GPS Invention: Satellites, Atomic Clocks, and Relativity

Photo: Asad Photo Maldives / Pexels
GPS was not invented in a single lab or by a single person. It was assembled over three decades from three separate ideas: a Navy satellite-tracking trick from 1957, a set of orbiting atomic clocks built by physicist Roger Easton, and a 1970s Air Force program under Bradford Parkinson that stitched them together into NAVSTAR-GPS. The system only works today because engineers also had to correct for a strange consequence of Einstein's relativity — without that fix, GPS would be useless within minutes. Another invention with a contested origin is covered in the telephone invention and the patent race between Bell, Gray and Meucci.
The problem GPS was built to solve
Before GPS, finding your exact position on Earth required visible landmarks, stars, or radio beacons within range. Ships used sextants and dead reckoning. Aircraft used ground-based radio beacons like VOR and LORAN, both of which had limited range and could be blocked by terrain, weather, or distance from the coast. None of these methods worked reliably in open ocean, over the poles, or underwater for a submerged submarine.
The US military had a specific, urgent version of this problem in the 1950s and 1960s: how do you tell a nuclear submarine or an intercontinental missile crew exactly where they are, anywhere on Earth, at any time, without them broadcasting a signal that reveals their position to an enemy? Existing systems like LORAN-C offered accuracy of roughly a few hundred meters at best. A ballistic missile launched from a submarine needs to know its exact starting position to within a few tens of meters, because even a small position error becomes a massive targeting error over a few thousand kilometers of flight.
That problem sat unsolved through most of the Cold War, worked on in parallel by the Navy, the Air Force, and the Applied Physics Laboratory, each with its own partial system and its own acronym. GPS became possible only when those separate threads were forced together into one program.
Where the idea actually came from
The origin story usually credited to GPS starts with an accident of physics, not a plan. In October 1957, scientists at Johns Hopkins University's Applied Physics Laboratory were listening to the radio signal from Sputnik, the Soviet Union's first satellite, when two physicists named William Guier and George Weiffenbach noticed something useful: the pitch of the signal shifted as the satellite approached and receded, a Doppler effect. By tracking that frequency shift over time, they realized they could calculate the satellite's exact orbit from a single fixed point on the ground.
Their supervisor, Frank McClure, flipped the logic around within months. If you already know a satellite's orbit precisely, he reasoned, you can run the calculation backward: measure the Doppler shift from a receiver of unknown location, and use it to calculate that receiver's position instead. That insight became the TRANSIT system, the US Navy's first satellite navigation network, operational by 1964 and used to help Polaris submarines fix their position before firing.
TRANSIT worked, but it was slow. A ship or submarine using it had to wait for a usable satellite pass, then run a calculation that could take up to fifteen minutes and only worked if the vessel stayed still. That was tolerable for a submarine surfacing briefly to check its position, but useless for a fighter jet or a moving convoy. Solving that speed problem is what led to GPS's second essential ingredient: extremely precise, continuously broadcasting atomic clocks.
Who is credited with inventing GPS?
Three people are most often named as GPS's principal architects: Roger L. Easton of the Naval Research Laboratory, who developed the TIMATION satellite clock system; Ivan Getting, founding president of the Aerospace Corporation, who proposed a satellite-based navigation concept in the early 1960s; and Bradford Parkinson, who led the Air Force's NAVSTAR Global Positioning System Joint Program Office from 1972 to 1978 and merged the competing designs into one working system. Parkinson is often called the "father of GPS" for that integration work, though he has consistently credited Easton and Getting alongside himself as co-inventors (History Cooperative; Wikipedia: Bradford Parkinson).
How atomic clocks turned satellites into a positioning system
GPS positioning is not based on triangulation, the method most people assume it uses. It is based on trilateration combined with extremely precise timing, and the timing part is what makes the whole system possible.
Here is the mechanism in plain terms. Each GPS satellite continuously broadcasts a radio signal that includes the exact time the signal was sent, generated by an onboard atomic clock accurate to within a few billionths of a second. A GPS receiver on the ground — in your phone, your car, or a ship — picks up that signal and compares the broadcast time to its own clock. The difference tells the receiver how long the signal took to travel, and because radio waves move at the speed of light, that travel time converts directly into a distance: roughly 300,000 kilometers per second of travel time.
