WO2009130305A1

WO2009130305A1 - Method of positioning using satellites

Info

Publication number: WO2009130305A1
Application number: PCT/EP2009/054962
Authority: WO
Inventors: Ty Lewis; Fredrik Lindstrom
Original assignee: Nordnav Technologies Ab
Priority date: 2008-04-24
Filing date: 2009-04-24
Publication date: 2009-10-29
Also published as: GB2459334B; GB0807526D0; GB2459334A

Abstract

A method is provided of calculating the position of a receiver with respect to a transmitter on-board each of at least four satellites. The method comprises synchronising a receiver time of the receiver with a transmission time of a first transmitter of the plurality of transmitters. The time of transmission of signals received from each of the other satellites is then estimated based upon the receiver time, an estimated user position and orbit data relating to the orbit of each satellite. Pseudoranges for each satellite are then calculated and then used to calculate a revised user position. The method is repeated until the user position meets a predetermined convergence criterion.

Description

Method of Positioning using Satellites

Field of the Invention The present invention relates to a method of calculating the position of a receiver with respect to a transmitter on-board each of at least four satellites.

Background to the Invention

Satellite positioning systems, known as "Global Navigation Satellite Systems" (GNSS), are well known throughout the world for providing accurate positioning of personnel and vehicles on land, sea and in the air. These systems rely upon obtaining extremely accurately timed signals from a number of satellites, each signal including an accurate transmission time together with data describing the orbit of the satellite in question. This information is used by a receiver to calculate the position of the person or vehicle in question to an accuracy of a few metres.

The increasing use of GNSS places ever increasing demands upon the speed of calculation of an accurate position solution for the user and the increasingly poor signal reception conditions under which a position solution may be obtained.

One specific limitation upon GNSS receivers is the time that is required for a receiver to achieve synchronisation with each particular satellite. This may take up to 6 seconds in good signal reception conditions and longer in poor reception conditions. The 6 second delay is caused by the format of the transmitted signal from each satellite. Specifically, the time stamp of the signal which is known as the Time of Week (TOW) is only transmitted every 6 seconds (once per subframe) and thus with known methods it may take 6 seconds to receive this signal before synchronisation can be achieved, the average time being 3 seconds.

Known methods of satellite-based positioning rely upon establishing sychronisation with at least 4 satellites followed by the formation of a position solution using "pseudoranges" calculated for each satellite. This produces excellent positioning results but the need for synchronisation being established with each satellite can mean that extended delays are incurred in finding a position when there are difficulties in establishing synchronisation due to poor signal conditions for example.

It is therefore extremely desirable to reduce the length of time needed to establish a position solution.

Summary of the Invention In accordance with the invention we provide a method of calculating the position of a receiver with respect to a transmitter on-board each of at least four satellites, comprising:- a) synchronising a receiver time of the receiver with a transmission time of a first transmitter of the plurality of transmitters; b) estimating the time of transmission of signals received from each of the other satellites based upon the receiver time, an estimated user position and orbit data relating to the orbit of each satellite; c) calculating a pseudorange for each of the satellites, wherein for the other satellites the pseudorange is calculated using the estimated time of transmission and the receiver time; d) calculating a revised user position and a revised receiver time using the pseudoranges; e) repeating steps (b) to (d) until the user position meets a predetermined convergence criterion.

The applicants have realised that it is not necessary to establish synchronisation with four satellites in order to obtain an accurate position solution. The present invention therefore enhances the speed at which a user position can be established. In comparison with known methods, importantly the present invention only requires synchronisation with one satellite. A very basic estimate of the position of the user is also needed, together with a knowledge of the content of signals which are transmitted by each satellite. Neither of these additional requirements is onerous since most users are able to provide an approximate estimate of their location, for example the region of a country in which they are located. Furthermore, the data transmitted by positioning satellites are readily available.

The method allows a position solution to be achieved without receiving a full repetition of the transmission of each satellite and indeed, for three of the satellites, without deciphering the content of the signal. A regular time stamp is not required to be received from each of the four satellites. It will be appreciated that the method provides many advantages in low signal conditions since continuous data are not required. The method is therefore not reliant upon a particular section of the transmitted signal being received.

