Embed Size (px)
Citation preview
1
H
GPS and Precision Timing
Applications
Application Note 1272
2
3
Table of Contents
Page
An Introduction to GPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
GPS Structure, Control and Operation . . . . . . . . . . . . . . . . . . . . 6
The Theory of GPS Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
GPS Nonmilitary User Concerns . . . . . . . . . . . . . . . . . . . . . . . . 21
Hewlett-Packard GPS Synchronization Products and
Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
GPS Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Abbreviations and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4
For thousands of years adventur-ers, high priests, kings, emperors,explorers, sailors and otherprivileged folk have found theirway from place to place, pre-dicted the future, furtheredreligions and developed entirecultures, in part, by looking to thestars. Mankind has grouped thestars into constellations to aidrecognition and to measure thechange of seasons. Using the starsas a measure of place and time isa very old science. History tells usthat study of the heavens wasdeveloped independently by anumber of civilizations widelyscattered on the earth. Eventoday, a prudent navigator will notleave a seaport without a sextantand the compiled tables andknowledge needed to find the wayin the event of a disaster.
December of 1993 marked thedebut of a full constellation ofmanmade stars (or satellites)orbiting the earth and transmittingthe data needed for determiningprecise time and position. Today’straveler, astronomer, air trafficcontroller, delivery fleet manager,surveyor, map maker, aviator orsailor can attain position and timeinformation with a stability,accuracy and reliability unheardof just a few years ago.
It is now possible to obtainprecise timing informationanywhere on earth with highreliability and at a low cost. It iseven possible to synchronize thetiming of geographically dispersedsystems. This enables importantapplications in wireless and wiredcommunication systems.
is continuously available world-wide to anyone with a GPSantenna and receiver. It is usablein all weather conditions, any-where on or near the earth (on theground, at sea, in the air, and innear space). No user fees areeither charged or anticipated. As aresult, international civil use of theGlobal Positioning System iswidespread and growing rapidly.
The U.S. Department of Defense(DoD) developed GPS over aperiod of almost 20 years at a costof more than $10 billion. GPS isintended to serve as the basis ofthe U.S. DoD’s primary means ofradio navigation well into thetwenty-first century.
Hewlett-Packard has a broadrange of products which useprecise time to improve theperformance and reliability ofcommunication systems, powerdistribution systems and telecom-munication systems. For moreinformation, see page 22.
The NAVSTAR Global
Positioning System
We’re referring to a manmadeconstellation called the NAVSTARGlobal Positioning System (GPS),a space-based radio positioningutility. GPS provides accuratethree-dimensional position(latitude, longitude, and altitude),velocity and precise time trace-able to Coordinated UniversalTime (UTC). All this information
An Introduction to GPS
GPS Delta
VLV Launch
(USAF photo)
5
a few. GPS will change the waymany people do their work. It willfundamentally alter many existingindustries and provide opportuni-ties for new industries we havenot yet imagined.
GPS as a Source of Precise
Time, Time Interval and
Frequency
Most descriptions of GPS focus onits use as a system to provideprecise latitude, longitude andaltitude information. Often it isused to determine speed as well.GPS is depicted as a dynamicpositioning system which pro-vides the raw information neededto navigate, that is, to find wherewe are and to figure how to getfrom there to some desired place(or, perhaps, to avoid someundesired place). This is a funda-mental use for GPS but it is farfrom the only use of the system.
We live in a four-dimensionalworld and the fourth dimension istime. Without an accurate esti-mate of time, finding position as itis understood today is not pos-sible. The GPS delivers time, timeinterval and frequency anywherein the world with precision andaccuracy more than adequate formany applications.
With GPS timing, precision ofbillionths of a second is nowpossible. A billionth of a second iscalled a nanosecond (ns). Suchprecision has opened up all kindsof opportunities.
Also, in accordance with theFederal Radionavigation Plan(FRP) prepared by the U.S. DoDand U.S. Department of Transpor-tation (DOT):
“...many existing navigationsystems are under considerationfor replacement by GPS beginningin the mid to late 1990s. GPS mayultimately supplant less accuratesystems such as LORAN-C, Omega,VOR, DME, TACAN, and Transit,thereby substantially reducingfederal maintenance and operatingcosts associated with these currentradio navigation systems.”
GPS is more than a military aid tonavigation. Civil applicationsabound and outnumber militaryuses in terms of the range ofapplications, number of users andtotal market value. New commer-cial applications arise rapidly asthe decreasing cost, size andpower consumption of GPSreceivers alter the economics ofimplementing GPS. Meanwhile,receiver capabilities continue toimprove at a pace that rivals therate of change in microprocessors.Small multi-channel receivers withsophisticated tracking, filteringand diagnostic features are makingpossible even more advancedapplications.
Among the already well estab-lished civil uses of GPS are marineand aviation navigation, precisiontimekeeping, surveying, fleetmanagement (rental cars, taxis,delivery vehicles), aircraftapproach assistance, geographicinformation systems (GIS), wildlifemanagement, natural resourcelocation, disaster management,meteorologic studies and recre-ation (hiking and boating) to name
Continuous access to precise timeand frequency, at low cost andanywhere it is needed, is a revolu-tionary development. It allows, forexample, improved synchroniza-tion and timing of both wired andwireless communications systems.Users see higher quality (fewerdropped calls), increased capacity(no delays getting on), improveddata transmission (low error rates)and new services (lifetime phonenumber).
Or, consider timing electricaltransients arriving at substationsin a geographically dispersedpower delivery system. A fault(a downed line, for instance) canbe precisely located and crews canbe transported to the precisegeographic spot without delay.Similar statements can be madefor wide-area computer networks.
GPS allows precise transfer oftime between the world’s timingcenters ensuring we all tick on thesame clock. The tracking of deepspace vehicles, timing require-ments of astronomical labs (suchas millisecond pulsars), militaryand intelligence uses of precisetime — the list goes on. In general,wide availability of precise timeand frequency at low cost willimprove many scientific, manufac-turing, business, R&D and justplain fun activities.
6
II-1 14 14 02/14/89 04/14/89 operatingII-2 13 2 06/10/89 07/12/89 operatingII-3 16 16 08/17/89 09/13/89 operatingII-4 19 19 10/21/89 11/14/89 operatingII-5 17 17 12/11/89 01/11/90 operatingII-6 18 18 01/24/90 02/14/90 operatingII-7 20 20 03/25/90 04/19/90 operatingII-8 21 21 08/02/90 08/21/90 operatingII-9 15 15 10/01/90 10/20/90 operatingIIA-10 23 23 11/26/90 12/10/90 operatingIIA-11 24 24 07/03/91 08/30/91 operatingIIA-10 25 25 02/23/92 03/24/92 operatingIIA-10 26 26 04/10/92 04/25/92 operatingIIA-14 28 28 07/07/92 07/23/92 operatingIIA-15 26 26 09/09/92 09/30/92 operatingIIA-16 27 27 11/22/92 12/11/92 operatingIIA-17 32 1 12/18/92 01/05/93 operatingIIA-18 29 29 02/03/93 04/04/93 operatingIIA-19 22 22 03/30/93 04/13/93 operatingIIA-20 37 7 05/13/93 06/12/93 operatingIIA-21 39 9 06/26/93 07/20/93 operatingIIA-22 35 5 08/30/93 09/28/93 operatingIIA-23 34 4 10/26/93 11/22/93 operatingIIA-24 36 6 03/10/94 03/28/94 operating
NOTE: SVN refers to the satellite number and PRN indicates the satellite’spseudorandom noise code.
GPS Structure, Control and Operation —Or, how does this all work?
Block & SVN PRN Launch Date Operational Date StatusNAVSTAR #
Figure 1.
GPS Segments
The Global Positioning System isdivided into three segments.These segments have beendefined by the U.S. DoD andmight be thought of as indepen-dent. In practice, all segmentswork in a very integratedmanner and none could existwithout the others. Theofficial names are theSpace, Control, and UserSegments. Figure 1diagrams the relationshipbetween segments.
The Space Segmentconsists of the satelliteconstellation — twenty-oneactive satellites and threein-orbit operating spares.
