A method for ground clutter recognition and weather echo restoration in dual-polarization weather radar

Abstract

The present invention relates to a method for identifying ground clutter and restoring meteorological echoes of a dual-polarization meteorological radar, which belongs to the field of radar detection, and includes: receiving time domain I / Q data collected by a radar in real time, and calculating identification factors based on radial I / Q data to identify ground clutter; for a distance library I / Q identified as ground clutter, adopting a time domain or frequency method to suppress ground objects, directly assigning values ​​to unidentified data, obtaining I / Q data before and after filtering, and performing basic product calculations and reflectivity texture calculations before and after filtering; judging whether ground clutter is insufficiently filtered or mistakenly filtered, and restoring the filtered product output of the distance library identified as mistakenly filtered using unfiltered products. The present invention solves the data quality problems caused by residuals due to insufficient filtering of most ground objects, and meteorological losses due to mistaken filtering of some meteorological echoes, and greatly improves the data quality compared to the traditional full-process filtering or CMD method identification or the method using ground clutter maps.

Description

Technical Field

[0001] The invention relates to the field of radar detection, and in particular to a method for identifying ground clutter and restoring meteorological echoes of a dual-polarization meteorological radar. Background Art

[0002] Doppler weather radar is a meteorological detection equipment that detects current weather conditions by radio. It generally has two detection forms: single polarization or dual polarization. The echo scanned by the radar can reflect whether there is rainfall or possible rainfall in the current weather. Single polarization detection radar generally outputs basic reflectivity, basic velocity and basic spectrum width products. Dual polarization detection radar adds differential reflectivity factor, differential phase, zero delay correlation coefficient and differential phase shift rate products on the basis of single polarization radar. Based on the basic products, it can further identify and classify weather echoes, carry out precipitation inversion and calculate weather radar QPE and QPF products. Ground clutter is manifested as the ground object echo superimposed on the meteorological echo on the radar echo map detected by the Doppler weather radar, which is regarded as an interference echo and will affect the detection and identification of weather echoes.

[0003] Ground clutter filtering requires first identifying whether it is a ground object, and then filtering the ground object. Traditional methods use full-range filtering or CMD methods to identify ground objects or use ground clutter map methods, which may cause some meteorological echoes to be mistakenly filtered out or not filtered out cleanly. Therefore, how to use the product characteristics of dual-polarization radar to filter ground clutter and reduce the impact of ground clutter on meteorological echoes is a problem that needs to be considered at present. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art, provide a dual-polarization weather radar ground clutter recognition and weather echo restoration method, and solve the shortcomings of the prior art.

[0005] The object of the present invention is achieved by the following technical scheme: a method for identifying ground clutter and restoring meteorological echoes of a dual-polarization meteorological radar, the method comprising:

[0006] Step 1: Receive the time domain I / Q data collected by the radar in real time, and calculate the identification factor based on the radial I / Q data to identify the ground clutter;

[0007] Step 2: For the distance library I / Q marked as ground clutter, the time domain or frequency method is used to suppress the ground clutter, and the unmarked data is directly assigned to obtain the I / Q data before and after filtering, and the basic product calculation and reflectivity texture calculation before and after filtering are performed;

[0008] Step 3: Determine whether the ground clutter is insufficiently filtered or mistakenly filtered out, and restore the filtered product output using the unfiltered product from the distance library marked as mistakenly filtered.

[0009] The step 1 specifically includes the following contents:

[0010] Receive time domain I / Q data collected by radar in real time;

[0011] Based on the radial I / Q data, the original echo power, the sudden change of echo power in radial distance, and the clutter phase are calculated to determine whether the square mean value of the echo power and the phase conform to the phase characteristics of the ground object, and the distance library that meets the conditions is identified as ground clutter.

[0012] The qualified distance library includes: if the original echo power of a distance library is greater than the noise power, the square mean of the echo power is greater than the set value, and the clutter phase is greater than the set value, then the distance library is identified as ground clutter.

