METHOD FOR POSITIONING A MOVING ELEMENT ON A RADAR IMAGE

The present invention relates to a method for positioning a moving target on a radar image representing the reflectivity levels of the various contributors (reflectors) belonging to an area of the ground illuminated by a radar beam. The invention applies notably to the production of simultaneous, high-resolution SAR-GMTI images from an aircraft equipped with a radar antenna.

In this SAR-GMTI representation, the "image background", made up of the fixed reflectors, is supplied by an SAR (Synthetic Aperture Radar) image and on this image background, the position of each of the moving reflectors present whose detection is sought (hereinafter in the description referred to as target, for convenience) over the area of interest is signaled by a blip, assigned a velocity vector corresponding to the velocity vector of the moving target at the chosen reference instant (generally the central instant of the illumination). The positioning of each moving target is made possible by the GMTI (Ground Moving Target Indicator) mode.

As a reminder, figure 1 illustrates a phase of data acquisition by a moving carrier in "Spot" SAR mode, that is to say a mode in which the antenna beam is permanently locked onto the area to be imaged. A radar fixed to an aircraft 101 illuminates an imaged area 102 for an illumination time Te by locking the antenna beam 103 onto the center 104 of said area 102 throughout the trajectory 105 of the aircraft 101. This time Te is inversely proportional to the resolution aimed for on the transverse axis 106, the resolution on the radial axis 108 being, for its part, inversely proportional to the band emitted by the radar antenna. The imaged area 102 is meshed by a : -grid 110 of cells for each of which the association of at least one reflectivity level is sought.

The radar detections make it possible to create an image on the radial axis 108 and the transverse axis 106, respectively designated hereinafter by the terms "Distance" axis 108 and "Doppler" axis 106. This image, hereinafter referred to as "Distance-Doppler" image, delivers, for each cell M situated inside the imaged area 102, a distance value DM and a Doppler frequency value fM, these two values DM and fwi being referenced relative to a given instant tref corresponding, for example, to the elapsing of half of the overall illumination time.

By describing a given angular segment around the imaged area 102, the radar periodically collects a series of N distance profiles with a recurrence frequency fr equal to N/Te. Each of the N distance profiles offers a one-dimensional representation of the imaged area 102 on the distance axis 108. Furthermore, the distance axis 108 is divided into a number of partitions, each of said partitions preferably having a size slightly less than the distance resolution. For a given distance partition, a spectral analysis on the transverse axis 106 performed on the collected signal makes it possible to make a Doppler discrimination between the different echoes contained within that partition. This spectral analysis makes it possible to discriminate the echoes with the desired resolution if certain conditions are met.

The difficulty in generating an SAR-GMTI image stems from the fact that: - In a conventional SAR mode (in which the fixed reflectors are positioned in azimuth according to a measurement of their Doppler frequency at a referenced instant), the moving echoes appear at a different position from their real position in the imaged scene: for these echoes, the measurement of the Doppler frequency can no longer be directly linked to a position in azimuth. In practice, the expression of the Doppler frequency of the target involves not only its azimuth (or, to be more precise, its angle relative to the velocity vector of the carrier), but also the radial velocity of the target, information that is not a priori known.

Furthermore, the moving target is significantly "out-of-focus": its energy is attenuated over a significant number of Doppler partitions, even Distance partitions if the target is ambiguous Doppler-wise. Consequently, a moving target does not benefit from the high compression gain of the SAR processing (wide emitted band and long integration time), which makes its detection difficult, all the more so if the Doppler (ambiguous) of the target lies within the Doppler band (ambiguous) covered by all the fixed reflectors illuminated by the beam. Such targets are called "endo-clutter" targets. These targets correspond to the slow targets or to the fast targets with a radial velocity that induces a Doppler frequency offset corresponding to the recurrence frequency of the SAR mode (the value of which defines the Doppler ambiguity).

- A conventional GMTI (Ground Moving Target Indicator) mode makes it possible to detect and locate the moving targets but does not supply any "azimuth-resolved" representation of the land area illuminated by the radar beam. This is because the integration times are generally too short for an azimuth discrimination of the fixed reflectors on the ground with a satisfactory resolution.

