Holmes, along with Easton, was a Navy representative to the tri-service meetings discussing the DNSS and had the contacts to convince the Navy to agree with the solution reached. A two-day meeting was held in June 1973 at the Space and Missile Systems Organization in California. Four alternatives were discussed for the DNSS (Defense Navigation Satellite System) and a draft decision coordinating paper (DCP 121) was written, dated June 7, 1973. Alternative I called for the Air Force to launch one NES satellite and the Navy to launch NTS-1. Alternative II was the same as the instructions in Deputy Secretary Clements’s April 17 memorandum. During 1977 the Air Force would design and deploy a single constellation of four synchronous NES satellites and in 1974 the Navy would deploy NTS-1, a midaltitude satellite testing the Timation approach. Alternative III was similar to the second alternative, except that the Navy was to launch another satellite, NTS-2. NTS-1 would be launched in calendar year 1976 with a crystal oscillator to test signal processing in a subsynchronous eight-hour orbit. NTS-2 would be launched in 1977 with a cesium atomic clock. The fourth alternative was to proceed directly with a global, operational DNSS. No additional testing would be conducted prior to developing the satellites for launch in calendar year 1978. NTS-1 would be launched in 1974 to provide limited information on propagation effects (radio signal variability) and ephemeris (orbital variability), prediction methods required for the DNSS.
A Tale of Two Meetings
In August the newly founded joint program office, with Air Force colonel Bradford Parkinson as its head, attempted to get Project 621B adopted as the DNSS. However, the required approval from all of the services was not forthcoming. Parkinson claims that he and about twelve other Air Force and Aerospace Corporation personnel formulated GPS at the Pentagon in the “Lonely Halls ” meeting over Labor Day 1973.17 Roger Easton asserts that during the Labor Day weekend an important meeting occurred at a motel on Spring Hill Road in Virginia.18 There, Easton and Captain Holmes, representing the Navy, met Parkinson and other Air Force representatives. Ron Beard and James Buisson from NRL recall Easton discussing the Spring Hill meeting the week after Labor Day. Beard recently questioned why Parkinson and other Air Force personnel from California would travel to the Washington DC area only to meet among themselves.19 A meeting with people residing near the capital makes sense of their trip to the East Coast. Other evidence exists substantiating the Spring Hill meeting. Captain Holmes’s daughter, Dian Moulin, found in his papers transparencies of a presentation about GPS’S origins titled “Another Navigation System? Why GPS Wouldn’t Sell. ” The last slide, reprinted in figure 4.4, corroborates Easton’s recollections about the motel meeting.
Easton asserts that in the Spring Hill meeting, Parkinson said that 621B was rejected because it was too expensive and Holmes offered him the Timation system. Easton recalls that Parkinson listened but did not comment on the offer. Ron Beard recently wrote, “There were several meetings at the motel and elsewhere. I remember going to the motel also. The meeting that turned the tide was one that [Easton] and Holmes were at. ”20
Obviously, recollections of events that occurred decades earlier can honestly differ. Whether history should trace the compromise that led to the ultimate synthesis of GPS architecture to a meeting with two NRL representatives at a motel, to the Lonely Halls Pentagon meeting of Air Force and Aerospace personnel, or to its genesis over many months of tri-service negotiations depends on whom one asks. Sonnemann, who was part of those negotiations before the formation of the joint program office, offered this assessment: “In the short interval between the rejection of the 621B system by the DSARC in August 1973, the September 1973 Labor Day meeting, and the second DSARC in December 1973, there was not enough time to do any more than make sure that the elements of the system as then constituted by the NAVSEG/NAVSMO [Navigation Satellite Management Office] were 1) Technically sound and 2) Would indeed satisfy the requirements that the synthesized Global Position System was the product of the NAVSEG joint service program and its supporting joint services NAVSMO staff. ”21
Fig. 4.3. Col. Bradford W. Parkinson, first manager of the Joint Program Office, about a year before he retired from the Air Force. (USAF photograph)
Keith McDonald, who headed NAVSMO, which provided technical and administrative support to NAVSEG, has written that the Air Force Space and Missile Systems Organization (SAMSO) “appropriate[d] the product of NAVSEG’S concept development effort. ” In a telephone interview he observed that the final GPS architecture so closely resembled what NAVSEG members had already agreed upon, crafting the specifications would not have required extensive effort. “You could do it in half a day ,” he said. “It wouldn’t take a weekend. ”22
Fig. 4.4. Holmes’s presentation on GPS origins, slide 13. This reproduction of a transparency used in a presentation by Capt. David Holmes references a meeting at the Spring Hill Motel at Bailey’s Crossroads in Virginia, where Navy representatives recommended that the Joint Program Office use the Timation system as the basis for GPS. (Courtesy Dian Moulin)
Whichever viewpoint one accepts, it is clear that after the Labor Day 1973 weekend, the GPS system proposal reached a formulation with enough performance characteristics of the Timation system that Navy opposition ceased. The media reported on the compromise, and military officials touted it.
