CN112255607A - Sea clutter suppression method - Google Patents
Sea clutter suppression method Download PDFInfo
- Publication number
- CN112255607A CN112255607A CN202011062961.7A CN202011062961A CN112255607A CN 112255607 A CN112255607 A CN 112255607A CN 202011062961 A CN202011062961 A CN 202011062961A CN 112255607 A CN112255607 A CN 112255607A
- Authority
- CN
- China
- Prior art keywords
- frequency
- sea clutter
- point
- radar echo
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/414—Discriminating targets with respect to background clutter
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a sea clutter suppression method, belonging to the radio frequency technology and radar technology field, comprising the following steps of S1 Fourier transform of radar echo time domain sequence, firstly transform to frequency domain, then transform to Doppler frequency spectrum, S2 divide the radar echo into frequency windows, a plurality of radar echoes correspond to a plurality of frequency windows correspondingly, and obtain the median point, med (A) of the radar echo in each frequency windown[xij]) Representation versus frequency window An[xij]All points in the interior are taken as the median value, yijIs a pixel point xijTwo-dimensional median filtered output value yijGet dij＝|xij‑yij|，dijThe gray value at (i, j) in the detail image of the Doppler spectrogram, S3 if dijIf the value is larger than the threshold T, the point (i, j) of the corresponding position is considered as the point of the sea clutter, and otherwise, the point without the sea clutter at the corresponding position (i, j) is considered. The method can improve the target detection probability, and has less calculation amount and higher processing speed.
Description
Technical Field
The invention belongs to the technical field of radio frequency technology and radar, and particularly relates to a method for suppressing sea clutter of a high-frequency ground wave radar of a miniaturized antenna array.
Background
Scholars at home and abroad propose various sea clutter suppression methods with different effects under different conditions. The traditional cancellation techniques are iterative and subspace decomposition. The iterative sea clutter cancellation algorithm proposed by Root in 1998 utilizes the iterative cancellation algorithm to effectively remove sea clutter after sea clutter is modeled as a sine model, and adopts short-time Doppler discrete Fourier transform to extract ship target information from a short time sequence. Primarily for identifying vessels in the marginal areas of sea clutter. The method is essentially an improved Clean algorithm, has high calculation speed and is suitable for the field of processing mass data in real time. However, the iteration times of the algorithm are difficult to control, the performance of the algorithm is limited in identifying the ship with the velocity spectrum just near the center of the first-order sea clutter spectral line, and the effect is not good when multiple targets are detected.
Researchers in 1993 such as m.w.poon, r.h.khan, Son Le-ngcc, etc. use subspace decomposition to implement sea clutter cancellation, usually singular value decomposition. And the first-order sea clutter and singular values partially reflecting noise are set to be zero, and signal components and noise of the first-order sea clutter can be eliminated through signal reconstruction. However, the clutter order of the method is not easy to determine, and the engineering implementation still has difficulty.
In addition, learners are happy bin, monetary vibration, Demarty and the like are engaged in sea clutter suppression according to the characteristics of sea clutter adjacent to each other. By utilizing the characteristics of the sea clutter, if the Doppler frequency change of the sea clutter is more than 1 and less than 2 Doppler resolution units in the azimuth and the adjacent distance units of the same wave beam, the echo data of the adjacent units can be used for inhibiting the clutter. Within a transmitting synthetic beam, amplitude-frequency characteristics of adjacent distance units of sea clutter necessarily have certain correlation. Or the average sea clutter time-frequency distribution graph is subtracted from the total time-frequency echo graph, so that the sea clutter can be weakened. Or the clutter cancellation preprocessing of the adjacent distance unit is improved, and the detection algorithm is verified by using a conventional target.
Before and after 2008, enyon bin considers that multi-angle and multi-dimensional joint information such as the distance, Doppler frequency shift, amplitude and shape structure of a Bragg peak and multi-dimensional characteristics such as continuity and symmetry of the amplitude of a sea wave Bragg peak are used for inhibiting first-order sea clutter, and provides an improvement method of a Clean algorithm. The method can effectively remove the first-order clutter with high intensity, has high calculation speed, but has the cost that the stronger target signal can be used as a sea clutter signal to be inhibited.