One satellite gives you a sphere of possible positions, all points that are the calculated distance away from that satellite. A second satellite gives you another sphere; where the two spheres intersect is a circle. A third satellite narrows that circle down to two points, one of which is usually impossible (in orbit, or underground). A fourth satellite is what actually makes the system work in practice: because a cheap receiver's internal clock is not atomic-clock accurate, that fourth measurement lets the receiver solve for its own clock error at the same time it solves for position. That is why GPS receivers need a minimum of four satellites in view, not three, to get a reliable 3D fix (GPS.gov, official US government GPS information site).
The analogy that makes this click: imagine four friends standing at fixed points around a large dark field, each shouting the exact time out loud once a second. If you know exactly how fast sound travels and you can hear all four shouts, you can work out exactly where you are standing in the field just from how long each shout took to reach you, even though nobody ever "aimed" anything at you. GPS satellites are those four friends, radio waves are the shouts, and the atomic clocks are what make the "exact time" part trustworthy enough to bet a position calculation on.
Why does GPS need atomic clocks specifically?
Because the math depends on timing accuracy at the nanosecond level. Light travels about 30 centimeters in one nanosecond, so a clock error of even one microsecond translates into a position error of roughly 300 meters (984 feet). Ordinary quartz clocks drift by far more than that over a day, which is why each satellite instead carries cesium or rubidium atomic clocks, accurate to about one second of error every 30,000 to 1 million years depending on the clock type (National Institute of Standards and Technology).
The relativity problem nobody expected to matter
This is the part of the GPS story most people never hear, and it is the part that makes the invention genuinely remarkable as physics, not just engineering. GPS satellites orbit at about 20,200 kilometers (12,550 miles) altitude, traveling at roughly 14,000 kilometers per hour (8,700 mph) relative to the ground. At that speed and altitude, two separate relativistic effects act on the satellites' atomic clocks, and they pull in opposite directions.
Special relativity says that a clock moving relative to an observer runs slower from that observer's point of view. Because GPS satellites move fast relative to the ground, special relativity alone would make their clocks run about 7 microseconds slower per day than identical clocks on Earth.
General relativity says the opposite in this case, and its effect is larger. Time runs faster the farther you get from a large mass's gravitational field, because gravity bends spacetime itself. GPS satellites sit in much weaker gravity than the surface of the Earth, so general relativity alone would make their clocks run about 45 microseconds faster per day than ground clocks.
Combine both effects and the net result is that an uncorrected GPS satellite clock gains roughly 38 microseconds per day compared to a clock on the ground (NIST). That sounds trivial. It is not. Thirty-eight microseconds of timing error, run through the speed-of-light distance calculation described above, works out to a position error that grows by roughly 10 kilometers (6 miles) every single day the correction is missing. A navigation system that drifts 10 kilometers a day is not a navigation system.
GPS engineers solved this before the first operational satellites launched, not by patching it in software after the fact but by building the correction into the hardware itself. Each satellite's atomic clock is deliberately manufactured and calibrated to tick slightly slower than a correct clock would on the ground — offset by about 4.465 parts in 10 billion — so that once relativity speeds the clock up in orbit, it settles into ticking at exactly the right rate as seen from Earth. It is one of the very few pieces of consumer technology in daily use whose correct operation depends on relativistic corrections that Einstein's equations predicted decades before anyone could test them with this kind of precision.
From military system to program everyone can use
The first NAVSTAR-GPS satellite, Navstar 1, launched on February 22, 1978. Through the late 1970s and 1980s, the Department of Defense slowly built out the constellation, one Air Force launch at a time, with the full system intended purely for military targeting, submarine navigation, and troop positioning.
Civilian access happened because of a disaster, not a plan. On September 1, 1983, Soviet fighter jets shot down Korean Air Lines Flight 007 after it drifted off course into restricted Soviet airspace, killing all 269 people aboard. Investigators later attributed the navigational error to a cockpit autopilot mode mistake. Sixteen days later, President Ronald Reagan issued a directive stating that GPS would be made available to civilian aircraft and ships worldwide, free of charge, once the system became fully operational, specifically to prevent similar navigational tragedies (Vice, "How an Airline Tragedy Brought GPS to the Masses").
