C H A P T E R 3▼ ▼ ▼

GPS System SegmentsL. F. WiederholtIntermetrics, Inc.

E. D. KaplanThe MITRE Corporation

3.1 OVERVIEW OF THE GPS SYSTEM

The Global Positioning System (GPS) is comprised of three segments: satellite constel-lation, ground control/monitoring network, and user receiving equipment. FormalGPS Joint Program Office (JPO) programmatic terms for these components arespace, operational control, and user equipment segments, respectively. The satelliteconstellation contains the satellites in orbit that provide the ranging signals and datamessages to the user equipment. The operational control segment (OCS) tracks andmaintains the satellites in space. The OCS monitors satellite health and signal integrityand maintains the orbital configuration of the satellites. Furthermore, the OCSupdates the satellite clock corrections and ephemerides as well as numerous otherparameters essential to determining user position, velocity, and time (PVT). Lastly,the user receiver equipment performs the navigation, timing, or other related func-tions (e.g., surveying). Elaboration on each of the system segments is provided next.

3.1.1 GPS Satellite Constellation

The satellite constellation consists of the nominal 24-satellite constellation. Thesatellites are positioned in six Earth-centered orbital planes with four satellites in

59

60 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

each plane. The nominal orbital period of a GPS satellite is one-half of a siderealday or 11 hr 58 min [1]. The orbits are nearly circular and equally spaced aboutthe equator at a 60° separation with an inclination relative to the equator of nominally55°. Figure 3.1 depicts the GPS constellation. The orbital radius (i.e., nominaldistance from the center of mass of the Earth to the satellite) is approximately 26,600km. This satellite constellation provides a 24-hr global user navigation and timedetermination capability.

Figure 3.2 presents the satellite orbits in a planar projection referenced to theepoch time of 0000 hr July 1, 1993 UTC(USNO). Thinking of the orbits as a ‘‘ring,’’this figure opens each orbit and lays it flat on a plane. Similarly for the Earth’sequator, it is like a ring that has been opened and laid on a flat surface. The slopeof each orbit represents its inclination with respect to the Earth’s equatorial plane,

Figure 3.1 GPS satellite constellation.

GPS System Segments 61

Figure 3.2 GPS constellation planar projection.

which is nominally 55°. The orbital plane locations with respect to the Earth aredefined by the longitude of the ascending node while the location of the satellitewithin the orbital plane is defined by the mean anomaly. The longitude of theascending node is the point of intersection of each orbital plane with the equatorialplane. The Greenwich meridian is the reference point or point where the longitudeof the ascending node has the value of zero. Mean anomaly is the angular positionof each satellite within the orbit with the Earth’s equator being the reference orpoint with a zero value of mean anomaly. It can be observed that the relative phasingbetween most satellites in adjoining orbits is approximately 40°.

Several different notations are used to refer to the satellites in their orbits. Onenomenclature assigns a letter to each orbital plane (i.e., A, B, C, D, E, and F) witheach satellite within a plane assigned a number from 1 to 4. Thus, a satellite referencedas B3 refers to satellite number 3 in orbital plane B. A second notation used is a

62 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

NAVSTAR satellite number assigned by the U.S. Air Force. This notation is in theform of space vehicle number (SVN) 11 to refer to NAVSTAR satellite 11. The thirdnotation represents the configuration of the pseudorandom (PRN) code generatorsonboard the satellite. These PRN code generators are configured uniquely on eachsatellite, thereby producing unique versions of both C/A-code and P(Y)-code. Thus,a satellite can be identified by the PRN codes that it generates.

3.1.2 Operational Control Segment (OCS)

The OCS has responsibility for maintaining the satellites and their proper functioning.This includes maintaining the satellites in their proper orbital positions (called stationkeeping) and monitoring satellite subsystem health and status. The OCS also moni-tors the satellite solar arrays, battery power levels, and propellant levels used formaneuvers and activates spare satellites (if available). The OCS updates each satel-lite’s clock, ephemeris, and almanac and other indicators in the navigation messageonce per day or as needed. The ephemeris parameters are a precise fit to the GPSsatellite orbits and are valid only for a time interval from 4 to 6 hr depending onthe time from the last control segment upload based on the once-per-day normalupload schedule.

Depending on the satellite version, the navigation message data can be storedfor a minimum 14 days to a maximum of 210 days duration in intervals of 4 or 6hr. The almanac is a reduced precision subset of the ephemeris parameters. Thealmanac consists of 7 of the 15 ephemeris orbital parameters. Almanac data is usedto predict the approximate satellite position and aid in satellite signal acquisition.

