System and method for simulating ground clutter for radar - Patents.com

Abstract

An improved method of simulating ground or sea clutter for radars with utility for radar testing and design. The method comprises: providing a terrain model comprising discrete plain patches, each having reflectivity, area, surface normal vector, and position in a global coordinate frame; calculating triplet values ​​of reflected energy, range, and Doppler shift for each discrete patch of the terrain model for a given position and attitude of an antenna with known gain and phase characteristics in the global coordinate frame; and resampling and integrating over all patches through a 2D transformation using a non-uniform Fast Fourier Transform to derive, for a given radar receiver sample rate and pulse repetition frequency (PRF), either i) the energy in each discrete range-Doppler pair bin of a set defined by the PRF and sample rate, and / or ii) the receiver signal level at discrete time samples per pulse. The method corrects for the lack of spatial correlation, allowing for simulations that have the benefits of the fixed Earth method but are as accurate as using the Iso-range method. This significantly increases the speed at which realistic ground reflection modeling can be generated.

Description

[Background technology]

[0001] Radar ground reflection, also known as ground clutter return, is a persistent challenge for airborne radar. Ground clutter returns depend on terrain, platform attitude, antenna steering, and the radiation patterns of both the transmitter and receiver antennas. Pulsed Doppler radar allows targets of interest to be differentiated from ground reflections by their Doppler shift as well as range. However, the periodic nature of the pulse train creates ambiguities in range and, in some cases, Doppler shift for both potential targets and ground clutter.

[0002] With the advent of array antennas and multi-channel receivers, techniques exist for the suppression of ground clutter based on the angle of arrival, such as space-time adaptive processing (STAP) and more generally adaptive detectors. Real-world radar data is expensive to collect and sets up a circular dependency between the radar antenna and receiver design and the data collection. Therefore, a realistic simulation of ground clutter is a desirable tool for the development of multi-channel radar systems.

[0003] A traditional approach to simulating ground clutter returns for airborne pulsed Doppler radar systems, referred to herein as the Iso-range method, is to first select a sample rate and a pulse repetition frequency (PRF), which together define discrete range and Doppler "bins" in the radar detector. These can then be projected onto a ground model that defines ground patches separated by the intersection of the iso-range curve with the iso-angle line. Pulsed Doppler radar is generally ambiguous in range, especially for high PRF (HPRF) modes. It can also be ambiguous in Doppler, especially for medium PRF (MPRF) modes. This means that each range-Doppler bin receives power reflected from multiple distinct ground patches whose boundaries are defined by the platform position and attitude, as well as by the PRF and sample rate of the radar. The energy reflected by each patch depends on the illumination angle, patch ground area, and material reflectivity, which can be statistically generated from a probability distribution, e.g., exponential or Weibull.

[0004] Ground patches defined in this way map unambiguously to range gates by construction. Although such patches are defined with respect to a constant azimuth angle, an equal division of the angle does not map to an equal division in Doppler. Thus, the contribution of one ground patch may affect multiple Doppler bins, and this interpolation is managed by first converting to a time-domain waveform representing each pulse and then applying Doppler compression. In this way, spatial correlation between nearby Doppler bins is maintained.

[0005] However, defining ground patches in terms of the platform position and attitude makes it difficult to maintain consistent models for ground altitude and reflectivity, which is particularly important for bistatic radar systems, and means that long-term correlations for ground clutter are difficult to maintain when accompanied by changes to the platform position or attitude, as required for ground imaging applications such as Synthetic Aperture Radar (SAR).

[0006] A more recent approach to clutter modeling, referred to herein as the fixed Earth method, uses fixed, persistent models for the ground geometry and reflectivity, which allows long-term correlation to be maintained in addition to providing the opportunity to decouple transmitter illumination from the receiving antenna position and attitude required for bistatic radar modeling.

[0007] A common approach across both the Iso range and fixed Earth methods is to model the ground as a discrete set of flat "patches", calculate the expected radar return from each using a radar equation, and then sum the contributions from every patch. Where the methods differ is in how the ground patches are defined and over what domain this integration takes place.