That promise came with a catch that lasted almost two more decades. The Department of Defense deliberately degraded the civilian GPS signal through a feature called Selective Availability, which introduced random timing errors into the public signal so that civilian receivers landed within roughly 100 meters (330 feet) of true position, while military receivers using an encrypted signal got much better accuracy. GPS reached full operational capability, 24 operational satellites, in 1995, still with Selective Availability switched on.
That changed on May 2, 2000, when President Bill Clinton ordered Selective Availability turned off across the entire satellite constellation simultaneously, just after midnight Eastern time. Civilian GPS accuracy improved overnight from roughly 100 meters down to about 10 meters or better (GPS.gov, official US government resource on Selective Availability). That single policy decision is arguably what made GPS a consumer technology rather than a professional or military tool. Car navigation systems, hiking GPS units, and eventually smartphone location services all became commercially viable within a few years of that switch flipping off.
How GPS compares to other global navigation systems
GPS was the first satellite navigation system to reach global, continuous coverage, but it is no longer the only one. Three other systems now provide comparable service, and modern smartphones typically use signals from all of them at once to improve accuracy and speed up the initial position fix.
| System | Operator | Satellites (nominal) | First launch | Civilian accuracy (typical) |
|---|---|---|---|---|
| GPS | United States (US Space Force) | 24, expandable | 1978 | 3–5 m with modern receivers |
| GLONASS | Russia | 24 | 1982 | 4–7 m |
| Galileo | European Union | 24 | 2011 | 1–4 m |
| BeiDou | China | 35 | 2000 (regional), 2020 (global) | 2–5 m |
Table: Major global satellite navigation systems, satellite counts, and typical civilian-grade positioning accuracy. Figures are nominal constellation sizes and representative accuracy ranges reported by each system's operating agency.
These systems all use the same core physics — trilateration from atomic-clock-timed signals, with relativistic corrections built into each satellite's clock — but they differ in orbital altitude, signal frequencies, and how many satellites each government has chosen to fund. A receiver capable of reading multiple constellations at once, which most modern smartphones are, can typically achieve a faster and more reliable fix than a GPS-only receiver because it has more satellites to choose from at any given moment, especially in cities where tall buildings block a clear view of the sky.
What people get wrong about GPS
The most common misconception is that GPS uses triangulation, measuring angles between the receiver and known points, the way old-fashioned surveying works. It does not. GPS uses trilateration, measuring distances (derived from signal travel time) from multiple known points and finding where those distance measurements overlap. No angles are measured or needed.
A second common myth is that your phone's GPS chip "talks to" the satellites in the sense of sending a signal up to space. It does not, and this is worth understanding because it explains GPS's biggest practical limitation. GPS satellites only broadcast; your receiver only listens. That is why GPS works even in places with no cellular signal at all, as long as the receiver has a clear view of enough sky, and it is also why GPS cannot easily "find you" if you are lost and your phone has no separate way to transmit your calculated position out to someone else.
A third misconception is that GPS accuracy is a fixed, universal number. In practice, accuracy depends heavily on how many satellites are visible, their geometric spread across the sky, atmospheric interference, and whether the receiver has line of sight or is working around reflections off buildings ("multipath" error). That is why the same phone can report a location accurate to 3 meters while standing in an open field and 30 meters while standing between two skyscrapers.
A fourth misconception, ironically the opposite kind of error, is assuming GPS is "basically free" hardware that needs no active management. The satellites themselves have limited operational lifespans, typically 10 to 15 years, and the US Space Force continuously launches replacement satellites (the GPS III series began launching in 2018) to keep the constellation's atomic clocks fresh and the signal strong. GPS is a maintained, actively operated system, not a set-and-forget piece of 1970s infrastructure.
How accurate is GPS today?
Standard civilian GPS receivers, including modern smartphones, typically achieve horizontal accuracy of about 3 to 5 meters (10 to 16 feet) under open sky conditions, according to the US government's official GPS performance standards. Specialized differential and survey-grade GPS equipment, which corrects for atmospheric and satellite-position errors using a nearby reference station, can achieve accuracy down to a few centimeters (GPS.gov, GPS Accuracy).
Does GPS still use Selective Availability today?