[0013] The step 2 specifically includes the following contents:

[0014] For the distance library I / Q marked as ground clutter, the time domain or frequency method is used to suppress the ground clutter, and the unmarked data is directly assigned to obtain the I / Q data before and after filtering;

[0015] Based on the unfiltered and filtered I / Q data, the basic reflectivity, unfiltered reflectivity, signal-to-noise ratio, velocity, spectral width, differential reflectivity factor, differential phase, zero-delay correlation coefficient are calculated, and the basic reflectivity radial texture and unfiltered reflectivity radial texture are calculated.

[0016] The determination of whether there is any missing or unfiltered ground clutter specifically includes the following contents:

[0017] The distance library that has not been filtered for ground clutter suppression is judged by the single radial distance library loop judgment method. If the signal-to-noise ratio of the current distance library is less than 5dB and the zero-delay correlation coefficient is less than 0.25, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. When the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated.

[0018] If the signal-to-noise ratio of the current distance library is less than 3dB and the zero-delay correlation coefficient is less than 0.9, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. When the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated.

[0019] The determination of whether ground clutter is misfiltered specifically includes the following:

[0020] The distance library that does not perform ground clutter suppression filtering is judged by the single radial distance library cyclic judgment method. When the zero-delay correlation coefficient of a filtered distance library and the unfiltered zero-delay correlation coefficient are both greater than 0.98, the standard deviation of the unfiltered reflectivity of the five distance libraries before and after is calculated. When the standard deviation of the unfiltered reflectivity is less than 3dB, the filtered distance library is marked as an incorrectly filtered distance library.

[0021] The present invention has the following advantages: a dual-polarization meteorological radar ground clutter identification and meteorological echo restoration method solves the data quality problems caused by residual due to insufficient filtering of most ground objects and meteorological loss due to erroneous filtering of some meteorological echoes. Compared with the traditional full-process filtering or CMD method identification or the method using ground clutter map, the data quality is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the process of the present invention;

[0023] Figure 2 It is a schematic diagram of real-time radial I / Q data arrangement;

[0024] Figure 3 This is a schematic diagram of the analysis of filtering indicators of the traditional method and the new method for a pure meteorological echo case;

[0025] Figure 4 Schematic diagram of filtering index analysis of traditional and new methods for a case of meteorological overlay objects. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the drawings is not intended to limit the scope of protection of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application. The present invention is further described below in conjunction with the drawings.

[0027] like Figure 1 As shown, the present invention specifically relates to a method for identifying ground clutter and restoring meteorological echoes of a dual-polarization meteorological radar. The method can effectively filter out ground object echoes while reducing the impact of filtering on meteorological echoes, and specifically includes the following contents:

[0028] S1. First object recognition: Real-time reception of time domain I / Q data collected by the radar, based on Figure 2 The real-time radial I / Q data shown in the figure calculates the identification factors, which mainly include the original echo power T, the sudden change of echo power in radial distance (9-point echo power square mean value )、clutter phase , determine whether the echo power changes suddenly and whether the phase conforms to the phase characteristics of the ground object, and the distance library that meets the conditions is marked as ground clutter.

[0029] Furthermore, the original echo power T is calculated as follows:

[0030] ,

[0031] in, , are the real and imaginary parts of the complex time series, i is the pulse number, and N is the total number of pulses.

[0032] Use the first 4 points and the last 4 points of the current distance library, a total of 9 points, to calculate the square mean of the echo power The calculation formula is as follows:

[0033] ,

[0034] in, is the square mean of the echo power of the current distance library, is the echo power of the kth range bin.

[0035] Clutter Phase The calculation formula is as follows:

[0036] ,

[0037] in, is a complex time series, , j is the imaginary unit.

[0038] Identification conditions: original echo power T> noise power, echo power square mean , clutter phase If it is greater than 0.7, the three conditions are met at the same time, and the distance library is marked as ground clutter.

[0039] S2. Ground clutter suppression filtering: For the distance library I / Q data marked as ground clutter, the time domain or frequency method is used to suppress the ground objects, and the distance library I / Q data not marked as ground clutter is directly assigned to obtain the I / Q data before and after filtering.

[0040] S3. Calculation of basic products before and after filtering, and calculation of product texture: Based on the unfiltered and filtered I / Q data, calculate the two channels, basic reflectivity, unfiltered reflectivity, signal-to-noise ratio, speed, spectral width, differential reflectivity factor, differential phase, zero-delay correlation coefficient, as well as basic reflectivity texture and unfiltered reflectivity texture.

