Multi-view InSAR (Interferometric Synthetic Aperture Radar) phase noise analysis method, system and equipment

Abstract

The invention discloses a multi-view InSAR (Interferometric Synthetic Aperture Radar) phase noise analysis method, system and equipment. The multi-view InSAR phase noise analysis method comprises the following steps: acquiring imaging parameters of an InSAR system and single-view complex data of main and auxiliary SAR channels; generating a single-vision interferogram based on the registered single-vision complex data of the main and auxiliary SAR channels, and converting the single-vision interferogram to a frequency domain through Fourier transform so as to obtain single-vision interferogram spectrum information; decomposing frequency spectrum information of the single-view interference pattern into a co-band part with crossed frequencies and an irrelevant part with non-crossed frequencies, and calculating the power spectrum density of a signal relevant part and the power spectrum density of a noise irrelevant part in the single-view interference pattern subjected to flat ground removal according to the co-band part and the irrelevant part; carrying out average filtering on the single-view interference pattern to obtain a multi-view interference pattern, and converting the multi-view interference pattern to a frequency domain through Fourier transform to obtain multi-view interference pattern frequency spectrum information; calculating the signal correlation part power and the noise ircorrelation part power in the multi-view interference pattern spectrum information in the average filter window through integration, and calculating the phase noise caused by geometric decorrelation of the multi-view interference InSAR according to the signal correlation part power and the noise ircorrelation part power; the phase noise analysis method, system and device are used for describing multi-view InSAR phase noise caused by geometric decorrelation in an interferometric SAR imaging system under the conditions of small incidence angles and large processing views.

Description

Technical Field

[0001] The present invention relates to the technical field of synthetic aperture radar interferometry, and particularly to a multi-look InSAR phase noise analysis method, system and device. Background Art

[0002] Synthetic Aperture Radar Interferometry (InSAR) is a technology that uses multiple (two or more) SAR complex image data of the same observation area to perform interference information processing to invert surface elevation or its deformation information. The existence of interference phase noise directly affects the measurement accuracy of the interference system, such as surface elevation accuracy or ground deformation accuracy. Therefore, it is crucial to perform high-precision error analysis on the phase noise of the InSAR system.

[0003] In the analysis of the coherence of the InSAR system and the system phase error, the Cramer-Rao expression is usually used to calculate the standard deviation of the interference pattern phase noise, and this expression depends on the total coherence and the number of independent looks. However, when the phase noise caused by geometric decorrelation dominates, especially in the imaging and interference processing cases with small incidence angles and large number of processing looks, the Cramer-Rao expression cannot accurately describe the spectral characteristics of the phase noise. In the prior art, it is usually assumed that the phase noise is white noise and its spectral shape is the same as the spectral shape of the signal in the range dimension. However, the phase noise caused by geometric decorrelation actually originates from the uncorrelated part of the single-channel radar data frequency band, and its spectral characteristics are different from the signal spectrum, resulting in faster attenuation of the phase noise in multi-look processing. Thus, it is different from the error analysis of the conventional Cramer-Rao expression, affecting the high-precision error analysis in the design of the SAR interference system. Summary of the Invention

[0004] Based on the technical problems existing in the background art, the present invention proposes a multi-look InSAR phase noise analysis method, system and device for describing the multi-look InSAR phase noise caused by geometric decorrelation in the SAR interference imaging system with small incidence angles and large number of processing looks.

[0005] A multi-look InSAR phase noise analysis method proposed by the present invention includes:

[0006] Obtain the imaging parameters of the InSAR system and the single-look complex data of the master and slave SAR channels;

[0007] Generate a single-look interference pattern based on the registered single-look complex data of the master and slave SAR channels, convert the single-look interference pattern to the frequency domain through Fourier transform, and obtain the spectral information of the single-look interference pattern accordingly;

[0008] Decompose the spectral information of the single - look interferogram into a co - band part with intersecting frequencies and an uncorrelated part with non - intersecting frequencies, and calculate the power spectral density of the signal - related part and the power spectral density of the noise - uncorrelated part in the single - look interferogram after removing the flat ground accordingly;

[0009] Perform mean filtering on the single - look interferogram to obtain a multi - look interferogram, and transform the multi - look interferogram to the frequency domain through Fourier transform to obtain the spectral information of the multi - look interferogram accordingly;

[0010] Calculate the power of the signal - related part and the power of the noise - uncorrelated part in the spectral information of the multi - look interferogram within the mean filter window through integration, and calculate the phase noise caused by geometric decorrelation in multi - look InSAR accordingly.

