A time-frequency domain joint pulse compression processing method for long pulse high resolution wind measurement

Abstract

The application relates to the technical field of radar signal processing, and discloses a time-frequency domain combined pulse compression processing method for long-pulse high-resolution wind measurement, which comprises the following steps: long-pulse echo data is acquired and subjected to local time-frequency block processing; a theoretical track of a candidate distance unit in a time-frequency plane is constructed according to a propagation time delay; track time-frequency coefficients are extracted, and a frequency domain consistency factor, a local time-frequency energy focusing factor and a Doppler consistency factor are calculated; combined compression weights are generated; the amplitude of a time domain matched filter output is adjusted and the phase information is kept; combined pulse compression output is obtained; and Doppler spectrum estimation is carried out to obtain radial wind speed information. According to the application, the long-pulse echo data is subjected to local time-frequency block processing, and a theoretical track is constructed, so that the frequency domain consistency factor and the local time-frequency energy focusing factor are calculated, thereby reducing distance sidelobe leakage and improving distance resolution capability.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and in particular to a time-frequency domain combined pulse compression processing method for long-pulse high-resolution wind measurement. Background Technology

[0002] Wind radar detects wind speed and direction at different altitudes by emitting electromagnetic waves into the atmosphere and receiving the scattered echoes from air turbulence, aerosols, and precipitation particles.

[0003] In existing technologies, linear frequency modulated long pulses are typically used as the transmitted signal, and time-domain matched filtering is applied to the received echo to obtain the compressed range response. Some technical solutions also reduce the range sidelobe level through windowing, mismatched filtering, or sidelobe suppression algorithms. In actual processing, the echo signal is first arranged into fast and slow time data according to the pulse repetition period, and then matched filtering is applied to each pulse period to obtain the compression result of each range cell. Subsequently, Doppler spectral analysis is performed on the compressed result to obtain wind speed information.

[0004] However, the inventors of this application discovered in the process of implementing the above technical solution that when using overall matched filtering to process long pulse echoes, the echo energy of different range cells will be extended to adjacent range cells through range sidelobes, resulting in energy aliasing between adjacent range cells, thereby affecting the resolution of the range compression result. Summary of the Invention

[0005] To overcome the above deficiencies, this invention provides a time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement, which aims to improve the problem that the echo energy of different distance cells can be extended to adjacent distance cells through the range sidelobe, resulting in energy aliasing between adjacent distance cells.

[0006] This invention provides the following technical solution: a time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement, comprising the following steps: S1. Acquire the long pulse echo signal received by the wind measuring radar, and perform down-conversion, sampling and pulse period adjustment on the long pulse echo signal to form echo data arranged according to fast time and slow time. S2. Perform local time-frequency block processing on the echo data to obtain time-frequency coefficients, and construct the theoretical trajectory of each candidate distance unit in the time-frequency plane according to the long pulse frequency modulation law and the propagation delay corresponding to the candidate distance unit; S3. Extract the time-frequency coefficients of the trajectory along the theoretical trajectory corresponding to each candidate distance cell, and calculate the frequency domain consistency factor and local time-frequency energy focusing factor corresponding to each candidate distance cell based on the phase change relationship and energy distribution of the trajectory time-frequency coefficients. S4. Perform time-domain matched filtering on the echo data to obtain the time-domain matched filter output corresponding to each candidate range unit, and calculate the Doppler consistency factor corresponding to each candidate range unit based on the phase change relationship between adjacent pulse periods. S5. Based on the frequency domain consistency factor, the local time-frequency energy focusing factor and the Doppler consistency factor, generate joint compression weights corresponding to each candidate distance unit, and apply the joint compression weights to the time domain matched filter output to obtain joint pulse compression output; S6. Construct a slow time series based on the joint pulse compression output and perform Doppler spectrum estimation to obtain the radial wind speed information corresponding to each candidate distance unit.

[0007] Preferably, in S2, the step of performing local time-frequency block processing on the echo data includes: The echo data is segmented and truncated using a local analysis window; By sliding the local analysis window along the fast time direction, multiple local time positions are obtained; The time-frequency coefficients are obtained by performing frequency domain transformation on the data corresponding to each local time position.

