DE102011016337A1 - Method for imaging land vehicle from aircraft, involves performing motion estimation and image formation based on signal components, where image formation is implemented based on observation period such that imaging of object is enabled - Google Patents

Abstract

The method involves emitting pulses with a radar or sonar sensor (3) toward a moving object i.e. land vehicle (2), and detecting echo signals. A radar/sonar image of the object is produced on basis of motion estimation of the object. Disturbing signal components are suppressed by processing of spatiotemporal correlated echo signals. Motion estimation and image formation are implemented on basis of signal components that remain after suppression of the disturbing components. The image formation is implemented based on a coherent observation period such that imaging of the object is enabled. An independent claim is also included for a device for imaging a moving object by a moving radar or sonar sensor.

Description

Technical application

The present invention relates to a method for imaging moving objects by means of a moving radar or sonar sensor, wherein the radar or sonar sensor emitted pulses in the direction of a moving object and radar or Sonarechosignale are detected, and from the echo signals based on a motion estimation of Object a radar or sonar image of the object is generated. The invention also relates to a device which is designed to carry out the method. The method and associated apparatus can be used in areas where radar or sonar imaging of moving objects, e.g. As of land vehicles, aircraft or ships, is required.

State of the art

In the field of radar imaging, it is known that images of stationary scenes with very fine resolution can be generated with a coherently operating moving radar system with high bandwidth. This is possible because on the one hand the high bandwidth allows a fine resolution in the range direction and on the other hand, the coherent recording of numerous echoes for forming a synthetic aperture in the signal processing can be used, which in turn leads to a fine resolution across the direction of removal. This principle of synthetic aperture radar (SAR) is already being used in many stationary scene imaging devices.

The mapping of moving objects with SAR requires extensions that were first introduced for the inverse constellation - moving radar forms moving object - and therefore are described under the term ISAR method (inverse SAR). The signal components to be processed for the image generation are in this case all non-stationary echo signal contributions which can be easily separated. While in imaging a stationary scene only the known or measurable movement of the radar sensor is to be considered, the ISAR constellations are characterized by the fact that the movement of the object to be imaged itself has to be considered for the signal processing. In many applications, object motion is unknown and must be estimated from the data itself. For some moving objects, methods for estimating object motion are known. To show eg. M. Stuff et al., "Imaging Moving Objects in 3D from Single Aperture Synthetic Aperture Radar", Radar Conference, 2004, Proceedings of the IEEE, 26.-29. April 2004, pages 94 to 98 , or D. Pastina, "Radiation Conference 2008, RADAR '08, IEEE, 26.-30. May 2008, pages 1 to 6 , various methods that can be used for the motion estimation of land vehicles, ships and aircraft. In addition to the previously described constellation (imaging of a moving object by means of stationary radar), the recording of moving objects with a moving radar sensor is nowadays also treated under the term ISAR method.

The radar technique is used not only to depict scenes or objects but also in other fields. For example, in the field of moving target indication (MTI), it is well known that a coherent, moving, multi-channel radar system is well-suited for the detection of moving objects. In this case, the channels can be formed by the assignment to real receiving antennas or antenna subgroups or as virtual channels by antenna switching strategies. Using space-time-adaptive processing (STAP), stationary ground scene echoes (ground clutter) can be detected in the case of land vehicle detection (GMTI: ground MTI) or flying objects (AMTI: airborne MTI). or the slow-moving lake (sea clutter) in the case of the discovery of maritime objects (MMTI: maritime MTI) are suppressed.

Here, two main classes of MTI methods can be distinguished, depending on the recording mode. In the first class of procedure ("Scan / MTI Class"), the direction of view of the antennas is changed at regular time intervals, so that a large area on the ground is illuminated and monitored. The multi-channel data is divided into small data sets (CPI: coherent processing interval) for coherent processing. Usually, the CPI, hereinafter also referred to as the coherent observation period, is chosen such that a moving target remains in a distance Doppler cell during this time. Clutter suppression and moving-target detection are performed on the distance-compressed raw data in either the distance-time or distance Doppler range. Benefits of this type of processing are efficient clutter suppression and good target location if the number of receive channels is sufficient. However, the achieved signal-to-noise ratio of the moving targets is smaller than in the following method class, since only a part of the synthetic aperture is used coherently.

The second class ("SAR / MTI class") requires a multi-channel SAR system. In contrast to the above first method class in this case broadband signals used to achieve a fine spatial resolution. Several SAR images are processed and clutter suppression and motion target detection are performed in the image area. However, the clutter suppression and the target localization are only suboptimal.

Out D. Cerutti-Maori et al., "Optimum GMTI Processing for Space-based SAR / GMTI Systems - Theoretical Derivation", EUSAR 2010, Aachen, Germany, June 2010, pages 390-393 , a method for the detection of moving objects is known that combines the advantages of image processing ("SAR / MTI class") with the advantages of Doppler clutter suppression ("Scan / MTI class"). The method is based on adaptive processing of multichannel SAR data, e.g. B. were recorded in the strip, sliding or headlight mode. This process offers many advantages. By maximizing the signal-to-clutter-plus-noise ratio (SCNR) of the moving targets using azimuth compression and efficient clutter suppression, a higher target detection probability and an improved estimation of the target parameters can be achieved. Moving targets with lower radar backscatter cross-section, ie targets with less backscattered power, can be detected. Furthermore, thanks to the fine spatial resolution and the target detection in the image area, this method improves the separability of targets which move close together at the same speed. Finally, the data can also be used to image the stationary scene in addition to the movement target detection.

In the field of radar imaging, in R. Klemm, "Interrelated Problems in Space-Time Processing for SAR and ISAR", Radar, Sonar and Navigation, IEEE Proceedings, Vol. 5, Oct. 1998, pages 297-302 , Presented problems that might occur in connection with clutter suppression in ISAR. Only cases in which the object moves at a constant high speed and processing with a small coherent observation time (CPI) are considered. The CPI is chosen so that the variation of the Doppler frequency of a clutter scatterer during this time is less than the Doppler resolution. This corresponds to the usual CPI of the first method class in the MTI methods described above and does not allow imaging. In addition, slowly moving objects that are in the clutter band are not considered in this publication.

Basically, radar imaging of moving objects by means of a moving radar sensor results in problems due to interfering signal components, for example due to clutter or jamming, which lead to substantial limitations of the imaging. If the movement of the object to be imaged is unknown, it must be determined from the recorded radar data as part of the signal processing. The known methods fail when the echoes of the object to be imaged in the Doppler area are superimposed by echoes of the rest of the scene. An illustration is then not possible. This means that objects with an unfavorable, unknown motion profile can not be imaged using the known means. In cases where the object movement is known, an image of the object can be calculated, but this is greatly affected by the overlapping clutter and possibly unusable. This case does not only occur when the relative velocity between the radar and the object to be imaged is in the order of the relative velocity between the radar sensor and the clutter-causing scene portions, but also objects with greater relative speed are affected because of the ambiguity of the sampled Doppler range.

The object of the present invention is to provide a method and a device for radar or sonar imaging, which allow the imaging of a moving object even if its echoes in the Doppler region wholly or partially surrounded by echoes of stationary or slowly moving surrounding Scene are superimposed.

Presentation of the invention

The object is achieved with the method and the device according to claims 1 and 16. Advantageous embodiments of the method are the subject of the dependent claims or can be found in the following description and the embodiment.

In the proposed method for imaging moving objects by means of a moving radar or sonar sensor, pulses are emitted in the known manner with the radar or sonar sensor in the direction of the moving object and radar or sonar echo signals are detected, from which a radar or sonar echo signal is acquired. or sonar image of the object is generated. The method is characterized by the fact that disturbing signal components, which do not originate from the object, are suppressed by a processing of spatio-temporally correlated echo signals. The motion estimation and imaging is then performed on the basis of remaining after the suppression of the interfering signal components signal components. The coherent observation time (CPI) is chosen so large in the processing for image generation that the variation of the aspect angle within the coherent observation period under which the object with the radar or sonar pulses is illuminated, a picture of the object even allows.

The spatiotemporal correlation of the echo signals can be obtained by recording the echoes with a multi-channel receiver, i. H. by recording with multiple antennas or antenna subsets can be achieved. In a further embodiment, multiple virtual channels are formed by appropriate antenna switching strategies, i. H. by switching antennas or antenna groups when transmitting the pulses or receiving the echo signals. For this purpose, only a single-channel receiver is required. These techniques are familiar to those skilled in the space-time-based processing (STAP) art.

The invention is explained below with reference to the radar image of moving objects, but can be transferred in the same way to an image by means of sonar. In the method, techniques from two different fields of radar signal processing, the motion target detection and the radar imaging, combined to achieve in the radar imaging suppression of spurious signal components, in particular clutter or jammer. First, with the methods from the field of motion target detection, based on multi-channel radar data, for example, ground or sea clutter suppressed, which usually interferes with the calculation of a moving target image. In this case, the method known from the above-mentioned publication by D. Cerutti-Maori is preferably used without azimuth focusing. Subsequently, the movement of a moving object is estimated using the methods from the areas of target area detection or radar imaging. Finally, an ISAR map is calculated based on the resulting data.