Monitor Station
Master Control Station
Ground Antenna
Control Segment
Space Segment
User Segment
Satellite Constellation
History
The U.S. Coast Guard providesthe table, at left, of operationalsatellite launches (in date order):The Control Segment includes theMaster Control Station (MCS), theMonitor Stations (MS) and theGround Antennas (GA). TheMaster Control Station is locatedat Falcon Air Force Base inColorado Springs, Colorado, USA.Monitor Stations are located atHawaii, Ascension, Diego Garciaand Kwajalein. Figure 2 on thenext page is a diagram showingMCS, MS and GA sites.
The User Segment includes all ofthe military and civil equipmentintended to provide and processposition, velocity and time (PVT)information.
7
The Space Segment
The GPS Space Segment consistsof the 24 satellites in semi-synchronous orbits around theearth. This is commonly referredto as the GPS satellite constella-tion. The constellation has fouroperational satellites in each ofsix orbital planes.
The orbital planes of the satellitesare inclined 55 degrees (relative tothe equator). The orbital planesare spaced at 60-degree incre-ments around the equator asviewed from the poles. Thesatellites are at an altitude of20,200 km (12,550 miles) and eachcompletes an orbit in 12 siderealhours, which is about 2 minutesless than 12 normal hours. Thereare 24 sidereal hours in the time ittakes the earth to rotate once withrespect to the fixed stars (not thesun). This positioning of the GPSsatellites ensures that a minimumof four are in view (observable)by a user anywhere worldwide atall times. Usually, more than fourare observable, and often as manyas eight are observable from mostsites. See cover illustration.
The GPS satellites transmit radiofrequency signals toward theearth. These transmissions conveythe required information for theuser equipment to make use ofthe GPS.
The provision of precise time andfrequency signals from the space-craft is aided by the presence oftwo cesium and two rubidiumatomic clocks on each satellite.Only a single standard is used atany given time, so each satellitehas good redundancy.
13
1. Falcon AFB — MCS, MS, GA 2. Hawaii — MS 3. Proposed Cape Canaveral Site 4. Ascension — MS, GA 5. Diego Garcia — MS, GA 6. Kwajalein — MS, GA
24
6
5
Figure 2. Locations
of GPS Control
Segment Stations
The Control Segment
The GPS Control Segment con-sists of a Master Control Station(MCS) at Falcon AFB (12 mileseast of Colorado Springs, Colo-rado in the U.S.), and five MonitorStations (MS) around the world.See Figure 2 above.
The Master Control Station (MCS)is the central processing facilityfor GPS. It is manned 24 hours aday, 7 days per week — noholidays allowed. The primarytask of the MCS is to track,monitor, and manage the GPSSatellite Constellation. The MCScollects data from the MonitorStations. It continuously estimateseach satellite’s position (ephem-eris) and clock parameters. Theseestimates are uploaded throughone of the Ground Antennas (GAs)providing the ephemeris and clockdata to each satellite for retrans-mission in the navigation datamessage.
The Monitor Stations (MS) havehighly accurate radio receiverslocated at precisely-surveyedlocations. They passively track allGPS satellites in view, up toeleven at any one time, and collect
ranging data and timing from each.Little data processing actuallyoccurs at an MS. The raw measure-ments and navigation data mes-sage observations are transmittedto the Master Control Station forprocessing.
Uploads to the satellites are sentby radio transmission through theGround Antennas (GAs). GAs arealso used for transmitting andreceiving satellite-control informa-tion. Each Monitoring Station isequipped with a Ground Antennato communicate with the satellites.
The User Segment
The User Segment is comprisedof a variety of user equipmentwith receiver/processors specifi-cally designed to receive, decode,and process the GPS satelliteranging codes and navigationdata messages.
GPS provides services for twolevels of users. These are referredto as the Standard PositioningService (SPS) and the PrecisePositioning Service (PPS). Thelatter is reserved, almost entirely,for the exclusive use of the DoD.
8
The Theory of GPSOperation
military use and will be denied tounauthorized users by use ofcryptography. PPS will be madeavailable to U.S. Federal and AlliedGovernment (civilian and military)users through special agreementswith the DoD. Limited, non-Federal Government, civilian useof PPS, both domestic and foreign,will be considered upon requestand authorized on a case-by-casebasis.”
User Equipment
NAVSTAR GPS provides userswith three dimensional position,velocity determination, andprecision time. The term receiver/processor indicates a GPS receiverand/or other user equipmentdesigned to track the GPS satelliteradio signals and provide positionand time information. Astonishingaccuracy is provided to the user.
Generalized GPS applicationsinclude positioning service,providing users with highlyaccurate navigation and geodesycapabilities, and extremely accu-rate time reporting for device andevent control.
GPS user equipment varies signifi-cantly in design and function.There is wide potential for special-ized and varied applications andreceiver designs are available formost user markets, such asaviation navigation, marinenavigation, geodesy and timing.Many of these use databases andelectronic charts to enhance theusefulness of raw latitude, longi-tude and time.
Hewlett-Packard has developed aunique set of timing equipmentutilizing GPS as an SPS user. UsingHP SmartClock Technology andspecial algorithmic filtering, PPS-like performance can be achievedfor timing. These techniques donot help positioning accuracy.
GPS provides position and velocityinformation with respect to theWorld Geodetic System 1984(WGS-84) map datum or coordi-nate system. Any other coordinatesystem or user-generated datumwill require a translation. This isoften done. GPS is not based onthe location of ground-basedtransmitters. Therefore, the typicalGPS receiver’s location outputscannot be directly used fornavigation in relation to a fixedpoint, as with a Very High Fre-quency Omnidirectional Range ornondirectional beacon or LORAN.Instead, an area navigationscheme with waypoint informationis used. The matter is easilyhandled by data processing andoutputs can be customized for anyspecific need.
Technically there is a distinctionbetween the terms positioning andnavigation. However, for thepurposes of the GPS user, the twoterms are employed interchange-ably. Remember, the GlobalPositioning System providesposition, velocity, and time (PVT)information which can be used fornavigation purposes.
The U.S. DoD states very clearly inthe Federal Radionavigation Plan(FRP) what SPS and PPS provide:
“SPS is a positioning and timingservice which will be available toall GPS users on a continuous,worldwide basis with no directcharge. SPS will be provided onthe GPS L1 frequency whichcontains a coarse acquisition(C/A) code and a navigation datamessage. SPS is planned toprovide, on a daily basis, thecapability to obtain horizontalpositioning accuracy within100 meters (2 drms, 95 percentprobability) and 300 meters(99.99 percent probability),vertical positioning accuracywithin 140 meters (95 percentprobability), and timing accuracywithin 340 ns (95 percent probabil-ity). The GPS L1 frequency alsocontains a precision (P) code thatis reserved for military use and isnot a part of the SPS. Althoughavailable during GPS constellationbuild-up, the P code will be alteredwithout notice and will not beavailable to users that do not havevalid cryptographic keys.
“PPS is a highly accurate militarypositioning, velocity, and timingservice which will be available ona continuous, worldwide basis tousers authorized by the DoD. PPSwill be the data transmitted onGPS L1 and L2 frequencies. PPSwas designed primarily for U.S.
9
Ranging Theory
GPS position determination isbased on a concept called timeof arrival (TOA) ranging. TOAis a complex way of saying signaltravel time from one point toanother.
Simply explained, TOA ranginginvolves transmitting a signal ata known time and measuring thearrival (reception) of that signalat a later known time. Theinterval between the known timeof transmission and the knowntime of reception is the TOAvalue.
TOA Ranging Examples
As a simple example, the TOAranging concept (based on thevelocity of sound) is frequentlyused by adults and childrenworldwide. It is generally under-stood that in a thunderstorm, thedistance from a lightning flash toyour location can be simplyfigured by counting the secondsfrom when the visual flash is seen(transmission) until the audiothunder report is heard (recep-tion). In this example, we canassume the perception of thelightning flash is instantaneousbecause the velocity of light isnearly a million times faster thansound traveling in air. The TOA isthe time between the flash andthe thunder.
