Interference Signal Detection and Suppression Method Applicable to Azimuth Multi-Channel SAR System

Abstract

The present invention provides a method and system for interference signal detection and suppression in an SAR system. It includes: (1) performing range Fourier transform on the original data; (2) selecting two receiving channels with the farthest distance as interference detection channels; (3) calculating the interference detection threshold of the coherence coefficient; (4) using the interference detection channels to calculate the coherence coefficient along azimuth samples to achieve time-stationary narrowband interference; (5) using the interference detection channels to calculate the coherence coefficient along range samples to achieve time-varying wideband interference detection; (6) fusing the interference mask to output the interference detection result. The present invention solves the problem of interference detection in azimuth multi-channel high-resolution broadband systems and has an important supporting role in the imaging processing of low-frequency multi-channel SAR systems.

Description

Technical Field

[0001] The present invention belongs to the technical field of signal and information processing. Specifically, it relates to a method and system for detecting and suppressing interference signals applicable to an azimuth multi-channel SAR system. Background Technique

[0002] Spaceborne synthetic aperture radar (SAR) has been widely used in military and civilian fields due to its all-weather, all-day, high-resolution, and wide-swath earth observation capabilities. Currently, it has become one of the remote sensing detection means widely studied and focused on by countries around the world. As an important part of scientific big data, earth remote sensing observation scientific data has become a national basic and strategic resource. However, with the development of wireless networks and systems, radio frequency interference (RFI) has become a major problem in SAR imaging, especially in low-frequency bands such as the P and L bands, which pose a major obstacle to the acquisition of raw data, imaging processing, and subsequent interpretation processing. Therefore, the problem of radio frequency interference detection and suppression is one of the key technologies in current ground data processing.

[0003] Terrestrial radiation source radio frequency interference will have a negative impact on spaceborne SAR data, specifically reflected in: (1) Affecting the echo signal acquisition process and amplitude dynamic range, and it is extremely easy to cause missed detection and false alarms of targets. The existence of strong energy interference will significantly reduce the echo signal-to-interference-plus-noise ratio, weaken the detection ability of low-observable targets in complex scenes, easily cause missed detection of targets, and even increase the false alarm probability of targets. (2) Affecting the coherent imaging focusing process, resulting in blurred and defocused images or the appearance of covering artifacts. Interference will cause deviations in the estimation of imaging parameters, leading to mismatches in matched filtering, inaccurate compensation of envelope phase errors, and defocusing of imaging results. In addition, interference appears as covering artifacts with specific texture features in the image, resulting in distortion of the spatial and radiation response information of real targets, and greatly affecting the reliability of qualitative interpretation applications such as land cover classification, change detection, and target recognition. (3) Reducing the quality of image-derived products and affecting the accuracy of parameter extraction and inversion in subsequent quantitative remote sensing applications. The amplitude and phase distortion errors caused by interference will affect the accuracy of derived parameters such as polarization correlation and interferometric phase diagrams, restricting the accuracy of remote sensing applications such as high-precision deformation observation and refined quantitative parameter inversion. If the interfered data is directly discarded, it will not only cause a great waste of scientific data resources but also delay the emergency response and rapid observation and analysis tasks for key regions and targets.

[0004] Patent document CN113064122B proposes a performance evaluation method, system and medium for a P-band SAR interference suppression algorithm. This patent document proposes an evaluation method and system for the P-band SAR interference suppression algorithm, but does not conduct a detailed analysis of the applicability to multi-channel systems; Patent document CN110221256B proposes a SAR interference suppression method based on a deep residual network. This patent document proposes using a deep convolutional neural network to detect whether there is interference in the original echo and perform suppression, but does not analyze the suppression effect on multi-channel systems; Patent document CN105974376B proposes a SAR radio frequency interference suppression method. This patent document proposes a radio frequency interference suppression method using subspace projection, but does not consider the multi-channel system; Patent document CN108318865A provides a multi-channel SAR deception interference identification and adaptive suppression method. This patent document uses the spatio-temporal characteristic differences between interference and SAR echoes for adaptive filtering processing, but this requires the system to have sufficient redundant channels. Patent document CN111239697A proposes a multi-dimensional domain joint SAR wideband interference suppression method based on low-rank matrix decomposition. This patent document proposes using the idea of vectorizing the short-time Fourier transform matrix to separate the interference signal and the SAR echo signal, which can effectively avoid the loss of the echo signal caused by interference suppression, but this method does not consider multi-channel applications. The article "Narrowband RFI Suppression on High-Resolution Wide-Swath SAR Systems via Low-Rank Recovery", 2022 IEEE International Geoscience and Remote Sensing Symposium, pp.5137-5140, 2022 proposes a narrowband RFI mitigation scheme for HRWS SAR systems, but it is not applicable to wideband interference signals. For a conventional spaceborne azimuth multi-channel high-resolution wide-swath SAR system without special design, there is currently no effective interference detection method applicable to all scenarios. Summary of the Invention