Rather than using synchronised signals with each satellite, the present method used a procedure of repeatedly calculating estimates of distance to each satellite based upon the time-of-flight of the transmitted signals (pseudorange) and the distance based upon the positions of the user and the satellite (geometric range). Therefore step (b) typically comprises estimating a geometric range of a satellite based upon the user position and a satellite position calculated from an estimate of the transmission time of the received signal and the orbital data. The position of a satellite can be provided with a high degree of accuracy using orbital data in a form which is a function of time.

Although global navigation satellites are each equipped with highly accurate "atomic" clocks, the clocks in receivers are not sophisticated since they are often designed for mass market production and may be required to be fitted into compact handheld devices. Preferably the receiver time used in the method comprises an accurate receiver clock time and a receiver clock bias representing the inaccuracy. A measure of the receiver clock bias therefore gives the accuracy of the receiver clock. Preferably therefore any revisions of the receiver time comprise revisions to the receiver clock bias.

A number of different data transmission protocols are envisaged by the present invention. It is preferred however that the protocol used is that of a code division multiple access (CDMA) signal having a "chip rate" of preferably 1.023 MHz, where the 1023 bit code repeats every millisecond. This protocol is that presently used by the Global Positioning System (GPS) and typically the present invention is contemplated as being used in the GPS system. However, the invention may equally be used in other satellite systems such as Galileo, Compass, Quasi Zenith Satellite System (QZSS), and so on. Other transmission protocols which comprise data bits transmitted at a constant frequency are envisaged as suitable for use with the invention.

The signal received from each satellite may comprise data having "code epochs" and the calculation of each pseudorange may comprise estimating a transmission time of the signal in the form of an estimate of the number of code epochs that have elapsed since a reference time which is common to each satellite. Such a reference time might be a Time of Week synch point or other reference point in the signal. This estimate of the number of code epochs is preferably based upon the receiver clock time, the geometric range, the receiver clock bias, the reference time, an estimate of the bias of a clock on-board the satellite and an estimate of errors caused by conditions affecting the propagation of the signal between the satellite and receiver. Typically such conditions are atmospheric conditions.

The initial estimate of the user position is preferably provided to within a distance from the actual receiver position of one half of a code epoch multiplied by the speed of propagation of the signal. Since the speed of propagation is essentially that of the constant "c" (the speed of light in a vacuum) with a typical CDMA transmission having a code epoch of 1 millisecond, the worst case scenario is that the user position must be estimated to within 150 kilometres of the actual position.

It will be appreciated that the repetition of the steps of the method according to step (e) is preferably performed using an iterative procedure. The repetition of the steps is performed until the predetermined convergence criterion is met which might include a comparison between user positions calculated as a result of subsequent steps (d) and terminating the procedure when the difference between position solutions is smaller than a predetermined quantity. Alternatively a fixed number of repetitions may be performed which is large enough to always ensure that a sufficiently accurate solution is arrived at. As a further sanity check upon the solution the method may also comprise a test of the user position against other criteria, such that, for example, the final position is within the initial range estimate.

The orbit data may take a known format according to GNSS standards. It typically gives the precise position of the respective satellite at any given time. The time of transmission of the signal may therefore be used to calculate the precise position of the satellite at the time corresponding to the transmission time. The orbit data preferably take the form of position data, such as data according to accepted astronomical or GNSS conventions. The orbit data may therefore take the form of ephemeris data defining the orbit of the satellite. As will be appreciated, in GNSS a "constellation" of satellites is provided to give sufficient geographical coverage for users and to provide a sufficient number of signals (of which four are needed for positioning) to enable a user position to be calculated.

The satellite signal preferably comprises a number of transmitted bits, each of which has an associated Time of Week. The signal is typically subdivided into time partitions. Therefore the transmitted signal preferably comprises time divisions as superframes, each superframe comprising a number of frames and each frame comprising a number of subframes. Typically each subframe is divided into words and each word into bits. Each subframe typically comprises time data as Time Of Week data acting as a time stamp. Since synchronisation is not achieved with each satellite, the ephemeris data for the satellites is preferably obtained from a separate source in a bit true form. This might be an on-board store of the receiver containing pre-loaded data, or it might be an assistance server which provides the data in a format via a network such as the Internet. This data can be communicated to the receiver via a suitable data connection including wirelessly using electromagnetic radiation. A mobile telephone or satellite telephone network may be used for this purpose. Typically such a server is in communication with a receiver which receives the transmitted signals of one or more satellites and these are then forwarded over the Internet for use. In general, each signal comprises a repeated transmission of time data defining a time stamp and orbit data relating to the orbit of the satellite such that the synchronisation step (a) comprises establishing the part of the repeated transmission which is being received at any time by the receiver. Thus the step may comprise establishing the relationship in time between the receiver time and the transmitter time at any particular instance in time.

