GB2170672A - Satellite navigation receiver - Google Patents

Abstract

An improved satellite navigation receiver suitable for receiving signals from satellites travelling in low earth orbits over the poles of the earth is arranged to perform a preliminary position estimating process prior to carrying out a relatively accurate position fix by conventional matching of measured Doppler shifts with corresponding calculated shifts based on a trial position. The receiver monitors the Doppler shift, detects the time at which zero Doppler shift occurs (which corresponds to the position of the satellite at peak elevation with respect to the receiver), measures the rate of change of the Doppler shift in the region of zero Doppler shift and, using a look-up table stored in memory, generates a signal value representative of the angle subtended at the earth's surface by the satellite at peak elevation and the receiver. The position of the satellite is determined from data transmitted by the satellite, and this position, together with the said angle enables the receiver to provide two preliminary position estimates on opposite sides of the satellite orbit plane. These are then used as trial positions for the know Doppler profile matching process, which is performed to a required degree of precision beginning with the trial position which yields the smallest initial error residual. A receiver able to operate in this manner has the advantage that it does not require the user to provide an initial position estimate in order to obtain a fix.

Description

SPECIFICATION Improved satellite navigation receiver This invention relates to a satellite navigation receiverandtoa position fixing method.

In a known satellite navigation system position fixing is carried out in the following manner. The satellite orbits the Earth in a plane which may contain the North-South axis of the Earth and transmits a data stream which contains a time signal at periodic intervals, details ofthe satellite orbit and details ofthe transmitting carrierfrequency.Areceiveronthe Earth's surface determines its position by first locking onto the satellite signal when the satellite appears above the horizon, monitoring the data stream and measuring the carrierfrequency over a period of time, monitoring the frequency variation due to the Doppler effect as the satellite moves along its orbit.Using the orbit information given by the satellite data stream the values of samplers of the received carrier frequency are compared with calculated values based on a number of trial positions until one position is found which gives the best match between the calculated and measured Doppler profiles. This process enables the receiver to fix its position to an accuracy of about + or - 0.05 nautical miles. However,forthe Doppler profile matching process to operate it is necessaryforthe receiver initially to hold information asto its current position to an accuracy of about 50 nautical miles and astothetime to within 15 minutes GMT.

The present invention provides a receiverwhich dispenses with the requirementforan initial estimate of approximate position. The receivercarriesouta position fixing process having two stages, the second of which is the Doppler profile matching process referred to above. In the first stage the receiver makes use ofthe satellite transmissions to compute a relatively crude estimate of its position by a process which includes estimating the angle subtended at the centre of the earth by a chord connecting the position ofthe observer with the position of the satellite at its maximum elevation angle with respect to the obser verasthesatellite moves across the sky, or by estimating a corresponding mathematically related dimension such as the elevation angle itself.

The first stage preferably comprises the following operations. Firstlythe receiver estimates or reads the time accurately by correct interpretation of the time data in the data transmission from the satellite, and uses the estimated or read time to correct its internal clock to an accuracy betterthan 100 us. The next activity of the receiver is to estimate the precise time at which the satellite reaches its peak elevation with respect to the location of the receiver. This time corresponds to that at which the received signal frequency is equal to the (known) transmitted carrier frequency; in otherwords, when the Dopplerfrequen cyiszero.

Atthe same time the receiver measures or estimates the rate of change of Dopplerfrequency. A signal corresponding to the rate of change of Doppler frequency is passed to a non-linear circuit which converts the input signal into one which approximate a signal equivalentto the said angle subtended at the centre of the earth corresponding to the maximum elevation angle of the satellite on that pass. The non-linearcircuit may include a look-up table for yielding a subtended angle value from an input value of the maximum rateofchangeofDopplerfrequency forthe pass.

