Overview Of The Global Positioning System And Current Or Planned Augmentations
Origins And Development Of The NAVSTAR GPS Program
The navigation, positioning, and timing system that is known today as the Global Positioning System (GPS) is a combination of several satellite navigation systems and concepts developed by or for the DOD (Department of Defense). The predecessors to GPS include the following satellite systems: (1) Transit, an operational system developed for the U.S. Navy by the Johns Hopkins Applied Physics Laboratory that is still in use today1 ; (2) Timation, an experimental program developed for the Navy by the Naval Research Laboratory that demonstrated the ability to operate atomic clocks on board orbiting satellites and was used as a system concept for GPS;2 and (3) Project 621B, an Air Force study program originated in 1964 by Aerospace Corporation and the Air Force's Space and Missile Organization.3 In addition, a DOD Four Service Executive Steering Group was established in 1968 to investigate the development of a Defense Navigation Satellite System that would satisfy all of the DOD's satellite navigation requirements.
By 1972, the best characteristics of each of these four programs had coalesced to form the general system characteristics and initial design parameters for the system now known as the NAVSTAR Global Positioning System.4 The system configuration and a request for developmental funding was submitted to the Director of Defense Research and Engineering, and the Air Force agreed to become the Executive Agent for this joint system.
The GPS program was approved in 1973, and a Joint Program Office (JPO) located at the Air Force Space and Missile Organization in El Segundo, California was established.
From its inception, GPS was designed to meet the radionavigation requirements of all the military services and those of civilian users as well. On February 22, 1978, the Air Force began launching experimental GPS satellites, termed Block I satellites, on Atlas F launch vehicles. After the third satellite successfully achieved orbit, testing of the system's capabilities began at Yuma Testing Grounds, Arizona. Using a portable receiver mounted in a truck moving at 80 kilometers per hour, the Air Force showed that the desired positioning accuracy of 10 meters in two dimensions was easily achievable. After tests with the first three experimental satellites proved successful, eight additional Block I satellites were launched to complete the design and testing phase of the GPS program.5 Although these satellites, designed and built by Rockwell International, were intended to have a 3-year life span, they achieved an average operational life of almost 7 years, and one of the Block I satellites was still operating as of the date of this report.
The next series of satellites, termed Block II, was designed to be fully operational. The first Block II satellite was launched aboard an Air Force Delta II rocket on February 14, 1989.6 The current GPS constellation consists of 24 Block II/IIA operational satellites, and as previously mentioned, 1 Block I experimental satellite.
The GPS JPO has done an outstanding job of developing and testing the systems and equipment for GPS, as well as acquiring the hardware and software needed to deploy the system. This excellent effort was recognized in 1994 with the award of the Collier Trophy to the JPO and several of the major contractors involved in the GPS program. 7
GPS Policy, Management, And Operations
Department of Defense
Responsibility for the day-to-day management of the GPS program and operation of the system continues to rest with the Department of Defense, and is carried out primarily
by the Air Force.8 GPS research and development is managed by the Space and Missile Systems Center at Los Angeles Air Force Base. Testing and evaluation is conducted jointly by the Air Force Operational Test and Evaluation Center at Kirtland Air Force Base, New Mexico, and Air Force Space Command at Falcon Air Force Base, Colorado. Operations and maintenance also are managed by Air Force Space Command. Procurement and budgetary oversight for GPS are managed by Program Element Monitors within the space systems office of the Assistant Secretary of the Air Force for Acquisition. Through fiscal year 1994, the cumulative procurement budget for the space and ground control segments of the GPS is approximately $3.5 billion; and research, development, testing and evaluation spending totals approximately $3.7 billion.9
DOD policy for the GPS program is set by the Under Secretary of Defense for Acquisition and Technology, with the help of the DOD Positioning/Navigation Executive Committee. This committee receives input from all of the DOD commands, departments, and agencies, and coordinates with the Department of Transportation (DOT) Positioning/Navigation Executive Committee.
Department of Transportation
In response to a request from the DOD, and in order to meet the needs of civil GPS users, the DOT established the Civil GPS Service (CGS) in 1987. The CGS is operated and managed within the DOT by the Coast Guard and consists of the following: (1) the Navigation Information Service, which provides GPS status information to civilian users; (2) the Civil GPS Interface Committee, which provides a forum for exchanging technical information in the civil GPS community; and (3) the Civil PPS Program Office, which administers the program that gives qualified civil users access to the Precise Positioning Service (PPS) signal, used primarily by the U.S. and allied armed forces.
In May of 1993, the Secretary of Defense and the Secretary of Transportation agreed to examine the operational, technical, and institutional implications of increased civil use of GPS in order to satisfy both military and civilian needs. The resulting joint DOD/DOT task force concluded its work in December 1993 with the release of a report titled The Global Positioning System: Management and Operation of a Dual Use System - A Report to
the Secretaries of Defense and Transportation.10 In response to management recommendations made in the report, the DOT has established a DOT Positioning/Navigation Executive Committee to interface directly with the DOD Positioning/Navigation Committee. The duties of the committee chair have been assigned to the Assistant Secretary for Transportation Policy who, along with the Under Secretary of Defense for Acquisition and Technology, will co-chair the newly formed joint DOD/DOT GPS Executive Board. This management structure is illustrated in Figure C-1. The DOT Positioning/Navigation Executive Committee and the Assistant Secretary for Transportation Policy will act as the focal point for GPS plans and policies developed by a number of DOT agencies involved in the use of GPS. These organizations include the U.S. Coast Guard, The Federal Aviation Administration (FAA), the Federal Highway Administration (FHWA), and the Federal Railroad Administration (FRA). The Executive Committee will also receive input from the Civil GPS Service Interface Committee.