[0041] Furthermore, the calculation of basic products uses standard weather radar formulas. The difference is that it is necessary to calculate the basic products based on the pre-filtering I / Q output of S1, the basic products based on the post-filtering I / Q output of S1, and the radial texture of the basic reflectivity before and after filtering. The texture characteristics are reflected by calculating the standard deviation of the five libraries before and after calculation.

[0042] Furthermore, the reflectivity before and after filtering is calculated as:

[0043] ,

[0044] ,

[0045] Where R is the I / Q complex time series The autocorrelation of R is calculated, and the logarithm of R is used to obtain the basic reflectivity dB. express The conjugate of is the I / Q complex time series before filtering, then the output is the reflectivity dB before filtering. is the I / Q complex time series after filtering, and the output is the reflectivity dB after filtering.

[0046] The reflectivity texture (5-point standard deviation) before and after filtering of the same distance library is calculated as follows:

[0047] ,

[0048] Among them, std is the standard deviation, and m is the sequence number of the five distance libraries before and after the current distance library.

[0049] The zero-delay correlation coefficient is calculated as:

[0050] ,

[0051] in , is the cross-correlation result of the complex time series of the horizontal and vertical channels with zero delay, , It is the autocorrelation result of the complex time series of the zero-delay horizontal and vertical channels.

[0052] The signal-to-noise ratio SNR is calculated as:

[0053] ,

[0054] Wherein, Noise is the noise power.

[0055] S4, judging whether the ground clutter is insufficiently filtered or the meteorological echo is mistakenly filtered out: judging the mistakenly filtered distance library and the omitted unfiltered distance library by means of a single radial distance library cycle;

[0056] Specifically include: processing of omitted unfiltered distance library: the distance library that does not perform the ground clutter suppression filter of S2 is judged as follows:

[0057] If the signal-to-noise ratio of the current distance library is less than 5dB and the zero-delay correlation coefficient is less than 0.25, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. If the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated.

[0058] If the signal-to-noise ratio of the current distance library is less than 3dB and the zero-delay correlation coefficient is less than 0.9, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. If the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated.

[0059] Processing of mis-filtered distance library: The distance library that has executed the ground clutter suppression algorithm makes the following judgment: when the zero-delay correlation coefficient and the unfiltered zero-delay correlation coefficient of a filtered distance library are both greater than 0.98, the unfiltered reflectivity standard deviation of the five distance libraries before and after is calculated. When the unfiltered reflectivity standard deviation is less than 3dB, the filtered distance library is marked as a mis-filtered distance library.

[0060] S5. The omitted unfiltered distance library is marked as invalid, and the filtered reflectivity is output invalidly, and the unfiltered reflectivity remains unchanged. The distance library marked as misfiltered uses the unfiltered product to restore the filtered product output.

[0061] The data quality is evaluated by calculating the average deviation and standard deviation of the basic reflectivity before and after filtering, and the correlation coefficient between the filtered and unfiltered reflectivity. The smaller the average deviation and standard deviation, the smaller the impact of the filtering process on the meteorological echo. The higher the correlation coefficient between the filtered and unfiltered reflectivity, the smaller the impact of the filtering process on the meteorological echo.

[0062] Figure 3 , Figure 4 The effects of the traditional identification filtering method and the filtering method on the meteorological echo before and after filtering are respectively shown. Figure 3 This is an example of a pure weather echo (without mixed ground echo). Figure 3Where a is the basic reflectivity product, b is the basic velocity product, c is the deviation of the basic reflectivity before and after filtering after traditional object recognition and suppression, with an average deviation of 0.046dB, a standard deviation of 0.455dB, and a correlation coefficient of 0.998; d is the deviation of the basic reflectivity before and after filtering after object recognition and suppression by this method, with an average deviation of 0.015dB, a standard deviation of 0.282dB, and a correlation coefficient of 0.999. d reflects the difference between the reflectivity before filtering and the reflectivity after filtering. It can be seen that the present invention reduces spot echoes, and the average deviation of echoes is reduced by 0.031dB, the standard deviation is reduced by 0.173dB, and the correlation coefficient is increased by 0.001.