Furthermore, the OCS resolves satellite anomalies, controls SA and AS (seeSections 1.3.1 and 4.1.1), and makes pseudorange and delta range measurementsat the remote monitor stations to determine satellite clock corrections, almanac, andephemeris.

To accomplish the above functions, the control segment is comprised of threedifferent physical components: the master control station (MCS), monitor stations,and the ground antennas. Each of these facilities is described in the following sections.

3.1.2.1 OCS Operations

An overview of control segment operations is presented in Figure 3.3. The MCS isthe center of the control segment operations and is located at Falcon Air Force Base,Colorado Springs, CO. The monitor stations passively track the GPS satellites asthey pass overhead by making pseudorange and delta range measurements. Thesemeasurements are made using both the L1 and L2 GPS satellite downlink frequencies.(L1 and L2 and their associated modulation formats are described in Section 4.1.1.)This raw data, in addition to the received navigation message and local weather

GP

SSystem

Segments

63Figure 3.3 Overview of the control segment operations.

64 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

data, is transmitted to the MCS via the Defense Satellite Communications Systemand other ground communications systems.

The MCS processes data from the monitor stations for satellite navigationpayload control. Data from all the monitor stations is used to form satellite clockcorrections, ephemeris and almanac data for each satellite. This processing isdescribed below in Section 3.1.2.4. The MCS also monitors the configuration statusof satellites and ground stations. Satellite processor diagnostics can be requestedand the satellite clock can be updated. The geographic distribution of the controlsegment facilities is depicted in Figure 3.4. A backup master control station is locatedat Gaithersburg, MD. This backup MCS is a temporary contractor facility. At thetime of this writing, there is a DOD effort underway to move it to VandenbergAFB, CA.

3.1.2.2 Monitor Stations Description

The monitor stations form the data collection component of the control segment.A monitor station contains a dual-frequency (L1/L2) GPS receiver that continuouslymakes pseudorange and delta range measurements to each satellite in view. Thelocation of the phase center of the receiver’s antenna is precisely known. The monitorstation also contains two cesium clocks referenced to GPS system time. Pseudorangeand delta range measurements made to each satellite in view by each monitor stationreceiver update the master control station’s precise Kalman filter statistical estimateof each satellite’s position, velocity, and timing (PVT).

The satellite transmissions are refracted and, hence, delayed by both the iono-sphere and the troposphere. (Elaboration on these effects is found in Section 7.1.2.)The monitor station receiver dual-frequency measurements enable the MCS to deter-mine the ionospheric delay for satellites within the monitor station satellite field ofview. Temperature, barometric pressure, and humidity data are provided to the MCSby U.S. government weather services to aid in tropospheric delay determinationwithin the region of each monitor station.

There are five monitor stations located at Colorado Springs, Kwajalein, Ascen-sion Island, Hawaii, and Diego Garcia. Within the near future, a monitor stationwill be operational at Cape Canaveral, FL. A photograph of the monitor stationand the ground uplink antenna at Diego Garcia is presented in Figure 3.5. Thesefacilities are unmanned and provide approximately 92% tracking coverage of theGPS satellites [2]. That is, the satellites are not continuously visible to the monitorstations. As stated above, all collected data are transmitted to the MCS for processing.

3.1.2.3 Ground Uplink Antenna Facility Description

The ground uplink antenna facility provides the means of commanding and control-ling the satellites and uploading the navigation messages and other data. A ground

GP

SSystem

Segments

65

Figure 3.4 Geographic distribution of the control segment facilities.

66 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Figure 3.5 Photo of monitor station and ground uplink antenna at Diego Garcia.

uplink antenna facility stores and uploads what is called the TT&C (telemetry,tracking, and command) data. A unique TT&C data set (which includes the naviga-tion message) is prepared by the MCS for each satellite. This data is forwarded tothe ground antenna from the MCS and stored until a particular satellite is in view.Once in view, an S-band data communication uplink is used to transmit data to thesatellite for forwarding to the satellite’s navigation processor. Ground antennas arecollocated with monitor stations at Ascension Island, Kwajalein, Diego Garcia, andCape Canaveral. These locations have been selected to maximize satellite coverage.

