imi
`
`Pe
`US005736960A
`
`i
`
`j
`
`United States Patent 9
`Murphyet al.
`
`{1143 Patent Number:
`
`[45] Date of Patent:
`
`5,736,960
`Apr. 7, 1998
`
`[54] ATOMIC CLOCK AUGMENTED GLOBAL
`POSITIONING SYSTEM RECEIVERS AND
`GLOBAL POSITIONING SYSTEM
`INCORPORATING SAME
`
`[75]
`
`Inventors: John H. Murphy. Churchill Boro, Pa.;
`Trent A. Skidmore, Athens, Ohio
`
`[73] Assignee: Northrop Grumman Corporation, Los
`Angeles, Calif.
`
`[21] Appl. No.: 530,553
`
`[22] Filed:
`Sep. 19, 1995
`[51]
`Int, CMe eceecseccsnees HO4B 7/185; GO1S 5/02
`[52] U.S. CU. naeecssessecossecencstsensesssecseransanessecnsesvecenenses 342/357
`
`
`(58] Field of Search ...........c000. 342/357; 364/449.7
`
`Electronics and Communication Engineering Journal, vol. 7,
`No. 1, 1 Feb. 1995, pp. 11-22, XP000500767; Morgan-O-
`wen G J et al: “Differential GPS Positioning”, see p. 11, col.
`2, line 20-p. 14. col. 1, line 36.
`
`M.A. Sturza, GPS Navigation Plane, published by the
`Department of Transportation and the Department of
`Defense. DOT—~VNTSC-RSPA-92-2/DOD-4650.5 (1992).
`
`P. Misra and M.Pratt, Role ofthe Clock ina GPS Navigation
`Receiver, ATC Project Memorandum No. 42PM-—SAT-
`NAV-0008, Massachusetts Institute of Technology, Lincoln
`Laboratory (May 1994).
`
`Primary Examiner—Theodore M. Blum
`Attorney, Agent, or Firm—Walter G. Sutcliff
`
`[57]
`
`ABSTRACT
`
`[56]
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`A GPS radio navigation system utilizes an atomic clock in
`each user’s GPS receiver station and a ground reference
`station located at a fixed, precisely known position which
`..
`5,155,490 10/1992 Spradley, Jr. et al.
`342/357
`determines GPS time from satellite information and trans-
`
`5,192,921
`3/1993 Chantry etal. ....
`we 3313
`mits an absolute time signal together with time-of-flight
`......cssecessceeseneeare 342/387
`5,319,374
`6/1994 Desai ct al.
`information over a communications path of precisely known
`5,364,093
`11/1994 Huston etal. .
`5,600,329=2/1997 Bremtter ........sccsecsersseseareeeee 342/357
`length to the user stations, each at a fixed position. Once the
`atomic clock in a GPS receiver has been set to the precise
`GPS time,
`the user station is free to maneuver for an
`extended period of time during which its position is calcu-
`lated from pseudorange and time information received from
`as few as three satellites and the time maintained by its
`atomic clock.
`
`OTHER PUBLICATIONS
`
`Proceedings of the Rural Electric Power Conference, New
`Orleans, May 3-5, 1992, No. 3, May, 1992, Institute of
`Electrical and Electronics Engineers, pp. D2.01—-D2.10,
`XP000300323; M.N. Zeiler Integrated GIS and GPS For
`Mapping and Analysis of Electric Distribution Circvuits, see
`p. D2-4, line 29—p. D2-5, line 21.
`
`14 Claims, 4 Drawing Sheets
`
`Google Exhibit 1020
`Google Exhibit 1020
`Google v. Mullen
`Google v. Mullen
`
`