[0005] Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide a method and system for detecting and suppressing interference signals applicable to azimuth multi-channel SAR systems.

[0006] According to a method for detecting and suppressing interference signals applicable to azimuth multi-channel SAR systems provided by the present invention, it includes:

[0007] Step S1, performing range-Doppler transform on the original data;

[0008] Step S2, selecting an interference detection channel;

[0009] Step S3: Calculate the coherence coefficient interference detection threshold according to the interference detection channel.

[0010] Step S4: Perform time-stationary narrowband interference detection according to the coherence coefficient interference detection threshold, and generate a time-stationary narrowband interference mask; perform time-varying wideband interference detection according to the coherence coefficient interference detection threshold, and generate a time-varying wideband interference mask.

[0011] Step S5: Fuse the time-stationary narrowband interference mask and the time-varying wideband interference mask respectively for interference mask fusion, and output the interference detection result to obtain an image after interference signal suppression.

[0012] Preferably, in step S1, the azimuth multi-channel SAR data is transformed from the two-dimensional time domain to the range-frequency - azimuth-time domain; during transmission, the antenna transmits with the full azimuth aperture, and during reception, m sub-apertures of the antenna receive simultaneously. When there is RFI, the SAR echo received by the m-th channel in the range-frequency - azimuth-time domain is expressed as:

[0013] y m (η,f) = s m (η,f) + x m (η,f) + n m (η,f)

[0014] where η is the azimuth time, f is the range frequency, s m is the useful signal of the m-th channel, x m is the interference signal of the m-th channel, n m is the system noise of the m-th channel, y m is the SAR echo received by the m-th channel, and the subscript m represents the m-th channel.

[0015] The interference signal received by the m-th channel is expressed as the time delay of the signal received by the first channel x m :

[0016]

[0017] In the formula, θ RFI represents the angle of the interference signal, C is the speed of light, x1 is the interference signal received by the first channel; d represents the distance between the phase centers of the two-channel antennas.

[0018] Preferably, in step S2, two receiving channels with a theoretical coherence coefficient of zero are used as the interference detection channels, and two receiving channels with the farthest phase centers apart are selected; the calculation process of the theoretical coherence coefficient is as follows:

[0019] The spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel is expressed as

[0020]

[0021] γ 1,m represents the spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel, E{·} represents the mathematical expectation, y1 represents the SAR echo received by the first channel, the symbol * represents the conjugate operation. Since s m 、x m 、n m are independently distributed, the subscript m represents the m-th channel, and the correlation coefficient is simplified to:

[0022]

[0023] Among them, is the cross-correlation function of the useful signals s1 and s m ; is the interference phase of the useful signals s1 and s m ;

[0024]

[0025] Among them, P(f a ) is the Doppler power spectral density function, f a is the Doppler frequency; Δη m is the time delay between the echo received by the m-th channel and the first channel;

[0026] If rectangular weighting is used in the azimuth direction, then is expressed as

[0027]

[0028] Among them v s is the satellite velocity.

[0029]

[0030] Among them, c is the speed of light, E() represents the expectation;

[0031] Two channels with zero coherence are used for interference detection, and the first channel and the two channels farthest from it are selected.