Sound travels at about 330 meters(1,100 feet) per second. Hence,the thunder takes about threeseconds per kilometer (fiveseconds per mile) to travel.Simply divide the TOA value(time between flash and thunderclap) by 3 to determine the rangein kilometers between you (thereceiver) and the lightning flash(the transmitter), or multiply by0.2 to get the distance in miles.See Figure 3.
of sound (1,087.1 feet per secondat sea level at 32°F). The range(distance from the foghorn) is10,871 feet. The mariner sets acompass to draw an arc corre-sponding to 10,871 feet at the scaleof his chart. This is TOA-BasedRange No.1, as shown in exampleA of Figure 4 on next page.
The mariner then locates a secondfoghorn on the chart and repeatsthe procedure. The mariner islocated at one of the two positionswhere the range circles intersect.See example A in Figure 4.
This simple example of the TOAranging concept is not without itsdifficulties. The range circles givetwo intersections (possible posi-tions) so the mariner must knowhis position to eliminate ambiguity.
As a further example of the TOAranging concept using soundwaves, assume the transmitter is afoghorn that sounds precisely onthe minute mark (precise andknown time of transmission) andthe receiver is a mariner listeningfor the foghorn. The mariner isequipped with a chronometer(accurate timepiece) to be usedfor timing purposes and a chartwhich shows the location of thefoghorn.
For this example, the foghorn’ssound arrives at the mariner’sposition exactly 10 seconds afterthe minute mark (precise andknown time of reception) asindicated by the chronograph.The mariner’s position relative tothe foghorn (range) is thencalculated by multiplying the TOAvalue of 10 seconds by the speed
The time interval between the time of transmission (flash of lightning) and the time of reception (man hears the thunder) is called time of arrival (TOA).
TOA = 6 seconds between the flash and the thunder. Range = TOA/3 = 2 kilometers (1.2 miles), since the speed of sound is 1 kilometer per 3 seconds.
Figure 3. A basic
lesson in TOA
ranging
10
Figure 4. Examples
of the Mariner’s
TOA ranging
(U.S. Govt. Pub.)
As a result of using three foghorns(transmitters) with certain knownfacts about each (exact positionand time of sounding), the ranginginformation can be applied to amap to determine latitude andlongitude coordinates. Further, thechronometer’s clock bias errorbecomes a known and correctablequantity.
In simple terms, the same line ofthinking follows for GPS. Thesignal velocity becomes the veloc-ity of light instead of sound. Threeposition coordinates (latitude,longitude and altitude) are neededinstead of two. Hence, four satel-lites are needed instead of the threefoghorns.
Intersection ambiguity canbe eliminated by makingrange measurements tothree foghorns. This willincrease the number ofintersection points, butonly one intersection pointwill be common to allthree TOA-based rangecircles. All other intersec-tions are false positions.See example B in Figure 4.
The Need for
Accurate Timing
The chronometer used fortime measurement must bevery accurate, or precisionpositioning will not bepossible. Clock bias error,as an error in clockaccuracy is called, deliversincorrect position.
If the mariner’s chronom-eter in the first examplewas running one second fast, thenthe sound of the foghorn thatarrived 10 seconds past the minutemark would have the appearanceof arriving at 11 seconds past theminute mark on the mariner’schronometer. The additional onesecond would cause the marinerto compute an erroneous range of11,958 feet from that foghorn(1,087-foot error). When rangingto a second foghorn, both of theTOA-based range observationswould have range errors of1,087 feet. Although the marinerwould now have an erroneousposition fix, the problem of clockbias errors and ambiguous inter-sections can be solved. Seeexample C in Figure 4.
A. TOA-Based Position Determination B. Eliminating False Intersection Ambguity
D. TOA-Based Position and Time DeterminationC. Effect of a Chronometer Bias
Latitude
Longitude Location of Foghorn No. 1
Location of Foghorn No. 2
TOA-Based Range No. 1 TOA-Based
Range No. 2
USER
USER
USER
USER
E
E
Erroneous Position Fix
Foghorn No. 1
Foghorn No. 1
Foghorn No. 1
Foghorn No. 3
Foghorn No. 2
Foghorn No. 2
Foghorn No. 3
Foghorn No. 2
TOA-Based Range No. 1 + E
TOA-Based Range No. 2 + E
TOA-Based Range No. 3
TOA-Based Range No. 1 TOA-Based
Range No. 2
TOA-Based Range No. 3
TOA-Based Range No. 1 TOA-Based
Range No. 2
A , B , and C are False Intersections
A
B
C
Using three foghorns, note thatthree dual-foghorn intersectionsare near the mariner’s true posi-tion. Distance E between theintersections of range circles fromfoghorns 1 and 2, foghorns 1 and 3,and foghorns 2 and 3 is a functionof the chronometer’s clock biaserror. By moving the clock for-ward or backward until the threedual-foghorn intersections con-verge at the true position, thechronometer’s clock bias has beendetermined. See example D inFigure 4.
11
Time and Position Theory
At the heart of GPS is the timingaccuracy available from atomicclocks. Albert Einstein gave usthe relationship between spaceand time — the four dimensionsof relativity. These four dimen-sions may be thought of aslatitude, longitude, altitude andtime, or in shorthand x, y, z and t.GPS is the first engineeringimplementation of relativity andwould not work without it.Fortunately, an understanding ofrelativity is not necessary to gainan understanding of how GPSworks. The relativistic terms areaccounted for in the design of thesatellite clocks and in the receiv-ers that properly process the data.
The orbit period for GPS placesthem at a distance of about4.2 earth radii from the center ofthe earth as illustrated by the feetof the tripod in Figure 5. Therelativistic velocities of thespace vehicle (SV) clocks causethem to lose about 7.2 millionthsof a second (7.2 microseconds)per day with respect to the earth.On the other hand, their altitude(often called the gravitationalredshift) causes them to gain45.6 microseconds per day. Thenet is a gain of 38.4 microsecondsper day. This accumulation isenormous compared to the fewnanoseconds synchronizationaccuracy desired for the system,since a microsecond is 1,000 timesa nanosecond. The SV clocks areconstructed on earth to lose 38.4microseconds per day, so thatwhen they are in space theyappear to be running at theright rate.
Because the earth is rotating andgenerally all receivers are moving,the relative velocities and posi-tions between SV clock andreceiver clock have to be accom-modated in the receiver software.It is not known that all receiversare performing this calculationcorrectly. This is a receivermanufacturer’s responsibility.Just due to the rotating earthalone, hundreds of nanosecondsof error can occur in synchroniz-ing clocks around the globe unlessthe calculations are done correctly.
The U.S. Air Force has created aninterface control document (ICD200-B) which describes how toperform these calculations.
Once the relativity is properlyincluded, GPS can be thoughtof in a traditional time-of-flightway. By measuring the timeof arrival (TOA) of a signal —knowing the time it began and itsvelocity — the range or distancebetween the sender and thereceiver can be easily calculated.
r
4.2r
Figure 5. Three
fixed points in
space given by
three satellites and
the lengths (given
by time of flight of
the signals from
the satellites)
determine the
position of the GPS
receiver’s antenna
like the tripod
fixes the position
of the telescope.
12
For GPS, the velocity of light isused instead of the velocity ofsound for determining range.As Einstein taught us, time andspace are interlinked (seeFigure 5). Consider the tripodshown in the illustration with theGPS receiver on top. The positionof the receiver antenna on earth isdetermined by the positions of thethree satellites at the bases of thetripod legs. The positions of theGPS satellites are determined bythe GPS control segment and eachsatellite broadcasts, as part of itsmessage, its own position.
The relativistic times broadcast bythe satellites are kept synchro-nous by the control segment aswell. Suppose there is also asynchronous atomic clock in theGPS receiver. Since all four clocksare ticking together and the ticksfrom the satellites are broadcasttoward the earth at the velocity oflight (30 cm — about 1 foot — perns), the earth GPS receiver clockcan measure how long it takeseach signal to arrive (like thepropagation of the sound signalfrom the lightning strike). Know-ing the propagation time of eachof the three legs allows thereceiver to calculate the length ofeach leg (the TOA times thevelocity of light). If the positionsof the satellites are known to onemeter and all four clocks aresynchronous to better than threenanoseconds, the position of theGPS receiver could be calculatedto an accuracy of about one meter.