`

`Sheet 1 of 4
`
`5,736,960
`
`US. Patent
`
`Apr. 7, 1998
`
`LOA
`
`

`

`US.
`
`Patent
`
`Apr. 7, 1998
`
`Sheet 2 of 4
`
`5,736,960
`
`CORRELATION = 0.0060
`
`|
`
`1
`
`3 |
`15+
`
`9 +
`
`15+
`
`14
`
`
`
`
`
`
`
`
`
`VERTICALDILUTIONOFPRECISION(VDOP)
`
`oes
`1p eeeeeEee
`
`0
`
`5
`
`10
`(HOURS)
`
`TIMES.
`
`15
`
`20
`FIG.2
`
`05
`
`357
`
`3
`
`29
`
`2
`
`154
`
`05
`
`Oy
`17
`1
`peee
`29
`-
`
`
`05
`
`0
`
`5
`
`10
`TIMES (HOURS)
`
`15
`
`<j
`9
`FIG.S
`
`0
`
`05
`
`
`
`
`
`VERTICALDILUTIONOFPRECISION(VDOP)
`
`

`

`US. Patent
`
`Apr. 7, 1998
`
`Sheet 3 of 4
`
`5,736,960
`
`UNAVAILABILITY
`
`UNAVAILABILITY
`
`19-4. +--+ t
`1.5
`2
`2.5
`5
`3.5
`
`4
`
`4.5
`
`
`
`
`HORIZONTAL DILUITION (HDOP WITH & WITHOUT CLOCK) FIC 4
`
`1.5
`
`2
`2.5
`3
`3.5
`4
`45
`VERTICAL DILUTION OF PRECISION (VDOP WITH & WITHOUT CLOCK) FIG 5
`
`