In order to minimize the spectral spread of the "ground clutter" (signals backscattered by all the fixed reflectors on the ground illuminated by the radar beam) and therefore reduce the number of endo-clutter targets (whose detection is problematical), the GMTI acquisitions favor the so-called "on-axis" acquisition geometries: the velocity vector of the carrier and the observation vector directing the radar beam are colinear. Now, these configurations are configurations specifically prohibited for SAR, since they prevent any azimuth discrimination of the fixed reflectors illuminated by the beam. Consequently, the conventional solution consisting in imaging the area of interest with an SAR mode (pointed away from the target) to then superimpose thereon the detections obtained from a GMTI mode with on-axi§, pointing has a major drawback: it requires the carrier to perform a turn between the SAR acquisition and the GMTI acquisition.

Moreover, the azimuth location of the moving targets (performed by angle error measurement) and the azimuth location of the fixed reflectors (deduced from a Doppler frequency measurement converted into angle relative to the velocity vector of the carrier) are skewed by two biases with different origins:

For the fixed reflectors, the location accuracy is mainly linked to the measurement accuracy on the direction of the velocity vector of the carrier (since the angular location of each fixed reflector is ultimately referenced relative to this direction).

For the moving targets, the bias on the azimuth positioning aggregates namely, on the one hand, a bias on the angle error measurement, and on the other hand, the error on the measurement of the spatial position of the antenna (the angle error measurement supplies an angle within a reference frame linked to the antenna. It is then necessary to know the coordinates of this "antenna" reference frame in space in order to ultimately estimate an azimuth).

Consequently, the ultimate positioning of the moving targets on the SAR image adds together these various measurement biases.

The aim of the invention is to resolve the abovementioned problems.

More specifically, the invention relates to a method for positioning a moving target on a radar image representing the reflectivity echoes of an area illuminated by a radar beam. Advantageously, it comprises at least the following steps:

- a first step of acquisition of two radar images simultaneously, one radar image representing the reflectivity level of the reflectors positioned in the illuminated area as a function of their distance relative to the reception chanels and their Doppler frequency, the first and the second radar images being acquired respectively on a first and a second azimuth reception chanel of the radar. In addition to the reflectivity information associated with each of the reflectors present in the imaged area, these two radar images also provide phase information. For a given reflector and a given reception chanel, this phase is expressed 2TT/AXDA/R (modulo 2TT), where DA/R designates the go and return distance traveled by the emitted-reflected electromagnetic wave received at the phase center of the reception chanel concerned (A designates the wavelength of the electromagnetic wave),

- a second step of calculation, executed on the basis of the first and the second radar images, of a mapping table correlating, for a given distance, a measurement of a phase difference between the two reception chanels with a position on said radar images,

- a third step of supplying a moving target present in the first and second radar images and defined at least by its distance relative to the reception chanels and its Doppler frequency,

- a fourth step of calculation, executed on the basis of the first and the second radar images, for the moving target, of the phase difference between the first and the second radar images and of the distance relative to the reception chanels,

- a fifth step of positioning the moving target on the radar image on the position of the pixel defined, in the mapping table, by the same phase difference and the same reflector distance as said moving target.

Advantageously, the calculation second step comprises the following substeps:

- calculation of a first image representing a phase difference measured between the first and second reception chanels as a function of the distance and Doppler data of the reflectors,

- calculation of a second image representing a theoretical phase difference from the acquisition geometry and from the characteristics of the reception chanels,

- calculation of a third image representing the residual phase difference error between said first image and said second image,

- parametric estimation of the residual phase difference error, by rephrasing of the echoes of the reflectors on this third image.

Advantageously, the calculation fourth step comprises- the following substeps:

- focusing of the moving target on the first and the second radar images,

- extraction of a reduced window from the first and the second radar images centered on the focused moving target,

- measurement of the phase difference for the moving target on the basis of the reduced windows obtained from the first and the second radar images.

Advantageously, in the third step, the moving target is part of a list of moving targets.

The method for positioning moving reflectors according to the invention makes it possible to obtain a radar image of SAR-GMTI type by a single acquisition. This is achieved by no longer requiring the moving platform to execute a change of trajectory in order to perform a first acquisition pointed away from the target followed by a second on-axis acquisition of displacement of the platform.

Furthermore, this method is not affected by the bias of the measurements conventionally applied in the prior art solutions. This method makes it possible to cancel the bias on the relative positioning of the moving reflectors and of the fixed reflectors. In other words, the relative positioning error of the moving targets and of the fixed reflectors in the imaged area is characterized by a zero average.

The invention will be better understood and other advantages will become apparent on reading the following description, given as a nonlimiting example, and through the following figures:

figure 1 represents an illustration of a data acquisition phase by a moving carrier (prior art);

figure 2 is a diagram representing radar image acquisition means for implementing the method according to the invention. In particular, this figure illustrates the two azimuth reception chanels (chanels A and B) as well as the path difference AR that exists between these two chanels if the signal backscattered by a reflector for which the direction is characterized by an angle 0 is considered;

figure 3 is a diagram representing the steps of the method for positioning a target according to the invention.