[0032] Preferably, in the step S3, it is assumed that the echo received by the m-th channel is completely decorrelated from the first channel; the coherence coefficient is estimated using the following formula

[0033]

[0034] In the formula, is the estimated value of the coherence coefficient, N is the total number of samples, N represents the nth estimated sample, y1(n), y m (n) represent the echo signals of the first channel and the mth channel respectively, the superscript * represents the conjugate operation, and the probability density function pdf() of the estimated value of the coherence coefficient is shown as follows:

[0035]

[0036] In the formula, 2F1 is the hypergeometric function, is the estimated value of the coherence coefficient, γ is the coherence coefficient, and L is the number of independently and identically distributed samples.

[0037] Preferably, in the step S4, the coherence coefficient between the mth channel and the first channel is calculated pixel by pixel using the azimuth upward samples:

[0038]

[0039] is the estimated value of the coherence coefficient between the mth channel and the first channel, f l is the lth range frequency unit, η k is the kth azimuth time unit;

[0040] Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det , it is determined that there is a time-stationary narrowband interference in the current unit; otherwise, it is determined that there is no interference; for the detected time-stationary narrowband interference, perform pixel-by-pixel detection of the time-stationary narrowband interference to generate a time-stationary narrowband interference detection mask map;

[0041] In order to detect time-varying broadband interference, the coherence coefficient between the mth channel and the first channel is calculated pixel by pixel using the range upward samples.

[0042]

[0043] represents the coherence coefficient between the mth channel and the first channel, l represents the lth unit, and Δθ represents the direction of arrival of the interference wave.

[0044] Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ detWhen it is determined that there is time-varying wideband interference in the current unit; otherwise, it is determined that there is no interference. The time-varying wideband interference is detected pixel by pixel to generate a time-varying wideband interference detection mask image.

[0045] An interference signal detection and suppression system applicable to an azimuth multi-channel SAR system provided by the present invention includes:

[0046] Module M1, performing range Fourier transform on the original data;

[0047] Module M2, selecting an interference detection channel;

[0048] Module M3, calculating a coherence coefficient interference detection threshold according to the interference detection channel;

[0049] Module M4, performing time-stationary narrowband interference detection according to the coherence coefficient interference detection threshold and generating a time-stationary narrowband interference mask; performing time-varying wideband interference detection according to the coherence coefficient interference detection threshold and generating a time-varying wideband interference mask;

[0050] Module M5, respectively fusing the time-stationary narrowband interference mask and the time-varying wideband interference mask, outputting an interference detection result, and obtaining an image after interference signal suppression.

[0051] Preferably, in the module M1, the azimuth multi-channel SAR data is transformed from the two-dimensional time domain to the range frequency-azimuth time domain; when transmitting, the antenna emits with the full azimuth aperture, and when receiving, m sub-apertures of the antenna receive simultaneously. When there is RFI, the SAR echo received by the m-th channel is expressed in the range frequency-azimuth time domain as:

[0052] y m (η,f) = s m (η,f) + x m (η,f) + n m (η,f)

[0053] Where η is the azimuth time, f is the range frequency, s m is the useful signal of the m-th channel, x m is the interference signal of the m-th channel, n m is the system noise of the m-th channel, y m is the SAR echo received by the m-th channel, and the subscript m represents the m-th channel;

[0054] The interference signal received by the m-th channel is expressed as the time delay of the signal received by the first channel x m :

[0055]

[0056] In the formula, θ RFI represents the angle of the interference signal, C is the speed of light, x1 is the interference signal received by the first channel; d represents the distance between the phase centers of the two-channel antennas.

[0057] Preferably, in the module M2, two receiving channels with a theoretical coherence coefficient of zero are used as interference detection channels, and two receiving channels with the farthest phase centers are selected; the calculation process of the theoretical coherence coefficient is as follows:

[0058] The spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel is expressed as

[0059]

[0060] γ 1,m represents the spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel, E{·} represents the mathematical expectation, y1 represents the SAR echo received by the first channel, the symbol * represents the conjugate operation, since s m , x m , n m are independently distributed, the subscript m represents the m-th channel, and the correlation coefficient is simplified to:

[0061]

[0062] Among them, is the cross-correlation function of the useful signals s1 and s m , is the interference phase of the useful signals s1 and s m ;

[0063]

[0064] Among them, P(fa) is the Doppler power spectral density function, f a is the Doppler frequency; Δη m is the time delay between the echo received by the m-th channel and the first channel;

[0065] If rectangular weighting is used in the azimuth direction, then is expressed as

[0066]

[0067] Among them v s is the satellite speed.