The main elements of compromise involved the signal, the onboard atomic clocks, and the orbital configuration. In a contentious exchange of letters to the editor of a technical journal in 1985, Parkinson responded, “As Roger Easton has noted, the Timation orbital configuration was, in fact, the basis for the Navstar system design. ”23 But after the disagreement, Parkinson began asserting that the orbits GPS used were not the same orbits Timation used. Although the Block I experimental satellites were placed into orbits at a higher angle of inclination from the equator to facilitate testing and tracking, the Block II operational satellites used the fifty-five-degree inclination specified for Timation. Parkinson often focuses on the techniques used to determine the precise locations of the satellites in their orbits, rather than the configuration itself, as he did in a 1999 interview:
For example, Transit gave us orbit determination. They really knew how to do that, and we needed it, because GPS satellites have to know where they are. Roger Easton brought the atomic clock technology forward, and we needed very stable time because in essence GPS acts as a one-way radar. That is enabled by knowing precisely when the GPS signal was generated. The third contributions are the digital signal structure and the concept of operation. The Air Force was pretty close, and we took their signal structure and refined it further, added some features to it that weren’t there. All these ideas contributed to the final GPS of 1973 (which is still essentially unchanged).24
This may overstate Transit’s role in GPS. Two Timation satellites had already been launched, and NRL tracked their positions precisely enough to estimate the positions of receivers. Easton had designed two major space-tracking systems, Minitrack and Space Surveillance, so he knew how to track satellites. And the problems posed by gravitational variations are far less for satellites at higher, eight-thousand- or twelve-thousand-mile orbits than they were for the Transit satellites at roughly seven-hundred-mile orbits. “The GPS high altitude satellites will avoid the drag which continuously changes the orbits of the TRANSIT birds ,” Holmes wrote in an article published in the Naval Institute’s Proceedings magazine.25 Aviation Week & Space Technology, which closely covered satellite navigation progress, reported, “The ability of USAF and Navy to resolve their long-standing differences over the orbital configuration by basically adopting the Navy-proposed constellation arrangement has eliminated one of the major obstacles to Pentagon approval for the program. ”26
Three alternatives are presented in the subsequent Development Concept Paper Number 133, dated November 26, 1973. The first was to stop development leading to GPS. The second was to launch four synchronous repeater Navigation Development Satellites (a 621B constellation) and
three NTS satellites. The third, which the paper’s authors strongly advocated, was to launch two NTS satellites and three subsynchronous Navigation Development Satellites. This third option, with modifications, is what occurred. The 621B satellites were canceled, and the Timation satellites were launched under different names.