[0008] Preferably, in S2, the step of constructing the theoretical trajectory of each candidate distance unit in the time-frequency plane includes: The theoretical frequency position corresponding to each local time position is determined based on the propagation delay corresponding to each candidate distance unit. Associate each local time position with its corresponding theoretical frequency position; The theoretical trajectory is formed by connecting the theoretical frequency positions in local time sequence.

[0009] Preferably, in S3, the steps for calculating the frequency domain consistency factor and local time-frequency energy focusing factor corresponding to each candidate distance cell include: Extract the time-frequency coefficients of the trajectory along the theoretical trajectory; Calculate the phase change between adjacent local time positions; The frequency domain consistency factor is generated based on the degree of dispersion of each phase change. Calculate the energy value of the time-frequency coefficients of the trajectory; The local time-frequency energy focusing factor is generated based on the overall energy and energy dispersion of each energy value.

[0010] Preferably, the step of extracting the time-frequency coefficients of the trajectory along the theoretical trajectory includes: The target sampling points in the time-frequency coefficients are determined based on the theoretical trajectories corresponding to each candidate distance unit. Read the complex values ​​corresponding to each target sampling point in local time sequence; The reading results are used to assemble the trajectory time-frequency coefficients according to the candidate distance units.

[0011] Preferably, in step S4, the step of calculating the Doppler consistency factor corresponding to each candidate distance cell includes: Calculate the phase change between adjacent pulse periods; The Doppler frequency corresponding to each pulse period is determined based on the phase change between adjacent pulse periods. Calculate the Doppler frequency change between adjacent pulse periods; The Doppler consistency factor is generated based on the degree of dispersion of each Doppler frequency variation.

[0012] Preferably, in S5, the step of generating the joint compression weights corresponding to each candidate distance unit includes: The frequency domain consistency factor, the local time-frequency energy focusing factor, and the Doppler consistency factor are combined to obtain a comprehensive evaluation value; The comprehensive evaluation values ​​corresponding to each candidate distance unit are normalized. The normalization result is used as the joint compression weight.

[0013] Preferably, in S5, the step of applying the joint compression weights to the time-domain matched filter output to obtain the joint pulse compression output includes: The amplitude of the corresponding time-domain matched filter output is adjusted according to the joint compression weights corresponding to each candidate distance unit; Preserve the phase information of the adjusted time-domain matched filter output; The time-domain matched filter output after amplitude and phase adjustment is used as the joint pulse compression output.

[0014] Preferably, in S6, the step of constructing a slow time series based on the joint pulse compression output and performing Doppler spectrum estimation includes: Arrange the joint pulse compression outputs corresponding to each candidate distance unit according to the pulse period order to construct a slow time series; Spectral analysis was performed on the slow time series to obtain the Doppler spectrum; Determine the dominant frequencies in the Doppler spectrum; The radial wind speed information is generated based on the dominant frequency.

[0015] Preferably, the step of determining the dominant frequency in the Doppler spectrum includes: Determine the frequency component with the highest power in the Doppler spectrum; The frequency corresponding to the frequency component with the highest power is taken as the dominant frequency.

[0016] The present invention has the following beneficial effects: 1. This invention reduces range sidelobe leakage and improves range resolution by performing local time-frequency block processing on long pulse echo data, constructing theoretical trajectories, and calculating frequency domain consistency factors and local time-frequency energy focusing factors.

[0017] 2. This invention improves the stability of radial wind speed estimation results by calculating the Doppler consistency factor and generating joint compression weights together with the frequency domain consistency factor and the local time-frequency energy focusing factor.

[0018] 3. This invention improves the reliability of long-pulse high-resolution wind measurement results by generating a joint pulse compression output based on the joint compression weight and performing Doppler spectrum estimation based on the joint pulse compression output. Attached Figure Description

[0019] Figure 1 This is a flowchart of a time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement proposed in this invention; Figure 2 This is a time-frequency processing schematic diagram of the time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement proposed in this invention; Figure 3 This is a schematic diagram of the joint compression and wind speed estimation of the time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement proposed in this invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Reference Figure 1 This invention provides a time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement. S1: Acquire the long-pulse echo signal received by the wind measurement radar, and perform down-conversion, sampling and pulse period adjustment on the long-pulse echo signal to form echo data arranged according to fast time and slow time.