[0068]

[0069] Among them, c is the speed of light, E() represents the expectation;

[0070] Interference detection is performed using two channels with zero coherence, and the first channel and the two channels farthest from it are selected.

[0071] Preferably, in the module M3, it is assumed that the echo received by the m-th channel is completely decorrelated from the first channel; the coherence coefficient is estimated using the following formula

[0072]

[0073] where is the estimated value of the coherence coefficient, N is the total number of samples, N represents the n-th estimated sample, y1(n), y m (n) represent the echo signals of the first channel and the m-th channel respectively, the superscript * represents the conjugate operation, and the probability density function pdf() of the estimated value of the coherence coefficient is shown as follows

[0074]

[0075] where 2F1 is the hypergeometric function is the estimated value of the coherence coefficient, γ is the coherence coefficient, and L is the number of independently and identically distributed samples.

[0076] Preferably, in the module M4, the coherence coefficient between the m-th channel and the first channel is calculated pixel by pixel using the azimuth upward samples:

[0077]

[0078] is the estimated value of the coherence coefficient between the m-th channel and the first channel, f l is the l-th range frequency unit, η k is the k-th azimuth time unit;

[0079] The calculated coherence coefficient is compared with the interference detection threshold γ det , and when the calculated coherence coefficient is greater than γ det , it is determined that there is time-stationary narrowband interference in the current unit; otherwise, it is determined that there is no interference; for those detected to have time-stationary narrowband interference, time-stationary narrowband interference is detected pixel by pixel to generate a time-stationary narrowband interference detection mask image;

[0080] In order to detect time-varying wideband interference, the coherence coefficient between the m-th channel and the first channel is calculated pixel by pixel using the range upward samples.

[0081]

[0082] Denote the coherence coefficient between the m-th channel and the 1st channel, l denote the l-th unit, and Δθ denote the direction of arrival of the interference wave.

[0083] Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det , it is determined that there is time-varying wideband interference in the current unit; otherwise, it is determined that there is no interference. Detect time-varying wideband interference pixel by pixel to generate a time-varying wideband interference detection mask image.

[0084] Compared with the prior art, the present invention has the following beneficial effects:

[0085] 1. The present invention solves the interference detection problem of azimuth multi-channel high-resolution wideband systems and plays an important supporting role in the imaging processing of low-frequency multi-channel SAR systems.

[0086] 2. The present invention utilizes the spatial decorrelation between the original echoes of the receiving channels and the high coherence between the interferences, and detects the interference by calculating the correlation between two channels, improving the possibility of interference detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] By reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings, other features, objects, and advantages of the present invention will become more apparent:

[0088] Figure 1 It is a schematic diagram of the interference detection processing steps for the SAR system.

[0089] Figure 2 It is a comparison graph of the variation curves of the signal correlation coefficient with respect to the signal-to-interference ratio and the interference phase,

[0090] Figure 3 It is a comparison graph of the SAR images before and after interference detection and suppression, (a) before interference suppression, (b) after interference suppression. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0091] The present invention will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those of ordinary skill in the art can make several changes and improvements without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0092] According to an interference signal detection and suppression method applicable to an azimuth multi-channel SAR system provided by the present invention, it includes the following steps:

[0093] Step 1: Perform range Fourier transform on the original data;

[0094] Step 2: Select the interference detection channel;

[0095] Step 3: Calculate the coherence coefficient interference detection threshold;

[0096] Step 4: Perform time-stationary narrowband interference detection according to the calculation result of the coherence coefficient interference detection threshold, and generate a time-stationary narrowband interference mask;

[0097] Step 5: Perform time-varying wideband interference detection according to the calculation result of the coherence coefficient interference detection threshold, and generate a time-varying wideband interference mask;

[0098] Step 6: Fuse the obtained time-stationary narrowband interference mask and time-varying wideband interference mask respectively, and output the interference detection result to obtain an image after interference signal suppression.