In about 2008, a french scholar Demarty also proposed a sea clutter modeling technology based on an electromagnetic scattering mechanism, and calculated the interaction between electromagnetic waves and the environment, thereby providing a new way for sea clutter suppression.
The above plans all have the problem of low sea clutter suppression effect or too large antenna array, and a novel sea clutter suppression method needs to be developed to overcome the above defects in the prior art.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a method for suppressing sea clutter, and aims to solve the problems of low target detection azimuth resolution and complex data processing method process in the prior art.
In order to achieve the above object, the present invention provides a sea clutter suppression method, which comprises the following steps:
s1: firstly, the time domain sequence of the radar echo is Fourier transformed to the frequency domain, then the Fourier transform is carried out again to the Doppler frequency spectrum,
s2: dividing a frequency window for the radar echo, corresponding to multiple frequency windows, finding the median point, med (A) of the radar echo in each frequency windown[xij]) Presentation pairFrequency window An[xij]All points in the interior are taken as the median value, yijIs a pixel point xijThe two-dimensional median filtered output value is:
wherein the content of the first and second substances,
the gray value of the image point at the (N +1) th position in the frequency window, N is a positive integer, N represents the point number of the frequency window, and d is takenij＝|xij-yij|，dijIndicating the gray value at (i, j) in the detail image of the Doppler spectrogram, wherein i and j are the ith row and the jth column of the detail image of the Doppler spectrogram, i and j are natural numbers,
s3: if d isijIf the value is larger than the threshold T, the point (i, j) of the corresponding position is considered as the point of the sea clutter, and otherwise, the point without the sea clutter at the corresponding position (i, j) is considered.
Further, in step S1, when performing fourier transform, a longer FFT formula is adopted for the time-domain sequence data of the radar echo with higher signal-to-noise ratio:
wherein i and j are natural numbers, N is the total length of the time domain sequence of the radar echo with higher signal-to-noise ratio, and xi,j(n) refers to time domain data of the radar echo signal of the nth row, k refers to the kth frequency domain, and the higher signal-to-noise ratio refers to the signal-to-noise ratio of 10dB to 50 dB.
Further, in step S1, when performing fourier transform, a shorter FFT formula is used for the time-domain sequence data of the radar echo with a lower signal-to-noise ratio:
wherein i and j are natural numbers, M is the total length of the time domain sequence of the radar echo with lower signal-to-noise ratio, and xi,j(n) refers to time domain data of the radar echo signal of the nth row, k refers to the kth frequency domain, and the lower signal-to-noise ratio refers to the signal-to-noise ratio of-20 dB to 10 dB.
Further, in step S2, on the two-dimensional range-doppler spectrogram, an[xij]It means that a window operation of N × N-2N +1 is performed on a point in the image with the point (i, j) as the center, where N is an odd integer.
Further, in step S2, the detail image of the doppler spectrogram is represented by D,
D＝|Y(k)-S(k,θ)|
y (k) is obtained by converting Y (ω), and the relationship between k and ω satisfies the following equation relationship ω 2 pi k,
wherein Y (ω) ═ med (X (ω)),
where N (p) is a normalization factor, p is a spreading factor, and is set to 15, θ0Is the wind direction, theta is the observation direction, the wave spectral function s (k).
Further, in step S2, a plurality of frequency windows are defined for the radar echo according to a frequency range of the first-order sea clutter, wherein the frequency range of the first-order sea clutter is 0 to 8 Hz.
Further, the frequency window is divided for the radar echo in the following manner: determining a frequency h in the frequency range of the first-order sea clutter, wherein the radar echo is w, and the frequency windows divided by the radar echo are (w-h) - (w + h).
Through the technical scheme, compared with the prior art, the invention can obtain the following beneficial effects:
the method for suppressing the sea clutter is suitable for target detection under a miniaturized radar antenna array, is beneficial to enhancing the high-frequency ground wave radar target detection performance, and realizes the detection accuracy and detection efficiency of small targets such as ships under different sea conditions. Specifically, during Fourier transformation, FFT (fast Fourier transform) formulas with different lengths are adopted for time domain sequence data of radar echoes with higher signal-to-noise ratios and time domain sequence data of radar echoes with lower signal-to-noise ratios, the transformation accuracy is improved, a median filtering method is adopted for sea clutter suppression, the hardware requirement is low, and engineering implementation is easy.
