LOCATING A TRANSMITTER BY MEANS OF A PLURALITY OF GEOGRAPHICALLY REMOTE RECEIVING STATIONS USING KNOWN OBJECT PATHS

Abstract

The invention relates to a method for locating a transmitter, which is implemented in a processing unit of a processing station of a locating system.

Claims

1. A method for locating a transmitter, implemented in a processing unit of a processing station of a locating system comprising the following steps: receiving signals acquired by geographically remote receiving stations, said signals being dated by local time bases of each receiving station and corresponding to signals from a transmitter to be located and from at least one known object; determining, on the basis of the dated signals, the measured TDOAs relating to the transmitter to be located and to the object; and/or measured FDOAs relating to the transmitter to be located and to the known object; determining, on the basis of known ephemerides relating to the known object and of geographical positions of the stations; theoretical TDOAs relating to the known object; and/or theoretical FDOAs; determining, by taking the difference of the measured and theoretical TDOAs and/or FDOAs relating to the known object, a residual error affecting the measured TDOAs and/or FDOAs in such a way that each receiving station corrects its local clock or the processing station corrects the timestamps of the portions of the signals from each receiving station; determining the location of the transmitter using measured TDOAs and/or FDOAs once the receiving stations have corrected their local time base or the processing station has corrected the timestamps of the portions of the signals from each receiving station.

2. The method as claimed in claim 1, comprising a step of determining, on the basis of the measured TDOAs and/or FDOAs relating to several objects, an indicator of reliability of said measured time-domain or frequency-domain separations, said reliability indicator making it possible to determine whether or not the ephemerides of the known object can be used for the correction of the local clock and for, in particular, determining whether or not the known object is in the process of maneuvering.

3. The method as claimed in claim 1, comprising a step of determining, on the basis of the residuals affecting the measured TDOAs and/or FDOAs, a number of data of correction of the local time bases and a step of transmitting to each receiving station data of correction of the local time bases determined in such a way that each receiving station corrects its local clock.

4. The method as claimed in claim 1, comprising a step of keeping data of correction of the local time bases determined in such a way that the processing station corrects the timestamps of the portions of the signals received from each receiving station.

5. The method as claimed in claim 1, comprising a step of determining the location of the known object or known objects on the basis of measurements taken by the.

6. The method as claimed in claim 2, comprising a step of keeping data of correction of the local time bases determined in such a way that the processing station corrects the timestamps of the portions of the signals received from each receiving station, wherein the location is used to determine the indicator of reliability.

7. The method as claimed in claim 2, wherein said determining consists in comparing several measured TDOAs and/or FDOAs for several objects which are known but different to check the consistency of the residuals with one another and determine all the objects that can be used for the correction of the local clock.

8. The method as claimed in claim 1, wherein the receiving stations are mutually synchronized by means of a signal coming from a satellite positioning system, preferably intermittently.

9. A system for locating a transmitter to be located, said locating system comprising a processing station comprising a processing unit configured to implement a method as claimed in claim 1, and at least two receiving stations each comprising: a local clock configured to provide a local time base; a first receiver configured to acquire signals from an object to be located, said first receiver further configured to acquire signals from a known object, said acquired signals being dated by the local time base

10. The system as claimed in claim 9, wherein each receiving station further comprises a second receiver configured to acquire signals from a satellite positioning system, said second receiver being further configured to demodulate the signals acquired by the second receiver to extract therefrom an absolute time base in order to correct the local time base of each receiving station.

11. The system as claimed in claim 9, wherein the processing station is constituted by one of the receiving stations.

12. A computer program product comprising code instructions for implementing a locating method as claimed in claim 1, when the latter is executed by a computer.

Description

OVERVIEW OF THE FIGURES

[0053] Other features, aims and advantages of the invention will become apparent from the following description, which is purely illustrative and non-limiting, and which must be read with reference to the appended drawings wherein:

[0054] FIG. 1 illustrates a locating system according to the invention

[0055] FIG. 2 illustrates a receiving station of a locating system according to the invention,

[0056] FIG. 3 illustrates a locating method according to the invention.

[0057] In all the figures similar elements bear identical reference numbers.