A latitude value is obtained by processing the orbit position data transmitted by the satellite and measuring the time at which the satellite reaches peak elevation. Having obtained the subtended angle at peak elevation, two values forthe latitude and longitude can be computed by solving the triangle containing the receiver position, the satellite at peak elevation, and the centre of the earth. (The height of the satellite is known from the orbit details.) One latitude and longitude is to the east of the satellite track whilst the other is to the west. The detailed information contained in the Doppler profile distinguishes one position from the other. The normal position fixing routine as described above is applied to each ofthe two trial positions (one east and one west).The quality of the Doppler profile match can be determined from the mean square error between the measured and calculated profiles. The position which givesthe minimum mean square error (or equivalent quality indicator) is taken to be correct.

The invention will now be described byway of example with reference to the drawings in which: Figure lisa block diagram of a satellite navigation receiver embodying the invention; and Figures 2Ato 2H are a flow chart illustrating the operation of the receiver of Figure 1.

Forthe purposes ofthis description it will be assumed that the signals to be received are transmitted by a satellite travelling in an earth orbit which contains the axis of rotation ofthe earth, i.e. in a North-South orbit. The satellite transmits data containing orbit information allowing the position of the satellite at predetermined time intervals to be computed in a relatively straight-forward and known manner.The satellite may be one oftheTransitgroup of satellites, and details ofthe signals transmitted and basic orbit information forthis group of satellites are contained in Technical Memoranda dated July1973 and December 1976 published by The John Hopkins University Applied Physics Laboratory of Silver Spring, Maryland, U.S.A. with the titles "Program Requirements for Two-Minute Integrated Doppler Satellite Navigation Solution" and "The Transit System, 1975". The disclosure of both Memoranda is contained herein by reference.The invention is not, however, limited to use with this particular group of satellites, nor indeed to satellites which travel in a North-South orbit, although use of a satellite travell- ing in an orbit which is nota North-South orbit will require additional computation facilities to yield position fixes expressed in terms of latitude and longitude.

Referring to Figure la a satellite navigation receiver in accordance with the invention has an antenna 10 coupled to an R.F. amplifier 12 and first mixer 14. The frequencies given in Figure 1 are those which are used when receiving signals from a satellite ofthe group referred to above. The first mixer 14 received a local oscillator signal at 375 MHzfrom a source comprising a 5MHz crystal oven reference oscillator 16 and multipliers 18 and 20 having a combined multiplication factor of75. Assuming the incoming signal has a frequency of 399.968 MHz, as shown, the signal at the output ofthefirst mixer 14 is at a first intermediate frequency of 24.968 MHz.However, as a result of the Doppler effect of the satellite moving towards, and then awayfrom,the antenna 10,the receivedfrequency, and hence also the signal frequency atthe output of the first mixer 14, are initially higherthan the figures given and, as the satellite moves away, lower. The bandwidth ofthe intermediate frequency stages ofthe receiver allows for a 8kHzshiftup or down in frequency.

The signal obtained from the first mixer 14 is fed to a first l .F. filter 22 and an amplifier 24, before being frequency converted a second time in a second mixer 26. The local oscillator signal for this mixer is obtained from the first multiplier 18 connected to the reference oscillator 16, and has a frequency of 25 MHz. It will be seen that, since the subtraction of the first l.F. signal frequencyfrom the second local oscillatorfrequency results in a negative quantity, the Doppler shift in the signal obtained atthe output ofthe second mixer 26 is inverted, so thatthe approaching satellite produces a signal frequency which is less than the nominal second intermediate frequency of 32 kHz, while, converselythe receding satellite produces a signal frequency greaterthan 32 kHz.

A phase sensitive detector 28 forms part of a phase locked loopwith avoltagecontrolledoscillator30 (VCO)supplying in-phase and quadrature signals to the detector in conventional mannervia line 32 and 34.

This detector arrangement receives the signal obtained from the second mixer26 and extracts data and synchronisation pulses from the incoming signal which are then fed to subsequent stages ofthe receiver by lines 36 and 38. As a result ofthe action of the phase locked loop, the frequency of the VCO 30 follows the frequency ofthe second l.F. signal and can be used to monitor the Doppler shift. For this reason, theVCO signal is produced at two outputs, one for feeding the phase sensitive detector, and one for feeding the processing arrangement, to be described below, via line 40.