[0099] The specific implementation steps of the present invention are as Figure 1 shown, specifically including:

[0100] Step 1 specifically includes the following content:

[0101] Perform range Fourier transform on the original data to transform the azimuth multi-channel SAR data from two-dimensional time domain to range frequency-azimuth time domain;

[0102] Based on the principle that spaceborne SAR usually adopts azimuth multi-channel system to achieve high-resolution wide-swath SAR imaging. When transmitting, the antenna emits with full aperture in the azimuth direction, and when receiving, m sub-apertures of the antenna receive simultaneously in the azimuth direction. When there is RFI, the SAR echo received by the mth channel can be expressed in the range frequency-azimuth time domain as:

[0103] y m (η,f)=s m (η,f)+x m (η,f)+n m (η,f)

[0104] where η is the azimuth time, f is the range frequency, s m is the useful signal of the mth channel, x m is the interference signal of the mth channel, n m is the system noise of the mth channel, y m is the SAR echo received by the mth channel, and the subscript m represents the mth channel.

[0105] The interference signal received by the mth channel can be expressed as the time delay of the signal received by the first channel x m :

[0106]

[0107] In the formula, θ RFI represents the angle of the interference signal, C is the speed of light, x1 is the interference signal received by the first channel. d represents the distance between the phase centers of the two-channel antennas.

[0108] Step 2 includes the following content:

[0109] Select the interference detection channels. Use two receiving channels with a theoretical coherence coefficient of zero as the interference detection channels, and select two receiving channels with the farthest phase centers.

[0110] The calculation process of the theoretical coherence coefficient is as follows:

[0111] The spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel can be expressed as

[0112]

[0113] γ 1,m represents the spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel, E{·} represents the mathematical expectation, y1 represents the SAR echo received by the first channel, the symbol * represents the conjugate operation, since s m , x m , n m are independently distributed, the subscript m represents the m-th channel, and the correlation coefficient can be simplified to

[0114]

[0115] Among them, is the cross-correlation function of the useful signals s1 and s m , is the interference phase of the useful signals s1 and s m .

[0116]

[0117] Among them, P(fa) is the Doppler power spectral density function, f a is the Doppler frequency; Δη m is the time delay between the echo received by the m-th channel and the first channel.

[0118] If rectangular weighting is used in the azimuth direction, then can be expressed as

[0119]

[0120] Among them v s is the satellite speed.

[0121]

[0122] Among them, c is the speed of light, and E() represents the expectation.

[0123] The coherence coefficient between channels can be recommended not to be expressed as:

[0124]

[0125] where SIR m is the signal-to-interference ratio of the m-th channel, and SINR m is the signal-to-interference-plus-noise ratio of the m-th channel.

[0126] SIR = E{|s m | 2} / E{|x m | 2}

[0127] SINR = E{|s m | 2} / (E{x m 2}+E{|n m | 2})

[0128] Assuming that each receiving channel has the same SIR and SINR, then

[0129]

[0130] Here, the influence of interference on the coherence of the original echo is simulated when the signal-to-noise ratio is 10 dB and the signal correlation coefficients are 0.6, 0.25, and 0, respectively. In these three cases, the coherence between the two receiving channels without RFI is 0.545, 0.23, and 0, respectively. The mutual correlation coefficient between the useful signal and RFI changes with the SIR and differential interference phase as Figure 2 shown. It can be seen from Figure 2 that:

[0131] When , in most cases, the coherence with RFI is higher than that without RFI, corresponding to the white area in Figure 2 (b). At the same time, in some cases, the coherence with RFI is lower than that without RFI, corresponding to the black area in Figure 2 (b). This increases the difficulty of setting the threshold for detecting and mitigating RFI.

[0132] When , the white area with larger RFI coherence is improved, as shown in Figure 2(as shown in (c)).

[0133] When the coherence is only determined by the SIR, and the coherence is greater than zero in the presence of RFI.

[0134] Therefore, for an azimuth multi-channel high-resolution wide-swath SAR satellite system, two channels with zero coherence can be used for interference detection, and the first channel and the two channels farthest from it are selected.

[0135] Step three specifically includes:

[0136] Calculate the interference detection threshold of the coherence coefficient.