It will be appreciated that, typically, the method is largely computer-implemented and therefore the invention contemplates the embodiment of the methods discussed herein as a computer program product comprising computer program code for performing some or all of the method steps of the invention when such code is executed upon a computer. The invention also extends to a satellite receiver adapted to perform the method of the invention so as to determine the position of a user.

Brief Description of the Drawings

An example of a method according to the invention will now be described with reference to the accompanying drawings, in which :-

Figure 1 is a flow diagram of the example method; Figure 2 is a schematic representation of the method; and,

Figure 3 shows the relationship between required estimate of the initial user position and the satellite elevation.

Description of Examples We now describe an example implementation of the invention in which a global positioning system (GPS) receiver, such as a handheld or vehicle mounted receiver, acquires synchronisation with a signal from a remote transmitter onboard a first satellite and uses this together with signals from three other satellites to produce a calculated user position.

A GPS signal comprises a series of "frames" (also called "pages"), each of 30 seconds in length. Each frame is divided into five subframes of 6 seconds. Each subframe comprises a preamble followed by the TOW "time stamp" data. The TOW is the GPS time according to the satellite's clock at the time of transmission from the satellite. This is extremely accurate since each satellite has an on-board atomic clock. Thus every transmitted bit of data has an associated TOW and the TOW time stamp itself is transmitted every 6 seconds. There are two primary components of the remainder of the data within each frame, these being ephemeris data (which is high accuracy orbital information of the particular satellite) and almanac data (which is part of a larger sequence of data giving the orbits of each of the satellites in the GPS constellation). The ephemeris data is the same in each frame, but the almanac data is different between frames. It takes 25 consecutive frames to transmit the full almanac data of all satellites in the constellation (this being known as a "superframe").

In order to determine the position of a receiver on Earth, it is essential to know the time at the receiver and the time of transmission of signals from a number of satellites. The difference between the time of transmission and the time of receipt allows the calculation of the distance to the satellite in each case. In known methods calculating this time difference requires a synchronisation to be performed by the receiver with each of at least four satellites. This involves the receiver determining which part of a frame is being decoded at any particular time.

The normal method for establishing synchronisation between a GPS receiver and a satellite transmission, known as a "Time of Week sync" is to decode the data received from the satellite until the Time of Week (TOW) signal is obtained. The TOW is obtained by first waiting for the preamble sequence (8 bits in word 1 of a subframe) and then decoding the complete first and second words (word 1 and word 2). Word 2 contains the TOW count. This means that a normal Time Of Week synch for the worst case (under good signal conditions) might take up to six seconds after data bits have started to be decoded (since a subframe is six seconds long). Of course this time is longer in the event that there is a weak signal. It will be recalled that the known method of establishing a receiver position involves performing this method for each of the four satellites that are required for positioning.

The method of the invention now described in association with the GPS system allows a user position to be calculated more rapidly since synchronisation to each satellite is not required prior to forming the position solution. The invention importantly only requires an initial time of week (TOW) synchronisation to one satellite but does require valid current ephemeris data for all satellites used in the positioning (at least 4). The approximate position of the user is required to within 150 km (as a worst case) although in many cases distances in excess of 150 km can be tolerated.

Referring now to the flow diagram of Figure 1 , a user equipped with a GPS receiver wishing to determine their position initialises the receiver at step 100. This involves turning the receiver on or undergoing a "power up" procedure. The receiver is assumed to have knowledge of the present time which is approximate since an on-board clock of the receiver is typically based upon quartz crystal resonance. The receiver also may include pre-loaded satellite orbit "alamanac" information giving a reasonably accurate estimate of the position of each satellite in the GPS constellation as a function of time.

At step 101 , the user of the receiver enters their approximate location, for example using a grid reference, this being assumed to be preferably within 150 kilometres of their actual position.

The receiver then attempts to acquire any signals that are being transmitted from any of the GPS satellites at step 102. The method requires at least 4 satellite signals to be received although preferably more may be used if available. Any pre-loaded almanac information on-board the receiver may be used to assist in this signals search. The strongest signal is then selected by the receiver, the satellite identified and a synchronisation procedure is performed. This may involve decoding the particular signal of the satellite in question for a number of seconds until the time of week signal ("synch point") has been received and thus the time of week of each received data bit in the signal is known. Equally, any alternative method for achieving synchronisation may be used.

