Strabismus quasi-static bistatic SAR imaging method based on improved PCS

Abstract

The invention discloses a squint quasi-static bistatic SAR imaging method based on an improved PCS, and the method comprises the steps: calculating a scale transformation factor in each azimuth according to a radar parameter, and determining a scale factor in a distance direction; multiplying an echo pulse by an improved phase factor in a time domain, and converting an echo signal into a frequency domain; multiplying the frequency domain signal by the improved matched filtering function, and then converting the signal back to the time domain; multiplying a time domain signal by an improved phase factor in a time domain, and converting an echo signal into a frequency domain; multiplying the signal transformed back to the frequency domain by an improved compensation function in the frequency domain; calculating an equivalent phase center of the quasi-static bistatic SAR, and performing azimuth interpolation processing to obtain data uniformly distributed in a wavenumber domain; and performing two-dimensional Fourier transform on the data uniformly sampled in the wavenumber domain to obtain a final imaging result. The method can effectively solve the problem that under the strabismus condition, a quasi-static bistatic SAR uses an overlarge scale transformation factor in an original PCS, so that ghosting occurs in an imaging result, and the imaging quality is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of radar imaging, and in particular relates to a squint quasi-static bistatic SAR imaging method based on improved PCS. Background Art

[0002] Microwave radars that can perform imaging are called synthetic aperture radars (SAR). Synthetic aperture radars place a small antenna on a moving carrier. When the carrier moves, the antenna also moves. Within a period of time, the position of the antenna is different at each moment. The movement trajectory of the antenna during the entire period of time is equivalent to a large antenna, which effectively improves the radar's azimuth resolution. Compared with traditional optical imaging technology, SAR has the advantages of all-weather and all-day, that is, it can still work normally under poor lighting conditions, and its resolution is independent of the range of action; compared with traditional radars, SAR has the characteristics of high resolution and can achieve two-dimensional or even three-dimensional imaging. With the continuous development of SAR, the shortcomings of single-base SAR, such as susceptibility to interference, poor survivability, and only being able to obtain scattering characteristics of a single target direction, have gradually been exposed. Bistatic SAR has gradually gained widespread attention for its strong anti-interference ability and strong survivability.

[0003] Polar Format Algorithm (PFA) is mainly used in spotlight synthetic aperture radar imaging to obtain better imaging effects through longer signal coverage. When the radar works in the spotlight mode, the carrier must constantly adjust the direction of the transmitting antenna during the movement to keep the imaging area always within the radar illumination. Due to the large turning angle of target observation, the spotlight mode can have a higher azimuth resolution. In the process of using PFA, the most computationally intensive part is the interpolation processing. The use of Principle of Chirp Scaling (PCS) can effectively reduce the computational complexity of PFA, but the traditional PCS will cause ghosting in the imaging results under squint conditions, so it needs to be improved. Summary of the invention

[0004] Purpose of the invention: The present invention provides a squint quasi-static bistatic SAR imaging method based on improved PCS, which eliminates the problem of ghosting in imaging results under squint conditions.

[0005] Technical solution: The invention discloses a squint quasi-static bistatic SAR imaging method based on improved PCS, comprising the following steps:

[0006] (1) Calculate the scale transformation factor δ for each azimuth according to the radar parameters r ;

[0007] (2) Determine the scale transformation factor β in the distance direction through the scale transformation factor;

[0008] (3) Multiplying the echo pulse in the time domain by the phase factor improved based on β, and then performing a range-to-Fourier transform to transform the echo signal into the frequency domain;

[0009] (4) multiplying the signal transformed into the frequency domain by the improved matched filter function, and then performing a range-direction inverse Fourier transform to transform the signal back into the time domain;

[0010] (5) multiplying the signal transformed back to the time domain by the phase factor after the improved matched filtering in the time domain, performing a range-to-Fourier transform again, and transforming the echo signal into the frequency domain;

[0011] (6) multiplying the signal in step (5) by the improved compensation function in the frequency domain to obtain an echo signal that has been processed in the range direction;

[0012] (7) Calculate the equivalent phase center of the quasi-static bistatic SAR, and use the equivalent phase center to perform azimuth interpolation processing on the echoes that have been processed in the range direction to obtain data that is uniformly distributed in the wavenumber domain;

[0013] (8) Perform a two-dimensional Fourier transform on the uniformly sampled data in the wavenumber domain to obtain the final imaging result.

[0014] Furthermore, the scale transformation factor in step (1) is:

[0015] α(t)=[cosφ t (t)cosθ t (t)+cosφ r (t)cosθ r (t)]cosφ b +[sinφ t (t)cosθ t (t)+sinφ r (t)cosθ r (t)]sinφ b

[0016]

[0017]

[0018] In the formula, in order to improve the utilization of data, the data is rotated in the wave number domain, and the rotation angle is φ b ,φ t ,θ t is the instantaneous azimuth and elevation angle of the transmitter phase center; φ r ,θ rare the instantaneous azimuth and elevation angles of the receiver phase center; t0 is the time of the synthetic aperture center.

[0019] Furthermore, the upward scale factor of the distance in step (2) is:

[0020] β=max(β1,β2,1)

[0021]

[0022]

[0023] Where β1 is the scale transformation factor in the distance calculated by the first pulse emission time, β2 is the scale transformation factor in the distance calculated by the last pulse emission time, and f c is the carrier frequency of the transmitted signal, fs is the sampling frequency in the range direction, δ r,1 is the scale transformation factor of the azimuth corresponding to the first pulse, δ r,na is the scale transformation factor of the azimuth direction corresponding to the last pulse.

[0024] Furthermore, the improved phase factor in step (3) is:

[0025]

[0026] Where τ represents the fast time variable, c is the speed of light, and k is the frequency modulation slope of the linear frequency modulation signal; δ r is the azimuth scale transformation factor, R a It is the sum of the distance from the transmitting antenna phase center to the scene center and the distance from the receiving antenna phase center to the scene center.

[0027] Furthermore, the improved matched filter function described in step (4) is:

[0028]

[0029] In the formula, f τ is the frequency variable corresponding to the distance fast time variable τ.

[0030] Furthermore, the phase factor after the improved matched filtering described in step (5) is:

[0031]

[0032] Furthermore, the improved compensation function described in step (6) is:

[0033]

[0034] Furthermore, the implementation process of step (7) is as follows:

[0035]

[0036] Among them, g(x) is the directional interpolation corresponding to the interpolation point x, sinc(x) is the interpolation kernel, sinc is the sinc function in mathematics, x is the offset in the convolution calculation, and i is the timing in the convolution calculation;

[0037] The coordinates before and after interpolation are calculated as:

[0038]

[0039]

[0040] Among them, t is the original azimuth time variable, t' is the azimuth time variable after Keystone transformation; R g is the length of the projection of the equivalent phase center position on the xoy plane, v is the common motion speed of the transmitting platform and the receiving platform, θ s is the equivalent oblique angle.

[0041] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention eliminates the ghosting in the imaging results under the condition of strabismus through the improved PCS algorithm, and performs improved PCS processing in the range direction and interpolation processing in the azimuth direction on the quasi-static bistatic SAR echo signal, so as to obtain an image with good focusing effect and no ghosting, and has the characteristics of high efficiency and easy implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the present invention;

[0043] Figure 2 The distribution of point targets in the simulation experiment of the present invention;

[0044] Figure 3 The pulse pressure lines after processing in two directions; (a) is the pulse pressure line processed by the original PCS; (b) is the pulse pressure line processed by the improved PCS method;

[0045] Figure 4 Schematic diagram of PFA imaging results; wherein (a) is a schematic diagram of the original PFA imaging result based on PCS; (b) is a schematic diagram of the PFA imaging result based on the improved PCS of the present invention;

[0046] Figure 5 Point target analysis of the PFA center point O for improving PCS; (a) the frequency spectrum corresponding to the center point; (b) the interpolated contour map corresponding to the center point O.