[0137] The calculation process is as follows: Assume that the echo received by the m-th channel is completely decorrelated from the first channel. The coherence coefficient is estimated using the following formula

[0138]

[0139] where is the estimated value of the coherence coefficient, N is the total number of samples, N represents the n-th estimated sample, y1(n) and y m (n) represent the echo signals of the first channel and the m-th channel respectively, and the superscript * represents the conjugate operation. The probability density function pdf() of the estimated value of the coherence coefficient is shown as follows

[0140]

[0141] where 2F1 is the hypergeometric function is the estimated value of the coherence coefficient, γ is the coherence coefficient, L is the number of independently and identically distributed samples. Taking L = 50 and a false alarm probability of 0.5% as an example, the interference detection threshold of the coherence coefficient is 0.32.

[0142] Step four, time-stationary narrowband interference detection

[0143] To detect time-stationary narrowband interference (TSNB RFI), the coherence coefficient between the m-th channel and the first channel is calculated pixel by pixel using the azimuth upward samples.

[0144]

[0145] is the estimated value of the coherence coefficient between the m-th channel and the first channel, f l is the l-th range frequency unit, η k is the k-th azimuth time unit.

[0146] Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient Greater than γ det When it is greater than γ, it is determined that there is time-stationary narrowband interference in the current cell; otherwise, it is determined that there is no interference. For those detected to have time-stationary narrowband interference, pixel-by-pixel detection of time-stationary narrowband interference is performed to generate a time-stationary narrowband interference detection mask image.

[0147] Step Five, time-varying wideband interference detection:

[0148] To detect time-varying wideband interference (TVWB RFI), the coherence coefficient between the m-th channel and the 1st channel is calculated pixel-by-pixel using the distance to the upward samples.

[0149]

[0150] represents the coherence coefficient between the m-th channel and the 1st channel, l represents the l-th unit, and Δθ represents the direction of arrival of the interference wave.

[0151] Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det it is determined that there is time-varying wideband interference in the current cell; otherwise, it is determined that there is no interference. Pixel-by-pixel detection of time-varying wideband interference is performed to generate a time-varying wideband interference detection mask image.

[0152] The specific content of Step Six is as follows:

[0153] Fuse the time-stationary narrowband interference detection mask image and the time-varying wideband interference detection mask image for interference mask fusion, and output the interference detection result to obtain the image after interference suppression:

[0154] Fuse the time-varying wideband interference detection mask image and the time-varying wideband interference detection mask image, such as using the Hough transform method, etc., to generate the final interference detection result.

[0155] Here, the method provided by the present invention is verified using azimuth multi-channel high-resolution wide-swath SAR measured data superimposed with simulated interference data. The SAR system is a five-channel airborne SAR system, and the time-domain interference-to-signal ratio of the added narrowband signal is -40 dB, and the time-domain interference-to-signal ratio of the wideband signal is -5 dB. Figure 3 SAR images before and after interference suppression are given. As Figure 3 can be seen, the interference is effectively suppressed after being processed by the method proposed by the present invention.

[0156] The present invention also provides an interference signal detection and suppression system for an SAR system. The interference signal detection and suppression system for the SAR system can be implemented by executing the process steps of the interference signal detection and suppression method for the SAR system. That is, those skilled in the art can understand the interference signal detection and suppression method for the SAR system as a preferred embodiment of the interference signal detection and suppression system for the SAR system.