[0022] Specifically, the wind-measuring radar transmits long-pulse frequency-modulated signals, receives atmospheric scattered echo signals, and performs radio frequency amplification, down-conversion, and analog-to-digital conversion processing through the receiver. The digital signal after analog-to-digital conversion is segmented according to the pulse repetition period. Each segment corresponds to fast-time sampling data within one pulse period, and the pulse periods are arranged in chronological order to form a slow-time sequence, thus constituting echo data arranged in fast and slow time sequences.

[0023] The transmitted long-pulse frequency-modulated signal is represented as: ; in, Indicates a long pulse frequency modulated signal; Represents the pulse envelope function; Indicates a fast time; Indicates the starting frequency; Indicates the frequency modulation slope; It represents the imaginary unit.

[0024] The echo signal is represented as: ; in, Indicates the first The echo signal corresponding to each pulse cycle; Indicates the pulse cycle number; Indicates the number of scattering units; Indicates the first The scattering unit at the th ... Complex scattering coefficients within one pulse cycle; Indicates the first The propagation delay corresponding to each scattering unit; Indicates the first The Doppler frequencies corresponding to each scattering unit; Indicates the pulse repetition period; This indicates the noise term.

[0025] No. The distance corresponding to each scattering unit is: ;in, Indicates the first The distance corresponding to each scattering unit; This represents the speed of electromagnetic wave propagation. In digital processing, the echo signal is discretized according to a fixed sampling interval. Let the fast-time sampling point number be... Then the first The discrete echo data corresponding to each pulse period is represented as follows: ;in, Indicates the first In the nth pulse cycle, the th The complex sampled values ​​corresponding to each fast time sampling point; This indicates the sequence number of the fast-time sampling point.

[0026] The discrete echo data corresponding to multiple consecutively received pulse cycles are arranged in order of pulse repetition period to form a two-dimensional echo data matrix: ; in, Represents the echo data matrix; This indicates the number of fast-time sampling points corresponding to each pulse cycle; This indicates the number of pulse cycles involved in the processing.

[0027] In the echo data matrix, each column corresponds to fast-time sampling data within one pulse period, and each row corresponds to sampling data at the same fast-time position across different pulse periods. The echo data matrix serves as input data for subsequent local time-frequency block processing, theoretical trajectory construction, joint pulse compression processing, and Doppler spectrum estimation.

[0028] Furthermore, S2, the echo data is processed by local time-frequency block segmentation to obtain time-frequency coefficients, and the theoretical trajectory of each candidate distance unit in the time-frequency plane is constructed according to the long pulse frequency modulation law and the propagation delay corresponding to the candidate distance unit. Preferably, in S2, the steps for performing local time-frequency block processing on the echo data include: The echo data is segmented and truncated using a local analysis window; Slide the local analysis window along the fast time direction to obtain multiple local time positions; The time-frequency coefficients are obtained by performing frequency domain transformation on the data corresponding to each local time position.

[0029] Preferably, in S2, the steps for constructing the theoretical trajectories of each candidate distance cell in the time-frequency plane include: The theoretical frequency position corresponding to each local time position is determined based on the propagation delay corresponding to each candidate distance unit. Associate each local time position with its corresponding theoretical frequency position; By connecting the theoretical frequency positions in local time sequence, a theoretical trajectory is formed.

[0030] Specifically, the echo data matrix obtained in step S1 Each column is processed using local time-frequency blocks. Let the local analysis window function be... , No. The center time corresponding to each local time position is ,in, Indicates the local time position number. Extracting the corresponding fast-time data from the center, multiplying it with the local analysis window function, and then performing a frequency domain transformation yields the time-frequency coefficients: ; in, Indicates the first Each pulse cycle at a local time position and frequency The corresponding time-frequency coefficients at that location; Represents frequency variables; Indicates the first The echo signal corresponding to each pulse cycle; Indicated by The local analysis window function is centered. The above process is repeated for all local time positions and all pulse periods to form a set of time-frequency coefficients containing three dimensions: local time position, frequency position, and pulse period.