The remainderofthe receiver is divided into four sections which are: (i) a Z80 microcomputer 100 for processing data received from the satellite, monitor ing the Doppler shift, carrying out calculations, and controlling the receiver operation; (ii) a power supply unit 102; (iii) a display module 104for displaying position fixes in latitude and longitude and other information, and (iv) a keyboard 106.

The microcomputer 100 has a CPU 110, an erasable programmable read-only memory (EPROM) 112for storing an operating program, a random access memory (RAM) 1 14forstoring information from the satellite and Doppler counts representative ofthe Doppler shift in the received signal frequency, as well for storing general information and quantities resulting from operation ofthe receiver controls, and values generated by calculation routines, etc. A real time clock (RTC) 116 acts as the receiver's internal time reference which can be updated using time information transmitted by the satellite or by keyboard input, and so-called counteritimer controller (CTC) 118 is used to count the cycles ofthe VCO output on line 40.

The CTC is a device manufactured by Zilog, Inc, type No. Z80A-CTC. Use ofthe CTC 118 avoids occupying excessive processortimeincountingfastexternal events. Instead, the CTC counts up to 256 cycles ofthe VCO output, overflows, and again counts up to 256 cycles, this process being repeated as many times as necessaryto count the VCO output for a predetermined time interval (30 seconds).The number of overflows is counted by the microcomputer and the final count added on at the end of the predetermined period so that a "Doppler count" to the nearest cycle of the VCO output can be stored in the RAM 114 for each of several consecutive 30 second periods as the satellite moves across the sky.Data and address busses, shown in Figure 1 as a single connection 120, connect all ofthe above elements ofthe microcomputer 100 together, as weli as a control logic element 122 coupled to the high stability reference oscillators 16 whichcontrolsthetiming ofthe microcomputer, a dual asynchronous receiver and transmitted (DART) 124 acting as an input and an outputforthe reception and transmission of serial data from and to external devices, and an analogue-to-digital (nod) converter 126 for converting incoming data and synchronisation pulses, and analogue information from external devices, into parallel digital data.

The DART 124 and the AID converter 126 are coupled to three input'output parts 128,130 and 132 which enablethe receiverto be connected to other marine equipment such as a magnetic compass and a speed log so thatthe indicated location can be updated automatically by dead reckoning (DR). In addition, the digital parts 128 and 130 may be coupled to,forexample, a printer, repeaterdisplayand an auto-pilot.

All of the Transit group ofsatellitetransmit on 399.968 MHz although the received signal can be + or 8kHz differentdue tothe Doppler shift. To pickup the signal from a satellite as it appears overthe horizon, the microcomputer 100 is programmed to cause the VCO 30, via control lines 42, to sweep in frequencyfrom 24 kHzto 4Q continuously until the satellite signal is detected, whereafterthe phaselocked loop is allowed to control the VCO.

The receiver presents information to the user by means of a display unit 104 having a 2 x 16 character vacuum fluorescent display device 134 coupled to a display driver device 136 which is, in turn, coupled to the microcomputer busses 120 and a keyboard encoder 138. The keyboard 106 is connected to the keyboard encoder 138, and has function and numeric keys for inputting a variety of information. Typical keyboard inputs are the date and approximate time.

This information alone is sufficient to enable the receiverto provide a position fix using the preliminary position estimating method in accordance with the present invention. It is also possible to enter a start position, height above sea level (for land based receivers), and way points, so that the position fix obtained from satellite transmissions can be used to compute bearings and point-to-pointdistancesforthe user. With regard to the input of time information, normally the real time clock will maintain an accuracy of plus or minus a few minutes over several months without updating using an internal small battery. The satellites in the Transit group provide an accurate time reference only insofarastheytransmit minutes and seconds within each half hourwithouttransmitting the hour and the date.It is therefore necessaryforthe receiverto containtime data accurate to plus or minus 15 minutes. The keyboard entryfacilitythus provides a backupfortime updating should the real time clock become de-energised. The computer 100 is programmed automatically to lock its own time reference onto the received time signals when a satellite signal is picked up.

The computer is also programmed to enter orbit data received from the satellite into its RAM 114.

