WO2015081519A1

WO2015081519A1 - Method and system for weather radar signal processing

Info

Publication number: WO2015081519A1
Application number: PCT/CN2013/088548
Authority: WO
Inventors: 李柏; 陈大任; 高玉春; 李喆; 王崇文
Original assignee: 中国气象局气象探测中心
Priority date: 2013-12-04
Filing date: 2013-12-04
Publication date: 2015-06-11

Abstract

A method and system for weather radar signal processing are provided. The method for weather radar signal processing includes the following steps: an emission step (S101) used for emitting main wave beams of radar waves to a target area through an antenna, wherein the main wave beams have a preset irradiation power, a preset wave-beam width and a preset number of pulses, and each pulse has a preset pulse width and a preset irradiation power; a receiving step (S102) used for receiving return waves from the target area through an antenna; and a return-wave processing step (S103) used for processing the return waves so as to obtain a weather radar signal that includes the return-wave power of the return waves and information related with the azimuth, pitch angle and distance of the target area, wherein the main wave beams are divided equally into n sub-wave beams according to the angle of the azimuth, and/or the main wave beams are divided equally into q sub-wave beams according to the angle of the pitch angle, and/or each pulse is divided equally into m sub-pulses according to the pulse width so as to explore the target area so that the weather radar signal of super resolution is obtained.

Description

Weather radar signal processing method and system

Technical field

The invention relates to the field of meteorological radar, in particular to a method and system for processing weather radar signals.

Background technique

Weather radar It is a kind of radar equipment used to detect the type of precipitation in the atmosphere (rain, snow, hail, etc.), distribution, movement and evolution, and to predict its future distribution and intensity. In recent years, with the change of climate environment, especially the increase of severe weather, the application of meteorological radar has become more and more important.

Figure 15 shows a schematic diagram of the meteorological radar detection target zone.

As shown in Figure 15, in a weather radar application, the location of any target in space can be determined by the following three coordinates:

(1) Distance R: the straight line distance from the radar to the target O P;

(2) azimuth angle α: the angle between the projection OB of the distance R on the horizontal plane and the horizontal plane in the true north direction;

(3) Pitch angle β: The angle between the distance R and its projection OB on the horizontal plane on the vertical plane, sometimes called the inclination angle or the high and low angle.

Currently, if you want to improve the resolution of the meteorological radar's detection signals (azimuth, elevation and distance), especially to obtain super-resolution detection signals, the traditional means is to improve the performance of the weather radar system hardware to achieve high resolution. Rate signal. That is, the traditional means is to reduce the beamwidth of the main beam of the radar wave and / Or pulse width to improve the resolution of the weather radar signal. However, due to limitations of current technical means, it is difficult to improve the resolution of the sounding signal by improving the performance of the weather radar hardware. This is because the resolution of the current meteorological radar signal depends on the beam width of the main beam of the radar wave and the pulse width of the pulse in the main beam. The beam width determines the azimuth and elevation angle of the target area detected by the meteorological radar. Resolution, while the pulse width determines the resolution of the distance of the target zone. For example, if the main beam of the radar wave of the meteorological radar has With a beamwidth of 3dB and a pulse within the main beam with a pulse width of 1ms, the resolution of the azimuth and elevation angle of the target zone is limited by the 3dB beamwidth and the resolution of the distance is limited to 1ms. The pulse width is therefore not improved.

Moreover, the existing weather radar signal processing system cannot process the information related to the azimuth, elevation angle and distance detected by the meteorological radar to obtain super-resolution high-precision data, and the super-resolution refers to The resolution of the processed data exceeds the resolution of the echo signal limited by beamwidth and pulse width.

technical problem

The meteorological radar signal processing method provided by the present invention is directed to a partial convolution result obtained by the meteorological radar detecting at least one of the azimuth angle, the elevation angle and the distance of the target area, that is, for the obtained partial convolution The resulting observations are processed to implement the deconvolution function and obtain super-resolution high-precision data.

Technical solution

The present invention provides a weather radar signal processing method and system.

In one aspect, an embodiment of the present invention provides a weather radar signal processing method, including the following steps: a transmitting step for transmitting a main beam of a radar wave to a target area through an antenna, the main beam having a predetermined illumination power, predetermined a beam width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse width and a predetermined illumination power; and a receiving step of receiving an echo from the target region through the antenna; And an echo processing step for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone; Wherein, the main beam is equally divided by azimuth angle n sub-beams to detect the target area to obtain an azimuth super-resolution weather radar signal, and/or to divide the main beam into pitch angle angles by q a sub-beam to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or to divide each of the pulses into pulse widths into m a sub-pulse to detect the target region to obtain a meteorological radar signal with a distance super-resolution, where n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is greater than or equal to 2 The integer.

According to an aspect of the invention, the main beam is equally divided by azimuth angle n a sub-beam to detect the target area, the target area is divided into a plurality of azimuth resolution elements, each azimuth angle corresponding to the azimuth angle is equal to one sub-beam width, and the n The sum of the illumination powers of the sub-beams is equal to the predetermined illumination power of the main beam, and the main beam is equally divided into pitch angle angles. a sub-beam to detect the target area, the target area is divided into a plurality of elevation angle resolution elements, and a pitch angle angle corresponding to each elevation angle resolution element is equal to one sub-beam width, and the q The sum of the illumination powers of the sub-beams is equal to the predetermined illumination power of the main beam, and each of the pulses is equally divided into pulses according to the pulse width. a sub-pulse to detect the target area, the target area is divided into a plurality of distance resolution elements, and the distance span of each distance resolution element corresponds to 1/2 of a pulse width of one sub-pulse, and The m The sum of the illumination powers of the sub-pulses is equal to the predetermined illumination power of each of the pulses.

According to an aspect of the invention, the main beam is equally divided by azimuth angle n In the case where a sub-beam is used to detect the target area, the n Each sub-beam sequentially illuminates each azimuth resolution element, and each azimuth resolution element has a reflection coefficient, and the reflection coefficient of each azimuth resolution element is the echo power of the beam irradiated to the azimuth resolution element. The ratio of the illumination power of the beam, In the case where the main beam is equally divided into q sub-beams by the pitch angle angle to detect the target area, the q Each of the sub-beams sequentially illuminates each of the elevation angle resolution elements, and each of the elevation angle resolution elements has a reflection coefficient, and the reflection coefficient of each of the elevation angle resolution elements is the echo power of the beam irradiated to the elevation angle resolution element. The ratio of the irradiation power of the beam is divided into equal divisions by pulse width Where m sub-pulses are used to detect the target zone, the m Each sub-pulse sequentially illuminates each distance resolution element, and each distance resolution element has a reflection coefficient, and the reflection coefficient of each distance resolution element is the echo power of the pulse irradiated to the distance resolution element and the irradiation of the pulse The ratio of power.

According to an aspect of the invention, the main beam is equally divided by azimuth angle n In the case of sub-beams to detect the target area, the echo processing step further includes: utilizing predetermined illumination power and echo power of the main beam and the n The illumination power of the sub-beams is used to calculate a reflection coefficient of one of the plurality of azimuth resolution elements, and the main beam is equally divided into a pitch angle angle. In the case of sub-beams to detect the target area, the echo processing step further includes: utilizing predetermined illumination power and echo power of the main beam and the q The illumination power of the sub-beams is used to calculate a reflection coefficient of one of the plurality of pitch angle resolution elements, and the pulse is equally divided into m by pulse width In the case where a sub-pulse is used to detect the target area, the echo processing step further includes: utilizing a predetermined illumination power and echo power of the pulse and the m The illumination power of the sub-pulses calculates a reflection coefficient of one of the plurality of distance-resolving elements.

According to an aspect of the invention, the step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements comprises: The sub-beams illuminate the echo power obtained by the same azimuth resolution element, and the step of calculating the reflection coefficient of one of the plurality of pitch angle resolving elements comprises: The sub-beams illuminate the echo power obtained by the same pitch angle resolution element, and the step of calculating the reflection coefficient of one of the plurality of distance resolution elements includes: The sub-pulses illuminate the echo power obtained by the same distance resolution element.

According to an aspect of the invention, the step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements comprises: when the n When the sub-beams are simultaneously irradiated to n azimuthal resolution elements, the n Calculating the plurality of pitch angle resolutions by dividing a sum of echo powers of the sub-beams by a predetermined illumination power of the main beam as an estimated initial value of a reflection coefficient of an azimuth resolution element where the main beam center line is located The step of reflecting the coefficient of one of the elements includes: when said When q sub-beams are simultaneously irradiated to q pitch angle resolution elements, use the q Calculating the plurality of distance resolution elements by dividing the sum of the echo powers of the sub-beams by the predetermined illumination power of the main beam as an estimated initial value of the reflection coefficient of the elevation angle resolution element of the main beam center line The step of one of the reflection coefficients includes: when When m sub-pulses are simultaneously irradiated to m distance-resolving elements, the m is used The sum of the echo powers of the sub-pulses divided by the predetermined illumination power of the pulses is used as an estimated initial value of the reflection coefficient of the distance resolution element in which the pulse width of the pulse is located.