3.1.2.4 Master Control Station Processing

As mentioned above, the MCS performs a multitude of functions to support theoperation of GPS as a system. One principal activity is the processing of the datacollected at the remote monitor stations to form estimates of the GPS satellite clock,ephemeris, and almanac data. Many steps are involved in this processing [3, 4], withthe major steps outlined here. The detailed processing is continually evolving basedon control segment experience. With the data collected from the remote monitorstations, the processing starts with the correction of the pseudorange measurementsfor tropospheric and ionospheric delays. The corrected pseudorange and delta rangemeasurements from all the remote monitor stations are then processed by an epoch-state Kalman filter to form a precise satellite ephemeris and clock offset solution.An epoch-state filter is a filter that maintains its estimates at a time different thanthe measurement times. For this filter, the state is maintained at the time of applicabil-ity of the ephemeris while measurements are collected at different times. This filter

GPS System Segments 67

is updated every 15 min with the satellite positions computed in the ECEF Cartesiancoordinate system. This process provides an accurate estimate of a satellite’s ephem-eris and clock offset at the time of the data collection. To be of use to a GPS userat some later time, these states must be predicted forward. The satellite position andclock corrections are predicted forward in time using precise models of the spacecraftand its environment. This model, called force integration, is a set of differentialequations describing the dynamic behavior of the satellite. The force integrationmodel includes the significant forces operating on the space vehicle over the predictioninterval that have been identified as the gravity field perturbations, the sun andmoon third-body mass attractions, spacecraft solar radiation pressure, and EarthUT1 and polar motion perturbations.

The prediction interval described above is subdivided into either 4- or 6-hrtime intervals since the start of the prediction time. For each subdivided interval,the predicted satellite Cartesian position data is transformed to 15 orbital elementsusing a least squares fit algorithm. (These elements are described in Section 2.3) Theephemeris data is thus valid only over that interval. The almanac and clock dataare also formed from this accurately predicted data. From this least squares fit data,the content of the navigation message in accordance with [5] is formed by scalingand truncating the ephemeris, almanac, and clock data to the described format. Thefitted data set for each time interval is then uploaded to the satellites’ navigationpayload for storage and transmission to the user.

Another important element of the MCS processing is monitoring the reliabilityof the system. The control segment must take meticulous care to ensure that allclock and ephemeris data uploads and other signal transmissions are correct. Thismonitoring is done principally through the MCS data processing. The MCS computesthe upload navigation messages, maintains an image of that satellite message forcomparison, monitors the uploading of data, and verifies the correct transmissionby the satellite. The OCS also monitors satellite L-band signal behavior and issuesan alarm to MCS personnel within 60 sec of a detected failure.

3.1.3 User Receiving Equipment

The user receiving equipment, typically referred to as a ‘‘GPS receiver,’’ processesthe L-band signals transmitted from the satellites to determine user PVT. There hasbeen a significant evolution, almost revolution, in the technology of GPS receivingsets, paralleling that of the electronics industry in general. The move has been fromanalog to digital solid state devices and surface-mount technology wherever feasible.The initial receiving sets manufactured in the mid-1970s as part of the conceptvalidation phase were principally analog devices for military application, which werelarge, bulky, and heavy. With today’s technology, a GPS receiver of comparable ormore capability typically weighs a few pounds (or ounces) and occupies a small

68 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

volume. The smallest sets today are those of ‘‘wrist watch’’ size while probably thelargest is a naval shipboard unit with a footprint of 1,550 cm2 (232.5 in2) and weightof 70 lb (mass of 31.8 kg). Selection of a GPS receiver depends on the user’sapplication (e.g., civilian versus military, platform dynamics, etc.). Following adescription of a typical receiver’s components, selection criteria are addressed.

3.1.3.1 GPS Set Characteristics

A block diagram of a GPS receiving set is shown in Figure 3.6. The GPS set consistsof five principal components: antenna, receiver, processor, input/output (I/O) devicesuch as a control display unit (CDU), and a power supply.

Antenna

Satellite signals are received via the antenna, which is right-hand circularly polarized(RHCP) and provides near hemispherical coverage. Typical coverage is 160° withgain variations from about 2.5 dBic at zenith to near unity at an elevation angle of10°. (The RHCP antenna unity gain also can be expressed as 0 dBic = 0 dB withrespect to an isotropic circularly polarized antenna.) Below 10°, the gain is usually

Figure 3.6 Principle GPS receiver components.

