Transit: the system before GPS
The first operational satellite navigation system was not GPS. It was Transit, developed by the U.S. Navy with the Johns Hopkins Applied Physics Laboratory. Its origin is often told as a story about Sputnik. Soon after the first satellite was launched in October 1957, two laboratory physicists, William Guier and George Weiffenbach, found that they could follow its radio signal and work out its orbit from the Doppler shift, the change in frequency caused by the satellite’s motion. Their colleague Frank McClure is credited with suggesting the reverse: if the satellite’s orbit were known, a receiver could calculate its own position from the same shift.
The Navy had good reason to want this. It was building fleets of missile-carrying submarines, which needed reliable position updates at sea. Test launches began in the early 1960s, and the system was in operational Navy use in the mid-1960s. It was later made available to civilian ships, and it remained in service until the 1990s, when GPS replaced it. Transit used a small number of satellites in low polar orbits, so a user might wait a while between passes and then spend a few minutes on calculations to get a fix. It was accurate enough for ships but not for fast-moving aircraft.
Competing ideas in the 1960s and early 1970s
Transit’s limits prompted other services to propose their own satellite systems. The Navy also worked on a project called Timation, which put precise clocks on satellites and is associated with Roger Easton at the Naval Research Laboratory. The Air Force studied a concept sometimes referred to as Project 621B, which aimed at three-dimensional positioning. The Army had its own interests. Having several overlapping programs was costly, and defense planners were under pressure to combine them.
The 1973 merger
In 1973 the Defense Department created a joint program office, led by Air Force Colonel Bradford Parkinson, to bring the work together under the name Navstar GPS. A well-known story says the basic plan was agreed at the Pentagon over a holiday weekend that year. The details are often told with some embellishment, so it is safest to say that the key design decisions were made in 1973. The plan took the pieces that worked best from the earlier efforts: precise clocks on every satellite, signals that allowed a receiver to measure distance from time delay rather than relying on Doppler shift, and a high orbit that kept each satellite in view for hours.
Building the constellation
The design settled on satellites in medium Earth orbit, about 20,200 kilometers above the surface, each orbit taking close to 12 hours. The constellation was to consist of 24 satellites arranged in six orbital planes, which is intended to let a user almost anywhere see at least four satellites at once. The first generation was developed in stages.
- Block I: experimental satellites, with the first launched in February 1978. These tested the signals, the clocks and the ground control system. Several were launched in the late 1970s and the 1980s.
- Block II and later: operational satellites, launched after the Space Shuttle program’s delays caused by the 1986 Challenger accident pushed the program toward expendable rockets. The first Block II satellite reached orbit in 1989.
- Operational milestones: initial operational capability was declared in December 1993 and full operational capability in 1995.
Satellite clocks were a central part of the design. Early satellites used rubidium clocks, and later ones added cesium clocks. The technology behind them is explained in our article on the atomic clock. The ground side included a master control station in Colorado and monitoring stations around the world.
Costs and uncertainty
Reasonable people still disagree about how to credit the project. It was expensive, had difficult moments in Congress, and was designed first as a military tool, with civilian use treated as a secondary matter. The accident that killed 269 people on a Korean airliner in 1983 shifted that thinking, as described in our account of civilian GPS access. The system worked out well, but in the mid-1970s its future was uncertain, and a proposal that satellites might fail early or that the military might not need them was a real concern.
Why this was a breakthrough
Transit proved that satellite signals could be used for navigation. Navstar made that idea practical for a much larger range of users by providing continuous, three-dimensional position and timing across the world. The precise timing was the main advance, and it made many later applications possible, from aviation to the self-driving cars now being tested.
The two systems also differ in how they were retired. Transit was switched off in 1996 after a long period of service, when GPS made it redundant. By that time a new generation of receivers, discussed in our article on GPS receivers, had been built to use Navstar signals. The Navstar constellation itself has been replaced and upgraded many times since, so the satellites in orbit today are very different from those launched in 1978.