According to an aspect of the invention, the step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements comprises: In the case where the reflection coefficient of the azimuth resolution element changes linearly along the azimuth angle within the predetermined beam width, the estimated initial value of the reflection coefficient of the azimuth angle resolution element of the main beam center line is equal to the azimuth angle resolution element The actual value of the reflection coefficient, the step of calculating the reflection coefficient of one of the plurality of pitch angle resolution elements, comprising: q In the case where the reflection coefficient of the pitch angle resolution element changes linearly along the pitch angle within the predetermined beam width, the estimated initial value of the reflection coefficient of the pitch angle resolution element where the main beam center line is located is equal to the pitch angle resolution element The actual value of the reflection coefficient, the step of calculating the reflection coefficient of one of the plurality of distance resolution elements includes: m In the case where the reflection coefficient of the distance resolution element changes linearly along the distance within the predetermined pulse width, the estimated initial value of the reflection coefficient of the distance resolution element where the pulse width of the pulse is located is equal to the reflection coefficient of the distance resolution element Actual value.

According to an aspect of the invention, the step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements comprises: The reflection coefficients of the azimuth resolution elements have a piecewise linear change and/or a sine wave variation along the azimuth angle within the predetermined beam width. Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. Pushing a calculation to obtain a change in the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor, and then correcting the estimated initial value with the correction factor to obtain a final estimated value of the reflection coefficient, The step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: The reflection coefficients of the q pitch angle resolvers exhibit a piecewise linear change and/or a sine wave change along the pitch angle within the predetermined beam width Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. Pushing a calculation to obtain a change in the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor, and then correcting the estimated initial value with the correction factor to obtain a final estimated value of the reflection coefficient, Calculating a reflection coefficient of one of the plurality of distance resolution elements includes: The reflection coefficients of the m distance resolution elements vary linearly and/or sinusoidally along the distance within the predetermined pulse width and / Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. A calculation is performed to obtain a change in the reflection coefficient caused by the pulse actually illuminating the target region, thereby obtaining a correction factor, and then the estimated initial value is corrected by the correction factor to obtain a final estimated value of the reflection coefficient.

According to an aspect of the invention, the step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements comprises: In the case where the reflection coefficients of the azimuth resolution elements vary randomly along the azimuth angle within the predetermined beam width, at least one level weighted summation operation is performed on the estimated initial values to calculate each stage weighted summation a change of the reflection coefficient before and after the calculation, and extrapolating the change of the reflection coefficient to obtain a change of the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor according to the estimated initial value and the The correction factor obtains a predicted value by using a prediction algorithm, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error And less than or equal to the preset error value, the predicted value is used as a final estimated value of the reflection coefficient of the azimuth resolution element, and if the comparison error is greater than the preset error value, according to the comparison error To correct the correction factor, the step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: q In the case where the reflection coefficient of the pitch angle resolution element varies randomly along the pitch angle within the predetermined beam width, at least one weighted summation operation is performed on the estimated initial value to calculate a weighted summation of each level. a change of the reflection coefficient before and after the calculation, and extrapolating the change of the reflection coefficient to obtain a change of the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor according to the estimated initial value and the The correction factor obtains a predicted value by using a prediction algorithm, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error And less than or equal to the preset error value, the predicted value is used as a final estimated value of the reflection coefficient of the pitch angle resolution element, and if the comparison error is greater than the preset error value, according to the comparison error To correct the correction factor, the step of calculating a reflection coefficient of one of the plurality of distance resolution elements includes: m In the case where the reflection coefficients of the distance resolution elements vary randomly along the distance within the predetermined pulse width, at least one level of weighted summation operation is performed on the estimated initial values to calculate each stage before and after the weighted summation operation a change in the reflection coefficient, an extrapolation calculation using the change in the reflection coefficient to obtain a change in the reflection coefficient caused by the pulse actually illuminating the target region, thereby obtaining a correction factor, which is utilized according to the estimated initial value and the correction factor The prediction algorithm obtains a predicted value, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error is less than or equal to Presetting the error value, the predicted value is used as a final estimated value of the reflection coefficient of the distance resolution element, and if the comparison error is greater than the preset error value, correcting the Correction factor.

According to an aspect of the invention, n is equal to a predetermined number of pulses within the main beam, and / or q Equal to the predetermined number of pulses within the main beam.

In another aspect, an embodiment of the present invention provides a weather radar signal processing system, including: a transmitter for transmitting a main beam of a radar wave to a target area through an antenna, the main beam having a predetermined illumination power, a predetermined beam a width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse width and a predetermined illumination power; a receiver for receiving an echo from the target zone through the antenna; An echo processor for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone; , dividing the main beam into azimuth angles n sub-beams to detect the target area to obtain an azimuth super-resolution weather radar signal, and/or to divide the main beam into pitch angle angles by q a sub-beam to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or to divide each of the pulses into pulse widths into m a sub-pulse to detect the target region to obtain a meteorological radar signal with a distance super-resolution, where n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is greater than or equal to 2 The integer.

Beneficial effect

The weather radar signal processing method and system described above have the following advantageous effects: the present invention performs deconvolution processing on observations (partial convolutions) of weather radars. Therefore, the beamwidth can be compressed to 1/n by azimuth (n It is an integer greater than or equal to 2, preferably the number of sub-beams or the number of pulses in the beam), and / or the beam width can be compressed to 1/q by the pitch angle (q is greater than or equal to 2 Integer, preferably the number of sub-beams or the number of pulses in the beam), and / or the pulse width can be compressed to 1/m by distance (m For the number of sub-pulses). In particular, the meteorological radar signal processing method and system according to an embodiment of the present invention can increase the azimuth resolution, the elevation angle resolution, and the range resolution of the meteorological target area by an order of magnitude, respectively, and can be based on beam width and pulse width. The volume resolution is reduced to one thousandth.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present invention, and other drawings can be obtained from those skilled in the art without any creative work.

1 is a flow chart of a weather radar signal processing method according to a first embodiment of the present invention.

figure 2 It is a schematic diagram of a convolution process in a weather azimuth radar signal processing method for an azimuth according to a second embodiment of the present invention (i.e., the response of the linear time invariant system is the superposition of the impulse response).

3 is a schematic diagram of a partial convolution process in a weather radar signal processing method for azimuth according to a second embodiment of the present invention.

4 is a graph showing an estimated initial value x _{0n of} a sample coefficient x _n and a normalized reflection coefficient of a reflection coefficient of each resolution element along azimuth in a piecewise linear (triangular waveform).

Figure 5 is a graph of sample values x _n and normalized estimated initial values x _0n for which the reflection coefficients of the respective resolution elements vary sinusoidally along the azimuth.

6 is a schematic diagram of a method of performing an estimation using an extrapolation search algorithm for estimating an initial value x _0n of a reflection coefficient according to a second embodiment of the present invention.

7 is an extrapolated search shown in FIG. 6 for the sample values x _n of the reflection coefficients of the respective resolution elements shown in FIG. 5 which vary sinusoidally along the azimuth angle and the estimated initial value x _0n shown in FIG. A diagram of the predicted values obtained after the algorithm is corrected.

Figure 8 is a graph of sample values x _n and normalized estimated initial values x _0n for which the reflection coefficients of the respective resolution elements are sinusoidally varied along the azimuth angle.

Figure 9 is an extrapolation of the sample values x _n of the reflection coefficients of the respective resolution elements shown in Figure 8 along the azimuthal sine wave and the estimated initial value x _0n shown in Figure 8 A diagram of the predicted values obtained by the search algorithm correction.

Figure 10 Is a schematic diagram of a method for processing reflection coefficients of resolution elements randomly varying by azimuth according to a second embodiment of the present invention, wherein the processing uses linear estimation, extrapolation search, and cyclic recursive algorithm.

Figure 11 (a) and Figure 11 (b An example of the result obtained by processing the acquired observation data of a certain weather radar by various algorithms according to the second embodiment of the present invention is shown.

Figure 12 Is a schematic diagram of a convolution process in a weather radar signal processing method for distance according to a fourth embodiment of the present invention (ie, the response of the linear time invariant system is the superposition of the impulse response).

Figure 13 is a schematic diagram of a partial convolution process in a weather radar signal processing method for distance according to a fourth embodiment of the present invention.

Figure 14 is a schematic block diagram of a weather radar signal processing system in accordance with a ninth embodiment of the present invention.

Figure 15 is a schematic illustration of a meteorological radar detection target zone in accordance with the prior art.

Embodiments of the invention

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

The present invention proposes a weather radar signal processing method and system that can implement super-resolution signals. The weather radar signal processing method and system are directed to azimuth, pitch angle, and/or Or target areas distributed along distance. The process by which the meteorological radar beam scans the target area to obtain echo power is a convolution process, or complete convolution. If the process of beam entry and exit from the target zone is removed (where the beam partially overlaps the target zone), the remainder is referred to as the 'partial convolution' process. The present invention analyzes this 'partial convolution' process and proposes that the beamwidth can be compressed to azimuth to 1/n (n is an integer greater than or equal to 2, preferably the number of sub-beams or the number of pulses in the beam), and the beamwidth is compressed to 1/q by the pitch angle (q is greater than or equal to 2 Integer, preferably the number of sub-beams or the number of pulses in the beam), and / or compress the pulse width to 1/m by distance (m Meteorological radar signal processing method and system for sub-pulse number). The present invention is capable of increasing the azimuth, elevation and range resolution by an order of magnitude, reducing the volume resolution based on beamwidth and pulse width to one thousandth.

First embodiment

The first embodiment of the present invention is a weather radar signal processing method 100 according to the present invention as shown in FIG.