Details ofthe manner in which the time and orbit data may be processed by a navigation receiver are given in the document referred to above, together with data formats and the theory of operation ofthe Transit satellite navigation system.

The operation ofthe receiverto obtain a preliminary position fix will now be described. This operation makes use ofthe time and orbit data stored in RAM 114 (fed to the microcomputer 100 on line 36) and the VCO output signal on line 40 (Figure 1).

As mentioned above, when a satellite signal is picked up, the frequency of the received signal is monitored over a number of consecutive 30 second periods as the satellite firstly movestowardsthe receiver, passes overhead orto one side, and then recedes. With the Transit system the satellite may be "visible" for sixto twenty minutes, thereby yielding 12 to 40 'Doppler counts" whose values are initially low and finally high, with a central value in the region of 960,000. this being the number of cycles of the 32 kHz second l.F. in 30 seconds when there is no relative approaching or receding movement between the satellite and the receiver.These Doppler counts, together with theirstart and finish times are stored in the RAM 114, and it is only afterthe satellite pass is finished that processing of the stored information to obtain a position fix begins.

Before considering this process with reference to the flow chart of Figures 2Ato 2H in detail, it is convenientto summarise the preliminary position estimation method as a series of basic steps.

Since it is the peak elevation ofthe satellite with respect to the observer that is examined in carrying out the position estimation, one ofthefirst basic steps isto establish which of the Doppler counts stored in RAM 114 represent the central section ofthe satellite pass under consideration. It will be appreciated that the received frequency at peak elevation will be approximately equal to the transmitted frequency; in other words, the Doppler shift is approximatelyzero.

Thus, the central section is chosen as being a group of consecutive Dopplercountscentred on the count which corresponds most closely to a VCO frequency of 32 kHz. This count is called d5 and the corresponding 30 second period over which the count ways obtained is identified as period S5. The counts representing the central section are then chosen as the nine counts dl to d9 centred on d5.

The second main step is to compute the satellite's position at the centre of period S5 in terms of a series of co-ordinates in a Cartesian co-ordinate system centred on the centre ofthe earth.

Thirdly, the difference (d9 - dl) between the Doppler counts d9 and dl is calculated, and the value obtained looked up in a look-uptable stored in EPROM 112 which contains nine values of an angleAcorrespond ingto nine values of d9 -d1.TheangleAis an approximation to the angle subtended at the earth's centre by a chord from the satellite to the receiver. In fact thins look up process involves looking forthefirst value in thetable greaterthan the measured d9 -dl and extractingthecorresponding angleA.

As a fourth step, the latitude and longitude (LAT1, LON1) ofthe point on the surface ofthe earth immediately below the satellite halfwayth rough period S5 is calculated.

Step 5 is to compute a position obtained by taking a greatcirclethrough (LATi, LON1) ata bearing of90% i.e. due East at (LAT1, LON1) andfollowing itaround the earth's surface by the angle A extracted from the look up tableto a position designated (LAT2, LON2A).

This constitutes a first estimated position.

Step 6 is a similar computation based on the same great circle but starting at a bearing of 2700from (LAT1, LON1), i.e. initially due West, using the same angle to establish a second position estimate (LAT2, LON2B).

The above completes the main portion of the preliminary position estimation method in accordance with the invention. Thereafter, the known curve matching process is carried out twice, once with the starting fix as (LAT2, LON2A) and once with the starting fix as (LAT2, LON2B). The residual errors resulting from these processes are compared, and the fix giving the smaller error is adopted as the one on the correct side of the satellite track. Further iterations on this adopted fix are performed in the known mannerto the required degree of precision.