It is important to note that the Timation Development Plan specified using both sidetone (or continuous wave) signals and spread-spectrum signals (see chapter 5). This report was submitted on November 25, 1970. Thus the Timation Development Plan proposed using a spread-spectrum signal two years before Colonel Parkinson assumed leadership of Project 621B. This fact has been overlooked in debates about the origin of GPS. An August 6, 1971, memo from Air Force lieutenant colonel Paul S. Deem states that a System 621B signal modulator would be flown aboard Timation III (later launched as NTS-1). This occurred; thus there was cooperation between the Navy and Air Force long before the formation of the joint program office. There has been so much emphasis on the incorrect story that only the Air Force planned to use a spread-spectrum signal that even some of the people involved in Timation have forgotten the plan’s intention to use both types of signals. Sonnemann, the former NAVSEG chairman, commented, “Once the orbit decision was made, a navy ‘victory’ in the eyes of the navy and the Air Force, the signal structure became the next hurdle. The use of the pseudo-random sequence structure proposed by the Aerospace Corporation for its 621B system had a number of attractive features, and no significant detrimental characteristics that would suggest the Timation signal structure should prevail. Adopting the 621B signal structure had the further advantage of balancing the slate by giving the Air Force a ‘victory.’ ”27
Easton offered his assessment in a 1996 interview with Naval Research Laboratory historian Dr. David K. van Keuren:
Dr. van Keuren: Parkinson agreed after these meetings to take the Timation system and manage it [?]
Easton: Essentially.
Dr. van Keuren: Essentially. How much of the combined system was Air Force and how much was Navy?
Easton: I say, “What percentage do you want to put on it? ” You could say it was ninety percent Navy and ten per cent Air Force, or you could use some other thing, but it was the Timation system with a different modulation. Essentially that is what it amounted to. That was in a way good because it gave the Air Force something to say that, well, it was a combination of two systems. The modulation, I thought, was the least essential thing because it could have been pulses FM, AM, sidetones, pseudo random noise, all kinds of ways you could have done it. We used the sidetones, and they worked, and we had hundreds of thousands of passes… . The original agreement was that we would have both, side tones and pseudo random noise. To me that made the most sense. You could have the sidetones for civilians and pseudo random noise for the military. If you needed to turn off the civilians, you just turn off that modulation.28
Former NAVSEG chairman Sonnemann was asked to provide his perspective on the elements that were synthesized to create GPS:
Question: In your estimation what percentage of the GPS system that emerged owes its technological origin to 621B?
Answer: The signal structure—100 percent.
Question: The same question for Timation?
Answer: The orbiting satellites (Timation) with precision atomic clocks—100 percent.29
However, the spread-spectrum signal structure was invented in World War II by the actress Hedy Lamarr and avant-garde composer George Antheil.30 They deserve at least some of the credit for the signal used by GPS.
The Atomic Clocks
The Defense Systems Acquisition Review Council (DSARC) accepted the compromise on December 17, 1973. A major change was augmenting crystal oscillators with rubidium atomic clocks in NTS-1. Holmes described how that came about in an article for the defense industry magazine Countermeasures: “However, six months before the NTS-1 scheduled launch date a rubidium clock made by Efratom [a German company—the clock was developed by Ernst Jechart and Gerhard Huebner] was brought to the attention of Mr. Easton. Being an atomic clock, the rubidium oscillator offered the promise of both higher accuracy and higher, long-term stability. With joint program office approval, NRL decided to take the risk and install it, even though time available for testing and installation of space-qualified parts was minimal. ”31
For the second NTS satellite, Robert Kern and Arthur McCoubrey, who were leaders in developing atomic clock technology, built the first cesium atomic clock placed in orbit. NTS-2, launched in June 1977, provided useful information for GPS’S further development, as reported at the time by Aviation Week & Space Technology: “Development test version of the Defense Department’s NavStar global positioning system tracked an instrumented USAF/Lockheed C-141 with three-dimensional accuracies of 3–4 meters (10–13 feet) during a recent test which synchronized the NavStar inverted range at Yuma Washington with the orbiting Navigation Technology Satellite (NTS-2). This constituted the first true navigational test of the NavStar-GPS concept, using a satellite and proved that the system is a workable one, according to officials from the joint program office at Air Force Space and Missile Systems Organization (SAMSO). ”32
Fig. 4.5. Diagram of NTS-2 satellite. NTS-2, the final NRL navigation satellite, carried the first cesium atomic clock into orbit in 1977. (Courtesy Naval Research Laboratory)
Harvard professor Peter Galison has written that Einstein’s development of special relativity was at least partially inspired by observing synchronized clocks in Europe.33 In turn, relativistic clock corrections are critical to GPS’S provision of worldwide clock synchronization. NRL’s Vince Folen and Don Lynch estimated correctly the relativistic corrections used in NTS-2’s cesium clocks and University of Maryland professors Joe Weber and Carroll Alley were also very helpful in this calculation. The NRL’S Peter Wilhelm and his staff built all four Timation satellites and developed innovative techniques to minimize launch costs. The 1971 Timation Development Plan proposed launching all twenty-seven satellites using three rockets with an estimated completion date of 1984.
