Multi-channel radar sea clutter suppression method and system based on time-frequency analysis

Abstract

The application provides a multi-channel radar sea clutter suppression method and system based on time-frequency analysis, comprising the following steps: S1, calculating the coherent time of sea clutter by using sea state series; S2, transforming the sea clutter data into a time-frequency domain by using the coherent time of sea clutter for windowing; S3, suppressing the sea clutter in the time-frequency domain by using a time-domain sliding window-based subspace projection method; and S4, restoring the suppressed sea clutter data to a Doppler domain and then performing a second-stage suppression. The application can realize effective compensation of multi-channel radar sea clutter, fully considers the time de-coherent characteristics of sea clutter, and thus can realize more robust suppression of sea clutter in the space-time-frequency domain.

Description

Technical Field

[0001] This invention relates to the field of radar clutter processing technology, specifically to a multi-channel radar sea clutter suppression method and system based on time-frequency analysis. Background Technology

[0002] Due to the high-speed movement of the radar platform, the geographic clutter spectrum will be severely expanded, causing useful moving target echo signals to be submerged in the main lobe clutter and unable to be effectively detected. Therefore, STAP processing technology is required to push the target into the main lobe region for detection.

[0003] However, for moving targets on the sea surface, complex random internal disturbances cause sea clutter to exhibit spatiotemporal characteristics completely different from land clutter. These characteristics are reflected in changes in the statistical distribution of clutter amplitude, the generation of spatiotemporal decorrelation characteristics, and further expansion of the spatiotemporal power spectrum. Due to the time-varying and space-varying characteristics of sea clutter, the echo received by radar undergoes further spectral expansion, making sea clutter suppression and detection of moving targets on the sea surface extremely difficult. Especially at high sea states, the target echo signal will be completely submerged in random sea clutter signals, and the signal-to-clutter-to-noise ratio deteriorates even more severely. The traditional STAP method is unable to complete the task of suppressing severely expanded sea clutter, thus the traditional spatiotemporal adaptive processing method cannot achieve good clutter suppression capability, affecting target detection performance.

[0004] Patent document CN109061598A (application number: CN201810984125.0) discloses a STAP clutter covariance matrix estimation method. This method first reconstructs the clutter in the target frequency channel of the CUT, then extracts clutter components other than the target frequency from the CUT, and finally calculates the clutter covariance matrix of the CUT from the reconstructed clutter. However, this invention does not perform clutter signal suppression processing in the space-time-frequency multidimensional domain. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide a multi-channel radar sea clutter suppression method and system based on time-frequency analysis.

[0006] A multi-channel radar sea clutter suppression method based on time-frequency analysis provided by the present invention includes:

[0007] Step S1: Calculate the coherence time of sea clutter using sea state series;

[0008] Step S2: Utilize the coherent time windowing of sea clutter to transform the sea clutter data into the time-frequency domain;

[0009] Step S3: Suppress sea clutter using a subspace projection method based on time-domain sliding windows in the time-frequency domain;

[0010] Step S4: Restore the suppressed clutter data to the Doppler domain, and then perform the second stage of suppression.

[0011] Preferably, in step S1:

[0012] The coherence time of sea clutter is expressed as:

[0013]

[0014] in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

[0015] Preferably, in step S2:

[0016] Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows:

[0017]

[0018] in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit;

[0019] Then the first n The signal received by each array element, after range pulse compression, is represented as:

[0020]

[0021] in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element;

[0022] To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as:

[0023]

[0024] in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to:

[0025]

[0026] in, This represents the physical distance between adjacent directional channels, satisfying... , d Indicates the distance between adjacent array elements;

[0027] The signal form extracted within the coherence time is as follows:

[0028]

[0029] in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period;

[0030] The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function:

[0031]

[0032] in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation.

[0033] Preferably, in step S3:

[0034] Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of r The coherent data blocks within the distance ring, obtained by sliding window K m The data set is represented as follows:

[0035]

[0036] in:

[0037]

[0038] in, express In the coherent data block, for the first N The data received by the receiving channel in each azimuth direction is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of r The Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is:

[0039]

[0040] in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose.

