A weather radar range resolution calibration method and system

Abstract

The present application relates to the field of radar technology, especially to a weather radar range resolution calibration method and system. The method comprises the following steps: collecting oversampling original baseband I / Q data of a point target in multiple pulse periods; determining pulse signal-to-noise ratio, and screening effective pulses according to a threshold value; performing time alignment of the effective pulses in a sub-sampling stage based on cross-correlation analysis, performing non-coherent average of power modulus of all the aligned pulses, and generating an average pulse power envelope; converting the average pulse power envelope to a logarithmic domain, and selecting discrete data points located in a main lobe region; using a nonlinear least squares method to iteratively fit a generalized super-Gaussian mathematical model according to the discrete data points, and obtaining a continuous analytical function describing the pulse envelope; determining a pulse time width at which the power drops to a peak value-6dB according to the continuous analytical function; and determining a real range resolution according to the pulse time width. The measurement robustness is improved, and the calibration accuracy is sub-meter.

Description

Technical Field

[0001] This invention relates to the field of radar technology, and in particular to a method and system for calibrating the range resolution of a weather radar. Background Technology

[0002] Range resolution is one of the most fundamental performance indicators of weather radar, defining its ability to distinguish two or more adjacent targets in the radial range dimension. High range resolution is crucial for modern meteorological observation, especially in the monitoring and research of small- and medium-scale severe convective weather. For example, to accurately identify intricate hazardous weather structures such as tornado vortex characteristics, gust fronts, and hail three-body scattering, a radar range resolution better than 250 meters is typically required. If the radar's actual range resolution is insufficient, echoes from adjacent meteorological targets will overlap in the range dimension, leading to ambiguity and distortion of key weather structures, which in turn severely affects the accuracy of quantitative precipitation estimation and the timeliness and accuracy of severe convective weather warnings.

[0003] In theory, the distance resolution is determined by the compressed effective pulse width: ;

[0004] In the formula, For theoretical distance resolution, At the speed of light, This represents the effective pulse width after compression. However, in actual radar operation, the signal undergoes complex processes such as transmission, propagation, reception, and processing. Distortions introduced at each stage, such as amplifier nonlinearity, filter bandwidth limitations, and quantization noise, lead to pulse waveform broadening and distortion, resulting in actual range resolution significantly worse than the theoretical value. Accurately measuring the true range resolution is crucial for evaluating radar performance, data fusion, and system verification.

[0005] Currently, the industry mainly relies on two distance resolution calibration methods, as follows:

[0006] 1. Dual-target field test method: Place two corner reflectors or standard metal spheres in the radar's far field. By adjusting the distance between the two corner reflectors or standard metal spheres, observe whether the radar echoes can be distinguished. This method can only make a qualitative judgment, obtaining a vague conclusion that "it can be distinguished at a distance of X meters, but cannot be distinguished at a distance of Y meters," without providing precise values. Furthermore, it is cumbersome to operate, inefficient, and easily affected by environmental interference. Moreover, it cannot pinpoint faulty components, such as transmitter, receiver, or processor problems, when the resolution is substandard.

[0007] 2. Single-pulse envelope measurement method: This method directly measures the -3dB width of a single echo pulse from a point target to calculate the range resolution. However, this method has several inherent drawbacks: a) Picket fence effect: The range sampling interval of traditional weather radar is usually set close to the theoretical resolution, such as 150 meters. Although this sampling satisfies the Nyquist sampling theorem in the frequency domain, it is extremely sparse in the time domain. This results in only one or two sampling points within a pulse width, causing the true pulse peak and the key edge points for determining the width to often fall between adjacent sampling points. Directly reading the pulse width based on discrete sampling data introduces a quantization error of up to half a sampling interval, which can result in an error of 75 meters at a 150m length. This is problematic for radars requiring high precision. a) This is completely unacceptable for modern calibration requirements; b) The industry has long been accustomed to using the -3dB standard of the transmitted pulse, i.e., the half-power point, to measure the width of the received pulse, ignoring the fact that radar detection is a two-way physical process of "transmission-reflection-reception". Using the -3dB standard will lead to a systematic deviation in the measurement of the received pulse width, and will not accurately reflect the true physical limit of radar in distinguishing two adjacent targets; c) Relying solely on the measurement of a single pulse makes the results highly susceptible to the influence of system thermal noise, transient interference, and other factors, resulting in large fluctuations, poor repeatability, and low confidence in the measurement data, which cannot meet the high reliability calibration requirements; d) Traditional methods often use standard Gaussian model fitting, which cannot accurately describe the "flat-top" pulse characteristics generated when modern transmitters operate in the saturation region.

[0008] Therefore, current methods for calibrating the range resolution of weather radar cannot meet the high-precision and quantitative requirements of modern weather radar. There is an urgent need for a high-precision range resolution calibration method that can overcome the limitations of discrete sampling, adapt to complex waveform characteristics, and conform to the physical mechanisms of radar.