Referring to Figures 2Ato 2H,the method described above will now be considered in more detail. It will be recalledthatthefirst main step inthe preliminary position estimating operation is the identification of a group of Dopplercounts representing the central section ofthe satellite pass, centred on the count corresponding approcimately to the peak elevation of the satellite. This is performed by the operations shown in Figure 2A and thefirsttwo instructions in Figure 2B. A series of Doppler counts have already been stored in RAM 114 by counting the number of cycles ofthe VCO output in a number of consecutive 30 second periods. lithe VCO frequency were to be constant at 32 kHz,the Dopplercountwould be 960,000. However, in reality, each count reaches a differentvalue due to the changing Doppler shift, and, as mentioned above, the counts produced by the approaching satellite will be lowerthan those produced bythe receding satellite. The first main operation is identifying the first Doppler count to exceed 960,000 by means of loop 200 as shown in Figure 2A. Traps 202,204 and 206 prevent further execution ofthe program if an invalid series of Doppler counts has been stored.When exiting from loop 200, a pointerwhich was initiallysetto the first count points at the first count to exceed 960,000.

Referring to Figure 2B, this pointer is now decremented back by four counts in step 208 to identify the first count (d1 ), this count and the next eight counts (i.e. up to d9) then being moved to a portion of the RAM 114called a 'work area' (step 210).

We now enterthe second main step referred to in the summary above with the subtration of do from d9 (instruction 212) to obtain a measure of the rate of change of the Dopplerfrequency. This is significant because this rate of change is large when the satellite passes overhead the receiver but relatively small when it passes well to one side. Following this calculation, the program hasfourtraps, 214,215,216, 217 to ensure firstlythatthe first Doppler count d1 is not null orthatthe receiver has not locked onto another satellite with a lower Doppler count (the latter should have been caught by the previous errortraps), and secondly that important counts in the selected group of counts are not invalid.

The following instructions 218,220 yield the position ofthe satellite (Xs, Ys, Zs) in Cartesian coordinates corresponding to the centre of period S5.

This is carried out in a known manner using time and orbit data transmitted by the satellite to establish the co-ordinated forthe beginning and end of period S5 and calculating the mean.

The third main step in the estimation process makes use of a look-up table shown in Table I below.

Difference Angle (d1 - d9) (Degrees) 300000 4 280000 7 250000 11 200000 13 180000 15 160000 18 140000 21 120000 25 90000 30 TABLE I Since there are 9 sets of values in the Table, the next pair of instructions 222 and 224 are to set up a count of 9 and a pointer pointing tothefirst (the highest)value of d9 - dl in the table. In normal circumstances, the program nowfollows loop 226 in which the value of d9-d1 obtained from the Doppler counts in the work area is compared with values of d9 - dl in the look-up table in steps 228 and 230.If d9 - dl is greaterthan the highestvalue in the table, it is assumed thatthe satellite is passing almost directly overhead and that angle A approximates to 2" (instruction 232 in Figure 2D).Valuesofd9-d1 which are intermediatetothe highest and lowestvalues in the look-up table cause the program ;o break out of loop 226 at step 230 so thatthe next instruction 234 is looking up the corresponding angle A in the table. If the program breaks out of loop 226 on instruction 228 after having passed once or more than once around the loop, there is a fault (the Doppler count runs 'backwards', for instance), and this is detected by trap 236 in Figure 2D.If the measured valueofd9-d1 is less than all of the values in the table, it is assumed that the satellite is very low overthe horizon to the East ortheWest and that consequently angle A is approximately 30 (step 238 in Figure 2C).

Having obtained an approximate value ofthe angle Asubtended at the earth's centre by the chord extending between the satellite and the receiver, there is sufficient data to compute the two position fixes to the East orthe West respectively of the satellite track.This is carried out in the microcomputer by means of steps 240,242,244,246,248 and 250 (Figures 2D and 2E) which make use of the following relationships: LAT1 = arctan (Zs/V(Xs2 + Ys2)) LON1 = arctan (Ys/Xs) (where LAT1 and LON 1 represent a point on the earth's surface directly below the satellite at the middle of period S5) sin LAT2 = cos A. sin LAT1 cos LAT2 = ++ - sin2 LAT2) LAT2 + arctan (sin LAT2/cos LAT2) cos (LON 1 - LON2) = (cos LAT1. cos A)/(cos LAT 2) sin (LON1 - LON2) = +5/[1 - cos2(LON1 - LON2)] jLON1-LON2i= arctan {[+V(1 - cos2 (LON 1 - LON2))]I[cos (LON 1 - LON2)1) (where LAT2 and LON2 represent the receiver's position) Itwill be seen that this yields two values forthe longitude of the receiver, LON2A and LON2B, as follows: LON2A= LON1 + jLON1 - LON2j LON2B = LON1 - jLON1 - LON2I Although the look-up table contains only nine values of dO - dl, for comparison with the measured value of d9 - dl, the preliminary position fixes obtained are sufficiently accurate as starting positions forca rrying outthe known curve matching process using the Doppler counts stored in RAM 114.

