A burst of radio interference, caught on an autumn evening in suburban Washington, became, in seven years, a system that let submarines know their place in the ocean to within a city block. This is the story of how the curiosity of two physicists with nothing to do with satellites gave rise to an entire industry.
October 4, 1957 — the world learned that something man-made was flying overhead. Sputnik 1 transmitted its signal at 20.005 MHz — monotonous, rhythmic, utterly useless for transmitting data. Just “beep-beep-beep.” Ham radio operators picked it up everywhere; newspapers printed instructions for tuning receivers; the Pentagon was nervous. And at the Johns Hopkins University Applied Physics Laboratory (APL) in Silver Spring, Maryland, three days later, two physicists — William Guier and George Weiffenbach — did what physicists do when they hear an interesting signal: they began measuring it.
Guier and Weiffenbach did not work on satellites. They were engaged in entirely different projects. But they had access to a good receiver, and Weiffenbach had a sensitive frequency counter. And they noticed what every ham operator had noticed, but few had thought to use: Sputnik’s signal frequency changed as it passed overhead. As the satellite approached, the frequency was slightly higher than the transmitted one; as it receded, slightly lower. The Doppler effect, familiar to anyone who has heard the pitch of an ambulance siren change as it goes by. Except an ambulance moves at sixty kilometers an hour, while a satellite moves at about eight kilometers a second.
They began recording the Doppler curve — a graph of frequency change over time. The shape of that curve contained information: how close the satellite had passed, how fast it was moving, at what angle above the horizon. In essence, each pass left a unique frequency signature on the chart recorder’s paper, and from that signature it was possible to reconstruct the orbital parameters. Guier and Weiffenbach did exactly that — calculated Sputnik 1’s orbit from its Doppler trail. For a laboratory that had worked for the Navy for decades, it was a beautiful but seemingly academic exercise: others were calculating Sputnik’s orbit too, by other methods. The real turn came several months later, made by a man who had not been sitting at the receiver himself.
Frank McClure headed a research center at APL and followed Guier and Weiffenbach’s work closely. In March 1958, he asked a question that seems obvious in retrospect, but was not at the time: what if you turned the problem around? If a Doppler curve could reveal a satellite’s orbit when you knew the receiver’s position, then the same curve could reveal the receiver’s position when you knew the satellite’s orbit.
The logic is flawless, and yet it required an intellectual effort — that moment when a familiar scheme has to be turned one hundred and eighty degrees in your mind. Guier and Weiffenbach were solving the direct problem: a fixed point on Earth, an unknown orbit. McClure proposed the inverse one: a known orbit, an unknown point on Earth. Mathematically, both problems rested on the same Doppler curve, but their practical implications differed radically. The direct problem gave you science — more precise orbital calculations. The inverse one gave you navigation — a position fix.
What gets me about this story is that McClure was not a radio engineer and did not spend his nights at a receiver. He was a manager who could see a problem from the other side. And he had a specific motivation: the U.S. Navy desperately needed a way to determine the coordinates of submarines carrying Polaris ballistic missiles. The missiles were guided by inertial systems, and the accuracy of their fire depended directly on how precisely the boat’s launch position was known. Surfacing for astronomical observations was slow and dangerous. Radio-navigation systems such as LORAN did not cover every ocean and were vulnerable to interference. A satellite passing overhead and broadcasting its orbit solved the problem elegantly: the receiver could be raised to periscope depth, record the signal for a few minutes, and dive again.
APL submitted a proposal, and in 1958 development began on a system called Transit. Funding came through ARPA — the Advanced Research Projects Agency of the Department of Defense, created that same year, also in response to the shock of Sputnik. The program was led by Richard Kershner, an APL engineer. Eighteen months passed between McClure’s idea and the first launch — a pace that seems impossible today.
Transit’s principle looks deceptively simple, but behind that simplicity lies a whole cascade of engineering decisions. The satellite flew in a polar orbit at an altitude of roughly 1100 kilometers (about 600 nautical miles) and continuously transmitted two kinds of information: its precise ephemerides — data on its current position and orbital parameters — and a stable frequency signal. Transmission took place on two frequencies: 150 MHz and 400 MHz, and the second frequency solved a specific physical problem — accounting for ionospheric refraction. The ionosphere slows radio waves, differently at different frequencies; by comparing the delays, the receiver could calculate and subtract the ionospheric error.
The user on the ground — or at sea — switched on the receiver and waited for a satellite to appear above the horizon. During the pass — which took roughly ten to sixteen minutes, depending on the geometry — the receiver recorded the changing signal frequency. Then the onboard computer (on submarines, the specialized AN/UYK-1 computer, also known as the TRW-130) processed the Doppler curve, matched it against the satellite’s ephemerides, and calculated the receiver’s coordinates.
The key word is “calculated,” not “displayed in real time.” Transit did not provide instant navigation. The solution appeared after the pass had ended and the data had been processed. Typical accuracy for civilian users was around 200 meters. The military, with access to encrypted high-precision ephemerides, could count on 25 meters or better. For a submarine that needed to know its position in order to launch a ballistic missile, twenty-five meters meant the difference between a hit and a miss over the warhead’s range of thousands of kilometers.
Satellite ephemerides did not last forever: an orbit is disturbed by Earth’s gravitational anomalies, solar radiation pressure, and drag from the residual atmosphere. So ground tracking stations constantly refined the orbits and uploaded fresh ephemerides to the satellites. The whole system resembled a clockwork mechanism whose mainspring had to be wound regularly — otherwise, accuracy degraded within hours.