Drawings
FIG. 1 is a schematic flow chart of a sea clutter suppression method according to the present invention;
FIG. 2 is a schematic diagram of sample points and range bins in an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a sea clutter suppression method under a small antenna array, which is improved aiming at the sea clutter suppression method for the target detection of a miniaturized high-frequency ground wave radar. In the method: the near real-time sea state information which can be provided by the radar itself is used as the prior knowledge of sea clutter suppression, and clutter is removed by applying a two-dimensional median filtering and a cancellation hybrid algorithm of a nearest distance unit. The invention aims to recover the discovery probability level under the noise background as much as possible, thereby improving the suppression performance of first-order sea clutter and background noise and improving the detection probability. The high-frequency radar sea clutter suppression method has feasibility after the radar antenna is miniaturized, can improve the signal-to-clutter ratio to a certain extent, and can further improve the detection performance of offshore sea surface ship targets after the radar antenna is miniaturized.
The design idea of the invention is as follows: the range-doppler space is examined for samples of the input signal, and an observation window consisting of an odd number of sample values is used to examine and determine whether the sample values represent a signal. Specifically, the numerical values in the observation window are subjected to size arrangement, and the value at the middle position of the observation window is output as a median. The amplitude and phase of the signal corresponding to the median are compared and judged according to the principles described herein, and if so, the signal is representative of the signal that must be retained, otherwise, the signal is filtered out. Values are discarded continuously while new window samples are taken continuously, and the above analysis process is repeated for all actual received data. The sea clutter suppression is completed by adopting the process.
Median filtering is a nonlinear signal processing technique based on the ordering statistical theory to effectively suppress noise. The median filter functions to divide the amplitude of each range gate signal by the average of the amplitudes of the doppler frequency bins over all range gates. The median filtering method is from image processing, can effectively restrain impulse noise and salt and pepper noise, solve the relevant interference problem, and can effectively protect boundary information. However, like other sorting algorithms, the method has the defect that the calculation times are uncertain because the method has a loop iteration structure, so that the classic median filter is only suitable for software implementation and is not suitable for hardware implementation with high real-time requirements.
In the following, the reason why the conventional median filtering algorithm is complicated is explained in detail. For example, after setting the register size, a sort calculation is performed to determine the median value of the data in the data register. In sorting, the sorting is carried out until a certain median is determined, and stopping is carried out, namely the (m × n-1)/2 th maximum value is obtained. The number of the sorting operations is:
therefore, conventional algorithms to derive the values of the mxn filter window require 3/[8 (n) (2×m2-1)]2-fold comparison calculation. For example, a 3 × 3 window, this number is 30; 7 x 7 windows, this number is 900. If the window size is increased continuously, the number of operations is increased sharply, which makes the hardware processing speed slow, increases the size of the resource register sharply, increases the resource occupancy rate sharply, and the like hardware difficulties.
Based on the consideration, the invention provides a rapid two-dimensional median filtering algorithm introduced into the data processing of the miniaturized high-frequency ground wave radar. One advantage is that the computation speed is increased, and the sorting basic unit is two numbers arranged in an ascending (or descending) order. The traditional processing method needs a large amount of iteration, and the method is different from the traditional processing method, only has a very small amount of loop iteration, eliminates the uncertainty of time delay from input to output, occupies small memory, is suitable for a field programmable gate array, and is beneficial to realizing the butt joint with engineering application. The butt joint with engineering practicality is realized.
The process of the present invention is described in more detail below. Before that, it is added that the target appears as a white dot on the range-doppler plot, corresponding to a gray scale value of 255. Sea clutter appears on the range-doppler plot as close to a white point, corresponding to a gray value of approximately 255. Sea clutter and targets have some confusion.
