Patent Grant
8179743
This is a U.S. National Phase Application under 35 U.S.C. §371 of International Application no. PCT/EP2007/063102, filed Nov. 30, 2007, and claims benefit of French Patent Application No. 06 10528, filed Dec. 1, 2006, both of which are incorporated herein. The International Application was published in French on Jun. 3, 2008 as WO 2008/065197 under PCT Article 21 (2).
This invention concerns the field of passive sonar systems simultaneously operating several linear antennas for coast, harbor or port zone surveillance. This invention mainly concerns harbor surveillance systems having several passive linear antennas installed on the seabed and located around the roadstead.
This invention relates to the data processing field. More precisely, the problem handled concerns the elimination of ghost sources that appear when processing angles measured from a linear antenna.
Standard linear antennas, also called “streamers”, generally comprise a set of omni-directional unit acoustic sensors or hydrophones located along the antenna. This structure makes angular measurements ambiguous with such antennas. The direction of the incoming acoustic signal corresponding to a bearing in relation to a given antenna, of the object, which emitted the signal, is not directly accessible as the field of ambiguity is defined around the axis of the antenna by a cone of revolution. The conical angles measured in this manner are still called “pseudo-bearings”. When the source is far, or even at a shallow depth, the incoming elevation of the acoustic waves is close to zero. Then there is, from the angle at which the source is detected, two possible values symmetrical in relation to the reference axis, being the antenna axis (right/left ambiguity) which are converted into two azimuth values θ and θ′ also called Az+ and Az−. The azimuth conventionally represents here the value of the angle defined by the north and the signal incoming direction, which corresponds to the “source-to-antenna” axis. These two azimuth values define two separate directions in space, of which only one corresponds to the effective position of the source.
A known method to eliminate the right/left ambiguity in the case of linear antennas consists of splitting the antenna using two or more parallel linear antennas and by combining the signals received by each antenna to make up a reception channel, for which the diagram is not ambiguous.
Another known method is to equip the streamer with directive acoustic sensors or groups of omni-directional elementary sensors, for example, hydrophone triplets, devices enabling some right/left rejection.
In the case of a surveillance network comprising linear antennas laid on the sea bed, the use of hydrophone triplet linear antennas is not necessarily an appropriate solution to efficiently eliminate the ambiguity, as the phases of the signals received may be disturbed by reflections on the sedimentary layers. Consequently, the right/left rejection obtained may be of a lower quality than in deep water. The same applies when two streamers are used in parallel. In this case, there is either an additional supplementary problem in ensuring the parallelism if the streamers are not precisely side-by-side or the disadvantage of a possible loss of sensitivity if they are.
These two known methods have, in substantially all cases, the disadvantage of increasing the number of hydrophones required to make up a linear antenna. The effect of this increase is to increase the manufacturing cost of the linear antenna implemented and this of the associated receiver, and therefore the global cost of the system required for complete coverage of a zone.
Another method is to operate coherently the two antennas simultaneously, i.e. make up a channel formation encompassing substantially all sensors of the two antennas. In a less costly manner, partial coherent processing may be made by, for example, time correlation of signals from the source and from the two antennas. The disadvantage of these up-line signal-processing techniques is that the sensors must be sampled in a perfectly synchronous manner. This assumes that it is possible to distribute a common clock signal to the two antennas, which is not always possible. Furthermore, this assumes that access to signals transiting in the up-line processing system of the sonar is available, which is not always possible if already existing sonars are used.
In the case of a towed antenna, therefore mobile, it is in addition possible to eliminate the right/left ambiguity by varying the orientation of the antenna. However, to do this the ship must maneuver in a manner so that the operation eliminating the ambiguity takes a relatively long time (typically around 10 minutes). Furthermore, this possibility is not available when working with fixed antennas.