[0031] For the Each candidate distance unit is first determined based on its corresponding distance. Calculate propagation delay: ;in, Indicates the first The propagation delay corresponding to each candidate distance unit; Indicates the first The distance corresponding to each candidate distance unit; Indicates the speed of electromagnetic wave propagation; This indicates the candidate distance cell number. Then, based on the starting frequency of the long pulse frequency-modulated signal... FM slope and local time position Determine the theoretical frequency position corresponding to the candidate distance cell at each local time position: ; in, Indicates the first The candidate distance unit in the th ... The theoretical frequency position corresponding to each local time position; Indicates the starting frequency; Indicates the frequency modulation slope. The local time positions are... Corresponding theoretical frequency position By matching the corresponding points and connecting them according to the chronological order of their local time positions, we obtain the first... The theoretical trajectories of candidate distance units in the time-frequency plane. The above process is repeated for all candidate distance units to form a set of theoretical trajectories corresponding to each candidate distance unit, which serves as the basis for subsequent extraction of trajectory time-frequency coefficients.

[0032] By performing local time-frequency block processing on the echo data and constructing the theoretical trajectory of each candidate distance unit in the time-frequency plane based on the propagation delay and long pulse frequency modulation law, each candidate distance unit has a corresponding sampling path in the time-frequency plane, providing input data for subsequent extraction of trajectory time-frequency coefficients along the theoretical trajectory and calculation of frequency domain consistency factor and local time-frequency energy focusing factor.

[0033] Reference Figure 2Furthermore, S3, extract the trajectory time-frequency coefficients along the theoretical trajectory corresponding to each candidate distance unit, and calculate the frequency domain consistency factor and local time-frequency energy focusing factor corresponding to each candidate distance unit based on the phase change relationship and energy distribution of the trajectory time-frequency coefficients.

[0034] Preferably, in S3, the steps for calculating the frequency domain consistency factor and local time-frequency energy focusing factor corresponding to each candidate distance cell include: Extracting the time-frequency coefficients of the trajectory along the theoretical trajectory; Calculate the phase change between adjacent local time positions; A frequency domain consistency factor is generated based on the degree of dispersion of each phase change. Energy values ​​of the time-frequency coefficients of statistical trajectories; A local time-frequency energy focusing factor is generated based on the overall energy and energy dispersion of each energy value.

[0035] Preferably, the steps for extracting the time-frequency coefficients of the trajectory along the theoretical trajectory include: The target sampling points in the time-frequency coefficients are determined based on the theoretical trajectories corresponding to each candidate distance unit. Read the complex values ​​corresponding to each target sampling point in local time sequence; The reading results are used to assemble trajectory time-frequency coefficients according to the candidate distance units.

[0036] Specifically, for the first Each candidate distance cell is used to determine the theoretical trajectory obtained in step S2. The corresponding target sampling points are determined from the time-frequency coefficient set, and the complex values ​​corresponding to each target sampling point are read in chronological order of local time positions to form the trajectory time-frequency coefficients: ; in, Indicates the first The candidate distance unit in the th ... The pulse cycle and the first The time-frequency coefficients of the trajectory corresponding to each local time position; This represents the time-frequency coefficient of the corresponding target sampling point; Indicates the candidate distance cell number; Indicates the local time position number; Indicates the pulse cycle number.

[0037] By taking the phase of the time-frequency coefficients of the trajectory, we obtain: ;in, This represents the phase value corresponding to the trajectory's time-frequency coefficients. The phase change between adjacent local time positions is calculated according to the local time sequence. ;in, This represents the phase change between adjacent local time positions. The average of all phase changes is calculated, and the deviation of each phase change from the average is statistically analyzed. A frequency domain consistency factor is generated based on the degree of deviation. ; in, Represents the frequency domain consistency factor; Indicates the total number of local time positions; This represents the average value of the phase change.

[0038] Calculate the energy value for the time-frequency coefficients of the trajectory: ;in, Indicates the first The energy values ​​correspond to a local time position. All energy values ​​are summed to obtain the total energy. The degree of energy dispersion is obtained by calculating the deviation of each energy value from the average energy. A local time-frequency energy focusing factor is generated based on the overall energy and the degree of energy dispersion. ; in, Indicates the local time-frequency energy focusing factor; Represents total energy; It indicates the degree of energy dispersion.

[0039] The above processing is repeated for each candidate range cell and each pulse period to obtain the corresponding frequency domain consistency factor and local time-frequency energy focusing factor. The frequency domain consistency factor is used to characterize the consistency of phase changes on the theoretical trajectory, and the local time-frequency energy focusing factor is used to characterize the energy concentration on the theoretical trajectory. Through the above processing, the subsequent joint compression weight generation process utilizes both phase distribution information and energy distribution information on the theoretical trajectory, thereby providing input parameters for reducing range sidelobe leakage.