1. Signal detection and conversion
Firstly, Fourier transform is carried out on a time domain sequence S (n) of the signal, and the time domain sequence S (n) is transformed into a frequency domain to obtain a complex data sequence S (omega). Fourier transform is again performed to transform to Doppler (Doppler) spectrum. Specifically, FFTs with different lengths and multiple sliding windows are adopted simultaneously in the Doppler processing (FFTs refers to fast fourier transform). Typical FFT lengths are from 8 to 1024 points. Research finds that for a low-speed sailing ship, the shorter FFT has better detection performance for detection, and after the ship reaches high speed and advances at a constant speed, the longer FFT with the median background is more suitable for detection. It is noted that in radar observation measurement, the distance and speed errors are small, and the estimation of distance and speed parameters can be optimized by adopting a multi-length FFT algorithm. For data with high signal-to-noise ratio, longer data is used, and for data with lower signal-to-noise ratio, shorter FFT is used.
The shorter FFT formula is as follows:
the longer FFT formula is as follows:
in the formula, Zi,j(k)、Fi,j(k) Representing the radar return signal. Because ω is 2 π k, Zi,j(k)、Fi,j(k) Namely, it is
i. j is a natural number, M, N has the same meaning, and is the total length of the respective sequence, xi,j(n) refers to time domain data of the radar echo signal of the nth row, and k refers to the kth frequency domain.
2. Median filtering
On a two-dimensional range-Doppler spectrum, An[xij]Means that a window operation is performed on a point in the image with N × N-2N +1 (where N is an odd integer and N is a positive integer) centered on the point (i, j), med (a)n[xij]) Represents window An[xij]All points within the cluster are taken as the median. y isijIs a pixel point xijThe median filtered output value, namely:
wherein the content of the first and second substances,
the gray value of the image point at the (N +1) th position in the window. For xijMedian filtering was performed with a 3 × 3 window:
yij(ω)＝med(xij(ω))，ω＝2πk
for the entire data sequence, the output of the median filtering can represent:
Y(ω)＝med(X(ω))，ω＝2πk
the absolute values are:
D＝|Y(k)-S(k,θ)|
d is a detail image of the Doppler spectrogram,
the calculation of Y (k) is obtained from Y (ω), the relationship of k to ω satisfies the following equation relationship ω 2 pi k,
wherein the wave spectrum function s (k) is described as follows:
in the formula (I), the compound is shown in the specification,
g is the acceleration of the gravity and,
is the wind speed measured at 19.5 meters off-shore, k is the wave number of the ocean waves,
then, by multiplying an angle function of a radar emission wave by the ocean wave spectrum s (k), a directivity spectrum function is obtained as follows:
where N (p) is a normalization factor and p is a spreading factor, empirically set to 15, θ0Is the wind direction and θ is the observation direction. For example: to theta0The waves propagate along the line of sight of the waves emitted by the radar.
For a single pixel, consider:
dij＝|xij-yij|
dijthe gray value at (i, j) in the detail image D of the doppler spectrogram is represented.
3. Judgment of
Using a threshold empirical value T, and comparing it with dijAnd (6) comparing. If d isijAnd if the position x is larger than the threshold T, the position (i, j) of the corresponding position is considered as a point of the sea clutter. And conversely, considering that x is (i, j) as a point without the sea clutter.
The invention also provides a sea clutter suppression method, which comprises the following specific steps:
1. signal detection and conversion
The time domain sequence S (n) of the signal is first fourier transformed to the frequency domain to obtain the complex data sequence S (ω), and then fourier transformed again to the Doppler (Doppler) spectrum. Specifically, FFTs with different lengths and multiple sliding windows are adopted simultaneously in the Doppler processing (FFTs refers to fast fourier transform). Typical FFT lengths are from 8 to 1024 points. Research finds that for a low-speed sailing ship, the shorter FFT has better detection performance for detection, and after the ship reaches high speed and advances at a constant speed, the longer FFT with the median background is more suitable for detection. It is noted that in radar observation measurement, the distance and speed errors are small, and the estimation of distance and speed parameters can be optimized by adopting a multi-length FFT algorithm. For data with high signal-to-noise ratio, longer data is used, and for data with lower signal-to-noise ratio, shorter FFT is used.
The shorter FFT formula is as follows:
the longer FFT formula is as follows:
in the formula, Zi,j(k)、Fi,j(k) Representing the radar return signal. Because ω is 2 π k, Zi,j(k)、Fi,j(k) Namely, it is
i. j is a natural number, M, N has the same meaning, and is the total length of the respective sequence, xi,j(n) refers to time domain data of the radar echo signal of the nth row, and k refers to the kth frequency domain.