The current methods making it possible to eliminate the ambiguity with fixed antennas, therefore appear as being costly solutions requiring in addition the implementation of more powerful means of computation to handle the increase of information flow following the implementation of a greater number of sensors. The use of more complex antennas associated with more powerful processing means entail a significant increase to the global cost of the surveillance sonar system. Therefore, the preference it to use conventional antennas fitted with omni-directional sensors, which are ambiguous.
One of the consequences from the ambiguity of angle measurement is that when there is a need to locate the source by triangulation, there may be several intersections in the lines of sight. Only one corresponds to the effective source, the others being false solutions, called ghost sources that are to be eliminated even though the antennas constituting the network remain right/left ambiguous.
One purpose of the invention is to propose a solution making it possible to use omni-directional sensor passive linear antennas, naturally ambiguous, to make up an underwater surveillance system capable of eliminating ghost sources appearing during the detection of sources located in the monitored zone or starting to enter this zone.
To this end, the object of the invention is a procedure for eliminating the ambiguity in the localization of a signal source and the rejection of “ghost” sources due to this ambiguity by means of a passive sonar including at least two linear antennas, A and B, separate one from the other. The procedure according to an embodiment of the invention mainly includes:
A second step (step 22) for determination of the speed values corresponding to the different potential sources, and possibly the corresponding frequency f0 of the signal emitted by these sources. This estimation of the parameters characterizing the potential sources is made using the frequency values measured f(ti) and azimuth values θ(ti) measured during the first step (step 21) at different moments ti.
A third step (step 23) includes determining the real source and eliminating the wrong solutions (ghost sources). During this step, the coherence of the values associated with each potential source are assessed in relation to a model represented by a set of measurement equations linking incoming direction and the received signal frequency to the speed, position and possibly to the transmission frequency of the potential source involved.
Then the ambiguity is eliminated by determining the potential source for which the speed, position and frequency values, satisfy these equations for each measurement moment ti. This potential source is considered as the real source. The other potential sources, considered as ghost sources, are rejected.
According to an initial method of implementation of an embodiment of the invention for two antennas A and B, the measurement equations used during step 23, are equations linking the Doppler frequency of the signal received to the frequency f0, to the azimuth θ and to the Cartesian components vx and vy of speed V of the source.
According to an initial variant of this first implementation mode of the procedure, according to an embodiment of the invention, during step 22, the components vx and vy of the potential source speed involved, whether real or ghost, will be determined by triangulation (intersection of the lines of sight from two antennas) at different moments.
According to a second variant of this first implementation mode, the possible values of the frequency f0 and components vx and vy of the speed of the object involved are calculated during step 22, by performing a series of N frequency measurements f(ti) and azimuth measurements θ(ti) received by each of the antennas at consecutive moments ti. These three unknowns are computed by solving an over-determined linear system by the least square method.
To this end, the azimuth measurements θA(ti) and θB(ti) obtained by means of antennas A and B are used, and the linear equation system represented by the 2N relations expressed in a synthetic manner by the following equation is solved:
F=M·X
where X represents the status vector X=[f0; f0vx; f0vy]T and where M represents a measurement matrix with 2N lines, in which each line is expressed as follows (c, celerity of sound assumed known):
MA(ti)=[1; −sin θA(ti)/c; −cos θA(ti)/c]
MB(ti)=[1; −sin θB(ti)/c; −cos θB(ti)/c],
each line MA(ti) or MB(ti) corresponding to a measurement moment ti, and F to a vector with 2N components constituted by the measurement pairs of frequencies received (fA(ti), fB(ti)) measured at N moments ti.
According to these two variants, the measurement equations used during step 23 are expressed as follows:
fA(ti)=f0[1−(vx·sin θA(ti)+vy·cos θA(ti))/c]
fB(ti)=f0[1−(vx·sin θB(ti)+vy·cos θB(ti))/c]
where c represents the celerity of sound, assumed known, and where vx and vy represent the components of speed {right arrow over (V)} of the source in a Cartesian reference system.