The weather radar signal processing method 100 shown in FIG. 1 includes the following steps: transmitting step S101 And a main beam for transmitting a radar wave to the target area through an antenna, the main beam having a predetermined illumination power, a predetermined beam width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse Width and predetermined illumination power; receiving step S102, configured to receive an echo from the target area by using the antenna; and an echo processing step S103 Means for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone, wherein The main beam is equally divided by azimuth angle n a sub-beam to detect the target area to obtain an azimuth super-resolution weather radar signal, and/or to divide the main beam into pitch angles a sub-beam to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or to divide each of the pulses into pulse widths into m A sub-pulse is used to detect the target region to obtain a meteorological radar signal with a distance super-resolution, n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is greater than or equal to 2 The integer.

It should be noted that in the present invention, signal processing may be performed for at least one of azimuth, pitch angle and distance to obtain azimuth super-resolution weather radar signal and pitch angle super-resolution weather radar signal of the target area, respectively. And at least one of the super-resolution weather radar signals. That is, in accordance with the present invention, those skilled in the art can perform deconvolution processing only for one of azimuth, pitch angle, and distance to obtain a super-resolution weather radar signal of one of azimuth, elevation, and distance, respectively. It is also possible to process both of them to obtain super-resolution weather radar signals for both, and it is also possible to process all of them to obtain super-resolution weather radar signals for the three.

In the following, the invention is directed to azimuth, pitch angle and / respectively in different embodiments. Or the method of processing weather radar signals at a distance.

Second embodiment

The second embodiment is a weather radar signal processing method for azimuth based on the first embodiment.

2.1 Convolution and deconvolution of antenna azimuth scanning of meteorological radar

For the sake of discussion, the main beam of the radar wave of the meteorological radar and the target area of the observation are discretized. The beam width of the main beam is equally divided into n sub-beams by azimuth angle, where n is an integer greater than or equal to 2. Preferably, n is the number of pulses in the main beam. The beamwidth of the main beam can be set as needed, for example, can be set to 2.8 dB beamwidth, 2.9 dB beamwidth, 3 dB beamwidth, and the like. The beam width of each sub-beam is equal to 1/n of the beam width of the main beam, and the illumination power of each sub-beam is named h ₁ , h ₂ , h ₃ in the direction (for example, clockwise direction). . . h _n .

In the case where the main beam is equally divided into n sub-beams by azimuth angle to detect the target area, the target area is divided into a plurality of azimuth resolution elements (hereinafter referred to as resolution elements), each The azimuth angle corresponding to the resolution element is equal to one sub-beam width, and the illumination powers h ₁ , h ₂ , h _{3 of} the n sub-beams. . . The _sum of h _n is equal to the predetermined illumination power (or total illumination power) h of the main beam. Specifically, in the target area, starting from a certain direction (for example, a true north direction), the target area is equally divided into a plurality of resolution elements in a certain direction (for example, a clockwise direction), and each resolution element is correspondingly The azimuth angle is equal to the width of the sub-beam.

And, n The sub-beams sequentially illuminate each of the resolving elements, and each of the resolving elements has a reflection coefficient, and the reflection coefficient of each of the resolving elements is the ratio of the echo power of the beam irradiated to the resolving element to the illumination power. That is, the reflection coefficient of each resolution element is not only equal to the ratio of the echo power of each sub-beam irradiated to the resolution element to the illumination power, but is also numerically equal to the echo power of each sub-beam that is irradiated to the resolution element in turn. And the ratio of the total illumination power to the main beam. Moreover, the reflection coefficient of one of the plurality of resolution elements can be calculated using the total illumination power and the echo power of the main beam and the illumination power of each sub-beam.

In the present embodiment, the reflection coefficients of the respective resolution elements are sequentially named x ₁ , x ₂ , and x ₃ . . . Wait. When the antenna of the meteorological radar performs the azimuth scanning, the echo power obtained is the illumination power h _i of each sub-beam (where i is the sub-beam number, i=1, 2, 3 ... n) and the corresponding resolution element The result of the addition of the reflection coefficient x _j (where j is the resolution element number, j = 1, 2, 3 ...) is multiplied. That is, the process of obtaining the echo power by the antenna scanning target area is a convolution process.

From the point of view of system analysis, the reflection coefficient of each resolution element in the meteorological target area can be regarded as the input signal of the system. The illumination power of each sub-beam is regarded as the unit impulse response of the system, and the output of the receiver of the meteorological radar. Power (eg, video output power) is the output signal of this system. Since the illumination power of each sub-beam is fixed, this system is a linear time-invariant system with limited impulse response. In this physical model, the output signal and the unit impulse response are known, and the input signal, that is, the reflection coefficient of each resolver, is required. Therefore, the nature of the problem of solving the reflection coefficient of each resolution element is a deconvolution problem.

It is to be noted that the system referred to in this embodiment is not the weather radar signal processing system of the present invention, but the linear time invariant system.

Figure 2 illustrates the convolution process when the antenna is scanned in a certain direction (for example, clockwise) with five sub-beams as an example. According to the superposition principle of the linear system, the system response is the superposition of the response of each input signal. For discrete sequences, one input data is an impact, so the system response is the superposition of the shock response of each resolution element, as shown in Figure 2. In Fig. 2, the first row h ₁ , h ₂ , h ₃ , h ₄ and h ₅ represent the illumination power of the five sub-beams, and the second behavior is the reflection coefficients x ₁ , x ₂ , x _{3 of the} respective resolution elements. . . Wait. The first column is time t ₁ , t ₂ , t ₃ . . . Corresponding to the radar pulse repetition period, P ₃ , P ₄ , P ₅ , P ₆ and P _{7 of} the second column correspond to the output power of the receiver (for example, video output power). Each small square illuminates the echo power of the corresponding resolution element with h _i x _j on behalf of the sub-beam.

In FIG 2, the reflection coefficient of the first resolution element in x ₁ sequence of the h below ₅ x _1, h ₄ x _1, _{h. 3} x _1, h ₂ x ₁ and h ₁ x ₁ is the fifth sub-beam, the fourth sub The beam, the third sub-beam, the second sub-beam, and the first sub-beam respectively illuminate the echo power generated by the first resolution element, that is, the impulse response sequence of the first resolution element. The second resolution element of the reflection coefficient of x below ₂

sequence

_{_{_{h 5 x 2, h 4 x}}} 2, h 3 x 2, h 2 x 2 and h ₁ x ₂ is the fifth sub-beam, the fourth sub-beam, the third sub- The beam, the second sub-beam and the first sub-beam respectively illuminate the echo power generated by the second resolution element, that is, the impulse response sequence of the second resolution element. The sequence below the reflection coefficient of the remaining resolution elements is deduced by analogy.

The sequence of impulse response of two adjacent elements is different in time by one radar repetition period. The echo power P ₃ obtained at the time corresponding to the time t ₅ is that the first sub-beam, the second sub-beam, the third sub-beam, the fourth sub-beam, and the fifth sub-beam simultaneously illuminate the corresponding first resolution element, and the second The sum of the echo powers h ₁ x ₁ , h ₂ x ₂ , h ₃ x ₃ , h ₄ x ₄ and h ₅ x ₅ obtained by the resolution element, the 3rd resolution element, the 4th resolution element and the 5th resolution element. At this time, the center line of the main beam is located at the third resolution element (ie, coincides with the center line of the third resolution element), so it is named P ₃ . At time t ₆ , when each sub-beam simultaneously illuminates the second resolution element, the third resolution element, the fourth resolution element, the fifth resolution element, and the sixth resolution element, the obtained echo power P ₄ is equal to the echo power h _{1 .} The sum of x ₂ , h ₂ x ₃ , h ₃ x ₄ , h ₄ x ₅ and h ₅ x ₆ . The case of time t ₇ , t ₈ , t _{9 ,} etc. and so on.

The convolution process of the antenna scanning target area starts from the main beam entering the target area and ends when the main beam leaves the target area. This convolution process is called linear convolution or full convolution, or simply convolution. If the convolution process of beam entry and exit from the target zone (in which case the main beam partially overlaps the target zone) is omitted, the rest is referred to as the 'partial convolution' process. In Figure 2, starting at time t ₅ , the five sub-beams coincide with the five resolved elements, which is the beginning of the 'partial convolution' process. The partial convolution is a weighted summation operation, and the weights of the sub-beams are weighted and summed to the reflection coefficients of the target area to obtain an observation value, that is, an output power of the target area echo (for example, video output power). The present invention focuses on the problem of deconvolution of 'partial convolution'.

2.2 Deconvolution and beam compression

The meteorological radar usually performs (video) processing by taking the number of pulses in the beam, and the number of pulses in the beam is defined as the number of pulses transmitted by the main beam midline sweeping over a beam width, which is preferably equal to the number of sub-beams.

The partial convolution process with a pulse number equal to 5 in the beam is shown in Figure 3. The partial convolution process shown in Figure 3 takes five echo powers P ₃ , P ₄ , P ₅ , P ₆ and P ₇ for processing. Moreover, this partial convolution process involves a total of 9 resolution elements whose reflection coefficients are x ₁ , x ₂ , x ₃ , x ₄ , x ₅ , x ₆ , x ₇ , x ₈ and x _{9 , respectively} . Among the 9 resolution elements, only the 5th resolution element with reflection coefficient x ₅ is irradiated once by each sub-beam, and the remaining 8 resolution elements are incomplete illumination. In this case, the deconvolution process is performed with 5 echo powers, and only the reflection coefficient of one resolution element can be obtained. This is because only one resolution element x ₅ meets the conditions of full illumination (same as the actual scan in Figure 2) or beam compression, and incomplete illumination does not achieve the required beam compression. Therefore, with 5 echo powers, only one reflection coefficient of the resolution element can be obtained, which can be called partial deconvolution.