The Federal Radionavigation Plan
The Federal Radionavigation Plan is the official source of planning and policy information for each radionavigation service provided by the U.S. government, including GPS. It is jointly developed by the DOD and the DOT, and is updated biennially.11 The Federal Radionavigation Plan represents an attempt to provide users with the optimum mix of federally-provided radionavigation systems, and reflects both the DOD's responsibility for national security, and the DOTs responsibility for public safety and transportation economy. It was first released in 1980 to Congress in response to the International Maritime Satellite Telecommunications (Inmarsat) Act of 1978 (P.L. 95-564).12
GPS Technical Overview
The technical and operational characteristics of GPS are organized into three distinct segments: the space segment, the operational control segment, and the user equipment segment. The GPS signals, which are broadcast by each satellite and carry data to both user equipment and the ground control facilities, link the segments together into one system. Figure C-2 briefly characterizes the signals and segments of the Global Positioning System, which are discussed in some detail below.
The GPS constellation consists of 24 satellites, arranged in 6 orbital planes of 55-degree inclination, 20,051 kilometers (12,532 miles) above the Earth. Each satellite completes one orbit in one half of a sidereal day and therefore passes over the same location on earth once every sidereal day, or approximately 23 hours and 56 minutes. This particular orbital configuration and number of satellites allows a user at any location on the earth to have at least four satellites in view 24 hours per day. The constellation described above currently consists of 24 Block II/IIA satellites and one Block I satellite, which have been built for the U.S. Air Force by Rockwell International Satellite and Space Electronics Division, Seal Beach, California. Based on a fixed price, multi-year procurement contract totalling approximately $1.5 billion for 28 satellites, the unit cost of each satellite is approximately $53.8 million (1995 dollars).13 Each Block II/IIA satellite is designed to operate for 7.5 years, but may operate beyond this life span based on the success of the Block I series. Figure C-3 shows a typical Block II/IIA GPS satellite.
Each Block II/IIA satellite weighs 1,881 kg (4,147 pounds) when fueled and is designed for a solo launch aboard an Air Force Delta II rocket.
The follow-on Block IIR replenishment satellite contract was competitively awarded in 1989 to Martin Marietta Astro Space Division, East Windsor, New Jersey for a total of 20 satellites. The estimated unit cost of each Block IIR satellite is $30.1 million (1995 dollars).14 Recently, the Air Force exercised an option in the Block IIR contract to purchase one additional satellite. These satellites will also be carried into orbit by the Delta II rocket, with the first launch currently scheduled for 1996. Figure C-4 represents a typical Block IIR satellite.
Although the Block IIR satellites are very different in appearance from the Block II/IIA satellites, they have been built to the same basic specifications and comprise the same kinds of components and subsystems. Many of the subsystems and components,
however, have been designed for improved performance and reliability, including the solar arrays, the gyroscopes, the batteries, and the nuclear-detonation detection system payload. In addition, the navigation payload on board the Block IIR satellites carries one cesium and two rubidium clocks, rather than the two rubidium and two cesium clocks present on the Block II/IIA spacecraft.15 The Block IIR satellites also have two important operational capabilities not available from the Block II/IIA satellites. First, each subsystem and payload has been designed to allow on-orbit software reprogramming, allowing for much greater operational flexibility and upgrading, and second, the satellites can maintain specified positioning accuracy without contact from the operational control segment (OCS) on the ground for up to 180 days. This mode of operation, known as autonomous navigation or autonav, is accomplished by relaying positioning information between satellites using ultrahigh frequency (UHF) inter-satellite links. 16
The draft request-for-proposal (RFP) for the next generation of satellites beyond the Block IIR design, known as the Block IIF, is currently scheduled to be released in the spring of 1995, and the final version is currently scheduled for release in the summer. The first launch is anticipated in 2001.17
Operational Control Segment
The GPS operational control segment (OCS) consists of the master control station (MCS), located at Falcon Air Force Base in Colorado Springs, Colorado; remote monitor stations, located in Hawaii, Diego Garcia, Ascension Island, and Kwajalein; and uplink antennas located at three of the four remote monitor stations and at the MCS.18 The four remote monitor stations contribute to satellite control by tracking each GPS satellite in orbit, monitoring its navigation signal, and relaying this information to the MCS. These four stations are able to track and monitor the whereabouts of each GPS satellite 20 to 21 hours per day. Land-based and space-based communications are used to connect the remote monitoring stations with the MCS.
The MCS is responsible for overall satellite command and control, which includes maintaining the exact orbits of each satellite and determining any timing errors that may be present in the highly accurate atomic clocks aboard each satellite. Errors in a satellite's orbital position or in a satellite's timing are determined by analyzing the same signal and
navigation message from each satellite that is used by GPS receiver equipment. Using a Kalman Filter, computers at the MCS process the data collected at all the monitor stations in order to estimate these errors.19 Updated orbits and clock corrections are relayed once a day to each satellite by the four ground antennas.