[0011] Furthermore, generate a single - look interferogram based on the registered single - look complex data of the master - slave SAR channels, and transform the single - look interferogram to the frequency domain through Fourier transform to obtain the spectral information of the single - look interferogram. Specifically:

[0012] Assume that the single - look complex data of the master - slave SAR channels after range compression for the \(i\in\{1,2\}\)th channel is denoted as \(y\) i (r):

[0013] y i (r)=[s(r)exp(j2πfr ri r)+n i (r)]*h i (r);

[0014] Generate a single - look interferogram \(z\) u (r):

[0015] z u (r)=y 1 * (r)y 2 (r);

[0016] Transform the single - look interferogram to the frequency domain through Fourier transform to obtain the spectral information \(Z\) of the single - look interferogram u (f r ) is:

[0017] Z u (f r ) = Y q * (-f r )*Y 2 (f r ) = Y 1 (f r )·Y 2 (f r );

[0018] Y i (f r ) is the single-look complex data y of the primary and secondary SAR channels i (r) of the Fourier transform:

[0019] Y i (f r ) = [S(f r +f ri ) + N i (f r )]H i (f r );

[0020] Among them, s(r) is the SAR surface complex reflectivity function, f ri represents the spatial frequency of the local linear phase of the i-th channel, h i (r) is the PTR of the i-th channel, and the PTR is the baseband range point target response of the primary and secondary SAR channels, * represents the convolution operation, the · operator represents the cross-correlation, n i (r) represents the system noise term of the i-th channel, the superscript * represents the complex conjugate, y i (r) takes 1 and 2 for i corresponding to y 1 (r) and y 2 (r), Y i (f r ) takes 1 and 2 for i corresponding to Y 1 (f r ) and Y 2 (f r ), f r is the Fourier domain spatial frequency transformation variable of the slant range r, S(f r ), N i (f r ) and H i (f r ) are the Fourier transforms of s(r), n i (r) and h i (r) respectively, and S(f r +f ri ) is the Fourier transform of the surface complex reflectivity function corresponding to the slant range and the local range difference of the i-th channel.

[0021] Furthermore, the calculation process of the power spectral density of the single-look interferogram is as follows:

[0022] Define the non-zero frequency band as the common band between channels, that is, the overlapping part of the frequency bands;

[0023] Decompose each spectrum H i (f r ) into the common band part Hci (f r ) and the uncorrelated part H of non - overlapping frequencies di (f r ), based on which the power spectral density P of the single - look interferogram after removing the flat ground is calculated u (f r );

[0024]

[0025] Among them, S 0 is the spectral density of the signal - related part, N 0 is the spectral density of the noise - uncorrelated part, the angle brackets <·> represent statistical expectation, Δf r is the linear phase difference, K S (f r - Δf r ) represents the frequency - response function related to the signal, K N (f r - Δf r ) is the frequency - response function related to the noise, δ(f r ) is the impulse - response function, used to describe the intensity concentration characteristic of the signal or noise at a certain set frequency point, P c (f r ) is the power spectral density of the coherent part, which is the contribution of the signal - related part to the power spectral density, P d (f r ) is the power spectral density of the incoherent part, which is the contribution of the noise - uncorrelated part to the power spectral density.

[0026] Furthermore, the calculation formulas for the power spectral density <|Z f (f r )∣ 2 > c of the signal - related part and the power spectral density <|Z f (f r )∣ 2 > uc in the spectral information of the single - look interferogram are as follows:

[0027]

[0028] Among them, P h (f r ) represents the conjugate - symmetric part contributing to the signal in P c (f r ), and P a (f r ) represents the asymmetric part contributing to the noise in P c (f r ).

[0029] Furthermore, the calculation formula for the phase noise caused by geometric decorrelation in multi-look interferometric InSAR is as follows:

[0030]

[0031] where σ φ is the standard deviation of the phase noise, Q N is the power of the uncorrelated part of the noise in the spectral information of the multi-look interferogram, and Q S is the power of the signal-correlated part in the spectral information of the multi-look interferogram, and its calculation formula is as follows:

[0032]

[0033] where G(f r ) is the Fourier transform of the multi-look processing window function of the multi-look interferogram.