The invention applies to moving carriers of aircraft or satellite type with a device of radar type onboard for monitoring an area of interest.

The invention makes it possible to obtain a synthetic image showing the moving targets with their direction of displacement. Such an image is obtained from the method according to the invention. On a radar image resulting from the method according to the invention, the moving targets have been previously detected and characterized as a function of their distance relative to the aircraft and of their Doppler frequency. On this Distance-Doppler image of the area of interest, the position of each of the moving targets detected is represented by a colored blip. The color of the blip gives an indication as to the sign of the radial component of the velocity vector of the target (moving away or converging). Such information is directly deduced from the Doppler frequency and from the position of the target. Typically, this type of information can help to determine the traffic lane on which the detected target is located.

Ultimately, such a mode provides an SAR image of the area of interest, on which are superimposed blips associated with the moving targets. These blips give the positions of the moving targets present in the imaged area at a reference instant corresponding (generally) to the central instant of the acquisition.

Typically, the waveform used is that of an SAR mode of metric resolution using the highest possible recurrence frequency, in order to maximize the width of the clear area Doppler-wise and increase the width of the Doppler ambiguity.

The method according to the invention is implemented by means of a radar device coupled to electronic and information processing capabilities for an SAR imaging application (ideally of metric-class resolution) and comprising at least two azimuth reception chanels.

These electronic and information processing capabilities for an SAR imaging application may be installed on the carrier or located on a remote station. In the latter case, the carrier also has means for transmitting measured data to the remote station. In the case of an application for an aircraft as illustrated by figure 1, the two reception chanels can, for example, be two antenna panels positioned on a surface of the carrier facing a direction that is non-colinear with the trajectory of the carrier. This is in order to be able to obtain Doppler (and therefore azimuth) discrimination for the fixed echoes from the ground illuminated by the radar beam and thus be able to construct an SAR image. The principle of the invention is based on the fact thatj if two focused reflectors are situated at the same distance R from the reception chanels (at the central instant of the illumination) and exhibit the same phase difference Acp, their position is identical at the central instant of the illumination, regardless of their nature (fixed or moving). The phase difference of reflectors in the area of interest can be calculated by measuring SAR images on two reception chanels. From the acquisition of two SAR images measured simultaneously on two reception chanels A and B of the radar in a first step, the aim of the method as represented in figure 3 is to calculate, in a next step 306, a mapping table correlating a given distance D with a measurement of a phase difference Acp, in another step 302 executed in parallel or following the step 306, to calculate, for a target, its phase difference Acp, measured between the two reception chanels A and B, and the distance D relative to the reception chanels A and B and, in a final step 308, to position the target on the radar image on the position of the pixel defined, in the mapping table, by the same phase difference and the same reflector distance as the target.

More specifically, the method comprises a first step 300 of acquisition of two radar images simultaneously, the first and the second radar images being acquired respectively on a first and on a second azimuth reception chanel A and B of the radar. The method may comprise a repetition of acquisitions of images taken at different positions in the trajectory of the carrier. From the acquisition of at least a pair of images of an area of interest, the method comprises a number of image processing steps in order to obtain information relating to the reflectors of the images.

Figure 2 schematically represents the reception chanels A and B of the radar device and the two straight lines linking a reflector M (placed at a very great distance relative to the spacing separating the two phase centers of the two reception chanels) to the two phase centers of the two reception chanels. If the signal backscattered by this reflector M is considered, a signal for which the direction of arrival is characterized by an angle 9, the phase difference Acp between the two paths is deduced directly from the path difference AR: A

. This phase difference between the two reception chanels is one of the essential elements exploited by the method of the invention. In practice, a measurement of phase difference A

_ V-V U -> expressed Ay c J , where Vc designates the velocity vector of the target at the central instant of the illumination. Unlike the case of the fixed reflector, knowledge of the Doppler frequency, of the distance R, of the velocity vector y and of the relative carrier-reflector altitude are no longer sufficient to characterize accurately the unitary vector jy pointing toward the target: there is in fact no information available concerning the velocity vector Vc of the target. In other words, it now becomes impossible to reliably link a measurement of Doppler frequency with an azimuth position.