DETAILED DESCRIPTION

[0058] In relation to FIG. 1, a system for locating 1 a transmitter 3 of a radio frequency signal comprises at least two receiving stations 2a, 2b, 2c.

[0059] The receiving stations 2a, 2b, 2c are geographically remote, separate and distant and are in a link with a processing station 6 which is used to process the signals from these receiving stations 2a, 2b, 2c to determine time offsets and frequency offsets of one and the same signal received by each of these stations and, on the basis of these offsets, deduce therefrom a trajectory of the transmitter 3 and therefore its location. These time and frequency offsets are the TDOA and FDOA explained in the introduction. The processing station 6 is, in FIG. 1, separate from the receiving stations but in a particular embodiment, one of the receiving stations can also be the processing station.

[0060] The transmitter 3 to be located is for example a satellite but it can be any object as long as it transmits a radio frequency signal which can be received by at least two receiving stations.

[0061] In FIG. 1, three stations are shown, two stations may be enough but the greater the number of stations the better the accuracy of the computations.

[0062] In relation to FIG. 2, each receiving station 2x (x=a or b or c) comprises a first receiver 21x configured to acquire signals from the transmitter 3 to be located and a second receiver 23x configured to acquire signals from one or more satellites of one or more GNSS constellations 5.

[0063] Furthermore, the first receiver 21x is configured to acquire signals from the known object 4. Specifically, since the processing of the known object serves to evaluate the characteristic parameters of the imperfections inherent to the receiving stations and degrading the accuracy of location of the transmitter 3, it is consequently essential that the process of receiving the signals coming from the known object 4 undergoes the same degradations and does not go via a dedicated receiving line.

[0064] The signals from natural or artificial celestial objects make it possible to palliate the absence of the GNSS signal as will be seen further on.

[0065] These celestial or artificial objects are for example stars or else geostationary satellites. In the following description the expression “known object” will be used to denote these objects. These known objects have the advantage that their ephemerides are well-known and that their trajectory can therefore be computed reliably and the TDOAs and/or FDOAs can be predicted. Specifically, since the TDOA and FDOA are predictable it is possible, by comparing values computed on the basis of the measurements and values resulting from the predictions, to evaluate any errors in the computations of the TDOA and/or FDOA of the transmitter to be located.

[0066] Furthermore, a signal emitted by a radiating celestial body is noise related to its equivalent temperature which is therefore detectable if its temperature is relatively high in relation to the cosmic radiation at 3 Kelvins (Sun, moon, quasar, etc.). Since the terrestrial stations are relatively close by comparison with their distance from these celestial bodies, they see this celestial body from a virtually identical angle and therefore receive the same thermal noise, but at times that are slightly shifted due to the difference in mutual separation of the stations.

[0067] Hence, the correlation of the two signals will be at a maximum when these two signals have been realigned: the autocorrelation function of wideband white noise is indeed a “Dirac” pulse at time 0. Thus, it will be understood that a noise has correlation properties (the Fourier transform of the autocorrelation function of a noise gives its spectrum by definition).

[0068] Returning to FIG. 2, each receiving station 2a, 2b, 2c further comprises a local clock h2x configured to provide a local time base tlocal2x. Furthermore, the receivers 21x. 23x are clocked to this clock. The term “local clock” is understood to mean an oscillator providing a stable frequency signal which makes it possible, on its rising or falling edges, to trigger and clock the sampling of the acquisitions by each receiver. Moreover, the counting of the clock edges provides a common timestamping of the samples of each acquisition.

[0069] When the GNSS signal is available, the local time base is synchronized on an absolute time base resulting from the demodulation of the GNSS signal. In this regard, the second receiver 23x is configured to acquire signals from a satellite positioning system 5, and is further configured to demodulate the acquired signals to extract therefrom an absolute time base in order to correct each local time base of each receiving station.

[0070] Thus, the receiving stations 2x are mutually synchronized by means of the signal from the satellite positioning system using this absolute time base. Note that when the GNSS signal is not available, the local time base which is no longer slaved to the absolute time will drift weakly but independently for all the receiving stations such that the separation in synchronization increases with time.