[0034] Furthermore, the imaging parameters of the InSAR system and the single-look complex data of the primary and secondary SAR channels are specifically: the interferometric measurement geometric configuration parameters of the InSAR system, the SAR imaging system parameters, the baseband range point target response of the primary and secondary SAR channels, the SAR surface complex reflectivity function, and the channel thermal noise function.

[0035] A multi-look InSAR phase noise analysis system includes an InSAR system parameter acquisition module, a single-look interferogram calculation module, a first power spectral density calculation module, a multi-look interferogram calculation module, a second power calculation module, and a phase noise calculation module;

[0036] The InSAR system parameter acquisition module is used to acquire the imaging parameters of the InSAR system and the single-look complex data of the primary and secondary SAR channels;

[0037] The single-look interferogram calculation module is used to generate a single-look interferogram based on the registered single-look complex data of the primary and secondary SAR channels, convert the single-look interferogram to the frequency domain through Fourier transform, and thereby obtain the spectral information of the single-look interferogram;

[0038] The first power spectral density calculation module is used to decompose the spectral information of the single-look interferogram into a co-band part with intersecting frequencies and an uncorrelated part with non-intersecting frequencies, and thereby calculate the power spectral density of the signal-correlated part and the power spectral density of the noise-uncorrelated part in the single-look interferogram after removing the flat ground;

[0039] The multi-look interferogram calculation module is used to perform average filtering on the single-look interferogram to obtain a multi-look interferogram, convert the multi-look interferogram to the frequency domain through Fourier transform, and thereby obtain the spectral information of the multi-look interferogram;

[0040] The second power calculation module is used to calculate the power of the signal-related part and the power of the noise-uncorrelated part in the multi-look interferogram spectrum information of the average filter window through integration;

[0041] The phase noise calculation module is used to calculate the phase noise caused by geometric decorrelation in multi-look interferometric InSAR.

[0042] Furthermore, the single-look interferogram calculation module is specifically used for:

[0043] Assume that the single-look complex data of the master and slave SAR channels after range compression in the i-th (i ∈ {1, 2}) channel is denoted as y i (r):

[0044] y i (r) = [s(r) exp(j2πf ri r) + n i (r)] * h i (r);

[0045] Generate a single-look interferogram z u (r) based on the registered single-look complex data of the master and slave SAR channels:

[0046] z u (r) = y 1 * (r) y 2 (r);

[0047] Convert the single-look interferogram to the frequency domain through Fourier transform to obtain the single-look interferogram spectrum information Z u (f r ) is:

[0048] Z u (f r ) = Y 1 * (-f r ) * Y 2 (f r ) = Y 1 (f r ) · Y 2 (f r );

[0049] Y i (f r ) is the Fourier transform of the single-look complex data y i (r) of the master and slave SAR channels:

[0050] Y i (f r ) = [S(f r + f ri ) + N i (fr )]H i (f r );

[0051] Among them, s(r) is the SAR surface complex reflectivity function, f ri represents the spatial frequency of the local linear phase of the i-th channel, h i (r) is the PTR of the i-th channel, and the PTR is the baseband range point target response of the primary and secondary SAR channels, * represents the convolution operation, and the · operator represents the cross-correlation, n i (r) represents the system noise term of the i-th channel, Y 1 * (-f r ) with the superscript * representing the complex conjugate, y i (r) with i taking 1 and 2 corresponding to y 1 (r) and y 2 (r) respectively, Y i (f r ) with i taking 1 and 2 corresponding to Y 1 (f r ) and Y 2 (f r ) respectively, f r is the Fourier domain spatial frequency transformation variable of the slant range r, S(f r )、N i (f r ) and H i (f r ) are the Fourier transforms of s(r), n i (r) and h i (r) respectively, and S(f r +f ri ) is the Fourier transform of the surface complex reflectivity function corresponding to the slant range and the local range difference of the i-th channel.

[0052] Furthermore, the obtaining module specifically obtains the following data:

[0053] InSAR system interferometric geometry configuration parameters, SAR imaging system parameters, baseband range point target responses of primary and secondary SAR channels, SAR surface complex reflectivity function, and channel thermal noise function.

[0054] A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the phase noise analysis method as described above.

