How GPS Works

The United States spent approximately $12 billion building GPS primarily so nuclear missiles could hit targets within a few meters Economy Insights. The technology was eventually opened to civilians in 1983, after a Korean Air flight was shot down after straying into Soviet airspace due to navigation error, killing 269 people. Today, the same system that guides guided munitions also tells you where the nearest coffee shop is. The physics behind it is stranger than the origin story.

The short answer

GPS works by measuring the time it takes for signals from satellites to reach your device. Your phone knows where each satellite is at any given moment and uses the travel time of signals from at least four satellites to calculate exactly where you must be. The precision required is extraordinary: an error of one microsecond in the timing would place you off by 300 meters.

The full picture

What GPS actually is

GPS stands for Global Positioning System — formally known as NAVSTAR GPS, a name coined by the U.S. Department of Defense when the programme launched in the 1970s. It’s a U.S. military system, with roots tracing back to NRL’s (Naval Research Laboratory) TIMATION research programme begun in 1964; NRL’s Navigation Technology Satellite II, launched in 1977, was the first satellite in the NAVSTAR GPS constellation. The system was made available to civilians in the 1980s, though with intentional accuracy limits.

The system has three parts:

The space segment is the constellation of satellites. There are at least 24 operational satellites at all times, currently 31, orbiting at about 20,200 kilometers above Earth GPS.gov. They’re arranged so that at least four satellites are visible from almost any point on the planet at any time. Each satellite broadcasts a continuous signal containing its position and a precise timestamp.

The control segment is a network of ground stations that monitor the satellites, correct their orbits, and update their clocks to keep everything synchronized.

The user segment is your GPS receiver: your phone, your car’s navigation system, a hiking watch. The receiver does the math. It doesn’t transmit anything. GPS is entirely one-way communication. The satellite broadcasts; your device listens.

Trilateration (not triangulation)

Here’s a distinction worth knowing, because the two terms mean different things.

Triangulation uses angles. Surveyors measure angles from known points to calculate where something is.

Trilateration uses distances. GPS uses trilateration. Your receiver measures how far it is from multiple known points (the satellites), then calculates where you must be based on those distances.

The intuition: imagine you’re somewhere in a city and someone tells you you’re exactly 3 miles from the Eiffel Tower. That’s not very helpful. You could be anywhere on a circle 3 miles around the tower. But if they also tell you you’re exactly 2 miles from Notre-Dame Cathedral, now you’re at one of two points where those circles intersect. Add a third location, say 1.5 miles from the Louvre, and there’s only one spot where all three circles overlap. That’s you.

GPS does this in three dimensions, which is why you get latitude, longitude, and altitude. In 3D space, you need at least three satellites to get a position fix.

Why you need a fourth satellite

GPS needs a fourth satellite to correct your phone’s clock. Your receiver’s clock is too inaccurate for GPS’s microsecond precision, so the four satellite signals create four equations that let the receiver solve for latitude, longitude, altitude, and its own clock error simultaneously.

Three satellites should be enough geometrically. So why does GPS need four?

Timing error. Your phone’s clock is not nearly accurate enough. A cheap quartz clock drifts by milliseconds. But GPS relies on microsecond precision, errors smaller than one millionth of a second. The satellites have atomic clocks accurate to about 20-30 nanoseconds. Your phone does not.

If your receiver’s clock is even slightly off, all the distance calculations are wrong by the same consistent error. A fourth satellite gives the receiver enough information to solve for its own clock error mathematically. The four measurements create four equations with four unknowns: latitude, longitude, altitude, and clock error. The receiver solves all four simultaneously.

More satellites (modern receivers often use 8-12) make this calculation more accurate and more robust.

Atomic clocks and why nanoseconds matter

GPS signals travel at the speed of light: about 300,000 kilometers per second. Tiny time errors produce large position errors.

• One microsecond of timing error = 300 meters of position error
• One nanosecond of timing error = 30 centimeters of position error

Each GPS satellite carries between one and four atomic clocks, accurate to about 20-30 nanoseconds. Atomic clocks work by counting the oscillations of atoms (usually cesium or rubidium). Cesium atoms oscillate about 9.19 billion times per second Ohio State University. Counting those oscillations is far more stable than any mechanical clock.