[0157] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for detecting and suppressing interference signals applicable to an azimuth multi-channel SAR system, characterized in that Including: Step S1, performing range-direction Fourier transform on the original data; Step S2, selecting an interference detection channel; Step S3, calculating a coherence coefficient interference detection threshold according to the interference detection channel; Step S4, performing time-stationary narrowband interference detection according to the coherence coefficient interference detection threshold, and generating a time-stationary narrowband interference mask; Performing time-varying wideband interference detection according to the coherence coefficient interference detection threshold, and generating a time-varying wideband interference mask; Step S5, respectively fusing the time-stationary narrowband interference mask and the time-varying wideband interference mask for interference mask fusion, and outputting an interference detection result to obtain an image after interference signal suppression; In the said Step S1, transforming the azimuth multi-channel SAR data from two-dimensional time domain to range-direction frequency-azimuth direction time domain; during transmission, the antenna transmits with full aperture in the azimuth direction, and during reception, m sub-apertures of the antenna receive simultaneously in the azimuth direction. When there is RFI, the SAR echo received by the m-th channel is expressed in the range-direction frequency-azimuth direction time domain as: y m (η, f) = s m (η, f) + x m (η, f) + n m (η, f) where η is the azimuth time, f is the range frequency, and s m is the useful signal of the m-th channel, x m is the interference signal of the m-th channel, n m is the system noise of the m-th channel, y m is the SAR echo received by the m-th channel, and the subscript m represents the m-th channel; The interference signal received by the m-th channel is expressed as the time delay x of the signal received by the first channel m : where θ RFI represents the angle of the interference signal, C is the speed of light, x1 is the interference signal received by the first channel; d represents the distance between the phase centers of the two-channel antennas; In the said Step S2, using two receiving channels with a theoretical coherence coefficient of zero as the interference detection channels, and selecting two receiving channels with the farthest phase centers; the calculation process of the theoretical coherence coefficient is as follows: The spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel is expressed as γ 1,m represents the spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel, E{·} represents the mathematical expectation, y1 represents the SAR echo received by the first channel, the symbol * represents the conjugate operation. Since s m , x m , n m are independently distributed, the subscript m represents the m-th channel, and the correlation coefficient is simplified to: Among them, is the cross-correlation function of the useful signal s1 and s m , is the interference phase of the useful signal s1 and s m . where P(f a ) is the Doppler power spectral density function, and f a is the Doppler frequency; Δη m is the time delay between the echo received by the m-th channel and the first channel; If azimuth rectangular weighting, then It is expressed as Among them v s is the satellite velocity; Among them, c is the speed of light, and E() represents the expectation; Using two channels with zero coherence for interference detection, selecting the first channel and the two channels farthest from it.

2. The interference signal detection and suppression method applicable to an azimuth multi-channel SAR system according to claim 1, characterized in that In the said Step S3, assuming that the echo received by the m-th channel is completely decorrelated from the first channel; the coherence coefficient is estimated using the following formula, In the formula, is the estimated value of the coherence coefficient, N is the total number of samples, n represents the nth estimated sample, y1(n) and y m (n) represent the echo signals of the first channel and the mth channel respectively. The superscript * represents the conjugate operation. The probability density function pdf() of the estimated value of the coherence coefficient is shown as follows: where 2F1 is the hypergeometric function, is the estimated value of the coherence coefficient, γ is the coherence coefficient, and L is the number of independently and identically distributed samples.

3. The interference signal detection and suppression method applicable to an azimuth multi-channel SAR system according to claim 1, characterized in that In the said Step S4, calculating the coherence coefficient between the m-th channel and the first channel pixel by pixel using up-sampling in the azimuth direction: is the estimated coherence coefficient between the m-th channel and the 1-st channel, f l is the l-th range-frequency cell, η k is the k-th azimuth-time cell; Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det , it is determined that there is a time-stationary narrowband interference in the current cell; Otherwise, it is determined that there is no interference; for those detected to have time-stationary narrowband interference, perform pixel-by-pixel detection of time-stationary narrowband interference to generate a time-stationary narrowband interference detection mask image; To detect time-varying wideband interference, calculate the coherence coefficient between the m-th channel and the first channel pixel by pixel using up-sampling in the range direction; represents the coherence coefficient between the m-th channel and the 1st channel, l represents the l-th unit, and Δθ represents the direction of arrival of the interfering wave; Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det , it is determined that there is time-varying wideband interference in the current unit; otherwise, it is determined that there is no interference; the time-varying wideband interference is detected pixel by pixel to generate a time-varying wideband interference detection mask image.