[0009] It should be noted that Chinese patent application CN120143069A discloses a method and system for calibrating observation errors in multi-beam scanning mode of X-band phased array weather radar based on statistical results. This method uses statistical results to collect observation base data from S-band and X-band radars, performs data quality control and preprocessing, and utilizes different radar temporal and spatial sampling volume coordinate matching methods, using S-POL observations as a benchmark, to statistically analyze and correct the observation errors of X-PAR. Chinese patent application CN119846577A discloses a method for quantitatively generating the reflectivity factor of simulated targets used in far-field calibration of weather radar. This method obtains radar antenna and transmission parameters, the transmit / receive ratio of the target simulator, and the distance between the installation point. Using a pre-constructed quantitative model of the equivalent cross-sectional area of ​​the simulated target, the equivalent cross-sectional area of ​​the simulated target is calculated, and based on this value and other parameters, the expected value of the radar reflectivity factor is calculated. Chinese patent application CN117930156A discloses a weather radar calibration method and device based on a metal ball. The method uses a metal ball to suspend a metal ball by a UAV, and uses radar beam directional scanning and reflectivity factor calculation to acquire radar calibration data in real time. The method calculates the beamwidth and pulse width of the radar antenna, determines the radar calibration coefficient, and improves the radar detection accuracy. Summary of the Invention

[0010] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method and system for calibrating the range resolution of weather radar, which improves the measurement robustness and has sub-meter level calibration accuracy.

[0011] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0012] In a first aspect, the present invention provides a method for calibrating the range resolution of a weather radar, based on aligning the radar antenna with a far-field point target, comprising the following steps:

[0013] Oversampled raw baseband I / Q data of the target point over multiple consecutive pulse cycles;

[0014] Based on the oversampled raw baseband I / Q data, the signal-to-noise ratio (SNR) of each pulse is determined; effective pulses are selected based on the SNR and a preset threshold; subsampling-level time alignment is performed on the effective pulses based on cross-correlation analysis; and the power magnitude of all time-aligned pulses is noncoherently accumulated and averaged to generate the average pulse power envelope.

[0015] The average pulse power envelope is transformed to the logarithmic domain, and discrete data points located within a predefined main lobe region are selected. Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using a nonlinear least squares method with boundary constraints to obtain a continuous analytic function describing the pulse envelope.

[0016] Based on the continuous analytical function, the effective pulse duration when the power drops to -6dB from the peak value is determined; based on the effective pulse duration, the true range resolution of the weather radar is determined.

[0017] Optionally, the oversampled raw baseband I / Q data is obtained by: the radar receiver directly oversampling at a sampling interval smaller than the theoretical range resolution; and / or, the radar receiver samples to obtain the raw baseband I / Q data, and performs frequency domain interpolation oversampling processing on the raw baseband I / Q data to obtain the oversampled raw baseband I / Q data.

[0018] Optionally, frequency domain interpolation oversampling processing is performed on the original baseband I / Q data, including: performing a fast Fourier transform on the original baseband I / Q data of length N to obtain its frequency domain spectrum; constructing an extended spectrum sequence of length N×K according to the target oversampling factor K, wherein the positive frequency part of the frequency domain spectrum is placed at the low-frequency end of the extended spectrum sequence, the negative frequency part of the frequency domain spectrum is placed at the high-frequency end of the extended spectrum sequence, and zero values ​​are filled between the low-frequency end and the high-frequency end to maintain the conjugate symmetry of the Nyquist frequency components; performing an inverse fast Fourier transform on the extended spectrum sequence, and multiplying the transform result by a gain factor K to obtain the original baseband I / Q data oversampled by a factor of K.

[0019] Optionally, subsampling-level time alignment of the effective pulses is performed based on cross-correlation analysis, including: selecting the effective pulse with the highest signal-to-noise ratio as the reference pulse; determining the cross-correlation function between the remaining effective pulses and the reference pulse, and determining the time offset of each pulse relative to the reference pulse based on the peak position of the cross-correlation function; and performing frequency domain phase shift correction on each effective pulse according to the time offset to achieve subsampling-level time alignment.

[0020] Optionally, the discrete data points within the pre-defined main lobe region are: all discrete data points whose power values ​​decrease along the falling edges on both sides of the center from the peak point of the average pulse power envelope to a preset power level.

[0021] Optionally, based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method to obtain a continuous analytic function describing the pulse envelope, including:

[0022] The generalized superGaussian mathematical model is expressed as: ;

[0023] The boundary constraints are that during the iteration process, the parameter w must be a positive number and n must be within a preset range;

[0024] In the formula:

[0025] : Pulse power envelope function, representing the time axis Above is the instantaneous power value of the radar echo after incoherent accumulation;

[0026] Peak power refers to the theoretical maximum power value of the main lobe of the pulse envelope obtained by fitting.