[0041] The clutter suppression weight vector is expressed as:

[0042]

[0043] in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.:

[0044]

[0045] in Indicates the Kronecker product. The pulse repetition time, The target guidance vector;

[0046] For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows:

[0047]

[0048] Preferably, when performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression, but rather the original data is processed separately. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

[0049] A multi-channel radar sea clutter suppression system based on time-frequency analysis provided by the present invention includes:

[0050] Module M1: Calculates the coherence time of sea clutter using sea state series;

[0051] Module M2: Utilizes coherent time windowing of sea clutter to transform sea clutter data into the time-frequency domain;

[0052] Module M3: Employs a subspace projection method based on time-domain sliding windows to suppress sea clutter in the time-frequency domain;

[0053] Module M4: Restores the suppressed clutter data to the Doppler domain, and then performs the second stage of suppression.

[0054] Preferably, in module M1:

[0055] The coherence time of sea clutter is expressed as:

[0056]

[0057] in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

[0058] Preferably, in module M2:

[0059] Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows:

[0060]

[0061] in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit;

[0062] Then the first n The signal received by each array element, after range pulse compression, is represented as:

[0063]

[0064] in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element;

[0065] To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as:

[0066]

[0067] in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to:

[0068]

[0069] in, This represents the physical distance between adjacent directional channels, satisfying... , d Indicates the distance between adjacent array elements;

[0070] The signal form extracted within the coherence time is as follows:

[0071]

[0072] in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period;

[0073] The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function:

[0074]

[0075] in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation.

[0076] Preferably, in module M3:

[0077] Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of rThe coherent data blocks within the distance ring, obtained by sliding window K m The data set is represented as follows:

[0078]

[0079] in:

[0080]

[0081] in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of r The Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is:

[0082]

[0083] in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose.

[0084] The clutter suppression weight vector is expressed as:

[0085]

[0086] in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.:

[0087]

[0088] in Indicates the Kronecker product. The pulse repetition time, The target guidance vector;

[0089] For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows:

[0090]

[0091] Preferably, when performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression, but rather the original data is processed separately. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

[0092] Compared with the prior art, the present invention has the following beneficial effects:

[0093] 1. This invention suppresses clutter signals in the space-time-frequency multidimensional domain, making comprehensive use of information from multiple dimensions to achieve better clutter suppression performance.

[0094] 2. This invention divides the data into multiple sets of highly coherent data based on the statistically obtained sea clutter coherence time, thereby effectively improving the adverse effect of sea clutter decoherence characteristics on clutter suppression;

[0095] 3. This invention can effectively compensate for sea clutter in multi-channel radar, and fully considers the time-decoherence characteristics of sea clutter, thereby achieving more robust suppression of sea clutter in the space-time-frequency domain. Attached Figure Description

[0096] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0097] Figure 1 This is a schematic diagram illustrating the reason for using suppression in the time-frequency domain;

[0098] Figure 2 This is a flowchart illustrating a multi-channel radar sea clutter suppression method based on time-frequency analysis proposed in this invention.

[0099] Figure 3 This is a schematic diagram of the spatial-temporal-frequency domain signal composition provided in the embodiments of the present invention;

[0100] Figure 4 This is a schematic diagram illustrating the restoration of the signal after the first suppression, as proposed in this invention.

[0101] Figure 5 The distance-Doppler spectrum before sea clutter suppression;

[0102] Figure 6 Range-Doppler spectrum image after sea clutter suppression using the method proposed in this invention.

[0103] Figure 7 This is a flowchart of the present invention. Detailed Implementation

[0104] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0105] Example 1:

[0106] The present invention provides a multi-channel radar sea clutter suppression method based on time-frequency analysis, such as... Figures 1-7 As shown, it includes:

[0107] Step S1: Calculate the coherence time of sea clutter using sea state series;

[0108] Specifically, in step S1:

[0109] The coherence time of sea clutter is expressed as:

[0110]

[0111] in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

[0112] Step S2: Utilize the coherent time windowing of sea clutter to transform the sea clutter data into the time-frequency domain;

[0113] Specifically, in step S2:

[0114] Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows:

[0115]

[0116] in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit;

[0117] Then the first n The signal received by each array element, after range pulse compression, is represented as:

[0118]

[0119] in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element;

[0120] To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as:

[0121]

[0122] in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to:

[0123]

[0124] in, This represents the physical distance between adjacent directional channels, satisfying... , d Indicates the distance between adjacent array elements;

[0125] The signal form extracted within the coherence time is as follows:

[0126]

[0127] in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period;

[0128] The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function:

[0129]

[0130] in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation.

