Rain detection radar clutter suppression and high-precision echo recovery method and related device

Abstract

The invention discloses a rain detection radar clutter suppression and high-precision echo recovery method and a related device, and belongs to the technical field of weather radar signal processing. According to the method, a non-uniform Doppler filter bank design is adopted, a wide and deep null trap is formed near zero frequency to suppress ground clutters, and meanwhile, a clutter channel of a DFT filter bank is reserved to be used for detecting a low-speed / zero-speed target; self-adaptive cancellation processing is carried out in combination with the dynamically constructed azimuth-distance-velocity three-dimensional clutter map; and finally, recovering a rainfall echo frequency spectrum submerged by clutters through a Gaussian fitting algorithm to realize high-precision estimation of rainfall parameters. According to the method, the problems of insufficient weak rainfall echo detection capability, low-speed / zero-speed target information loss and large spectral moment estimation error of a traditional method under a strong clutter background are effectively solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of weather radar signal processing, and relates to a method for suppressing clutter and recovering high-precision echoes of a rain measuring radar and a related device. Background Art

[0002] As one of the core monitoring tools of the rainfall monitoring system, weather radar systems play an irreplaceable and critical role in real-time rainfall monitoring and flood control in river basins. To enhance flood prevention and disaster reduction capabilities, there is an urgent need to develop a new generation of weather radar systems with higher detection accuracy, effectively extending flood forecast periods and significantly improving flood prevention decision-making capabilities.

[0003] The core detection area of ​​weather radar for rainfall monitoring is typically concentrated in the atmospheric boundary layer, from cloud base to the ground surface. To effectively observe the motion characteristics of rain particles near the ground, the system must use a low-elevation detection mode. However, this observation method causes ground clutter and rain particle echo signals to overlap, seriously affecting the accuracy of rainfall parameter estimation. Therefore, how to effectively suppress clutter and accurately detect and recover rain echo signals in the presence of strong ground clutter interference has become a key technical challenge for improving the accuracy of rainfall parameter estimation.

[0004] Existing technologies often employ IIR elliptic filters combined with PPP / DFT methods and Gaussian model-based adaptive filtering (GMAP) methods. The IIR elliptic filter combined with PPP / DFT methods employs an infinite impulse response (IIR) elliptic filter for ground clutter suppression, followed by pulse pair processing (PPP) or discrete Fourier transform (DFT) for meteorological parameter estimation. The IIR elliptic filter acts as a high-pass filter, removing ground clutter by setting a low cutoff frequency and a deep suppression notch. The PPP algorithm performs time-domain correlation analysis to calculate Doppler velocity and spectral width, while the DFT performs spectral analysis to extract the frequency domain characteristics of meteorological targets. The Gaussian model-based adaptive filtering (GMAP) method, based on the spectral characteristics of weather echoes and ground clutter, assumes that both conform to a Gaussian distribution. Because the power spectrum of ground clutter is concentrated near zero frequency and has a narrow spectral width, this method dynamically identifies spectral boundaries, suppressing clutter while simultaneously performing spectral fitting on the weather echo near zero frequency to compensate for the filtered-out valid signal, thereby improving the accuracy of spectral moment estimation.