[0027] Time variable;

[0028] The pulse center moment refers to the moment when the pulse envelope reaches its peak power. The corresponding time position;

[0029] : Pulse width scaling parameter. This parameter characterizes the characteristic width of the pulse, and its specific physical meaning is determined by the shape order. Joint decision;

[0030] Shape order, a dimensionless positive real number, controls the flatness of the top and the steepness of the edges of the pulse waveform;

[0031] : Basis noise power, refers to the average power of thermal noise and background interference in the receiver system, and is used as the DC bias term of the fitting model;

[0032] Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method to solve for the parameters. , , , and , thus obtaining a continuous analytic function.

[0033] Optionally, the effective pulse duration is determined based on the continuous analytical function, including: determining the time parameter that satisfies the power drop to 1 / 4 of the peak power based on the continuous analytical function, and determining the effective pulse duration based on the time parameter.

[0034] Optionally, the weather radar range resolution calibration method further includes: performing diagnostic analysis on the performance status of the radar system based on the morphological characteristics of the continuous analytical function.

[0035] Optionally, based on the morphological characteristics of the continuous analytic function, a diagnostic analysis of the radar system's performance status is performed, including at least one of the following:

[0036] The main lobe broadening factor is determined based on the ratio of the actual distance resolution to the theoretical distance resolution. Based on the main lobe broadening factor and the first threshold, it is determined whether the intermediate frequency filter bandwidth of the receiver is too narrow and / or whether the matched filter coefficients are mismatched.

[0037] Based on the -6dB points on both sides of the peak of the continuous analytical function, the waveform symmetry factor is determined. Based on the degree to which the waveform symmetry factor deviates from the set value, the I / Q channel balance and / or filter group delay characteristics are judged.

[0038] The difference between the peak value of the main lobe and the peak value of the first side lobe is determined by the continuous analytical function. Based on the difference and the second threshold, it is determined whether the compression coefficient of the pulse compression system is mismatched and / or whether the transmitter is operating in the nonlinear region, resulting in spectrum regeneration.

[0039] The fitting residual is calculated based on the difference between the continuous analytical function and the discrete data points. Based on the fitting residual and the third threshold, it is determined whether there is multipath interference and / or hardware fault.

[0040] In a second aspect, the present invention provides a weather radar range resolution calibration system, comprising:

[0041] The acquisition module is used to acquire oversampled raw baseband I / Q data of a point target over multiple consecutive pulse cycles;

[0042] The signal enhancement module is used to determine the signal-to-noise ratio of each pulse based on the oversampled original baseband I / Q data; to filter out effective pulses based on the signal-to-noise ratio and a preset threshold; to perform subsampling-level time alignment on the effective pulses based on cross-correlation analysis; and to perform incoherent cumulative averaging on the power magnitude of all time-aligned pulses to generate an average pulse power envelope.

[0043] The waveform reconstruction module is used to convert the average pulse power envelope to the logarithmic domain and select discrete data points located in the pre-defined main lobe region. Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method with boundary constraints to obtain a continuous analytic function describing the pulse envelope.

[0044] The range resolution determination module is used to determine the effective pulse duration when the power drops to -6dB of the peak value based on a continuous analytical function; and to determine the true range resolution of the weather radar based on the effective pulse duration.

[0045] The beneficial effects of this invention are:

[0046] The range resolution calibration method and system provided by this invention mathematically and losslessly recover the waveform information between sampling points through frequency domain interpolation oversampling, completely overcoming the picket fence effect caused by the ADC sampling rate limitation in traditional methods. It replaces the single Gaussian model with a generalized super-Gaussian model, adaptively fitting non-ideal waveforms such as flat tops and steep edges generated by modern radar transmitters (such as klystrons and solid-state power amplifiers), significantly reducing fitting residuals and improving range resolution measurement accuracy to sub-meter level, achieving true quantitative measurement. Following the -6dB criterion established by the radar two-way detection physical mechanism, it corrects the industry's misuse of the -3dB width, strictly based on the two-way propagation mechanism of the radar equations, establishing -6dB as the calibration benchmark for range resolution, ensuring that the calibration results truly reflect the radar's ability to distinguish adjacent targets and guaranteeing the physical authenticity of the calibration results. With the help of multi-pulse alignment and incoherent accumulation techniques, it significantly improves anti-interference capability and measurement robustness. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating the weather radar range resolution calibration method according to a specific embodiment of the present invention;

[0048] Figure 2 The image shows a pulse waveform obtained by frequency domain interpolation and oversampling of the original baseband I / Q data according to a specific embodiment of the present invention.

[0049] Figure 3 This is a schematic diagram of the pulse power envelope after performing an incoherent cumulative average of all effective pulses according to a specific embodiment of the present invention.