- A focusing law. This law characterizes the migration (in Distance and in Doppler) of the target during the illumination, or at least during a part of the illumination centered on the reference instant (center of the illumination). More generally, the information needed to be able to obtain a "focused" representation of the target in the Distance-Doppler domain (representation stripped of any residual Distance or Doppler migration) is available, over an integration time that is sufficient to guarantee a significant compression gain: after correction of the Distance and Doppler migrations, the energy of the target can thus be concentrated over a limited number of pixels in the Distance-Doppler domain.

More specifically, the step 301 of the method for detecting and characterizing a moving target on a radar image comprises the following substeps:

A first substep of elimination of the standing fixed echoes of strong reflectivity measured on the radar image, in the case where interest is focused on the so-called "endo-clutter" moving targets, that is to say the moving targets for which the Doppler frequency (ambiguous) falls within the Doppler band (ambiguous) covered by the fixed echoes illuminated by the radar beam, a second substep of reduction of the integration time by extraction of the temporally central portion of illumination on the radar image, a third substep of focusing of the echo of the moving target by the application of a correction of the distance migration and of the Doppler migration of the moving target on the radar image, a fourth substep of extraction of the moving target from the radar image.

More specifically, the first substep of elimination of the standing fixed echoes comprises a lowering of the reflectivity of the moving echoes that consists in retaining, for each of the pixels of the radar image, the minimum reflectivity over all of the repetition of radar acquisitions and the detection of the fixed echoes by a thresholding operation.

More specifically, in the third substep, the correction of the distance migration of a target is executed by testing a number of possible ranks of Doppler ambiguity for the target.

More specifically, in the third substep, the correction of the Doppler migration of a target is performed by testing a number of quadratic phase law hypotheses to compensate the residual phase characterizing the residual Doppler migration of the moving target, the quadratic term of this residual phase being defined by the following formula:

\v-v \ - Jd

t2C(f) =—*- C) I 2 > x;2

,cw A 2RC

in which:

->

- v: velocity vector of the carrier,

-c: position of the target,

->

- c: velocity vector of the target,

D

- c: carrier-target distance,

f c

- Jd . Doppler frequency of the target.

More specifically, the third substep comprises a filtering of the out-of-focus points.

More specifically, the fourth substep comprises an operation of thresholding of the pixels of the radar image.

Following the step 301, the method comprises a main step 302 of calculation, for the characterized target or for each characterized target of the list of moving targets, on the one hand of the phase difference measured between the first and the second radar images A and B and, on the other hand, of the distance R of the target or targets relative to the reception chanels.

The step 301 has supplied a list of targets detected, in which each of the targets is characterized by its distance R and its Doppler frequency f at the central instant of the illumination, and a focusing law over a given integration time T, a time centered on the central instant of the complete illumination.

For a given target, the phase difference measurement Acp between the two reception chanels A and B is performed from the two focused Distance-Doppler representations of the target, obtained on the two reception chanels A and B.

For each of the targets, in a substep 3021 and 3023 of the step 302, this information makes it possible to obtain a focused Distance-Doppler representation, that is to say a representation in which the energy of the target is concentrated over a restricted number of pixels centered around the position in the Distance-Doppler plane meshed with a distance resolution $R (radial resolution of the waveform) and a Doppler resolution equal to vT. This focusing on a limited number of pixels in the Distance-Doppler domain makes it possible to isolate the signal associated with the target from the signals obtained from the other reflectors and to benefit from a significant compression gain (inversely proportional to the radial resolution and proportional to the integration time T for a one-off target), in other words an advantageous signal-to-noise ratio.

From these two representations, two analysis windows of reduced size are extracted, for the images originating from the images A and B respectively in a substep 3022 and a substep 3024, around the position (#,/), in order to isolate the pixels on which the energy of the focused target is concentrated. Typically, the analysis window contains 3x3 or 5X5 pixels.

These analysis windows are denoted WindowchanelAv"'"' and WindowchanelB(m'").

The phase difference Acp is estimated, in a substep 3025 of the step 302, by calculating the phase of the following complex number:

~ ( WindowchanelA^") x

Conjugate[WindowchanelBVw'")] ).

A maximum weight is thus given to the pixels where the energy of the target is concentrated. The quality of this measurement of Acp is directly linked to the signal-to-noise ratio that the focused target benefits from.

On completion of this operation, performed on each of the targets detected, in a step 303, a list of moving targets is provided in which the different targets are now characterized by their distance R and their phase difference Acp.