P ₃ from the extracted long h _₅ x _5, P ₄ are extracted from h ₄ x _5, h ₃ x ₅ extracted from _{P. 5} extracted from h ₂ x ₅ P _6, P is extracted from h ₁ ₇ x ₅ , the reflection coefficient x _{5 of} the 5th resolution element can be obtained. The formula is derived as follows. Equation (1) shows that the sum of the illumination powers h ₁ , h ₂ , h ₃ , h ₄ and h ₅ of each sub-beam is equal to the illumination power h of the main beam:

h=h ₁ +h ₂ +h ₃ +h ₄ +h ₅ -----( 1 )

Multiplying both sides of equation (1) by the reflection coefficient x _{5 of} the fifth resolution element yields equation (2):

Hx ₅ =h ₁ x ₅ +h ₂ x ₅ +h ₃ x ₅ +h ₄ x ₅ +h ₅ x ₅ -----( 2 )

The h on the left side of equation (1) represents the total illumination power when the main beam is at a predetermined beamwidth (for example, 3 dB beamwidth), and the sum on the right is the sum of sub-beam illumination powers h ₁ , h ₂ , h ₃ , h _{4 ,} and h ₅ . . Equation (2) embodies the concept of beam compression, because the right side is the sum of the echo powers of the 5th resolution elements whose reflection coefficients are x ₅ in turn, and the left side is the total illumination power of the main beam h. The echo power of the 5th resolution of x ₅ . Note that the total illumination power h of the main beam corresponds to a beamwidth equal to the sub-beamwidth. Since the total illumination power h on the left side of the equation (2) corresponds to the sub-beam width, if the sum of the echo powers of the fifth resolver with the reflection coefficient of x ₅ on the right side of the equation (2) can be obtained, the sum can be realized. Beam compression. If the illumination power and echo power of the main beam are normalized, that is, h=1, then equation (3) is obtained:

x ₅ =h ₁ x ₅ +h ₂ x ₅ +h ₃ x ₅ +h ₄ x ₅ +h ₅ x ₅ -----( 3 )

Formula (3) shows that: each sub-beam of the reflection coefficient of 5 x ₅ of each resolution element illuminated a normalized sum of echo power, equal to the resolution element of 5 x ₅ in the reflection coefficient value. Therefore, deconvolution processing or beam compression is to require the sum of the echo powers of the respective sub-beams to illuminate the same resolution element. Since the proportion of the echo power when the sub-beams are irradiated to the 5th resolution element in the corresponding observations P ₃ , P ₄ , P ₅ , P ₆ and P ₇ is unknown (see Figure 3), the constant to be determined is used. C means that we get the formula (4):

x ₅ =C ₁ P ₃ + C ₂ P ₄ + C ₃ P ₅ + C ₄ P ₆ + C ₅ P ₇ -----( 4 )

In formula (4), C ₁ P ₃ represents the extraction of h ₅ x ₅ from P ₃ , C ₂ P ₄ represents the extraction of h ₄ x ₅ from P ₄ , and C ₃ P ₅ represents the extraction of h ₃ x ₅ from P ₅ . , C ₄ P ₆ represents the extraction of h ₂ x ₅ from P ₆ , and C ₅ P ₇ represents the extraction of h ₁ x ₅ from P ₇ .

The following describes the specific way to solve the deconvolution problem.

2.3 Linear Estimation, Extrapolation Search and Cyclic Recursive Algorithm

Let's try to analyze from the airspace.

2.3.1 Linear Estimation Algorithm: Assume that the reflection coefficients of each resolution element change linearly along the azimuth

If the reflection coefficient varies linearly along the azimuth within one beam width, the relationship between the resolution elements x ₁ , x ₂ , x ₃ , x ₄ and x ₅ is as shown in equation (5):

x ₁ + x ₂ = x ₂ + x ₄ = 2x ₃ -----( 5 )

The relationship between the echo power of the main beam and the echo power of the sub-beam in Figure 3 is as shown in the following equation (6):

P ₃ = h ₁ x ₁ + h ₂ x ₂ + h ₃ x ₃ + h ₄ x ₄ + h ₅ x ₅ -----( 6 )

Moreover, according to the characteristics of the radar wave, the illumination powers h ₁ , h ₂ , h ₃ , h ₄ and h _{5 of the} respective sub-beams are symmetric with respect to the center line of the main beam, that is, h ₁ = h ₅ and h ₂ = h ₄ ,

The relationship between the reflection coefficients x ₁ , x ₂ , x ₃ , x ₄ and x _{5 of} the respective resolution elements and the relationship between the illumination powers h ₁ , h ₂ , h ₃ , h ₄ and h ₅ of the respective sub-beams Substituting (6) yields equation (7):

x ₃ = P ₃ /h -----( 7 )

That is, the observation value P ₃ is normalized by the total power of the main beam, and the reflection coefficient x _{3 of} the third resolution element is obtained. Similarly, by processing the echo powers P ₄ , P ₅ , P ₆ and P ₇ shown in Fig. 2 in the same way, the reflection coefficients x ₄ and the 5th resolution of the 4th resolution element can be obtained accordingly. The reflection coefficient x ₅ , the reflection coefficient x _{6 of the sixth} resolution element, and the reflection coefficient x _{7 of} the seventh resolution element.

In the present invention, when n sub-beams simultaneously illuminate n resolving elements, the value obtained by dividing the sum of the echo powers of the n sub-beams by the predetermined illumination power of the main beam is taken as the main beam center line. The estimated initial value of the reflection coefficient of the resolution element. In the case where the reflection coefficient x _{n of} the n resolution elements changes linearly along the azimuth angle within the predetermined beam width, the estimated initial value is equal to the actual value of the reflection coefficient of the resolution element, as in the above formula (7) ) shown. In this way, it is achieved that the width of the main beam is compressed to 1/n, thereby obtaining an azimuth super resolution.

Take the piecewise linear (triangular waveform) variation in Figure 4 as an example: the reflection coefficient sample value x _n (the dot shown in Figure 4) varies in a piecewise linear (triangular waveform) along the azimuth. The calculated value (estimated initial value x _0n ) (circle as shown in Figure 4) is well approximated in the linear portion.

2.3.2 Extrapolation search algorithm: Assume that the reflection coefficients of each resolution element are piecewise linear, sine wave and sine-like wave variation along azimuth.

The reflection coefficients of the n resolution elements are piecewise linearly varying along the azimuth and/or sinusoidal variations within a predetermined beamwidth and / Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. A calculation is performed to obtain a change in the reflection coefficient caused by the main beam actually illuminating the target region, thereby obtaining a correction factor, and then the estimated initial value is corrected by the correction factor to obtain a final estimated value of the reflection coefficient.

As shown in Fig. 4, in the linear portion of the triangular waveform, there is a linear relationship between the respective reflection coefficients, and in the inflection point portion of the triangular waveform, there is a wireless relationship between the respective reflection coefficients. In this case, the echo power is processed in the same way as in 2.3.1, and an estimated initial value sequence x _0n of the reflection coefficients x _n of the respective resolution elements is obtained, where n = 1...N, N is the processed Resolve the number of elements. For example, in the case of a wireless relationship between the respective reflection coefficients, the above equation (7) gives a value of x ₀₃ = P ₃ /h , that is, an estimated initial value of the reflection coefficient x ₃ of the third resolution element.

The process of obtaining x _0n from x _n can be regarded as a weighted summation operation, that is, using a normalized illumination function (ie, each sub-beam illumination power divided by the illumination power of the main beam) and 'partial convolution' with the input x _n , get the output x _0n . x _0n can be considered as an estimated initial value of the unknown sample sequence x _n of the reflection coefficient.

Figure 4 The process of estimating the initial value of the reflection coefficient at the inflection point of the piecewise linear (triangular waveform) change shown is referred to the following process for estimating the initial value of the reflection coefficient which is a sine wave or a sine wave-like change.

If the reflection coefficient x _{n of} each resolution element changes according to the sine wave law, the x _0n obtained by performing the weighted summation operation is still a sinusoidal waveform. As can be seen from Figure 5, the sample value x _{n of the} reflection coefficient (the dot shown in Figure 5) and the estimated initial value x _0n (the circle shown in Figure 5) are at the maximum and minimum values. Obviously, the error is minimal at the intermediate value.

In Fig. 6, x _0n is an estimated initial value of the reflection coefficient x _n . The weighted summation operation is a partial convolution of the weighting coefficients with the input sequence. Δ _{0n The} rate of change of the reflection coefficient (correction factor) caused by the main beam being irradiated to the target area.

In order to reduce the error, a search algorithm as shown in Fig. 6 is employed: a three-stage weighted summation operation is performed on the initial value sequence x _0n to obtain three sequences of x _1n , x _2n and x _3n . Find the change of the reflection coefficient before and after the weighted summation operation of each stage, and then use the extrapolation algorithm to obtain the change of the reflection coefficient caused by the first illumination (the actual target area of the radar beam), and obtain the correction factor △ _0n with △ _0n x _0n performs prediction correction point by point to obtain a predicted value of the sample sequence x _n of the reflection coefficient.

7 is an extrapolated search shown in FIG. 6 for the sample values x _n in which the reflection coefficients of the respective resolution elements shown in FIG. 5 are sinusoidally changed along the azimuth angle and the estimated initial value x _0n shown in FIG. A diagram of the predicted values obtained after the algorithm is corrected.