GPS System Segments 69

negative. Since the satellite signals are RHCP, a conical helix antenna or variation issuitable. GPS receivers that track P(Y)-code on both L1 and L2 need to accommodate20.46-MHz bandwidths on both frequencies. If the set only tracks C/A-code on L1,the antenna (and receiver) must have a bandwidth of at least 2.046 MHz. Antennadesigns vary from helical coils to thin microstrip (i.e., patch) antennas. High-dynamicaircraft prefer low profile, low air resistance patch antennas whereas land vehiclescan tolerate a larger antenna. Antenna selection requires evaluation of parameterssuch as antenna gain pattern, available mounting area, aerodynamic performance,multipath performance, and stability of the electrical phase center of the antenna [6].Another issue regarding antenna selection is the need for resistance to interference. (Inthe context of this discussion, any electronic emission whether ‘‘friendly’’ or hostilethat interferes with the reception and processing of GPS signals is considered aninterferer.) Some military aircraft employ beam steering or adaptive nulling arraysto resist interference. Beam steering techniques electronically concentrate the antennagain in the direction of the satellites to maximize link margin. An adaptive nullingarray is electronically steerable and creates nulls in the antenna pattern that are inthe direction of the interferer [7] (see also Section 6.1.2.5).

Receiver

Chapter 5 provides a detailed description of receiver signal acquisition and trackingoperation; however, some high-level aspects are described herein to aid ourdiscussion.

Two basic receiver types exist today: (1) those that track both P(Y)-code andC/A-code and (2) those that only track C/A-code. PPS users generally employ setsthat track P(Y)-code on both L1 and L2. These sets initiate operation with receiverstracking C/A-code on L1 tracking and then switch to tracking P(Y)-code on eitherL1 or L2. Y-code tracking occurs only with the aid of cryptographic equipment. (If thesatellite signal is encrypted and the receiver does not have the proper cryptographicequipment, the receiver generally defaults to tracking C/A-code on L1.) On the otherhand, SPS users employ sets that track the C/A-code exclusively on L1 since that isthe only frequency that the C/A-code is generally broadcast on. Of these two basicreceiver types, there are other variations such as codeless L2 tracking receivers,which track the C/A-code on L1 and carrier phase of both the L1 and L2 frequencies.Utilizing the carrier phase as a measurement observable enables centimeter-level(or even millimeter-level) measurement accuracy. Carrier-phase measurements aredescribed extensively in Section 8.3.

Most receivers have multiple channels whereby each channel tracks the trans-mission from a single satellite. A simplified block diagram of a multichannel genericSPS receiver is shown in Figure 3.7. The received RF CDMA satellite signals areusually filtered by a passive bandpass prefilter to reduce out-of-band RF interference.

70 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Figure 3.7 Generic SPS receiver.

This is normally followed by a preamplifier. The RF signals are then downconvertedto an intermediate frequency (IF). In a typical modern GPS receiver design, the IFsignals are sampled and digitized by an analog to digital (A/D) converter. The A/Dsampling rate is typically eight to twelve times the PRN code chipping rate(1.023 MHz for L1 C/A-code and 10.23 MHz for L1 and L2 P(Y)-code.) Theminimum sampling rate is twice the stopband bandwidth of the codes to satisfy theNyquist criterion. For L1 C/A-code only sets, the stopband bandwidth may beslightly greater than 2 MHz. On the other hand, the stopband bandwidth is slightlymore than 20 MHz for P(Y)-code sets. Oversampling reduces the receiver sensitivityto A/D quantization noise, thereby reducing the number of bits required in theA/D converter. The samples are forwarded to the digital signal processor (DSP). TheDSP contains N parallel channels to simultaneously track the carriers and codesfrom up to N satellites. (N generally ranges from 5 to 12 in today’s receivers.) Eachchannel contains code and carrier tracking loops to perform code and carrier-phasemeasurements as well as navigation message data demodulation. The channel maycompute three different satellite-to-user measurement types: pseudoranges, delta

GPS System Segments 71

ranges (sometimes referred to as delta pseudorange), and integrated Doppler,depending on the implementation. The desired measurements and demodulatednavigation message data are forwarded to the processor.

Navigation/Receiver Processor

A processor is generally required to control and command the receiver through itsoperational sequence, starting with channel signal acquisition and followed by signaltracking and data collection. (Some GPS sets have an integral processing capabilitywithin the channel circuitry to perform these signal processing functions.) In addition,the processor may also form the PVT solution from the receiver measurements. Insome applications, a separate processor may be dedicated to the computation ofboth PVT and associated navigation functions. Most processors provide an indepen-dent PVT solution on a 1-Hz basis. However, receivers designated for autolandaircraft precision approach and other high-dynamic applications normally requirecomputation of independent PVT solutions at a minimum of 5 Hz. The formulatedPVT solution and other navigation-related data is forwarded to the I/O device.