[0071] Each receiving station 2a, 2b, 2c also comprises a receiving antenna Aix, connected to each receiver 21x, 23x. Furthermore, the receiving stations each comprise a communication interface (not shown) to communicate with the processing station 6.

[0072] As regards the acquisition, each receiver is composed of a conventional radio frequency receiving unit. This receiving line includes a frequency converter slaved to the frequency reference, a multi-channel digitization line deriving from an analog-to-digital converter slaved to the frequency reference. This receiving line well-known to those skilled in the art will not be described in further detail here.

[0073] This concerns the case where the GNSS signal is not available such that the local time base is no longer reliable and provides an erroneous date, which slowly drifts as soon as the GNSS signal becomes unavailable.

[0074] In this particular situation, a method for locating a transmitter 3 is described hereinafter in relation to FIG. 3. Such a method is implemented in a processing unit 7 of the processing station 6.

[0075] At least two receiving stations 2a, 2b, 2c proceed to the acquisition (step E1) and the timestamping (step E2) of the portions of signals from the transmitter 3 to be located and of at least one known object 4. In particular, one obtains for the transmitter a signal Semetteur_x and for the object a signal Sobjet_x. These signals are dated using the local time base of each receiver station 2a, 2b, 2c.

[0076] These signals are transmitted (step E3) to the processing station 6 which will, after receiving these signals (step REC) for example correlate pairwise the signals from several stations in order to be able to compare identical portions of signals to deduce therefrom the TDOAij and/or FDOAij i.e. the time-domain and frequency-domain separations of identical signal portions (the indices i and j denote the stations a, b, c) determined for two stations i,j.

[0077] One of the objectives expected by the use of an object with a known ephemeris is to be able to correct the local time bases of the receiving stations, as soon as the TDOA of the known object or else the FDOAs of the known object are used.

[0078] As regards the TDOAij it is exactly the difference in propagation time taken by identical portions of the signal of the transmitter 3 to reach the station i and to reach the station j. Of course these time separations are measured in relation to the local time bases which are inaccurate given the absence of the GNSS signal.

[0079] Thus, on the basis of the timestamps of the received signals from at east two receiving stations 2a, 2b, 2c the processing station determines (step DET1) measured time separations TDOA_objet_.sub.ij.sup.MES. TDOA_emetteur.sub.ij.sup.MES corresponding to the received signals relating to the transmitter to be located and to the known object (or objects). Of course, similar processing is possible on the basis of the FDOAs.

[0080] Next, on the basis of known ephemerides and determined in an absolute time base in relation to at least one known object, theoretical time and/or frequency separations TDOA_objet_.sub.ij.sup.TH, FDO_Aobjet_.sub.ij.sup.TH relating to the known object are determined (DET2).

[0081] By taking the difference of the separations measured and the theoretical values, one determines (step DET3) a time-domain error RES_TDOA.sub.ij (or TDOA residual) affecting the TDOAs and which makes it possible to correct (step E4) the time bases of the receiving stations. In a similar manner a frequency error (or FDOA residual) could be computed on the bases of the measured FDOA values and the theoretical values.

[0082] According to an embodiment, on the basis of the residual errors affecting the measured TDOAs and/or FDOAs data of correction of the local time bases are determined (step DET5).

[0083] Then, these correction data are transmitted to each receiving station (step TRANS) which resets the date of its local clock.

[0084] According to an embodiment, the processing station keeps the residual and makes the timestamping corrections of the signal portions received from each station (step CONS).

[0085] Finally the locating (step LOC1) of the transmitter 3 using measured TDOAs and/or FDOAs once the receiving stations have corrected their local time base is carried out.

[0086] As regards the TDOA, this gives the following expression

TDOA_objet_.sub.ij.sup.MES=TDOA.sub.ij.sup.reel+CorrNoise+Δ.sub.ErrGNSS.sup.ij+Δ.sub.ErrTshort.sup.ij+Δ.sub.BiasCal

With

[0087] TDOA.sub.ij.sup.Reel the actual physical value, that one is seeking to measure; [0088] CorrNoise; the correlation noise, typically AWGN (Additive White Gaussian Noise) (which is white, Gaussian, of zero mean and predictable energy and determined by the channel); [0089] Δ.sub.BiasCal: the error incurred by the offsets of the physical device of the station and which can be calibrated (time of propagation through the equipment, uncertainty on the actual geographical position of the receivers), They are considered very stable on the scale of several weeks and are therefore considered as known since they are estimated by a calibration process; [0090] Δ.sub.ErrGNSS.sup.ij: the difference in timestamping error obtained by the use of the GNSS signal (typically low, tending to be of AWGN type). [0091] Δ.sub.ErrTshort.sup.ij: the difference in short-term timestamping error (which is not compensated for by the GPS correction process, thus the short-term clock jitter).