If this prediction is satisfactory, it can be used as the final estimate, as shown in the circle in Figure 7, which approximates the sample value x _{n of the} reflection coefficient, as shown by the dot in Figure 7. If this prediction is not satisfactory, cyclic recursive processing is used, as shown in Figure 10.

The mean square error of the estimated initial value x _0n and the sample value x _n shown in Fig. 5 is 6.4 × 10 ^-3 . The predicted value corrected by the extrapolated search algorithm shown in Fig. 7 is closer to the sample value x _n , and the mean square error is 3.5 × 10 ^-4 .

Figure 8 is a sample value x _n (the dot shown in Figure 8) and a normalized estimated initial value x _0n of the reflection coefficient of each resolution element along azimuth sinusoidal variation (as shown in Figure 8). Diagram of the circle).

Figure 9 is a sample value x _n (such as the dot shown in Figure 9) in which the reflection coefficients of the respective resolution elements shown in Figure 8 are sinusoidally varied along the azimuth angle and the estimated initial value shown in Figure 8 x _0n is a diagram in which the predicted value (circle shown in FIG. 9) obtained by the extrapolation search algorithm shown in FIG. 6 is performed.

The processing method when the reflection coefficient changes like a sine wave along the azimuth is the same as the processing method when the reflection coefficient changes sinusoidally along the azimuth.

The mean square error of the estimated initial value x _0n and the sample value x _n shown in Fig. 8 is 6.9 × 10 ^-4 . The predicted value corrected by the extrapolated search algorithm shown in Fig. 9 is closer to the sample value x _n , and the mean square error is 1.3 × 10 ^-4 .

In addition, when an extrapolation search algorithm is used to correct the estimated values of the reflection coefficients of the individual azimuths in a piecewise linear, sinusoidal, and sinusoidal variation along the azimuth, the method is particularly suitable for the reflection coefficients of the respective resolution elements. There is only one maximum value or one minimum value in the main beam width.

It should be noted that the extrapolation search algorithm of the present invention is not limited to performing a three-level weighted summation operation on the estimated initial value sequence x _0n . Depending on the actual situation, those skilled in the art can perform less or more levels of weighted summation operations when employing the extrapolation search algorithm of the present invention.

2.3.3 Cyclic recursive algorithm: Assume that the reflection coefficients of each resolution element vary randomly along the azimuth

In the case of a random change in the reflection coefficient, only Figure 6 is used. The result of the extrapolation search algorithm shown in the processing is unsatisfactory, and a cyclic recursive algorithm is needed to further reduce the error.

As shown in Figure 10, at n In the case where the reflection coefficients of the resolution elements vary randomly along the azimuth within the beamwidth of the main beam, at least one weighted summation operation can be performed on the estimated initial values of the reflection coefficients to calculate the weighted summation of each level. a change of the reflection coefficient before and after the calculation, and extrapolating the change of the reflection coefficient to obtain a change of the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor according to the estimated initial value and the The correction factor obtains a predicted value by using a prediction algorithm, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error is less than or equal to The error value is used, and the predicted value is used as the final estimated value of the reflection coefficient of the resolution element, and if the comparison error is greater than the preset error value, the correction factor is corrected according to the comparison error.

Explanation of the symbols in Figure 10:

x _n is the reflection coefficient of the resolution element, n=1, 2...N , where N is the number of resolutions of the target region being processed;

x _0n is an estimated initial value of the reflection coefficient x _n of the resolution element;

Δ _0n is a correction factor from the extrapolation search algorithm;

Δ _in is a correction factor obtained by correcting the correction factor Δ _0n with the comparison error e _in ;

y _in is a predicted value obtained by using a prediction algorithm according to the estimated initial value x _0n and the correction factor Δ _0n or Δ _in , where i=0, 1...K; for example, where y _{0n is represented} by x _0n and △ _0n obtained predicted value, the predicted value y _1n and by △ _1n x _0n obtained;

z _in is a test value obtained by performing a weighted summation operation on the predicted value y _in , where i=0, 1...K;

e _in = x _{0n -} y _in , that is, a comparison error or a test error obtained by comparing the estimated initial value x _0n with the test value z _in , where i = 0, 1, ... K;

ε is the error value preset according to the required calculation precision. When e _in > ε, the cyclic recursion is continued until the K recursive recursion causes e _Kn ≤ ε to end the cyclic recursion, and the predicted value y _Kn The final estimate x _En for a given error requirement, ie y _Kn = x _En .

Specific instructions for the processing performed in Figure 10:

1) The main beam of the meteorological radar with the total illumination power h is the target beam x _{n to} obtain the echo power P _n , which is normalized to obtain the estimated initial value x _0n . This process is equivalent to a weighted summation of x _n to obtain x _0n , n = 1 ... N , where N is the number of resolved elements.

2) Perform a three-level weighted summation operation on the estimated initial value x _0n to obtain x _1n , x _2n and x _3n , and obtain an correction factor Δ _0n for each resolution element by using an extrapolation algorithm. It should be noted that the extrapolation search algorithm of the present invention is not limited to performing a three-level weighted summation operation on the estimated initial value x _0n . Depending on the actual situation, those skilled in the art can perform fewer or more levels of weighted summation operations when performing the extrapolation search algorithm.

3) Based on the estimated initial value x _0n and the correction factor Δ _0n , the prediction value y _{0n of} the reflection coefficient x _n is obtained by the prediction algorithm. In order to test the prediction value y _0n, the predicted value y _0n weighted summation operation verification value obtained z _0n.

4) Compare the estimated initial value x _0n with the test value z _0n to obtain a comparison error or test error e _0n . If the test error e _{0n is} less than or equal to the preset error value ε , then y _{0n is} the final estimated value. If the test is greater than a preset error e _0n error value [epsilon], with the comparison error of the correction factor e _0n _0n correction △, △ correction factor to obtain the corrected _1N, then with _1N △ △ _0n instead of performing prediction processing.

5) Repeat steps 3) and 4) until the Kth time, so that the error between x _0n and z _Kn is less than or equal to the preset error value ε, then the loop is terminated, and y _Kn at this time is the final satisfying the given error. The estimated value x _En is y _Kn = x _En .

Figure 11 (a) shows the results of processing the observation data using a linear estimation algorithm. In Figure 11 (a In the middle, the abscissa is the resolution element number, and the ordinate is the amplitude of the normalized reflection coefficient. In this example, the width of the resolution element is equal to the sub-beam width and the number of sub-beams is five. In Figure 11 (a In the case, the ordinate value corresponding to the dot represents a normalized reflection coefficient with a resolution of 3 dB beam width, and the ordinate value corresponding to the circle represents a normalized reflection coefficient with a resolution of the sub-beam width.

Figure 11 (b) shows the results of processing the observation data using linear estimation, extrapolation search, and cyclic recursive algorithm. In Figure 11 In (b), the abscissa is the resolution element number and the ordinate is the amplitude of the reflection coefficient. In Figure 11 (b In the middle, the ordinate value corresponding to the dot represents the observed value or the estimated initial value of the reflection coefficient of the resolution element, and the ordinate value corresponding to the circle represents the predicted value of the reflection coefficient of the resolution element.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the meteorological radar signal according to the second embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beamwidth can be compressed to 1/n by azimuth (n is the number of sub-beams or the number of pulses in the beam) . In particular, the weather radar signal processing method according to the second embodiment of the present invention can increase the azimuth resolution of the weather target zone by an order of magnitude.

Third embodiment

The third embodiment is a weather radar signal processing method for a pitch angle based on the first embodiment.

In the third embodiment of the present invention, the beam width of the main beam is divided into q sub-beams according to the pitch angle angle (where q The third embodiment of the present invention is the same as the second embodiment except for an integer greater than or equal to 2.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the weather radar signal according to the third embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beam width can be compressed to 1/q by the pitch angle (q) The number of sub-beams or the number of pulses in the beam). Specifically, the weather radar signal processing method according to the third embodiment of the present invention can increase the pitch angle resolution of the weather target area by an order of magnitude.

Fourth embodiment

The fourth embodiment is a weather radar signal processing method for distance based on the first embodiment.

4.1 Convolution and deconvolution of the antenna radar's antenna distance scanning

For the sake of discussion, each pulse in the main beam of the radar wave of the meteorological radar and the target area of the observation are discretized. The pulse width of each pulse is equally divided into m sub-pulses, where m is an integer greater than or equal to 2. The pulse width of each sub-pulse is equal to 1/m of the pulse width of the pulse. The reverse order of the sub-pulses along the propagation direction is named as the first sub-pulse, the second sub-pulse, and the third sub-pulse. . . The mth sub-pulse (both the transmit pulse and the echo pulse), the illumination power of each sub-pulse is named h ₁ , h ₂ , h _{3 , respectively} . . . h _m .

In the case where the pulse is equally divided into m sub-pulses according to the pulse width to detect the target region, the target region is divided into a plurality of distance resolution elements (hereinafter referred to as resolution elements), each resolution The distance span of the element corresponds to 1/2 of the pulse width of one sub-pulse, and the illumination powers h ₁ , h ₂ , h _{3 of} the m sub-pulses. . . The sum of h _m is equal to the predetermined illumination power (total illumination power) h of each of the pulses. Specifically, in the target area, the distance is divided into continuous resolution elements along the propagation direction of the transmission pulse, and the distance span of each resolution element is made to correspond to 1/2 of the sub-pulse width.