2. Median filtering
On a two-dimensional range-Doppler spectrum, An[xij]Means that points in the image are centered at point (i, j) and N is 2N +1 (where N is N ═ 2N +1N is an odd integer, N is a positive integer) window operation, med (A)n[xij]) Represents window An[xij]All points within the cluster are taken as the median. y isijIs a pixel point xijThe median filtered output value, namely:
wherein the content of the first and second substances,
the gray value of the image point at the (N +1) th position in the window. For xijFor example, median filtering is performed using a 3 × 3 window:
yij(ω)＝med(xij(ω))，ω＝2πk
for the entire data sequence, the output of the median filtering can represent:
Y(ω)＝med(X(ω))，ω＝2πk
after the above processing, the amplitude, phase and frequency corresponding to the ith median point are respectively used as parameter estimation of the ith component signal, that is:
Ai＝A{|Y(ω)|}
wherein, the amplitude directly takes the module of the complex data to obtain the amplitude AiFrequency omega can be obtained from wave number, seawater permeability and dielectric constant of radar signaliTaking tangent function to the ratio of imaginary part to real part can obtain phase
A represents amplitude operation, mu represents seawater permeability, epsilon represents dielectric constant, and kiTo representThe wave number of the radar echo signal.
3. Judgment of
First, the amplitude A of the ith component signal is measurediIf the ith component signal amplitude A is compared with the average sea clutter poweriLarger, the retention, lower, the corresponding deletion from the original time domain sequence,
then, for the retained component signals, the phase of the ith component signal is required to be adjusted
And radar echo signals
Is compared, if
The component correspondence is removed from the original time domain sequence and the process is repeated for the remaining signal until it is time to remove the component correspondence from the original time domain sequence
The component signal is considered to correspond to a ship target,
then, the frequency ω of the i-th component signal is also measured simultaneouslyiWith frequency omega of radar echo signals0Making a comparison if omegai≠ω0Then the component signal is correspondingly deleted from the original time domain sequence,
and finally, if the phase and the frequency of the ith component signal are equal, the component signals are kept to be continuously compared, and the comparison is carried out until the power of a median point in the clutter area is smaller than the average clutter power, so that the sea clutter suppression is completed.
In this way, the sea clutter is identified and eliminated, and the suppression of the sea clutter is completed.
According to the method, a plurality of frequency windows are divided for the radar echo according to the frequency range of the first-order sea clutter, wherein the frequency range of the first-order sea clutter is 0-8 Hz. The way to divide the frequency bin for the radar echo is as follows: determining a frequency h in the frequency range of the first-order sea clutter, wherein the radar echo is w, and the frequency windows divided by the radar echo are (w-h) - (w + h).
In order to verify the superiority of the method of the invention, the measured data recorded in a certain experimental field is adopted for verification analysis:
the radar is placed parallel to the coast with the axis pointing 105 deg. in the true north direction, and the scans cover 15 deg. and 155 deg. in the true north direction. The measured data for a certain period of time is 256 frames, (i.e. one sweep period of 0.65s) and comprises 2560, and each channel comprises 80 data. For a total of 16 channels 1280 data. Therefore, each data sample represents the synthesis of the radiation energy of points at all positions on the circumference with the ith antenna element as the center and the 5km as the increasing radius, as shown in fig. 1, and fig. 1 is a schematic diagram of the sampling points and the distance elements in the embodiment of the present invention. Specifically, for comparison and analysis, the measured radar echo data of 13 th, 17 th, 19 th and 37 th range bins are processed, and the results before and after the first-order sea clutter suppression processing are mainly compared and analyzed. Firstly, denoising and suppressing the sea clutter by using a two-dimensional median filtering algorithm, selecting reference cancellation unit data, wherein the data consists of average data of 8 adjacent distance units, the data is used as a subtraction number, 13 th distance metadata is used as a subtraction number, the subtraction number and the subtraction number are subtracted, and then the amplitude is compensated to the difference of echo data, so that the data after the sea clutter suppression can be obtained. In the detection process, the distance dimension and Doppler dimension reference units are respectively set as a left sampling unit and a right sampling unit, the distance dimension and Doppler dimension constant false alarm detection reference units are set as a left unit and a right unit of the detection unit respectively, and the left unit and the right unit are taken as protection units respectively, so that higher detection probability and lower false alarm probability are obtained. Test results show that when the threshold is 11dB, after data are subjected to median filtering after two FFT in four different time periods, the density of salt-pepper noise can be actually controlled by the threshold of a constant false alarm. When the threshold value is 7dB, the threshold setting is reduced, and the more salt and pepper noise points are, the main body of the false alarm is formed. The "false alarm rate" is related to the threshold of the noise level. The higher the threshold, the lower the false alarm rate. After Gumbel distributed double thresholds (7dB and 11dB) are adopted, the comparison and analysis of the overall consistent effect can show that the first-order peak, the second-order peak and the island ground clutter of the sea clutter are greatly eliminated, and the target is obviously shown. After median filtering and different distance element clutter cancellation are carried out by utilizing the correlation of the sea clutter, the first-order sea clutter suppression performance is greatly improved.