Figure 2 is a schematic representation of the above. The user has a receiver 1 and is located upon the surface of the Earth 10 at the position indicated.

However, the user only knows their location approximately, this being represented by the region shown at 2. Four satellites are indicated at 3,4,5,6, these being in orbit around the Earth. In this example, satellite 4 has the strongest signal and this is the satellite with which synchronisation is achieved, this being indicated at 7. For the remaining satellites, signals are received at the receiver 1 , but synchronisation is not required.

Returning to Figure 1 , once the Time of Week (TOW) sync has been achieved for the signal broadcasted by a first satellite, the time of transmission of that signal can be derived at any point later in time. With the knowledge of both the transmission time and the reception time a pseudorange, p_v , can then be formed according to Equation 1.

Equation 1 :

P, = c(T_u -T,)= r + c(t_u +& + &_D)

In Equation 1 T₁₁ is the receiver (user) time which is the GPS time at the time of receipt of the signal, T₅ is the satellite GPS time (the TOW) at the time of transmission, r is the user to satellite range, c is the speed of light, t_u is a receiver clock bias (essentially any error in the time at the receiver), St is a satellite clock bias (due to any error in the atomic clock on-board the satellite), and St _D represents the combined signal delay due to propagation errors (such as atmospheric effects, multipath and interference). The satellite clock bias St , can be determined from the ephemeris information for the satellite and St _D is of the order of 100 metres or less. Synchronisation allows knowledge of which part of the signal is being received at any moment. However, this does not provide the exact time of receipt at the receiver 1 since the receiver includes a receiver clock bias. Thus the receiver clock bias must be evaluated.

In this method the initial position provided to the receiver 1 by the user as a starting point must have an accuracy as set out in Equation 2.

Equation 2:

£ p_o„s. — < 0.5cAt CE In Equation 2 r is the vector range and At_CE is a code epoch length. This is the length of time taken to transmit a data bit by the satellite. For a CDMA transmission, this is termed a "code epoch".

ε_pos is therefore the difference between the initially provided "guess" position and the actual position of the receiver (which is to be determined by the method). The inaccuracy of the estimated position must not be more than ε_pos since otherwise ambiguity would be introduced which would then require further information to be provided, such as by synchronization with the other satellites. This allows calculation of the receiver clock bias to the nearest code epoch length (1 millisecond). The method therefore relies upon evaluating which exact data bit is being transmitted and this is not possible if the error in the position of the user is larger than half the distance travelled at the speed of light during one code epoch.

In the worst case where the user position error is entirely parallel to the range vector the accuracy must be limited to roughly 150 km. However, the error in height (the user's altitude on Earth) can be assumed to be small in comparison with horizontal position error, in which case the incurred range error becomes largely dependent on elevation angle. For example, a 250km horizontal position error can be tolerated if the elevation angle is above 60 degrees. Figure 3 illustrates the implications of Equation 2 given that the height can be determined to within 5 kilometres. Thus, by synchronising at step 103 with a satellite having a strong signal (and therefore likely to have a large elevation angle) the inaccuracy in the user position estimate can be tolerated to hundreds of kilometres.

Given that the user position estimate provided is within the described bounds the receiver clock bias from Equation 1 can be estimated to an accuracy of the nearest data bit, specifically, better than 0.5 milliseconds in a CDMA transmission with a code epoch length of 1 millisecond. This estimate is accomplished at step 104 by first estimating the geometric range, r, based upon the satellite position for a given TOW (using valid ephemeris data) and the estimated user position. The ephemeris data for the satellite in question may be obtained from a store on-board the receiver 1. However, the ephemeris data is typically only valid for about 2 hours and therefore to ensure an accurate position solution it is preferable to load the ephemeris data to the receiver store shortly prior to using the receiver, including substantially simultaneously. For example, the ephemeris data may be uploaded via a communications link in the form of a mobile telephone or satellite telephone network. This might give access to an assistance server via the Internet. Such assistance servers are known for providing bit true ephemeris data for each satellite in the GPS constellation.

