Meteorological radar signal and data processing system and method based on software radio

Abstract

The invention relates to the technical field of meteorological detection, and discloses a meteorological radar signal and data processing system and method based on software radio. A meteorological radar signal and data processing system based on software radio comprises an antenna, a microwave front end, an ADC module, an SDR hardware platform and a display and control console. An excitation source signal is radiated by an antenna, an echo signal is converted into a 60MHz intermediate frequency signal through a microwave front end, the ADC module realizes echo digitization through high-speed sampling and then sends the signal to an SDR hardware platform for processing, and finally the signal is transmitted to a display and control console through a network so as to display wind field synthesis data. The radar detection precision, the anti-interference capability and the data processing efficiency can be improved, and accurate inversion and quality control of a wind field are realized; the device is suitable for signal acquisition, processing and wind field synthesis of the wind finding radar. The problems of sensitivity, anti-interference performance, real-time performance, detection precision and the like of a traditional meteorological radar are solved, and a more efficient and reliable solution is provided for meteorological detection.

Description

Technical Field

[0001] The present invention relates to the field of meteorological detection technology, and in particular to a system and method for processing meteorological radar signals and data based on software defined radio. Background Art

[0002] With the development of meteorological detection technology, traditional monopulse weather radar systems based on analog signal processing have gradually shown limitations in detection accuracy. These systems generally have the following technical defects: 1. Insufficient sensitivity: The inherent noise of the analog receiving link limits the system's ability to detect weak weather echo signals, making it difficult to effectively identify meteorological targets such as weak precipitation. 2. Weak anti-interference ability: Analog filters are unable to effectively suppress noise such as ground clutter and radio frequency interference, resulting in the power spectrum data being easily "contaminated"; 3. Data processing limitations: The traditional method of analog phase detection and FFT spectrum analysis is difficult to achieve accurate wind field inversion under complex meteorological conditions.

[0003] Although the existing digital improvement solution uses digital intermediate frequency sampling technology, it still has the following technical contradictions: 1. It is difficult to balance the requirements of detection height and distance resolution when selecting pulse width; 2. The Doppler velocity ambiguity problem limits the radial velocity measurement range; 3. The balance between real-time processing capability and computing accuracy is prominent.

[0004] Especially in an environment with strong ground clutter interference, the existing system's recognition accuracy of dangerous weather phenomena such as low-level wind shear is significantly reduced. Summary of the Invention

[0005] The present invention provides a system and method for processing weather radar signals and data based on software defined radio (SDR), thereby improving radar detection accuracy, anti-interference capability, and data processing efficiency, achieving accurate inversion and quality control of wind fields, and resolving the technical problems of low analog signal processing sensitivity, poor anti-interference capability, and insufficient data processing accuracy in traditional weather radar systems.

[0006] According to one aspect of the present invention, a system for processing weather radar signals and data based on software defined radio (SDR) is provided, comprising an antenna, a microwave front end, an ADC module, an SDR hardware platform, and a display and control console. The excitation source signal is radiated by the antenna, and the echo signal is converted into a 60 MHz intermediate frequency signal by the microwave front end. The ADC module digitizes the echo through high-speed sampling and sends the digitized echo to the SDR hardware platform for processing. The digitized echo is finally transmitted to the display and control console via a network, thereby displaying the synthetic wind field data.

[0007] Furthermore, the processing system also includes a timing system, which uses an Ethernet token ring trigger mechanism to ensure the control timing of the radar working mode.

[0008] According to another aspect of the present invention, a method for processing weather radar signals and data based on software defined radio (SDR) is provided. The method employs the above-mentioned weather radar signal and data processing system based on software defined radio (SDR), and includes the following steps: S100, detection mode design: generating control instructions based on meteorological observation requirements to dynamically configure the transmission parameters and scanning strategy of the weather radar; S200, processing system response: executing radio frequency signal transmission and echo acquisition through an SDR hardware platform based on the control instructions, and providing hardware resources required for real-time processing; S300, signal processing: receiving raw signal data transmitted by the processing system, and performing pulse compression, moving target detection, and Doppler spectrum analysis based on the parameter configuration of the detection mode design to generate preprocessed data; S400, data processing: performing wind field inversion and quality control on the preprocessed data, outputting three-dimensional wind field information, optimizing hardware resource allocation of the system structure based on quality feedback, and adjusting the parameters of the detection mode design according to performance requirements; Steps S100 to S400 form a closed-loop processing flow, wherein the quality feedback includes signal-to-noise ratio, velocity ambiguity, and data consistency indicators, and the performance requirements include detection accuracy, real-time performance, and anti-interference capability.