The method has a very small amount of loop iteration, effectively avoids the defect of a sorting algorithm with a loop iteration structure, eliminates the uncertainty of time delay from input to output, occupies small memory, can effectively protect boundary information, has low requirement on hardware and is easy to realize in engineering.
The invention aims to recover the probability level under the noise background as much as possible, thereby improving the first-order sea clutter suppression performance and improving the target detection probability. The high-frequency radar sea clutter suppression method has feasibility after the radar antenna is miniaturized, can improve the signal-to-clutter ratio to a certain extent, and can further improve the detection performance of offshore sea surface ship targets after the radar antenna is miniaturized.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A sea clutter suppression method is characterized by comprising the following steps:
s1: firstly, the time domain sequence of the radar echo is Fourier transformed to the frequency domain, then the Fourier transform is carried out again to the Doppler frequency spectrum,
s2: dividing a frequency window for the radar echo, corresponding to multiple frequency windows, finding the median point, med (A) of the radar echo in each frequency windown[xij]) Representation versus frequency window An[xij]All points in the interior are taken as the median value, yijIs a pixel point xijThe two-dimensional median filtered output value is:
wherein the content of the first and second substances,
the gray value of the image point at the (N +1) th position in the frequency window, N is a positive integer, N represents the point number of the frequency window,
get dij＝|xij-yij|，dijIndicating the gray value at (i, j) in the detail image of the Doppler spectrogram, wherein i and j are the ith row and the jth column of the detail image of the Doppler spectrogram, i and j are natural numbers,
s3: if d isijIf the value is larger than the threshold T, the point (i, j) of the corresponding position is considered as the point of the sea clutter, and otherwise, the point without the sea clutter at the corresponding position (i, j) is considered.
2. The method for suppressing sea clutter according to claim 1, wherein in step S1, in performing fourier transform, for the time domain sequence data of radar returns with higher signal to noise ratio, a longer FFT formula is used:
wherein i and j are natural numbers, N is the total length of the time domain sequence of the radar echo with higher signal-to-noise ratio, and xi,j(n) refers to time domain data of the radar echo signal of the nth row, k refers to the kth frequency domain, and the higher signal-to-noise ratio refers to the signal-to-noise ratio of 10dB to 50 dB.
3. The method of claim 2, wherein in step S1, in performing the fourier transform, a shorter FFT formula is used for the time domain sequence data of the radar echo with a lower signal-to-noise ratio:
wherein the content of the first and second substances,i. j is a natural number, M is the total length of the time domain sequence of the radar echo with lower signal-to-noise ratio, xi,j(n) refers to time domain data of the radar echo signal of the nth row, k refers to the kth frequency domain, and the lower signal-to-noise ratio refers to the signal-to-noise ratio of-20 dB to 10 dB.
4. The method for suppressing sea clutter according to claim 3, wherein in step S2, A is located on the two-dimensional range-Doppler spectrumn[xij]It means that a window operation of N × N-2N +1 is performed on a point in the image with the point (i, j) as the center, where N is an odd integer.
5. The sea clutter suppression method of claim 4, wherein in step S2, the detail image of the Doppler spectrogram is represented by D,
D＝|Y(k)-S(k,θ)|
y (k) is obtained by converting Y (ω), and the relationship between k and ω satisfies the following equation relationship ω 2 pi k,
wherein Y (ω) ═ med (X (ω)),
where N (p) is a normalization factor, p is a spreading factor, and is set to 15, θ0Is the wind direction, theta is the observation direction, the wave spectral function s (k).