[0005] However, the traditional coherent integration method based on windowed DFT suffers from spectral leakage in strong clutter backgrounds, drowning out weak rain echo signals and affecting the radar system's detection performance. Furthermore, in existing clutter suppression algorithms, IIR elliptic filters, due to their inherent frequency response characteristics, inevitably suppress valid rain echoes near zero frequency. While conventional adaptive clutter suppression filter banks can adaptively form nulls based on clutter characteristics, their high sidelobe levels can cause target signal spectrum leakage, thus affecting the accuracy of rain parameter estimation. Furthermore, existing GMAP-based signal processing methods have two major drawbacks: First, they fail to form effective nulls at zero frequency, resulting in insufficient clutter suppression performance. Second, when the rain echo spectrum peak is near zero frequency, the algorithm lacks the ability to distinguish between clutter and slow / zero-speed targets, often misclassifying valid rain echoes as clutter and filtering them out, thus affecting the accuracy of subsequent spectral moment estimation. This leads to inaccurate rain echo measurements, which can impact weather forecasting, disaster warning, public safety, and socioeconomic activities. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and related device for clutter suppression and high-precision echo recovery of a rain measuring radar, so as to solve the technical problem in the prior art that in the low-elevation-angle detection mode of weather radar, ground clutter and rain echo overlap, resulting in inaccurate rain echo measurement.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for suppressing clutter and recovering high-precision echoes of a rain measuring radar, comprising the following steps: Based on the weather radar system parameters and the intensity and spectral width of the actual scene clutter, an MTD filter bank based on a digital synthesis algorithm is constructed; the channels of the MTD filter bank all have nulling characteristics, forming a nulling channel group; Based on the MTD filter bank, the DFT filter bank is used to replace the channel where the clutter is located to obtain a non-uniform filter bank; The echo signal received by the weather radar system is preprocessed as the input signal of the non-uniform filter bank, and the matrix multiplication operation is performed on the filter coefficients to generate a two-dimensional range-Doppler spectrum matrix; Based on the two-dimensional range-Doppler spectrum matrix, a three-dimensional dynamic clutter map of azimuth, range, and velocity is iteratively constructed for the clutter channel during each azimuth beam dwell period. The clutter map cancellation method is then used to process each subsequent frame of CPI data to obtain slow / stationary targets. The moving target is extracted through the null channel, and the data of the moving target is fused with the slow / stationary target to reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal.

[0008] Furthermore, the sidelobe area of ​​the MTD filter can obtain sidelobe levels of different shapes by changing the size of the interference power placed, which specifically includes the following steps: Assuming the ground clutter power spectrum to be a Gaussian spectrum with the center of the spectrum being zero, the ground clutter covariance matrix is for:

[0009] Where, , , , is the ground clutter power spectrum variance; is the ground clutter power, is the filter order, is the weather radar pulse repetition period; With Interference signals are input to the filter, and their frequencies are And evenly cover the entire pulse repetition frequency range, then the covariance matrix of clutter, interference and noise is for:

[0010] Where, is the noise power, The frequency is The power of interference, is the pilot vector of frequency; but MTD filter bank The expression of the weight vector of a filter is:

[0011] Where, For the The weight vector of the filter.

[0012] Furthermore, the step of preprocessing the echo signal received by the weather radar system as the input signal of the non-uniform filter bank, and performing matrix multiplication operation in combination with the filter coefficients to generate a two-dimensional range Doppler spectrum matrix specifically includes: The echo signals received by the weather radar system are subjected to digital down-conversion and pulse compression processing in sequence to obtain a coherent pulse sequence; Multiply the filter coefficients with the coherent pulse sequence to obtain a two-dimensional range-Doppler spectrum matrix; Assume that the coherent pulse sequence of the input signal is:

[0013] Where, , , represents transpose; represents the zeroth moment, Indicates the time; Input Data , For signal; for clutter; is the noise vector; When performing MTD filtering, the filter output is:

[0014] Where, is the weight coefficient of the non-uniform Doppler filter bank.

[0015] Furthermore, the step of iteratively constructing an azimuth-range-velocity three-dimensional dynamic clutter map for the clutter channel within each azimuth beam dwell period based on the two-dimensional range Doppler spectrum matrix specifically includes: In each azimuth beam dwell period, the data of each range unit output by the clutter channel is stored in the storage unit. As the beam scans, the signal of each storage unit is recursively updated. The specific expression is:

[0016] Where, is a factor less than 1.

[0017] Furthermore, the step of using the clutter map cancellation method to process each subsequent frame of CPI data to obtain a slow / stationary target specifically includes: After the clutter map data is relatively stable, the clutter map threshold is set and the current signal amplitude is compared with the clutter map threshold. If the signal amplitude exceeds the clutter map threshold, it is determined to be a valid echo and the original data is retained. Otherwise, the signal is considered to be a clutter component and is replaced with average noise.

[0018] Furthermore, the calculation formula of the clutter map threshold is:

[0019] Where, is the threshold factor.

[0020] Furthermore, in the step of fusing the moving target and the slow / stationary target data to reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal, a Gaussian fitting algorithm is used for data fusion.

[0021] In a second aspect, the present invention provides a rain radar clutter suppression and high-precision echo recovery system, comprising: An MTD filter bank construction module is used to construct an MTD filter bank based on a digital synthesis algorithm based on weather radar system parameters and the intensity and spectral width of actual scene clutter; the channels of the MTD filter bank all have nulling characteristics, forming a nulling channel group; A non-uniform filter bank construction module is used to replace the MTD filter bank with a DFT filter bank for the channel where the clutter is located to obtain a non-uniform filter bank; The two-dimensional range-Doppler spectrum matrix generation module is used to pre-process the echo signal received by the weather radar system as the input signal of the non-uniform filter bank, and perform matrix multiplication operation with the filter coefficients to generate a two-dimensional range-Doppler spectrum matrix; The slow / stationary target detection module is used to iteratively construct a three-dimensional dynamic clutter map of azimuth, range, and velocity for the clutter channel within each azimuth beam dwell period based on the two-dimensional range-Doppler spectrum matrix, and then use the clutter map cancellation method to process each subsequent frame of CPI data to obtain slow / stationary targets; The moving target detection and fusion module is used to extract the moving target through the null channel, fuse the moving target with the slow / stationary target data, and reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal.

[0022] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for suppressing clutter and recovering high-precision echoes of a rain measuring radar are implemented.

[0023] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a method for suppressing clutter and recovering high-precision echoes of a rain measuring radar.

[0024] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a method and related device for clutter suppression and high-precision echo recovery of rain radar. The method adopts a non-uniform Doppler filter bank design to achieve efficient suppression of ground clutter while significantly improving the signal-to-noise ratio of rain echoes and reducing the target sidelobe effect. This design not only enhances the detection capability of weak rain echoes, but also fully preserves the signal characteristics of slow and stationary targets by adding a clutter channel. Prior information of ground clutter is introduced to pre-optimize the filter bank design, and the coefficient preloading method is used to significantly improve the real-time processing performance of the algorithm, solving the problem of high computational complexity of traditional methods. Based on the dynamic clutter map construction technology, the present invention realizes the real-time update and precise cancellation of clutter characteristics, effectively solving the problem of distinguishing slow-moving targets from ground clutter. In addition, the present invention adopts the strategy of replacing clutter points with system average noise. The algorithm is simple and significantly improves the reliability of the rain echo spectrum after Gaussian fitting, thereby greatly improving the accuracy of rainfall parameter inversion. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 is a flow chart of the method of the present invention; Figure 2 This is a graph showing the amplitude-frequency response characteristics of a 43rd-order MTD filter bank based on a digital synthesis algorithm according to an embodiment of the present invention; Figure 3 This is a graph showing the amplitude-frequency response characteristics of a non-uniform filter bank according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a primary recursive filtering principle according to an embodiment of the present invention; Figure 5 is a flow chart for designing a non-uniform filter bank in an embodiment of the present invention; Figure 6 Flowchart of a method for detecting and recovering rainfall echoes according to an embodiment of the present invention; Figure 7 A result diagram of constructing a zero-channel clutter map when a weather radar resides at a certain azimuth in an embodiment of the present invention; Figure 8 This is a comparison diagram of the amplitude-frequency characteristics of a certain distance unit after passing through a non-uniform filter bank and a DFT filter bank under a rainfall background in an embodiment of the present invention; Figure 9 1 is a comparison diagram of the amplitude-frequency characteristics before and after processing using the Gaussian fitting method in an embodiment of the present invention; Figure 10This is a comparison diagram of the output spectrum of the non-uniform filter and the spectrum after noise map cancellation and Gaussian fitting restoration in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0028] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise indicated, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0029] See also Figure 1 The embodiment of the present invention discloses a method for suppressing clutter and recovering high-precision echoes of a rain measuring radar, comprising the following steps: S1, based on the weather radar system parameters and the intensity and spectral width of the actual scene clutter, constructs a moving target detection (MTD) filter bank based on a digital synthesis algorithm; the frequency response of this MTD filter bank not only has a wide and deep null near zero frequency, but also the sidelobe area can be designed to any shape according to the actual needs.

[0030] By placing interference signals of different frequencies in the sidelobe region of its frequency response, the power intensity of the interference signals is changed, thereby adaptively controlling the sidelobe level of the filter's frequency response. Compared to traditional point-optimal or equally spaced Doppler filter banks that use conventional window functions and cannot meet the requirements of filters with passbands far from ground clutter, this method can design filters with lower sidelobes in the low-frequency range, thereby better detecting moving targets. The specific implementation process is as follows: In practical applications, the ground clutter power spectrum is usually assumed to be a Gaussian spectrum with the center of the spectrum being zero, so the ground clutter covariance matrix is It can be written as:

[0031] Where, , , , is the ground clutter power spectrum variance; is the ground clutter power, is the filter order, is the weather radar pulse repetition period; With Interference signals are input to the filter, and their frequencies are And evenly cover the entire pulse repetition frequency range, then the covariance matrix of clutter, interference and noise is for:

[0032] Where, is the noise power, The frequency is The power of interference, is the pilot vector of frequency.

[0033] but MTD filter bank The expression of the weight vector of a filter is:

[0034] Where, For the The weight vector of the filter.

[0035] By changing the power of the interference placed within the filter sidelobe frequency range, sidelobe levels of different shapes can be obtained.

[0036] When designing a filter bank, the ideal main-to-sidelobe ratio (MSR) is usually determined first. Based on this ratio, the required interference power is determined by calculating the difference between the ideal MSR and the actual MSR. Figure 2 The amplitude-frequency response characteristics of the 43rd-order MTD filter bank designed using a digital synthesis algorithm are shown when the ideal main-sidelobe ratio is 40dB (the zero channel and its adjacent channels have been removed): It can be seen that the filter group not only has a low sidelobe level, effectively avoiding spectrum leakage, but also forms a deeper null at zero frequency, which can effectively suppress ground clutter interference and improve the moving target detection capability.

[0037] S2: Based on the MTD filter bank designed in step S1, a discrete Fourier transform (DFT) filter bank is used to replace the channels where the clutter is located (the zero-Doppler channel and its adjacent channels). This results in a new Doppler filter bank with a non-uniform amplitude-frequency response (non-uniform filter bank). The specific number of replacement channels is dynamically determined based on the actual clutter intensity and spectral width.

[0038] In this step, the constructed non-uniform filter bank is mainly composed of the null filter bank designed by S1 and the clutter channel (zero Doppler channel and its adjacent channels) of the DFT filter bank. Figure 3As shown in Figure 1, the filter bank's amplitude-frequency response exhibits non-uniform characteristics, characterized by uneven distribution of mainlobe peaks across the entire spectrum (reflected primarily by the difference in spacing between the clutter channel and the null channel). This design effectively detects moving targets while fully preserving the amplitude information of slow and zero-speed target signals.

[0039] S3, pre-processes the echo signal received by the pulse weather radar system as the input signal of the non-uniform filter bank, and then performs matrix multiplication operation on the data of the same range unit in each coherent processing period (CPI) and the filter coefficient to generate a two-dimensional range Doppler spectrum matrix; The echo signals received by the weather radar system are subjected to digital down-conversion and pulse compression processing in sequence to obtain a coherent pulse sequence; Multiply the filter coefficients with the coherent pulse sequence to obtain a two-dimensional range-Doppler spectrum matrix; Assume that the coherent pulse sequence of the input signal is:

[0040] Where, , , Indicates transpose. represents the zeroth moment, Indicates the time; Input Data , For signal; for clutter; is the noise vector; When performing MTD filtering, the filter output is:

[0041] Where, is the weight coefficient of the non-uniform Doppler filter bank.

[0042] S4, based on the two-dimensional range-Doppler spectrum matrix, iteratively constructs a three-dimensional dynamic clutter map of azimuth-range-velocity for the clutter channel during each azimuth beam dwell period (one azimuth dwell period contains one frame of CPI data). It then uses the clutter map cancellation method to process each subsequent frame of CPI data to obtain slow / stationary targets. In this step, a dynamic clutter map is constructed iteratively based on the clutter channel output during each azimuth beam dwell period. The specific construction method is as follows: In each azimuth beam dwell period, the data of each distance unit output by the clutter channel is stored in the storage unit. As the beam scans, the signal of each storage unit is recursively updated. The updating principle diagram is as follows: Figure 4 As shown, It represents the delay of the beam scanning cycle and represents the clutter map memory.

[0043] The corresponding expression is:

[0044] Where, is a factor less than 1.

[0045] After repeated scanning of the beam, the clutter mean of the azimuth range unit is stored in the clutter map, and the detection threshold is according to ( is the threshold factor, which determines the detection probability and false alarm probability) calculation. If the detected signal Greater than threshold , it is judged as a target and the target amplitude is retained, otherwise it is replaced by the system average noise power.

[0046] S5, extracts the moving target through the null channel, fuses the moving target with the slow / stationary target data, and reconstructs the complete amplitude-frequency characteristics of the rainfall echo signal based on the Gaussian fitting algorithm, ultimately achieving the estimation of key parameters such as rainfall intensity and spectrum width.

[0047] In this step, since a portion of the rain echo may still be submerged in the clutter channel and replaced by noise power by the above-mentioned clutter pattern detection method, it is necessary to restore the rain echo in the spectrum.

[0048] Since rainfall echoes have wide spectrum width and relatively random Doppler frequency, their power spectrum can be approximately expressed by a Gaussian probability density function as follows:

[0049] Where, is the echo power, is the average Doppler frequency, is the Doppler spectrum width, is the noise power.

[0050] Gaussian fitting is performed on the clutter map and the spectrum output by the null filter bank. After the rain echo is restored, parameters such as rainfall intensity and radial velocity are inverted through spectral moments. This process can effectively restore rain echoes obscured by clutter, thereby improving the accuracy of rainfall parameter estimation.

[0051] An embodiment of the present invention also discloses a rain measuring radar clutter suppression and high-precision echo recovery system, which includes an MTD filter bank construction module, a non-uniform filter bank construction module, a two-dimensional range Doppler spectrum matrix generation module, a slow / stationary target detection module, and a moving target detection and fusion module.

[0052] Among them, the MTD filter group construction module is used to construct an MTD filter group based on a digital synthesis algorithm based on the weather radar system parameters and the intensity and spectral width of the actual scene clutter; the channels of the MTD filter group all have null-pitching characteristics, forming a null-pitching channel group; the non-uniform filter group construction module is used to replace the channel where the clutter is located with a DFT filter group based on the MTD filter group to obtain a non-uniform filter group; the two-dimensional range Doppler spectrum matrix generation module is used to pre-process the echo signal received by the weather radar system as the input signal of the non-uniform filter group signal, and performs matrix multiplication operations in combination with the filter coefficients to generate a two-dimensional range-Doppler spectrum matrix; the slow / stationary target detection module is used to iteratively construct an azimuth-range-speed three-dimensional dynamic clutter map for the clutter channel within each azimuth beam dwell period based on the two-dimensional range-Doppler spectrum matrix, and use the clutter map cancellation method to process each subsequent CPI frame data to obtain the slow / stationary target; the moving target detection and fusion module is used to extract the moving target through the null channel, fuse the moving target with the slow / stationary target data, and reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal.

[0053] This invention addresses the problem of overlapping ground clutter and rain echoes in the low-elevation-angle detection mode of weather radars. It innovatively applies a non-uniform Doppler filter bank to rain echo detection. This filter bank improves the detection capability of low signal-to-clutter ratio (SCR) rain echoes by generating nulls near zero frequency for rain particles with a certain velocity. For zero-speed and slow rain echoes, the low-frequency signal components are fully preserved. By constructing a three-dimensional dynamic clutter map of azimuth, range, and velocity, a clutter map cancellation method is employed to effectively separate rain echoes from clutter interference. Subsequently, Gaussian fitting is performed on the clutter map and rain echoes detected by the filter bank. Compared to the traditional GMAP method, which searches for clutter boundaries and replaces clutter points, this method not only significantly improves the accuracy of rain echo parameter estimation but also enhances the timeliness of weather radar systems. This provides a technical approach that combines theoretical innovation with practical value for the implementation of weather radar projects.

[0054] Example: This embodiment uses measured data collected by a weather radar as the verification basis. The data used is divided into two groups: the first group is the original echo data obtained by multi-cycle scanning of the radar servo system under clear weather conditions, which is used to establish background clutter characteristics; the second group is the original echo data obtained by single-cycle scanning of the radar under rainy weather conditions.

[0055] Step S1: Design a non-uniform filter bank based on the spectral distribution characteristics of ground clutter in the measured data. Figure 5 This is a flow chart for designing a non-uniform filter bank in an example of the present invention, and the specific steps are as follows: First, the collected raw intermediate frequency echo data undergoes frequency mixing, low-pass filtering, pulse compression, and MTD processing in sequence. The null depth and notch width of the filter bank are adaptively set by analyzing the amplitude and variance characteristics of the ground clutter power spectrum. Second, a digital synthesis algorithm is used to optimize the filter bank's performance. In this embodiment, 40 dB is selected as the ideal main-to-sidelobe ratio. The interference power is calculated by taking the difference between the filter's current sidelobe level and the ideal level. After uniformly adding interference across the sidelobe frequency region, a null filter bank with a main-to-sidelobe ratio that meets the required requirements is obtained. Finally, this embodiment replaces the null filter bank's -1, 0, and 1 Doppler channels (determined by the null filter bank's frequency coverage) with the corresponding channels of a DFT filter bank capable of detecting zero-speed and low-speed targets. This results in a non-uniform filter bank structure that combines strong clutter suppression with low-speed target detection performance.

[0056] Step S2: Iteratively construct a clutter map for the first set of data under a clear sky background according to the servo scanning cycle. The specific process is as follows: The collected raw intermediate frequency echo data is sequentially processed through mixing, low-pass filtering, pulse compression, and non-uniform filtering to obtain a two-dimensional range-Doppler matrix. The output of the clutter channels (-1, 0, and 1 channels) is then extracted as input data for clutter map construction. A three-dimensional azimuth-range-velocity clutter map is constructed through a sequential iteration process. This embodiment specifically selects 50 consecutive servo scan cycle data and iteratively updates them in chronological order. Figure 6 The statistical results of 50 iterative updates of each range unit of the zero channel at a fixed azimuth are shown. This result verifies the characteristic that the clutter map gradually converges with the increase of the number of scans under clear weather conditions, providing a reliable environmental background reference for subsequent clutter suppression.

[0057] Step S3: For the second set of measured data under rainfall environment, non-uniform filtering is used in combination with clutter cancellation technology to effectively extract rainfall echoes, and then the Gaussian fitting algorithm is used to restore the complete rainfall echo spectrum characteristics. Figure 7 The general steps of the rainfall echo detection and recovery method in an embodiment of the present invention are shown.

[0058] First, the collected raw intermediate frequency echo data undergoes frequency mixing, low-pass filtering, pulse compression, and non-uniform filtering to obtain a two-dimensional range-Doppler matrix containing the clutter and rainfall characteristics at each azimuth. This example selects the spectral distributions of typical clutter and rainfall echoes from a radar servo at two specific azimuth angles. The following describes these two sets of data and uses them as a basis to verify the effectiveness of this method.

[0059] The rainfall echo at the first azimuth angle has a certain Doppler frequency, and its spectrum peak is located at the 35th Doppler unit. Figure 8The comparison of the amplitude-frequency characteristics of the range unit after processing with a nonuniform filter bank and a DFT filter bank with a Hamming window is shown. As can be seen, the nonuniform filter bank effectively suppresses ground clutter against a strong ground clutter background due to its wide and deep notch at zero frequency, improving the signal-to-clutter ratio of the rainfall echo.

[0060] Since the clutter channel signal amplitude of this range unit is lower than the corresponding clutter map threshold, it is replaced by the system average noise, and the replaced spectrum is restored by Gaussian fitting. Figure 9 The comparison results of the amplitude-frequency response before and after processing using the Gaussian fitting method are shown.

[0061] At the second azimuth angle, the peak of the rainfall echo spectrum is close to the clutter channel. After the clutter map cancellation detection, the signal amplitude at channel 1 is greater than the corresponding clutter point threshold and is retained. The signal amplitudes at channels -1 and 0 are less than the corresponding clutter point threshold and are therefore replaced by noise power. Subsequently, the three channel data are Gaussian fitted with the full channel data of the null filter to recover the zero-speed rainfall echo submerged by the clutter. Figure 10 As shown in the figure, by comparing the spectrum output by the non-uniform filter with the spectrum restored by clutter pattern cancellation and Gaussian fitting, it can be seen that this method can effectively handle the overlap of slow rainfall echoes and clutter: on the one hand, the slow rainfall echo component is accurately detected through clutter pattern cancellation, and on the other hand, the amplitude of the zero-speed target signal is restored using Gaussian fitting. Based on this result, the spectral moment is calculated, which significantly improves the estimation accuracy of rainfall parameters.

[0062] In one embodiment of the present invention, a computer device is provided, comprising a processor and memory. The memory is configured to store a computer program, the computer program including program instructions, and the processor is configured to execute the program instructions stored in a computer storage medium. The processor may be a central processing unit (CPU), or may also be another general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. It serves as the computing and control core of a terminal and is adapted to implement one or more instructions, specifically, to load and execute one or more instructions from a computer storage medium to implement a corresponding method flow or function. The processor described in this embodiment of the present invention can be used in a method for clutter suppression and high-precision echo recovery of a rain radar.

[0063] The present invention also provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device, used to store programs and data. It is understood that the computer-readable storage medium herein may include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides storage space, which stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for being loaded and executed by a processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer-readable storage medium herein may be high-speed RAM memory or non-volatile memory, such as at least one disk drive. The processor may load and execute the one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the method for rain radar clutter suppression and high-precision echo recovery described in the above-mentioned embodiment.

[0064] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0066] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0067] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for suppressing clutter and recovering high-precision echoes of a rain radar, characterized in that: The following steps are involved: Based on the weather radar system parameters and the intensity and spectral width of the actual scene clutter, an MTD filter bank based on a digital synthesis algorithm is constructed; the channels of the MTD filter bank all have nulling characteristics, forming a nulling channel group; Based on the MTD filter bank, the DFT filter bank is used to replace the channel where the clutter is located to obtain a non-uniform filter bank; The echo signal received by the weather radar system is preprocessed as the input signal of the non-uniform filter bank, and the matrix multiplication operation is performed on the filter coefficients to generate a two-dimensional range-Doppler spectrum matrix; Based on the two-dimensional range-Doppler spectrum matrix, a three-dimensional dynamic clutter map of azimuth, range, and velocity is iteratively constructed for the clutter channel during each azimuth beam dwell period. The clutter map cancellation method is then used to process each subsequent frame of CPI data to obtain slow / stationary targets. The moving target is extracted through the null channel, and the data of the moving target is fused with the slow / stationary target to reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal.

2. The method for suppressing clutter and recovering high-precision echoes of a rain radar according to claim 1, characterized in that: The sidelobe area of ​​the MTD filter can obtain sidelobe levels of different shapes by changing the size of the interference power placed, which specifically includes the following steps: Assuming the ground clutter power spectrum to be a Gaussian spectrum with the center of the spectrum being zero, the ground clutter covariance matrix is for: Where, , , , is the ground clutter power spectrum variance; is the ground clutter power, is the filter order, is the weather radar pulse repetition period; With Interference signals are input to the filter, and their frequencies are And evenly cover the entire pulse repetition frequency range, then the covariance matrix of clutter, interference and noise is for: Where, is the noise power, The frequency is The power of interference, is the pilot vector of frequency; but MTD filter bank The expression of the weight vector of a filter is: Where, For the The weight vector of the filter.