[0009] Furthermore, the parameters of the detection mode design in step S100 are optimized, specifically including: height resolution: using a combination of short pulses and long pulses, balancing the height resolution and detection altitude through pulse compression technology, and ensuring the overlap of detection altitudes between modes; maximum unambiguous Doppler velocity: set to 15m / s below 5km and 20m / s above 5km to reduce the probability of velocity ambiguity; velocity resolution: ≤0.2m / s, guaranteed by the number of FFT points, the number of FFT points ≥2×(maximum unambiguous Doppler velocity / 0.2); time resolution: combining the number of beams, the number of modes, the number of time domain accumulations, the number of FFT points and the spectral average calculation to achieve parallel data processing and transmission.

[0010] Furthermore, the detection mode design in step S100 specifically includes: S101, determining the pulse width according to the height resolution; S102, determining the pulse repetition period according to the detection height and the detection range ≥ 1.5 times the target height; S103, determining the number of coherent averages according to the maximum unambiguous speed and the pulse repetition period; S104, determining the number of FFT points; S105, determining the spectral average according to the beam dwell time of 20-30s.

[0011] Furthermore, the signal processing in step S300 is as follows: saturation detection is performed on the digitized signal, and the gain of the receiving channel is dynamically adjusted through AGC to ensure the dynamic range of the signal; a linear phase filter is used to suppress interference; in-phase accumulation is performed in the time domain according to the phase relationship to enhance the signal-to-noise ratio; FFT spectrum analysis is performed on the accumulated signal after windowing, with the number of FFT points ≥ effective bandwidth / 1.6 Hz, and a frequency domain filter with a good rectangular coefficient and a small notch width is used to suppress ground clutter; and the corresponding frequency points of P power spectral density functions are averaged to further reduce noise.

[0012] Furthermore, the signal processing flow of the SDR hardware platform specifically includes: high-speed A / D sampling and saturation detection, first performing saturation detection on the intermediate frequency direct sampling signal, adjusting the receiving channel through the AGC control loop to ensure that the input data is not saturated before sampling; time domain filtering, designing a linear phase filter to suppress interference such as secondary range folding echoes and rainfall echoes, improving the signal-to-noise ratio and reducing the FFT calculation amount; time domain averaging, adding the echoes of adjacent cycles in phase according to the phase relationship, improving the echo signal-to-noise ratio and target detection capability; spectrum analysis, windowing the time domain average signal, combining the I and Q signals into a complex signal and then performing FFT analysis, and the number of FFT points is determined according to the effective bandwidth and Doppler spectrum interval; frequency domain filtering, using a filter with a good rectangular coefficient and a small notch width to suppress ground clutter and retain meteorological echoes; spectrum averaging, averaging the values ​​at the corresponding frequencies of P power spectral density functions to further improve the signal-to-noise ratio.

[0013] Furthermore, the data processing in step S400 specifically includes: S401, target detection and spectral moment calculation: identifying the echo signal of the meteorological target through the power spectrum peak detection algorithm, and calculating the noise level, Doppler velocity, spectral width and signal-to-noise ratio based on the power spectrum peak; S402, consistency averaging processing: performing time domain averaging on multiple Doppler velocity measurements with the same beam pointing and within the same range gate to suppress random measurement errors; S403, wind vector inversion: adopting a three-beam or five-beam scanning strategy, and solving the horizontal wind vector component based on the Doppler velocity measurements with different beam pointings by the least squares method; S404, data quality control: marking the data credibility through wind shear check, continuity check and two-dimensional median check; S405, anti-interference processing: adopting half-plane cancellation, data smoothing, pattern recognition, multi-peak extraction, median filtering and cluster analysis to process the power spectrum.

[0014] Furthermore, the data quality control in step S404 specifically includes: S4041, spatial consistency check based on wind field shear threshold; S4042, outlier removal based on time continuity constraint; S4043, smoothing the wind field data using a two-dimensional median filter.

[0015] Furthermore, the anti-interference processing in step S405 specifically includes: S4051, suppressing ground clutter through the half-plane cancellation algorithm; S4052, using the Savitzky-Golay filter for spectral smoothing; S4053, identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S4054, combining morphological median filtering to eliminate pulse interference noise.

[0016] The present invention has the following beneficial effects: 1. Improve detection accuracy and sensitivity: The high-speed ADC module is used to directly digitize the intermediate frequency signal, avoiding the noise accumulation and distortion problems in the traditional analog signal processing link, ensuring the complete acquisition of weak meteorological echo signals, and thus improving the accuracy of wind field inversion.

[0017] 2. Enhanced anti-interference capability: The SDR hardware platform uses algorithms such as digital filtering and adaptive clutter suppression to effectively eliminate noise pollution to the power spectrum caused by ground clutter, radio frequency interference, etc., ensuring data reliability under complex meteorological conditions.

[0018] 3. Optimizing the contradiction between pulse width and detection height: Based on the programmable signal processing capabilities of SDR, the system can flexibly adjust the pulse width and sampling strategy to take into account the needs of long-distance detection and high resolution, breaking through the limitations of traditional radar between detection range and accuracy.

[0019] 4. Solve the Doppler velocity ambiguity problem: The SDR hardware platform supports real-time Doppler deambiguation algorithm, which expands the measurement range of radial velocity, avoids spectrum aliasing caused by traditional FFT analysis, and improves the continuity of wind field data.