The pseudorange delay can be evaluated using the receiver time (which is the true time at the receiver in addition to the receiver clock bias) and the GPS TOW for the transmission. This information can be used in Equation 1 together with estimates for δt and δt_D to evaluate the receiver clock bias t_u . Note that t_u is typically the largest source of error in Equation 1.

With this first estimate of receiver clock bias t_u , a reasonably accurate estimate of the GPS time at the receiver (this being effectively the TOW at the time of receipt of the data bit in question) is then calculated.

The calculated receiver TOW is then used to produce a first estimate of the positions of any other satellites, specifically the remaining three satellites, 3,5,6, since their ephemeris data is known to the receiver, for example due to assistance server data.

Positioning requires the measurement of pseudoranges to 4 satellites in total and therefore the method requires to firstly estimate the position of the remaining three satellites. As stated earlier the ephemeris data of each of the satellites is assumed to be known.

A first estimate of the position and velocity vectors of each of the remaining three satellites (without synchronisation for each) is then performed at step 105. This is performed using the receiver time, the estimated clock bias and an initial estimated range delay (difference between transmission and receipt time) of 70 milliseconds.

At this point one true pseudorange (with a range delay accuracy of to the nearest millisecond) is available for satellite 4. For the remaining three satellites

3,5,6, the signals are being received but TOW synch has not been established, thus we only have an inaccurate estimate of the TOW for each transmitted bit since the range delay is assumed as 70 milliseconds. All that is known precisely from these signals is the difference in phase between the receiver clock and the received transmission. This is some millisecond fraction (code epoch phase) between the receiver clock and the bits received from each other satellite.

At step 106, the satellite positions are used to calculate a geometric range for each satellite 3,5,6. This is the physical distance between the calculated position of the satellite and the estimated position of the user.

In order to determine the pseudorange for satellites 3,5,6, firstly the code epoch integer ambiguities between the transmit time and the time of a known code epoch starting point (e.g. the time of week (TOW) sync point, or bit edge) must be resolved so as to determine the precise TOW of the transmitted bit received from each satellite. The time of transmission of a particular bit is the time of the TOW synch point plus the number of bits that have occurred since that TOW divided by 1000 (since they are each 1 millisecond in length). Substituting this relationship into Equation 1 , gives the relationship set out in Equation 3.

Equation 3:

n CE = wund(l000 - (τ_u - /r + (_K + St ₊ δt_D )-T_TOW J

The estimated number of code epochs n_CE is thus a float estimate of the elapsed time between the last time of week sync T_τow and the time of transmission rounded to the nearest integer. It will be recalled that T_τow is known from the synchronisation with satellite 4, because this is the last TOW synch time. The receiver clock bias t_u is common to the calculations for each satellite. An initial value for n_CE is determined for each of the satellites 3,5,6 at step 107 using Equation 3.

The integer value for n_CE is then corrected if necessary due to the measured code phase φ . If the measured code phase φ minus the remainder of code epochs is greater than 0.5 code epochs or less than -0.5 code epochs one code epoch must be added or subtracted from n_CE accordingly to determine the correct number of bits (code epochs) since the TOW synchronisation point.

Having determined the initial estimates of n_CE for each satellite, a pseudorange for each satellite can then be calculated. The pseudoranges are formed at step 108 by adding the measured code phase φ to the number of code epochs n_CE to the last synchronisation time T_τow and subtracting the resulting transmission time from the receiver time. This is shown in Equation 4.

Equation 4:

A = c{T_u -T_s) = c{T_u -(τ_{τow +} 0.001(n_{CE +} φ)))

Once the three outstanding pseudoranges have been formed a new three dimensional user position solution can be found using least squares or closed form methods (known in the art). This is performed at step 109.

Steps 104 to 109 are then repeated. Once a revised user position has been established this can then be used to produce an enhanced measure of the receiver clock bias by comparing the pseudorange delay of the synched satellite 4 with the geometric range delay (repeated step 104). The receiver clock bias is used at the repeated step 105 to calculate the positions of the satellites. Now that there exists an updated estimate of user position and a corresponding set of satellite positions (from 105, previous iteration), instead of using a 70ms delay the propagation delay is estimated as the geometric range divided by the speed of light. Values for the geometric ranges are in practice estimated as part of the position solution calculation in previous step 109. As the solution converges during future repeats of steps 104 to 109, the estimated geometric ranges converge to the true geometric range and thus the satellite positions will be corrected as well.