[0050] Figure 4 This is a schematic diagram of the pulse envelope obtained by fitting discrete data points in the logarithmic domain using the generalized super-Gaussian mathematical model according to a specific embodiment of the present invention. Detailed Implementation

[0051] To better explain and facilitate understanding of the present invention, it is described in detail below with reference to the accompanying drawings and specific embodiments. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a clearer and more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0052] like Figure 1 The diagram shown is a flowchart illustrating the weather radar range resolution calibration method provided by this invention. This weather radar range resolution calibration method includes the following steps:

[0053] Step S1: Oversample the raw baseband I / Q data of the target point for multiple consecutive pulse cycles.

[0054] Specifically, select a flat, open area with a wide field of view, and choose a high-precision point target with isotropic scattering characteristics, such as a precision-machined metal sphere or a standard corner reflector. Position the point target in the far-field region of the radar antenna to ensure plane wave illumination conditions. Simultaneously, ensure that there are no strongly reflective objects around the point target to avoid clutter interference.

[0055] Preferably, the point target range radar should meet the far-field conditions. In the formula, For far-field distance, For radar antenna aperture, For wavelength. As an example, S-band weather radar, For radar antenna aperture, For wavelength. As an example, S-band weather radar, It is 8.5m. If it is 0.1m, then The range is 1.445 km. Generally speaking, for S-band weather radar, a point target distance of 2-5 km from the radar is recommended.

[0056] Specifically, the antenna azimuth and elevation angles are locked so that the beam center is directly facing the point target.

[0057] Specifically, the radar is placed in the calibration mode, such as a specific volume scan mode (VCP), and all back-end signal processing algorithms that may alter the original pulse shape are temporarily disabled, including but not limited to ground clutter suppression, range smoothing, and signal threshold clipping. The purpose of this operation is to obtain raw baseband I / Q data that accurately reflects the inherent characteristics of the radar's transmission, reception, and front-end processing links.

[0058] To overcome the "pick-fence effect" in traditional range resolution calibration methods, high-sampling-density raw baseband I / Q data is required. Depending on the radar hardware capabilities, one or a combination of the following two strategies can be employed:

[0059] If the radar's digital intermediate frequency receiver supports a variable decimation rate, then the data output rate after the analog-to-digital converter (ADC) can be directly increased, thus improving the actual distance sampling interval. Significantly smaller than the theoretical distance resolution This yields oversampled raw baseband I / Q data. Preferably, this allows... , where M≥2, usually taken as 4 to 10.

[0060] If the radar hardware sampling rate is fixed, the raw baseband I / Q data will first be acquired at the conventional sampling rate. The length is N. Then, frequency domain interpolation and oversampling are performed on the original baseband I / Q data to obtain oversampled original baseband I / Q data. The specific process is as follows: [The text abruptly ends here, likely due to an incomplete sentence or missing information.] Perform an N-point Fast Fourier Transform (FFT) to obtain its spectrum. Construct a length of (K is the oversampling factor, K≥2) extended spectrum sequence ,Will The positive frequency part is placed The low-frequency end will The negative frequency part is placed The high-frequency end, and fill the middle of the high-frequency end. Each zero value preserves the conjugate symmetry of the Nyquist frequency components of the spectrum; for conduct The point-wise inverse fast Fourier transform (IFFT) is then multiplied by a gain factor K to obtain the original baseband I / Q data after K times oversampling. This process is mathematically equivalent to time-domain interpolation using an ideal low-pass filter with the Sinc function, which can accurately reconstruct the pulse waveform without introducing high-frequency harmonic distortion, such as... Figure 2 As shown.

[0061] Specifically, the gain or attenuator of the radar receiver is adjusted so that the peak power of the point target echo signal is within the optimal range of the analog-to-digital converter's (ADC) linear dynamic range, such as 70% to 85% of full scale. This operation aims to balance high signal-to-noise ratio with avoiding signal saturation, preventing saturation clipping caused by excessively strong signals, and preventing nonlinear distortion from affecting the accuracy of subsequent pulse waveform measurements.

[0062] Step S2: Determine the signal-to-noise ratio (SNR) of each pulse based on the oversampled original baseband I / Q data; filter out effective pulses based on the SNR and a preset threshold; perform subsampling-level time alignment on the effective pulses based on cross-correlation analysis; perform incoherent cumulative averaging on the power magnitude of all time-aligned pulses to generate an average pulse power envelope.

[0063] Specifically, as an example, the signal-to-noise ratio (SNR) threshold is 20dB, and pulses with an SNR lower than this threshold are discarded. This operation aims to exclude low-quality pulse data affected by factors such as sudden radio frequency interference, target flicker, or transient system instability, ensuring the purity and consistency of subsequent processed data.

[0064] Due to inherent timing jitter in radar systems or minor fluctuations caused by wind and other loads on point targets, different pulse echoes exhibit minute random shifts in time at the sub-sampling interval level. To eliminate the adverse effects of this shift on subsequent averaging processing, high-precision time alignment is required.