The day-to-day operations at the MCS are carried out by personnel belonging to the 2nd Space Operations Wing of Air Force Space Command. Routine maintenance is also conducted by the Air Force and its contractors. Remote monitoring stations are largely automated, but a small number of contract personnel do monitor and maintain each station's equipment. Average annual personnel and maintenance cost for the MCS, the four remote monitoring stations, and all their associated equipment is approximately $30 million.20
GPS user equipment varies widely in cost and complexity, depending on the receiver design and application. Receiver sets, which currently vary in price from approximately $400 or less to $30,000, can range from simple one-channel devices that only track one satellite at a time and provide only basic positioning information, to complex multi-channel units that track all satellites in view and perform a variety of functions. Most GPS receivers, however, consist of the same three basic components: (1) the antenna, which receives the GPS radio signal and in some cases provides anti-jamming capabilities; (2) the receiver-processor unit, which converts the radio signal to a useable navigation solution; and (3) a control/display unit, which displays the positioning information and provides an interface for receiver control.
The subsections of a typical GPS receiver-processor unit include the front-end section, the digital signal processor, and the microprocessor. The front-end section translates the frequency of a GPS signal arriving at the antenna into lower or intermediate frequency (IF) and converts the signal from analog to digital. This more manageable signal is then passed to the digital signal processor, which "tunes in" to these signals using tracking loops that compare incoming signal data to internally generated models of the satellite signals. GPS receivers normally track more than one signal at a time using multiple channels, but also can track multiple signals using either a single channel sequenced between satellite signals or a multiplexing channel. Once the digital signal processor is successfully tracking a set of GPS signals, the ranging data it extracts is passed to the microprocessor, where
computer software converts it into information that can be usefully displayed for a user, such as position coordinates, or input to another type of user equipment, such as an inertial navigation system.21
Although the functions of a current GPS receiver are the same as those present in user equipment tested in the 1970s, they have little else in common. The size and cost of user equipment has decreased dramatically, while capabilities and the size of the commercial market continue to increase. In 1993, the total value of the GPS user equipment market was estimated to be $420 million, with over 100 companies marketing GPS receivers.22 U.S. manufacturers maintain a competitive advantage over their Japanese counterparts, who are currently the principal competitors. However, the advantage could easily be lost. Larger U.S. companies, like Trimble Navigation, Ltd., invest as much as $25 million per year in GPS research to maintain their technological advantage. At the present time, U.S. domestic sales per unit represent less than 50 percent of the worldwide GPS market, and 45 percent of U.S. industry sales are to overseas markets.23
GPS Signal Characteristics and Operational Concepts
The GPS relies on the principle of "pseudoranging" to provide accurate positioning to its users. Each satellite in orbit continuously transmits a radio signal with a unique code, called a pseudorandom noise (PRN) code, that includes data about the satellite's position and the exact time that the coded transmission was initiated, as kept by the satellites' onboard atomic clocks. A pseudorange measurement is created by measuring the distance between a user's receiver and a satellite by subtracting the time the signal was sent by the satellite from the time it is received by the user.24
Once three ranges (or distances) from three known positions are measured, a position in all three dimensions can be determined. In the case of GPS, however, a fourth satellite is generally needed in order to eliminate a common bias in the pseudoranges to all satellites caused by a lack of synchronization between the satellite and receiver clocks. Once this clock bias is eliminated by the presence of a fourth signal, a highly accurate three-dimensional position can be determined. Figure C-5 below further illustrates the GPS pseudoranging concept.
Instead of transmitting one PRN code on one radio signal as described above, each satellite actually transmits two distinct spread spectrum signals that contain two different PRN codes, called the Coarse Acquisition (C/A) code and the Precision (P) code. The C/Acode is broadcast on the L-band carrier signal known as L1, which is centered at 1575.42 MHz. The P-code is broadcast on the L1 carrier in phase quadrature with the C/A carrier
and on a second carrier frequency designated as L2, that is centered at 1227.60 MHz. Figure C-6 illustrates the characteristics of both the L1 and the L2 signals.
The L1 C/A-code provides free positioning and timing information to civilian users all over the world, and is known as the Standard Positioning Service (SPS). The timing information on the C/A-code is also used by some receivers to aid the acquisition of the more accurate P-code. The P-code is normally encrypted using National Security Agency cryptographic techniques, and decryption capability is available only to the military and other authorized users as determined by the DOD. When encrypted, the P-code is normally referred to as the Y-code. The encryption process utilized, known as Anti-Spoofing (A-S), denies unauthorized access to the Y-code, and also significantly improves a receiver's ability to resist locking onto mimicked GPS signals, which could potentially provide incorrect
positioning information to a GPS user.25 Y-code availability through authorized decryption capability is known as the Precise Positioning Service (PPS).
Selective Availability and Other Positioning Errors
Before the PPS and SPS were established by the DOD with their current specified accuracy levels, the designers of GPS had anticipated that use of the Y- and C/A-codes would produce very different levels of positioning accuracy. Use of the Y-code was expected to result in 10-meter accuracy, whereas the C/A-code was expected to provide accuracy of 100 meters. Developmental testing of Block I GPS satellites, however, revealed that the accuracy difference between the two codes was not this significant. A report developed by the Joint Chiefs of Staff in the late 1970s highlighted this fact, and recommended that the GPS accuracy made available to civilians should be limited to 300 meters to 500 meters for national security reasons.26 The precise positioning and standard positioning services were soon established, with PPS accuracy officially specified as 16 meters (SEP), and SPS accuracy specified as 500 meters (2 drms).27 The SPS accuracy level was later changed to 100 meters (2 drms), as announced by the Under Secretary of Defense for Research and Engineering on June 28, 1983. This two-level accuracy arrangement is made possible on the Block II/IIA satellites through an accuracy denial method known as SA (Selective Availability), which was activated on March 25, 1990.