[0047] Figure 6Point target analysis of an edge point of PFA to improve PCS; (a) the frequency spectrum corresponding to the point; (b) the interpolated contour map corresponding to the point.

[0048] Figure 7 The flight paths of the two platforms are shown below. DETAILED DESCRIPTION

[0049] The present invention is further described in detail below with reference to the accompanying drawings.

[0050] Under the condition of squint, the echo signal is processed by PCS in the range direction and interpolation in the azimuth direction, which will cause ghosting in the final imaging result. The reason is that the azimuth scale transformation factor is too large, causing the frequency domain range of some azimuth directions to exceed the sampling range, resulting in spectrum aliasing, which is manifested as ghosting in the final image. In addition, in the quasi-static bistatic SAR, the flight speeds of the two platforms are the same, that is, the equivalent phase center does not change with time. Therefore, in the process of using the improved PCS, the range direction is basically the same as the single-base improved PCS method, while in the azimuth direction, the quasi-static bistatic SAR is equivalent to a single-base SAR and processed using the interpolation method. Compared with the use of the PFA algorithm for imaging, the PCS method can effectively reduce the amount of calculation in the range direction. Based on the above, the present invention provides a method such as Figure 1 The squint quasi-static bistatic SAR imaging method based on improved PCS includes the following steps:

[0051] Step 1: Calculate the scale transformation factor δ in each azimuth according to the radar parameters r .

[0052] The signal echo of the bistatic SAR can be expressed (ignoring the envelope effect) as:

[0053]

[0054] In the formula, τ represents the fast time variable, that is, the time variable in the range direction, t represents the slow time variable, that is, the time variable in the azimuth direction, c represents the speed of light, k represents the frequency modulation slope of the linear frequency modulation signal, and f represents the speed of light. c is the carrier frequency of the transmitted signal, and R is the sum of the distance from the phase center of the transmitting antenna to the target point and the distance from the receiving phase center to the target point.

[0055] The azimuth scale transformation factor δ r The calculation method is:

[0056]

[0057] α(t)=[cosφ t (t)cosθ t (t)+cosφ r (t)cosθr (t)]cosφ b +[sinφ t (t)cosθ t (t)+sinφ r (t)cosθ r (t)]sinφ b

[0058]

[0059] In the formula, φ b The angle of rotation is; φ t ,θ t is the instantaneous azimuth and elevation angle of the transmitter phase center; φ r ,θ r is the instantaneous azimuth and elevation angle of the receiver phase center; t0 is the time of the synthetic aperture center, δ r is a function of the azimuth time t. In order to improve the utilization of the data, the data are rotated in the wavenumber domain.

[0060] Step 2: Determine the distance upward scale factor β through the azimuth scale transformation factor, and optimize and improve the parameters of the subsequent steps through this new scale transformation factor:

[0061]

[0062]

[0063] β=max(β1,β2,1)

[0064] Where β1 is the scale transformation factor in the range calculated at the time of the first pulse emission and β2 is the scale transformation factor in the range calculated at the time of the last pulse emission, f c is the carrier frequency of the transmitted signal, fs is the sampling frequency in the range direction, δ r,1 is the scale transformation factor of the azimuth corresponding to the first pulse, δ r,na is the scale transformation factor of the azimuth direction corresponding to the last pulse.

[0065] Step 3: Multiply the echo pulse by the improved phase factor in the time domain, and then perform a range-to-Fourier transform to transform the echo signal into the frequency domain.

[0066] Improved Phase Factor The calculation method is:

[0067]

[0068] Where τ represents the fast time variable, c is the speed of light, and k is the frequency modulation slope of the linear frequency modulation signal; δr This is the azimuth scale transformation factor mentioned in 2, R a It is the sum of the distance from the transmitting antenna phase center to the scene center and the distance from the receiving antenna phase center to the scene center.