After receiving approval in December 1973, Parkinson and the JPO implemented the plan by working on the GPS Block I satellites and the receiver equipment. A major problem GPS managers continually faced was convincing some military leaders and many in Congress that this satellite navigation system’s costs were worthwhile. Civilian applications would eventually be plentiful, but no civilian money supporting the program was forthcoming. Parkinson retired from the JPO and the military in 1978. Sonnemann commented about Parkinson’s time leading the GPS program: “The Air Force JPO under Col. Parkinson did an excellent job in implementing GPS. Until the capability of the system was demonstrated and verified, no significant funds could be expected to be committed from non-DOD [Department of Defense] government organizations or the civilian potential users of GPS, so it would have been very difficult to speed up the process. ”34
Fig. 4.6. NTS-2 satellite and first GPS demonstration constellation. An artist’s rendering of NTS-2 and three subsequent navigation development satellites, which together created the first demonstration GPS constellation, giving users longitude, latitude, altitude, and time. (Courtesy Naval Research Laboratory)
In sum, when the documentary record and the recollections of those who differ with Parkinson’s narrative are considered, a more complicated picture emerges. These sources reveal that the synthesis of competing satellite navigation approaches resulted from a long, complex interservice negotiation. That process culminated in a compromise that incorporated the 621B spread-spectrum signal into a constellation of satellites carrying atomic clocks in a worldwide configuration of midaltitude orbits that resembles Timation, not the regional constellations of 621B. During the final, critical period of negotiations leading to the compromise, Parkinson appears to have played the role of an adroit broker, navigating the differing approaches of the two services into one aligned course. Although the two sides successfully merged the technical specifications
into a single system, their stories about which system contributed each element and how they achieved the compromise remain divergent.
Fig. 4.7. NRL NTS-2 team. The NRL team that launched NTS-2 included (standing, left to right) Dr. Bruce Faraday, Richard Statler, Guy Burke, and Roger Easton; (seated, left to right) Al Bartholomew, Cdr. Bill Huston, Red Woosley, Ron Beard, Woody Ewen, and Pete Wilhelm. (Courtesy Naval Research Laboratory)
5
Invisible Stars How GPS Works
Any sufficiently advanced technology is indistinguishable from magic.
Arthur C. Clarke, Profiles of the Future: An Inquiry into the Limits of the Possible, 1973
Throughout history new technologies have intimidated those unschooled in their operation, and inventions sometimes have been branded as sorcery or witchcraft. This was true of early mechanical clocks; many eleventh-century “commoners ” attributed their origin to the devil, so the introduction of clocks for general use flowed through the church, where monasteries found them useful for waking monks at prayer hour.1
Noted astronomer Herbert Hall Turner, former chief assistant at the Royal Greenwich Observatory and president of the Royal Astronomical Society from 1903 to 1905, once observed that to secure the Longitude Prize John Harrison “first had to disprove a charge of witchcraft by showing that his wonderful clock could be duplicated by another workman. ”2 Since the Witchcraft Act of 1736 had repealed prior laws treating witchcraft as real and recast the crime as one akin to fraud or con artistry, the charge brought against Harrison was undoubtedly a tactic aimed at forcing him to divulge his secret design to the Board of Longitude.3
GPS Declassified Page 9
