 meters of positional error — radio signals travel at the speed of light, so even tiny miscalculations cascade hard through the math. A well-functioning MLAT network in a dense receiver area lands somewhere between 100 and 300 meters of accuracy. Bad conditions — sparse receivers, weak geometry, clock drift — and you’re nudging past 500 meters without much warning.

ADS-B pulls position directly from the aircraft’s onboard GPS. Modern WAAS-enabled aviation GPS — the kind inside a Garmin GTX 345 or an L3 Lynx NGT-9000 — delivers accuracy under 10 meters in typical conditions. The aircraft isn’t estimating. It knows exactly where it is and broadcasts that directly. That’s not a marginal accuracy difference. It’s an order of magnitude. That’s what makes ADS-B so endearing to us tracking enthusiasts.

Where Each System Breaks Down

Neither system is bulletproof. I learned this the embarrassing way — spending the better part of an evening convinced a regional turboprop was flying a genuinely bizarre zigzag pattern over my neighborhood. It wasn’t. I was watching MLAT position errors near the edge of a thin receiver network. Don’t make my mistake.

MLAT failure modes cluster around geometry and infrastructure. Sparse receiver networks are the most common culprit. When only three stations are hearing a particular aircraft, the positional geometry is weak — and any clock drift in one of those three stations punches above its weight on the final result. Rural areas suffer badly here. Low altitudes too, where terrain blocks line-of-sight to distant stations. An aircraft cruising at 1,500 feet AGL over rolling hills might only be visible to two nearby receivers. Two isn’t enough for reliable MLAT at all.

Urban environments introduce a different headache: multipath interference. Dense cities mean transponder signals can bounce off buildings before reaching a receiver — arriving via an indirect route that’s slightly longer than the straight-line path. The receiver has no way to distinguish a reflection from a direct signal. That extra travel time gets fed into the TDOA calculation as genuine data, and the resulting position estimate skews in ways that are genuinely hard to predict or correct for.

ADS-B failure modes are fewer but can be dramatic. GPS spoofing — broadcasting fake satellite signals to fool an onboard receiver — can convince an aircraft’s GPS that it’s somewhere completely different, with high confidence. The system doesn’t know it’s being deceived. This has been documented near conflict zones and politically sensitive airspace, where tracking sites have shown commercial aircraft apparently teleporting to airport coordinates hundreds of miles from their actual position.

Altitude accuracy in ADS-B deserves its own callout, separate from horizontal position. Geometric altitude from GPS is quite solid. But many ADS-B implementations also broadcast barometric altitude — which depends on proper altimeter calibration and correct local pressure settings. A poorly maintained or misconfigured transponder installation can broadcast altitude that’s off by several hundred feet even when horizontal position is perfectly accurate.

Then there’s intentional blocking. Some operators — private owners, certain government operations — disable or modify transponders to limit tracking visibility. When ADS-B goes dark, tracking sites fall back to MLAT if coverage exists. If it doesn’t, the aircraft simply disappears from the map entirely.

How Flight Trackers Blend Both Signals

So, without further ado, let’s dive into what sites like Flightradar24 and FlightAware are actually doing with all of this. They’re not showing you one clean data type. They’re aggregating feeds from thousands of volunteer-operated receivers worldwide — mixing ADS-B and MLAT into a single continuous track — and labeling the source, if you know where to look.

On Flightradar24, source labels appear in the flight detail panel. “ADS-B” means the aircraft is reporting its own GPS position. “MLAT” means ground stations are triangulating it. Mid-flight switching between the two is completely normal. A departure out of a well-covered urban airport might show ADS-B through cruise, drop into MLAT as it passes through a gap in receiver density somewhere over a rural stretch, then pick ADS-B back up on final approach.

That handoff is exactly what causes the dot to jump on the map. Not a glitch. Not the aircraft teleporting. Just a transition between a self-reported GPS position and a ground-calculated approximation — two different answers to the same question, separated by the full width of their respective error margins.

Which One Should You Trust and When

I’m apparently a “check the source label before drawing conclusions” type of person, and that habit works for me — while just trusting whatever dot appears on screen never quite worked out.

For casual tracking — watching a family member’s flight land, monitoring traffic near a home airport, figuring out what that loud jet was overhead — both ADS-B and MLAT are directionally accurate enough. The aircraft is roughly where the dot says it is, give or take a few hundred meters.

For anything requiring precision, source matters enormously. ADS-B might be the best option, as serious flight path analysis requires position accuracy that MLAT simply can’t guarantee. That is because a 300-meter error sounds minor until you’re trying to determine whether an aircraft actually crossed a restricted zone boundary — at which point it’s the only thing that matters. Filter for ADS-B sources. Treat MLAT as approximate.

First, you should check the source label before interpreting unusual flight path behavior — at least if you want to avoid spending an evening convinced a turboprop has lost its mind. Pilots noticing discrepancies between cockpit displays and what tracking sites show should do the same. Odds are the tracking site is working from MLAT while the aircraft’s own GPS knows the position to within 10 meters.

The deeper you dig into this infrastructure, the more two things happen simultaneously: you get genuinely impressed by how well it works, and you stop being surprised by jumping dots. Two different methods solving the same problem. Both doing their best with fundamentally different tools.