How GPS Works and the Satellites That Run It

01 The Military Origins

The Global Positioning System began as a Cold War project. In the early 1960s, the U.S. Navy deployed Transit, the first satellite navigation system, which used Doppler shift measurements from low-earth-orbit satellites to determine a submarine's position. Transit was slow — a fix could take ten to fifteen minutes and was only accurate to about 100 meters — but it proved the concept. The Air Force, meanwhile, was developing its own system called 621B, which used pseudorandom noise codes and ranged from satellites in higher orbits.

In 1973, the Department of Defense consolidated these efforts into a single program: the Navigation Technology Satellite program, soon renamed Navstar GPS. The first Block I GPS satellite launched on February 22, 1978, from Vandenberg Air Force Base atop an Atlas F rocket. The system was designed from the outset as a dual-use utility: precise encrypted signals for the military, and a deliberately degraded civilian signal called Selective Availability, which limited non-military accuracy to roughly 100 meters. The military saw navigation as a force multiplier — bombs that could find their targets, troops that always knew their position, and a common timing reference for encrypted communications.

Selective Availability was turned off by presidential order on May 1, 2000. Overnight, civilian GPS accuracy improved from ~100 meters to ~10 meters. This single decision opened the door to consumer GPS navigation, location-based services, and the entire mapping economy that now runs on every smartphone. The U.S. Space Force, which inherited operational responsibility for GPS in 2019, continues to maintain and upgrade the constellation, ensuring that the system remains freely accessible to anyone with a receiver.

02 The Constellation in Orbit

The operational GPS constellation consists of 31 active satellites deployed across six orbital planes, each inclined at 55 degrees to the equator. The satellites orbit at a semi-major axis of approximately 26,560 kilometers — about 20,200 kilometers above Earth's surface — completing two orbits for every one rotation of the Earth. This 2:1 orbital resonance means each satellite passes over the same ground track every sidereal day, repeating its coverage pattern. The arrangement ensures that at least four satellites are visible from virtually any point on Earth at any time, which is the minimum needed for a three-dimensional position fix.

The satellites themselves have grown substantially more capable across five generations, known as Blocks. The original Block I satellites, launched from 1978 to 1985, were experimental and had design lifetimes of only five years. Block II and IIA, deployed from 1989 to 1997, brought the system to full operational capability with radiation-hardened electronics and the ability to detect nuclear detonations as a secondary mission. Block IIR and IIR-M satellites, launched from 1997 to 2004, added the modernized military M-code signal and the L2C civilian signal. Block IIF satellites, launched from 2010 to 2016, introduced the L5 signal designed for safety-of-life applications like aviation. The newest Block III satellites, first launched in 2018, bring the most powerful signals yet, including a new L1C signal compatible with Europe's Galileo system.

03 Trilateration: The Geometry of Position

The core principle behind GPS is trilateration — not triangulation, despite the common confusion. Trilateration works by measuring distances, not angles. Each satellite broadcasts a signal containing its precise orbital position (ephemeris data) and a timing pulse. When your receiver picks up that signal, it compares the arrival time with the transmission time encoded in the signal. Because the signal travels at the speed of light — approximately 300,000 kilometers per second — the time difference directly yields the distance to that satellite. A one-millisecond delay equals roughly 300 kilometers of range.

One satellite tells you that you are somewhere on the surface of a sphere centered on that satellite, with a radius equal to the measured distance. Two satellites narrow it to the intersection of two spheres — a circle. Three satellites narrow it further to two points, and one of those points can usually be discarded because it is absurdly far from Earth's surface. A fourth satellite is needed not for geometry alone, but to solve for the one variable the receiver cannot know on its own: the precise time. Your phone's quartz clock is far too imprecise compared to the satellite's atomic clocks, so the receiver treats its own clock bias as a fourth unknown and uses the fourth satellite's signal to solve for it simultaneously. Four satellites, four unknowns (x, y, z, and clock bias), one elegant system of equations.

This is why GPS is fundamentally a time-delivery system. Every position fix is also a time transfer: the receiver locks onto atomic-clock-grade timing as a byproduct of computing its location. This side effect underpins financial transaction timestamping, cellular network synchronization, and power grid phase coordination worldwide.

04 Einstein's Relativity at Work

GPS is the first engineering system that must account for both special and general relativity to function correctly. The satellites orbit at 20,200 kilometers, where Earth's gravitational field is weaker than on the surface. According to general relativity, clocks in weaker gravitational fields run faster. The effect is small but not negligible: the satellites' clocks gain about 45.8 microseconds per day compared to ground-based clocks. Meanwhile, special relativity predicts that the satellites' orbital speed — about 3.87 kilometers per second — causes their clocks to lose about 7.2 microseconds per day. The net relativistic drift is approximately +38.6 microseconds per day.