And m Each sub-pulse sequentially illuminates each of the resolving elements, and each of the resolving elements has a reflection coefficient, and the reflection coefficient of each of the resolving elements is the ratio of the echo power of the pulse irradiated to the resolving element to the irradiation power. That is, the reflection coefficient of each resolution element is not only equal to the ratio of the echo power of each sub-pulse irradiated to the resolution element to the illumination power, but is also numerically equal to the echo power of each sub-pulse that is irradiated to the resolution element in turn. And the ratio of the total illumination power to the pulse. Moreover, the reflection coefficient of one of the plurality of resolution elements can be calculated using the illumination power and the echo power of the pulse and the illumination power of each sub-pulse.

In the present embodiment, the reflection characteristic of each resolution element is equivalent to a point target, and the reflection coefficients are named x ₁ , x ₂ , x ₃ . . . Wait. In the process of transmitting the pulse forward, the echo power generated by the illumination target region is the superposition of the impulse response sequence generated by the sub-pulse sequence irradiating the respective resolution elements, as shown in FIG. In other words, when the antenna of the meteorological radar performs the distance scanning, the echo power of each pulse obtained is the illumination power h _i of each sub-pulse (where i is the sub-pulse number, i=1, 2, 3 ... m ) The result of addition after multiplying the reflection coefficient x _j (where j is the resolution element number, j = 1, 2, 3 ...) of the corresponding resolution element. That is, the process of obtaining the echo power by the antenna scanning target area is a convolution process.

From the point of view of system analysis, the reflection coefficient of each resolution element in the meteorological target area can be regarded as the input signal of the system. The illumination power of each sub-pulse is regarded as the unit impulse response of the system, and the output of the receiver of the meteorological radar. Power (eg, video output power) is the output signal of this system. Since the illumination power of each sub-pulse is fixed, this system is a linear time-invariant system with finite impulse response. In this physical model, the output signal and the unit impulse response are known, and the input signal, that is, the reflection coefficient of each resolver, is required. Therefore, the nature of the problem of solving the reflection coefficient of each resolution element is a deconvolution problem.

Figure 12 illustrates the convolution process by taking 9 sub-pulses to illuminate 9 resolved elements. According to the superposition principle of the linear system, the system response is the superposition of the response of each input signal. For discrete sequences, one input data is an impact, so the system response is the superposition of the impulse response of each resolution element, as shown in Figure 12. In Fig. 12, the first row h ₅ , h ₄ , h ₃ , h ₂ and h ₁ represent the irradiation power of 5 sub-pulses, and the second behavior is the reflection coefficients x ₁ , x ₂ , x _{3 of the} respective resolution elements. . . Wait. The first column is time t ₁ , t ₂ , t ₃ . . . Corresponding to the radar pulse repetition period, the second column P ₃ , P ₄ , P ₅ , P ₆ and P ₇ corresponds to the output power of the receiver (for example, video output power). Each small square illuminates the echo power of the corresponding resolution element with h _i x _j on behalf of the sub-pulse.

In Fig. 12, when the sub-pulse sequence propagates and illuminates the first resolver, the sequences h ₁ x ₁ , h ₂ x ₁ , h ₃ x ₁ , h ₄ x ₁ below the reflection coefficient x _{1 of the first} resolver And h ₅ x ₁ are the echo powers generated by the first sub-pulse, the second sub-pulse, the third sub-pulse, the fourth sub-pulse, and the fifth sub-pulse respectively illuminating the first resolving element, that is, the first resolving element Shock response sequence. When the sub-pulse sequence propagates and illuminates the second resolution element, the sequences h ₁ x ₂ , h ₂ x ₂ , h ₃ x ₂ , h ₄ x _{2 ,} and h ₅ x ₂ below the reflection coefficient x2 of the second resolution element are The first sub-pulse, the second sub-pulse, the third sub-pulse, the fourth sub-pulse, and the fifth sub-pulse respectively illuminate the echo power generated by the second resolution element, that is, the impulse response sequence of the second resolution element. The sequence below the reflection coefficient of the remaining resolution elements is deduced by analogy.

The impulse response sequences of two adjacent bins differ in time by one sub-pulse width. Since the illumination of each sub-pulse expands a point of space into an impulse response sequence, the impulse response sequences of adjacent resolution elements are staggered by a sub-pulse width and are superimposed on each other. The result is: each sample of the system response The value is the superposition of five impulse response sequences, each of which responds to a sub-pulse that illuminates the echo power of one resolver. For example, the echo power P ₃ obtained at the time corresponding to time t ₅ is h ₅ x ₁ contributed by the first impulse response sequence, h ₄ x ₂ contributed by the second impulse response sequence, and the third impulse response sequence contribution h ₃ x ₃ , the fourth impulse response sequence contributes h ₂ x ₄ and the fifth impulse response sequence contributes h ₁ x ₅ . The echo power P ₃ is numerically the fifth sub-pulse, the fourth sub-pulse, the third sub-pulse, the second sub-pulse, and the first sub-pulse, and the corresponding first, second, and third resolutions are simultaneously irradiated. The sum of the echo powers h ₅ x ₁ , h ₄ x ₂ , h ₃ x ₃ , h ₂ x ₄ and h ₁ x ₅ obtained by the element, the 4th resolution element and the 5th resolution element. At this time, the center line of the pulse width is located at the third resolution element (that is, coincides with the center line of the third resolution element), so it is named P ₃ . At time t ₆ , when each sub-pulse simultaneously illuminates the second, third, fourth, fifth, and sixth resolved elements, the resulting echo power P ₄ is numerically equal to the echo. The sum of power h ₅ x ₂ , h ₄ x ₃ , h ₃ x ₄ , h ₂ x ₅ and h ₁ x ₆ . . . . The case of time t ₇ , t ₈ , t _{9 ,} etc. and so on. At this time, FIG. 12 can also be understood as a linear convolution process in which each sub-pulse illuminates each resolution element.

4.2 Deconvolution and beam compression

When dealing with azimuth super-resolution problems, the number of pulses processed by (video) can be equal to the number of sub-beams n . Here, the number of pulses processed by (video) is equal to the number m of sub-pulses. For example, a partial convolution process with a number of subpulses equal to 5 is shown in Figure 13.

The partial convolution process shown in Figure 13 takes 5 echo powers P ₃ , P ₄ , P ₅ , P ₆ and P ₇ for processing. Moreover, this partial convolution process involves a total of 9 resolution elements whose reflection coefficients are x ₁ , x ₂ , x ₃ , x ₄ , x ₅ , x ₆ , x ₇ , x ₈ and x _{9 , respectively} . Among the 9 resolution elements, only the 5th resolution element whose reflection coefficient is x ₅ is irradiated once by each sub-pulse, and the remaining 8 resolution elements are incomplete illumination. In this case, the deconvolution process is performed with 5 echo powers, and only the reflection coefficient of one resolution element can be obtained. This is because only one resolution element x ₅ meets the conditions of full illumination (same as the actual illumination in Figure 12) or pulse compression, and complete illumination is not possible to achieve the desired pulse compression. With 5 echo powers, only one reflection coefficient of the resolution element can be obtained, which can be called partial deconvolution.

The conceptual reasoning of pulse compression is as follows. Equation (8) shows that the sum of the illumination powers h ₁ , h ₂ , h ₃ , h ₄ and h ₅ of each sub-pulse is equal to the total illumination power h of the pulse:

h = h ₁ + h ₂ +h ₃ + h ₄ + h ₅ -----( 8 )

Multiplying both sides of equation (8) by the reflection coefficient x5 of the fifth resolution element yields equation (9):

Hx ₅ = h ₁ x ₅ + h ₂ x ₅ + h ₃ x ₅ + h ₄ x ₅ + h ₅ x ₅ -----( 9 )

The h on the left side of equation (9) represents the total illumination power of the pulse, and the right side is the sum of the illumination powers h ₁ , h ₂ , h ₃ , h ₄ and h ₅ of each sub-pulse. Equation (9) embodies the concept of pulse compression, because the right side is the sum of the echo powers of the 5th resolution elements whose reflection coefficients are x ₅ in turn, and the total illumination power h of the pulses on the left is the reflection coefficient of one illumination. The echo power of the 5th resolution of x ₅ . Note that the pulse width corresponding to the total illumination power h of the pulse at this time is equal to the sub-pulse width. Since h on the left side of equation (9) corresponds to the sub-pulse width, pulse compression can be achieved by finding the sum of the echo powers of the fifth resolver with the reflection coefficient of x ₅ on the right side of equation (9). If the total illumination power and echo power of the pulse are normalized, that is, let h=1, then the equation (10) is obtained:

x ₅ = h ₁ x ₅ + h ₂ x ₅ + h ₃ x ₅ + h ₄ x ₅ + h ₅ x ₅ -----( 10 )

Formula (10) shows that: for each sub-pulse reflection coefficient of 5 x _5, respectively irradiating a resolution element of the normalized sum of echo power, equal to the resolution element 5 x ₅ reflection coefficient numerically. Therefore, deconvolution or pulse compression is the sum of the echo powers that each sub-pulse is irradiated to the same resolution element. As can be seen from FIG. 13: P ₃ from the extracted long h ₁ x _5, extracted from h ₂ x ₅ P ₄ extracted from h ₃ x _{_5. 5} P extracted from h ₄ x ₅ P _6, By extracting h ₅ x ₅ from P ₇ , the reflection coefficient x _{5 of the fifth} resolver can be obtained. Since the ratio of the echo power at the 5th resolution of each sub-pulse to the corresponding observations P ₃ , P ₄ , P ₅ , P ₆ and P ₇ is unknown (see Figure 13), the constant to be determined is used. C means that we get the formula (11):

x ₅ = C ₁ P ₃ + C ₂ P ₄ + C ₃ P ₅ + C ₄ P ₆ + C ₅ P ₇ -----( 11 )

In formula (11), C ₁ P ₃ represents the extraction of h ₁ x ₅ from P ₃ , C ₂ P ₄ represents the extraction of h ₂ x ₅ from P ₄ , and C ₃ P ₅ represents the extraction of h ₃ x ₅ from P ₅ . , C ₄ P ₆ represents the extraction of h ₄ x ₅ from P ₆ , and C ₅ P ₇ represents the extraction of h ₅ x ₅ from P ₇ .