The revised positions allow a revised estimate of the geometric range to be made at repeat step 106, this using the revised positions and the previous user position solution. This in turn leads to revisions of the number of code epochs since a given synch point for each satellite at repeated step 107. A revised pseudorange is then formed at repeated step 108 and then a new position solution is calculated at step 109.

Since the precise range of each satellite r, the receiver clock bias and the number of code epochs that have elapsed are estimates, an iterative procedure is needed in which the estimates are revised repeatedly until a satisfactory solution is achieved. These iterations are indicated at repeat step 110 and steps 104 to 109 are repeated many times.

At each repeat step the user position is tested to see whether convergence of the user position solution has been achieved. This may involve a comparison of the difference between the position solution of the present step with that of one or more previous steps. Alternatively a sufficient number of repetitions of step 110 may be used such that a solution is always achieved by the last repeat with the convergence test in this case simply being a count which is compared against a predetermined number.

Once the solution has been achieved at step 111 , a final test is performed to ensure that the solution is sensible. This may involve a comparison with the original estimated position to ensure that the solution is within the original range error.

By use of the method discussed above, it will be appreciated that a receiver may establish the user position more quickly than by using known techniques since synchronisation with each of the four satellites is not required. Whilst the method has been described with the use of four GPS satellites, it will be appreciated that is may be used with a number of satellites in excess of four and also in association with a non-GPS satellite navigation system.

Claims

1. A method of calculating the position of a receiver with respect to a transmitter on-board each of at least four satellites, comprising:- a) synchronising a receiver time of the receiver with a transmission time of a first transmitter of the plurality of transmitters; b) estimating the time of transmission of signals received from each of the other satellites based upon the receiver time, an estimated user position and orbit data relating to the orbit of each satellite; c) calculating a pseudorange for each of the satellites, wherein for the other satellites the pseudorange is calculated using the estimated time of transmission and the receiver time; d) calculating a revised user position and a revised receiver time using the pseudoranges; e) repeating steps (b) to (d) until the user position meets a predetermined convergence criterion.

2. A method according to claim 1 , wherein step (b) comprises estimating a geometric range of a satellite based upon the user position and a satellite position calculated from an estimate of the transmission time of the received signal and the orbital data.

3. A method according to claim 1 or claim 2, wherein the receiver time comprises an accurate receiver clock time and a receiver clock bias.

4. A method according to claim 3, wherein revisions to the receiver time comprise revisions to the receiver clock bias.

5. A method according to any of the preceding claims, wherein the signal receiver from each satellite comprises data having code epochs and the calculation of each pseudorange comprises estimating a transmission time of the signal in the form of an estimate of the number of code epochs that have elapsed since a reference time which is common to each satellite.

6. A method according to claim 5, wherein the initial estimate of the user position is provided to within a distance from the actual receiver position of one half of a code epoch multiplied by the speed of propagation of the signal.

7. A method according to each of the preceding claims, wherein the estimate of the number of code epochs is based upon the receiver clock time, the geometric range, the receiver clock bias, the reference time, an estimate of the bias of a clock on-board the satellite and an estimate of errors caused by conditions affecting the propagation of the signal between the satellite and receiver.

8. A method according to any of the preceding claims, wherein step (e) is performed iteratively.

9. A method according to any of the preceding claims, wherein the predetermined convergence criterion includes a comparison between user positions calculated as a result of subsequent steps (d).

10. A method according to any of the preceding claims, wherein orbit data is satellite ephemeris data.

11. A method according to any of the preceding claims, further comprising testing the user position.

12. A method according to any of the preceding claims, wherein the signal transmitted by each satellite comprises a number of transmitted bits, each of which has an associated Time of Week.

13. A method according to any of claims 1 to 6, wherein the orbit data are provided to the receiver by an assistance server.

14. A method according to any of the preceding claims, wherein each signal comprises a repeated transmission of time data defining a time stamp and orbit data relating to the orbit of the satellite and wherein the synchronisation step (a) comprises establishing the part of the repeated transmission which is being received at any time by the receiver.

15. A method according to claim 14, wherein the synchronising step comprises establishing the relationship in time between the receiver time and the transmitter time at any particular instance in time.

16. A computer program product comprising computer program code for performing the method of any of the preceding claims when such code is executed upon a computer.

17. A receiver adapted to receive a signal from a plurality of remote transmitters and further adapted to perform a method according to any of claims 1 to 15 when in use.