`

`US. Patent
`
`Apr. 7, 1998
`
`Sheet 4 of 4
`
`5,736,960
`
`

`

`5,736,960
`
`1
`ATOMIC CLOCK AUGMENTED GLOBAL
`POSITIONING SYSTEM RECEIVERS AND
`GLOBAL POSITIONING SYSTEM
`INCORPORATING SAME
`
`BACKGROUND OF THE INVENTION
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`50
`
`55
`
`1, Field of the Invention
`
`This invention relates to radio navigation systems, and in
`particular, to the Global Position System (“GPS”) in which
`multiple satellites transmit position and time information
`which is received by user stations and processed to deter-
`mine the time and location at the user station.
`
`2. Background Information
`The Global Positioning System (GPS) is a navigation
`system in which a constellation of orbiting satellites emit
`spread spectrum signals which contain information aboutthe
`location of the satellites and the time at which the signal was
`transmitted.
`
`The GPS Navigation System has been designed to use
`receivers in the users stations that are based on low-cost
`crystal oscillator technology, and are not dependent upon a
`highly accurate time piece to do the timeof flight calculation
`of the signal transmission. This is accomplished by solving
`for the three dimensions of position and time from four or
`more satellite pseudorange measurements.
`The GPSnavigation system is owned and operated by the
`Department of Defense (DOD). It is considered to be a
`mnajor military tool for precision location which can be used
`to locate assets and to assist in the accurate delivery of
`munitions to a target. As such, they have created a system
`that operates in two modes. One mode. the PPS or Precise
`Positioning Service mode, is highly secure and has tremen-
`dous accuracy (21 m horizontally, 29 m vertically, and 200
`ns temporally). The other mode, the SPS or Standard Posi-
`tioning Service mode, is publicly available but has a lower
`accuracy (100 m horizontally, 140 m vertically, and 340 ns
`temporally). The DOD controls the accuracy that the pub-
`licly available system can achieve. This is currently accom-
`plished through selective availability (SA). There are two
`ways to degrade the system accuracy. The first way is
`introduce errors in GPS time and the second way is to
`introduce errors in the GPS satellite positions. At present,
`SA is on and the DOD is degrading the system performance
`by introducing errors in the GPS time.
`The FAA has negotiated with the DOD about the use of
`the GPS navigation system by the aviation industry to
`perform non-precision and precision navigation. The pub-
`licly available mode of operation is nearly adequate to meet
`these needs. However, to allow navigation under extreme
`operating conditions some improvements/augmentations to
`the GPS navigation system are required.
`The key augmentations to GPS being considered by the
`FAA are: the wide area augmentation system (WAAS) and.
`local-area augmentations. WAAS consists of geostationary
`satellites and a supporting ground network used to increase
`the integrity, availability, and based on DOD approval, the
`accuracy of GPS. It is the goal of the FAA for WAAS to
`provide the primary means of navigation for all domestic
`operations down to Category I requirements (stated as 32
`feet vertical and 110 feet horizontal error at the 200 foot
`decision height). The most-likely technologies being con-
`sidered for local-area augmentations for Category II and IIT
`operations are a) code-phase differential GPS, b)
`pseudolites, which transmit a GPS-like signal, and c) kine-
`matic carrier phase tracking, which obtains centimeter-level
`
`2
`accuracy based on carrier phase tracking as opposedto just
`GPS code phase tracking.
`It is a knownfactthat the introduction of an atomic clock
`in the GPS receiver can improve the navigation system
`availability. The introduction of a precise time piece in the
`receiver reduces the numberof satellites neededto establish
`the location by one. Thatis, if the time is given by a precise
`elock then 3 or more satellite measurements can be used to
`establish the three dimensional position of the receiver. The
`augmentation of the GPS receiver by a precise clock there-
`fore has a major impact on the performance parameters
`being used by the FAA to develop standards for satellite-
`based navigation in the national air space. Availability is
`defined as the percentage ofthe time that the GPS navigation
`services are available for use. Continuity of service is
`defined as the ability of the total system to provide accept-
`able performance throughouta phase of flight, given that the
`performance was acceptable at the initiation of the phase of
`flight. The possibility of performing this type of augmenta-
`tion of GPS receivers for aviation navigation systems has
`not been seriously pursued or even considered up to now
`because commercially available, precision atomic clock
`costs have been too high.
`Even if it is determined that a precision atomic clock .
`should be used in the GPS receiver, the task remains to
`provide apparatus and a technique for synchronizing the
`precision atomic clock in the receiver with the time standard
`used by the satellites.
`There is a need,therefore, for an improved GPS naviga-
`tion system.
`There is a particular need for a GPS navigation system
`which provides improved availability and continuity at a
`reasonable cost. There is also a need for apparatus and a
`technique for synchronizing the precision atomic clock used
`in the GPS receiver.
`
`SUMMARYOF THE INVENTION
`
`These needs and others are satisfied by the invention
`which is directed to a global positioning system in which the
`user stations are receivers incorporating a precision clock so
`that the position of the receiver can be determined using the
`position signals and time of position signals from as few as
`three satellites. In order to precisely set the time maintained
`by the precision clock, a synchronization meansis provided
`at a precisely known fixed position close to the initial
`position of the receivers. This synchronization meanscal-
`culates the absolute time established by a masterstation and
`transmitted to the satellites. Where the satellites introduce
`etror into their time ofposition signals, the synchronizing
`meansintegrates the time signal continuously to removethis
`variation and provide a precision time signal accurately
`synchronized to the absolute time.
`The user stations are initially positioned at a fixed, pre-
`cisely knownposition relative to the fixed precisely known
`position of the synchronizing means. The synchronizing
`means then sends the precision time signal to the receiver
`together with timeofflight information so that the receiver
`can accurately set its precision clock with respect to the
`absolute time. Once the precision clock in the receiver has
`been accurately set from a precision time signal with adjust-
`mentfor the timeofflight between the synchronizing means
`andtheinitial position of the receiver, the receiver can leave
`the initial position and maneuver. The subsequentpositions
`of the receiver can then be calculated from the signals
`received from at least three satellites and the time main-
`tained by its precision clock. The accuracy of these position
`
`