Since, in general, the vessel in which the receiver is housed is moving during the satellite pass, tut is necessary to update the final accurate fix obtained by the conventional methodtoobtainafixforthe position ofthe receiver atthe end ofthe satellite pass, i.e. when the fix is calculated. Conventionally, a series of 33 position estimates corresponding to 33 time instants defined by 32 thirty second periods S1 to S32 are provided as the basis forthe accurate curve matching process. Thetrack of the vessel may be calculated by, for example, entering on the keyboard an estimate (1, Al) ofthe position at the start of period S1 ofthe satellite pass togetherwith the bearing and speed ofthe vessel. Alternatively, the track may be computed automaticallyfrom the speed log and compass read-outs fed to the ports 128, 130 or 132, again using an initial position estimate. The advantage ofthe present receiver is thatthe initial position estimate is not required, and therefore an entirely arbitrary starting position may be used since the next part ofthe program to be described simply substitutes(LAT2, LON2A} or (LAT2, LON2B) for (l, Al). Thirty-three position estimates are required, so a count of 33 is set up in step 252. Instruction 254 sets up two pointers to the arbitrary or non-arbitrary position (1,A1)stored in RAM 114.RAM 114also contains 32 further positions (2, A2)... (33, A33), corresponding to the track as calculated from bearing and speed inputs referenced to (l ,Al). These latitudes and longitudes are updated throughout by the substitution firstly of (LAT2, LON2A) for (l , Al) in loop 256 (Figure 2E). Trap 258 prevents gaps from appearing in the event of null readings. Having up-dated the track on the basis of (LAT2, LON2A), these positions are then used to perform the conventional satellite fix to a first approximation using the curve matching process as shown in Figure 2F (step 260), and the resulting residual is stored (step 262).

Next, another loop 264 is set up to substitute LON2A as the reference longitude at the start of the track. This produces a second set of positionsfforthe basis of a second conventional satellitefix266 (Figure 2G).

Again the residual is saved (step 268). This residual is compared with the residual obtained at step 262 to establish which is the lesser residual, in other words, which is the correct fix (step 270). (The difference in the residuals arises from the effect ofthe earth's rotation which is allowed for in the reference values.) If the smallest residual is the one resulting from step 266, the conventional position fixing process is completed to the required precision on the basis of the stored track data (step 272). the other residual is smaller,thetrack longitudes are again updated by substituting LON2A for LON2B using loop 274 shown in Figure 2H, and the correct fix is completed in step 276.

Claims (28)

1. A method of obtaining a position fix in a satellite navigation system in which a signal including time and orbit information is transmitted at a predetermined frequency by an earth orbiting satellite and is picked up by satellite navigation receiver, wherein the method includes (1) a preliminary estimation process comprising generating, in the receiver, signals representative of the Doppler varia- tion ofthe frequency ofthe received signal, demodulating the time and orbit information, feeding the said generated signals and the said information into a processing circuit arrangement, and applying, in the said circuit arrangement, an operator to the said generated signals and information to provide position signals representing at least one approximate estimate ofthe position ofthe receiver, and (ii) a further estimating process comprising generating from the said position signals and from signals representing the Doppler variation ofthe received signals, output signals representative of a more accurate position estimate.

2. A method according to claim 1, wherein the preliminary process is non-iterative and the further process is iterative.

3. A method according to claim 1, wherein: at least two position estimates are produced by the preliminary process; part of the said further process is repeated for each estimate to provide respective sets of signal values which are then compared with a set of reference signals values; one ofthe said position estimated is selected on the basis of its respective set the calculated set of values; and the remainder ofthe further process is performed with an input derived from the selected position estimate to generate the said output signals.