[0065] Specifically, from the effective pulses, the pulse with the highest signal-to-noise ratio is selected as the reference pulse. The cross-correlation function between the remaining effective pulses and the reference pulse is determined. Based on the peak position of the cross-correlation function, the time offset of each pulse relative to the reference pulse is determined. According to the time offset, a linear phase is applied to each pulse in the frequency domain using the phase shift characteristic of the Fourier transform. Then, an inverse Fourier transform is performed to correct the frequency domain phase shift of each valid pulse, achieving sub-sampling-level time alignment. This process is executed one by one for each of the selected valid pulses. The imaginary unit, It is a frequency vector. This is the time offset of the i-th valid pulse.

[0066] Preferably, to improve estimation accuracy, sub-pixel interpolation methods such as parabolic fitting can be used to process data points near the peak of the cross-correlation function, thereby obtaining the time offset with sub-sampling interval accuracy.

[0067] Based on all time-aligned valid pulses, determine the instantaneous power sequence of each pulse. In the formula For the first The instantaneous power of a pulse signal, I[ x ] is the first x The in-phase component of a pulse signal, Q[ x ] is the first x The orthogonal components with a 90° phase difference of each pulse signal are used to incoherently accumulate and average all pulse power sequences to obtain the average pulse power envelope. In the formula, M represents the incoherent accumulation number. This power averaging operation is called incoherent accumulation. Compared to coherent accumulation, which directly adds I / Q data, incoherent accumulation is less sensitive to local oscillator phase noise and small phase changes caused by target micro-motions, exhibiting better robustness. By averaging M statistically independent pulses, the signal-to-noise ratio can be significantly improved, yielding an extremely smooth power envelope curve that clearly characterizes the pulse response shape of the radar system, such as... Figure 3 As shown, this lays the foundation for subsequent high-precision waveform parameter extraction.

[0068] Step S3: Transform the average pulse power envelope to the logarithmic domain, and select discrete data points located within a pre-defined main lobe region; based on the selected discrete data points, iteratively fit the generalized super-Gaussian mathematical model using a nonlinear least squares method with boundary constraints to obtain a continuous analytic function describing the pulse envelope, such as... Figure 4 As shown.

[0069] Transforming the average pulse power envelope to the logarithmic domain allows for a clearer representation of the dynamic range differences between the pulse's main lobe characteristics, side lobe structure, and noise floor, which is beneficial for subsequent accurate identification and analysis of key pulse waveform features.

[0070] As an example, the average pulse power envelope is transformed to the logarithmic domain as follows: ,in, for Average power at time t, for The average power after taking the logarithm at time step 1.

[0071] Specifically, the discrete data points within the predefined main lobe region are defined as all discrete data points whose power values ​​decrease along the falling edges on both sides of the peak point of the average pulse power envelope to a preset power level. As an example, the preset power level is a value within the range of 20 dB to 30 dB. This operation aims to focus on the critical region determining distance resolution—the pulse main lobe—while effectively eliminating noisy baseline data points far from the main lobe to avoid interfering with fitting accuracy and improving computational efficiency.

[0072] This invention employs a generalized super-Gaussian mathematical model to fit the pulse envelope. By introducing the shape order n as a degree of freedom, the generalized super-Gaussian mathematical model can more flexibly and accurately describe various morphological characteristics of the output pulse of actual radar systems, such as top flattening and edge steepening, compared to the standard Gaussian model with a fixed n=2.

[0073] Specifically, the generalized superGaussian mathematical model is expressed as: ;

[0074] In the formula:

[0075] : Pulse power envelope function. Represented on the time axis The instantaneous power value of the radar echo after incoherent accumulation is usually expressed in watts (W) or milliwatts (mW).

[0076] Peak power. Refers to the theoretical maximum power value of the main lobe of the pulse envelope obtained from the fitting.

[0077] Time variable. Refers to discrete time points after sampling or interpolation.

[0078] The pulse center moment. This refers to the point at which the pulse envelope reaches its peak power. The corresponding time position.

[0079] : Pulse width scaling parameter. This parameter characterizes the characteristic width of the pulse (corresponding to energy decay to...). (location), its specific physical meaning is determined by the shape order. A joint decision.

[0080] Shape order. A dimensionless positive real number that controls the flatness of the top and the steepness of the edges of the pulse waveform. When it follows a standard Gaussian distribution; when It exhibits a flat-top (super-Gaussian) characteristic.

[0081] : Basis noise power. Refers to the average power of thermal noise and background interference in the receiver system, used as the DC bias term in the fitting model.