A specific manner of solving the deconvolution problem in the fourth embodiment of the present invention and 2.3 in the second embodiment The way of solving this deconvolution problem is the same in the section using linear estimation, extrapolation search and cyclic recursive algorithm.

In the weather radar signal processing method for deconvolving the observation value (partial convolution) of the weather radar signal according to the fourth embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the pulse width can be compressed to 1/m by distance . Specifically, the weather radar signal processing method according to the fourth embodiment of the present invention can increase the distance resolution of the weather target area by an order of magnitude.

Fifth embodiment

The fifth embodiment is a weather radar signal processing method for azimuth and elevation angles based on the first embodiment. The weather radar signal processing method can achieve beam compression by performing corresponding processing by using the processing method for the azimuth in the second embodiment and the processing method for the elevation angle in the third embodiment, respectively.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the weather radar signal according to the fifth embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beamwidth can be compressed to 1/n by azimuth and the beamwidth can be compressed to 1/q by the pitch angle. . Specifically, the weather radar signal processing method according to the fifth embodiment of the present invention can increase the azimuth resolution and the elevation angle resolution of the weather target area by an order of magnitude.

Sixth embodiment

The sixth embodiment is a weather radar signal processing method for azimuth and distance based on the first embodiment. The weather radar signal processing method can realize beam compression and pulse compression by performing corresponding processing by using the processing method for azimuth in the second embodiment and the processing method for distance in the fourth embodiment, respectively.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the meteorological radar signal according to the sixth embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beam width can be compressed to 1/n by azimuth and the pulse width can be compressed to 1/m by distance. . Specifically, the weather radar signal processing method according to the sixth embodiment of the present invention can increase the azimuth resolution and the distance resolution of the weather target area by an order of magnitude.

Seventh embodiment

The seventh embodiment is a weather radar signal processing method for the pitch angle and the distance based on the first embodiment. The weather radar signal processing method can perform beam compression and pulse compression by performing corresponding processing by using the processing method for the elevation angle in the third embodiment and the processing method for the distance in the fourth embodiment, respectively.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the weather radar signal according to the seventh embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beam width can be compressed to 1/q by the pitch angle and the pulse width can be compressed to 1/m by distance. . Specifically, the weather radar signal processing method according to the seventh embodiment of the present invention can increase the pitch angle resolution and the distance resolution of the weather target area by an order of magnitude.

Eighth embodiment

The eighth embodiment is a weather radar signal processing method for azimuth, elevation angle, and distance based on the first embodiment. The weather radar signal processing method can be respectively performed by using the processing method for the azimuth angle in the second embodiment, the processing method for the pitch angle in the third embodiment, and the processing method for the distance in the fourth embodiment, respectively. The processing implements beam compression and pulse compression.

In the meteorological radar signal processing method for deconvolving the observation value (partial convolution) of the weather radar signal according to the eighth embodiment of the present invention, linear estimation, extrapolation search, and/or Or the cyclic recursive algorithm implements the deconvolution function. Therefore, the beam width can be compressed to 1/n by azimuth, the beam width can be compressed to 1/q by the pitch angle, and the pulse width can be compressed to 1/m by distance. . Specifically, the meteorological radar signal processing method according to the eighth embodiment of the present invention can increase the azimuth resolution, the pitch angle resolution, and the range resolution of the meteorological target area by an order of magnitude, and can make the beam width and the pulse width based on the beam width and the pulse width. The volume resolution is reduced to one thousandth.

Ninth embodiment

Figure 14 is a schematic block diagram of a weather radar signal processing system 200 in accordance with a ninth embodiment of the present invention.

The weather radar signal processing system 200 shown in FIG. 14 includes: a transmitter 201 And a main beam for transmitting a radar wave to the target area through an antenna, the main beam having a predetermined illumination power, a predetermined beam width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse Width and predetermined illumination power; receiver 202 for receiving an echo from the target zone through the antenna; and an echo processor 203 Means for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone; wherein The main beam is equally divided by azimuth angle n a sub-beam to detect the target area to obtain an azimuth super-resolution weather radar signal, and/or to divide the main beam into pitch angles a sub-beam to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or to divide each of the pulses into pulse widths into m a sub-pulse to detect the target region to obtain a meteorological radar signal with a distance super-resolution, where n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is greater than or equal to 2 The integer.

Weather radar signal processing system 200 according to ninth embodiment of the present invention The weather radar signal processing method described in the second to eighth embodiments described above can be realized.

Tenth embodiment

A tenth embodiment of the present invention provides a computer readable storage device storing instructions for executing the weather radar signal processing method according to the first to eighth embodiments. Software modules can be stored in RAM Memory, Flash, ROM Memory, EPROM Memory, EEPROM Memory, Registers, Hard Disk, Removable Disk, CD-ROM Or any other form of storage medium in the art. Illustratively, the storage medium can be coupled to the processor such that the processor can read information from the storage medium and can write information to the storage medium. Alternatively, the storage medium can also be integrated into the processor. The processor and storage medium can be set to In an ASIC, an ASIC can be placed in a user terminal. Alternatively, the processor and the storage medium may also be disposed in different components in the user terminal.

The present invention performs deconvolution processing on observations (partial convolutions) of meteorological radars. Therefore, the beamwidth can be compressed to 1/n by azimuth ( n is an integer greater than or equal to 2, preferably the number of sub-beams or the number of pulses in the beam), and the beam width is compressed to 1/q by the pitch angle (q is greater than or equal to 2 Integer, preferably the number of subbeams or the number of pulses in the beam), and / or compress the pulse width to 1/m by distance For the number of sub-pulses). In particular, the meteorological radar signal processing method and system according to an embodiment of the present invention can increase the azimuth resolution, the elevation angle resolution, and the range resolution of the meteorological target area by an order of magnitude, respectively, and can be based on beam width and pulse width. The volume resolution is reduced to one thousandth.

The various illustrative logical blocks, elements, and steps listed in the embodiments of the invention may be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, various illustrative components, units and steps described above have generally described their functions. Whether such functionality is implemented by hardware or software depends on the design requirements of the particular application and the overall system. In one or more exemplary designs, the above-described functions described in the embodiments of the present invention may be implemented in hardware, software, firmware, or any combination of the three. A person skilled in the art can implement the described functions using various methods for each specific application, but such implementation should not be construed as being beyond the scope of the embodiments of the present invention.

The specific embodiments of the present invention have been described in detail with reference to the preferred embodiments of the present invention. All modifications, equivalent substitutions, improvements, etc., made within the spirit and scope of the invention are intended to be included within the scope of the invention.

Claims

A weather radar signal processing method, comprising the steps of:

a transmitting step for transmitting a main beam of a radar wave to a target area through an antenna, the main beam having a predetermined illumination power a predetermined beam width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse width and a predetermined illumination power;

a receiving step for receiving an echo from the target zone through the antenna;

An echo processing step for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone;

Wherein the main beam is equally divided into n sub-beams by azimuth angle to detect the target area to obtain an azimuth super-resolution weather radar signal, and Or dividing the main beam into q sub-beams by a pitch angle to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or equally dividing each pulse by pulse width For m a sub-pulse to detect the target area to obtain a meteorological radar signal with a distance super-resolution,

Where n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is an integer greater than or equal to 2.
A weather radar signal processing method according to claim 1, wherein

Dividing the main beam into azimuth angles by n a sub-beam to detect the target area, the target area is divided into a plurality of azimuth resolution elements, each azimuth angle corresponding to the azimuth angle is equal to one sub-beam width, and the n The sum of the illumination powers of the sub-beams is equal to the predetermined illumination power of the main beam,

Dividing the main beam into pitch angles by q a sub-beam to detect the target area, the target area is divided into a plurality of elevation angle resolution elements, and a pitch angle angle corresponding to each elevation angle resolution element is equal to one sub-beam width, and the q The sum of the illumination powers of the sub-beams is equal to the predetermined illumination power of the main beam,

In the case where each of the pulses is equally divided into m sub-pulses by pulse width to detect the target area Dividing the target area into a plurality of distance resolution elements, the distance span of each distance resolution element corresponding to 1/2 of a pulse width of one sub-pulse, and the m The sum of the illumination powers of the sub-pulses is equal to the predetermined illumination power of each of the pulses.
A weather radar signal processing method according to claim 2, characterized in that

In the case where the main beam is equally divided into n sub-beams by azimuth angle to detect the target area, the n Each sub-beam sequentially illuminates each azimuth resolution element, and each azimuth resolution element has a reflection coefficient, and the reflection coefficient of each azimuth resolution element is the echo power of the beam irradiated to the azimuth resolution element. Ratio of illumination power of the beam ,