According to a third variant of this first implementation mode, the components vx and vy of the potential source speed considered are determined, during step 22, by applying an extended Kalman filter on the status vector X=[x; vx; y; vy; f0] subject to a certain status noise defined beforehand to estimate the possible values of the Cartesian components of the speed of the object emitting the measured signal and of the corresponding frequency of the emitted signal.
According to this variant, the remainder or the innovations issued from the filter are analyzed during step 23. Subsequently, the real source then corresponds to the residuals mostly centered around zero.
According to a second mode of implementation of the invention, for two antennas A and B, the measurement equations used during step 23 are equations linking the Doppler frequency of the signal received to the frequency and to the radial speed {dot over (ρ)} of the source. At reference moment tR (any reference moment) they are expressed as follows:
fA(tR)=f0[1−{dot over (ρ)}A(tR)/c]
fB(tR)=f0[1−{dot over (ρ)}B(tR)/c]
the values {dot over (ρ)}A(tR) and {dot over (ρ)}B(tR) are determined separately from the measurements collected at each antenna by solving the equation system:
sin θA(tR)·{dot over (ρ)}A(tR)−sin θB(tR)·{dot over (ρ)}B(tR)=−cos θA(tR)·ρA(tR)·{dot over (θ)}A(tR)+cos θB(tR)·ρB(tR)·{dot over (θ)}B(tR)
cos θA(tR)·{dot over (ρ)}A(tR)−cos θB(tR)·{dot over (ρ)}B(tR)=sin θA(tR)·ρA(tR)·{dot over (θ)}A(tR)−sin θB(tR)·ρB(tR)·{dot over (θ)}B(tR)
where {dot over (θ)}A(tR) and {dot over (θ)}B(tR) represent the derivative values of the azimuth measurements θA(tR) and θB(tR) at this same reference moment tR.
According to an first variant of this second implementation mode, the derivatives {dot over (θ)}A(tR) and {dot over (θ)}B(tR) are determined by linear Kalman filtering of angle measurement made during step 21 (status vector Kalman filter (θ, {dot over (θ)}), or alpha-beta filtering)
According to another variant of this second implementation mode, {dot over (θ)}A(tR) and {dot over (θ)}B(tR) are determined by a graphic method.
Whatever the method of implementation of the invention, the procedure according to an embodiment of the invention makes it possible to advantageously eliminate the ghost sources by means of a sonar system using conventional passive linear antennas, which present inherently a right/left ambiguity, and this without additional equipment being necessary.
The procedure according to an embodiment of the invention is advantageously adaptable to a large range of sonar systems that may include a variable number of linear antennas distant one from the other.
Advantageously, the procedure according to an embodiment of the invention is implemented using conventional measurements of frequency and bearing. The elimination of ambiguity to the measured bearing by different antennas is obtained by using the results from measurements made on signal samples obtained at different measurement moments.
The characteristics and the advantages of the different aspects of the invention will clearly appear in the following description. The description is associated with the attached figures representing
First of all, consider
This ambiguity, which appears at each antenna considered separately from the others is naturally eliminated when the system involved includes at least three antennas detecting the source. In fact, as illustrated by the figure, the real position of the object, which emitted the detected signal, may be determined simply by triangulation. Only axes 15, 16 and 17, corresponding to the directions in which the object is effectively located in relation to each of the antennas, incorporate an intersection corresponding approximately to the real position of the object, whereas axes 18, 19 and 111, corresponding to the virtual directions, do not incorporate any common intersection. On the other hand, as we see in
Then consider
As can be seen in this figure, the procedure according to an embodiment of the invention includes three main steps:
The first step of the procedure according to an embodiment of the invention, step 21, includes analyzing the signals received by the two linear antennas A and B, by applying a conventional narrow band passive sonar processing operation on these signals. The spectrum of the signal received may be a pure frequency or even a narrow band line spectrum.