[0055] The advantages of a multi-look InSAR phase noise analysis method, system, and device provided by the present invention are as follows: By calculating the power spectral density of the signal-correlated part and the noise-uncorrelated part in the spectrum information of the single-look interferogram, and combining the average filtering operation in the multi-look interferogram spectrum, the signal spectrum component and the noise spectrum component in the multi-look interferogram can be integrally calculated, and then the multi-look interferometric SAR phase noise error can be calculated, so as to realize the precision analysis of the decorrelation phase error caused by geometric spectrum offset in the interferometric SAR system with a large number of processing looks, and provide theoretical support for obtaining high-precision surface and sea surface elevation measurements. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic structural diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] Next, the technical solution of the present invention will be described in detail through specific embodiments. Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0058] As Figure 1 shown, a multi-look InSAR phase noise analysis method proposed by the present invention not only considers the system coherence and the effective number of looks in the interferometric processing, but also considers the spectral characteristics of a single channel and the spectral characteristics of the multi-look average window, and can more accurately predict the spectral characteristics of the phase noise, especially in the case of a small incidence angle and a large number of looks in the InSAR interferometric processing system. At the same time, it helps to give a high-precision InSAR system performance evaluation and analysis, filling the gap in the existing analysis that does not comprehensively consider the influence of the interferometric geometry decorrelation factor and the high-precision analysis of the system random phase error under a small incidence angle and a large number of processing looks, and providing an effective way and theoretical support for the parameter design and performance analysis of the subsequent interferometric measurement system. The phase noise analysis method includes steps S1 to S5:

[0059] S1. Obtain the imaging parameters of the InSAR system and the single-look complex data of the master and slave SAR channels;

[0060] The imaging parameters of the InSAR system and the single-look complex data of the master and slave SAR channels are specifically: the interferometric measurement geometric configuration parameters of the InSAR system, the SAR imaging system parameters, the baseband range point target response (PTR) of the master and slave SAR channels, the SAR surface complex reflectivity function (s(r)), and the channel thermal noise function.

[0061] S2. Generate a single - look interferogram based on the registered single - look complex data of the primary and secondary SAR channels, and transform the single - look interferogram to the frequency domain through Fourier transform to obtain the spectral information of the single - look interferogram;

[0062] Assume that the single - look complex data of the primary and secondary SAR channels after range compression for the \(i\in\{1,2\}\) - th channel is denoted as \(y i (r)\):

[0063] y i (r)=[s(r)\(\exp(j2\pi f ri r)+n i (r)]\(*h i (r);(1)

[0064] where \(s(r)\) is the SAR surface complex reflectivity function, \(f ri represents the spatial frequency of the local linear phase of the \(i\) - th channel, \(h i (r)\) is the PTR of the \(i\) - th channel, and the PTR is the base - band range point - target response of the primary and secondary SAR channels, \(*\) represents the convolution operation, \(r\) represents the range - direction coordinate, \(n i (r)\) represents the noise term of the \(i\) - th channel.

[0065] Generate a single - look interferogram \(z u (r)\):

[0066] z u (r)=y 1 * (r)y 2 (r);(2)

[0067] The superscript \(*\) represents complex conjugate. At this time, the wavelength - scale distance difference between the two channels will generate an interference phase. This is captured by the \(f ri terms in each image; the difference \(\Delta f r =f r2 -f r1 is the spatial frequency of the flat - earth interference phase, and this frequency term is equivalent to the well - known spectral shift, which causes geometric decorrelation. Where \(f r1 and \(f r2 are the parameters corresponding to \(i = 1\) and \(i = 2\) in \(f ri , and \(y i (r)\) with \(i = 1\) and \(i = 2\) corresponds to \(y 1 (r)\) and \(y 2 (r)\) respectively.

[0068] Let \(Y i (f r )\) be the single - look complex data of the primary and secondary SAR channels \(yi Fourier transform of (r):

[0069] Y i (f r ) = [S(f r + f ri ) + N i (f r )]H i (f r ); (3)

[0070] where f r is the Fourier domain spatial frequency transformation variable of the slant range r.