Though the above example outlinesthe essential principles, the actualoperation of GPS is somewhatdifferent. In the above example,the receiver already knew the time,and from the measurements couldcalculate the latitude, longitude,and altitude of the GPS receiver’santenna, which is the receivingpoint of the signals. Since havingan atomic clock at the receiverwould be both expensive andbulky, adding a fourth satellitemakes it possible to calculate thetime as well as the latitude, longi-tude, altitude of the receiver clock.In other words, the four legs fromthe four satellites to the receiverantenna yield four equations, andthere are four unknowns: x, y, zand t. Having four satellitesprovides a significant cost savingsin the user equipment.
The receiver only needs to have ashort-term stable quartz clock,which can easily be used to mea-sure the time differences of theTOA of each of the received signalsto a precision of a nanosecond.The clock bias for this inexpensivereceiver-clock can be calculatedusing the data from all four satel-lites (as in the foghorn exampleearlier). In other words, there willbe a conceptual sphere aroundeach of the four SVs. The radius ofeach sphere will be the calculatedrange. The only way the fourspheres will intersect at a point is ifthe time used in the receiver’scalculation is synchronous with
GPS time. Hence the receiver’scomputer can calculate clock biasfor inexpensive clock at the timeof the measurement — effectivelygiving it the accuracy of theatomic clocks in the GPS.
Potential error sources in GPSarise from sundry sources. Thebroadcast positions (ephemerides)of the satellites are not knownprecisely. The velocity of thesignal from the satellite is per-turbed by the ionosphere and thetroposphere. The broadcast timesfrom the four or more satellitesare not synchronized perfectly.Multipath distortion affects thereceived signal at the antennacausing errors in the calculatedTOA value. Antennaand receiver electronic noise andcable delay variations with tem-perature, for example, add to theuncertainties in the accuratedetermination of x, y, z and t.
The accuracies and precisionsavailable from GPS vary dramati-cally, depending on the methods ofutilizing the GPS signal informa-tion. Positioning accuracies varyfrom sub-centimeter for differen-tial measurements to over 100meters for a simple SPS user.Timing precisions vary from sub-nanosecond to hundreds ofnanoseconds, and accuracies oftendepend on user and/or manufac-turer’s care in setting up andcalibrating the equipment.
13
The above levels of accuracy arenot available to the SPS userwithout special processingbecause of the degradationpurposely imposed by the DoDfor security reasons. This degra-dation is called Selective Avail-ability (SA). Hewlett-Packardimplements a unique SA filteringalgorithm which removes most ofthe effects of SA for timingpurposes. Hence, the aboveaccuracies are potentially avail-able with the proper choice ofHewlett-Packard equipment andwith proper set up procedures.
GPS time has no leap seconds andits time is given by TAI minus19 seconds. The GPS compositeclock is steered to be synchronousto a very high accuracy with theinternational UTC scale’s moduloone second. This will be discussedin more detail later in this applica-tion note. All that is required foraccurate positioning with GPS isthat its time be consistentlyavailable from all the satellites.This is currently being done at thelevel of about 10 ns. By steeringprocedures, GPS time is now keptaccurate within about 20 ns withrespect to TAI–19s.
GPS also provides as part of itsdata message the corrections toapply to GPS time to obtainUTC. The reference in this caseis the master clock at the USNaval Observatory (USNO MC).When these corrections areapplied, the time accuracy achiev-able is typically less than 10 ns.
Accurate Time
The accuracy of a clock can bethought of as the degree ofagreement with some acceptedreference clock. GPS time can bethought of as a software clock.GPS time (often called thecomposite clock) is computed asa weighted set of all the readingsof the current GPS clocks. TheMaster Control Station (MCS)continually estimates the clockbias for each of the SV hardwareclocks so that the satellites canbroadcast its estimate of GPSsoftware time. The errors in thecomputed GPS time will beaffected by the above sources andwill consist of both systematicand environmental perturbingsources as well as randomvariations.
The official reference clocks forthe world are InternationalAtomic Time (TAI) and Coordi-nated Universal Time (UTC).These time scales are generatedby the Bureau International desPoids et Mesures (BIPM). Bothscales use exactly the same valuefor the System Internationale(SI) second based on the cesiumatom and differ in their readingsby an integer number of seconds(called leap seconds). Underinternational agreement asannounced by the InternationalEarth Rotation Service (IERS),leap seconds are added to UTCto keep it within 0.9 seconds ofearth time (UT1).
14
GPS Transmitted Message
The orbiting GPS satellites arethe broadcast beacons (transmit-ters). The satellite’s radio signalstravel at the speed of light(299,792,458 meters per second,about 30 cm/ns or 186,282 milesper second) and consist ofpseudorandom noise (PRN)modulated L-band radio waves.The PRN sequences, coarse/acquisition codes (C/A-codes),and precision codes (P-codes)are predetermined strings of oneand zero data bits generated byan onboard clock. This clockprovides the exact transmit timeof the broadcast signals (preciseand known time of transmission).
For determining x, y, z and t, theGPS satellites transmit (usingspread spectrum techniques) ontwo frequencies referred to asL1 and L2:
L1 = 1575.42 MHzL2 = 1227.60 MHz
The GPS satellites transmit the1 and 0 data bits by the L1 and L2
GPS Clock Bias
If the GPS receiver clock wassynchronized to the onboard GPSsatellite clocks (GPS system time),the TOA values observed by thereceiver would be equal to thegeometric ranges between thesatellites and the user, divided bythe speed of light.
In practice there are clock biasesin the receiver’s clock as well as inthe SV clock. Radio signals(traveling at the speed of light)require clock accuracy to within afew nanoseconds. Part of theinformation in the message beingtransmitted is the precise timedifference between the
carrier signals, digitally modulat-ing, using binary phase shiftkeying (BPSK). BPSK moves thecarrier phase in one direction for adigital PRN code change from 1 to0 and in the other direction for achange from 0 to 1. GPS differsfrom AM/FM radio, and most otherradio systems, in that all satellitestransmit on exactly the sameL1 and L2 frequencies. Eachsatellite has a different PRN codesequence. Hence, by locking ontoa particular code sequence, areceiver can select a uniquesatellite. The rest of the satellitesignals look like white noisebecause the codes are randomly(in a known way) distributed.
The P- and C/A-codes are alsoknown and predictable relativeto the start time of the codesequence. Therefore, by use of aproperly designed receiver /processor, the user can preciselyreplicate the same code thesatellite will send at any giventime. When the replicated P- orC/A-code sequences are mixedwith the L-band radio signalstransmitted from the satellites
(offsetting the replica sequenceforward or backward in relation totime), a match of the incomingsignal with the replica will bemade. This match is called acorrelation, and the amount oftime offset computed by thereceiver to achieve correlation isdirectly proportional to the rangebetween the receiver and thesatellite. The result is called theobserved TOA value. See Figure 6.
SIS From The Satellite
Time Interval Receiver Predicted Signal Correlated Signals
Time Offset
Time Marks
TOA Is The Time Interval Between the Satellite’s SIS Transmission and Reception at the GPS Receiver's Antenna.
The Amount of Time Offset Made by the Receiver's Code Generator to Make the Correlation is Directly Proportional to the Range Between the Receiver and the Satellite.
Figure 6.
Offsetting by Time
15
satellite’s free running clock andGPS system time. The computerin the receiver uses the foursatellites to calculate its internalclock bias. These two pieces ofinformation are used by thereceiver in order to estimate thetrue range to the satellite.
When the clock biases are can-celled, the receiver’s calculatedTOA value times the velocity oflight becomes the range from eachsatellite. The raw measurementsbefore correcting for clock biasare called pseudorange (PR)measurements.
Note how the user’s receiverperforms all the functions of themariner’s navigation system. Thereceiver’s antenna (ear), accurateclock (chronometer), plottingtable (charts), and calculator(data processor), are all replaced
The receiver’s data processorobtains the necessary informationto make these compensationsfrom the navigation data message(NAV-msg). The NAV-msg issuperimposed on both the P-codeand C/A-code satellite signals at adata rate of 50 bits/s and contains25 data frames. Each data frameconsists of 1,500 bits divided intosubframes of 300 bits each. A GPSreceiver requires thirty seconds toreceive one data frame and 12.5minutes to receive all the 25 dataframes available. See Figure 7.
by one device. This single piece ofequipment, consisting of a radioreceiver and a data processor,does all the positioning calcula-tions and reports time with nooperator intervention.