`

`5,736,960
`
`3
`calculations over time depends upon the stability of the
`clock and the inaccuracies of the initial setting of the clock.
`Use of an atomic clock set to within 3 ns of absolute time
`and having a drift of no more than 6 ns per day would
`maintain accuracy for Category I operations for up to about
`six hours.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`A full understanding of the invention can be gained from
`the following description of the preferred embodiments
`when read in conjunction with the accompanying drawings
`in which:
`FIG. 1 is a schematic view of a GPS system incorporating
`the invention.
`
`10
`
`15
`
`FIG.2 is a plot comparing vertical dilution of precision
`(VDOP) versus time for a GPS receiver with the invention
`to one without the invention during an example of enroute
`navigation.
`FIG. 3 is a plot comparing VDOP versus time for a
`precision landing for a GPS receiver, without the invention,
`with the invention and with a perfect clock.
`FIG.4 is a plot comparing unavailability versus horizon-
`tal dilution of precision (HDOP) for a precision landing
`without an atomic clock and one with a perfect clock.
`FIG. Sis a plot comparing unavailability versus VDOPfor
`a precision landing without an atomic clock and with a
`perfect clock.
`FIG.6 is a schematic diagram of a system for setting the
`atomic clocks in receiver stations of a GPS in accordance
`with the invention.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`
`FIG. 1 illustrates schematically a Global Positioning
`System (GPS) 1 augmented in accordance with the inven-
`tion. The conventional GPS system includes a master station
`3 which maintainsan absolute time standard using an atomic
`clock 5. This master station 3 transmits an absolute time
`standard signal to a number oforbiting satellites 7 -7,, each
`of which also have an atomic clock 9,-9,. Each of the
`satellites 7-7,, continually calculatesits position and repeti-
`tively transmits a signal containing three dimensional sat-
`ellite position information and time information. User
`stations, such as for example the aircraft 11, have a GPS
`receiver 13. The satellite signals to reach the GPS receiver
`13 must be taken into account in order to determine the
`position of the user station 11 to the accuracy desired.
`Presently the user stations contain a crystal oscillator for use
`as a clock. Such a clock is not sufficiently accurate to make
`the time of flight calculations. Accordingly, the receiver
`station 13 must derive the time with the required accuracy
`from the satellite signal. Thus, the receiver station must
`solve for four unknowns, the three position coordinates (two
`horizontal and one vertical) and the time. This requires the
`receipt of signals from at least four satellites. While the
`number of satellites and their orbits are established in an
`attempt to optimize coverage,
`there can be times when
`signals from less than four satellites are available. This can
`be especially true for land-based GPS receivers wherein
`natural obstacles 15 such as mountains or man-made
`obstacles such as large buildings can block the receipt of
`signals, especially from satellites which are low on the
`horizon. In the case where signals from more than four
`satellites are received by the GPS receiver 13, the additional
`information is used, such as in a Kalman filter or a least
`
`25
`
`30
`
`35
`
`45
`
`50
`
`55
`
`65
`
`4
`squares estimation routine, to determine the location of the
`receiver with improved accuracy.
`
`In accordance with the invention, the GPS system 1 is
`augmented by the use of a GPS receiver 17 with an atomic
`clock 19 in the user station 11’. We have discovered that a
`precise time standard such as provided by the atomic clock
`19 in a GPS receiver significantly improves its vertical
`position accuracy. This comes about because when the
`atomic clock is used as a time reference,
`the position
`solution is uncoupled from the time solution whichleadsto
`an improvement in the position accuracy. Part of the
`improvement comes of the fact that we are now solving n
`equations in three unknowns as opposed to n equations in
`four unknowns. There is greater redundancy in the
`information, which when solved for, using for example a
`least squares or Kalman filter, leads to an improvementin
`the accuracy. The other part of the improvement comes from
`the fact that we can, in principle, set the atomic clock time
`more precisely than the time from the satellites can be
`instantaneously determined. The precision of the time piece
`is related to how closely the time can be set and the amount
`of drift in time that the physical device experiences.
`
`Dilution of Precision (DOP) is a measure of the errors or
`accuracies in the position and time calculations in a GPS
`navigation system. Typically, a Horizontal Dilution of Pre-
`cision (HDOP), a Vertical Dilution of Precision (VDOP),
`and a Time Dilution of Precision (TDOP)are ofinterest. The
`present invention provides an improvementin the instanta-
`neous dilution of precision in all of these regimes which has
`particular impact for use of the augmented GPS navigation
`system for airborne vehicles. In particular, it has a tremen-
`dous influence on oceanic and enroute navigation which will
`impact the number of required geostationary satellites to
`achieve the availability needs for precision landing and radio
`navigation. It is also likely to lo influence the needs for
`pseudolites in the pseudolite augmented GPS system.
`
`Oneofthe benefits of using atomic clock augmentationis
`thatit reduces the burden on the governmentin establishing
`a reliable navigation system. That
`is,
`the cost for this
`augmentation is placed on the users rather than on the
`public. Fortunately, developments in the production of
`small, moderately priced atomic clocks promise to make this
`practical.
`The improvements in the dilution of precision provided
`by use of an atomic clock in a GPS receiver can be
`appreciated from the following: Let Y be the known changes
`in the pseudorange measurements, H be the measurement
`matrix (composed of direction cosines determined by the
`satellite geometries), and X be the unknown parameters. For
`the case of n-equations in 4 unknowns, Y is a (nx1) matrix,
`H is a (nx4) matrix, and x is a (41) matrix
`
`Y=HX
`
`where
`
`@)
`
`Y=1P: P2--- Pal”
`p=the changesin the pseudorange measurement to the i
`th satellite (meters)
`X=[x y zb]*
`b=the clock bias (meters) [note that the clock bias can be
`converted from a distance to time by dividing by the
`speed of light (3x10* meters per second)|,
`
`