[0131] Step S3: Suppress sea clutter using a subspace projection method based on time-domain sliding windows in the time-frequency domain;

[0132] Specifically, in step S3:

[0133] Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of r The coherent data blocks within the distance ring, obtained by sliding window K m The data set is represented as follows:

[0134]

[0135] in:

[0136]

[0137] in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is fa And at a slope distance of r The Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is:

[0138]

[0139] in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose.

[0140] The clutter suppression weight vector is expressed as:

[0141]

[0142] in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.:

[0143]

[0144] in Indicates the Kronecker product. The pulse repetition time, The target guidance vector;

[0145] For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows:

[0146]

[0147] Step S4: Restore the suppressed clutter data to the Doppler domain, and then perform the second stage of suppression.

[0148] Specifically, when performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression; instead, each channel is used separately for the original data. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

[0149] Example 2:

[0150] Example 2 is a preferred example of Example 1, and is used to illustrate the present invention in more detail.

[0151] Those skilled in the art can understand the multi-channel radar sea clutter suppression method based on time-frequency analysis provided by the present invention as a specific implementation of a multi-channel radar sea clutter suppression system based on time-frequency analysis. That is, the multi-channel radar sea clutter suppression system based on time-frequency analysis can be implemented by executing the steps of the multi-channel radar sea clutter suppression method based on time-frequency analysis.

[0152] A multi-channel radar sea clutter suppression system based on time-frequency analysis provided by the present invention includes:

[0153] Module M1: Calculates the coherence time of sea clutter using sea state series;

[0154] Specifically, in module M1:

[0155] The coherence time of sea clutter is expressed as:

[0156]

[0157] in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

[0158] Module M2: Utilizes coherent time windowing of sea clutter to transform sea clutter data into the time-frequency domain;

[0159] Specifically, in module M2:

[0160] Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows:

[0161]

[0162] in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit;

[0163] Then the first nThe signal received by each array element, after range pulse compression, is represented as:

[0164]

[0165] in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element;

[0166] To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as:

[0167]

[0168] in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to:

[0169]

[0170] in, This represents the physical distance between adjacent directional channels, satisfying... ,d Indicates the distance between adjacent array elements;

[0171] The signal form extracted within the coherence time is as follows:

[0172]

[0173] in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period;

[0174] The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function:

[0175]

[0176] in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation.

[0177] Module M3: Employs a subspace projection method based on time-domain sliding windows to suppress sea clutter in the time-frequency domain;

[0178] Specifically, in module M3:

[0179] Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of r The coherent data blocks within the distance ring, obtained by sliding window K m The data set is represented as follows:

[0180]

[0181] in:

[0182]

[0183] in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of rThe Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is:

[0184]

[0185] in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose.

[0186] The clutter suppression weight vector is expressed as:

[0187]

[0188] in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.:

[0189]

[0190] in Indicates the Kronecker product. The pulse repetition time, The target guidance vector;

[0191] For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows:

[0192]

[0193] Module M4: Restores the suppressed clutter data to the Doppler domain, and then performs the second stage of suppression.

[0194] Specifically, when performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression; instead, each channel is used separately for the original data. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

[0195] Example 3:

[0196] Example 3 is a preferred example of Example 1, and is used to illustrate the present invention in more detail.

[0197] This invention provides a radar sea clutter suppression method based on time-frequency analysis, the flowchart of which is shown below. Figure 1 As shown, the specific steps of this method are as follows:

[0198] Step S1: Calculate the coherence time of sea clutter using sea state series. The coherence time of sea clutter can be statistically expressed as:

[0199]

[0200] in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the smallest resolution unit on the radar ground.

[0201] Step S2: Using coherent time windowing, transform the sea clutter data to the time-frequency domain. Assume the radar transmits a linear frequency modulated signal, its first... n The signal form of 0 azimuth receiving channels can be represented as:

[0202]

[0203] in N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the radiation pattern gain of the full-element array, For the first i Clutter amplitude at each scattering point B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. and For the first i The scattering point is relative to the radar's downward angle and azimuth. This represents the physical distance between adjacent directional channels, satisfying... ,in d Indicates the distance between adjacent array elements. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, d T This represents the distance from the equivalent launch center to the reference array element.

[0204] Therefore, the form of the sea clutter signal within the coherence time can be obtained as follows:

[0205]

[0206] in For window functions, Let represent the intermediate pulse time of the selected coherent time interval. Its value should be such that the window function intersects with the pulse sequence at least once. Therefore, the window length of the window function can be obtained as . , Let be the pulse repetition time. Transforming it to the time-frequency domain, the signal form is:

[0207]

[0208] in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a The frequency is the Doppler frequency.

[0209] Step S3: In the time-frequency domain, a subspace projection method based on time-domain sliding windows is used to suppress sea clutter. It is assumed that for each coherent block, a... K m Sub-temporal sliding window. For And at a slope distance of r The coherent data blocks within the distance ring, obtained by sliding window K m Group data can be represented as:

[0210]

[0211] in:

[0212]

[0213] in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of r The Doppler domain discrete data within the range loop. The clutter subspace projection matrix at this point is...

[0214]

[0215] in, Since the spatial steering vector is the equivalent after the time-domain sliding window, the clutter suppression weight vector can be expressed as:

[0216]

[0217] in The equivalent target orientation vector is:

[0218]

[0219] in Representing the Kronecker product, we can ultimately obtain the result for the th... The sea clutter coherent block clutter suppression results corresponding to each time point are as follows:

[0220]

[0221] Step S4: Restore the suppressed clutter data to the Doppler domain, and then perform the second stage of suppression. This is done by processing the original data before... Channel and back After performing time-frequency domain clutter suppression as described in step S3 of the channel execution, and obtaining two sets of suppression results, based on... Figure 4 The restoration method restores the suppressed clutter data to the range-slow time domain, considering the slant range as... r The distance unit requires the recovery of the first signal. k pulse data As can be seen from the aforementioned inhibition methods, it participates in the first... The coherent blocks up to the first In the weighted operation of each coherent block, therefore Need to start from the first , , … , The signal is extracted from the given information. Based on the weighting relationship during windowing, the restored signal has the following form:

[0222]

[0223] Reconstruct the restored signal

[0224]

[0225] in , and They represent and In executing the i The distance-Doppler domain data is obtained after the time-domain sliding window. Therefore, the clutter suppression weight vector at this time can be obtained as:

[0226]

[0227] in The dimension is The identity matrix, If this is the equivalent target steering vector at this point, then the result of the second-stage clutter suppression is:

[0228]

[0229] The simulation was performed according to the operation steps of this embodiment. The simulation system parameters used are as follows: platform height is 508km, radar carrier frequency is 2.6 GHz, pulse repetition frequency is 6000 Hz, signal bandwidth is 3 MHz, and the number of azimuth receiving channels is 16. All implementation steps of this embodiment are performed on the MATLAB2020 simulation platform. Figure 5 The range-Doppler spectrum before sea clutter suppression. Figure 6 The image shows the range-Doppler spectrum after sea clutter suppression using the method proposed in this invention. The results demonstrate that the method provided by this invention can achieve good suppression of sea clutter for multi-channel radar.

[0230] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0231] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, 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. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A multi-channel radar sea clutter suppression method based on time-frequency analysis, characterized in that, include: Step S1: Calculate the coherence time of sea clutter using sea state series; Step S2: Utilize the coherent time windowing of sea clutter to transform the sea clutter data into the time-frequency domain; Step S3: Suppress sea clutter using a subspace projection method based on time-domain sliding windows in the time-frequency domain; Step S4: Restore the suppressed clutter data to the Doppler domain, and then perform the second stage of suppression; In step S2: Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows: in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit; Then the first n The signal received by each array element, after range pulse compression, is represented as: in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element; The operating wavelength of the radar; To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as: in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to: in, This represents the physical distance between adjacent directional channels, satisfying... , d Indicates the distance between adjacent array elements; The signal form extracted within the coherence time is as follows: in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period; The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function: in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation; When performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression; instead, each channel is used separately for the original data. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

2. The multi-channel radar sea clutter suppression method based on time-frequency analysis according to claim 1, characterized in that, In step S1: The coherence time of sea clutter is expressed as: in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

3. The multi-channel radar sea clutter suppression method based on time-frequency analysis according to claim 1, characterized in that, In step S3: Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of r The coherent data blocks within the distance ring, This represents the intermediate pulse time of the selected coherent time period; its sliding window yields... K m The data set is represented as follows: in: in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of r The Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is: in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose. The clutter suppression weight vector is expressed as: in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.: in Indicates the Kronecker product. The pulse repetition time, The target guidance vector; For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows: 。 4. A multi-channel radar sea clutter suppression system based on time-frequency analysis, characterized in that, include: Module M1: Calculates the coherence time of sea clutter using sea state series; Module M2: Utilizes coherent time windowing of sea clutter to transform sea clutter data into the time-frequency domain; Module M3: Employs a subspace projection method based on time-domain sliding windows to suppress sea clutter in the time-frequency domain; Module M4: Restores the suppressed clutter data to the Doppler domain, and then performs the second stage of suppression; In module M2: Based on the coherence time of sea clutter, at a time interval of The pulse sequences of sea clutter echoes are correlated, and the echo intervals are extracted. The radar transmits a linear frequency modulated signal to detect sea clutter signals. The transmitted signal is represented as follows: in, T p Indicates the pulse duration. f c The carrier frequency for radar transmission signals. t For fast time variables, Represents a rectangular window function. µ The modulation frequency of a linear frequency modulated signal; The imaginary unit; Then the first n The signal received by each array element, after range pulse compression, is represented as: in, For slow time variables, The pulse repetition time, N c The slant distance is r The number of sea surface scattering points included in the distance ring. For the first i Clutter amplitude at each scattering point For the radiation pattern gain of the full-element array, B The bandwidth of the radar transmitted signal. c At the speed of light, R 0,i Indicates the initial slope distance. The instantaneous slant range random term introduced for sea clutter. and These represent the yaw magnitude and yaw angle, respectively. For the platform's movement speed, and For the first i The scattering point is relative to the radar's downward angle and azimuth. d n and d T They represent the first n The distance of each array element and the equivalent launch center from the reference array element; The operating wavelength of the radar; To obtain the signal model at the receiving channel level, azimuth channel synthesis is performed, with the azimuth direction uniformly divided into... N 0 receive channels, each receive channel contains The array element, the first n The signal from 0 azimuth receiving channels is represented as: in, For the first n The weights of the i-th channel are determined, and it is assumed that the pattern weighting function in the array element is slowly varying. Since the random motion term of sea clutter changes slowly between different channels, the weights of the i-th channel are... n The signal format of 0 azimuth receiving channels can be further simplified to: in, This represents the physical distance between adjacent directional channels, satisfying... , d Indicates the distance between adjacent array elements; The signal form extracted within the coherence time is as follows: in, Therefore A normalized window function with an odd length and a midpoint. This indicates the midpoint pulse time of the selected coherent time period; The windowed sea clutter data is transformed to the time-frequency domain, and a rectangular window is selected as the window function: in, Represents convolution. This represents the amplitude of the signal in the time-frequency domain. f a For Doppler frequency, To t k Perform a Fourier transform operation; When performing time-frequency domain sea clutter suppression, not all channels are used for the first stage of suppression; instead, each channel is used separately for the original data. Channel and back The channel is subjected to time-frequency domain clutter suppression. After being restored to the range-Doppler domain, the residual clutter of the two channels is suppressed again using the subspace projection method based on time-domain sliding window to obtain the suppression result.

5. The multi-channel radar sea clutter suppression system based on time-frequency analysis according to claim 4, characterized in that, In module M1: The coherence time of sea clutter is expressed as: in, For radar operating wavelength, The wind speed under the current sea conditions. Let be the error function, satisfying , This represents the area of ​​the radar's smallest ground resolution unit. It is the integral variable.

6. The multi-channel radar sea clutter suppression system based on time-frequency analysis according to claim 4, characterized in that, In module M3: Take for each coherent block K m Sub-temporal sliding window, for And at a slope distance of r The coherent data blocks within the distance ring, This represents the intermediate pulse time of the selected coherent time period; its sliding window yields... K m The data set is represented as follows: in: in, express In the coherent data block, for the first N The data received by the 0th azimuth receiving channel is processed in the first... k m The Doppler frequency obtained by the second sliding window is f a And at a slope distance of r The Doppler domain discrete data within the range loop; the clutter subspace projection matrix at this time is: in, H is the equivalent spatial guiding vector after the time-domain sliding window, where H is the conjugate transpose. The clutter suppression weight vector is expressed as: in, For the equivalent target orientation vector, For dimension is The identity matrix, i.e.: in Indicates the Kronecker product. The pulse repetition time, The target guidance vector; For the first The sea clutter coherent block clutter suppression results corresponding to each time point are as follows: 。

Images