The method and the associated radar system allow the imaging of moving objects, eg. As of land vehicles, aircraft or ships, using radar according to the ISAR method, due to their direction of movement and speed in the Doppler range of radar echoes of stationary or slow-moving scene (clutter), signals from jammers or a combination of Clutter and jammer are superimposed. In this way, in comparison to known methods, the ISAR imaging of moving objects is improved and the imaging of additional moving objects which have a movement profile unfavorable to the known processing is made possible. The present description explains the proposed method for suppressing clutter signals. Disturbing signal contributions from whining, however, can be treated in a similar way.

The recording and evaluation of radar echoes according to the proposed method prove to be advantageous in two respects. On the one hand, they make it possible to significantly reduce the impairments of signal processing by ground or sea clutter, by avoiding the problems that the clutter causes during the estimation of the movement of the object to be imaged by the clutter suppression. In particular, the estimation of the movement of the object to be imaged is performed with much better accuracy and in many cases enabled by the clutter suppression. On the other hand, in the cases where the image of the object overlaps the clutter without the clutter suppression, the final image generation leads to significantly better results, since the image of the object no longer cluttered with the clutches after application of the processing proposed in the method -Echos competes. The ratio between signal power and clutter-plus-noise power (signal-to-interference ratio) in the image is thus significantly improved. Thus, the classification of the imaged objects can be greatly improved.

The proposed radar or sonar system for imaging moving objects by means of a moving radar or sonar sensor comprises one or more antennas for emitting pulses in the direction of a moving object and for receiving echo signals as well as a control and evaluation device, which the antennas according to the proposed method for emitting the pulses and for receiving the echo signals drives and performs the image generation according to this method. The radar system can thus have both a multi-channel receiver and a single-channel receiver in conjunction with virtual channels generated by the control and evaluation device. The images of the moving objects generated by the radar or sonar system are displayed in a suitable manner on a screen.

Brief description of the drawings

The proposed method will be explained in more detail below in conjunction with the drawings and with reference to an embodiment. Hereby show:

1 an example of a radar system for acquisition of multi-channel ISAR data in a schematic representation; such as

2 an example of the processing stages of the proposed method.

Ways to carry out the invention

The proposed method is exemplified by a radar sensor 3 explained on a moving platform, an airplane 1 , and a radar image of a vehicle moving on the ground 2 should generate. 1 shows the aircraft 1 with the radar sensor 3 as well as the moving vehicle located on the ground 2 , The radar sensor 3 sends broadband pulses (radar lobe 4 ) in the direction of the vehicle 2 , and receives the echo signals with multiple antennas or antenna arrays. The signals of the different antennas are assigned to different channels. If only one channel or only one antenna is available, switching strategies can also be used to generate virtual channels, with the possibility of switching over partial apertures or the antenna feed. Taking into account the spatio-temporal correlation of the received echo signals of the different channels, the signal components of the stationary ground scene (ground clutter) in imaging of moving land vehicles, as in the present example, or low-flying objects or the signal components of the slowly moving sea (sea clutter). Clutter) when imaging moving maritime objects. It is exploited that the signal components of the clutter in the different channels depend only on the location, but not on the date of acquisition.

For the suppression of the clutter, different clutter suppression filters can be used. As a rule, the inverses of the clutter covariance matrix or a suitable projection matrix are used, which are adaptively calculated on the basis of changing statistical properties of the clutter. Other adaptive or theoretical filters or an FIR filter (FIR: finite impulse response) can also be used.

After suppressing the clutter, the remaining signal components contain only echoes of the moving objects and noise. In order to image one of the objects, the distance profile between the radar sensor and the object as well as the profile of the aspect angle at which the object is illuminated by the radar must be known exactly. Since in most applications the movement of the object is unknown, it must be estimated taking into account the preprocessed radar echoes. Both the methods known in the MTI field, which are based on the multi-channel filtered data (eg maximum likelihood method), and the methods known from the ISAR field for estimating the motion of the various object types (land vehicles, maritime Objects and planes). If available, additional knowledge can also be considered. Thus, for an object that moves along a road, for example, the knowledge about the course of the road proves to be helpful. This knowledge of the road can then be used as additional knowledge to support the motion estimation of the object.

To generate an image of the object, its echoes freed from the clutter, i. H. coherently superimposed the signal components or data remaining after the clutter suppression, taking into account the determined distance profile and the course of the aspect angle. For this purpose, from the literature known reconstruction methods in the local and frequency domain can be used. Examples of such methods are the methods of backprojection, fast backprojection, range-doppler, polar-format, range-migration, or chirp-scaling. For an optimum signal-to-noise ratio, a combination of the channels can also take place, taking into account the relative movement between the radar sensor and the object. For this purpose, several channels can be superimposed before or after image formation with a beamformer.

2 shows an example of a concrete realization of the method. The processing of the received radar signals takes place here in three stages. In this case, the vector Z designates the multi-channel data after preprocessing. Preprocessing in this example includes range compression and calibration to approximate the channels. The parameter r denotes the distance and the parameter f _d the Doppler frequency.