`

`x. y, Z=unknown location (meters), and
`
`OO
`Or Cy
`G2
`Ogy Oy
`H=] Gy Osy Gaz
`
`Onx Oy Oe
`
`1
`1
`1
`
`1
`
`where 0,,. &,,, and o,, are the direction cosines between the
`unknown location (x, y, Z) and the i th satellite.
`For the case when we synchronize a clock to the GPS time,
`we need only to solve for the location. That is, we have
`n-equations in 3 unknowns, and a dock with some known
`bias b. where Y is a (n<1) matrix. H is a (n<3) matrix, X is
`a (3x1) matrix, and 1 is a (nx1) matrix.
`
`Y=HX+b1
`
`2)
`
`where
`- Pal”
`Y=[P1 P2- +
`p=the changes in the pseudorange measurementto the i
`th satellite (meters)
`X=[x y z]?
`x. y. Z =unknownlocation (meters)
`b=the dock bias (meters)
`I=[2 1... 17
`
`Ot, Ay Oty
`Cz,
`Czy Ory
`osx Gay Gaz
`
`H=]|
`
`One
`
`Ory
`
`One
`
`Since equations (1) and (2) represent the same GPS system
`operating under different scenarios, it should be no surprise
`that these equations are identical with the only difference
`being the interpretation of what is known and what is
`unknown.
`:
`The following solutions have been east using the least
`squares estimation approachto resolve the over specification
`resulting from n-equations in m-unknowns, where n>m.
`Similar results can be found using other estimation tech-
`niques. For n-equations in 4 unknowns, we have
`
`cov (X}-Ga,71,
`
`where
`
`J, is the (nxn) identity matrix
`6,” is the variance in the changes in the pseudorange
`{where the variance on each of the pseudoranges are
`assumed to be equal for each of the satellites] and
`G=(H7H]*
`where the vertical dilution of precision (VDOP)is given by
`
`15
`
`25
`
`35
`
`45
`
`55
`
`VDOPHG 33)°*
`
`the horizontal dilution of precision (HDOP) is given by
`
`HDOPSG ,4G ,2)°5
`
`and the time dilution of precision (TDOP) is given by
`
`TDOPHG ,,)°5
`
`for the n-equations in 3 unknowns, we have
`
`@)
`
`GS)
`
`65
`
`(6)
`
`5,736,960
`
`6
`cov (X}Go,7[,+k7H"1,H G]
`
`where
`
`Q)
`
`I, is the (nxn) identity matrix
`6,7 is the variance in the changes in the pseudorange
`(meters)
`1, is an (nxn) matrix of ones
`k=(0,/6,)"=(G,o/0,)"(,/0,)°(t-toMtg
`G,, is the variance in the temporal bias (meters)
`O;9 is the initial variance in the temporal bias (meters)
`G,, is the drift variance in the temporal bias (meters)
`t, is the time atinitialization (seconds)
`t is the time (seconds)
`and
`G={H7Ay!
`where the vertical dilution of precision is given by
`
`VDOPH(GIL,+H71,HG)3)°*
`
`and the horizontal dilution of precision is given by
`
`HDOP=(G[LHCH"1,H G),,+G[L,+k?H71,HG)2)"
`
`(8)
`
`9)
`
`The difference between equations (4) and (8) lead to the
`improvement in instantaneous VDOP that comes from the
`use of a precise time standard. Likewise the difference
`between equations (5) and (9) lead to the small improvement
`in instantaneous HDOPthat also comes from the use of a
`precise time standard.
`In conjunction with Ohio University, a simulator was
`developed based on their Satellite Coverage Research
`Analysis Model (SCRAM)code which calculates the DOPs
`of GPS receivers with and without precision time references.
`This model allows investigation of GPS receiver accuracies
`and dilution of precision characteristics at a specific location
`Or on a route between two locations for a period of time of
`interest. FIG. 2 illustrates a comparison of VDOPs for
`enroute navigation between Pittsburgh, Pa. and Seattle,
`Wash. using a mask angle (minimum angle above the
`horizon to visible satellites with acceptable signal errors) of
`5 degrees. The curve 21 is the instantaneous VDOPfor a
`code—phase GPS receiver without an atomic clock and
`assuming a measurement accuracy of o,=100 m (300 ns).
`The curve 23 is the instantaneous VDOP for a GPS receiver
`with an atomic clock that is resynchronized every 5.2 hours
`to within 3 ns [(6,./6,)°=10~*] of absolute time and drifts 6
`ns per day [(o,,/0,)7=4x10~]
`VDOPcalculations were also performed over a 24 hour
`period for Seattle, Wash. to illustrate the impact of an atomic
`clock during a precision landing. FIG.3 illustrates this case.
`The atomic clock was assumed to be synchronized to within
`3 os and drifts on an average of no more than 6 ns per day.
`Thus, 6,"=2 m (6 ns) which corresponds to a code phase
`differential GPS system, the expected mode of operation
`during a precision landing. A mask angle of 7.5 decrees was
`used. The curve 25 is the instantaneous VDOP for a GPS
`receiver without an atomic clock, assuming a differential
`code phase GPS receiver which has a 6,=2 m (6 9 ns). The
`curve 27 is the instantaneous VDOPfor a GPS receiver with
`an atomic clock that is resynchronized every 6 hours to
`within 3 ns [(Gyo/6,)?=0.25] of absolute time and drifts 6 ns
`per day [(0,,/0,)"=1]. The curve 29 is the instantaneous
`VDOPfor a GPS receiver with a perfect clock. That is, an
`atomic clock that has no drift and is exactly synchronized to
`absolute time.
`Availability improvement is a central result that comes
`from clock augmentation of GPS receivers. To verify this we
`
`