Position and Time
Computations
As the GPS receiver beginstracking four satellites, receivingthe PRN sequences for each, andgenerating TOA values, thereceiver’s data processor takesover. It samples the TOA valuesfor each of the four satellites andmultiplies them by the speed oflight to produce four pseudorangemeasurements. The processor thencompensates the PR measure-ments for deterministic errorsincluding time differences be-tween an individual satellite’sclock and GPS time, atmosphericradio signal distortion, effects ofrelativity, and internally-generatedreceiver noise.
Bit No.Subframe
1
Subframe 2
Subframe 3
Subframe 4
Subframe 5
Telemetry Word
Handover Word
Telemetry Word
Handover Word
Telemetry Word
Handover Word
Telemetry Word
Handover Word
Telemetry Word
Handover Word
0 30 60 300
600
900
1200
1500
300 330 360
600 630 660
900 930 960
1200 1230 1260
(Multiples)
(Multiples)
Clock Correction
Ephemeris
Ephemeris
Message (Changes Through 15 Frames)
Almanac-Health Status (Changes Through 25 Frames)
6 Sec
12 Sec
18 Sec
24 Sec
30 Sec
*12.5 minutes before the entire message reports
Figure 7. The
NAV-msg
(U.S. Govt. Pub.)
16
The navigation data message(NAV-msg) contains GPS time oftransmission, ephemeris or orbitalposition data, clock data of theparticular satellite being tracked,and almanac data for the remain-ing satellites in the constellation.Coefficients for calculatingUniversal Time Coordinated(UTC) and the ionospheric delaymodel (propagation delays andatmospheric radio signal distor-tion) for C/A-code users are alsopart of the NAV-msg. The NAV-msg is normally valid for a four-hour period.
The broadcast ionospheric delaymodel is about 50 percent accu-rate. This can give rise to errorsof several nanoseconds. PPSreceivers have the potentialthrough the use of the L1 and L2frequencies to measure theionospheric delay, reducing this
are in coordinate terms, latitude,longitude, and altitude.
Special timing receivers, depend-ing upon their style and parametersettings, will either produce anestimate of GPS system time orUTC, or they will measure a localreference clock against either orboth GPS time and UTC.
Position and Time Reporting
The adjustments and mathemati-cal solution are, in reality, a veryhigh-speed version of the marinerplotting out a two-dimensionalposition fix and zeroing thechronometer clock bias errorusing the intersection of threeTOA-based range circles. GPS is afour-dimensional system (latitude,longitude, altitude and time), sofour TOA-based range spheres areneeded for a complete solution.
Figure 8. Simplified
Computations
(U.S. Govt. Pub.)
Simplified User Position/Time Computation Process
A. Data processor obtains pseudorange measurements (PR1, PR2, PR3, Pr4) from four satellites Time-coded signals transmitted by each satellite
Time each signal received by user
T1T2
T3T4
PR1 = T1 x cPR2 = T1 x cPR3 = T1 x cPR4 = T1 x c
B. Data processor applies deterministic corrections PRI = Pseudorange (I = 1, 2, 3, 4) • Pseudorange includes actual distance between satellite and user plus satellite clock bias, atmospheric distortions, relativity effects, receiver noise, and receiver clock bias • Satellite clock bias, atmospheric distortions, relativity effects are compensated for by incorporation of deterministic adjustments to pseudoranges prior to inclusion into position/time solution process
C. Data processor performs the position/time solution four ranging equations
(X1 - UX ) 2 + (Y1 - UY )2 + (Z1 - UZ )2 = (PR1 - CB x c)2
(X2 - UX )2 + (Y2 - UY )2 + (Z2 - UZ )2 = (PR2 - CB x c)2
(X3 - UX )2 + (Y3 - UY )2 + (Z3 - UZ )2 = (PR3 - CB x c)2
(X4 - UX )2 + (Y4 - UY )2 + (Z4 - UZ )2 = (PR4 - CB x c)2
XI, YI, ZI = satellite position (I = 1, 2, 3, 4) • Satellite position broadcast in 50 Hz navigation message Data processor solves for: • UX, UY, UZ = user position • CB = GPS receiver clock bias
(c = Speed of light)
PR2
PR 1
PR3PR 4
inaccuracy to the order of anonosecond.
Adjustments to the pseudorangemeasurements are made by thedata processor portion of the GPSreceiver, then the position/timesolution calculations are made.The position/time solution processconsists of a mathematical solu-tion of four ranging equations,using the four adjustedpseudorange measurements todetermine the four unknownquantities of the x, y, z positioncoordinates and the receiver clockbias. See Figure 8.
The NAV-msg contains the infor-mation required by the receiver’sdata processor to compute thesatellite’s position at any point intime. The results of the receiver’sprocessor, sometimes called a fix,
17
Time and Time Transfer and
Frequency
GPS provides time traceableto UTC as produced by the U.S.Naval Observatory MasterClock (USNO MC). With its ratesyntonized with the internationaldefinition of the second, and withUTC (USNO MC) time synchro-nized well within 20 ns of UTC, theGPS system is a worldwideresource for precision time.
Note, these accuracy values onlyapply to time results based upon areceiver’s solution for the fourranging equations and do not applywhen using specialized time-transfer receivers designed toexploit GPS as a precise time andfrequency transfer total. There areat least six different methods ofusing GPS for timing. The threemost popular ones will be touchedon in this application note.
The GPS common-view method isused to transfer the time of about300 clocks to the BIPM for thegeneration of UTC. It is the mostaccurate operational method todate. Two locations in commonview of a GPS satellite measurethe TOA over the same interval.Subtracting these two measure-ments cancels the GPS clock biasand much of the ephemeris error.This leaves the time differencebetween the clocks at the twolocations. Precision and accura-cies of a few nanoseconds arecommonly achieved.
This basic concept requires thereceiver to be fixed at a knownlocation, thus allowing the re-ceiver to use all pseudorangemeasurements solely for timing.
In addition, this common-viewmethod is used for comparing thetime and frequency of clocks atinternational timing centers for
A positioning receiver does notneed accurate time, since it onlyneeds to remove time or receiverclock bias in solving for x, y and z.In fact, time delay variations inthe antenna and receiver that arecommon to all four satellites usedin the solution will cancel incommon-mode fashion.
A receiver/processor candetermine its position usingpseudorange measurementsfrom fewer than four satellitesif external aiding information(exact altitude measurementsfrom a barometric altimeter orthe exact GPS time) is available.An aiding source can replace onesatellite-based PR measurementin the solution process. If bothexact altitude and exact GPSsystem time aiding were available,position, velocity, and time(PVT) could be solved withmeasurements from only twosatellites.
GPS positions are referenced tothe electrical phase center of thereceiver’s antenna, not thereceiver’s location. The antenna isthe actual point of signal recep-tion. Therefore it is the antenna’sposition that is determined.
Some receivers don’t producetiming information and are onlyused for position or navigationpurposes. Accurate timingreceivers require some specialcare in their design.
millisecond pulsar timing, syn-chronizing and syntonizing NASAJPL’s Deep Space Network, andfor several other time and fre-quency calibration facilities.
Other applications using precisetiming include radio pagingsystems designed to share chan-nels (as required by a recent FCCruling regarding Private CarrierPaging), time division of frequen-cies (channels) in the cellular andother communication industries,and other radio spectrum efficientoperations. Many of these commu-nication applications base theirentire operation on one pulse-per-second (pps) timing rather thanon GPS or UTC.
GPS is an excellent resource forpps timing over wide geographicalareas. For example, the specificin-view satellites used for PVTinformation will normally be verysimilar, at any given time, over theentire continental U.S. This makesprecise time synchronization overan area this size relatively easy.
Signal Acquisition
The GPS receiver detects thesatellite constellation’s spreadspectrum broadcasts (SIS), whichare actually below the earth’snatural radio noise level. Afterinitial detection, the signal iseffectively amplified by use of theprocessor-predicted P- and C/A-codes and collapsed into theoriginal carrier frequency band,where it is concentrated andamplified above the natural noiselevel.