[0069] Step 4: Multiply the signal in step 3 by the improved matched filter function, and then perform inverse Fourier transform in the range direction to transform the signal back to the time domain;

[0070] Improved matched filter function H1(f τ ) is calculated as:

[0071]

[0072] In the formula, f τ is the frequency variable corresponding to the distance fast time τ.

[0073] Step 5: Multiply the signal in step 4 by the phase factor after the improved matched filtering in the time domain, perform range-to-Fourier transform again, and transform the echo signal into the frequency domain.

[0074] Improved Phase Factor The calculation method is:

[0075]

[0076] Step 6: Multiply the signal in step 5 by the improved compensation function in the frequency domain to obtain the echo signal that has been processed in the range direction.

[0077] Compensation function H2(f τ ) is calculated as:

[0078]

[0079] Step 7: Calculate the equivalent phase center of the quasi-static bistatic SAR, and use the equivalent phase center to perform azimuth interpolation processing on the echoes that have been processed in the range direction to obtain data that is evenly distributed in the wavenumber domain. The azimuth interpolation calculation formula is:

[0080]

[0081] Among them, sinc(x) is the interpolation kernel, sinc is the sinc function in mathematics, x is the offset in the convolution calculation, and i is the timing in the convolution calculation.

[0082] The coordinates before and after interpolation are calculated as t is the original azimuth time variable, and t' is the azimuth time variable after Keystone transformation.

[0083]

[0084] In the formula, R g is the length of the projection of the equivalent phase center position on the xoy plane, v is the common motion speed of the transmitting platform and the receiving platform, θ s is the equivalent oblique angle. Since the two platforms move at the same speed, this angle is also the oblique angle of each platform.

[0085] Step 8: Perform a two-dimensional Fourier transform on the azimuthally resampled signal to obtain the final imaging result.

[0086] The above echo and algorithm process are simulated on the simulation software MATLAB according to the echo model of quasi-static bistatic SAR. The point target distribution in the scene is as follows: Figure 2 As shown, the motion trajectory of the transceiver platform in the simulation experiment is shown as Figure 7 As shown in FIG. 1 , the simulated radar parameters are: the oblique angle is 45°, the radar signal carrier frequency is 10 GHz, the signal bandwidth is 400 MHz, the pulse duration is 2 us, the pulse repetition frequency is 1200 Hz, the sampling frequency is 500 MHz, and the carrier speed is 180 m / s. The original distance PCS processing flow and the improved distance PCS processing flow proposed in the present invention are used to process the echoes, and the pulse pressure lines of the point targets after the distance processing are shown in FIG. Figure 3 (a) and Figure 3 As shown in (b); Figure 3 The pulse pressure line in (a) obviously has an extra part, which is the ghost of each point target in the final image. Since the line has two parts, Figure 4 In (a), two parts of the double image appear. Figure 3 (b) only has the pulse pressure lines of each point target, without other components, so Figure 4 In (b) there are only corresponding point targets. Figure 3 The two images are not at the same distance because of the scale transformation factor β of the frequency axis. Figure 4 The center point of (b) is used to make a two-dimensional point target response contour map. The frequency spectrum and contour map of this point are as follows: Figure 5 (a) and Figure 5 As shown in (b); Figure 4 The edge point in (b) is plotted as a two-dimensional point target response contour map. The frequency spectrum and contour map of the point are shown in Figure 6 (a) and Figure 6 As shown in (b); because they are in the squint working mode, the spectrum of the two is not a rectangle but a parallelogram, and the obtained contour map is similar to the single-base squint state. This simulation example proves that the use of the original PCS under the dual-base quasi-static squint condition will cause ghosting in the image, while the use of the improved PCS makes the ghosting of the image disappear, verifying the feasibility of the algorithm.