[0071] The single-look interferogram is transformed to the frequency domain through Fourier transform to obtain the single-look interferogram spectrum information Z u (f r ) as follows:[[]]

[0072] Z u (f r ) = Y 1 * (-f r ) * Y 2 (f r ) = Y 1 (f r ) · Y 2 (f r ); (4)

[0073] where the · operator represents cross-correlation, and in Y i (f r ), i takes 1 and 2 corresponding to Y 1 (f r ) and Y 2 (f r ) respectively. S(f r ), N i (f r ) and H i (f r ) are the Fourier transforms of s(r), n i (r) and h i (r) respectively. S(f r + f ri ) is the Fourier transform of the surface complex reflectivity function corresponding to the slant range and local distance difference in the i-th channel.

[0074] S3. Decompose the single-look interferogram spectrum information of each channel into a co-band part with intersecting frequencies and an uncorrelated part with non-intersecting frequencies, and calculate the power spectral density of the single-look interferogram after removing the flat ground based on this, and decompose the power spectral density of the signal-related part and the power spectral density of the noise-uncorrelated part in the single-look interferogram spectrum information;

[0075] It should be noted that when \(i\in\{1,2\}\), the spectral requirements in the interference processing are two single-channel spectra \(H\) 1 (f r ) and \(H\) 2 (f r ) having an overlap. Without loss of generality, here a non-zero frequency band \(f\) min \(<f<f\) max is defined as the common band between the channels, that is, the overlapping part of the frequency bands. Each spectrum \(H\) i (f r ) is decomposed into the common band part \(H\) ci (f r ) and the uncorrelated part \(H\) di (f r ) as follows:

[0076]

[0077] where \(\Delta f\) r \(=f\) r2 \(-f\) r1 is the difference in spatial frequencies of the local linear phase, and \(f\) r1 , \(f\) r2 are the spatial frequencies of the local linear phase of the two single channels respectively.

[0078] Among them,

[0079]

[0080] Among them, \(f\) min , \(f\) max are the lower and upper limits of the overlapping part of the frequencies \(f\) of the two channels respectively.

[0081] The plus and minus signs in formulas (6) and (7) are positive when \(i = 1\) and negative when \(i = 2\).

[0082] Therefore, next, the power spectral density \(\langle|Z\) f (f r )|^2\rangle\) of the signal-related part and the power spectral density \(\langle|Z\) 2 \rangle\) of the noise-uncorrelated part in the single-look interferogram spectral information are calculated respectively as follows: c and \(\langle|Z\) f (f r )|^2\rangle\) 2 \rangle\) uc as follows:

[0083] The power spectral density \(P\) u (f r ) of the interferogram without flattening can be expressed as follows:

[0084] \(P\) u (f r) = <|Z u (f r )∣ 2 >= <|Z f (f r +Δf r )∣ 2 >; (8)

[0085] Therefore, it can be derived that the power spectral density of the single-look interferogram after removing the flat ground is as follows:

[0086]

[0087] where S 0 is the spectral density of the signal-related part, S 0 = <|S(f r )| 2 >, N 0 is the spectral density of the noise-uncorrelated part, N 0 = <|N i (f r )| 2 >, the angle brackets <·> represent the statistical expectation, Δf r is the linear phase difference, K S (f r -Δf r ) represents the frequency response function related to the signal, K N (f r -Δf r ) is the frequency response function related to the noise, δ(f r ) is the impulse response function, which is used to describe the intensity concentration characteristic of the signal or noise at a certain set frequency point, P c (f r ) is the coherent part power spectral density, which is the contribution of the signal-related part to the power spectral density, P d (f r ) is the incoherent part power spectral density, which is the contribution of the noise-uncorrelated part to the power spectral density.

[0088] Each component of formula (9) is expressed as follows:

[0089] K S (f r ) = |H 1 (f r )·H 2 (f r )| 2 ; (10)

[0090] K N (f r ) = [|H 1 (f r )|2 ·|H 2 (f r )∣ 2 ;(11)

[0091]

[0092] Among them, H 1 (f r ) and H 2 (f r ) are respectively two single-channel spectra, and S 0 is the spectral density of the signal-related part, and H c1 (f r ) and |H c2 (f r )| are the parameters corresponding to i taking 1 and 2 in H ci (f r ), and H d1 (f r ) and H d2 (f r ) are the parameters corresponding to i taking 1 and 2 in H di (f r ).

[0093] At this time, the spectral density of the signal-related part in the spectral information of the single-look interferogram is as follows:

[0094]

[0095] Among them, P h (f r ) represents the conjugate-symmetric part of the signal contributed in P c (f r ):

[0096]

[0097] And the power spectral density of the noise-uncorrelated part in the spectral information of the single-look interferogram is as follows:

[0098]

[0099] Among them, P a (f r ) represents the asymmetric part of the noise contributed in P c (f r ):

[0100]

[0101] Among them, The superscript * represents the complex conjugate, represents taking the real part of its parameter.