Input/Output Device

The I/O device is the interface between the GPS set and the user. I/O devices are oftwo basic types: integral or external. For many applications, the I/O device is aCDU. The CDU permits operator data entry, displays status and navigation solutionparameters, and usually accesses numerous navigation functions such as waypointentry, time-to-go, etc. Most handheld units have an integral CDU. Other installations,such as those onboard an aircraft or ship, may have the I/O device integrated withexisting instruments or control panels. In addition to the user and operator interface,applications such as integration with other sensors (e.g., INS) require a digital datainterface to input and output data. Common interfaces are ARINC 429, MIL-STD-1553B, RS-232, and RS-422.

Power Supply

The power supply can be either integral, external, or a combination of the two.Typically, alkaline or lithium batteries are used for integral or self-contained imple-mentations, such as handheld portable units; whereas an existing power supply isnormally used in integrated applications, such as board-mounted receivers installedinside personal computers. Airborne, automotive, and shipboard GPS set installationsnormally use platform power but typically have built-in power converters (ac to dcor dc to dc) and regulators. There usually is an internal battery to maintain data

72 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

stored in volatile random access memory (RAM) integrated circuits (ICs) and tooperate a built-in timepiece (date/time clock) in the event platform power isdisconnected.

3.1.3.2 GPS Receiver Selection

At the time of this writing, there are over 50 GPS set manufacturers producing over200 different GPS set versions [8]. GPS receiver selection is dependent on userapplication. The intended application strongly influences receiver design, construc-tion, and capability. For each application, numerous environmental, operational,and performance parameters must be examined. A sampling of these parameters isprovided below:

• Shock and vibration requirements, temperature and humidity extremes, as wellas atmospheric salt content.

• If the receiver is to be used by government and/or military personnel, PPSoperation may be required. PPS operation usually dictates that a dual-frequencyset with a cryptographic capability is needed.

• The necessary independent PVT update rate must be determined. As an example,this rate is different for aircraft precision approach than for marine oil tankerguidance.

• Under what type of dynamic conditions (e.g., acceleration, velocity) will theset have to operate? GPS sets for fighter aircraft applications are designed tomaintain full performance even while experiencing multiple ‘‘g’s’’ of accelera-tion, whereas sets designated for surveying are not normally designed for severedynamic environments.

• Is a differential GPS (DGPS) capability required? (DGPS is an accuracy enhance-ment technique covered in Chapter 8.) DGPS provides greater accuracy thanstandalone PPS and SPS. Most receivers are manufactured with a DGPScapability.

• Does the application require reception of a geostationary satellite-based overlayservice (e.g., INMARSAT) broadcasting GPS and/or GLONASS satellite integ-rity, ranging, and DGPS information? (The INMARSAT overlay is describedin Chapter 11.)

• Waypoint storage capability as well as the number of routes and legs need tobe assessed.

• Does the GPS set have to operate in an environment that requires enhancedinterference rejection capabilities? Chapter 6 describes several techniques toachieve this.

• If the receiver has to be interfaced with an external system, does the properI/O hardware and software exist?

GPS System Segments 73

• In terms of data input and display features, does the receiver require an externalor integral CDU capability. Some aircraft and ships use ‘‘repeater’’ units suchthat data can be entered or extracted from various physical locations. Displayrequirements such as sunlight-readable or night-vision-goggle-compatible mustbe considered.

• Are local datum conversions required, or is WGS-84 sufficient? If so, does thereceiver contain the proper transformations?

• Is portability for field use required?• Economics, physical size, and power consumption must also be considered.

As stated above, these are only a sampling of GPS set selection parameters.One must carefully review the requirements of the user application prior to selectinga receiver. In most cases, the selection will be a tradeoff that requires awareness ofthe impact of any GPS set deficiencies for the intended application.

3.2 SPACE SEGMENT PHASED DEVELOPMENT

The development of the control and space segments has been phased in over manyyears starting in the mid-1970s and is continuing. This development started with aconcept validation phase and has progressed to the production phase. The satellitesassociated with each phase of development are called a block of satellites. Characteris-tics of each phase and block are presented in the following sections.

3.2.1 Characteristics Summary of Satellite Block

Three satellite blocks have been deployed and two more blocks are planned. Theinitial concept validation satellites were called Block (BLK) I. The last remainingprototype BLK I satellite, PRN 12, was disposed of in the fall of 1995. Block IIsatellites are the initial production satellites while Block IIA refers to the upgradedproduction satellites. Blocks I, II, and IIA have been launched and are in service. Atthe time of this writing, Block IIR satellites, called the replenishment satellites, arein production. Block IIF satellites, referred to as the follow-on or sustainment satel-lites, are in the planning stage.