4. A method according to claim 1, wherein the preliminary estimation process includes generating signals representative ofthe rate of change ofthe frequency of the received signal when the satellite is art a selected location with respect to the receiver.

5. A method according to claim 1, including updating a time reference source in the receiver by combining time data from the source with the time information received from the satellite, and applying the said operatorto the combination ofthe generated signals representative ofthe Doppler variation, the orbit information and an output signal from the time reference source.

6. Amethod according to claim 1, wherein the position estimate obtained bythefurther process has an accuracy at least an order of magnitude better than the mean accuracy of the best estimate obtained by the preliminary estimation process.

7. A method according to claim 1, in which the preliminary process includes converting at least some ofthe signals representative ofthe Doppler variation ofthe received signal into signal values representative of a geometrical relationship between the position ofthe receiver and the position of the satellite with respect to the receiver.

8. A method according to claim 7,wherein the geometrical relationship is the angle subtended atthe centre ofthe earth by the receiver and the satellite at the selected location.

9. Amethod according to claim 7,wherein the selected location is the position ofthesatellitewhen it is at its peak elevation with respect to the receiver.

10. A method according to claim 1, wherein the preliminary estimation process includes generating a signal representative of the difference between first and second measurements ofthe received signal frequency, and comparing the said difference signal with values in a look-up table stored in the processing circuit arrangementto yield an outputvalue defining a geometrical relationship between, on the one hand, the satellite at a selected location with respect to the receiver, and, on the other hand, the receiver.

11. A method of obtaining a position fix in a satellite navigation system in which a signal transmitted at a predetermined frequency by an earth orbiting satellite is picked up by a satellite navigation receiver, and in which the Doppler variation of the frequency of the received signal, due to the relative movement of the satellite with respect to the receiver, is measured in the receiver and compared with calculated Doppler variations patterns based on trial positions to provide the required position fix, wherein, prior to the said comparison, an approximate preliminary position fix is obtained by the steps of:: (i) generating signals representative of the rate of change ofthe frequency ofthe received signal when the satellite is at a selected location with respect to the receiver; (ii) converting the rate of change signals into signals representative of a geometrical relationship ofthe position ofthe receiver relative to the position ofthe satellite when the latter is at the selected location; and (iii) performing a calculation whereby signals representing the position ofthe satellite and the signals representing the said geometrical relationship are used to generate signals representing the required position ofthe receiver.

12. A method according to claim 1, wherein the geometrical relationship is the angle of elevation of the satellite viewed from the receiver.

13. Amethodaccordingtoclaim 1,whereinthe geometrical relationship is the angle subtended at the centre ofthe earth by the satellite and the receiver.

14. A method according to claim 1, wherein the selected location is the position ofthesatellitewhen it is at its peak elevation with respectto the receiver.

15. A method according to claim 4, wherein the generation ofthe signals representative ofthe rate of change of the received signal frequency comprises generating a plurality of frequency counts representative ofthe received signal frequency at regular time intervals, selecting a pair ofthe said counts representing the received signal frequency before and afterthe point of zero Dopplershift, and forming signals representing the difference between the countsofthesaid pair.

16. A method of obtaining a position fix in a satellite navigation system in which a signal transmitted at a predetermined frequency by an earth orbiting satellite is picked up by a satellite navigation receiver, wherein a position estimate is obtained by generating in the receiversignalsrepresentative ofthe received signal frequency, generating from the said signals further signals representative ofthe rate of change of the received signal frequency in the region of zero Doppler shift, deriving from the said further signals signals representing the position ofthe receiver with respect to the position ofthe satellite at substantially zero Doppler shift, generating from stored data, signals representing the satellite position at substantiallyzero Doppler shift, and processing the said signals representing the positions ofthe satellite and ofthe receiver relative to the satellite to produce the required position fix.