[0102] S4. Perform mean filtering on the single-look interferogram to obtain a multi-look interferogram, and transform the multi-look interferogram to the frequency domain through Fourier transform to obtain the spectral information of the multi-look interferogram.

[0103] Multi-look interferogram z m (r) is generally obtained by performing mean filtering on the single-look interferogram z f (r):

[0104] z m (r) = z f (r) * g(r); (18)

[0105] where g(r) is the mean filter, and * represents the convolution operation. In the frequency domain, its spectrum is:

[0106] Z m (f r ) = Z f (f r )G(f r ); (19)

[0107] where Z m (f r ) and G(f r ) are the Fourier transforms of z m (r) and g(r) respectively. Therefore, the spectral information of the multi-look interferogram is:

[0108] <|Z m (f r )| 2 >= <|Z f (f r )| 2 >G 2 (f r ). (20)

[0109] S5. Calculate the power of the signal-related part and the power of the noise-uncorrelated part in the spectral information of the multi-look interferogram within the mean filter window through integration, and calculate the phase noise caused by geometric decorrelation in multi-look InSAR accordingly.

[0110] Based on the derivation of the power spectral density of the signal-related part and the noise-uncorrelated part in the multi-look interference spectrum, calculate the power Q of the signal-related part in the spectral information of the multi-look interferogram within the mean filter window through integration S and the power Q of the noise-uncorrelated part in the spectral information of the multi-look interferogram N :

[0111]

[0112] where G(f r) is the Fourier transform of the multi-look processing window function of the multi-look interferogram.

[0113] According to the standard deviation formula of phase noise, the phase noise caused by geometric decorrelation in multi-look interferometric InSAR is calculated as follows:

[0114]

[0115] Formula (23) is the general form of the interferometric phase noise. When the PTR spectrum is rectangular, the above formula (23) can be simplified. For example, when the spectrum G(f r ) of the average filter can be approximated as a rectangular function centered at f r = 0, the simplification in this special case is as follows:

[0116]

[0117] Among them, is the single-channel signal-to-noise ratio, B g is the spatial bandwidth of the filter, B r is the range bandwidth of the SAR imaging system, N eff is the number of effective looks, which is expressed as the integral of the spectra H 1 (f r ) and H 2 (f r ), and N eff is expressed as follows:

[0118]

[0119] Among them, The superscript * of indicates the complex conjugate.

[0120] Through steps S1 to S5, the phase noise analysis method proposed in this embodiment can calculate the power spectral densities of the signal-related part and the noise-unrelated part in the spectral information of the single-look interferogram, and combine the average filtering operation in the multi-look interferogram spectrum to integrally calculate the signal spectral power component and the noise spectral power component in the multi-look interferogram, and then calculate and obtain the multi-look interferometric SAR phase noise error, thereby realizing the precision analysis of the decorrelation phase error caused by geometric spectral shift in the interferometric SAR system with a large number of processing looks, and providing theoretical support for obtaining high-precision surface and sea surface elevation measurements.

[0121] Therefore, in this embodiment, the geometric decorrelation multi-look phase error analysis design of the InSAR system can be carried out according to the actual system imaging parameters and interference processing parameters, so as to obtain a high-precision analysis of the random phase error of the interference system under the current design parameters. If the interference phase accuracy does not meet the design requirements, the system parameters can be adjusted on the premise of radar system performance compromise, and then the system interference performance calculation and analysis are carried out again until the given InSAR system design requirements are met, thereby reducing the experimental cost.

[0122] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A multi-look InSAR phase noise analysis method, characterized in that: include: Obtain InSAR system imaging parameters and single-view complex data of main and auxiliary SAR channels; A single-view interferogram is generated based on the registered single-view complex data of the primary and secondary SAR channels, and the single-view interferogram is converted into the frequency domain through Fourier transform to obtain the spectrum information of the single-view interferogram; Decomposing the frequency spectrum information of the single-look interferogram into a common band part with intersecting frequencies and an irrelevant part with non-intersecting frequencies, and calculating the power spectrum density of the signal-related part and the power spectrum density of the noise-irrelevant part in the single-look interferogram after removing the flat ground; The single-view interferogram is averaged and filtered to obtain a multi-view interferogram, and the multi-view interferogram is converted into the frequency domain by Fourier transform to obtain the spectrum information of the multi-view interferogram; The signal-correlated power and the noise-irrelevant power in the spectrum information of the multi-look interferogram within the average filter window are calculated by integration, and the phase noise caused by geometric decorrelation in the multi-look interferometric InSAR is calculated accordingly.