[0008] Compared to the Iso-range method, the Fixed Earth method achieves greater processing speed improvements from running on a graphics processing unit (GPU). Furthermore, since the model does not need to be regenerated for changes in the platform's position, it allows for the simulation of consistent clutter returns over the flight path, thus enabling ground clutter modeling for bistatic radar or SAR. Furthermore, it is possible to incorporate altitude and ground type information of real landscapes, obtained for example from third-party datasets. The result of this is that the simulated ground clutter can be compared to real-world flight test data for accuracy. Summary of the Invention

[0009] However, a problem with using a fixed Earth model is the relationship between the boundaries of the ground patches, and the range and Doppler division of the receiver is not straightforward. Ignoring this leads to a lack of spatial correlation and therefore to an unrealistic modelling of ground clutter compared to the Iso-range method. The present invention was thought to remedy this problem.

[0010] According to a first aspect of the present invention there is provided a computer implemented method of simulating radar returns from ground and / or sea clutter for a pulsed Doppler radar comprising the steps of: a) providing a terrain model comprising a plurality of discrete flat land patches, all independent of a viewpoint, each having a respective reflectivity, area, surface normal vector, and position within a global coordinate frame; b) calculating triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of the antenna with known gain and phase characteristics in a global coordinate frame; c) For a given radar receiver sample rate and pulse repetition frequency (PRF): i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and the sample rate; and / or ii) The receiver signal level at discrete time samples for each pulse. and resampling and integrating over all patches to derive either A method comprising: d) the process of resampling and integrating comprises performing a first 2D transformation of the triplet values ​​using a Non-Uniform Fast Fourier Transform (NUFFT) followed by a further transformation of the output of the first 2D transformation using an Inverse Fast Fourier Transform; A method is provided.

[0011] Use of the present invention corrects for the lack of spatial correlation allowing simulations that have the benefits of fixed Earth methods but are as accurate as using the Iso-range method.

[0012] The method may further include resampling across all patches to normalize range values ​​such that the ambiguous range associated with the PRF is equal to 1 and the ambiguous Doppler associated with the PRF is equal to 1, and scaling the range and Doppler shift values ​​before integrating. The combination of using the NUFFT with the scaling factor implements the range and Doppler aliasing inherent in pulsed Doppler radar.

[0013] When implemented for a bistatic radar, the method may include calculating triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of both non-co-located transmit and receive antennas in a global coordinate frame, each having known gain and phase characteristics.

[0014] The pulse Doppler radar may be an airborne pulse Doppler radar.

[0015] The terrain model may simulate one or more realistic and / or real-world terrain features. For example, the terrain model may be generated by defining a fixed ground mesh by mapping digital real-world elevation map data to geocentric, earth-fixed coordinates. Alternatively, the terrain may be randomly generated, for example using Perlin noise.

[0016] In another aspect of the invention, there is provided a method of performance testing a radar component comprising: providing (which may comprise configuring) a computer model of the radar component (which may include selecting design parameters of the radar component, e.g., one or more of antenna gain characteristics, beam shape, PRF, operating frequency, type, number of clutter filters design), and simulating operation of the radar component by running the computer model of the radar component using simulated radar reflections from ground and / or sea clutter derived from the method of simulating radar reflections from ground and / or sea clutter for a pulsed Doppler radar described above as inputs to the computer model.

[0017] The invention may also be expressed in apparatus terms, and thus in yet another aspect of the invention there is provided a computer implemented return simulator for simulating radar returns from ground clutter for a pulsed Doppler radar, comprising: a computer readable store holding a terrain model consisting of a plurality of discrete plain patches, all independent of viewpoint, each having a respective reflectance, area, surface normal vector, and position within a global coordinate frame; One or more processors wherein the one or more processors a radar renderer configured to calculate triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of the antenna with known gain and phase characteristics in a global coordinate frame; For a given radar receiver sample rate and pulse repetition frequency (PRF), i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and the sample rate; and / or ii) The receiver signal level at discrete time samples for each pulse. and a model calculator adapted to resample and integrate over all patches to derive either A reflection simulator configured to perform the functions of characterised in that the resampling and integrating process performed by the model calculator comprises a first 2D transformation of the triplet values ​​using a NUFFT, followed by a further transformation of the output of the first transformation using an inverse FFT, A reflex simulator is provided.