Thirty-eight microseconds sounds trivial, but at the speed of light it corresponds to about 11.6 kilometers of ranging error per day. Without correction, GPS positions would drift by roughly 10 kilometers every day, rendering the system useless within minutes. The satellite clocks are pre-adjusted before launch — their oscillation frequency is set slightly slower than ground atomic clocks, by 10.23 MHz minus 0.00455 Hz, so that once in orbit they tick at the correct rate relative to receivers on the ground. Additional relativistic corrections, including effects from the satellites' orbital eccentricity, are applied in the receiver's software.

05 Signals, Frequencies, and the L-Band

GPS satellites transmit on multiple frequencies in the L-band, a portion of the radio spectrum between 1 and 2 GHz chosen for its ability to penetrate clouds, rain, and foliage while requiring a reasonably compact antenna. The original signals are L1 at 1575.42 MHz and L2 at 1227.60 MHz. L1 carries the Coarse/Acquisition (C/A) code — the civilian signal — and the encrypted Precision (P(Y)) code used by the military. L2 was originally military-only, but modernized satellites have added civilian signals on L2 (L2C) and a third frequency, L5 at 1176.45 MHz, specifically designed for safety-critical applications like aircraft navigation.

The genius of the signal design lies in code division multiple access (CDMA). Each satellite transmits on the same frequency but uses a unique pseudorandom noise (PRN) code — a gold code sequence — that is orthogonal to all other satellites' codes. The receiver correlates the incoming signal against each known PRN code; only the matching satellite's signal rises above the noise floor. This allows all satellites to share the same bandwidth without interfering with each other, a scheme conceived in the 1970s that remains elegant today.

Multiple frequencies matter because the ionosphere delays GPS signals by an amount that depends on frequency. By measuring the range on two frequencies, the receiver can calculate and subtract the ionospheric delay, improving accuracy from meters to centimeters. Dual-frequency receivers, once restricted to surveying equipment, are now standard in modern smartphones thanks to the L1 + L5 capability built into recent chipsets from Apple, Qualcomm, and Broadcom.

06 Augmentation, Jamming, and Vulnerability

Despite its remarkable accuracy, standalone GPS has limitations that have driven the development of augmentation systems. The Wide Area Augmentation System (WAAS), operated by the U.S. Federal Aviation Administration, uses ground reference stations to measure GPS errors and broadcast corrections via geostationary satellites, enabling precision approaches at airports. The European Geostationary Navigation Overlay Service (EGNOS) and Japan's MTSAT Satellite Augmentation System provide analogous services. Real-Time Kinematic (RTK) and Precise Point Positioning (PPP) techniques push accuracy to the centimeter level for surveying, agriculture, and autonomous vehicles by using carrier-phase measurements and correction streams from networks of reference stations.

GPS signals are weak. By the time they reach Earth, the signal strength is around −160 dBW — well below the thermal noise floor. This makes GPS remarkably susceptible to jamming, which has become a growing problem. Low-cost jammer devices, sold online and often used to defeat vehicle tracking, can disrupt GPS reception over a radius of several kilometers. More alarmingly, deliberate military jamming and spoofing — broadcasting fake GPS signals to mislead receivers — has been documented near conflict zones. In 2024, widespread GPS interference was reported across northern Europe, attributed to jamming from Russian military systems in the Baltic region. The Veritasium video in this article's references documents a firsthand investigation into this phenomenon.

The fragility of GPS has motivated interest in complementary systems. The U.S. eLORAN system, a modernized version of the old LORAN-C ground-based navigation network, provides an independent timing and positioning signal that is far harder to jam. Europe's Galileo, Russia's GLONASS, and China's BeiDou provide alternative GNSS constellations that modern receivers can use simultaneously, improving both accuracy and resilience through multi-constellation fusion.

07 The Invisible Backbone of Modern Civilization

GPS is the most widely used military system ever made available to civilians. Beyond navigation, its timing function has become a critical infrastructure dependency. Cellular networks use GPS-derived timing to synchronize base stations — without it, handoffs between towers would fail within hours. Financial trading systems timestamp transactions to the microsecond using GPS-disciplined clocks, and regulations like the SEC's Consolidated Audit Trail require sub-millisecond timing accuracy that only GNSS can provide. Power grids use GPS timing to synchronize phase measurement units across the transmission network, enabling real-time stability monitoring and fault detection.

The economic value of GPS has been estimated at over $1.4 trillion in the United States alone, a figure that continues to grow as autonomous systems, precision agriculture, drone delivery, and logistics tracking expand. The system's founders could not have anticipated that a military navigation project would become the invisible timing backbone of the global economy. The decision to make GPS freely available — and the subsequent decision to remove deliberate degradation — was one of the most consequential infrastructure policy choices of the late twentieth century, creating more economic value than nearly any government technology program in history.