17. A method according to claim 16, in which two position fixes are produced equally spaced from and on opposite sides ofthe plane ofthe satellite orbit, in which both ofthe said fixes are used by the receiver astrial positions forafurther position determining process in which signal values representative ofthe received signal frequency over a period oftime are compared with corresponding calculated signal values derived from the trial positions, in which error signals are obtained from each ofthe two comparisions, and in which the trial position producing the smaller error is selected as the basis for subsequent higher precision processing.

18. A satellite navigation receiver for receiving a signal transmitted from a satellite in earth orbit at a predetermined frequency and modulated by a data signal containing time and orbit data, wherein the receiver comprises atime reference source, means for monitoring the Doppler variation ofthe frequency ofthe received signal duetothe relative movement of the satellite with respect to the receiver, a demodula torfor recovering the data signal, and circuit means coupled to the monitoring means and the demodula torforapplying an operatorto at least some ofthe output signals thereof to provide at least one preliminary estimate ofthe position ofthe receiver, the circuit means being operable to generate output signals representing a further, more accurate, esti mate of the receiver position by comparison of Doppler signal values obtained from the monitoring means with reference Doppler signal values corres ponding to the or each preliminary estimate.

19. A receiver according to claim 18, wherein the circuit means includes a non-iinear circuit arrange mentforapplying the said operator.

20. A receiver according to claim 19, wherein the non-linear circuit arrangement includes a memory having stored therein a look-up table containing a plurality of signal values representative ofthe Dop plervariation ofthe received signal variation, and a plurality of values of a geometrical relationship between the satellite at a selected location and the receiver.

21. A receiver according to claim 18, further comprising means coupled to the demodulatorfor processing the data received from the satellite and for altering the outputofthetime reference source in respect to the received data.

22. Areceiveraccording to claim 18,whereinthe said circuit means is operable to generate sets of reference Doppler signal values corresponding to a plurality of respective preliminary position estimates, and to compare each set of reference values with the values obtained from the monitoring means thereby to generate a plurality of respective error signals, the circuit means being further operable to selectthe estimate corresponding to the smallest error and to repeatedly comparethe values obtain from the monitoring means with values derived from the selected estimate to produce the said further esti mate ofthe receiver position.

23. Asatellite navigation receiver for receiving a signal transmitted at a predetermined frequency by an earth satellite which moves relative to the surface ofthe earth in a defined orbit, wherein the receiver comprises a time reference source, means for moni toring the frequency of the received signal which exhibits a Dopplervariation due to the relative movement ofthe satellite with respect to the receiver, rate determining means coupled to the frequency monitoring means and operable to generate signals representative ofthe rate of change ofthe received signal, and position determining means operable to convertthe rate of change signals into signals defining a geometrical relationship between the positions of the satellite and the receiver.

24. A receiver according to claim 9, wherein the geometrical relationship is the angle subtended at the centre of the earth by a chord joining the receiver with the satellite when the satellite is at its peak elevation with respect to the receiver.

25. A satellite navigation receiverfor receiving a signal transmitted at a predetermined frequency by an earth satellite travelling in an orbitwhich defines a plane containing the axis of rotation ofthe earth, wherein the receiver comprises a time reference source, means for generating signals representative ofthe frequency of the received signal, which frequencyexhibitsa Dopplervariation duetothe relative movement ofthe satellite with respect to the receiver, processor means operable to receive the frequency representative signals and to generate signals representative ofthe rate of change of the received signal frequency at the timethe received signal frequency is substantially equal to the transmitted frequency, means operable to store data defining the orbit of the satellite, the processor means being coupledtothe storage means and operable to calculate the position of the satellite at the said time, and being further operable to convert the rate representative signals into signals representative of a geometrical relationship ofthe position of receiver relative to the said satellite position, and, using this relationship and the calculated satellite position, to calculate an approximate position fixforthe receiver.

26. A receiver according to claim 25, wherein the processor means is operable to perform a second position fix based on the said approximate position.

27. A method of obtaining a position fix in a satellite navigation system,the method being substantially as herein described with reference to the drawings.

28. A satellite navigation receiver constructed and arranged substantially as herein described.