[0082] Specifically, based on the selected discrete data points, a nonlinear least squares algorithm is used to iteratively fit the generalized super-Gaussian mathematical model. Through iterative optimization, the optimal combination of model parameters that makes the objective function optimal is obtained. , , , and Once the optimal parameters are obtained, a continuous analytical function that can accurately describe the entire pulse envelope is obtained, successfully transforming the data points limited by discrete sampling into a continuous mathematical curve. This completely breaks through the "pick-up fence effect" limitation caused by the hardware sampling interval, laying a solid foundation for subsequent realization of sub-sampling level precision pulse width measurement.

[0083] Furthermore, the nonlinear least squares algorithm can be the Levenberg-Marquardt algorithm.

[0084] Step S4: Determine the effective pulse duration when the power drops to -6dB from the peak value based on the continuous analytical function; determine the true range resolution of the weather radar based on the effective pulse duration.

[0085] This invention abandons the traditional practice of using a -3dB standard to measure pulse width at the receiver, and instead adopts a -6dB criterion based on the physical mechanism of radar two-way detection as the criterion for determining the effective pulse width. The physical basis is as follows: the essence of radar range resolution is the ability to distinguish between two adjacent point targets. Radar detection is a two-way process of "transmission-reflection-reception." The voltage amplitude half-power point of the transmitted pulse, i.e., the -3dB point, after experiencing the two-way path of "transmission transmission → target reflection → reception transmission," has its voltage attenuation factor squared, causing the power at the corresponding point at the receiver output to drop to (1 / 2)² = 1 / 4 of the peak power, i.e., -6dB. Therefore, using -6dB as the criterion for determining the effective pulse width at the receiver strictly conforms to the physical essence of radar two-way detection and can accurately reflect the radar's actual ability to distinguish between two point targets of equal intensity.

[0086] Specifically, based on the continuous analytical function, the effective pulse duration at which the power drops to -6dB from the peak value is determined, including:

[0087] Let the normalized power term That is, -6dB, the power drops to 1 / 4 of the peak value.

[0088] Solving the equation, we get: ,Right now ;

[0089] Half width ;

[0090] Effective pulse duration, i.e., full width In the formula, Effective pulse width. Specifically refers to the -6dB criterion proposed according to this invention, i.e., power drops to one-quarter of the peak power ( The full width of the pulse corresponding to the location. The natural logarithm constant, with a value of approximately 1.386, is derived from the linear power ratio corresponding to -6 dB. w: Pulse width scaling parameter.

[0091] Specifically, the true range resolution of the weather radar is determined based on the effective pulse width, including: Substitute into the radar ranging formula Determine the true range resolution of the weather radar. .in, It is the speed of light.

[0092] Through this step, the present invention finally achieves high-precision range resolution calibration based on strict physical criteria and precise mathematical calculations, enabling sub-meter level calibration and providing a reliable quantitative basis for the performance evaluation and maintenance of weather radar.

[0093] Step S5: Based on the morphological characteristics of the continuous analytical function, perform a diagnostic analysis of the performance status of the radar system.

[0094] Based on the morphological characteristics of continuous analytic functions, a diagnostic analysis of the performance status of the radar system is performed, including at least one of the following:

[0095] The main lobe broadening factor is determined by the ratio of the actual distance resolution to the theoretical distance resolution. Based on the main lobe broadening factor and the first threshold, it can be indicated that there may be performance degradation in the receiving channel, specifically the receiver intermediate frequency filter bandwidth being too narrow, filter element aging, and / or mismatch in the design of the matched filter coefficients.

[0096] Based on the -6dB points on both sides of the peak value of the continuous analytical function, the waveform symmetry factor is determined. Based on the degree to which the waveform symmetry factor deviates from the set value, the I / Q channel balance, quadrature demodulator fault, and / or filter group delay characteristics can be judged.

[0097] The difference between the peak value of the main lobe and the peak value of the first side lobe is determined by the continuous analytical function. Based on the difference and the second threshold, it can be determined whether the compression coefficient of the pulse compression system is mismatched and / or whether the transmitter is operating in the nonlinear region, resulting in spectrum regeneration.

[0098] The fitting residual is calculated based on the difference between the continuous analytical function and the discrete data points. This residual reflects the degree of agreement between the continuous analytical function and the original discrete data points. Based on the fitting residual and a third threshold, it can be determined whether multipath interference and / or hardware faults exist. If the fitting residual exceeds the preset third threshold, it indicates that the pulse shape is severely distorted and cannot be well represented by the generalized super-Gaussian model.

[0099] As an example, the first threshold is 1.2 and the second threshold is 20dB.

[0100] Through the system performance diagnosis and analysis in step S5, this invention can not only provide high-precision range resolution calibration results, but also further reveal potential system faults or performance degradation links that cause resolution changes. It provides a powerful technical means for radar system status monitoring, preventive maintenance and fault location, and realizes the functional expansion from single parameter measurement to comprehensive performance diagnosis.