At this point the signal can bereferred to as locked and itbecomes useful.
18
The minimum required signalstrengths at a GPS receiverantenna are:
frequency L1 C/A-code: –160 dBWfrequency L1 P-code: –163 dBWfrequency L2 P-code: –166 dBW
Satellite tracking sequences beginwith a determination by thereceiver of which satellites arevisible (useful) for tracking. Asearch of the sky is made bysequencing through the possible orprobable PRN codes being trans-mitted by the satellites to locateand lock onto any single satellite inview. Once a satellite is locatedand locked, the receiver reads thenavigation data message (NAV-msg) to obtain current almanacinformation about the other GPSconstellation satellites.
Accuracy Limiting Factors
The potential accuracy of a GPSreceiver’s position solutions isdetermined by two very importantfactors:
1. Measurement errors in thepseudoranges.
2. Satellite-to-user geometry.
Knowing both these factors for aspecific GPS receiver at a particu-lar point in time and space allowsan understanding of the limitationsof GPS and enables forecasting ofthe receiver’s position and timeaccuracy.
The error in the receiver’s mea-surement of the pseudorangesfrom each satellite is called theuser equivalent range error(UERE) and is the product of:
1. Uncertainty in the GPS satellite’sclock bias.
2. Uncertainty in the satellite’sephemeris.
3. Errors in the satellite-broadcastNAV-msgs.
4. Imprecision of receiver PRNsequence tracking design.
5. Errors in calculation of theionospheric model.
6. Errors in location, antennadelay, antenna cable delay, tropo-spheric effects, solar flares andothers.
Figure 9. UERE
(U.S. Govt. Pub.)
User Equivalent Range Error (UERE):
If The True PR Is:
If The Measured PRs Are:
20,000,000 m Errors Are: When Plotted:
20,000,00220,000,004
19,999,99520,000,000
20,000,001
19,999,998
+2 m+4 m
–5 m
0 m
+1 m
–2 m
– +–1σ +1σ0
16% 68% 16%
The UERE is further affected bythe quality of the received satellitesignals and will vary betweensatellites and times. Due to differ-ences in receiver designs, theUERE will also vary betweendifferent user equipment. SeeFigure 9.
Dilution of Precision (DOP) is thesecond accuracy-limiting factor. Itis a geometric quantity dependingupon the relative positions of theuser and the selected satellites.High values of DOP cause smallerrors in range measurements tobecome large position errors. Forthis reason, the four satellitesselected by the receiver to deter-mine PVT must represent low DOPvalues. The DOP limiting factor isindependent of the quality of thesignals or the type of GPS receiver(providing the same four satellitesare selected). Basically, DOP is anamplification factor that multipliesthe UEREs and increases thereceiver’s PVT solution errors.
19
Figure 10.
Good Dilution of Precision (DOP)
Good DOP indicates that thesatellites exhibit good geometryrelative to the user. Good DOPsare indicated by low numbers andpoor DOPs by high numbers. Thelowest numbers occur when theselected satellites are widelyspaced in the sky above the GPSreceiver. See Figure 10.
Poor DOP occurs when thesatellites are close together orwhen they form a row or circle. Itis possible for poor DOP numbersto be so high that the receiver isunable to process a solution. SeeFigure 11.
Note: The best possible satellitegeometry for PVT solution requiresone satellite to be directly over-head and three other satellitesequally spaced around the horizon.
Each of the position and timesolution dimensions has a specifictype of DOP involved with satellitegeometry:
Vertical or spherical dimension(VDOP) — altitude errors.
Horizontal or circular dimension(HDOP) — latitude/longitudeerrors.
Position (PDOP) — combinationof HDOP and VDOP.
Time (TDOP) — relates to VDOP,HDOP, and PDOP.
Time transfer (TTDOP) — relativeto single unknown solution.
Position and Time Errors
Accuracy of the position and timesolutions obtainable from a GPSreceiver are the product of the twofactors UERE and DOP.
User
User
Figure 11.
Poor Dilution of Precision (DOP)
Knowing the UERE and DOPfactors for a particular receiver ata specific time and place allowsthe forecast of that unit’s positiontime errors. Forecasting is essen-tial to provide:
1. Real-time performance estima-tion within the GPS receiver.
2. Planning and analyzing the useof GPS.
It is easy to forecast position andtime errors when all the satellitesignals give the same UERE. Insuch a case, the forecast naviga-tion and time errors are computed
by simply multiplying the appropri-ate DOP value by the commonsatellite UERE value.
DOPs are forecast as a function ofthe user’s position/time, the numberof satellites visible, and the satellitelocations. The GPS receiver’sportion of the UERE is usuallycompensated for by the manufac-turer. The satellite portion of theUERE is forecast by the GPSControl Segment and is providedto the user’s receiver as a part ofthe NAV-msg.
20
The average UERE and DOPvalues for the composite globalsample space (upon which thefour GPS accuracy specificationsare based) are:
UERE value for PPS receivers is7.0 meters (22.96 feet) one sigma.
UERE value for SPS receivers isapproximately 32 meters (104.99feet) one sigma.
DOP numbers for PPS receivers:HDOP 1.5, VDOP 2.0, PDOP 2.5,and TDOP 1.1.
DOP numbers for SPS receivers:HDOP 1.6, VDOP 2.2, PDOP 2.7,and TDOP 1.3.
Methods of Increasing GPS
Accuracy
GPS can exhibit variations ofaccuracy; among the causes arepropagation anomalies, errors ingeodesy, accidental disturbancesof signal timing, or SelectiveAvailability (SA — official degra-dation by the DoD for nationalsecurity reasons). The SA canoccur in the form of dither onthe satellite clock signal, whichis not reported in the NAV-msg,and/or in an erroneous broadcastephemeris. The adverse effectsof these variations may bereduced or eliminated by usingthe common-view method, byDifferential GPS (DGPS) andEnhanced GPS (EGPS)techniques.
DGPS Operation
Differential GPS operation uses aprecisely located reference stationequipped with an all-in-view GPSreceiver to measure pseudorangesto each satellite. By comparing acalculated range and the measuredpseudorange, correction informa-tion is computed for eachpseudorange measurement. Thiscorrection information is thenbroadcast to nearby GPS users(equipped with Differential GPSreceivers).
Differential corrections to SPSimprove positioning accuracy from100 meters (328.1 feet) (2 drms) tobetter than 10 meters (32.8 feet)(2 drms). Time determination islikewise improved. The geographi-cal area over which DGPS correc-tions can be made from a singledifferential reference facility canvary from a small and specifictarget such as an airport, to anarea covering thousands of squaremiles.
DGPS stations are currently beingdeployed rapidly by the U.S. CoastGuard and many others. Also,private DGPS providers exist inmost areas of the U.S. Manytransmit in the FM radio band,charge for their services, andprovide DGPS coverage in areaswell-removed from the coastlineand river systems that the CoastGuard doesn’t serve.
Enhanced GPS
Enhanced GPS (EGPS) is ameans of algorithmic filtering theeffects of SA on SPS accuracy.This new approach developed atHewlett-Packard has been shownto provide real-time accuracy to 20nanoseconds, or better dependingon the quality of the reference
clock used in or with the GPSreceiver. With a quartz crystaloscillator for the clock, the EGPSdelivers about 20 ns (rms), with arubidian clock about 10 ns, andwith a cecium clock about 3 ns.Note that receivers operating inthe EGPS mode must be station-ary, as EGPS is not dynamic innature and cannot operate onmoving platforms.
GPS Operational Capability
Attainment
Initial Operational Capability(IOC) was attained in December1993, when the system configura-tion met the configuration definedin the 1992 FRP (24 GPS Block I/II/IIA type satellites operating intheir assigned orbits, available fornavigation use and providingpositioning and time service). Thesatellite total of 24 includes threeoperational spares, in orbit, thatcan be relocated to replace a failedor missing satellite.
The IOC notification was made bythe Secretary of Defense followingan assessment by the systemoperator (USAF) that the satelliteconstellation could sustain therequired levels of accuracy andavailability.