`

`5,736,960
`
`7
`constructed a computer model based on the SCRAM codeof
`the GPS location system to determine the magnitude ofthis
`improvement in availability. For the case we studied, we
`used a Markov modelfor the state probabilities, in which the
`probability of no failures is 0.659858, the probability of a
`single failure is 0.230474, and a probability of twofailures
`is 0.076341. This gives a 96.7% accuracy of having 22 or
`more satellites. FIG. 4 illustrates the impact of a perfect
`clock on the unavailability (one minus the availability)
`plotted against
`the instantaneous HDOP. This plot was
`generated for precision landing at the Seattle Airport over a
`24 hour period assuming a mask angle of 10 degrees. The
`curve 31is for a GPS receiver without a precision clock and
`the curve 33 illustrates the results for a receiver augmented
`with an atomic clock. This figure illustrates that the clock
`has the potential to decrease the unavailability by between a
`factor of 2 and 10 for HDOPs between 1.5 and 4.5. This is
`of particular importance to enroute and oceanic navigation.
`More striking is the effect on unavailability when ploted
`against VDOP. FIG.5 illustrates these results for a precision
`landing at Seattle airport using a mask angle of 10 degrees.
`The VDOPand unavailability are determined over a speci-
`fied 24 hour period at the Seattle airport based upon the
`calculated satellite geometries. The curve 35 is the unavail-
`ability versus VDOP for a GPS receiver without a perfect
`clock. The curve 37is the unavailability versus VDOPfor a
`GPS receiver with a perfect clock. It can be seen that the
`atomic clock augmentation improves the unavailability by
`over 3 orders of magnitude for a VDOPbetween1.5 and 4.5.
`Thus, we see that under Special Category 1 (SCAT-D
`precision approach conditions (VDOP~4), the unavailability
`for a standard GPS receiver is 2.0%, whereas when aug-
`mented by a perfect clock the unavailability approaches
`0.003%. (This implies that the availability for SCAT-I land-
`ings goes from 98% to 99.997%.) Furthermore, under CAT-I
`landing conditions (VDOP-~2.3), we find that the unavail-
`ability for a standard GPS receiver is 13%, whereas when
`augmented by a perfect clock the unavailability approaches
`0.006%. (This implies that the availability for CAT-I land-
`ings goes from goes from 87% to 99.94%.) Current thinking
`at a FAA is that a target availability is 99.999% for enroute
`navigation and they will tolerate an availability as large as
`99,9% for precision approaches under CAT-I conditions.
`Currentinstrument landing systems (ILS) for CAT-T land-
`ings at single ILS airports (~500 in the U.S.) have an
`availability of 95-99.5%, and for multiple ILS airports
`(~120 in the U.S.) have an availability of 99.95-99.998%,
`and for multiple-ILS airports (~55 in the U.S.) have avail-
`abilities greater than 99.999%.
`Thus, we see that the availability for SCAT-I precision
`approaches with GPS receivers without an atomic clock is
`comparable to CAT-I precision approaches at single-ILS
`airports. However, the GPS navigation system is inadequate
`for precision approaches at airports requiring 99.95% or
`better availability without some augmentation (precision
`clocks, pseudolites, or geostationary satellites). Clock aug-
`mentation of the GPS navigation system promises to extend
`CAT-I landings to airports requiring 99.998% without addi-
`tional augmentation from either pseudolites or gcostationary
`satellites. This implies that in the U.S. CAT-I landings can be
`made at 92% of the airports for airplanes operating with only
`GPS receivers augmented by precision clocks.
`Thus, it can be seen that a GPS receiver with a perfect
`clock could have a significant improvementin the dilutions
`of precision. Furthermore,it has been shownthat even with
`areal atomic clock which is resynched to within 3 ns every
`six hours a significant improvementin the accuracy can be
`achieved. The challenge, therefore, is to devise a means of
`precision time transfer which permits realization of the types
`of improvements discussed above.
`
`8
`In accordance with the invention, a system 39 for syn-