[0018] The model calculator may include a range-Doppler resampler to derive the energy within each discrete range-Doppler pair bin of a set defined by the PRF and the sample rate. The model calculator may include a time-domain resampler to derive the receiver signal level at discrete time samples for each pulse.

[0019] The computer-implemented reflectance simulator may be incorporated into a radar test apparatus, thus providing a radar test apparatus comprising a signal generator adapted to generate a signal derived from a specified ground clutter model generated by a model calculator of the computer-implemented reflectance simulator to mimic expected signal levels from ground clutter.

[0020] The present invention may also find application for simulating synthetic aperture radar returns and therefore, according to a further aspect of the invention, there is provided a method of simulating synthetic aperture radar returns for a pulsed Doppler radar, comprising the steps of: a) providing a terrain model comprising a plurality of discrete flat land patches, all independent of viewpoint, each having a respective reflectivity, area, surface normal vector, and position within a global coordinate frame; b) calculating triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of the antenna with known gain and phase characteristics in a global coordinate frame; c) For a given radar receiver sample rate and pulse repetition frequency (PRF): i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and the sample rate; and / or ii) The receiver signal level at discrete time samples for each pulse. and resampling and integrating over all patches to derive either A method comprising: d) the process of resampling and integrating comprises performing a first 2D transformation of the triplet values ​​using a Non-Uniform Fast Fourier Transform (NUFFT) followed by a further transformation of the output of the first transformation using an Inverse Fast Fourier Transform; A method is provided.

[0021] The invention will now be described, by way of example only, with reference to the following drawings, in which: [Brief description of the drawings]

[0022] [Figure 1] Schematic diagram of a system for simulating target reflections and ground clutter to test radar component performance. [Diagram 2] Figure 1. Schematic of the ground clutter simulator. [Diagram 3] Process steps performed by the ground clutter simulator. [Figure 4A] FIG. 10 shows details of the processing performed during the resampling and integration processing steps by the range Doppler resampler. [Figure 4B] FIG. 4 shows details of the processing performed during the resampling and integration processing steps by the time domain resampler. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Referring to Figure 1, there is shown a system 1 for simulating target reflections and ground clutter for testing radar system components. The system 1 comprises a ground clutter simulator 2 adapted to output a range-Doppler clutter map 3 and / or a time-domain clutter waveform 4 based on terrain data in the form of a land use map 5, a digital elevation map 6, platform position and attitude information 7, and a model 8 of the gain characteristics of the (each, if there is more than one) antenna.

[0024] These simulated ground data can be used along with simulations of potential target returns 9 to test or evaluate the performance of radar system components such as a radar detector 10, a radar receiver 11 (or computer models thereof, i.e., an emulated radar detector or an emulated radar receiver) to measure radar performance metrics 12, 13.

[0025] 2 shows the functional components of the ground clutter simulator 2. The ground clutter simulator 2 comprises a ground model generator 20, a radar renderer 21, a range-Doppler resampler 22 and a time domain resampler 23.

[0026] Figure 3 shows the method performed by the ground clutter simulator 2. A ground model generator 20 generates a ground model 30. A radar renderer 21 specializes the ground model to a particular radar, platform position and attitude 31. Finally, a resample and integrate process 32 further specializes to a particular PRF, represented as either a range-Doppler energy map 33 or receiver signal samples 34 for the pulse train. This final step can be repeated for different PRF bursts.

[0027] Returning to Figure 2, the ground model generator 20 outputs a ground mesh model 24 consisting of contiguous triangular (triangular is preferred, but not required) patches represented in a fixed global coordinate frame, with each patch having associated values ​​of reflectivity, area, surface normal vector, and position in the global coordinate frame.