Thus, for each antenna, the frequencies (fA and fB) of the signal received as well as bearings GA and GB are determined, in which the detected object is located in relation to each of the antennas. As stated previously, this bearing is determined with a right/left ambiguity. Therefore the two possible bearing values for each antenna are obtained; bearing values, which are translated by two azimuth values θA+ (θB+) and θA− (θB−) for each antenna. As stated previously, the azimuth represents here, in a conventional manner, the value of the angle defined by the north direction and the direction linking the center of symmetry of the antenna to the signal source, as these angles are counted down positively clockwise. Thus, for each antenna two azimuth values θ+ and θ− are obtained, of which only one corresponds to the real direction in which the detected source is located. According to an embodiment of the invention, the frequency and bearing measurements are made at different moments ti.
The second step of the procedure according to an embodiment of the invention, step 22, includes determining the movement speed of the source by means of its Cartesian components vx and vy for example, in a fixed reference.
This determination is by means of azimuth values measured in the previous step 21, knowing that the source is considered as being in an approximately uniform rectilinear motion (MRU) during the measurement time. As said previously, this azimuth measurement is an ambiguous measurement, giving two possible azimuth values for each antenna. Thus, step 22 naturally provides several possible values of the source speed; values of which one truly corresponds to the source and of which the others can be called ghost sources.
As shown in this figure, the implementation of the procedure according to an embodiment of the invention assumes that the surveillance system involved includes at least two linear antennas of which the axes are not co-linear, i.e. not aligned on the same straight line. This restriction represents the general case of the surveillance systems considered, in view of the laying constraints of antennas in a roadstead. Nevertheless, for other cases of use, it is necessary to take this restriction into account when the positioning of the antennas making up the system is defined.
Note that there are numerous cases where the intersection of the lines of sight corresponding to the segments, materializing the incoming direction of the signal on each antenna, do not always exist for the two possible directions of each antenna. The case illustrated by
The third step of the procedure according to an embodiment of the invention, step 23 includes, as previously explained, eliminating the ambiguity resulting from steps 21 and 22. During this step, the real source is determined and the ghost sources eliminated. To achieve this, the speed composition and possibly the frequency values are considered for each potential source and then these values are verified to check whether they satisfy, for each moment the measurement equations linking the incoming directions and the frequency of the signals received to the speed, the position of the source and possibly the frequency of the signal emitted by the potential source involved.
The potential source satisfying these measurement equations is accepted. The other sources are rejected, as being ghost sources.
Depending on the implementation method selected and the reference selected to express the potential source position and speed measurements, the measurement equations are expressed differently. In the rest of the descriptions, several implementation methods are presented as examples.
Now,
As this figure illustrates, in this first implementation method presented as an example, the procedure according to an embodiment of the invention performs, in step 22, the estimation of the parameters associated with the movement of the object at the source of the detected signal, especially its movement speed or, more exactly in this case, the components of this speed defined in an absolute Cartesian reference of which one of the axes, axis y in the example in the figure, is plotted by the north direction.
The incoming directions determined are materialized on the figure by half-straight lines 31 to 38. The intersection of the half-straight lines (or lines of sight) corresponding to a given moment, make it possible to define the potential positions that may correspond to the position of the object at a given moment. Amongst these potential positions, only one corresponds to the real position of the object and the others correspond to virtual objects or ghost sources.
Thus, for example, the intersections of segments 31 and 32 on the one hand and segments 33 and 34 on the other hand make it possible to estimate the two possible positions M2 and M′2 for the source at moment t2, positions identified by their Cartesian coordinates x and y. Then using the estimated positions at moments t1 and t2, it is possible to determine the two corresponding speeds v and v′.
Note here that, because amongst the two positions Mi and M′i estimated at a given moment one is real and the other virtual, the same applies to the estimated speeds {right arrow over (V)} and {right arrow over (V)} and therefore determining which of the speeds is virtual, is the same as determining what is the true incoming direction of the signal.
In a first variant of this first implementation method, the Cartesian components vx and vy of the speed vector are estimated by triangulation, according to a method called indirect or semi-graphic. As illustrated in
Thus in the example of the
Using these positions, the possible corresponding values {right arrow over (V)} and {right arrow over (V)}′ of the speed vector are estimated. Movement of the source is assumed uniform and rectilinear. In practice, the Cartesian components (vx, vy) and (v′x, v′y) of vectors {right arrow over (V)} and {right arrow over (V)}′ are determined by means of the following relation:
In a second variant of this first implementation mode, the Cartesian components vx and vy of the speed vector estimated by means of a method called the direct method, includes defining a transposed vector XT of the status vector X=[f0; f0vx; f0·vy], and a measurement matrix M where each line is defined for a measurement moment ti given by one of the following relations:
MA(ti)=[1; −sin θA(ti)/c; −cos θA(ti)/c] [3]
and
MB(ti)=[1; −sin θB(ti)/c; −cos θB(ti)/c] [4]
where θA(ti) corresponds for the moment t1 considered, to one of the possible incoming directions of the signal on antenna A, whereas θB(ti) corresponds to one of the possible incoming directions of the signal on antenna B; c representing here, as previously stated, the celerity of sound.
In such a manner, a vector F, of dimension 2N, of which the components correspond to N pairs of frequency measurements (fA(ti), fB(ti)) of the signals received by antennas A and B, obtained for N consecutive moments ti, may be defined by the following relation:
F=M·X [5]
where M is the measurement matrix including 2N lines MA(ti) and MB(ti).
The expression of the components of vector F is then:
FA(ti)=MA(ti)·X [6]
and
FB(ti)=MB(ti)·X [7]
Subsequently, the estimation of X will then be made by resolution of this over determined linear system, as per the least square method.
In this first implementation mode of the procedure according to an embodiment of the invention, especially for the two variants described previously, the expression of the equation of measurements considered is:
fA(ti)=f0[1−(vx·sin θA(ti)+vy·cos θA(ti))/c] [8]
fB(ti)=f0[1−(vx·sin θB(ti)+vy·cos θB(ti))/c] [9]
where c represents the celerity of sound, assumed known, and where vx and vy represent the components of the speed {right arrow over (V)} of the source in a Cartesian reference system.
Subsequently, step 23 of the evaluation of the coherence includes, first of all, calculating for each moment ti considered, the value of magnitude νm defined by the relation:
νm(ti)=(fA(ti)−fB(ti))/(fA(ti)+fB(ti)) [10]
where fA(ti) and fB(ti) represent the frequency measurements of the signal received by each of the antennas at measurement moment ti.
According to this method, the pair of estimated values (vx, vy) or (v′x, v′y) of the Cartesian components of the speed of the object is considered and for each antenna the computation is made for each pair of magnitude ν defined by the relation:
where θA(ti) and θB(t) represent the azimuths measured at the moment ti considered.
Thus, for each potential source, a given value of magnitude ν is obtained. Thus, for a system including two antennas A and B and for each measurement moment ti, there are four possible values of the magnitude θ, being values corresponding to the association of the different measured azimuth values:
Subsequently, the evaluation operation of the coherence itself includes then calculating for each value ν, the deviation value ν−νm. The value of ν given the lowest deviation ν−νm then corresponds to the position of the potential source representing the real source. The other potential sources are then considered as ghost sources and are then rejected.
In a third variant of this first implementation mode, step 21 includes, in an even more direct manner, applying an extended Kalman filter “EKF” to the status value X=[x; vx; y; vy; f0] subject to a certain previously defined status noise.
The measurement equations used during steps 23 are then as follows, where the symbol “Atan” represents the Arctangent function:
for the azimuth measurements,
fA(ti)=f0[1−(vx·sin θA(ti)+vy·cos θA(ti))/c] [18]
and
fB(ti)=f0[1−(vx·sin θA(ti)+vy·cos θA(ti))/c] [19]
for the frequency measurements.
The quantities fA, fB, θA and θB represent the values of the signal frequency received and the azimuth measured during step 21, for each of the two antennas A and B.