In the case where the main beam is equally divided into q sub-beams by the pitch angle angle to detect the target area, the q Each of the sub-beams sequentially illuminates each of the elevation angle resolution elements, and each of the elevation angle resolution elements has a reflection coefficient, and the reflection coefficient of each of the elevation angle resolution elements is the echo power of the beam irradiated to the elevation angle resolution element. The ratio of the illumination power of the beam, In the case where each of the pulses is equally divided into m sub-pulses by pulse width to detect the target region, the m Each sub-pulse sequentially illuminates each distance resolution element, and each distance resolution element has a reflection coefficient, and the reflection coefficient of each distance resolution element is the echo power of the pulse irradiated to the distance resolution element and the irradiation of the pulse Power ratio .
A weather radar signal processing method according to claim 3, wherein

Dividing the main beam into azimuth angles by n In the case of sub-beams to detect the target area, the echo processing step further includes: utilizing predetermined illumination power and echo power of the main beam and the n The illumination power of the sub-beams calculates a reflection coefficient of one of the plurality of azimuth resolution elements,

Dividing the main beam into pitch angles by q In the case of sub-beams to detect the target area, the echo processing step further includes: utilizing predetermined illumination power and echo power of the main beam and the q The illumination power of the sub-beams calculates a reflection coefficient of one of the plurality of elevation angle resolution elements,

Dividing each of the pulses into pulse widths by m In the case where a sub-pulse is used to detect the target area, the echo processing step further includes: utilizing a predetermined illumination power and echo power of the pulse and the m The illumination power of the sub-pulses calculates a reflection coefficient of one of the plurality of distance-resolving elements.
A weather radar signal processing method according to claim 4, characterized in that

The step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements includes: The sum of the echo powers obtained by the sub-beams illuminating the same azimuth resolution element,

The step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: The sub-beams are summed with the echo power obtained by the same pitch angle resolution element.

The step of calculating a reflection coefficient of one of the plurality of distance resolution elements includes: The sub-pulses illuminate the echo power obtained by the same distance resolution element.
A weather radar signal processing method according to claim 4, characterized in that

The step of calculating the reflection coefficient of one of the plurality of azimuth resolution elements includes: when the n sub-beams are simultaneously irradiated to the n azimuth resolution elements, using the n The sum of the echo powers of the sub-beams divided by the predetermined illumination power of the main beam is used as an estimated initial value of the reflection coefficient of the azimuth resolution element where the main beam center line is located,

Calculating a reflection coefficient of one of the plurality of elevation angle resolution elements includes: when the q sub-beams are simultaneously irradiated to q elevation angle resolution elements, using the q The sum of the echo powers of the sub-beams divided by the predetermined illumination power of the main beam is used as an estimated initial value of the reflection coefficient of the elevation angle resolution element of the main beam center line.

Calculating a reflection coefficient of one of the plurality of distance resolution elements includes: when the m sub-pulses are simultaneously irradiated to m distance resolution elements, using the m The sum of the echo powers of the sub-pulses divided by the predetermined illumination power of the pulses is used as an estimated initial value of the reflection coefficient of the distance resolution element in which the pulse width of the pulse is located.
A weather radar signal processing method according to claim 6, wherein

The step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements includes: In the case where the reflection coefficient of the azimuth resolution element changes linearly along the azimuth angle within the predetermined beam width, the estimated initial value of the reflection coefficient of the azimuth angle resolution element of the main beam center line is equal to the azimuth angle resolution element The actual value of the reflection coefficient,

The step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: at the q In the case where the reflection coefficient of the pitch angle resolution element changes linearly along the pitch angle within the predetermined beam width, the estimated initial value of the reflection coefficient of the pitch angle resolution element where the main beam center line is located is equal to the pitch angle resolution element The actual value of the reflection coefficient,

The step of calculating a reflection coefficient of one of the plurality of distance resolution elements includes: at the m In the case where the reflection coefficient of the distance resolution element changes linearly along the distance within the predetermined pulse width, the estimated initial value of the reflection coefficient of the distance resolution element where the pulse width of the pulse is located is equal to the reflection coefficient of the distance resolution element Actual value.
A weather radar signal processing method according to claim 6, wherein

The step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements includes: The reflection coefficients of the azimuth resolution elements vary in a piecewise linear and/or sinusoidal variation along the azimuth within the predetermined beamwidth and/or In the case of a sinusoidal-like change, the change of the reflection coefficient before and after the weighted summation operation of each stage is calculated by performing at least one-stage weighted summation operation on the estimated initial value, and extrapolating by the change of the reflection coefficient Calculating to obtain a change in the reflection coefficient caused by the main beam actually illuminating the target region, thereby obtaining a correction factor, and then correcting the estimated initial value with the correction factor to obtain a final estimated value of the reflection coefficient ,

The step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: at the q The reflection coefficient of the pitch angle resolution element has a piecewise linear change and/or a sine wave variation along the pitch angle within the predetermined beam width Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. Pushing a calculation to obtain a change in the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor, and then correcting the estimated initial value with the correction factor to obtain a final estimated value of the reflection coefficient,

The step of calculating a reflection coefficient of one of the plurality of distance resolution elements includes: at the m The reflection coefficients of the distance resolution elements vary linearly and/or sinusoidally along the distance within the predetermined pulse width and / Or in the case of a sinusoidal-like variation, the at least one-stage weighted summation operation is performed on the estimated initial value to calculate a change in the reflection coefficient before and after the weighted summation operation of each stage, and the variation of the reflection coefficient is used to perform the change. A calculation is performed to obtain a change in the reflection coefficient caused by the pulse actually illuminating the target region, thereby obtaining a correction factor, and then the estimated initial value is corrected by the correction factor to obtain a final estimated value of the reflection coefficient.
A weather radar signal processing method according to claim 6, wherein

The step of calculating a reflection coefficient of one of the plurality of azimuth resolution elements includes: In the case where the reflection coefficients of the azimuth resolution elements vary randomly along the azimuth angle within the predetermined beam width, at least one level weighted summation operation is performed on the estimated initial values to calculate each stage weighted summation a change of the reflection coefficient before and after the calculation, and extrapolating the change of the reflection coefficient to obtain a change of the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor according to the estimated initial value and the The correction factor obtains a predicted value by using a prediction algorithm, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error And less than or equal to the preset error value, the predicted value is used as a final estimated value of the reflection coefficient of the azimuth resolution element, and if the comparison error is greater than the preset error value, according to the comparison error To correct the correction factor,

The step of calculating a reflection coefficient of one of the plurality of pitch angle resolution elements includes: at the q In the case where the reflection coefficient of the pitch angle resolution element varies randomly along the pitch angle within the predetermined beam width, at least one weighted summation operation is performed on the estimated initial value to calculate a weighted summation of each level. a change of the reflection coefficient before and after the calculation, and extrapolating the change of the reflection coefficient to obtain a change of the reflection coefficient caused by the main beam actually illuminating the target area, thereby obtaining a correction factor according to the estimated initial value and the The correction factor obtains a predicted value by using a prediction algorithm, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error And less than or equal to the preset error value, the predicted value is used as a final estimated value of the reflection coefficient of the pitch angle resolution element, and if the comparison error is greater than the preset error value, according to the comparison error To correct the correction factor,

The step of calculating a reflection coefficient of one of the plurality of distance resolution elements includes: at the m In the case where the reflection coefficients of the distance resolution elements vary randomly along the distance within the predetermined pulse width, at least one level of weighted summation operation is performed on the estimated initial values to calculate each stage before and after the weighted summation operation a change in the reflection coefficient, an extrapolation calculation using the change in the reflection coefficient to obtain a change in the reflection coefficient caused by the pulse actually illuminating the target region, thereby obtaining a correction factor, which is utilized according to the estimated initial value and the correction factor The prediction algorithm obtains a predicted value, obtains a test value by performing a weighted summation operation on the predicted value, and compares the estimated initial value with the test value to obtain a comparison error, if the comparison error is less than or equal to Presetting the error value, the predicted value is used as a final estimated value of the reflection coefficient of the distance resolution element, and if the comparison error is greater than the preset error value, correcting the Correction factor.
The weather radar signal processing method according to claim 1, wherein n is equal to a predetermined number of pulses in said main beam, and / or q Equal to a predetermined number of pulses within the main beam.
A weather radar signal processing system, comprising:

a transmitter for transmitting a main beam of a radar wave to a target area through an antenna, the main beam having a predetermined illumination power, a predetermined beam width, and a predetermined number of pulses, each of the predetermined number of pulses having a predetermined pulse width and a predetermined illumination power;

a receiver for receiving an echo from the target zone through the antenna;

An echo processor for processing the echo to obtain a weather radar signal including echo power of the echo and information related to azimuth, elevation and distance of the target zone;

Wherein the main beam is equally divided into n sub-beams by azimuth angle to detect the target area to obtain an azimuth super-resolution weather radar signal, and Or dividing the main beam into q sub-beams by a pitch angle to detect the target area to obtain a pitch angle super-resolution weather radar signal, and/or equally dividing each pulse by pulse width For m a sub-pulse to detect the target area to obtain a weather radar signal with a super-resolution,

Where n is an integer greater than or equal to 2, q is an integer greater than or equal to 2, and m is greater than or equal to 2 The integer.