6. The method of claim 3, wherein in step S2, the radar echo is divided into a plurality of frequency windows according to the Doppler frequency range of the first-order sea clutter, wherein the frequency range of the first-order sea clutter is 0-8 Hz.
7. A method of sea clutter suppression according to claim 4, wherein the frequency windowing is performed on the radar echo by: determining a frequency h in the Doppler frequency range of the first-order sea clutter, wherein the radar echo is w, and the frequency windows divided by the radar echo are (w-h) to (w + h).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011062961.7A CN112255607B (en) | 2020-09-30 | 2020-09-30 | Sea clutter suppression method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011062961.7A CN112255607B (en) | 2020-09-30 | 2020-09-30 | Sea clutter suppression method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112255607A true CN112255607A (en) | 2021-01-22 |
| CN112255607B CN112255607B (en) | 2022-06-07 |
Family
ID=74234916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011062961.7A Active CN112255607B (en) | 2020-09-30 | 2020-09-30 | Sea clutter suppression method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112255607B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112946654A (en) * | 2021-01-27 | 2021-06-11 | 中国人民解放军国防科技大学 | Method, device, computer system and storage medium for radar filtering dynamic clutter |
| CN114755654A (en) * | 2022-06-14 | 2022-07-15 | 中达天昇(江苏)电子科技有限公司 | Damaged radar signal restoration method based on image mimicry technology |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014044193A (en) * | 2012-07-31 | 2014-03-13 | Mitsubishi Electric Corp | Clutter suppressing device |
| CN103885044A (en) * | 2014-03-31 | 2014-06-25 | 西安电子科技大学 | Method for suppressing clutter and noise of narrow-band radar echoes based on CLEAN algorithm |
| CN105388465A (en) * | 2015-12-17 | 2016-03-09 | 西安电子科技大学 | Sea clutter simulation method based on sea wave spectrum model |
| CN106249211A (en) * | 2016-08-05 | 2016-12-21 | 中国电子科技集团公司第二十八研究所 | A kind of sea clutter and sexual intercourse Clutter suppression algorithm |
| CN107678003A (en) * | 2017-09-15 | 2018-02-09 | 国家海洋局第海洋研究所 | Object detection method and device under a kind of ground wave radar sea clutter background |
| CN109116326A (en) * | 2018-09-27 | 2019-01-01 | 中国科学院电子学研究所苏州研究院 | A kind of adaption radar ocean clutter cancellation method based on medion estimator |
| CN109782251A (en) * | 2019-03-14 | 2019-05-21 | 北京航空航天大学 | A kind of slower-velocity target discrimination method after ocean clutter cancellation |
| CN111624573A (en) * | 2020-07-20 | 2020-09-04 | 上海无线电设备研究所 | Time domain self-adaptive target detection method under sea clutter background |
-
2020
- 2020-09-30 CN CN202011062961.7A patent/CN112255607B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014044193A (en) * | 2012-07-31 | 2014-03-13 | Mitsubishi Electric Corp | Clutter suppressing device |
| CN103885044A (en) * | 2014-03-31 | 2014-06-25 | 西安电子科技大学 | Method for suppressing clutter and noise of narrow-band radar echoes based on CLEAN algorithm |
| CN105388465A (en) * | 2015-12-17 | 2016-03-09 | 西安电子科技大学 | Sea clutter simulation method based on sea wave spectrum model |
| CN106249211A (en) * | 2016-08-05 | 2016-12-21 | 中国电子科技集团公司第二十八研究所 | A kind of sea clutter and sexual intercourse Clutter suppression algorithm |
| CN107678003A (en) * | 2017-09-15 | 2018-02-09 | 国家海洋局第海洋研究所 | Object detection method and device under a kind of ground wave radar sea clutter background |
| CN109116326A (en) * | 2018-09-27 | 2019-01-01 | 中国科学院电子学研究所苏州研究院 | A kind of adaption radar ocean clutter cancellation method based on medion estimator |
| CN109782251A (en) * | 2019-03-14 | 2019-05-21 | 北京航空航天大学 | A kind of slower-velocity target discrimination method after ocean clutter cancellation |
| CN111624573A (en) * | 2020-07-20 | 2020-09-04 | 上海无线电设备研究所 | Time domain self-adaptive target detection method under sea clutter background |