In this variant, a possible method to assess the coherence of measurements with regard to the model includes, when possible, estimating the transmission frequency f0 of the source and, on the conditions that there are a sufficient number of measurements, in testing the frequency remainders made up by the different deviations between calculated frequencies [fA(ti), fB(ti)] using estimated values of the speed and azimuth values at moments ti and the values [fAm(ti), fBm(ti)] measured on the signal samples received at different moments ti.
Subsequently, the real position of the source is then where the greatest amount of remainders centered around zero is obtained.
Then consider
According to this second implementation mode, measurement step 21 includes, as previously and for each of the antennas A and B, measuring the frequency fA or fB of the signal received (with the offset due to the Doppler), and on the other hand, of measuring the azimuths θA or θB corresponding to the incoming direction of the signal emitted by the source. According to this implementation mode, these measurements are made consecutively at different moments t.
According to this second implementation mode, step 22 includes determining the radial speed {dot over (ρ)} of the source, measured by each antenna, at reference moment tR. According to an embodiment of the invention, this speed {dot over (ρ)}(tR)) is calculated from the corresponding radial distance ρ(tR) and the drift of the azimuth measurement θ(ti) noted {dot over (θ)}(tR). Thus, for two antennas A and B, ρA(tR), ρB(tR), {dot over (θ)}A(tR), {dot over (θ)}(tR) is determined and subsequently {dot over (ρ)}A(tR) and {dot over (ρ)}B(tR).
According to an embodiment of the invention, the radial distances are calculated from azimuth measurements using the following relations:
x(tR)−xA=ρA(tR)·sin θA(tR)
Note that the system [20-21] of four equations with four unknowns (x(tR), y(tR), ρA(tR), ρB(tR) accepts, in a known manner, a unique solution as long as the azimuths θA (tR) and θ(tR) are different.
Also note that here the antennas A and B are assumed fixed and therefore that (xA, yA) and (xB, yB) are independent from time. However, an analogous method may apply to a system including mobile antennas, as long as their speed is known.
Subsequently, from relations [20] and [21], the following system is obtained by eliminating x(tR) and y(tR):
sin θA(tR)·ρA(tR)−sin θB(tR)·ρB(tR)=−xA+xB
cos θA(tR)·ρA(tR)−cos θB(tR)·ρB(tR)=−yA+yB [22]
then by deriving in relation to time:
sin θA(tR)·{dot over (ρ)}A(tR)−sin θB(tR)·{dot over (ρ)}B(tR)=−cos θA(tR)·ρA(tR)·{dot over (θ)}A(tR)+cos θB(tR)·ρB(tR)·{dot over (θ)}B(tR)
and
cos θA(tR)·{dot over (ρ)}A(tR)−cos θB(tR)·{dot over (ρ)}B(tR)=sin θA(tR)·ρA(tR)·{dot over (θ)}A(tR)−sin θB(tR)·ρB(tR)·{dot over (θ)}B(tR) [23]
Thus, a system with two equations with two unknowns {dot over (ρ)}A and {dot over (ρ)}B is obtained, accepting in a known manner a unique solution as long as θA is different from θB.
To solve the system [21] and determine the values of {dot over (ρ)}A and {dot over (ρ)}B it is necessary to know not simply the measurements of azimuths θA and θB, but also the values of the variations {dot over (θ)}A and {dot over (θ)}B of these azimuths over time. {dot over (θ)}A and {dot over (θ)}B may be determined by different known methods, of which two corresponding to the preferred implementation mode are presented below.
According to a preferred first form of this second implementation mode of the procedure according to an embodiment of the invention, {dot over (θ)}A and {dot over (θ)}B, are determined during step 22, by separately applying to the azimuth measurements obtained on antennas A and B a Kalman filter with dynamics order 1 (status vector (θB; {dot over (θ)}B). Thus linearization of the law of variation of these measurements is made.
In addition, an estimation of frequencies related to source movement in relation to antennas A and B is made by applying to frequency measurements fA and fB a Kalman filter with a dynamics order 0 (with as status vector X=(f)). Thus, smoothing is obtained during measured fA and fB frequency time.