[0101] The range resolution calibration method provided by this invention completely overcomes the picket fence effect and quantization error in traditional methods by combining frequency domain interpolation oversampling with generalized super-Gaussian model fitting, improving the range resolution measurement accuracy to sub-meter level and achieving true quantitative measurement. Following the -6dB criterion established by the physical mechanism of radar two-way detection, it corrects the long-standing standard confusion problem, ensuring the physical authenticity of the calibration results. With the help of multi-pulse alignment and incoherent accumulation techniques, it significantly improves anti-interference capability and measurement robustness. Simultaneously, through in-depth analysis of pulse waveform characteristics, this invention endows the method with fault diagnosis capabilities for the radar system's transmission, reception, and processing links, upgrading the calibration process from single-parameter measurement to comprehensive performance evaluation, greatly enhancing the intelligence level and application value of weather radar operation and maintenance support.

[0102] The present invention also provides a system for implementing the above-described weather radar range resolution calibration method, comprising:

[0103] The acquisition module is used to acquire oversampled raw baseband I / Q data of a point target over multiple consecutive pulse cycles.

[0104] The signal enhancement module is used to determine the signal-to-noise ratio of each pulse based on the oversampled original baseband I / Q data; to filter out effective pulses based on the signal-to-noise ratio and a preset threshold; to perform subsampling-level time alignment on the effective pulses based on cross-correlation analysis; and to perform incoherent cumulative averaging on the power magnitude of all time-aligned pulses to generate an average pulse power envelope.

[0105] The waveform reconstruction module is used to convert the average pulse power envelope to the logarithmic domain and select discrete data points located within a pre-defined main lobe region. Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method to obtain a continuous analytic function describing the pulse envelope.

[0106] The range resolution determination module is used to determine the effective pulse duration when the power drops to -6dB of the peak value based on a continuous analytical function; and to determine the true range resolution of the weather radar based on the effective pulse duration.

[0107] The various modules refer to the above-mentioned weather radar range resolution calibration method, which will not be repeated here.

[0108] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0109] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0110] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first and second features are in direct contact, or that they are in indirect contact through an intermediate medium. Furthermore, "above," "over," or "on top" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," or "beneath" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0111] In the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0112] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A weather radar range resolution calibration method, characterized by, Achieving this by aligning the radar antenna with a far-field point target includes the following steps: Oversampled raw baseband I / Q data of the target point over multiple consecutive pulse cycles; Based on the oversampled raw baseband I / Q data, the signal-to-noise ratio (SNR) of each pulse is determined; effective pulses are selected based on the SNR and a preset threshold; subsampling-level time alignment is performed on the effective pulses based on cross-correlation analysis; and the power magnitude of all time-aligned pulses is noncoherently accumulated and averaged to generate the average pulse power envelope. The average pulse power envelope is transformed to the logarithmic domain, and discrete data points located within a predefined main lobe region are selected. Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using a nonlinear least squares method with boundary constraints to obtain a continuous analytic function describing the pulse envelope. The generalized super-Gaussian mathematical model is expressed as: ; The boundary constraints are that during the iteration process, the parameter w must be a positive number and n must be within a preset range; In the formula: : Pulse power envelope function, representing the time axis Above is the instantaneous power value of the radar echo after incoherent accumulation; Peak power refers to the theoretical maximum power value of the main lobe of the pulse envelope obtained by fitting. Time variable; The pulse center moment refers to the moment when the pulse envelope reaches its peak power. The corresponding time position; : Pulse width scaling parameter. This parameter characterizes the characteristic width of the pulse, and its specific physical meaning is determined by the shape order. Joint decision; Shape order, a dimensionless positive real number, controls the flatness of the top and the steepness of the edges of the pulse waveform; : Basis noise power, refers to the average power of thermal noise and background interference in the receiver system, and is used as the DC bias term of the fitting model; Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method to solve for the parameters. , , , and To obtain continuous analytic functions; Based on the continuous analytical function, the effective pulse duration when the power drops to -6dB from the peak value is determined; based on the effective pulse duration, the true range resolution of the weather radar is determined.

2. The weather radar range resolution calibration method of claim 1, wherein, Oversampled raw baseband I / Q data is obtained in the following way: The radar receiver directly oversamples at a sampling interval smaller than the theoretical range resolution; and / or, the radar receiver samples to obtain the original baseband I / Q data, performs frequency domain interpolation oversampling on the original baseband I / Q data, and obtains oversampled original baseband I / Q data.

3. The weather radar range resolution calibration method of claim 2, wherein, Frequency domain interpolation and oversampling processing is performed on the raw baseband I / Q data, including: A Fast Fourier Transform (FFT) is performed on the original baseband I / Q data of length N to obtain its frequency domain spectrum. Based on the target oversampling factor K, an extended spectrum sequence of length N×K is constructed, in which the positive frequency part of the frequency domain spectrum is placed at the low-frequency end of the extended spectrum sequence, and the negative frequency part of the frequency domain spectrum is placed at the high-frequency end of the extended spectrum sequence. Zero values ​​are filled between the low-frequency end and the high-frequency end to maintain the conjugate symmetry of the Nyquist frequency components. An Inverse Fast Fourier Transform (IFFT) is performed on the extended spectrum sequence, and the transform result is multiplied by the gain factor K to obtain the original baseband I / Q data with K times oversampling.