[0040] Furthermore, S4, time-domain matched filtering is performed on the echo data to obtain the time-domain matched filter output corresponding to each candidate range cell, and the Doppler consistency factor corresponding to each candidate range cell is calculated based on the phase change relationship between adjacent pulse periods.

[0041] Preferably, in step S4, the step of calculating the Doppler consistency factor corresponding to each candidate distance cell includes: Calculate the phase change between adjacent pulse periods; The Doppler frequency corresponding to each pulse period is determined based on the phase change between adjacent pulse periods. Calculate the Doppler frequency change between adjacent pulse periods; Doppler consistency factors are generated based on the degree of dispersion of each Doppler frequency variation.

[0042] Specifically, for the first Each candidate distance unit is determined based on its corresponding propagation delay. The echo signal obtained in step S1 is correlated with the conjugate delayed signal of the long pulse frequency modulated signal to obtain the time-domain matched filter output: ; in, Indicates the first The candidate distance unit in the th ... Time-domain matched filter output corresponding to each pulse period; Indicates the first The echo signal corresponding to each pulse cycle; This represents the conjugate signal of a long-pulse frequency-modulated signal after propagation delay compensation; Indicates the first The propagation delay corresponding to each candidate distance unit; Indicates the candidate distance cell number; Indicates the pulse cycle number.

[0043] For the time-domain matched filter output of the same candidate distance cell in adjacent pulse periods, calculate the phase change between adjacent pulse periods: ;in, Indicates the first The pulse cycle and the first Phase change between pulse cycles; express The conjugate value; This indicates taking the complex phase.

[0044] The Doppler frequency corresponding to each pulse period is determined based on the phase change between adjacent pulse periods: ; in, Indicates the first The Doppler frequency corresponding to each pulse cycle; This represents the pulse repetition period. The difference between adjacent Doppler frequencies is calculated sequentially according to the pulse period to obtain the Doppler frequency variation. The average value and dispersion of all Doppler frequency variations are then calculated. Based on the deviation of each Doppler frequency variation from the average value, a Doppler consistency factor is generated. ; in, Indicates the first Doppler consistency factor corresponding to each candidate distance unit; Indicates the number of pulse cycles involved in the processing; This represents the change in Doppler frequency between adjacent pulse periods; This represents the average value of the Doppler frequency change.

[0045] The above process is repeated for all candidate range cells to obtain the Doppler consistency factor for each candidate range cell. The Doppler consistency factor is used to characterize the stability of the Doppler frequency change in the slow time direction for the same candidate range cell, providing input parameters for the subsequent generation of joint compression weights, thereby enabling the joint pulse compression process to include phase change information in the slow time direction.

[0046] Reference Figure 3 Furthermore, S5 generates joint compression weights corresponding to each candidate distance cell based on the frequency domain consistency factor, local time-frequency energy focusing factor, and Doppler consistency factor, and applies the joint compression weights to the time-domain matched filter output to obtain the joint pulse compression output.

[0047] Preferably, in S5, the step of generating the joint compression weights corresponding to each candidate distance unit includes: The frequency domain consistency factor, local time-frequency energy focusing factor, and Doppler consistency factor are combined to obtain a comprehensive evaluation value; The comprehensive evaluation values ​​corresponding to each candidate distance unit are normalized. The normalization result is used as the joint compression weight.

[0048] Preferably, in S5, the step of applying the joint compression weights to the time-domain matched filter output to obtain the joint pulse compression output includes: The amplitude of the corresponding time-domain matched filter output is adjusted according to the joint compression weights corresponding to each candidate distance unit; Preserve the phase information of the adjusted time-domain matched filter output; The time-domain matched filter output after amplitude and phase adjustment is used as the joint pulse compression output.

[0049] Specifically, for the first The candidate distance unit, at the nth candidate distance unit, Within each pulse period, the frequency domain consistency factor will be... Local time-frequency energy focusing factor And Doppler consistency factor Multiply by each product to obtain the overall evaluation value: ; in, Indicates the first The candidate distance unit in the th ... The comprehensive evaluation value corresponding to each pulse cycle; Represents the frequency domain consistency factor; Indicates the local time-frequency energy focusing factor; This represents the Doppler consistency factor.

