A system built for the military
GPS was designed by the U.S. Department of Defense, paid for by the U.S. government, and intended first of all for military users. Even so, its designers planned from the start to send two kinds of signal. One, a coarse and easy-to-acquire code, was meant to be available to a wider set of users. The other, a longer and more precise code, was reserved for authorized military receivers and was encrypted in a way that is usually described as protecting against spoofing. The question of how good the open signal should be, and who should be allowed to use it, was a policy decision rather than an engineering one, and it changed several times.
This article is about those decisions. The hardware story is in our article on GPS receivers, and the program’s wider military origins are covered in our look at military positioning.
The airliner shootdown of 1983
The most cited turning point is the destruction of Korean Air Lines Flight 007. On September 1, 1983, the Boeing 747 was flying from Anchorage to Seoul when it strayed far off its planned route and entered Soviet airspace. A Soviet fighter shot it down near Sakhalin Island, and all 269 people aboard were killed. Investigations later concluded that a navigation error, most likely linked to how the autopilot was set, had carried the aircraft off course. The long-range aircraft navigation of that time used inertial systems and radio aids, which could drift or be mis-set without anyone noticing.
In the weeks after the event, President Ronald Reagan announced that the United States would make GPS available for civilian use once the system was operational, so that similar accidents might be avoided. The promise is often described as the start of civilian GPS. It is fair to say it was a public commitment rather than an immediate change. At the time, only a few experimental satellites were in orbit, and a usable civilian service was still years away. The full effect of the decision for pilots is the subject of our article on aviation navigation.
Selective availability
As the system developed, military planners worried that an adversary could use the open signal to guide weapons or to navigate. Their solution was selective availability, usually shortened to S/A. This deliberately introduced small errors into the civilian signal, by dithering the satellite clock data and slightly altering the transmitted orbit data. It was activated in 1990. The accuracy available to civilian users was officially described as about 100 meters horizontally for most of the time. Authorized military receivers could remove the errors.
The policy was not without exceptions. During the Gulf War of 1991, the military had too few of its own receivers and many troops used commercial ones, and the government reportedly switched selective availability off for a period. That episode showed both the usefulness of civilian receivers and the contradiction in degrading a signal the military itself relied upon.
Working around the errors
Commercial users did not wait for the policy to change. They developed differential GPS, in which a receiver at a known fixed location compares its calculated position with its true one and broadcasts the difference as a correction to nearby users. Because selective availability errors were largely shared by users in the same area, corrections cut the error to a few meters. The U.S. Coast Guard built a network of beacons for marine users, and aviation authorities developed their own systems of corrections. Surveyors had earlier methods that avoided the problem. In this period, the degradation was less of a barrier than the policy implied, and it motivated companies to develop their own correction services.
The 1996 directive and the 2000 decision
By the mid-1990s the system was fully operating, and the use of GPS by civilians was growing quickly in transportation, surveying and timing. In 1996, a presidential policy directive stated that the United States would make GPS available worldwide for peaceful uses and said that selective availability would be discontinued within a decade. The reasoning was partly economic, since GPS-based industries were becoming important, and partly that the military had developed ways to deny GPS signals locally in a conflict zone without degrading them everywhere.
On May 1, 2000, President Bill Clinton announced that selective availability would be turned off. The change took effect in the early hours of May 2. Overnight, typical civilian accuracy improved from tens of meters to a few meters under good conditions. Later, from the satellites launched after 2007, the capability to turn the degradation on was removed from the design, which was intended to reassure users that it would not return.
Why the decision mattered
The decision removed a major barrier to consumer products. Within a few years, car navigation devices, hiking units and phone-based services grew rapidly, as described in our articles on handheld GPS and GPS chipsets. Applications built on the signal’s timing function, such as network synchronization, also depended on a signal that was freely available and stable.
The policy is still a balance. The government continues to operate the system and can limit access in specific regions, and military signals remain separate and encrypted. Additional civilian signals have been added to newer satellites to improve accuracy and reliability. There are also reasonable debates about dependence on a single service that can be jammed, which is why other countries built their own systems. In short, civilian access to GPS resulted from a series of political and practical decisions over about two decades, not from one announcement.