3. The method for suppressing clutter and recovering high-precision echoes of a rain measuring radar according to claim 1, characterized in that: The step of preprocessing the echo signal received by the weather radar system as the input signal of the non-uniform filter bank, and performing matrix multiplication operation in combination with the filter coefficients to generate a two-dimensional range Doppler spectrum matrix specifically includes: The echo signals received by the weather radar system are subjected to digital down-conversion and pulse compression processing in sequence to obtain a coherent pulse sequence; Multiply the filter coefficients with the coherent pulse sequence to obtain a two-dimensional range-Doppler spectrum matrix; Assume that the coherent pulse sequence of the input signal is: Where, , , represents transpose; represents the zeroth moment, Indicates the time; Input Data , For signal; for clutter; is the noise vector; When performing MTD filtering, the filter output is: Where, is the weight coefficient of the non-uniform Doppler filter group.

4. The method for suppressing clutter and recovering high-precision echoes of a rain radar according to claim 1, characterized in that: The step of iteratively constructing an azimuth-range-velocity three-dimensional dynamic clutter map for the clutter channel within each azimuth beam dwell period based on the two-dimensional range Doppler spectrum matrix specifically includes: In each azimuth beam dwell period, the data of each range unit output by the clutter channel is stored in the storage unit. As the beam scans, the signal of each storage unit is recursively updated. The specific expression is: Where, is a factor less than 1.

5. The method for suppressing clutter and recovering high-precision echoes of a rain measuring radar according to claim 1, characterized in that: The step of using the clutter map cancellation method to process each subsequent frame of CPI data to obtain a slow / stationary target specifically includes: After the clutter map data is relatively stable, the clutter map threshold is set and the current signal amplitude is compared with the clutter map threshold. If the signal amplitude exceeds the clutter map threshold, it is determined to be a valid echo and the original data is retained. Otherwise, the signal is considered to be a clutter component and is replaced with average noise.

6. The method for suppressing clutter and recovering high-precision echoes of a rain measuring radar according to claim 5, characterized in that: The calculation formula of the clutter map threshold is: Where, is the threshold factor.

7. The method for suppressing clutter and recovering high-precision echoes of a rain measuring radar according to claim 1, characterized in that: In the step of fusing the data of the moving target and the slow / stationary target to reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal, a Gaussian fitting algorithm is used for data fusion.

8. A rain radar clutter suppression and high-precision echo recovery system, characterized in that: include: An MTD filter bank construction module is used to construct an MTD filter bank based on a digital synthesis algorithm based on weather radar system parameters and the intensity and spectral width of actual scene clutter; the channels of the MTD filter bank all have nulling characteristics, forming a nulling channel group; A non-uniform filter bank construction module is used to replace the MTD filter bank with a DFT filter bank for the channel where the clutter is located to obtain a non-uniform filter bank; The two-dimensional range-Doppler spectrum matrix generation module is used to pre-process the echo signal received by the weather radar system as the input signal of the non-uniform filter bank, and perform matrix multiplication operation with the filter coefficients to generate a two-dimensional range-Doppler spectrum matrix; The slow / stationary target detection module is used to iteratively construct a three-dimensional dynamic clutter map of azimuth, range, and velocity for the clutter channel within each azimuth beam dwell period based on the two-dimensional range-Doppler spectrum matrix, and then use the clutter map cancellation method to process each subsequent frame of CPI data to obtain slow / stationary targets; The moving target detection and fusion module is used to extract the moving target through the null channel, fuse the moving target with the slow / stationary target data, and reconstruct the complete amplitude-frequency characteristics of the rainfall echo signal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for suppressing clutter and recovering high-precision echo of a rain measuring radar as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of a method for suppressing clutter and recovering high-precision echoes of a rain measuring radar as described in any one of claims 1 to 7 are implemented.

Images