In the first stage, suppression of the signal components caused by clutter occurs by processing spatiotemporal correlated echo signals. This clutter suppression (stage 1) takes place in the Doppler range by using a clutter filter matrix B. Subsequently, the filtered multichannel data Z _{MT are transformed back} into the time domain, wherein the variable t _s denotes the time.

In the next processing stage (stage 2), the parameters of the moving object (target) are estimated. These are the target position (x, y) and the target speed (v _x , v _y ) in the coordinate system of the mobile platform or of the moving radar sensor.

Subsequently, the data of the different channels are coherently combined by multiplication with the steering vector b (stage 3, first equation). The steering vector takes into account the relative movement between the radar sensor and the object determined in the preceding step.

Finally, the ISAR image is calculated (level 3, second equation). In this example, the backprojection algorithm is used to avoid approximations. Alternatively, however, it is also possible to use other known image reconstruction methods which, although they do not manage without approximations, work numerically more efficiently.

LIST OF REFERENCE NUMBERS

1

plane

2

ground vehicle

3

radar sensor

4

radar lobe

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited non-patent literature

• M. Stuff et al., "Imaging Moving Objects in 3D from Single Aperture Synthetic Aperture Radar", Radar Conference, 2004, Proceedings of the IEEE, 26.-29. April 2004, pages 94 to 98 [0003]
• D. Pastina, "Radiation Conference 2008, RADAR '08, IEEE, 26.-30. May 2008, pages 1 to 6 [0003]
• D. Cerutti-Maori et al., "Optimum GMTI Processing for Space-based SAR / GMTI Systems - Theoretical Derivation", EUSAR 2010, Aachen, Germany, June 2010, pages 390-393 [0007]
• R. Klemm, "Interrelated Problems in Space-Time Processing for SAR and ISAR", Radar, Sonar and Navigation, IEEE Proceedings, Vol. 5, Oct. 1998, pages 297 to 302 [0008]

Claims (16)

Method for imaging moving objects by means of a moving radar or sonar sensor, in which the radar or sonar sensor ( 3 ) Pulses in the direction of a moving object ( 2 ) and echo signals are detected, from which, on the basis of a motion estimation of the object ( 2 ) a radar or sonar image of the object ( 2 ), wherein spurious signal portions not originating from the object are suppressed by processing space-time correlated echo signals, the motion estimation and imaging are performed on the basis of the signal portions remaining after the suppression of the spurious signal portions, and the image formation based on a coherent observation period is chosen which is so large that an image of the object is made possible. A method according to claim 1, characterized in that the spatio-temporal correlation is achieved by recording the echo signals with a multi-channel receiver. A method according to claim 1 or 2, characterized in that the spatio-temporal correlation is achieved by switching of antennas or antenna groups when transmitting the pulses or when receiving the echo signals. Method according to one of claims 1 to 3, characterized in that as disturbing signal components caused by clutter and / or interferer signal components are suppressed. Method according to one of claims 1 to 4, characterized in that an adaptive or theoretical filter is used to suppress the interfering signal components. Method according to one of Claims 1 to 4, characterized in that the inverse of the covariance matrix of the disturbing signal components to be suppressed is used to suppress the interfering signal components. Method according to one of claims 1 to 4, characterized in that for the suppression of the interfering signal components, a projection matrix is used. Method according to one of claims 1 to 4, characterized in that for suppressing the interfering signal components, an FIR filter is used. Method according to one of claims 1 to 8, characterized in that for the motion estimation of the object ( 2 ) Additional information is used. Method according to one of claims 1 to 9, characterized in that individual channels of the radar or sonar sensor are further processed for image formation. Method according to one of claims 1 to 9, characterized in that to improve the signal-to-noise ratio before imaging multiple channels of the radar or sonar sensor are superimposed with a beamformer. Method according to one of Claims 1 to 10, characterized in that, in order to improve the signal-to-noise ratio after image formation, a plurality of channels of the radar or sonar sensor are superimposed with a beamformer. Method according to one of claims 1 to 12, characterized in that the image formation takes place by application of methods of backprojection or rapid backprojection. Method according to one of claims 1 to 12, characterized in that the image is generated in the frequency or wave number range. A method according to claim 14, characterized in that the image is formed with the range Doppler method, the polar format method, the range migration method or the chirp scaling algorithm. Device for imaging moving objects, comprising at least one radar or sonar sensor with one or more antennas for emitting pulses in the direction of a moving object ( 2 ) and for receiving echo signals, as well as a control and evaluation device which according to the method of the preceding claims, the antenna (s) for emitting the pulses and for receiving the echo signals drives and performs the processing of the echo signals according to this method.

Images