`chronizing the atomic clocks 19 in the user vehicles 11’ is
`provided. The synchronizing system 37 includes a ground
`reference station 41 located at a fixed, precisely known
`location. Precision time acquisition at ground reference
`stations is a well knownart. Briefly, the ground reference
`station comprises a multi-channel GPS receiver and an
`atomic clock 43. The referencestation antenna 45 is located
`at a surveyed location. The accuracy of the survey deter-
`mines the accuracy of the absolute time established at the
`reference station 41. Subcentimeter level resolution of the
`reference antenna location is possible with current GPS-
`based surveying equipment. The reference station 41 moni-
`tors all GPS satellites which are in view. From their almanac
`information and knowledge of the reference station antenna
`location, a continual stream of estimates of the GPS time can
`be established from the solution of the location equations at
`the ground reference station 41. These estimates are con-
`tinually fed to a Kalman filter, along with the current ground
`reference time to establish the absolute time. The absolute
`time output from the Kalman filter is used to condition the
`groundreference station atomic clock 43. The ground ref-
`erence station atomic clock 43 is therefore being corrected
`for its drift in time on a continual basis. Also, it was
`indicated that in the SPS mode,the satellites apply a bias to
`the time signal to degrade accuracy. The ground reference
`station 41 by integrating the calculated time signal continu-
`ally over time can removethis bias, and generate accurate
`absolute time information.
`The reference station atomic clock time becomes the
`absolute time which needs to be transferred to the GPS
`receivers 17 in the user stations 11' before they depart on a
`mission. The transfer of the precision time signal generated
`by the ground reference station 41 can be accomplished
`either wirelessly or over-the-wire. Wireless time transfer
`requires corrections for atmospheric conditions, direct path,
`multi-path and other possible sources of error. Because of
`the large number of error sources, this approach is not
`preferred. In the preferred embodimentof the invention, an
`over-the-wire communications network 47 is used to trans-
`fer the precision time signal between the ground reference
`station 41 andthe user stations 11’. As can be seen from FIG.
`6, the communications network 47 connects the ground
`reference station 41 with a number of nodes 49 to which user
`stations 11' can be connected. This type of time transfer
`requires a precision determination of the path between the
`reference station 41 and each individual node 49. For a 0.1
`to 1 ns accuracy,it is recommended that measurementof the
`communication paths be to precisions of 10 to 100 cm.
`Precision time transfer is achieved through the emission of
`a timing pulse along with a correction message (almanac)
`which defines the time-of-flight corrections for each of the
`nodes 49 in the communications network 47. The media for
`the over-the-wire time transfer needs to be taken into con-
`sideration. For a very rapid pulse and almanac information
`transfers, fiber optic cable is preferred. However, for airport
`applications where time transfer is done continually, and the
`almanac information is small, over-the-wire transfer via
`conducting materials can be considered. However, thermal
`effects on the properties of the conductor influence the time
`transfer and have to be accounted for in these alternative
`over-the-wire time transfer systems. The precision time
`signal transmitted by the groundreference station 41 is used
`by the user stations 11' to establish the time for on-board
`atomic clocks 19 in the GPS receiver 17. Simultaneously, an
`internal clock counter is set to the time of transfer plus the
`time delay appropriate to the node at which the GPS receiver
`is attached. Using this methodology, time transfers of sub-
`nano seconds are possible. The accuracy of the time transfer
`is therefore constrained to the accuracyof the absolute time
`determination at the ground reference station 41.
`
`20
`
`25
`
`30
`
`35
`
`45
`
`30
`
`55
`
`65
`
`