2. The multi-look InSAR phase noise analysis method according to claim 1, characterized in that: A single-view interferogram is generated based on the single-view complex data of the primary and secondary SAR channels after registration, and the single-view interferogram is converted to the frequency domain through Fourier transform to obtain the spectrum information of the single-view interferogram, specifically: Assume that the single-view complex data of the main and auxiliary SAR channels of the i∈{1,2}th channel after range compression is recorded as y i (r): y i (r)=[s(r)exp(j2πf ri r)+n i (r)]*h i (r); Generate single-look interferogram z based on the registered single-look complex data of primary and secondary SAR channels u (r): z u (r)=y1 * (r)y2(r); The single-view interferogram is converted to the frequency domain by Fourier transform to obtain the single-view interferogram spectrum information Z u (f r )for: Z u (f r )=Y1 * (-f r )*Y2(f r )=Y1(f r )·Y2(f r ); Y i (f r ) is the single-view complex data of the main and auxiliary SAR channels y i The Fourier transform of (r) is: Y i (f r )=[S(f r +f ri )+N i (f r )]H i (f r ); Where s(r) is the SAR surface complex reflectivity function, f ri represents the spatial frequency of the local linear phase of the ith channel, h i (r) is the PTR of the ith channel, and the PTR is the baseband range point target response of the primary and secondary SAR channels, * indicates the convolution operation, the · operator indicates the cross-correlation, and n i (r) represents the system noise term of the ith channel, the superscript * represents the complex conjugate, y i (r) where i is 1 and 2, corresponding to y1(r) and y2(r), respectively. i (f r ) where i is 1 and 2, corresponding to Y1(f r ) and Y2(f r ), f r is the Fourier domain spatial frequency transform variable of the slant range r, S(f r ), N i (f r ) and H i (f r ) are s(r), n i (r) and h i The Fourier transform of (r), S(f r +f ri ) is the Fourier transform of the surface complex reflectivity function corresponding to the slant range and local range difference of the i-th channel.

3. The multi-look InSAR phase noise analysis method according to claim 1, characterized in that: The power spectral density of the single-view interferogram is calculated as follows: Define the non-zero frequency band to represent the common band between channels, i.e., the overlapping part of the frequency band; Each spectrum H i (f r ) is decomposed into the common band part H ci (f r ) and the irrelevant part of the disjoint frequencies H di (f r ), and then calculate the power spectrum density P of the single-view interferogram after removing the flat ground u (f r ); Where S0 is the spectral density of the signal-correlated part, N0 is the spectral density of the noise-irrelevant part, and the angle brackets <·> represent the statistical expectation. Δf r is the linear phase difference, K S (f r -Δf r ) represents the frequency response function associated with the signal, K N (f r -Δf r ) is the frequency response function related to noise, δ(f r ) is the impulse response function, which is used to describe the intensity concentration characteristics of the signal or noise at a certain set frequency point. c (f r ) is the power spectral density of the coherent part, which is the contribution of the signal-related part to the power spectral density. d (f r ) is the power spectral density of the incoherent part, which is the contribution of the irrelevant part of the noise to the power spectral density.

4. The multi-look InSAR phase noise analysis method according to claim 3, characterized in that: The power spectral density of the signal-related part in the single-view interference pattern spectrum information <|Z f (f r )∣ 2 > c The power spectral density of the uncorrelated part of the noise <|Z f (f r )∣ 2 > uc The calculation formula is as follows: Among them, P h (f r ) indicates P c (f r ) is the conjugate symmetric part of the contribution signal, P a (f r ) indicates P c (f r ) contributes to the asymmetric part of the noise.