Prior to the IOC notification, GPSwas considered to be in theprocess of development, meaningthat signal availability and accu-racy were subject to change at anytime.
Full operational capability (FOC)defines the condition when theGPS system is capable of provid-ing full and supportable militaryservices.
21
In March of 1996, the White House,Office of Science & TechnologyPolicy, and U.S. National SecurityCouncil released a statementaddressing these very realconcerns. The release outlines“a comprehensive national policyon the future management anduse of the U.S. Global PositioningSystem (GPS) and related U.S.Government augmentations”approved by the President. Thegoal of this policy is to managethe use of GPS to “support andenhance our economic competi-tiveness and productivity whileprotecting U.S. national securityand foreign policy interests.”
The policy’s guidelines declarethat the U.S. will continue toprovide GPS “for peaceful civil,commercial, and scientific useon a continuous, worldwidebasis, free of direct user fees.” Inthe year 2000, the President willbegin making annual determina-tions on continued use of GPSSelective Availability (SA). Thepolicy’s guidelines state that SAis planned to be discontinuedwithin a decade, allowing timefor U.S. military forces to preparefor operations without it.
GPS Nonmilitary User Concerns
Although conceived and fundedby the United States Departmentof Defense (DoD), GPS has growninto such a technological giantthat changes and/or cancellationsare now virtually impossible,except in the case of a nationaldefense emergency. The DoD’ssystem of Selective Availability(SA) was put in place to degradethe accuracy of civilian GPSsignals in the event of such anemergency. This is a source ofconcern among commercial andinternational GPS users.
To understand the pervasivenessof GPS, consider a recent GPSWorld Receiver Survey which listsover 300 units available from 50companies. The number and widevariety of uses of these receiversare impressive. If GPS were to beturned off permanently tomorrow,we would lose most of ourprecision surveying capabilities,most of our long-distance commu-nication, a major portion of thesmall boat, plane and othernavigational uses, and therewould be significant interruptionof commerce. The GPS-aidedIntelligent Vehicle/HighwaySystem (IVHS) currently beingequipped in new automobiles byboth Detroit, Michigan and Japanwould be shut down. The FAAwould be disrupted. That agencyhas adopted GPS as its Class 3(all-weather) landing system, andit is also developing a wide areaaugmentation system (WAAS)based on GPS which will bedesigned to provide better than1-meter accuracy.
22
Hewlett-Packard GPS SynchronizationProducts and Applications
The HP 58503A
GPS Time and
Frequency Refer-
ence Receiver
Hewlett-Packard provides a rangeof timing and synchronizationproducts based on advanced GPStechnology. These productsinclude:
• Primary reference sources. Aprimary reference is oftenreferred to as the master clock.Top level synchronizationsolutions demand the highestaccuracy and best long-termstability since they are used todrive multiple lower levelsynchronization units.
• Synchronization units. Thesesolutions deliver synchroniza-tion signals at specific pointswithin a network or site, ordistribute a reference source tomultiple points. Since manyorganizations require largenumbers of these units, theymust be affordable whilemaintaining accuracy andstability throughout the syn-chronization infrastructure.
• Specialized products andcustom solutions. HP offerscustomization services andspecialized products, such asthe HP 59551A, for synchroniz-ing measurements for powertransmission systems.
General-Purpose Metrology
and Manufacturing
The HP 58503A GPS Time andFrequency Reference Receiver, alightweight compact module, fitscomfortably into not only con-trolled lab environments, but alsoa variety of general-purposecalibration and manufacturingapplications that require precisiontime or synchronization.
The HP 58503A source maintainsfrequency accuracy of better than1 × 10–12, even in the presence ofSelective Availability. Thisperformance, combined with theunit’s low cost, makes it anattractive lab alternative com-pared with more expensivecesium and rubidium solutions.Manufacturing companies aretaking advantage of the HP58503A. Because of its low cost,the HP 58503A can be locatedanywhere precision time andfrequency are needed. This caneliminate or simplify distributionsystems and expensive cable runs.
In operation, the HP 58503Arequires no periodic adjustmentsor calibration which, along withits greater than 100,000-hourMTBF, provides an extremely lowcost of ownership. If the GPSsignal is lost, the HP 58503Aautomatically goes into its intelli-gent holdover mode. An RS-232port and a TTL alarm output(BNC) allow easy performancemonitoring. Either of theseoutputs also provides automaticmonitoring of the status of theHP 58503A.
Synchronizing Wireless
Networks
Accurate timing is essential forthe evolving digital cellularapplications. Soft handoffs —passing a CDMA user from onebase station to another — requiresexact synchronization. TheHP 58503A general purposereference source can deliver anaccurate and stable synchroniza-tion signal for digital base stationsand other points throughout aservice provider’s network. Otherfeatures it brings to the wirelessmarket include:
• Shortened time to market.• Lowered costs.• Improved system and service
quality.• Reduced project risk.
Today, many wireless networksalso require customized designsfor timing and frequency synchro-nization and distribution. Fornetwork equipment manufactur-ers building volume quantities ofwireless communications basestations or other equipment, the
23
The HP 55300A GPS
Telecom Primary
Reference Source
Measuring and Monitoring
Power Transmission Systems
Accurate fault location for largepower transmission networksrequires accurately synchronizingmany points throughout thetransmission system. TheHP 59551A module provides alow-cost synchronization founda-tion for monitoring wide-areatransmission systems, or for real-time monitoring and control. Itincludes capabilities suited to thesynchronization requirements forpower system analysis andoptimization.
Time tagging, a standard featureof the HP 59551A, facilitates avariety of applications includingfault location, network distur-bance analysis, and detailedsequence-of-events analysis.Events can be time-stamped (with100-nanosecond resolution),recorded, and downloaded to acomputer system for review.
For distributing synchronizationreference sources, the HP 55400ANetwork Synchronization Unitcan monitor up to five differentin-house or GPS reference sourcesand deliver as many as 400 outputsto different points on the network.Configured as a master subrackand associated expansion sub-racks, the HP 55400A capacityallows the economical distributionof multiple output signals through-out a large facility.
The redundant architecture and“hot swapping” features of theHP 55400A minimize interruptionsto the network being synchro-nized. The active input can beswitched on-line without anyphase hits for the output signals.Redundant pairs of input refer-ences and output cards offer one-to-one protection, and redundantpower supplies further enhancethe robustness of the synchroniza-tion solution. The open design ofthe HP 55400A allows the units tobe integrated into high-levelnetwork administration solutions.An optional information manage-ment card supports local andremote system alarm reportingand performance monitoring.
HP customization services canshorten implementation cyclesand lower in-house developmentcosts without sacrificing qualityand affordability.
Custom designs use many of thesame technologies incorporatedin the HP 58503A and are tailoredto client specifications. A customdesign can accommodate theneeds of any analog or digitalcellular, special mobile radio, orpersonal communication systemmanufacturer. Clients specify theexact form, fit, and function ofthe equipment needed for syn-chronizing base station operation.Custom products are manufac-tured to HP’s quality standardsand are supported worldwide.
Synchronizing Telecom-
munications Networks
Telecom digital networks requirehighly-reliable, low-cost sourcesof precision frequencies in orderto maintain data integrity andguarantee the delivery of high-quality services. The HP 55300AGPS Telecom Primary ReferenceSource gives service providers alightweight, compact module thatfits comfortably into controlledlab environments, base stations,and remote or unattendedapplications.
The HP 55300A source maintainsfrequency stability of greater than1 × 10–12, even in the presence ofSelective Availability. The module“flattens” existing digital net-works and delivers Level 1-typeperformance to Level 2 and 3locations. The HP 55300A moduleprovides a low-cost referencesource that improves networkservices and provides scalabilityto synchronization solutions.
24
Input/output capabilities of theHP 59551A simplify its use with avariety of existing event and faultrecorders. Standard functionsinclude time tagging of condi-tioned TTL input signals, IRIG-Boutput, and an Alarm BITE outputthat indicates a system fault or lossof satellite lock. The low cost ofthe HP 59551A makes it practicalto place one at each of the criticalpoints in a power system. Bysynchronizing multiple points inthe distribution network, reliableand meaningful field data can becollected and crucial systemperformance and operatingcharacteristics can be extracted.