[0092] When the known object is tracked by the receiver stations (nominal rating) its position and therefore the actual TDOA.sub.ij.sup.Reel values are known to the nearest error ERR_PROPAG_TDOA.sub.ij of the propagator which allows it to compute the theoretical value TDOA.sub.ij.sup.TH. By eliminating the terms that are assumed to be known, one therefore defines the TDOA residual by: RES_TDOA.sub.ij=TDOA.sub.ij.sup.Mes−TDOA.sub.ij.sup.TH=ERR_PROPAG_TDOA.sub.ij+CorrNoise+Δ.sub.ErrGNSS.sup.ij+Δ.sub.ErrTshort.sup.ij. (Here, it has been considered that Δ.sub.BiasCal is known and has been removed).

[0093] Most of these terms are of negligible intensity by comparison with the drift that one is seeking to estimate and fall in the category of noise that can be approached by a low zero-mean noise. Once the stations are in nominal mode (start-up phase finished, time base slaved for the first time to the GNSS, then continuous slaving to the GNSS, etc.) the drift of REF_FREQ.sub.i (and therefore the term Δ.sub.ErrTshort.sup.ij) and the associated time base depends only on the characteristics of REF_FREQ.sub.i in non-slaved mode. Its characteristics are chosen to be of very good quality. In nominal mode (namely when the time-domain synchronization making use of the GNSS constellations is operational), the tracking of a known object therefore makes it possible to estimate the short-term timestamping error (incurred by the short-term jitter in the local clocks) of the system which can be used to also correct the local clocks of the receiving stations. In one implementation of the invention, the processing loops using the known objects are active and used even when the GNSS synchronization is active and operational. These loops are then used solely in order to correct the short-term jitter in the local clocks.

[0094] In the event of the GNSS synchronization no longer being possible, the invention compensates both for mid-term and short-term drifts in the local clocks.

[0095] As described, the obtainment of the time-domain error is based on the tracking of a known object. The reliability of the measurements concerning it is therefore critical.

[0096] Specifically, when these are geostationary satellites, these latter can be in maneuvering phases such that their trajectories are not predictable on the basis if the ephemerides. The method of location of the transmitter to be located supposes the prior (and where applicable simultaneous) location of the reference artificial objects (step LOC2).

[0097] Hence, the locating method comprises a step of determining (step DET4) on the basis of the measured TDOAs and/or FDOAs relating to the known object, an indicator of reliability of the measured time separations, said reliability indicator having the aim of determining whether or not the ephemerides of the known object can be used for the local clock correction. This reliability indicator in particular makes it possible to determine whether or not the known object is in the process of maneuvering.

[0098] Of course the known object, when it is a celestial object (for example the sun) is not concerned by these concepts of reliability. Specifically, these known objects are classified and easily identifiable and extremely accurate ephemerides are available.

[0099] However, the classified natural objects may not be in permanent visibility (for example if one uses the sun as the known object, its visibility is of course subject to day/night alternation) for the high-reliability resetting measurements (not dependent on the propagation error) and the use of the artificial objects (for example geostationary) in constant visibility can systematically be used in relative resetting (subject to propagation error) in the phases of invisibility of the natural objects.

[0100] Furthermore, and advantageously, the reliability of the measurements resulting from the known objects consists in comparing several TDOA residuals obtained for several objects which are known but different to check the alignment of these time-domain residuals with one another since they are not meant to depend on the known object. If some of these objects diverge too far from the others then it can be deduced therefrom that their ephemeris is not reliable and the object can then be removed from the list of reference objects that can be used for maintaining the synchronization between the stations.