[0063] Figure 4 This is an example of a weather echo mixed with ground object echo. Figure 4 In the figure, a is the basic reflectivity product, b is the basic velocity product, c is the deviation of the basic reflectivity before and after filtering after traditional object recognition and suppression, with an average deviation of 0.087dB, a standard deviation of 0.560dB, and a correlation coefficient of 0.998; d is the deviation of the basic reflectivity before and after filtering after the new method of object recognition and suppression, with an average deviation of 0.052dB, a standard deviation of 0.437dB, and a correlation coefficient of 0.999. d reflects the difference between the reflectivity before and after filtering. It can be seen that this method reduces spot echoes, and the average deviation of echoes is reduced by 0.035dB, the standard deviation is reduced by 0.123dB, and the correlation coefficient is increased by 0.001.

[0064] The results of the two case data analysis both show that the method adopted by the present invention reduces the impact of ground object filtering on meteorological echoes, and the data quality is improved.

[0065] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used for various other combinations, modifications and improvements, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for identifying ground clutter and restoring meteorological echoes using a dual-polarization weather radar, characterized in that: The method comprises: Step 1: Receive the time domain I / Q data collected by the radar in real time, and calculate the identification factor based on the radial I / Q data to identify the ground clutter; Step 2: For the distance library I / Q marked as ground clutter, the time domain or frequency method is used to suppress the ground clutter, and the unmarked data is directly assigned to obtain the I / Q data before and after filtering, and the basic product calculation and reflectivity texture calculation before and after filtering are performed; Step 3: determine whether there is any missing or mis-filtered ground clutter, mark the missing and mis-filtered distance library as invalid, and restore the filtered product output using the unfiltered product for the distance library marked as mis-filtered; The determination of whether there is any missing or unfiltered ground clutter specifically includes the following contents: The distance library that has not been filtered for ground clutter suppression is judged by the single radial distance library loop judgment method. If the signal-to-noise ratio of the current distance library is less than 5dB and the zero-delay correlation coefficient is less than 0.25, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. When the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated. If the signal-to-noise ratio of the current distance library is less than 3dB and the zero-delay correlation coefficient is less than 0.9, the reflectivity standard deviation of the five distance libraries before and after the current distance library is calculated. If the reflectivity standard deviation is greater than 3dB, the current distance library is ground clutter and is invalidated. The determination of whether ground clutter is misfiltered specifically includes the following: The distance library that does not perform ground clutter suppression filtering is judged by the single radial distance library cyclic judgment method. When the zero-delay correlation coefficient of a filtered distance library and the unfiltered zero-delay correlation coefficient are both greater than 0.98, the unfiltered reflectivity standard deviation of the five distance libraries before and after is calculated. When the unfiltered reflectivity standard deviation is less than 3dB, the filtered distance library is identified as an erroneously filtered distance library, and the unfiltered product data is replaced by the filtered product data.

2. The method for dual-polarization weather radar ground clutter identification and weather echo restoration according to claim 1, characterized in that: The step 1 specifically includes the following contents: Receive time domain I / Q data collected by radar in real time; Based on the radial I / Q data, the original echo power, the sudden change of echo power in radial distance, and the clutter phase are calculated, and the size and phase of the square mean of the echo power are judged to determine whether they meet the phase characteristics of the ground objects, and the distance library that meets the conditions is identified as ground clutter.

3. The method for dual-polarization weather radar ground clutter identification and weather echo restoration according to claim 2, characterized in that: The qualified distance library includes: if the original echo power of a distance library is greater than the noise power, the square mean of the echo power is greater than the set value, and the clutter phase is greater than the set value, then the distance library is identified as ground clutter.

4. The method for dual-polarization weather radar ground clutter identification and weather echo restoration according to claim 1, characterized in that: The step 2 specifically includes the following contents: For the distance library I / Q marked as ground clutter, the time domain or frequency method is used to suppress the ground clutter, and the unmarked data is directly assigned to obtain the I / Q data before and after filtering; Based on the unfiltered and filtered I / Q data, the basic reflectivity, unfiltered reflectivity, signal-to-noise ratio, velocity, spectral width, differential reflectivity factor, differential phase, zero-delay correlation coefficient are calculated, and the basic reflectivity radial texture and unfiltered reflectivity radial texture are calculated.

Images