SA is a purposeful degradation in GPS navigation and timing accuracy that controls access to the system's full capabilities. SA is accomplished in part by intentionally varying the precise time of the clocks on board the satellites, which introduces errors into the GPS signal. This component of SA is known as dither. A second component of SA, known as epsilon, can also add error to the signal by providing incorrect orbital positioning data. PPS receivers with the appropriate encryption keys can eliminate the effects of SA. SA-induced errors can be varied by the DOD at the request of the National Command Authority or eliminated altogether, as was the case during the Persian Gulf War and the initial
The process of sending incorrect information to an adversary's radio equipment (in this case a GPS receiver) without their knowledge, using mimicked signals, is known as spoofing.
Potential Military Exploitation of the NAVSTAR GPS by Adversary Nations (Washington, D.C.: Organization of the Joint Chiefs of Staff, date unknown)
SEP, or spherical error probable, represents an accuracy that is achievable 50 percent of the time in all three dimensions (latitude, longitude, and altitude). PPS accuracy is normally represented in this manner. SPS accuracy, however, is normally represented using a horizontal 2 drms measurement, or twice the root mean square radial distance error. Normally, 2 drms can be graphically represented as a circle about the true position containing approximately 95 percent of the position determinations. 2 drms, and other positioning accuracy definitions are discussed in greater detail in Appendix D.
In practice, there are several additional sources of error other than selective availability that can affect the accuracy of a GPS-derived position. These include unintentional clock and ephemeris errors, errors due to atmospheric delays, multipath errors, errors due to receiver noise, and errors due to poor satellite geometry. Each of these error sources is discussed below and summarized in Table C-1.
Table C-1 GPS Positioning Errorsa
Range Error Magnitude (meters, one sigma)
Clock and Ephemeris
Total User Equivalent Range Error (UERE)b
Typical Horizontal DOP (HDOP)c
Total Stand-Alone Horizontal Accuracy, 2 drmsd
a. The error budget figures included in this table are conservative estimates for a typical stand alone C/Acode receiver using standard correlation techniques, and a typical dual frequency Y-code receiver. This information was provided by the Jet Propulsion Laboratory (JPL) of the National Aeronautics and Space Administration (NASA), Pasadena, CA. Notes related to each component of this error budget, and the assumptions made to derive its value, are provided with Table 3-1 in Chapter 3.
b. The total UERE is determined by adding the squares of the individual error magnitudes and taking the square root of the total.
c. Dilution of precision (DOP) is discussed below, and HDOP is mathematically defined in Appendix D.
d. The 2 drms horizontal positioning error is equal to 2 times UERE times HDOP. This mathematical relationship is further defined in Appendix D.
Atmospheric error is caused by the delay of the GPS signal as it passes through the Earth's atmosphere. Part of this delay is due to the troposphere and part is due to the ionosphere. Because the ionospheric effect is dispersive and is a function of frequency, dual-frequency GPS receivers can calibrate this effect by measuring the differential delay and/or phase advance between the L1 and L2 frequencies, thus eliminating a great deal of the atmospheric error.
Civil users do not have direct access to dual frequency observations but have several means for reducing the ionospheric error contribution. For stand-alone navigation most C/A-code receivers apply an ionospheric correction, known as the Klobuchar Model, which
can correct approximately 50 percent of the total ionospheric delay. 30 The model parameters are transmitted in the navigation message and are updated infrequently. High performance C/A-code receivers often perform codeless or cross-correlation tracking of the L2 signal to permit them to derive ionospheric correction parameters. These techniques suffer from substantial signal-to-noise ratio losses and do not work well in high-blockage or high-dynamic situations.
Tropospheric delay cannot be eliminated through the use of two frequencies, but both C/A-code, and Y-code receivers can eliminate most of this error using software modelling.31
Clock and Ephemeris Error
As shown in Figure C-7, the atomic clocks on board each GPS satellite are designed to provide highly accurate timing specifications. Even a small amount of inaccuracy, however, combined with the fact that the estimated orbital positions, or ephemeris, of each satellite are also not exact, can cause a certain amount of error in a receiver's position solution.
Multipath errors occur when incoming GPS signals bounce off a reflective surface such as a building or a body of water before reaching a user's receiver. For highly specialized receivers that are able to eliminate other error sources, pseudorange and/or carrier-phase multipath is frequently a dominant error source.
GPS receivers themselves introduce several sources of error to the measurement of satellite ranges. Thermal noise produced by the environment and the various components within a receiver cause small random errors. Received signal to noise ratio, quantization of the analog to digital converter, and the type of tracking loop used by a receiver are also determining factors in the noise level. Typical receiver errors can be as little as 1 centimeter or as large as several meters. This error is quite random in nature and is often reduced by averaging or smoothing over a short period of time.
Dilution of Precision
Dilution of precision, or DOP, is a term that describes the effect of satellite geometry on positioning, timing, and velocity accuracy. Any positioning system that relies on pseudoranging will be affected by the angular spacing between the known points that are used to measure from. The GPS constellation has been designed to give users at least four satellites in view with good geometric spacing, but terrain and man-made structures can occasionally block a receiver's view of some satellites, especially those near the horizon, making the dilution of precision less than ideal.
Improving The Capabilities Of GPS
Even before the implementation of SA in 1990, many potential GPS users envisioned a need to improve the accuracy of the system, as well as some of its other specified characteristics. Although GPS accuracy has just been discussed, other characteristics such as integrity, availability, continuity of service, and resistance to radio frequency (RF) interference require further elaboration.
Integrity, as defined by the Federal Radionavigation Plan, is the ability of a navigation system to provide timely warnings to users if and when the system should not be used. The integrity function of a navigation system involves monitoring the system's errors and, if specified protection levels are estimated to be exceeded, giving a warning to the user that the system cannot be used for navigation. In the case of GPS, integrity is maintained by monitoring the signal emanating from each satellite and determining if the pseudorange accuracy meets specified performance criteria for a given application.
Two statistical measures of integrity are often used. One measure relates the probability that a hazardously misleading error will occur and the probability that this error will go undetected (1 minus PHE times PD, where PHE is the probability of hazardous error and PD is the probability of missed detection). The second measure of integrity is simply the time a navigation system takes to warn the user that a hazardous error exists (time-to-alarm) There is currently no specified integrity value for either the GPS SPS or the PPS.
The availability of a navigation system, which is also defined in the Federal Radionavigation Plan, is the percentage of time that the services of the system are useable. Availability is an indication of the ability of a system to provide useable service within the specified coverage area. For GPS, "useable service within the specified coverage area" means
that at least four satellites must be visible to a user's receiver anywhere on or near the Earth, and the satellites must be providing the required positioning accuracy for the user's application. Some GPS applications, such as static surveying, do not require continuous availability. Others, such as air navigation, can require that GPS signals be available 99.999 percent of the time. The average availability of four or more GPS satellites in view of a given receiver, at SPS accuracy levels, is currently specified as 99.85 percent.32
Continuity of Service
Continuity of service, which also is referred to as reliability, is the ability of a navigation system to provide required service over a specified period of time without interruption. The level of continuity is expressed in terms of the probability of not losing the radiated guidance signals.33 Where warranted, continuity of service is achieved by using redundant transmitters and monitors. Continuity of service and availability go together in that availability is the probability that a system will be in service when it needs to be used, and reliability is the probability that the system will continue to provide service. The global average reliability for GPS is specified as 99.97 percent.34
Resistance to RF Interference
The accuracy of a GPS receiver can be degraded in the presence of unwanted interfering signals from terrestrial or other sources. In extreme cases, the receiver is unable to provide any useful navigation or positioning capability. Unwanted and unintentional sources of interference exist, such as the third harmonic of some UHF transmitters, which many civilian users may be unaware of. Military users are also concerned with unintentional interference, but they are more concerned with deliberate efforts to prevent the use of navigation signals through jamming. While no receiver can be made entirely immune to interference (intentional or otherwise), steps can be taken in the design of the receiver to
provide some protection against interfering signals. Although quantitative measures of resistance to RF interference, such as jammer-to-signal ratio (J/S) measured in decibels (dB) do exist, these values are very specific to a user's equipment and the signal environment in which it is operating. Therefore, no meaningful specifications for GPS as a complete system can be given.
Augmentations and Enhancements
Many techniques and technical systems designed to improve the capabilities of the basic GPS have been proposed, are under development, or are already in operational use. These techniques range from the differential augmentation of the basic system, to software and hardware enhancements within GPS user equipment, to the integration of GPS user equipment with another navigation/positioning system. Examples of each of these major areas of GPS improvement are discussed below.
Differential GPS (DGPS) is the most widely used method of GPS augmentation and can significantly improve the accuracy, integrity, and availability of the basic GPS. In fact the term ''augmentation" has almost become synonymous with DGPS. DGPS is based upon knowledge of the highly accurate, geodetically surveyed location of a GPS reference station, which observes GPS signals in real time and compares their ranging information to the ranges expected to be observed at its fixed point. The differences between observed ranges and predicted ranges are used to compute corrections to GPS parameters, error sources, and/or resultant positions. These differential corrections are then transmitted to GPS users, who apply the corrections to their received GPS signals or computed position. Figure C-8 further illustrates this concept.
Depending on the user application, DGPS reference stations can be permanent, elaborate installations or small, mobile GPS receivers that can be moved to various well-surveyed locations. The equipment used to broadcast differential corrections, the type of radio datalink used, and the size of the geographic area covered by the DGPS system, also vary greatly with the application. No matter what type of system is used, however, the navigation and positioning capabilities that will be available to any DGPS user within the covered area will be much better than what is available from a stand-alone GPS receiver using either the standard positioning service or the precise positioning service.35
Carrier Phase (Interferometric) GPS
In addition to the use of C/A-code, Y-code, or both as measurements of pseudorange for obtaining a position solution, many GPS receivers also measure the L-band carrier phase itself. This enhancement technique can produce very high precision measurements, sometimes as good as 1 to 5 millimeters and, thus, is valuable for high- performance applications. The carrier phase data is used almost exclusively in an interferometric mode, where the phase data from two receivers are processed together to solve for the baseline between them. This eliminates atmospheric errors, and when combined with DGPS, can result in sub-centimeter positioning accuracies.
The difficulty with using carrier phase tracking is the necessity to solve for an unknown quantity termed the integer or cycle ambiguity. Reliable techniques for using carrier phase data in static surveying applications have existed, however, since the mid 1980s. More recently, ambiguity resolution techniques adapted to dynamic applications such as aircraft and ship navigation have also been developed. The success of these new algorithms hinges on the ambiguity resolution technique. One very effective technique, known as wide-laning, relies on carrier phase measurements from both the L1 and L2 frequencies.36
Multi-channel GPS receivers have recently been developed that take advantage of L1 and L2 wide-laning to resolve carrier phase cycle ambiguity by squaring the L2 signal or cross correlating L1 and L2 within a single receiver. The term "codeless" has been associated with these receivers because, as with earlier carrier phase techniques using two receivers, knowledge of the Y-code itself is not required. 37
A "pseudolite" or pseudo-satellite is a land-based GPS transmitter capable of generating a signal similar to that of an actual GPS satellite. This signal can be received by a user's GPS receiver without the need for additional frequency reception capability. Pseudolites can improve accuracy, integrity, availability, and continuity of service by simply increasing the number of satellite signals available to the receiver. Adding a differential correction to the broadcast signal makes pseudolites even more effective. Like GPS satellites, however, a pseudolite is only effective if it is within the line of sight of a GPS receiver. The signal power of a pseudolite must also be carefully adjusted to avoid interfering with actual GPS signals.
Receiver Autonomous Integrity Monitoring (RAIM)
Receiver Autonomous Integrity Monitoring (RAIM), as the name implies, is a method to enhance the integrity of a GPS receiver without requiring any external
augmentations. RAIM algorithms rely on redundant GPS satellite measurements as a means of detecting unreliable satellites or position solutions. All RAIM approaches look for inconsistencies in either the raw measurements or in the position solutions derived from these measurements. RAIM techniques are generally most effective when six or more satellites are in view of the receiver. This means that RAIM alone is not always the best way to improve GPS integrity, and other solutions are often required.38
Combined Use of GPS and GLONASS
GLONASS is often discussed as a potential means of augmenting the basic capabilities of GPS by providing additional ranging signals to a user, and integrated GPS/GLONASS receivers are available from a limited number of suppliers. GLONASS, or Global Navigation Satellite System, which is operated and managed by the military of the former Soviet Union, consists of three segments just as GPS does. The GLONASS space segment also is designed to consist of 24 satellites, but these satellites are to be arranged in three 64.8º orbital planes 19,100 kilometers (11,870 miles) above the Earth, rather than six planes. The full GLONASS constellation is currently scheduled to be completed in 1995.39
GLONASS differs most from GPS in the way that the user segment differentiates one satellite from another. Instead of each satellite transmitting a unique PRN code as GPS satellites do, GLONASS satellites all transmit the same PRN code on different channels or frequencies.40 All of these frequencies, however, are in the L-band spectrum near eitherthe GPS L1 or L2 signal, which simplifies the task of designing integrated receivers. There are still two additional differences between the two systems that must be taken into consideration by combined receiver designers. First, GPS and GLONASS use different time standards for system synchronization. GPS utilizes UTC (Coordinated Universal Time) maintained by the U.S. Naval Observatory (UTC[USNO]), whereas GLONASS uses the UTC standard kept in the former Soviet Union (UTC[SU]. Discrepancies between these two time scales can reach tens of microseconds, which is significant for systems that keep time with better than 1 microsecond accuracy. Secondly, GPS and GLONASS use different coordinate systems. GLONASS positioning is based on the Soviet Geodetic System (SGS 85), while GPS uses the World Geodetic System (WGS 84) for position determination. Discrepancies between these coordinate systems exist, and must be corrected by combined receivers.
GPS/Inertial Navigation System (INS) Integration
The present GPS can provide a suitably equipped user with a position, velocity, and time solution whose errors are generally smaller than those of most inertial navigation systems (INS).41 This performance is achieved in all weather, at any time of day, and under a wide range of signal availability and vehicle dynamics. Nevertheless, the integration of GPS with INS can provide a more robust and possibly more accurate navigation service than is possible with stand-alone sensors. In particular, integration may be the only way to achieve the following:
- Maintain a specified level of navigation performance during outages of GPS satellite reception.
- Reduce the random noise component of errors in the GPS navigation solution.
- Maintain the availability of a GPS solution in the presence of higher vehicle dynamics and radio interference than can be tolerated by GPS alone.
The technical basis for considering GPS/INS integration is the complementary nature of the navigation errors for each system operating in a stand-alone mode. The GPS solution is relatively noisy, but stays within its statistical accuracy boundaries (either CEP or 2 drms boundaries) over time. In contrast, inertial navigation errors are not noisy, but grow in proportion to the duration of a mission and the acceleration experienced by the system. One expects that an integrated navigation solution would perform like an inertial navigator whose errors were bounded by the GPS errors. Additional benefits as noted above are also achievable with more complex integration approaches.42
GPS and Loran-C
Loran-C, originally developed by the DOD, is a low frequency (90-110 KHz) radionavigation system that is used by the civil maritime and civil aviation communities. Chains of Loran-C transmitting stations cover the continental U.S. and the coastline of Alaska, as well as the coastlines of many other nations. A Loran-C receiver normally
determines its position by computing lines of position based on radio pulse transmissions from three stations within a chain.43
As with GPS/INS integration, the addition of another navigation system provides redundancy. If GPS signal reception is poor due to a lack of satellites in view or due to signal interference, an integrated system can maintain a specified level of navigation performance using only Loran-C. The system integrity and availability of a GPS/Loran-C system is also improved over GPS alone. A study focused on integrity and availability requirements for aviation non-precision approaches has shown that RAIM performance is significantly improved by the presence of Loran-C signals, and availability improves from 99 percent for a GPS receiver with RAIM and a barometric altimeter to 99.7 percent for a GPS/Loran-C receiver with RAIM.44
The integration of Loran-C with DGPS has also been proposed as a potential means of improving both integrity and accuracy. Integrity information and differential corrections could potentially be broadcast on Loran-C signals from existing ground-based transmitter stations to GPS/Loran-C receivers. If this proposal proves to be technically feasible, the entire continental United States and Western Europe could potentially be provided with DGPS capability using Loran-C signals.45
Permanent Differential GPS Augmentations
It is impossible to estimate the number of temporary DGPS networks in use around the world at any given time because of the ease with which they can be established, utilized, and then removed. GPS users such as surveyors and resource monitors may go through this process several times in one day. It is possible, however, to describe some of the permanent DGPS services that are currently operating or are under development by the U.S. government, state and local governments, foreign governments, and the private sector.
U.S. Government-Supported Differential GPS
There are currently at least a dozen U.S. federal agencies that operate or plan to operate permanent DGPS networks.46 Three agencies in particular, the FAA (Federal Aviation Administration), the U.S. Coast Guard, and NOAA (National Oceanic and Atmospheric Administration), plan to provide nationwide DGPS services. Each of these three programs is described briefly below.
FAA Wide-Area and Local-Area DGPS Concepts
The FAA plans to improve the accuracy, integrity, and availability of GPS to levels which support flight operations in the National Airspace System from en route navigation through Category I precision approaches by using a wide-area DGPS concept known as the Wide-Area Augmentation System (WAAS).47 In June 1994, the FAA released an RFP (request-for-proposal) for the WAAS that calls for a ground-based communications network and several geosynchronous satellites to provide nationwide coverage. The ground-based communications network will consist of 24 wide-area reference stations, two wide-area master stations, and two satellite uplink sites. Differential corrections and integrity data derived from the ground-based network, as well as additional ranging data, will be broadcast to users from the geostationary satellites using an "L1-like" signal with a frequency of 1575.42 MHz.48 The RFP calls for the WAAS to be in place by the end of 1997.
Local-area DGPS systems are also being considered by the FAA to support landing operations beyond Category I. The airline industry estimates that there are approximately 120 runways in the United States that will require this type of service through the year
U.S. Coast Guard DGPS Service
The Coast Guard currently is establishing a DGPS network that will for the first time, meet the extremely accurate navigation requirements of commercial and recreational mariners in our nation's environmentally sensitive harbor and harbor approach areas.51 When fully operational in 1996, the system is expected to reduce the number of navigation-related grounding and collision incidents by 50 percent over existing navigation methods. A total of 50 reference stations will be installed at sites along the coastal United States, the Great Lakes, Puerto Rico, Alaska, and Hawaii. Each site will use a marine radiobeacon to broadcast differential corrections and integrity information in the RTCM SC-104 message format.52 The radiobeacon signals can be received by a device about the size of a computer modem with an antenna similar in size to one used by a GPS receiver. By applying the broadcast differential corrections to a GPS position solution in real-time, a user can achieve navigational accuracy as good as 1.5 meters (2 drms) up to 460 kilometers (250 nautical miles) from the reference station.53
The Coast Guard hopes to eventually meet the stringent accuracy requirements of inland waterway navigation with their DGPS network as well. In order to achieve this goal, the Coast Guard has entered into a Memorandum of Agreement with the Army Corps of Engineers that will expand DGPS service throughout the navigable waters of the Mississippi River and its tributaries.54
NOAA Continuously Operated Reference Stations
The goal of NOAA's Continuously Operated Reference Station (CORS) program is to implement a single, consistent set of federally funded DGPS reference stations that would provide GPS data to all users in a single common format with continuous monitoring of
reference station position. Each reference site would measure coded and codeless L1 and L2 data. This data would then be sent to the CORS Central Facility, where it can be stored on computer disc. Users could then access this data electronically within one hour after it has been measured, providing post-processed positioning accuracy of 5 to 10 centimeters. All Coast Guard, Army Corps of Engineers, and FAA reference stations that are part of the DGPS services described above are designed to be CORS-compatible. In addition, a recent technical report to the Secretary of Transportation has recommended that all future federally provided DGPS reference stations should comply with the CORS standard.55
State and Local Government DGPS
A number of state and local governments either have established or plan to establish permanent DGPS reference sites. For example, Riverside County, California, has established two continuously operating, permanent DGPS reference stations as part of the Permanent GPS Geodetic Array. This array, whose participants also include federal agencies, state agencies, other local government agencies, and universities, is used primarily for earthquake monitoring and, perhaps, eventually will be used for earthquake prediction. Riverside County engineers and surveyors, however, also use the array for typical day-to-day surveying applications.
Differential Systems Supported by Foreign Governments and International Organizations
Foreign governments and public sector international organizations are actively developing and utilizing differential GPS networks. Several examples designed to support aviation, maritime, and survey/scientific applications are discussed below.
Maritime DGPS Services
Many countries are currently operating, prototyping, or planning maritime DGPS services similar to the U.S. Coast Guard's. The low cost, combined with the absence of any international frequency allocation problems makes these systems practical for all nations. Since most sea coasts and ports have medium-frequency radiobeacons for direction finding, DGPS services can be added quite simply with the purchase and installation of off-the-shelf GPS equipment. The International Association of Lighthouse Authorities (IALA) coordinates the assignment of frequencies and DGPS reference station identifying numbers,
and is compiling information on maritime DGPS broadcasts worldwide. Currently Sweden, Finland, The Netherlands, Denmark, Iceland, and Germany have complete or nearly complete coastal coverage. Several other countries have prototype or demonstration services including Australia, Canada, China, Norway, and Poland. India and South Africa are planning maritime DGPS services.
International Participation in the FAA's WAAS
In order to eventually develop the WAAS into a Global Navigation Satellite System (GNSS) that is useful to aircraft anywhere in the world, the FAA is encouraging other nations to participate in the program at any level they feel comfortable with.56 Nations involved at the lowest level will simply utilize the GPS-like WAAS signals without any contribution to the system in the form of ground based wide-area reference stations. Participation at a higher level would involve the installation of wide-area reference stations and possibly wide-area master stations within the sovereign territory of a nation. Even higher levels of involvement are possible if a nation is willing to provide a geostationary satellite for the space segment of the system. Several countries have expressed an interest in WAAS participation, including Canada, Australia, New Zealand, and Japan.57
Inmarsat (the International Maritime Satellite Organization), a not-for-profit international organization that provides global mobile satellite services to the maritime, land- mobile, and aviation markets, has firm plans to augment GPS by placing a navigation payload on board its third generation geostationary communications satellites. Plans call for this payload to broadcast GPS and GLONASS integrity information, ranging information, and wide-area differential corrections on a "GPS-like" L1 signal centered at 1575.42 MHz. These satellites and their navigation payloads may form the nucleus of the WAAS space segment if the winning team of contractors chooses to use them. Future Inmarsat plans include the possible development of a fully civil GNSS based on light satellite (lightsat) navigation payloads placed in intermediate circular orbits and geostationary orbits.58
The International GPS Service for Geodynamics
The International GPS Service for Geodynamics (IGS) is a network of more than 50 globally distributed GPS tracking sites that has been established by NASA and other organizations from various nations in order to support geodetic and geophysical research activities. 59 Rather than provide real-time differential corrections to users, the tracking sites are used to produce post-processed GPS orbits, or ephemerides, with an accuracy of 10 to 30 centimeters. Orbits are processed at the IGS central bureau at NASA's Jet Propulsion Laboratory in Pasadena, California, and at other sites within the United States and around the globe. These orbits are typically available on the Internet within a few days after they have been processed. 60
Private Sector DGPS Services
There are a number of private sector enterprises that now offer differential GPS services to the public at various levels of accuracy and at a wide range of prices. These systems use both space-based and land-based datalinks that are encrypted to provide access to only paying customers. Brief summaries of four of these services are provided below.61
Racal Survey of Surrey England (U.K.) has developed a worldwide, space-based differential GPS service known as SkyFix for use in a number of surveying applications. The ground segment of the SkyFix system currently consists of over 25 reference stations around the globe that determine differential corrections that are sent to users via geostationary satellite. The four satellites currently in use are owned and operated by Inmarsat, and provide worldwide coverage except for the polar regions. Users access the differential corrections broadcast in L-band (1530-1545 MHz) using either an Inmarsat terminal or a specialized SkyFix terminal. Racal Survey advertises a positioning accuracy of 3 to 5 meters using this system.
John E. Chance & Associates, Inc. (A member of the Fugro Group of Companies)
John E. Chance & Associates, Inc, now affiliated with the Dutch Fugro Group, provides DGPS services to North America and much of the rest of the world with a system known as Starfix II.62 Starfix II systems operate throughout the world by sending differential corrections from each of the reference sites to a central network control center using leased telephone lines, communications satellites, or both. Differential corrections are broadcast to users via L-Band and C-Band geostationary communications satellites and are received by user equipment that consists of a small (3.8 cm high, 7.6 cm diameter) omnidirectional antenna and a signal downconverter (5.0 x 7.6 x 25.4 cm in size).63 John E. Chance advertises real-time positioning accuracies of 53 centimeters (2 drms).
John E. Chance also provides continuous DGPS coverage to all of the continental United States and most of North America using the OMNISTAR system. The OMNISTAR system is essentially the same as Starfix II, except that differential corrections are broadcast to OMNISTAR users in RTCM SC-104 format, and an ionospheric model that takes the user's location into consideration is utilized in determining the corrections .64 This approach is a convenient mechanism for providing differential corrections to users with a variety of GPS receivers.
John E. Chance will also provide DGPS correction data via satellite to ACCQPOINT, an FM subcarrier-based DGPS service based on an alliance between Lecia of Torrance, California, and CUE Network based in Irvine, California. ACCQPOINT plans to eventually install receivers for the John E. Chance data at all 500 radio stations that currently are part of CUE's North American paging network. The pseudorange corrections received at the stations will then be broadcast to users within a reception range of 35 to 85 miles (56 to 136 kilometers) using mobile broadcast service (MBS) technology originally developed in Europe. MBS technology allows conventional FM radio broadcasts to carry digital data, such as differential corrections, by modulating the data on an inaudible subcarrier frequency of 57 KHz at approximately 1100 bits per second. The FM subcarrier signal is received by equipment that is only slightly larger than a standard pager and provides users with an advertised accuracy of approximately 1.5 meters.
Differential Corrections Inc.
Differential Corrections Inc. (DCI), of Cupertino, California, also provides DGPS services to its subscribers using a 57 KHz FM subcarrier. DCI utilizes radio data system (RDS) technology, also originally developed in Europe, that transmits data at 1187.5 bits per second. As of August, 1994, DCI had 46 FM stations operating in their DGPS network, with another 51 stations scheduled to begin service, effectively covering every major population area in the United States. DCI also operates in several foreign countries and hopes to expand their international service.
The differential correction broadcast to DCI users is determined by reference stations located at each of the FM stations within the network. As with ACCUPOINTs service, the user equipment required to receive the FM subcarrier frequency resembles a typical pager. DCI provides its customers with three levels of accuracy at three different prices. The premium service has an advertised accuracy of 1 meter (2 drms), the intermediate service provides 5-meter accuracy, and the basic service gives a customer 10-meter accuracy.