5. The multi-look InSAR phase noise analysis method according to claim 1, characterized in that: The calculation formula of the phase noise caused by geometric decorrelation in multi-look interferometric InSAR is as follows: Among them, σ φ is the standard deviation of phase noise, Q N is the power of the noise-irrelevant part of the spectrum information of the multi-view interferogram, Q S is the power of the signal-related part in the multi-view interference graph spectrum information, and its calculation formula is as follows: Among them, G(f r ) is the Fourier transform of the multi-look processing window function of the multi-look interferogram.

6. The multi-look InSAR phase noise analysis method according to claim 1, characterized in that: The imaging parameters of the InSAR system and the single-view complex data of the main and auxiliary SAR channels are specifically: InSAR system interferometric measurement geometric configuration parameters, SAR imaging system parameters, baseband range point target response of the main and auxiliary SAR channels, SAR surface complex reflectivity function and channel thermal noise function.

7. A multi-look InSAR phase noise analysis system, characterized in that: It includes an InSAR system parameter acquisition module, a single-view interferogram calculation module, a first power spectrum density calculation module, a multi-view interferogram calculation module, a second power calculation module and a phase noise calculation module; The InSAR system parameter acquisition module is used to acquire the InSAR system imaging parameters and the single-view complex data of the main and auxiliary SAR channels; The single-view interferogram calculation module is used to generate a single-view interferogram based on the registered primary and secondary SAR channel single-view complex data, and convert the single-view interferogram into the frequency domain through Fourier transform to obtain the single-view interferogram spectrum information; The first power spectrum density is used for the calculation module to decompose the frequency spectrum information of the single-view interferogram into a common band part with intersecting frequencies and an irrelevant part with non-intersecting frequencies, so as to calculate the power spectrum density of the signal-related part and the power spectrum density of the noise-irrelevant part in the single-view interferogram after removing the flat ground; The multi-view interference graph calculation module is used to perform average filtering on the single-view interference graph to obtain a multi-view interference graph, and convert the multi-view interference graph into a frequency domain through Fourier transform to obtain frequency spectrum information of the multi-view interference graph; The second power calculation module is used to calculate the signal-related part power and the noise-irrelevant part power in the multi-view interference pattern spectrum information within the average filter window by integration; The phase noise calculation module is used to calculate the phase noise caused by geometric decorrelation in multi-look interferometric InSAR.

8. The multi-look InSAR phase noise analysis system according to claim 7, characterized in that: The single-view interference pattern calculation module is specifically used for: Assume that the single-view complex data of the main and auxiliary SAR channels of the i∈{1,2}th channel after range compression is recorded as y i (r): y i (r)=[s(r)exp(j2πf ri r)+n i (r)]*h i (r); Generate single-look interferogram z based on the registered single-look complex data of primary and secondary SAR channels u (r): z u (r)=y1 * (r)y2(r); The single-view interferogram is converted to the frequency domain by Fourier transform to obtain the single-view interferogram spectrum information Z u (f r )for: Z u (f r )=Y1 * (-f r )*Y2(f r )=Y1(f r )·Y2(f r ); Y i (f r ) is the single-view complex data of the main and auxiliary SAR channels y i The Fourier transform of (r) is: Y i (f r )=[S(f r +f ri )+N i (f r )]H i (f r ); Where s(r) is the SAR surface complex reflectivity function, f ri represents the spatial frequency of the local linear phase of the ith channel, h i (r) is the PTR of the ith channel, and the PTR is the baseband range point target response of the primary and secondary SAR channels, * indicates the convolution operation, the · operator indicates the cross-correlation, and n i (r) represents the system noise term of the i-th channel, Y1 * (-f r ) indicates complex conjugation, y i (r) where i is 1 and 2, corresponding to y1(r) and y2(r), respectively. i (f r ) where i is 1 and 2, corresponding to Y1(f r ) and Y2(f r ), f r is the Fourier domain spatial frequency transform variable of the slant range r, S(f r ), N i (f r ) and H i (f r ) are s(r), n i (r) and h i The Fourier transform of (r), S(f r +f ri ) is the Fourier transform of the surface complex reflectivity function corresponding to the slant range and local range difference of the i-th channel.

9. The multi-look InSAR phase noise analysis system according to claim 7, characterized in that: The acquisition module specifically obtains the following data: InSAR system interferometry geometry parameters, SAR imaging system parameters, baseband range point target response of primary and auxiliary SAR channels, SAR surface complex reflectivity function and channel thermal noise function.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the phase noise analysis method according to any one of claims 1 to 6 is implemented.

Images