No. Selective Availability was permanently discontinued on May 2, 2000, and the United States government has stated it has no plans to reintroduce deliberate signal degradation for civilian users. Modern GPS satellites are also designed without the hardware needed to reinstate the old form of Selective Availability (GPS.gov).
Why do GPS satellites need atomic clocks instead of regular clocks?
Because GPS position accuracy depends on measuring signal travel time to within nanoseconds. Since light and radio waves travel about 30 centimeters per nanosecond, even a microsecond of clock drift produces roughly 300 meters of position error. Regular quartz clocks drift far too much for this; cesium and rubidium atomic clocks lose or gain only a few nanoseconds per day.
How many satellites does GPS need to determine a location?
A minimum of four satellites in view are needed for a full three-dimensional position fix (latitude, longitude, and altitude). Three satellites can narrow a position to two possible points, but the fourth is required to correct for the receiver's own imperfect clock, which is not atomic-clock accurate.
What would happen to GPS without relativistic corrections?
Position errors would accumulate at roughly 10 kilometers (6 miles) per day, because uncorrected atomic clocks aboard the satellites would drift about 38 microseconds ahead of ground time every day, a combined effect of special and general relativity. GPS would become unusable for navigation within hours to a couple of days.
Who first proposed satellite-based navigation?
Ivan Getting, founding president of the Aerospace Corporation, proposed a satellite-based system for continuous, all-weather navigation in the early 1960s. The concept combined ideas already emerging from the Navy's TRANSIT and TIMATION programs, developed separately by researchers including Roger Easton at the Naval Research Laboratory.
Is GPS the only global satellite navigation system?
No. Russia's GLONASS, the European Union's Galileo, and China's BeiDou all provide global coverage today, and most modern smartphones and vehicle navigation systems use signals from multiple systems simultaneously to improve accuracy and reliability.
When did GPS become free for civilian use?
President Reagan directed in September 1983 that GPS would eventually be made available to civilian users worldwide at no charge, following the shootdown of Korean Air Lines Flight 007. The system reached full civilian usability with comparable accuracy to military GPS on May 2, 2000, when Selective Availability was switched off.
What is the difference between GPS and GNSS?
GPS is the specific US-operated satellite navigation system. GNSS, or Global Navigation Satellite System, is the umbrella term for any satellite positioning system, including GPS, GLONASS, Galileo, and BeiDou. GPS is technically one GNSS among several, though the term "GPS" is often used loosely to mean satellite navigation in general.
The takeaway
GPS is a case study in how transformative technology actually gets built: not from a single insight, but from stitching together an accidental discovery about satellite radio signals, a military need for precise weapons targeting, a set of atomic clocks accurate enough to matter, and a physics correction that only became necessary because the system got precise enough for Einstein's relativity to show up as a real-world error. The 1983 decision to open GPS to civilians, forced by a tragedy rather than a business plan, is the reason it ended up in every phone rather than staying a classified military tool. What looks today like a simple blue dot on a map is the end product of Cold War satellite tracking, nanosecond-precision atomic engineering, and a relativistic correction built into hardware three decades before most people had ever heard of Einstein's name attached to anything practical. Compare the story of television's invention by Baird, Farnsworth and others.
Read next:
- /technology/history-of-computing-from-abacus-to-ai
- /technology/how-the-transistor-changed-everything
- /inventions/the-internet
Sources
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

Laser Invention: From Masers to Coherent Light
The laser grew from quantum theory, masers, optical engineering, and patent battles. This article explains how coherent light became a world-changing tool.

Camera Invention: From Camera Obscura to Digital Sensors
The camera evolved from the camera obscura to chemical photography and digital sensors. This guide explains the chain of inventions.

How GPS Uses Einstein's Relativity Every Single Day
Your phone's navigation relies on corrections from both special and general relativity. Without Einstein, GPS would be off by kilometers within hours.

How GPS Satellites Know Your Location
GPS uses clocks accurate to a billionth of a second and Einstein's relativity to pinpoint you anywhere on Earth within meters.

Microchip Invention: Kilby, Noyce, and the Integrated Circuit
The microchip emerged from parallel breakthroughs by Jack Kilby and Robert Noyce. This guide explains the invention that made modern computing possible.

Telephone Invention: Bell, Gray, Meucci, and the Patent Race
The telephone's invention involved Alexander Graham Bell, Elisha Gray, Antonio Meucci, patents, and the race to transmit speech electrically.