[0050] The comprehensive evaluation values ​​corresponding to all candidate distance units are normalized to obtain the joint compression weights: ;in, Indicates the first The candidate distance unit in the th ... The joint compression weights corresponding to each pulse period; Indicates the total number of candidate distance cells; This represents the candidate range cell number in the normalization calculation. Based on the joint compression weights corresponding to each candidate range cell, the amplitude of the time-domain matched filter output obtained in step S4 is adjusted while maintaining its phase information, resulting in the joint pulse compression output: ; in, Indicates the first The candidate distance unit in the th ... The combined pulse compression output corresponding to each pulse cycle; This indicates the output of the time-domain matched filter.

[0051] In the actual processing, the complex values ​​of the time-domain matched filter output corresponding to each candidate range cell are first read. Then, the magnitude of the complex values ​​is scaled according to the joint compression weights while keeping the argument of the complex values ​​unchanged. The adjusted complex values ​​are then written into the joint pulse compression output matrix. The above process is repeated for all candidate range cells and all pulse periods to form the joint pulse compression output data.

[0052] By using the frequency domain consistency factor, local time-frequency energy focusing factor, and Doppler consistency factor together to generate joint compression weights, and then using these joint compression weights to adjust the amplitude of the time-domain matched filter output, the candidate range cells with stable theoretical trajectory phase changes, concentrated trajectory energy, and stable Doppler changes retain a large amplitude in the joint pulse compression output, thus providing input data for subsequent Doppler spectrum estimation.

[0053] Furthermore, S6, a slow time series is constructed based on the joint pulse compression output, and Doppler spectrum estimation is performed to obtain the radial wind speed information corresponding to each candidate distance unit.

[0054] Preferably, in S6, the steps of constructing a slow time series based on the joint pulse compression output and performing Doppler spectrum estimation include: Arrange the joint pulse compression outputs corresponding to each candidate distance unit according to the pulse period order to construct a slow time series; Spectral analysis was performed on the slow time series to obtain the Doppler spectrum; Determine the dominant frequencies in the Doppler spectrum; Radial wind speed information is generated based on the dominant frequency.

[0055] Preferably, the steps for determining the dominant frequencies in the Doppler spectrum include: Identify the frequency component with the highest power in the Doppler spectrum; The frequency corresponding to the frequency component with the highest power is taken as the dominant frequency.

[0056] Specifically, for the first The candidate distance units will compress and output the joint pulse obtained in step S5. Arrange them according to the pulse period order to construct a slow time series: ; in, Indicates the first Slow time series corresponding to each candidate distance unit; Indicates the first The combined pulse compression output corresponding to each pulse cycle; This indicates the number of pulse cycles involved in the processing.

[0057] Spectral analysis of the slow time series yields the Doppler spectrum: ; in, Indicates the first The candidate distance unit in the th ... Doppler spectral power corresponding to each frequency point; Indicates the frequency point number.

[0058] By comparing various frequency points in the Doppler spectrum, the frequency component with the highest power is determined, and the frequency corresponding to this component is taken as the dominant frequency. . No. The radial wind speed information corresponding to each candidate distance cell is: ;in, Indicates the first Radial wind speed information corresponding to each candidate distance cell; Indicates the operating wavelength of the wind-measuring radar; Indicates the dominant frequency.

[0059] In the actual processing, slow time series were constructed and spectral analysis was performed on all candidate distance units to obtain the Doppler spectrum, dominant frequency and radial wind speed information corresponding to each candidate distance unit, and wind speed profile data were formed according to distance order.

[0060] By constructing a slow time series from the joint pulse compression output and performing Doppler spectrum estimation, the distance compression result and the wind speed estimation process use the same set of jointly constrained data, thereby obtaining the radial wind speed information corresponding to each candidate distance unit.