[0028] To generate the ground mesh model 24, the ground model generator 20 first defines a fixed ground mesh, which may be generated by mapping the digital elevation map data to Earth-centered, Earth-fixed coordinates using an Earth ellipsoid model or by using a Delaunay triangulation of the filtered LiDAR point cloud data. The terrain may alternatively be generated randomly, for example using Perlin noise. The ground model generator 20 applies a radar reflectivity value γ to each ground patch. This may be achieved by incorporating land use map data of terrain types, for example from a third party source, in combination with a table relating terrain types to their reflectivity.

[0029] The area, surface normal vector, and position of each patch may be determined as follows:

[0030] Let the vertices of a triangular ground patch have position vectors a, b, c in the global coordinate frame.

[0031] The center y of the patch is:

[0032]

number

[0033] The area A is:

[0034]

number

[0035] The normal n of a patch vector is given by:

[0036]

number

[0037] The radar renderer 21 is configured to use the ground mesh model 24 together with the given platform position and attitude data 7 and the antenna gain model G 8 to calculate, for each patch in the ground mesh model 24, a triplet value 25 of reflected energy, range and Doppler shift.

[0038] The ranges are as follows:

[0039]

number

[0040] The patch displacement direction vector d is given by:

[0041]

number

[0042] The elevation and azimuth angles as direction cosines u, v are:

[0043]

number

[0044]

number

[0045] where θ is the elevation angle and ψ is the azimuth angle. The grazing angle φ of the patch relative to the antenna beam is:

[0046]

number

[0047] Antenna gain model G and transmitted energy E t Together, the following radar equation gives the reflected energy value:

[0048]

number

[0049] Finally, the carrier frequency f c The Doppler shift δ for is:

[0050]

number

[0051] The output of the radar renderer 21 is a set of scalar triplets (r, δf, E) 25. These are converted by the time domain resampler 23 into time domain triplets 26 of sample time t, pulse number k and energy E′, or by the range-Doppler resampler 22 into range-gated triplet 27 of sample time t, pulse number k and energy E′.

[0052]

number

[0053] , Doppler Bin

[0054]

number

[0055] and clutter energy

[0056]

number

[0057] and integrated into a range-Doppler map 27 defined by a triplet of

[0058] The above applies to monostatic radar. The radar renderer 21 may be replaced with a variant for bistatic radar that employs two passes, an illumination pass and an observation pass. In the illumination pass, the unidirectional radar equation is used to determine the energy at each patch due to the transmitting antenna. The observation pass uses the unidirectional radar equation to determine the energy reflected to the observer. A reflection model is applied that uses both the illuminator angle and the observer angle along with the patch.

[0059] Referring to FIG. 4A, the range-Doppler resampler 22 performs a first, two-dimensional (2D) transformation of the triplet values ​​25 using a Non-Uniform Fast Fourier Transform (NUFFT) 40, and then performs a further transformation of the output of the first transform using a 2D inverse Fast Fourier Transform 41.

[0060] The 2D NUFFT 40 performs the functions of integration and resampling, but the output is in the transform domain. The 2D inverse FFT serves to bring the data back into the range and Doppler onset domain.

[0061] Referring to FIG. 4B, the time domain resampler 23 performs a first two-dimensional (2D) transformation of the triplet values ​​25 using a Non-Uniform Fast Fourier Transform (NUFFT) 40, and then further transforms the output of the transform using a 1D inverse Fast Fourier Transform 42 along the range dimension.

[0062] NUFFT40 is a computationally efficient method to implement the Nonuniform Discrete Fourier Transform, which can be written as:

[0063]

number

[0064]

number

[0065]

number

[0066] Where:

[0067]

number

[0068] NUFFT40 implements NUDFT as follows:

[0069] i) An input triplet 25 of value (energy) and two coordinates (range and Doppler) is resampled onto a regular grid of range-Doppler bins using a specified interpolation kernel function. This gives the interpolation function 43.

[0070] ii) The resampled values ​​that are mapped to the same bin are added together. This gives the integral function44.

[0071] iii) A regular 2D FFT is applied to all range-Doppler bins45.

[0072] iv) The bins are element-wise multiplied by a 2D weighting function46. This function varies over a regular grid and is the same size as the regular grid.

[0073] Further details of the implementation of NUFFT can be found in J. A. Essler, "On NUFFT-based gridding for non-Cartesian MRI", Journal of magnetic resonance, Vol. 188, No. 2, pp. 191-195, 2007, which uses Kaiser-Bessel functions as its interpolators, and K. Fourmont, "Non-equispaced fast Fourier transforms with applications to tomography", Journal of Fourier Analysis and Applications, Vol. 9, No. 5, pp. 431-450, 2003, and an extension is given in P. J. Beatty, D. G. Nishimura and J. M. Fauli, "Rapid gridding reconstruction with a minimal oversampling ratio", IEEE transactions on medical imaging, Vol. 24, No. 6, pp. 799-808, 2005, for choosing a reasonable scale parameter for the interpolation kernel.

[0074] Before applying the NUFFT 40, the range and Doppler coordinate values ​​of the triplets 25 may be scaled by a coordinate scaling operator 47. The previously computed ground patch triplet representation (r t , δf t ,E t ) and then scale it appropriately to the given PRF f r Render with a specific range and Doppler aliasing. The number of pulses is N p and the number of samples in the PRI is M r Let N be the p ×M r We apply a NUFFT with output dimensions of . The coordinate scaling required to mimic the aliasing induced by the PRF is

[0075]

number

[0076]

number

[0077] This has the effect of normalizing the Doppler shift so that the ambiguous Doppler shift becomes unity and similarly the ambiguous range becomes unity.

[0078] Next, these scaled triplets are p ×M r Apply the dimensional NUFFT.

[0079]

number

[0080] Variations to the previously described example are possible. For example, in an alternative embodiment, the ground clutter simulator may comprise only one of the resamplers 22, 23.

[0081] The methods described above can be used to simulate Synthetic Aperture Radar (SAR) returns by repeating the process for the positions and attitudes of multiple platforms spaced along a flight path.

[0082] Similarly, multi-burst dwells can be simulated by repeating the resampler stages 22, 23 with several different PRFs.

[0083] The ground clutter simulator 2 may be implemented by one or more processors programmed with suitable software. A typical combination of hardware and software may be a general-purpose computer system with a computer program that, when loaded and executed, controls the computer system to perform the methods described herein.

[0084] When used to evaluate the performance of an emulated radar component (e.g., a radar receiver or radar detector), system 1 may be based generally on one or more processors programmed with suitable software.

[0085] In one application, a computer model of a radar component is generated having a selected set of design parameters, including, for example, antenna gain characteristics, PRF, operating frequency, and one or more specific clutter filters. The radar component model is run with input data derived from a specified ground clutter model constructed by the previously described system that mimics the expected signal from ground clutter and simulated target reflections. The output of the radar component model is output to a resource, for example printed, or displayed on a visual display of a computer. The results can be used to determine whether (and if so, how well) targets can be differentiated from clutter. One or more of the design parameters of the emulated radar component can then be changed and testing repeated to identify differences in radar system performance with the overall goal of improving the radar component design.

[0086] When the system is used to performance test a physical radar component (as opposed to an emulated radar component), the system includes a signal generator, e.g., with a digital-to-analog converter, having an output connected to a test signal input point on the radar component under test. The signal generator is adapted to generate a (typically radio frequency) signal derived from a specified ground clutter model constructed by the previously described system that mimics a signal expected from ground clutter (and optionally also simulated target reflections) to test the performance of the radar component. As with an emulated component, the test may be repeated after changing one or more of the operating parameters of the radar component.

[0087] As described above, the system can be used to simulate realistic (but not real-world) ground clutter associated with one or more types of terrain relief, as well as to simulate ground clutter reflections expected from real-world locations. Although the above description is directed to simulating ground clutter, the method can also be applied to simulate sea clutter.