`

`5,736,960
`
`9
`Once the atomic clock 19 in a user station 11' such as an
`aircraft has been set using the precision time signal and
`almanac received at an associated node 49 over the com-
`munications network 47, the user station/aircraft disconnects
`from the communications network and can maneuver. As
`indicated previously, in the case of an aircraft 11 with its
`atomic clock 17 set to within 3 ns of absolute time by
`information received from the ground reference station 39,
`and assuming a drift of 6 ns per day the accuracy of the
`position calculations made by the receiver in the user
`stations/aircraft would be suitable for a Category I approach
`at an airport for up to 6 hours after the user station/aircraft
`disconnected from the communication system 43.
`While specific embodiments of the invention have been
`described in detail, it will be appreciated by those skilled in
`the art that various modifications and alternatives to those
`details could be developed in light of the overall teachings
`of the disclosure. Accordingly, the particular arrangements
`disclosed are meant to be illustrative only and not limiting
`as to the scope of invention which is to be given the full
`breadth of the claims appended and any andall equivalents
`thereof.
`Whatis claimed:
`1. A global positioning system comprising:
`a plurality of satellites each repetitively transmitting sig-
`nals indicating three dimensional satellite position
`information and time information,
`master station means for providing an absolute time
`reference and position reference to said satellites;
`a user station comprising a precision clockinitially set by
`a precision time signal to a precise timerelative to the
`absolute time reference, means for receiving said sig-
`nals generated by at least three of said satellites, and
`processing means for repetitively generating a user
`position signal representing a three dimensional posi-
`tion of said user station using satellite position infor-
`mation from at least three satellites and time informa-
`tion from said precision clockafter said precision clock
`is initially set; and
`synchronizing means comprising reference means at a
`fixed, known location generating said precision time
`signal from said signals transmitted by at least three of
`said satellites and means transmitting said precision
`time signal to said user station to provide the initial
`clock setting when the user station is located at a
`precisely knownposition relative to said fixed, known
`position of said synchronizing means,said user station
`further including means for moving from said precisely
`knownposition while said processing means continues
`to repetitively generate said user position signal in
`response to time info