Nine BLK II satellites and fifteen BLK IIA comprise the nominal constellation.Four other BLK IIA satellites have been purchased and will be used to replace failedsatellites or satellites deemed unusable by MCS personnel. When all four of theseextras are deployed, Block IIR satellites will then be used as replacements. BLK IIRsatellites are scheduled for replenishment in the 1997–2004 timeframe. Block IIFsatellite launches are planned for the post-2004 timeframe. A tabulation of theNAVSTAR satellites and their respective orbital locations is contained in Table 3.1.

74 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Table 3.1Satellite Constellation Configuration (as of July 1996)

Satellite NumberBlock PRN SVN Launch Date Orbit

II 14 14 Feb. 89 E1II 02 13 June 89 B3II 16 16 Aug. 89 E3II 19 19 Oct. 89 A4II 17 17 Dec. 89 D3II 18 18 Jan. 90 F3II 20 20 Mar. 90 B2II 21 21 Aug. 90 E2II 15 15 Oct. 90 D2IIA 23 23 Nov. 90 E4IIA 24 24 July 91 D1IIA 25 25 Feb. 92 A2IIA 28 28 April 92 C2IIA 26 26 July 92 F2IIA 27 27 Sept. 92 A3IIA 01 32 Nov. 92 F1IIA 29 29 Dec. 92 F4IIA 22 22 Feb. 93 B1IIA 31 31 Mar. 93 C3IIA 07 37 May 93 C4IIA 09 39 June 93 A1IIA 05 35 Aug. 93 B4IIA 04 34 Oct. 93 D4IIA 06 36 Mar. 94 C1IIA 03 33 Mar. 96 C2

3.2.2 Navigation Payload

The navigation payload is that part of the satellite which is responsible for receptionof data from the OCS, intersatellite ranging (only on the BLK IIR and BLK IIFversions) and the transmission of ranging codes and navigation data to the usercommunity. The navigation payload is only one part of the spacecraft with othersystems being responsible for such functions as attitude control and solar panelpointing. Figure 3.8 is a generic block diagram of a navigation payload. The OCSTT&C function is responsible for uploading data and command and control informa-tion to the satellite. This data is stored in processor memory. The frequency standardssubsystem contains the atomic frequency standards, of which there are currentlytwo cesium and two rubidium standards on each production satellite. (The BLK IIRreplenishment satellites contain one cesium standard and two rubidium standards.)

GPS System Segments 75

Figure 3.8 Satellite navigation payload.

Of these multiple atomic standards, one is designated as the primary and serves asthe timing reference onboard the space vehicle for navigation signal generation andtransmission. The other atomic standards are for redundancy. The code generatorgenerates the C/A-code and P(Y)-codes for modulo-2 addition with the navigationmessage data, which are then sent to the L-band subsystem for transmission to theuser. The L-band subsystem contains the L1/L2 synthesizers and associated antennas.The processor memory also interfaces to the receiver/transmitter for intersatelliteranging on BLK IIR and BLK IIF versions. This receiver/transmitter uses a separateantenna and feed system.

In addition to the GPS navigation payload, other missions have been addedto the space vehicle over the course of time. These non-GPS-related missions haveincluded a nuclear detonation detection and location payload, laser reflectors forsatellite laser ranging (i.e., validation of predicted ephemeris), and free electronmeasurement experiments.

3.2.3 Block I-Initial Concept Validation Satellites

Block I satellites were developmental prototypes to validate the initial GPS concept, soonly eleven satellites were built. The Block I satellites, built by Rockwell International,were launched between 1978 and 1985 from Vandenberg Air Force Base, California.A picture of the Block I satellite is presented in Figure 3.9. The onboard storagecapability was for 14 days of navigation messages. In this satellite version, thenavigation message data was only valid for a 1-hr period. Since there was no onboard

76 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Figure 3.9 BLK I satellite.

momentum management, frequent ground contact was required for momentummanagement. Without such management, the satellites would lose attitude determi-nation after a short time interval. Two cesium and two rubidium atomic frequencystandards were employed. These satellites were designed for a mean mission durationof 4.5 years, a design life of five years and inventory expendable (e.g. fuel, batterylife, and solar panel power capacity) of seven years. Reliability improvements weremade to the atomic clocks on later satellites based on failure analysis from earlierlaunches. Some Block I satellites operated for more than double their design life.