[0087] The above content has described in detail the method for enhancing the training data of the traceability link recovery model based on unlabeled data of the present invention, but it is obvious that the specific implementation form of the present invention is not limited thereto. For those skilled in the art, various obvious changes made to it without departing from the spirit and scope of the present invention are within the protection scope of the present invention.

Claims

1. A squint quasi-static bistatic SAR imaging method based on improved PCS, characterized in that: The following steps are involved: (1) Calculate the scale transformation factor δ for each azimuth according to the radar parameters r ; (2) Determine the scale transformation factor β in the distance direction through the scale transformation factor; (3) Multiplying the echo pulse in the time domain by the phase factor improved based on β, and then performing a range-to-Fourier transform to transform the echo signal into the frequency domain; (4) multiplying the signal transformed into the frequency domain by the improved matched filter function, and then performing a range-direction inverse Fourier transform to transform the signal back into the time domain; (5) multiplying the signal transformed back to the time domain by the phase factor after the improved matched filtering in the time domain, performing a range-to-Fourier transform again, and transforming the echo signal into the frequency domain; (6) multiplying the signal in step (5) by the improved compensation function in the frequency domain to obtain an echo signal that has been processed in the range direction; (7) Calculate the equivalent phase center of the quasi-static bistatic SAR, and use the equivalent phase center to perform azimuth interpolation processing on the echoes that have been processed in the range direction to obtain data that is uniformly distributed in the wavenumber domain; (8) Perform a two-dimensional Fourier transform on the uniformly sampled data in the wavenumber domain to obtain the final imaging result.

2. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The scale transformation factor in step (1) is: α(t)=[cosφ t (t)cosθ t (t)+cosφ r (t)cosθ r (t)]cosφ b +[sinφ t (t)cosθ t (t)+sinφ r (t)cosθ r (t)]sinφ b In the formula, in order to improve the utilization of data, the data is rotated in the wave number domain, and the rotation angle is φ b ,φ t ,θ t is the instantaneous azimuth and elevation angle of the transmitter phase center; φ r ,θ r are the instantaneous azimuth and elevation angles of the receiver phase center; t0 is the time of the synthetic aperture center.

3. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The upward scale factor of the distance in step (2) is: β=max(β1,β2,1) Where β1 is the scale transformation factor calculated at the first pulse emission time, β2 is the scale transformation factor calculated at the last pulse emission time, and f c is the carrier frequency of the transmitted signal, fs is the sampling frequency in the range direction, δ r,1 is the scale transformation factor corresponding to the first pulse, δ r,na is the scaling factor corresponding to the last pulse.

4. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The improved phase factor described in step (3) is: Where τ represents the fast time variable, c is the speed of light, and k is the frequency modulation slope of the linear frequency modulation signal; δ r is the azimuth scale transformation factor, R a It is the sum of the distance from the transmitting antenna phase center to the scene center and the distance from the receiving antenna phase center to the scene center.

5. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The improved matched filter function described in step (4) is: In the formula, f τ is the frequency variable corresponding to the distance fast time variable τ.

6. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The phase factor after the improved matched filtering described in step (5) is:

7. The squint quasi-static bistatic SAR imaging method based on improved PCS as claimed in claim 1, characterized in that: The improved compensation function described in step (6) is:

8. The squint quasi-static bistatic SAR imaging method based on improved PCS according to claim 1, characterized in that: The implementation process of step (7) is as follows: Among them, g(x) is the directional interpolation corresponding to the interpolation point x, sinc(x) is the interpolation kernel, sinc is the sinc function in mathematics, x is the offset in the convolution calculation, and i is the timing in the convolution calculation; The coordinates before and after interpolation are calculated as: Among them, t is the original azimuth time variable, t' is the azimuth time variable after Keystone transformation; R g is the length of the projection of the equivalent phase center position on the xoy plane, v is the common motion speed of the transmitting platform and the receiving platform, θ s is the equivalent oblique angle.

Images