Non-Patent Citations (5)
| Title |
|---|
| CAN LI 等: "Sea Land Clutter Recognition for Over The Horizon Radar via Deep CNN", 《2019 INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND INFORMATION SCIENCES (ICCAIS)》 * |
| 李健 等: "结合全局和随机局部频率调谐的复杂纹理表面缺陷检测", 《陕西科技大学学报(自然科学版)》 * |
| 温幼辉: "海防雷达仿真系统杂波抑制算法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王炜鹏 等: "采用改进型时频滤波的海杂波抑制方法", 《信号处理》 * |
| 王祎鸣 等: "高频地波雷达海杂波抑制的时频-矩阵联合法", 《中国海洋大学学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112946654A (en) * | 2021-01-27 | 2021-06-11 | 中国人民解放军国防科技大学 | Method, device, computer system and storage medium for radar filtering dynamic clutter |
| CN112946654B (en) * | 2021-01-27 | 2023-06-23 | 中国人民解放军国防科技大学 | Method, device, computer system and storage medium for filtering dynamic clutter by radar |
| CN114755654A (en) * | 2022-06-14 | 2022-07-15 | 中达天昇(江苏)电子科技有限公司 | Damaged radar signal restoration method based on image mimicry technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112255607B (en) | 2022-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104569948B (en) | Sub-band adaptive GLRT LTD detection methods under sea clutter background | |
| CN112255607B (en) | Sea clutter suppression method | |
| CN111624574A (en) | Target detection method, system, storage medium and device for weak target detection | |
| CN109212500A (en) | A kind of miscellaneous covariance matrix high-precision estimation method of making an uproar of KA-STAP based on sparse reconstruct | |
| CN106772352A (en) | A kind of PD radars extension Weak target detecting method based on Hough and particle filter | |
| CN109581516B (en) | Denoising method and system for data of curvelet domain statistic adaptive threshold value ground penetrating radar | |
| CN108983287B (en) | Curvelet transform anti-aliasing seismic data reconstruction method based on convex set projection algorithm | |
| Hu et al. | A new way to model nonstationary sea clutter | |
| CN113887398A (en) | GPR signal denoising method based on variational modal decomposition and singular spectrum analysis | |
| CN107229040B (en) | high-frequency radar target detection method based on sparse recovery space-time spectrum estimation | |
| CN108414992A (en) | A kind of object detection method based on phase information clutter map | |
| CN110888133A (en) | V frequency modulation signal ISAR sparse imaging method under low signal-to-noise ratio condition | |
| CN108896971B (en) | Simulation method for echoes of small targets floating on sea surface | |
| CN103885044B (en) | A kind of miscellaneous suppressing method of making an uproar of Narrow-band Radar echo based on CLEAN algorithm | |
| CN113406634B (en) | Spatial high-speed spinning target ISAR three-dimensional imaging method based on time domain phase matching | |
| CN114296046B (en) | HFSWR multi-sea-condition effective wave height extraction method and device based on artificial neural network | |
| CN116819432A (en) | Single-vector hydrophone underwater multi-target high-stability direction finding method and system based on characteristic spectrum tracking | |
| CN115561764A (en) | Moving target depth estimation method based on single-vector hydrophone | |
| CN115032601A (en) | Marine radar target detection algorithm for inhibiting sea clutter in image sequence based on space-time combined filtering technology | |
| CN115113208A (en) | Continuous wave radar clutter cancellation method based on accurate clutter feature recognition | |
| CN104463325A (en) | Noise suppression method for polar ice-penetrating radar original data | |
| Juan et al. | A new Wavelet Prediction method for GPR clutter elimination Based on LSTM network | |
| CN113156392A (en) | Clutter suppression method based on pitching domain self-adaptive processing | |
| CN112684425A (en) | Target secondary discrimination method after constant false alarm detection | |
| Shang et al. | Stationary time statistical property of ionospheric clutter in high-frequency surface-wave radar |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |