An airborne weather radar wind shear detection velocity deambiguating method

Abstract

The application discloses an airborne weather radar wind shear detection speed unblurring method, complete radar waveforms are generated based on A-wave and B-wave grouping waveforms; after receiving radar echo for digital sampling, rearrangement processing is carried out, and data of fast-slow time dimensions of A-group and B-group waveforms are obtained; then, along the fast time dimension, matching filter processing is carried out, and the fast time dimension is converted into a distance dimension; after Doppler components caused by the speed of an airplane itself are compensated, along the slow time dimension, FFT transformation processing is carried out, and distance-speed dimension spectrum data are obtained; along the distance dimension, first moment calculation processing is carried out on Doppler velocity spectrum for each distance gate to obtain blurred wind speed measurement values, and corresponding A-wave and B-wave data are extracted; unblurring operation processing is carried out based on the A-wave and B-wave data to obtain wind speed measurement values. The application has extremely small calculation amount, and can complete speed unblurring in a signal processing link, and does not affect the data refresh rate of wind shear detection.

Description

Technical Field

[0001] This invention relates to the field of airborne weather radar signal processing, and more particularly to the field of velocity defuzzification technology for airborne weather radar wind shear detection. Technical Background Wind shear is one of the most dangerous weather phenomena during aircraft takeoff and landing, historically causing numerous fatal air disasters and posing a serious threat to flight safety. Currently, wind shear detection and early warning primarily utilizes airborne weather radar for predictive wind shear detection. Its main working principle is based on pulse-Doppler radar, detecting changes in the Doppler frequency of radar echoes to assess the velocity variations of tiny particles in the wind field in the radar's illumination direction, thereby identifying potential wind shear hazard areas. Airborne weather radar wind shear detection provides crucial early warnings before aircraft enter hazardous areas, making it a vital technological means for modern civil aviation to ensure takeoff and landing safety and prevent accidents caused by wind shear.

[0002] Airborne weather radar is a pulse-Doppler radar system. The overall structure of the radar system is as follows: Figure 1 As shown, at the transmitting end, the baseband signal generated by the waveform transmitter is up-converted, amplified by a power amplifier, and then transmitted to the radar antenna via a transceiver isolator, from which the radar antenna radiates the signal towards the target. At the receiving end, the target echo signal is received by the radar antenna, amplified by a low-noise amplifier, mixed by a mixer, filtered by a low-pass filter, and then sampled by an ADC to generate two digital signals, I and Q, which undergo complete signal processing in subsequent signal processing stages.

[0003] The signal processing flow for airborne weather radar wind shear detection is as follows: Figure 3As shown, wind shear detection requires sequentially performing several processes, including wind speed estimation, wind speed gradient estimation, horizontal F-factor calculation, vertical F-factor calculation, overall F-factor calculation, and average F-factor calculation. This process demonstrates that accurate wind speed estimation is the primary task for obtaining correct wind shear detection results. According to the working principle of pulse Doppler radar, velocity resolution is determined by the number of pulses, the pulse repetition frequency period, and the carrier wavelength. Velocity range is determined by both the pulse repetition frequency period and the carrier wavelength. Since airborne weather radar operates in the X-band, with a typical operating frequency of 9.3 GHz and a corresponding wavelength of 3.23 cm, the carrier wavelength is essentially fixed. Simultaneously, the number of pulses is generally also relatively fixed. Therefore, a significant contradiction inevitably arises between velocity resolution and velocity range. In practical applications, a lower pulse repetition frequency period is typically set to ensure sufficient velocity resolution for wind speed measurements. However, this inevitably limits the velocity range, leading to velocity ambiguity. Therefore, researching feasible velocity defuzzification methods to solve the velocity fuzziness problem faced by airborne meteorological radar in wind shear detection during wind speed estimation has practical significance and engineering application value.

[0004] For velocity deambiguation in pulse Doppler radar, current methods mainly employ velocity deambiguation based on multi-frequency periodic pulse waveforms. A Chinese master's thesis, "Research on the Implementation Method of Deambiguation Processing in Weather Radar," uses a "staggered PRF deambiguation" method. This method uses two sets of pulse waveforms with different transmission pulse repetition frequencies and solves for the unambiguous velocity based on the autocorrelation function of the two sets of pulses. The main problem with this method is poor algorithm stability. Another Chinese thesis (Hu Yanzhe. Research on Low-Altitude Wind Shear Deambiguation Method Based on Millimeter-Wave LFMCW Radar [D]. Tianjin: Civil Aviation University of China, 2021) uses a "one-dimensional set algorithm" velocity deambiguation method. This method uses multi-frequency pulse waveforms and generates a velocity candidate set, searching for the most probable velocity value within the candidate set as the deambigued velocity. The main problem with this method is its high computational complexity. Chinese patent document CN 106291497 B, entitled "A Velocity Defuzzification Algorithm Based on Fast Lookup Table Method," proposes a "fast lookup table method" for velocity defuzzification. This method uses a lookup table to solve for the fuzzy velocity. While this saves time in velocity defuzzification, the pre-stored lookup table requires significant storage space. Furthermore, a Chinese dissertation (Liu Bingya. Design and Implementation of Pulse Doppler Radar Defuzzification Algorithm [D]. Xi'an: Xi'an University of Electronic Science and Technology, 2017) uses a "Kalman filter differential method" for velocity defuzzification. This method is based on the target's moving speed output by the Kalman filter and uses this speed for defuzzification. This method is suitable for target positioning and tracking but not for airborne weather radar wind shear detection scenarios where parameter analysis is the primary focus.

[0005] In summary, existing velocity deambiguation methods for pulse Doppler radar all have significant drawbacks. The "staggered PRF deambiguation" method suffers from poor algorithmic stability; the "one-dimensional set algorithm" involves extremely high computational complexity; the "fast lookup table method" requires substantial storage space for the lookup table; and the "Kalman filter differentiation method" requires target tracking, making it difficult to apply in airborne weather radar scenarios. Furthermore, this method requires results from different data periods, reducing the radar's data refresh rate. Therefore, addressing the shortcomings of existing velocity deambiguation methods and specifically targeting the wind shear detection scenario in airborne weather radar, it is necessary and valuable to research a feasible velocity deambiguation method with low computational complexity that can be integrated with the wind shear detection signal processing flow. Summary of the Invention

[0006] The technical problem to be solved by this invention is to provide a velocity de-ambiguity method that can be implemented in the signal processing domain and has low computational complexity, addressing the velocity ambiguity problem in the velocity estimation stage of airborne weather radar wind shear detection application scenarios.

[0007] To address the aforementioned technical problems, this invention proposes a method for deambiguing the velocity detection of wind shear in airborne meteorological radar, comprising the following steps: S1. Using a grouped waveform design approach, after determining the radar waveform parameters and generating grouped waveforms of A-wave and B-wave, they are combined to form a complete radar waveform; S2. Based on the design parameters of the combined radar waveform, the ambiguity number range, range resolution, maximum unambiguous range, velocity resolution, and maximum unambiguous velocity are determined; S3. The radar echo is received, and the echo sampling data of the pulse Doppler radar is rearranged to obtain the fast time dimension-slow time dimension 2D data of the A-group and B-group waveforms respectively; S4. The flight data is obtained from the local data source. The system first obtains the aircraft's own speed and its horizontal and pitch angles relative to the radar pointing direction; then, it performs matched filtering along the fast time dimension to obtain the processed signal data; finally, it performs compensation processing on the Doppler component of the aircraft's own speed; then, it performs FFT transformation processing on the compensated data along the slow time dimension to transform the slow time dimension into the speed dimension; finally, it performs first-order moment calculation processing on the Doppler speed spectrum of each range gate along the range dimension to calculate the target spectral peak speed and extract the A-wave and B-wave data at the corresponding Doppler frequency points; finally, it performs speed defuzzification operation based on the A-wave and B-wave data to obtain the defuzzified wind speed measurement value.

[0008] Furthermore, the radar waveform in step S1 is obtained through a grouped waveform design method, specifically including: firstly determining the fundamental period of the radar pulse waveform. T Time delay coefficient between wave A and wave B a Pulse width τ、Pulse Amplitude A 0 and pulse count M And according to the repetition frequency period. T d = 2 T +a, pulse width τ、 Pulse Amplitude A 0 and pulse count M The parameters are set, and the pulse pair waveforms of group A are obtained by starting the first pulse waveform from time zero; then the entire waveform of group A is processed according to the time delay. T + a The process yields waveform group B; finally, waveform group A is combined with waveform group B to obtain the final radar deambiguity waveform design; waveform group A and waveform group B have the same repetition frequency period. T d = 2 T +a, and at the same time, there is a fixed phase difference between the waveforms in group A and group B. T +a.

[0009] Furthermore, the size of the fuzzy number range in step S2 K From 2 T +a and T The +a determines this, and the corresponding expression is shown below: In the above formula, A / B represents (2 T +a) / ( T +a) is the smallest integer ratio, and gcd(·) represents the greatest common divisor solution function; the distance resolution in step S2 R ves Pulse width τ The decision, and the corresponding expression, is: ,in c Represents the speed of light constant; the maximum unambiguous distance in step S2. R max From the basic cycle T The decision, and the corresponding expression, is: The velocity resolution in step S2 V ves From the repetition frequency period T d and pulse count M The decision, and the corresponding expression, is: ,in λ The wavelength of the carrier wave is represented; the maximum unambiguous velocity in step S2 is determined by the repetition frequency period. T d The decision, and the corresponding expression, is: .

[0010] Further, in step S3, the echo signal is discretely sampled at a fixed sampling rate, and A-wave data and B-wave data are extracted at time intervals. Simultaneously, A-wave data is extracted at (m...) T d , mT d + T Sampling is performed within a time interval, and the B wave is within (m) T d +2 T , mT d +2 T + a Sampling is performed within a specified time period; here m Indicates the repetition frequency period T d Sequence numbers are used to ensure that the lengths of wave A and wave B data are aligned; wave A and wave B data are then processed according to the fast time dimension and... m Data is reorganized using a two-dimensional approach to obtain 2D data with a fast time dimension and a slow time dimension, where the length of the fast time dimension is used as... N The sampling time sequence number of the fast time dimension is represented using... n This indicates that the rearranged A-wave and B-wave data were respectively used S A ( n , m )as well as S B ( n , m )express.

[0011] Furthermore, step S4 obtains the aircraft's ground speed from its own onboard inertial navigation system. V F Simultaneously, the radar pointing angle relative to the aircraft is obtained from the radar's servo control. θ and pitch angle β This allows us to calculate the aircraft's ground speed along the radar's pointing direction. .

[0012] Furthermore, the matched filtering process used in step S5 is performed in the frequency domain. The signal is first transformed from the time domain to the frequency domain, and then inversely transformed back to the time domain. The corresponding calculation process can be summarized as follows: ; In the above formula x ( t () represents the radar fundamental pulse time-domain signal. x ( nT s )express x ( t The corresponding digital signal, express x ( nT s The complex conjugate signal of ) FFT n Indicates along n Perform FFT transformation on the dimension. IFFT n Indicates along n After performing inverse FFT transformation and matched filtering, the A-wave and B-wave data are respectively used... Y A ( n , m )as well as Y B ( n , m )express.

[0013] Furthermore, step S6 is in m Compensation is performed on the upper dimension, and the magnitude of the aircraft's ground speed along the radar pointing direction is: The corresponding compensation coefficient is determined to be The overall calculation process is as follows: ; In the above formula f 0 represents the radar carrier frequency. After compensation processing for the aircraft's own velocity Doppler component, the A-wave and B-wave data are used respectively. as well as express.

[0014] Furthermore, in step S7, the input data sampling is processed by FFT transformation in the slow time dimension. as well as along m The dimension is then subjected to an FFT transformation, which is represented by the following formula: ; In the above formula, p Representing spectral coefficients, corresponding to different velocities, after slow-time FFT transformation, the A-wave and B-wave data are respectively used... Z A ( n , p )as well as Z B ( n , p )express.

[0015] Furthermore, the first-order moment calculation process in step S8 is based on spectral data. Z A ( n , p ),Z A ( n , p First-order moment calculations are performed to obtain the average Doppler frequency, which in turn yields the fuzzy wind speed measurement. The calculation process for the average Doppler frequency is expressed by the following formula: ; Simultaneously extract A-wave data and B-wave data. L ( n Spectral location data D A ( n )as well as D B ( n The processing procedure is represented by the following formula: ; In the above formula Indicates searching for the closest L ( n )of p After determining the location and completing the above operations, the average Doppler frequency is further converted into an average wind speed measurement. The calculation process is expressed by the following formula: ; In the above formula V ( n ) indicates different correspondences n The wind speed measurement value was not defuzzified.

[0016] Furthermore, the speed defuzzification operation in step S9 affects the fuzzy numbers. k To solve this problem, we first determine the following rules: k The set of values : ; Then, the fuzzy numbers are calculated. The process of calculating the fuzzy numbers can be summarized by the following formula: ; In the above formula Indicates the corresponding number n Numerical solution for fuzzy numbers of distance gates. express D A ( n The complex conjugate of ), the real(·) function represents extracting the real part from the complex number, and based on this, further... V ( n The velocity defuzzing process can be performed using the following formula: ; In the above formulaV m ( n ) represents the wind speed measurement value after defuzzification.

[0017] Compared with existing technologies, this invention offers the following advantages: The proposed velocity de-ambiguity method for airborne weather radar wind shear detection is applicable to airborne weather radar wind shear detection scenarios. It requires only adjustments to the radar waveform without altering the existing airborne weather radar hardware architecture or the overall signal processing flow for wind shear detection. Velocity de-ambiguity processing can be performed during the signal processing stage. This velocity de-ambiguity method requires minimal computation and can complete the de-ambiguity process within the signal processing stage without affecting the data refresh rate of wind shear detection. Therefore, it possesses significant advantages and application value. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of an airborne weather radar system to which this invention is applicable; Figure 2 This is a waveform design schematic diagram of the present invention, wherein Figure (a) shows the overall combined waveform of the radar, Figure (b) shows the waveform schematic diagram of wave A, and Figure (c) shows the waveform schematic diagram of wave B. Figure 3 This is the overall process for wind shear detection using airborne weather radar; Figure 4 This is a flowchart of the overall processing procedure of the speed defuzzification method designed in this invention; Figure 5 This is an illustrative diagram illustrating the first-order moment calculation process used in this invention for the Doppler velocity spectrum; Figure 6 This is a description of the processing results of different processing steps in the preliminary verification of the embodiments of the present invention; Figure 7 This is a distance-velocity 2D thermal map obtained from the wind shear scenario simulation test of this invention; Figure 8 This is a comparison chart of velocity curves before and after defuzzification obtained in the wind shear scenario simulation test of this invention. Detailed Implementation

[0019] The following is a detailed description of a method for de-ambiguity detection velocity using airborne meteorological radar, which is disclosed in this invention, with reference to specific embodiments.

[0020] The proposed airborne weather radar wind shear detection velocity deambiguation method differs from the traditional "multi-frequency periodic pulse waveform" in its waveform design. This waveform design uses pulse groups with different repetition frequency periods; instead, it redesigns a combined waveform with A and B waves of the same repetition frequency period transmitted at intervals. Detailed waveform design specifications can be found in [reference needed]. Figure 2As can be seen, waves A and B are emitted at intervals and have the same repetition frequency period. T d Pulse width τ From a waveform design perspective, this design ensures that the target echo appears at the same distance and velocity position after A-wave and B-wave signal processing. Simultaneously, the pulses of A-wave and B-wave are fixed. T + a The time delay causes a signal difference between the A-wave data and the corresponding B-wave data. This signal difference is related to the target's velocity, so velocity deambiguation can be performed based on this signal difference.

[0021] Furthermore, relevant system coefficients (ambiguity range, range resolution, maximum unambiguous range, velocity resolution, maximum unambiguous velocity, etc.) can be calculated based on radar waveform parameters. (Range resolution for A-wave or B-wave) R ves Pulse width τ The decision, and the corresponding expression, is: ,in c Represents the speed of light constant; maximum unambiguous distance. R max From the basic cycle T The decision, and the corresponding expression, is: Speed ​​resolution V ves From the repetition frequency period T d and pulse count M The decision, and the corresponding expression, is: in λ The wavelength of the carrier wave is represented; the maximum unambiguous velocity is determined by the repetition rate period. T d The decision, and the corresponding expression, is: Size of the fuzzy number range K From the repetition frequency period T d = 2 T + a and T + a determines that, on the one hand, the observable velocity range of the Doppler spectrum of either A-wave or B-wave is: 2 On the other hand, due to A waves and B waves, there exists a magnitude of T + a The fixed time delay results in a fixed phase between waveform A and waveform B caused by the Doppler frequency, and the corresponding velocity measurement range is: The ratio of the speed measurement range of the two methods is ( T + a ): (2 T + aCombining these two methods can expand the speed measurement range; therefore, the expanded speed measurement range can be expressed as: ,here K Indicates the range of fuzzy numbers. K For positive integers, the range of fuzzy numbers is... K The calculation expression is as follows: ; In the above formula, A / B represents (2 T +a) / ( T The smallest integer ratio of +a), gcd(·) represents the function for finding the greatest common divisor.

[0022] Furthermore, the digital signal of the radar echo is modeled. (Reference) Figure 1 The diagram of the weather radar system is shown using... x ( t If ) represents the fundamental pulse waveform of a radar pulse, then the radar's transmitted signal can be represented as: ; In the above formula This indicates the transmitted signal of wave A. This indicates the transmitted signal of wave B. f 0 represents the carrier frequency. Based on this, a model of the target's echo signal is established: ; In the above formula This represents the echo signal of wave A. This represents the echo signal of the B wave. α This represents the attenuation coefficient of the echo. τ The echo delay can be represented by the following formula: ; In the above formula R Indicates the distance to the target. v Indicates the speed of the target. v f Indicates the speed of the aircraft. c This represents the speed of light constant. The echo signal undergoes down-conversion processing to eliminate... Components, while ignoring α The effect is substituted into the above echo delay. τ From the expressions, we can obtain the expressions for the intermediate frequency continuous signals of wave A and wave B as follows: ; For intermediate frequency signals according to the sampling period T s Digital sampling was performed, with A-wave sampling starting at time zero. T At that moment, wave B came from T +a Time starts sampling up to 2 T + a At that time, the A-wave and B-wave data have the same sampling length, such as Figure 4 As shown, the digital sampled signal of the intermediate frequency signal can be obtained as follows: ; In the above formula n Represents a fast time-dimensional sampling sequence with a length of N From the perspective of digital sampling signals, this design allows the digital sampling signals of the A wave and the B wave to be... n Weihe m The dimensions are consistent, thus ensuring the consistency of the correspondence between wave A and wave B data in subsequent processing, eliminating the need for further data matching between wave A and wave B data. Simultaneously, wave A and wave B data exist... The signal difference term is associated with the target's velocity, and therefore velocity defuzzification can be performed based on this signal difference.

[0023] Furthermore, matched filtering is performed on the intermediate frequency discrete signal. The corresponding processing procedure is as follows: ; In the above formula x ( n () indicates radar pulse signal. This represents its complex conjugate signal. FFT n Indicates along n Perform FFT transformation on the dimension. IFFT n Indicates along n After performing inverse FFT transformation and matched filtering, the A-wave and B-wave data are respectively used... Y A ( n , m )as well as Y B ( n , m )express.

[0024] make x ( t The time-domain waveform after matched filtering is: y ( t Then, the above formula can be further expressed as:

[0025] ; As can be seen from the above formula, after matched filtering, from n From a dimensional perspective, in The target peak appears at [location]. nThe time dimension reflects the target distance information, therefore n Dimension represents the distance dimension.

[0026] Furthermore, it is necessary to define the aircraft speed. v f The calculation formula shows that the weather radar can obtain the aircraft's ground speed from the aircraft's inertial navigation system. V F Simultaneously, the radar pointing angle relative to the aircraft's horizontal direction is obtained from the radar's servo control system. θ and pitch angle β This allows for the aircraft's speed. v f The calculation formula: .

[0027] Furthermore, based on v f The calculation formula requires... Y A ( n , m )as well as Y B ( n , m )middle v f The components are compensated to eliminate... v f The impact on target velocity detection is crucial; otherwise, accurate target velocity measurements cannot be obtained. This invention summarizes... v f The component processing procedure and the corresponding calculation expressions are as follows: ; In the above formula, the pair v f The A-wave and B-wave data after compensation processing of the resulting Doppler frequency components were respectively used as well as To represent. Because... R>> 2( R - vmT d - v f mT d ) / c , at the same time 2( vmT d + v f mT d ) / c <R ves Therefore, the above formula can be simplified to: ; In the above formula The first term represents the range component, the second term represents the velocity component, the third term represents the fixed phase caused by range, which does not affect detection and is not processed, and the fourth term represents the comparison. The difference component containing target velocity information is used for velocity defuzzification. As can be seen, and The distance and velocity components are exactly the same, therefore the target position corresponds identically in n-dimensional and m-dimensional dimensions. and The data can be directly matched, but there are differences between the two. The signal difference term is associated with the target's velocity, and velocity defuzzification can be performed based on this signal difference.

[0028] Furthermore, regarding and The calculation process for performing an FFT along the m-dimensional plane can be represented by the following formula: ; go through m After the 3D FFT transformation, the A-wave and B-wave data were respectively used Z A ( n , p )as well as Z B ( n , p )express, p Dimension represents the Doppler velocity dimension. As can be seen, Z A ( n , p )and Z B ( n , p )of n dimensional components and p The dimensional components are exactly the same, therefore in n Wei and p The target positions in the dimension are consistent. Z A ( n , p )and Z B ( n , p The data can be directly matched. Meanwhile... Z A ( n , p )and Z B ( n, p The difference exists The signal difference term is associated with the target's velocity, and velocity defuzzification can be performed based on this signal difference.

[0029] Furthermore, in wind shear detection scenarios, specifically... p The position of the spectral peak is estimated using the first-order moment calculation method. This invention proposes a comprehensive approach... Z A ( n , p ), Z A ( n , p The processing method of combining two sets of spectral data for spectral peak estimation, such as... Figure 5 As shown, the calculation process can be expressed by the following formula: ; Simultaneously extract the corresponding A-wave and B-wave data at the spectral peak. L ( n Spectral location data D A ( n )as well as D B ( n The processing can be represented by the following formula: ; In the above formula Indicates searching for the closest L ( n )of p After determining the position and completing the above operations, the Doppler frequency is further converted into a wind speed measurement value. The calculation process is expressed by the following formula: ; In the above formula V ( n ) indicates doors at different distances n The unclear wind speed measurement value differs from the correct wind speed measurement value. Further speed-based defuzzification processing is needed.

[0030] Furthermore, this invention summarizes a feasible speed defuzzification calculation process, as can be seen from the above analysis. D A ( n )and D B ( n The difference between ) The item is associated with the target's true velocity, which is related to the fuzzy velocity. V The relationship is as follows: ; In the above formula k Represents fuzzy numbers, fuzzy numbers k Size of the complete range of values K Determined by the following formula: ; consider K To ensure symmetry within the positive and negative velocity ranges, and considering the possibility of even numbers, the following set of fuzzy number values ​​was designed. Define the rules: ; For doors at different distances n Specific blur speed V ( n This invention summarizes an algorithm for solving fuzzy numbers, which can be summarized as follows: ; In the above formula Indicates the corresponding number n Numerical solution for fuzzy numbers of distance gates. express D A ( n The complex conjugate of ) is represented by the real(·) function, which extracts the real part from the complex number. This fuzzy number solving algorithm requires two complex multiplication operations per iteration, and the overall algorithm requires at most 2 K The algorithm performs multiple complex multiplications, and its time complexity is only [missing information]. O ( K It has the significant advantage of low computational cost.

[0031] Furthermore, based on Solving for distance gates n The calculation formula for the wind speed measurement after defuzzification is shown below: ; In the above formula V m ( n The figure () represents the wind speed measurement after defuzzification. This completes the defuzzification process for wind speed in wind shear detection. Then, based on... Figure 3 The wind shear detection process completes subsequent processing. The velocity de-ambiguity method disclosed in this invention, in Figure 3 The "wind speed estimation" step in the wind shear detection process shown is completed without extending the overall processing link of the wind shear detection process, and therefore will not affect the data refresh rate of wind shear detection.

[0032] Please continue reading Figure 2The radar waveform used in this invention is a combined waveform, designed by combining waveform group A and waveform group B to generate the radar waveform. The basic parameters of waveform group A and waveform group B are the same, but there is a fixed time delay. a The basic radar parameters and waveform parameters in this embodiment are shown in Table 1 below: Table 1. Basic Simulation Parameters and Radar Waveform Parameters:

[0033] Carrier frequencies in Table 1 f 0 is taken as 9.30 × 10 9 The Hz value represents the typical operating frequency of airborne weather radar, and the wavelength... λ Take 3.23 × 10 -2 m represents the corresponding wavelength value. The remaining waveform parameters are designed based on the application requirements of airborne weather radar in this embodiment.

[0034] Based on the parameter table in Table 1 and following the above explanation, the distance resolution can be further calculated. R ves Maximum unambiguous distance R max Speed ​​resolution Vves Maximum unambiguous speed V max fuzzy number range K The performance parameters and their corresponding results are summarized in Table 2 below.

[0035] Table 2 Performance Parameters:

[0036] The performance parameters in Table 2 are based on the waveform parameters in Table 1 and are calculated according to the formulas given above, where the fuzzy number... K In the calculation, because (2 T + a):( T The smallest integer ratio of + a) is 3:2, therefore gcd(3,2) = 1, and thus K = 3 / gcd(3,2) = 3. Therefore, by introducing velocity defuzzification processing, the measurement range of the target wind speed can be extended to (-3). V max , 3 V max It covers a wind speed measurement range of -48.3 m / s to +48.3 m / s, meeting the speed measurement range requirements of airborne meteorological scenarios. Meanwhile, the distance resolution... R ves Maximum unambiguous distance R maxSpeed ​​resolution Vves Maximum unambiguous speed V max The values ​​also meet the requirements for wind shear detection by airborne weather radar.

[0037] To verify the velocity deblurring algorithm disclosed in this invention, the effectiveness of the algorithm in this embodiment was verified in a simulation environment. First, to preliminarily verify the correctness of the velocity deblurring algorithm, this embodiment generated a single-point target signal with set distance and velocity in a simulation environment, obtained the ADC sampling data of the target echo, and then... Figure 4 The processing flow shown is used to process the data, analyze the results of each step of the speed defuzzification process of the present invention, and compare them with the target set value to verify the feasibility and correctness of the method of the present invention.

[0038] To verify the velocity deblurring algorithm disclosed in this invention, its effectiveness was tested in a simulation environment. First, to preliminarily verify the correctness of the velocity deblurring algorithm, this embodiment generated a single-point target signal with fixed distance and velocity in a simulation environment, obtained the ADC sampling data of the target echo, and then proceeded according to… Figure 4 The processing flow shown is executed, and the results of each step in the speed defuzzification process of this invention are analyzed. The results are then analyzed to verify the feasibility and correctness of the method of this invention. The distance to the target is set. R = 10000m, and simultaneously set the target speed. v = +30m / s, making the target velocity exceed V max To verify the speed defuzzification method disclosed in this invention.

[0039] The preliminary verification results are as follows Figure 6 As shown, Figure 6 The speed-based defuzzification process according to this invention is summarized as follows: Figure 4 As shown, the heat map is obtained from data from different stages. Figure 6 The upper part shows the processed results of waveform data from group A, and the lower part shows the processed results of waveform data from group B. The rearranged A and B wave data are shown below. S A ( n , m ), S B ( n , m The time-domain signal represented by ) is as follows Figure 6 As shown in the left image, you can see S A ( n , m )andS B ( n , m They correspond perfectly; the target bright lines are in the same position in the fast time dimension, after data rearrangement. S A ( n , m ), S B ( n , m The signal results are in line with expectations.

[0040] The results of A-wave and B-wave data after matched filtering are as follows: Figure 6 As shown in the upper half and the middle half of the lower half of the image, after matched filtering, the target data is transformed into a 2D spectrum of distance dimension-slow time dimension. It can be seen that the target presents obvious bright lines on the 2D spectrum. In addition, it can be seen that the matched filtering results of wave A and wave B data correspond to each other. The target position is the same in the distance dimension, and the result of matched filtering is as expected.

[0041] A-wave and B-wave data after m The result after 3D FFT processing is as follows Figure 6 The upper half and the lower right half of the image are shown. After... m After FFT processing, the target data is transformed into a 2D spectrum in the range-velocity dimension. It can be seen that the target exhibits a single-point spectral peak in the 2D spectrum. Furthermore, the matched filtering results for the A-wave and B-wave data are consistent, indicating that the target's position is the same in both the range and velocity dimensions. m The results of the 3D FFT processing are as expected.

[0042] After m After performing a 3D FFT, the spectral peak positions of the 2D spectrum in the distance-velocity dimension are estimated on a distance-by-distance basis to obtain the measured distance corresponding to the spectral peak positions of the target. R m = 10005m, fuzzy speed V = -2.01m / s, compared with the measured distance R m Distance from target R As can be seen, the measured distance is correct. Next, ambiguity numbers are calculated based on the A-wave and B-wave data, yielding the ambiguity numbers. κ = 1, thus obtaining the target's true measured velocity as V m = 30.18 m / s. Compare with measured velocity. V m relative to target speed vAs can be seen, the measured speed of the target is correct. The speed defuzzification algorithm of this invention correctly solves the fuzzy number and obtains the correct measured speed.

[0043] To further verify the velocity deambiguity algorithm designed in this invention, this embodiment performs simulation verification under a wind shear simulation scenario. Here, a wind shear numerical model is used to simulate wind field data, and radar echo signals are further obtained based on this. The velocity deambiguity algorithm of this invention is then used to process the echo signals, and finally, the processing results are analyzed. The wind shear wind field model parameters are set so that the velocity range in the radar observation direction covers the range from -38.2 m / s to 38.2 m / s; therefore, the target velocity will exceed the radar's maximum unambiguous velocity. V max Therefore, speed ambiguity will exist, which can be used to verify the speed deambiguity algorithm of the present invention.

[0044] Based on the above wind shear simulation scenario, according to Figure 4 The speed of defuzzification processing flow, m The 2D heatmap of wave A and wave B after FFT processing is shown below. Figure 7 As shown in the heatmap, the target's outline becomes significantly discontinuous, exhibiting a marked velocity blurring phenomenon.

[0045] The fuzzy velocity curve obtained by first-moment estimation based on 2D spectral data of A-wave and B-wave distance dimension-velocity dimension is shown below. Figure 8 As shown in the image above, a significant velocity blurring is clearly visible in the velocity curve. The velocity curve result after velocity deblurring is shown below. Figure 8 As shown in the image below, after velocity blurring, the velocity curve becomes continuous and exhibits the typical "S" curve characteristics of wind shear. (Comparison) Figure 8 As can be seen from the upper and lower figures, the velocity defuzzification algorithm of this invention effectively defuzzifies the target fuzzy velocity and correctly recovers the true measured velocity value.

[0046] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the claims.

Claims

1. A method for de-ambiguity detection velocity using airborne weather radar, characterized in that, Includes the following steps: S1. Using a grouped waveform design method, after determining the radar waveform parameters to generate grouped waveforms of A-wave and B-wave, they are combined to form a complete radar waveform. S2. Determine the ambiguity range, range resolution, maximum unambiguous range, velocity resolution, and maximum unambiguous velocity based on the design parameters of the combined radar waveform; S3. Receive the radar echo, rearrange the pulse Doppler radar echo sampling data to obtain fast time dimension-slow time dimension 2D data of group A and group B waveforms respectively. S4. Obtain the aircraft's own speed and horizontal and pitch angles relative to the radar from the local data source; S5. Perform matched filtering along the fast time dimension to obtain the processed signal data; S6. Perform compensation processing on the Doppler component of the aircraft's own velocity. S7. Perform FFT transformation along the slow time dimension on the compensated data to transform the slow time dimension into the velocity dimension. S8. The Doppler velocity spectrum of each range gate along the range dimension is processed by first-order moment calculation to calculate the target spectral peak velocity, and the A-wave and B-wave data of the corresponding Doppler frequency points are extracted. S9. Based on the A and B wave data, perform velocity defuzzification calculation to obtain the measured value of the defuzzified wind speed.

2. The airborne meteorological radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, The radar waveform in step S1 is obtained through a grouped waveform design method, specifically including: First, determine the fundamental period of the radar pulse waveform. T Time delay coefficient between wave A and wave B a Pulse width τ、 Pulse Amplitude A 0 and pulse count M And according to the repetition frequency period. T d = 2 T +a, pulse width τ、 Pulse Amplitude A 0 and pulse count M The parameters are set, and the pulse pair waveforms of group A are obtained by starting the first pulse waveform from time zero; then the entire waveform of group A is processed according to the time delay. T + a The process yields waveform group B; finally, waveform group A is combined with waveform group B to obtain the final radar deambiguity waveform design; waveform group A and waveform group B have the same repetition frequency period. T d = 2 T +a, and at the same time, there is a fixed phase difference between the waveforms in group A and group B. T +a.