4. The weather radar range resolution calibration method of claim 1, wherein, Subsampling-level time alignment of effective pulses based on cross-correlation analysis includes: The effective pulse with the highest signal-to-noise ratio is selected as the reference pulse; the cross-correlation function between the remaining effective pulses and the reference pulse is determined, and the time offset of each pulse relative to the reference pulse is determined based on the peak position of the cross-correlation function; according to the time offset, frequency domain phase shift correction is performed on each effective pulse to achieve sub-sampling level time alignment.

5. The weather radar range resolution calibration method of claim 1, wherein, The discrete data points within the pre-defined main lobe region are: all discrete data points whose power values ​​decrease along the falling edges on both sides of the center, centered on the peak point of the average pulse power envelope, down to the preset power level.

6. The weather radar range resolution calibration method of claim 1, wherein, The effective pulse duration is determined based on the continuous analytical function, including: determining the time parameter that satisfies the power drop to 1 / 4 of the peak power, and determining the effective pulse duration based on the time parameter.

7. The weather radar range resolution calibration method of claim 1, wherein, Also includes: Based on the morphological characteristics of continuous analytic functions, the performance status of radar systems is diagnosed and analyzed.

8. The weather radar range resolution calibration method of claim 7, wherein, Based on the morphological characteristics of continuous analytic functions, a diagnostic analysis of the performance status of the radar system is performed, including at least one of the following: The main lobe broadening factor is determined based on the ratio of the actual distance resolution to the theoretical distance resolution. Based on the main lobe broadening factor and the first threshold, it is determined whether the intermediate frequency filter bandwidth of the receiver is too narrow and / or whether the matched filter coefficients are mismatched. Based on the -6dB points on both sides of the peak of the continuous analytical function, the waveform symmetry factor is determined. Based on the degree to which the waveform symmetry factor deviates from the set value, the I / Q channel balance and / or filter group delay characteristics are judged. The difference between the peak value of the main lobe and the peak value of the first side lobe is determined based on the continuous analytical function. Based on the difference and the second threshold, it is determined whether the compression coefficient of the pulse compression system is mismatched and / or whether the transmitter is operating in the nonlinear region, resulting in spectrum regeneration. The fitting residual is calculated based on the difference between the continuous analytical function and the discrete data points. Based on the fitting residual and the third threshold, it is determined whether there is multipath interference and / or hardware fault.

9. A weather radar range resolution calibration system, characterized by, include: The acquisition module is used to acquire oversampled raw baseband I / Q data of a point target over multiple consecutive pulse cycles; The signal enhancement module is used to determine the signal-to-noise ratio of each pulse based on the oversampled raw baseband I / Q data; Valid pulses are selected based on the signal-to-noise ratio and a preset threshold. Based on cross-correlation analysis, subsampling-level time alignment is performed on the effective pulses, and the power magnitudes of all time-aligned pulses are incoherently accumulated and averaged to generate the average pulse power envelope. The waveform reconstruction module is used to convert the average pulse power envelope to the logarithmic domain and select discrete data points located within a pre-defined main lobe region. Based on the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted using the nonlinear least squares method with boundary constraints to obtain a continuous analytic function describing the pulse envelope. The generalized super-Gaussian mathematical model is expressed as: ; The bounding constraint is that during the iteration process, the parameter w is limited to be positive and n is within a preset range; In the formula: : Pulse power envelope function, representing the time axis Above is the instantaneous power value of the radar echo after incoherent accumulation; Peak power refers to the theoretical maximum power value of the main lobe of the pulse envelope obtained by fitting. Time variable; The pulse center moment refers to the moment when the pulse envelope reaches its peak power. The corresponding time position; : Pulse width scaling parameter. This parameter characterizes the characteristic width of the pulse, and its specific physical meaning is determined by the shape order. Joint decision; Shape order, a dimensionless positive real number, controls the flatness of the top and the steepness of the edges of the pulse waveform; : Basis noise power, refers to the average power of thermal noise and background interference in the receiver system, and is used as the DC bias term of the fitting model; According to the selected discrete data points, the generalized super-Gaussian mathematical model is iteratively fitted by using a nonlinear least square method to solve the parameters , , , and , and a continuous analytical function is obtained. The range resolution determination module is used to determine the effective pulse duration when the power drops to -6dB from the peak value based on a continuous analytical function; and to determine the true range resolution of the weather radar based on the effective pulse duration.

Images