Subsequently, the output from these filters obtained is an estimation at current moment of magnitudes (θA, θB, {dot over (θ)}A, {dot over (θ)}B, fA, fB), estimation used during the next step to determine the values {dot over (ρ)}A and {dot over (ρ)}B of the radial speeds.
According to a second preferred form of this implementation mode, the determination of {dot over (θ)}A(tR) and {dot over (θ)}B(tR) is obtained graphically. This form is in particular applicable in the case where means are available making it possible to represent measurement of signals received in an azimuth/time or frequency/time reference system on a screen, so as to have graphic representations such as those illustrated in
In this type of case, a representation of measurements 51, 61 of the received signal is available, made at consecutive moments ti from which it is possible using graphic tools such as mobile cursors 52 and 62, to determine in the best manner by smoothing or “visual filtering” the value of magnitudes (θA, θB, {dot over (θ)}A, {dot over (θ)}B, fA, fB) at a reference moment tR. These magnitudes are deduced from the positions and orientations of cursors in the view planes considered.
The potential values of {dot over (ρ)}A(tR) and of {dot over (ρ)}B(tR) and the corresponding values of fA(tR) and fB(tR) are thus determined, then step 23 includes, in this second implementation mode of the procedure according to an embodiment of the invention, in searching which of these values minimizes the criterion Δ defined by the following relation:
where f0A and f0B represent the estimations of frequency f0 of the signal when emitted (i.e. without Doppler effect), produced using measurements respectively related to the signal received on antenna A and the signal received on antenna B.
In the assumption where the potential values considered of {dot over (ρ)}A(tR)) and {dot over (ρ)}B(tR)) and the values of fA(tR) and fB(tR) correspond to the real location of the sources, the two values of the emitted frequency f0A and f0B obtained are similar and only differ one from the other due to measurement errors. The value of Δ is then minimal.
On the other hand, in the assumption where the potential values considered for {dot over (ρ)}A(tR)) and {dot over (ρ)}B(tR) and fA(tR) and fB(tR) correspond to a ghost source, these two values f0A and f0B are appreciably different so that Δ is not minimal. Thus, the ambiguity affecting measurements made by antennas is clearly eliminated.
For the ease of understanding of the description, the different steps of the procedure according to the invention have been described in the two modes and for substantially all their variants in the specific case where the system is in the presence of a single source in the monitored zone. Naturally, this specific case does not define the limit of the scope of the invention; it may be implemented advantageously to eliminate the ambiguity concerning the real position of several distinct sources moving in the monitored zone.
| Number | Date | Country | Kind |
|---|---|---|---|
| 06 10528 | Dec 2006 | FR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2007/063102 | 11/30/2007 | WO | 00 | 8/20/2009 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2008/065197 | 6/5/2008 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4648080 | Hargreaves | Mar 1987 | A |
| 4906940 | Greene et al. | Mar 1990 | A |
| 5089972 | Nachman et al. | Feb 1992 | A |
| 5528554 | Posch et al. | Jun 1996 | A |
| 6002645 | Booth | Dec 1999 | A |
| 6493636 | DeKok | Dec 2002 | B1 |
| 6681887 | Kragh et al. | Jan 2004 | B1 |
| 7075858 | Doisy et al. | Jul 2006 | B2 |
| 20040071046 | Kervern et al. | Apr 2004 | A1 |
| 20050099887 | Zimmerman et al. | May 2005 | A1 |
| 20060109743 | Kosalos et al. | May 2006 | A1 |
| 20060241914 | Barnard et al. | Oct 2006 | A1 |
| 20080106466 | Kelley et al. | May 2008 | A1 |
| Number | Date | Country |
|---|---|---|
| 079506 | May 1983 | EP |
| 0416992 | Mar 1991 | EP |
| 1099958 | May 2001 | EP |
| 2727765 | Jun 1996 | FR |
| 2366617 | Mar 2002 | GB |
| WO-03019224 | Mar 2003 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20100080080 A1 | Apr 2010 | US |