[0088] Although conceived for the purposes of simulating ground clutter, the inventors recognize that the method may also be applicable to simulating synthetic aperture radar returns and thus may be used in the design and / or performance testing of synthetic aperture radar.

Claims

1. 1. A computer-implemented method for simulating radar returns from ground and / or sea clutter for a pulsed Doppler radar, comprising: a) providing a terrain model comprising a plurality of discrete plain patches, all independent of viewpoint, each having a respective reflectivity, area, surface normal vector, and position within a global coordinate frame; b) calculating triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of said terrain model for a given position and attitude of an antenna with known gain and phase characteristics in said global coordinate frame; c) For a given radar receiver sample rate and pulse repetition frequency (PRF): i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and sample rate; and / or ii) the receiver signal level at discrete time samples for each pulse; Resample and integrate over all patches to derive A method comprising: d) the resampling and integrating process comprises performing a first 2D transformation of the triplet values ​​using a Non-Uniform Fast Fourier Transform (NUFFT), followed by a further transformation of the output of the first 2D transformation using an Inverse Fast Fourier Transform.

2. 2. The method of claim 1, further comprising: resampling across all patches to normalize range values ​​such that ambiguous range related to pulse repetition frequency (PRF) is equal to 1 and ambiguous Doppler related to PRF is equal to 1, and scaling the range and Doppler shift values ​​before integrating.

3. 2. The method of claim 1, comprising: calculating a triplet of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of both non-co-located transmit and receive antennas in the global coordinate frame, each having known gain and phase characteristics.

4. 1. A computer-implemented return simulator for simulating radar returns from ground and / or sea clutter for a pulsed Doppler radar, comprising: a computer readable store holding a terrain model consisting of a plurality of discrete plain patches, all independent of viewpoint, each having a reflectance, an area, a surface normal vector, and a position within a global coordinate frame; one or more processors; wherein the one or more processors: a radar renderer configured to calculate triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of the terrain model for a given position and attitude of an antenna with known gain and phase characteristics in the global coordinate frame; For a given radar receiver sample rate and pulse repetition frequency (PRF), i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and sample rate; and / or ii) the receiver signal level at discrete time samples for each pulse; and a model calculator adapted to resample and integrate over all patches to derive either A reflection simulator configured to perform the functions of 11. A reflectance simulator, wherein the resampling and integrating process performed by the model computer comprises performing a first 2D transformation of the triplet values ​​using a Non-Uniform Fast Fourier Transform (NUFFT), followed by a further transformation of the output of the first 2D transformation using an Inverse Fast Fourier Transform.

5. 5. A radar test apparatus comprising: a signal generator adapted to generate a signal derived from a specified ground clutter model generated by the model calculator of claim 4 to mimic signal levels expected from ground clutter.

6. 1. A method for simulating synthetic aperture radar returns for a pulsed Doppler radar, comprising: a) providing a terrain model comprising a plurality of discrete plain patches, all independent of viewpoint, each having a reflectivity, an area, a surface normal vector, and a position in a global coordinate frame; b) calculating triplet values ​​of reflected energy, range and Doppler shift for each discrete patch of said terrain model for a given position and attitude of an antenna with known gain and phase characteristics in said global coordinate frame; c) For a given radar receiver sample rate and pulse repetition frequency (PRF): i) the energy within each discrete range-Doppler pair bin of a set defined by the PRF and sample rate; and / or ii) the receiver signal level at discrete time samples for each pulse; and resampling and integrating over all patches to derive either A method comprising: d) the resampling and integrating process comprises performing a first 2D transformation of the triplet values ​​using a Non-Uniform Fast Fourier Transform (NUFFT), followed by a further transformation of the output of the first 2D transform using an Inverse Fast Fourier Transform.

7. 1. A method of performance testing a radar component, comprising: providing a computer model of the radar component; simulating operation of the radar component by running the computer model of the radar component using, as input to the computer model, simulated radar returns from ground and / or sea clutter derived from a computer-implemented method for simulating radar returns from ground and / or sea clutter for a pulse Doppler radar according to claim 1, 2 or 3; A method comprising:

Images