3.2.4 Block II-Initial Production Satellites

On-orbit operation of the Block I satellites provided valuable experience that led toseveral significant capability enhancements in subsystem design for the Block IIsatellites. These improvements included radiation hardening to prevent randommemory upset from events such as cosmic rays to improve reliability and survivability.Besides these enhancements, several other refinements were incorporated to supportthe fully operational GPS system requirements. While most of the changes affectedonly the control segment/space interface, some also affected the user signal interface.The significant changes are identified as the following. To provide security, SAand AS capabilities were added. System integrity was improved by the addition ofautomatic error detection for certain error conditions. After detection of these errorconditions, there is a changeover to the transmission of a nonstandard pseudorandomcode to prevent the usage of a corrupted signal or data.

Nine Block II satellites were built by Rockwell International and the first waslaunched in February 1989 from Cape Canaveral Air Force Station, Florida. Theonboard navigation message storage capacity is identical to the Block I version. Withno autonomous onboard momentum control, frequent ground contact is requiredfor its momentum management. With no momentum control, the satellites may startto tumble between 28 and 45 days after the last ground contact, thus their data

GPS System Segments 77

upload is only accurate for 14 days. Again, for reliability and survivability, multiplerubidium and cesium atomic standards are onboard. These satellites were designedfor a mean mission duration of 6 years, a design life of 7.5 years and inventoryexpendables (e.g., fuel, battery life, and solar pane power capacity) of 10 years.Figure 3.10 is a depiction of a BLK II satellite.

3.2.5 Block IIA-Upgraded Production Satellites

The Block IIA satellites are very similar to the Block II satellites, but with severalsystem enhancements. The onboard navigation message data storage capability hadbeen increased to 180 days of reference data. For the first 14 days on-orbit, navigationmessage data is valid over 4-hr intervals. Following this initial on-orbit period, thenavigation message data validity interval is extended to 6 hours. With this additionalonboard storage capability, the satellites can function continuously for a period ofsix months without ground support. The OCS is limited in its ability to forecastsatellite ephemeris and clock data because orbital perturbations are nonpredictable.Therefore, the accuracy of the navigation message data will gracefully degrade overtime such that the user range error (URE) will be bounded by 10,000m after 180days. (The URE is the contribution of the pseudorange error from the OCS andspace segment.) Typically, the URE is 5.5 m (1s) based on fresh uploads of navigationmessage data every day. Details of pseudorange errors are extensively discussed inSection 7.1.2. With no general onboard processing capability, no updates to storedreference ephemeris data are possible. So, as a result, full system accuracy is onlyavailable when the OCS is functioning properly and navigation messages areuploaded on a daily basis; otherwise, the prediction error in the ephemeris dataincreases. An autonomous onboard momentum management capability has beenadded; therefore, less frequent ground contact is required. BLK IIA electronics areradiation hardened. Nineteen Block IIA satellites were built by Rockwell Interna-tional and the first was launched in November 1990 from Cape Canaveral Air ForceStation, Florida. The life expectancy of the Block IIA is the same as that of the BlockII. A BLK IIA satellite is shown in Figure 3.11.

Figure 3.10 BLK II satellite.

78 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Figure 3.11 Block IIA satellite.

3.2.6 Block IIR-Replenishment Satellites

Twenty satellites have been purchased as replenishment satellites to replace inopera-tive Block II/IIA satellites as they reach the end of their design life or fail catastrophi-cally. The first Block IIR satellite is scheduled for launch in 1997. These satellitesare being built by Lockheed Martin. To the user, the signal and data transmissionswill be identical to the Block II/IIA satellites. While the changes are transparent tothe user, the system-level operations are much different from the earlier satellites.A new capability of the satellites called autonomous navigation (AutoNav) permitsthe satellites to maintain their own ephemeris and clock data for 180 days by rangingoff other visible satellites. To support this autonomous operation, the OCS uploadsa 210-day set of predicted ephemeris and clock elements every 30 days to have datafor a full 180 days of autonomous operation. With uploading of data only every30 days, it is necessary to upload 210 days of data to ensure a full 180 days ofoperation with no ground contact. This can be clarified by considering the worstcase scenario, which is if a failure of the OCS occurred on the last day before the30-day upload, there would still be 180 days of uploaded data available with thisapproach. An upload of less than 210 days would result in a capability of less than180 days of continued operation in this case.