Moreover, from the radar images obtained in the first step 300 on the two reception chanels A and B, the method comprises another main step 306 also comprising a number of substeps 3061, 3062, 3063, 3064. The function of this step 306 is to calculate, from the first and the second radar images supplied respectively by a step 304 and 305, a mapping table correlating, for a given distance, a measurement of a phase difference, calculated between the images obtained from the two reception chanels A and B, with a position on said radar images, that is to say coordinates of pixels on the radar image.

This table takes the form of a Distance-Doppler table called Table Acp(m,n), in which the indices m and n are associated respectively with the Doppler axis and with the Distance axis.

More specifically, it is calculated as follows:

The two Distance-Doppler SAR images associated with the two azimuth reception chanels A and B are respectively denoted lmagechanelA(m,n) and lmagechanelB(m,n).

The first step 3061 of creation of the mapping table consists in calculating the "measured phase difference" image. The expression of this image is:

Image An) is modeled by a Doppler affine and Distance affine function, the missing parameters can, for example,

(A A be estimated by searching for the pair slopef,slopeD which maximizes thev) norm of the following complex number: V

z{slopef,slopeD) = »••» [ \mageAnJ by granting a maximum weight to the strongly reflecting echoes.

Consequently, the next image, denoted Image ^^m'n>, exhibits a maximum phase consistency with the image ImageA = —exp - j slopefxm + slopeDxn

2 y V J)

A f A A ^

with Z-Z slope{,slopeD

V J

On completion of this last step 3064, the mapping table sought TableAcp(m,n) is deduced directly from the following Distance-Doppler image:

Image A(p(m,n) = Image A(ptheoretical(m,n) x ImageAn).

A mapping table is thus ultimately available that has the advantage of being unbiased, without noise and reliable over all of the imaged area.

Finally, the method according to the invention comprises a third main step 308 of positioning the target on the radar image on the position of the pixel defined, in the mapping table, by the same phase difference and the same reflector distance as the target.

At this stage, there is available, in an intermediate step 303, a list of moving targets, each characterized by a pair (Distance R, phase difference Acp) and, in another intermediate step 307, a mapping table, which correlates a pair (Distance R, phase difference Acp) and a position on the Distance-Doppler SAR image of the area imaged (pixel index (m,n)).

The final positioning of a target on the SAR image is therefore done simply by looking in the mapping table for the position of the pixel associated with the pair (Distance R, phase difference A^) characterizing the target.

CLAIMS

1. A method for positioning a moving target on a radar image representing the reflectivity echoes of an area illuminated by a radar beam, said radar being installed on a moving carrier, characterized in that it comprises at least the following steps:

- a first step (300) of acquisition of two radar images simultaneously, one radar image representing the reflectivity level and the phase of the reflectors positioned in the illuminated area as a function of their distance relative to the reception chanels and their Doppler frequency, the first and the second radar images being acquired respectively on a first (A) and a second (B) azimuth reception chanel of the radar,

- a second step (306) of calculation, executed on the basis of the first and the second radar images, of a mapping table correlating, for a given distance, a measurement of a phase difference between the two reception chanels with a position on said radar images,

- a third step (301) of supplying a moving target present in the first and second radar images and defined at least by its distance relative to the reception chanels and its Doppler frequency,

- a fourth step (302) of calculation, executed on the basis of the first and the second radar images, for the moving target, of the phase difference between the first and the second radar images and of the distance relative to the reception chanels,

- a fifth step (308) of positioning the moving target on the radar image on the position of the pixel defined, in the mapping table, by the same phase difference and the same reflector distance as said moving target.

2. The method as claimed in claim 1, characterized in that the calculation second step (306) comprises the following substeps:

- calculation (3061) of a first image representing a phase difference measured between the first and second reception chanels as a function of the distance and Doppler data of the reflectors,

- calculation (3062) of a second image representing a theoretical phase difference from acquisition geometry and from the characteristics of the reception chanels,

- calculation (3063) of a third image representing the residual phase difference error between said first image and said second image,

- parametric estimation of the residual phase difference error, by rephrasing (3064) of the echoes of the reflectors on this third image.

3. The method as claimed in claim 1, characterized in that the calculation fourth step (302) comprises the following substeps:

- focusing (3021; 3023) of the moving target on the first and the second radar images,

-" extraction (3022; 3024) of a reduced window from the first and the second radar images centered on the focused moving target,

- measurement (3025) of the phase difference for the moving target on the basis of the reduced windows obtained from the first and the second radar images.

4. The method as claimed in claim 1, characterized in that, in the third step (301), the moving target is part of a list of moving targets. .. . ' !■■