HP Technology: Enhancing
Quartz Performance With
HP SmartClock
Each of the HP GPS productscombines affordable quartzoscillator platforms withHP SmartClock Technology, abreakthrough design for improvedstability. The time base of Hewlett-Packard’s GPS synchronization
oscillator over time,HP SmartClock adjusts theoscillator output frequencyaccordingly and significantlyimproves accuracy.
Holdover Mode During GPS
Signal Loss
In normal operation, the GPSreference signal determines thelong-term stability of the quartzoscillator. In the unlikely event ofsatellite signal loss or interrup-tion, HP SmartClock enters anintelligent holdover mode tomaintain accuracy until the GPSreference is regained. In holdovermode, HP SmartClock Technologyadjusts the oscillator output basedon the most-recently-learnedcharacteristics.
Long-Term Accuracy and
Reliability
HP SmartClock Technologyeliminates the need for frequentcalibrations — its learningcapabilities and the GPS referencesource automatically ensure on-going accuracy. This accuracycomplements the field-provenreliability of the HP quartz-basedoscillators that offer a Mean TimeBetween Failures (MTBF) of morethan 500,000 hours.
The HP 59551A GPS
Measurements
Synchronization
Module
products, the HP 10811D QuartzOscillator, is a highly-reliablecrystal component characterizedby low sensitivity to temperaturechanges, low phase noise, andwell-understood aging characteris-tics. Integrated with the quartzoscillator, the HP SmartClockalgorithm improves the quartz-based system performance, makingit equal to or better than a rubidi-um-based solution. The algorithmcompares the oscillator frequencywith the GPS reference signal. By“learning” the aging behavior andthe environmental effects on the
HP SmartClock
Technology
Intelligent Clock Diagram
External Frequency Reference In Enhanced GPS/
Digital FilterHP SmartClock Technology Microprocessor
Oscillator
Clock Frequency Out
The HP SmartClock
block diagram
25
GPS Glossary
Circular Error Probable (CEP)
— In a circular normal distribu-tion (the magnitudes of the twoone-dimensional input errors areequal and the angle of cut is 90degrees), circular error probableis the radius of the circle contain-ing 50 percent of the individualmeasurements being made, or theradius of the circle inside ofwhich there is a 50 percentprobability of a point beinglocated.
Coordinated Universal Time
(UTC) — UTC, an atomic timescale, is the basis for civil time. Itis occasionally adjusted by one-second increments to ensure thatthe difference between theuniform time scale, defined byatomic clocks, does not differfrom the earth’s rotation by morethan 0.9 seconds.
Differential GPS — A techniqueused to improve radio navigationsystem accuracy by determiningpositioning error at a knownlocation and subsequently trans-mitting the determined error, orcorrective factors, to users of thesame radio navigation system,operating in the same area.
Distance Root Mean Square
(drms) — The root-mean-squarevalue of the distances from thetrue location point of the positionfixes in a collection of measure-ments. As used in this document,2 drms is the radius of a circlethat contains at least 95 percentof all possible fixes that can beobtained with a system at any oneplace. Actually, the percentage offixes contained within 2 drmsvaries between approximately95.5 percent and 98.2 percent,depending on the degree ofellipticity of the error distribution.
Enhanced GPS — A method ofdigital filtering to reduce theeffects of SA on a stationaryreceiver. Contributes to attaininghigh time accuracy.
Ephemeris — A table giving thecoordinates of a celestial body at anumber of specific times within aspecific period.
Full Operational Capability
(FOC) — For GPS, this is definedas the capability that occurs when24 operational (Block II/IIA)satellites are operating in theirassigned orbits and have beentested for military functionalityand meet military requirements.
Geodesy — The science related tothe determination of the size andshape of the earth (by such directmeasurements as triangulation,leveling, and gravimetric observa-tions) which determines theexternal gravitational field of theearth and, to a limited degree, theinternal structure.
Geometric Dilution of Preci-
sion (GDOP) — All geometricfactors that degrade the accuracyof position fixes derived fromexternally-referenced navigationsystems.
Initial Operational Capability
(IOC) — For GPS, this is definedas the capability that will occurwhen 24 GPS satellites (Block I/II/IIA) are operating in their assignedorbits and are available fornavigation use.
Navigation — The process ofplanning, recording, and control-ling the movement of a craft orvehicle from one place to another.
Spherical Error Probable (SEP)
— The radius of a sphere withinwhich there is a 50 percent prob-ability of locating a point. SEP isthe three-dimensional equivalent ofCEP.
Standard Deviation (sigma) —A measure of the dispersion ofrandom errors about the meanvalue. If a large number of mea-surements or observations of thesame quantity are made, thestandard deviation is the squareroot of the sum of the squares ofdeviations from the mean valuedivided by the number of observa-tions less one.
Time Interval — The duration ofa segment of time without refer-ence to where the time intervalbegins or ends.
World Geodetic System (WGS)
— A consistent set of parametersdescribing the size and shape ofthe earth, the positions in anetwork of points with respect tothe center of mass of the earth.
26
1 pps 1 pulse per second
A-S anti-spoofing
BIPM Bureau International des Poids et Mesures
BPSK binary phase shift keying
C/A-code coarse acquisition-code
CEP circular error probable
CS Control Segment
DGPS Differential Global Positioning System
DMA Defense Mapping Agency
DME distance measuring equipment
DoD Department of Defense
DOP dilution of precision
DOT Department of Transportation
drms distance root mean squared
EGPS Enhanced Global Positioning System
FOC full operational capability
FOM figure of merit
FRP Federal Radio Navigation Plan
GPS Global Positioning System
HDOP horizontal dilution of precision
IOC initial operational capability
L1 GPS primary frequency, 1575.42 MHz
L2 GPS secondary frequency, 1227.6 MHz
LADGPS Local Area Differential GPS
Abbreviations and Acronyms
LOP line of position
LORAN-C most recent version of Long RangeNavigation
MCS Master Control Station
NAV-msg navigation data message
P-Code precise code
PDOP position dilution of precision
PPS Precise Positioning Service
PRC pseudorange corrections
PRN pseudorandom noise
PTTI precise time and time interval
SA Selective Availability
SEP spherical error probable
SPS Standard Positioning Service
SV Space Vehicle (satellite)
TDOP time dilution of precision
TFOM time figure of merit
TOA time of arrival
UERE user equivalent range error
URE user range error
UTC Universal Time Coordinated
VDOP vertical dilution of precision
WADGPS Wide Area Differential GPS
WGS-84 World Geodetic System (1984 update)
27
28
H
Japan:
Hewlett-Packard Japan Ltd.Measurement Assistance Center9-1, Takakura-Cho, Hachioji-Shi,Tokyo 192, Japan(81) 426 48 3860
Latin America:
Hewlett-PackardLatin American Region Headquarters5200 Blue Lagoon Drive9th FloorMiami, Florida 33126U.S.A.(305) 267 4245/4220
Australia/New Zealand:
Hewlett-Packard Australia Ltd.31-41 Joseph StreetBlackburn, Victoria 3130Australia131 347 ext 2902
Asia Pacific:
Hewlett-Packard Asia Pacific Ltd.17-21/F Shell Tower, Time Square,1 Matheson Street, Causeway Bay,Hong Kong(852) 2599 7070
For more information on Hewlett-PackardTest and Measurement products,application or services please call yourlocal Hewlett-Packard sales ofices. Acurrent listing is available via Webthrough AccessHP at http://www.hp.com.If you do not have access to the internetplease contact one of the HP centerslisted below and they will direct you toyour nearest HP representative.
United States:
Hewlett-Packard CompanyTest and Measurement Organization5301 Stevens Creek Blvd.Bldg. 51L-SCSanta Clara, CA 95052-80591 800 452 4844
Canada:
Hewlett-Packard Canada Ltd.5150 Spectrum WayMississauga, OntarioL4W 5G1(905) 206-4725
Europe:
Hewlett-PackardEuropean Marketing CentreP.O. Box 9991180 AZ AmstelveenThe Netherlands
Data Subject to ChangePrinted in U.S.A. May 1996
Hewlett-Packard CompanyCopyright © 19965965-2791E
For more information:HP 58503A brochureHP 59551A brochure