The path from blueprints to orbit proved thorny, like almost everything in early spaceflight. The first craft, Transit 1A, was launched on September 17, 1959 — and failed to reach orbit. The launch vehicle let it down. The failure was painful, but not fatal: the program had strong Navy backing, and the second craft was being prepared in parallel.
On April 13, 1960, Transit 1B successfully reached orbit and became the first navigation satellite in history. It confirmed the crucial point: the method worked. The Doppler curve recorded by a receiver on the ground really did make it possible to calculate coordinates to the stated accuracy. A little over two years had passed between McClure’s idea and its experimental confirmation — even though, in parallel, dozens of side problems had to be solved: creating a sufficiently stable onboard frequency generator, developing a format for transmitting ephemerides, building a network of ground tracking stations.
The next four years went into turning the experiment into an operational system. Enough satellites had to be put into orbit to ensure regular passes over any point in the world’s oceans; ephemeris-update procedures had to be refined; receivers had to be integrated into the standard equipment of ships and submarines. In 1964, the U.S. Navy officially placed Transit in service. By then, a constellation of satellites in polar orbit was operating overhead, and receivers had been installed aboard George Washington- and Ethan Allen-class submarines carrying Polaris missiles.
For submariners, Transit was a quiet revolution. Before it, a boat that had spent several days underwater could accumulate an inertial-navigation error of several miles. Now it was enough to raise an antenna to periscope depth, wait for a satellite pass, and get an accurate fix in fifteen minutes. The boat remained virtually invisible — unlike a full surfacing for astronomical observations, periscope depth did not expose it to enemy radar.
Transit was conceived as a military system, but its civilian use proved unexpectedly extensive. In 1967, Vice President Hubert Humphrey announced that the system was open to merchant ships and aircraft of all friendly nations, and it became a working tool for surveyors around the world. Transit made it possible to tie points on different continents into a single coordinate system with an accuracy beyond the reach of classical triangulation. The geodetic networks of Europe, Africa, South America — all of them were “stitched together” with Transit satellite observations. That was how the first truly global geodetic framework emerged.
Fishing fleets, merchant vessels, research ships — all gained access to navigation that had previously been the military’s privilege. The receivers were bulky and expensive by consumer-electronics standards, but by maritime-equipment standards, entirely manageable.
But Transit had fundamental limitations, and they were no secret. The first and most important: there were few satellites. The operational constellation numbered only five or six craft plus spares, and between passes over a particular point, one and a half to several hours could go by. Navigation was discrete — you received coordinates in “portions,” one fix per pass. Between passes, you had to rely on dead reckoning.
The second limitation: for an accurate solution, the receiver had to remain stationary during the pass, or move at a precisely known speed and heading. Any unaccounted-for component of the receiver’s velocity introduced an error into the Doppler curve and, accordingly, into the calculated coordinates. For a submarine holding at periscope depth with minimal headway, that was tolerable; for a fast-moving target, accuracy fell apart quickly.
Third: the system provided two-dimensional coordinates — latitude and longitude. Altitude had to be known in advance. At sea, that was no problem — you were at sea level. But for land users, and especially aviation users, it was a serious limitation. Every one of these shortcomings pointed in the same direction: what was needed was a system with more satellites, visible simultaneously, providing an instant three-dimensional fix of position and velocity.
It would be convenient to draw a straight arrow from Transit to GPS and say: one grew out of the other. Reality is more complicated. GPS — a next-generation system, placed in service in 1995 — did inherit from Transit the very idea of satellite navigation and the experience of operating it. But GPS architecture is fundamentally different: it is based not on the Doppler method, but on measuring the arrival time of signals from several satellites at once. That required a constellation of 24 spacecraft in medium orbits, each with atomic clocks onboard, and entirely different mathematics.
GPS was shaped by at least three predecessor programs. Transit supplied the general concept and proved that satellite navigation worked. The Timation program, developed by the Naval Research Laboratory (NRL) from the mid-1960s onward, demonstrated the key idea — using highly accurate onboard clocks, first quartz and then atomic, to navigate by signal arrival time. The U.S. Air Force’s 621B program developed the method of pseudorandom noise codes for precise measurement of range to a satellite. GPS brought all three lines together into a single system — and that merger took place in the early 1970s, when the Department of Defense decided to end parallel development efforts and create one unified navigation program.
The genealogy is worth keeping intact. Transit was the world’s first operational satellite-navigation system and remained in service until 1996 — thirty-two years, outlasting many systems that came after it. But GPS was not “Transit 2.0”: it was the result of the convergence of several technological branches, each solving its own piece of the puzzle.
This whole story begins with two scientists deciding, on a Monday evening, to listen to a Soviet satellite beep. They had no assignment. No grant for the work. They had curiosity and a good receiver. Their measurements gave birth to an idea, the idea to a system, the system to an industry that now maps the route of every taxi and every container ship on the planet.
The Cold War played the role of catalyst in this story, not inventor. Without the launch of Sputnik 1, Guier and Weiffenbach would never have heard its signal. Without the Polaris program and the fear of nuclear war, the Navy would not have funded Transit. Without ARPA, created in panic after the Soviet satellite, the money would not have arrived so quickly. But the physics of Doppler shift did not depend on politics — it waited for someone to think of applying it.
Still, the specific path from “beep-beep-beep” to the navigator in your pocket took nearly forty years. Transit operated from 1964 to 1996. GPS reached full operational capability in 1995. Civilian users received unrestricted access to an accurate GPS signal only in 2000, when President Clinton ended the deliberate degradation of the signal (Selective Availability). Forty-two years passed between McClure’s insight in March 1958 and the moment when anyone with a hundred-dollar receiver could determine their coordinates to within ten meters. Technologies that change everyday life are rarely born quickly — but sometimes they are born from a sound someone decided not to ignore.