Einstein’s relativity complicates things slightly. Satellites are moving fast (about 14,000 km/h), which makes their clocks tick slightly slower due to special relativity. They’re also farther from Earth’s gravitational field, which makes their clocks tick slightly faster due to general relativity. Engineers correct for both effects continuously. Without those corrections, GPS would drift by about 10 kilometers per day.

Why GPS accuracy varies

Modern GPS can be accurate to a few meters under ideal conditions. Several factors degrade it.

Atmospheric interference: GPS signals pass through the ionosphere and troposphere. These layers slow the signals slightly and unpredictably, adding timing errors. Professional GPS systems use dual-frequency receivers that calculate and correct for this effect. Consumer GPS uses a model-based correction that’s less precise.

Multipath errors: In cities, signals bounce off buildings before reaching your receiver. Your device might receive the same signal twice, once directly and once after bouncing off a glass tower. The reflected signal arrives later, making the satellite appear farther away than it is. This is why GPS is less accurate in dense urban canyons.

Satellite geometry: If all visible satellites are clustered in one part of the sky, your position calculation is less accurate than if they’re spread out. A metric called DOP (Dilution of Precision) quantifies this.

Multiple systems: Your phone doesn’t rely only on GPS. It uses Russia’s GLONASS, Europe’s Galileo, China’s BeiDou, and combines that with Wi-Fi network positioning and cell tower data. The location dot on your map is a blended estimate from all of these.

A note on civilian accuracy: Selective Availability

For most of GPS’s early civilian life, the U.S. government intentionally degraded the signal available to non-military users — a feature called Selective Availability — keeping civilian accuracy at roughly 100 meters. This changed on May 1, 2000, when President Bill Clinton signed a policy directive turning off Selective Availability, making full precision (down to a few meters) available to civilians worldwide GPS.gov. Consumer GPS became dramatically more useful overnight.

What GPS doesn’t tell you: the assisted GPS revolution

Standalone GPS — your device talking directly to satellites — is slow and battery-hungry. In the open sky, getting an initial position fix can take 30-60 seconds while your receiver waits for satellite signals that repeat their data every 30 seconds. In a city, where buildings block portions of the sky, this can take even longer.

Modern smartphones don’t work this way. They use Assisted GPS (A-GPS), where cellular networks download a table of satellite positions directly to your phone. Instead of waiting for satellites to transmit their own position data, your phone already knows where the satellites are and can compute a fix in a few seconds using just timing signals.

This is why your map app shows your location almost instantly indoors, where GPS signals might be too weak to use alone. The phone blends multiple positioning systems: GPS satellites, GLONASS, Galileo, and BeiDou for outdoor precision; Wi-Fi positioning (comparing visible networks against a database of known network locations) for indoor use; and cell tower triangulation as a rough fallback. The blue dot on your map is a confidence-weighted blend of all these systems, updated constantly.

The practical consequence: that “GPS” app on your phone is rarely using only GPS. When you’re inside a shopping mall and the app still knows which floor you’re on, it’s using Wi-Fi positioning — your phone comparing the Wi-Fi networks it can see against a massive database that Google, Apple, and others have built by driving cars with Wi-Fi scanners down every street in the world. The location technology you use daily is far more layered, and stranger, than the satellite story alone suggests.

Why it matters

GPS is infrastructure now, and not just for navigation. Financial systems use GPS timestamps to synchronize transactions across continents. Power grids use it to coordinate. Autonomous vehicles depend on centimeter-level GPS. Mobile networks use GPS timing to coordinate cell handoffs.

Knowing that your location comes from timing and satellites gives you a better intuition for when GPS fails: inside buildings (signals blocked), in deep urban canyons (multipath), during solar storms (atmospheric disruption). The phone’s fallback to Wi-Fi positioning is a real downgrade in precision.

Common misconceptions

“GPS is triangulation.” GPS uses trilateration (distances), not triangulation (angles). The difference matters technically, even if both words get used colloquially.

“GPS knows where I am.” GPS satellites don’t know you exist. Your receiver calculates your position. The system is entirely passive from the satellite side.

“My phone’s GPS is powered by the phone’s battery.” The satellites power their own transmitters using solar panels. Your phone only powers its receiver, which listens passively.