[0061] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement, characterized in that, Includes the following steps: S1. Acquire the long pulse echo signal received by the wind measuring radar, and perform down-conversion, sampling and pulse period adjustment on the long pulse echo signal to form echo data arranged according to fast time and slow time. S2. Perform local time-frequency block processing on the echo data to obtain time-frequency coefficients, and construct the theoretical trajectory of each candidate distance unit in the time-frequency plane according to the long pulse frequency modulation law and the propagation delay corresponding to the candidate distance unit; S3. Extract the time-frequency coefficients of the trajectory along the theoretical trajectory corresponding to each candidate distance cell, and calculate the frequency domain consistency factor and local time-frequency energy focusing factor corresponding to each candidate distance cell based on the phase change relationship and energy distribution of the trajectory time-frequency coefficients. S4. Perform time-domain matched filtering on the echo data to obtain the time-domain matched filter output corresponding to each candidate range unit, and calculate the Doppler consistency factor corresponding to each candidate range unit based on the phase change relationship between adjacent pulse periods. S5. Based on the frequency domain consistency factor, the local time-frequency energy focusing factor and the Doppler consistency factor, generate joint compression weights corresponding to each candidate distance unit, and apply the joint compression weights to the time domain matched filter output to obtain joint pulse compression output; S6. Construct a slow time series based on the joint pulse compression output and perform Doppler spectrum estimation to obtain the radial wind speed information corresponding to each candidate distance unit.

2. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S2, the step of performing local time-frequency block processing on the echo data includes: The echo data is segmented and truncated using a local analysis window; By sliding the local analysis window along the fast time direction, multiple local time positions are obtained; The time-frequency coefficients are obtained by performing frequency domain transformation on the data corresponding to each local time position.

3. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S2, the step of constructing the theoretical trajectory of each candidate distance unit in the time-frequency plane includes: The theoretical frequency position corresponding to each local time position is determined based on the propagation delay corresponding to each candidate distance unit. Associate each local time position with its corresponding theoretical frequency position; The theoretical trajectory is formed by connecting the theoretical frequency positions in local time sequence.

4. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S3, the steps for calculating the frequency domain consistency factor and local time-frequency energy focusing factor for each candidate distance cell include: Extract the time-frequency coefficients of the trajectory along the theoretical trajectory; Calculate the phase change between adjacent local time positions; The frequency domain consistency factor is generated based on the degree of dispersion of each phase change. Calculate the energy value of the time-frequency coefficients of the trajectory; The local time-frequency energy focusing factor is generated based on the overall energy and energy dispersion of each energy value.

5. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 4, characterized in that, The step of extracting the time-frequency coefficients of the trajectory along the theoretical trajectory includes: The target sampling points in the time-frequency coefficients are determined based on the theoretical trajectories corresponding to each candidate distance unit. Read the complex values ​​corresponding to each target sampling point in local time sequence; The reading results are used to assemble the trajectory time-frequency coefficients according to the candidate distance units.

6. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In step S4, the step of calculating the Doppler consistency factor corresponding to each candidate distance cell includes: Calculate the phase change between adjacent pulse periods; The Doppler frequency corresponding to each pulse period is determined based on the phase change between adjacent pulse periods. Calculate the Doppler frequency change between adjacent pulse periods; The Doppler consistency factor is generated based on the degree of dispersion of each Doppler frequency variation.

7. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S5, the steps for generating the joint compression weights for each candidate distance unit include: The frequency domain consistency factor, the local time-frequency energy focusing factor, and the Doppler consistency factor are combined to obtain a comprehensive evaluation value; The comprehensive evaluation values ​​corresponding to each candidate distance unit are normalized. The normalization result is used as the joint compression weight.

8. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S5, the step of applying the joint compression weights to the time-domain matched filter output to obtain the joint pulse compression output includes: The amplitude of the corresponding time-domain matched filter output is adjusted according to the joint compression weights corresponding to each candidate distance unit; Preserve the phase information of the adjusted time-domain matched filter output; The time-domain matched filter output after amplitude and phase adjustment is used as the joint pulse compression output.

9. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 1, characterized in that, In S6, the steps of constructing a slow time series based on the joint pulse compression output and performing Doppler spectrum estimation include: Arrange the joint pulse compression outputs corresponding to each candidate distance unit according to the pulse period order to construct a slow time series; Spectral analysis was performed on the slow time series to obtain the Doppler spectrum; Determine the dominant frequencies in the Doppler spectrum; The radial wind speed information is generated based on the dominant frequency.

10. The time-frequency domain joint pulse compression processing method for long-pulse high-resolution wind measurement according to claim 9, characterized in that, The step of determining the dominant frequency in the Doppler spectrum includes: Determine the frequency component with the highest power in the Doppler spectrum; The frequency corresponding to the frequency component with the highest power is taken as the dominant frequency.

Images