4. A jamming signal detection and suppression system applicable to an azimuth multi-channel SAR system, characterized in that, Including: Module M1, performing range-direction Fourier transform on the original data; Module M2, selecting an interference detection channel; Module M3, calculating a coherence coefficient interference detection threshold according to the interference detection channel; Module M4, performing time-stationary narrowband interference detection according to the coherence coefficient interference detection threshold, and generating a time-stationary narrowband interference mask; Performing time-varying wideband interference detection according to the coherence coefficient interference detection threshold, and generating a time-varying wideband interference mask; Module M5, respectively fusing the time-stationary narrowband interference mask and the time-varying wideband interference mask for interference mask fusion, and outputting an interference detection result to obtain an image after interference signal suppression; In the said Module M1, transforming the azimuth multi-channel SAR data from two-dimensional time domain to range-direction frequency-azimuth direction time domain; during transmission, the antenna transmits with full aperture in the azimuth direction, and during reception, m sub-apertures of the antenna receive simultaneously in the azimuth direction. When there is RFI, the SAR echo received by the m-th channel is expressed in the range-direction frequency-azimuth direction time domain as: y m (η, f) = s m (η, f) + x m (η, f) + n m (η, f) where η is the azimuth time, f is the range frequency, s m is the useful signal of the m-th channel, x m is the interference signal of the m-th channel, n m is the system noise of the m-th channel, y m is the SAR echo received by the m-th channel, and the subscript m represents the m-th channel; The interference signal received by the m-th channel is expressed as the time delay x of the signal received by the first channel m : where θ RFI represents the angle of the interference signal, C is the speed of light, x1 is the interference signal received by the first channel; d represents the distance between the phase centers of the two-channel antennas; In the module M2, two receiving channels with a theoretical coherence coefficient of zero are used as interference detection channels, and two receiving channels with the farthest phase centers apart are selected; the calculation process of the theoretical coherence coefficient is as follows: The spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel is expressed as γ 1,m represents the spatial coherence coefficient between the SAR signal received by the first channel and the m-th channel, E{·} represents the mathematical expectation, y1 represents the SAR echo received by the first channel, the symbol * represents the conjugate operation, since s m , x m , n m are independently distributed, the subscript m represents the m-th channel, and the correlation coefficient is simplified to: Among them, is the cross-correlation function of the useful signal s1 and s m , is the interference phase of the useful signal s1 and s m . where P(f a ) is the Doppler power spectral density function, and f a is the Doppler frequency; Δη m is the time delay between the echo received by the m-th channel and the first channel. If azimuth rectangular weighting is applied, then is expressed as wherein v s is the satellite velocity; Among them, c is the speed of light, and E() represents the expectation; Use two channels with zero coherence for interference detection, and select the first channel and the two channels farthest from it.

5. The interference signal detection and suppression system applicable to the azimuth multi-channel SAR system according to claim 4, wherein In the module M3, it is assumed that the echo received by the m-th channel is completely decorrelated from the first channel; the coherence coefficient is estimated using the following formula In the formula, is the estimated value of the coherence coefficient, N is the total number of samples, n represents the nth estimated sample, y1(n) and y m (n) represent the echo signals of the first channel and the mth channel respectively, the superscript * represents the conjugate operation, and the probability density function pdf() of the estimated value of the coherence coefficient is shown as follows In the formula, 2F1 is the hypergeometric function, is the estimated value of the coherence coefficient, γ is the coherence coefficient, and L is the number of independent and identically distributed samples.

6. The interference signal detection and suppression system applicable to the azimuth multi-channel SAR system according to claim 5, characterized in that In the module M4, calculate the coherence coefficient between the m-th channel and the first channel pixel by pixel using azimuth upward samples: is the estimated value of the coherence coefficient between the m-th channel and the 1-st channel, f l is the l-th range-frequency cell, η k is the k-th azimuth-time cell; Compare the calculated coherence coefficient with the interference detection threshold γ det , and when the calculated coherence coefficient is greater than γ det , it is determined that there is a time-stationary narrowband interference in the current cell; Conversely, it is determined that there is no interference; for those detected to have time-stationary narrowband interference, perform pixel-by-pixel detection of time-stationary narrowband interference to generate a time-stationary narrowband interference detection mask map; To detect time-varying broadband interference, calculate the coherence coefficient between the m-th channel and the first channel pixel by pixel using range upward samples; represents the coherence coefficient between the m-th channel and the 1st channel, l represents the l-th unit, and Δθ represents the direction of arrival of the interference wave; Compare the calculated coherence coefficient with the interference detection threshold γ det , when the calculated coherence coefficient is greater than γ det , it is determined that there is time-varying broadband interference in the current cell; otherwise, it is determined that there is no interference; detect time-varying broadband interference pixel by pixel to generate a time-varying broadband interference detection mask image.

Images