AutoNav is designed to operate for the 180-day period during which the UREis bounded by 5.3m. In addition, AutoNav has the capability to monitor the integrityof the GPS system such that when AutoNav is operational, the reliability and integrityof the system are improved. The AutoNav function uses a crosslink capability forranging and data communication between satellites. This crosslink is a radiofre-quency link that receives and transmits digital data and performs precision intersatel-lite ranging. Two-way crosslink range measurements are used to update clock andephemeris data relative to the stored reference data set. The crosslink uses a frequen-cy-hopped time division multiple access (TDMA) structure. The AutoNav functioncan be activated or deactivated by the OCS. When AutoNav is inactive, the Block

GPS System Segments 79

IIR satellite operates like a Block IIA in that it transmits navigation message datafrom its computer memory. The crosslink data capability also permits relaying databetween satellites. The Block IIR satellites have an onboard processing capabilityusing a MIL-STD-1750A (i.e., space qualified) processor to implement the AutoNavand other functions. The onboard software was developed using the computer lan-guage Ada throughout.

One cesium and two rubidium atomic frequency standards are used forincreased reliability and survivability. The BLK IIR satellites are designed for a meanmission duration of 7.5 years, a design life of 10 years and inventory expendables(e.g., fuel, battery life, and solar panel power capacity) of 10 years.

An artist’s depiction of a BLK IIR satellite is shown in Figure 3.12. It can beobserved that the exterior of the spacecraft is composed of an array of elements andcomponents. The two large panels on each side extending from the main body ofthe spacecraft are solar panels to provide the satellite’s electrical power. The bottomof the spacecraft is referred to as the Earth panel because of its orientation to theEarth. On this panel are a collection of antennas. The L-band antennae are arrangedwith the UHF and S-band antennae. In a small box-like compartment of the spacecraftbody are located the thrusters used for maneuvering and stationkeeping. All otherelectronics, processors, and controllers are located inside the spacecraft body.

3.2.7 Block IIF-Follow-On ‘‘Sustainment’’ Satellites

As the Block IIR satellites are launched and reach their end of life, the Air Forcewill need a procurement to replace failing Block IIR satellites. This procurement isknown as the Block IIF Sustainment Program. The DOD has defined the requirementsfor this program and awarded a contract to Rockwell International. The developmenttimeframe will be such that the satellites will be available for launch in the 2004timeframe. The potential enhancements are many. One such enhancement is a cross-link capability for all satellite commands to update status and health thus reducingthe OCS workload.

3.2.8 Summary of Satellite Block Features and Error Budget

The key features of the various satellite Blocks are summarized in Table 3.2. Thistabulation depicts the progression of satellite capability and functionality as GPShas matured to an operational system.

80 UNDERSTANDING GPS: PRINCIPLES AND APPLICATIONS

Figure 3.12 Block IIR satellite. (Source: Lockheed Martin, Inc. Reprinted with permission.)

GPS System Segments 81

Table 3.2Summary of BLK II, IIA, and IIR Features

URE at EndPeriod of of Period of

Data Storage: Autonomous AutonomousBlock Ephemeris/ Momentum Operation OperationNumber AutoNav Clock (Days) Management (Days) (m)

II No 14 OCS 14 161.1IIA No 180 Onboard 180 < 10,000IIR Yes 210 Onboard 180 7.4

References

[1] Bates, R., et al., Fundamentals of Astrodynamics, New York, NY: Dover Publications, Inc., 1971.[2] Lavrakas, J., and C. Shank, ‘‘Inside GPS: The Master Control Station,’’ GPS World Magazine,

Advanstar Communications, Sept. 1994, pp. 46–54.[3] Brown, K. R., ‘‘Characterizations of OCS Kalman Filter Errors,’’ Proc. 4th ION Satellite Division

International Technical Meeting, Albuquerque, NM, Sept. 11–13, 1991, pp. 149–158.[4] Scardera, M. P., ‘‘The NAVSTAR GPS Master Control Station’s Kalman Filter Experience,’’ Flight

Mechanics/Estimation Theory Symposium 1990, NASA Conference Proceedings CP3102, 1991.[5] ARINC Research Corporation, NAVSTAR GPS Space Segment/Navigation User Interfaces ICD-

GPS-200, Public Release Version, April 16, 1993, Reprinted by Navtech Seminars, Arlington, VA.[6] Seeber, G., Satellite Geodesy: Foundations, Methods, and Applications, New York, NY: Walter De

Gruyter, 1993.[7] Kaplan, E., ‘‘The Global Positioning System (GPS),’’ Communications Quarterly, CQ Communica-

tions, Inc., Summer 1994.[8] ‘‘1995 GPS World Receiver Survey,’’ GPS World Magazine, Advanstar Communications, Jan. 1995,

pp. 46–67.[9] ‘‘GPS/GLONASS Almanac,’’ GPS World Magazine, Advanstar Communications, Oct. 1994,

pp. 60–61.