3. The airborne meteorological radar wind shear detection velocity deambiguation method as described in claim 1, characterized in that, The size of the fuzzy number range in step S2 K From 2 T +a and T The +a determines this, and the corresponding expression is shown below: ； In the above formula, A / B represents (2 T +a) / ( T +a) is the smallest integer ratio, and gcd(·) represents the greatest common divisor solution function; the distance resolution in step S2 R ves Pulse width τ The decision, and the corresponding expression, is: ,in c Represents the speed of light constant; the maximum unambiguous distance in step S2. R max From the basic cycle T The decision, and the corresponding expression, is: The velocity resolution in step S2 V ves From the repetition frequency period T d and pulse count M The decision, and the corresponding expression, is: ,in λ Indicates the wavelength of the carrier wave; In step S2, the maximum unambiguous speed is determined by the repetition frequency period. T d The decision, and the corresponding expression, is: .

4. The airborne weather radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, Step S3 involves discretely sampling the echo signal at a fixed sampling rate, extracting A-wave and B-wave data at time intervals, while simultaneously sampling the A-wave at (m... T d , mT d + T Sampling is performed within a time period, and the B wave is within (m) T d +2 T , mT d +2 T + a Sampling is performed within a specified time period; here m Indicates the repetition frequency period T d Sequence numbers are used to ensure that the lengths of wave A and wave B data are aligned. For A-wave and B-wave data, according to the fast time dimension and m Data is reorganized using a two-dimensional approach to obtain 2D data with a fast time dimension and a slow time dimension, where the length of the fast time dimension is used as... N The sampling time sequence number of the fast time dimension is represented using... n This indicates that the rearranged A-wave and B-wave data were respectively used S A ( n , m )as well as S B ( n , m )express.

5. The airborne meteorological radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, Step S4 obtains the aircraft's ground speed from its own inertial navigation system. V F Simultaneously, the radar pointing angle relative to the aircraft is obtained from the radar's servo control. θ and pitch angle β This allows us to calculate the aircraft's ground speed along the radar's pointing direction. .

6. The airborne meteorological radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, The matched filtering process used in step S5 is performed in the frequency domain. First, the signal is transformed from the time domain to the frequency domain, and then inversely transformed back to the time domain. The corresponding calculation process can be summarized as follows: ； In the above formula x ( t () represents the radar fundamental pulse time-domain signal. x ( nT s )express x ( t The corresponding digital signal, express x ( nT s The complex conjugate signal of ) FFT n Indicates along n Perform FFT transformation on the dimension. IFFT n Indicates along n After performing inverse FFT transformation and matched filtering, the A-wave and B-wave data are respectively used... Y A ( n , m )as well as Y B ( n , m )express.

7. The airborne meteorological radar wind shear detection velocity deambiguation method as described in claim 1, characterized in that, The step S6 is in m Compensation is performed on the upper dimension, and the magnitude of the aircraft's ground speed along the radar pointing direction is: The corresponding compensation coefficient is determined to be The overall calculation process is as follows: ； In the above formula f 0 represents the radar carrier frequency. After compensation processing for the aircraft's own velocity Doppler component, the A-wave and B-wave data are used respectively. as well as express.

8. The airborne meteorological radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, In step S7, the input data is sampled and processed by FFT transformation in the slow time dimension. as well as along m The dimension is then subjected to an FFT transformation, which is represented by the following formula: ； In the above formula, p Representing spectral coefficients, corresponding to different velocities, after slow-time FFT transformation, the A-wave and B-wave data are respectively used... Z A ( n , p )as well as Z B ( n , p )express.

9. The airborne meteorological radar wind shear detection velocity de-ambiguation method as described in claim 1, characterized in that, The first-order moment calculation process in step S8 is based on spectral data. Z A ( n , p ), Z A ( n , p First-order moment calculations are performed to obtain the average Doppler frequency, which in turn yields the fuzzy wind speed measurement. The calculation process for the average Doppler frequency is expressed by the following formula: ； Simultaneously extract A-wave data and B-wave data corresponding to... L ( n Spectral location data D A ( n )as well as D B ( n The processing procedure is represented by the following formula: ； In the above formula Indicates searching for the closest L ( n )of p After determining the location and completing the above operations, the average Doppler frequency is further converted into an average wind speed measurement. The calculation process is expressed by the following formula: ； In the above formula V ( n ) indicates different correspondences n The wind speed measurement value was not defuzzified.

10. The airborne meteorological radar wind shear detection velocity deambiguation method as described in claim 1, characterized in that, The velocity defuzzification operation in step S9 affects the fuzzy number. k To solve this problem, we first determine the following rules: k The set of values : ； Then, the fuzzy numbers are calculated. The process of calculating the fuzzy numbers can be summarized by the following formula: ； In the above formula Indicates the corresponding number n Numerical solution for fuzzy numbers of distance gates. express D A ( n The complex conjugate of ), the real(·) function represents extracting the real part from the complex number, and based on this, further... V ( n The speed defuzzing process can be performed using the following formula: ； In the above formula V m ( n ) represents the wind speed measurement value after defuzzification.

Images