[0020] 5. Improve real-time performance and processing efficiency: Through high-speed digital signal processing and networked data transmission, the system can quickly complete IQ demodulation, spectral moment calculation, and wind field synthesis, meeting the real-time requirements of meteorological monitoring while reducing the risk of performance degradation caused by aging of analog components.

[0021] 6. Achieve integrated and intelligent processing: The display and control console combines quality control algorithms to visualize and correct wind field data, further optimizing data availability. It is suitable for monitoring dangerous weather conditions such as low-altitude wind shear in strong ground clutter environments.

[0022] In addition to the above-described objects, features and advantages, the present invention has other objects, features and advantages. The present invention will be further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which constitute part of the present invention, are provided to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings: Figure 1 This is a block diagram of the principle of a simple wind measurement radar according to a preferred embodiment of the present invention; Figure 2 1 is a flow chart of wind profiler radar signal processing according to a preferred embodiment of the present invention; Figure 3 4 is a flow chart of wind profiler radar data processing according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0024] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways as defined and covered below.

[0025] like Figure 1As shown, the weather radar signal and data processing system based on software defined radio (SDR) of this embodiment includes an antenna, a microwave front end, an ADC (analog-to-digital converter) module, an SDR hardware platform, and a display and control console. The excitation source signal is radiated by the antenna, and the echo signal is converted into a 60MHz intermediate frequency signal by the microwave front end. The ADC module digitizes the echo through high-speed sampling and sends it to the SDR hardware platform for processing. It is finally transmitted to the display and control console through the network, and then the wind field synthetic data is displayed. The weather radar signal and data processing system based on software defined radio (SDR) of the present invention uses a high-speed ADC module to directly digitize the intermediate frequency signal, avoiding the noise accumulation and distortion problems in the traditional analog signal processing link, ensuring the complete acquisition of weak weather echo signals, and thus improving the accuracy of wind field inversion. The SDR hardware platform effectively eliminates the pollution of noise such as ground clutter and radio frequency interference on the power spectrum through algorithms such as digital filtering and adaptive clutter suppression, ensuring data reliability under complex meteorological conditions. Based on the programmable signal processing capability of SDR, the system can flexibly adjust the pulse width and sampling strategy to take into account the requirements of long-distance detection and high resolution. The SDR overcomes the limitations of traditional radar between detection range and accuracy. The SDR hardware platform supports real-time Doppler deambiguation algorithms, expanding the radial velocity measurement range, avoiding spectral aliasing caused by traditional FFT analysis, and improving the continuity of wind data. Through high-speed digital signal processing and networked data transmission, the system rapidly completes IQ demodulation, spectral moment calculation, and wind field synthesis, meeting the real-time requirements of meteorological monitoring while reducing the risk of performance degradation caused by analog component aging. The display console, combined with quality control algorithms, visualizes and corrects wind data, further optimizing data availability and making it suitable for monitoring hazardous weather conditions such as low-level wind shear in strong ground clutter environments. Through fully digital signal acquisition and a programmable processing architecture, the system overcomes the technical bottlenecks of traditional meteorological radar in terms of sensitivity, interference rejection, real-time performance, and detection accuracy, providing a more efficient and reliable solution for meteorological detection. An ADC (analog-to-digital converter) is an electronic device or functional module that converts continuous analog signals into discrete digital signals. Ethernet Token Ring is a deterministic network access control mechanism in which all nodes are connected into a logical ring, with tokens passed between nodes in a fixed sequence. Optionally, the excitation source signal is amplified and converted by a T / R switch before being radiated by the antenna.

[0026] like Figure 1As shown in this embodiment, the software-defined radio (SDR)-based weather radar signal and data processing system adopts a modular design and is divided into two major components: onboard and offboard. The onboard component includes the antenna, servo system, transmitter, and microwave front-end (with integrated amplification, mixing, and control functions). The antenna achieves beam steering through the servo system and forms a radio frequency (RF) signal transceiver link with the transmitter and microwave front-end. The microwave front-end down-converts the echo signal and outputs it to a digital intermediate frequency (IF) receiver (including signal processing). The core modules of the offboard component include a digital IF receiver, an integrated monitoring and timing system, and a data processing and display control unit. These modules work together: the timing system provides synchronous triggering for transmission, sampling, and processing; the integrated monitoring module comprehensively manages the status of modules such as the transmitter and digital IF receiver; and after the digital IF receiver completes signal processing, it transmits the data to the data processing and display control unit for wind field synthesis and visualization. The system embodies the deep integration of hardware acquisition and software-defined processing (SDR) through signal flow (such as antenna → microwave front end → digital intermediate frequency receiver) and control flow (such as integrated monitoring → transmitter, timing → data processing), achieving full digitalization and programmability.

[0027] In this embodiment, the processing system also includes a timing system that utilizes an Ethernet token ring trigger mechanism to ensure control timing for the radar's operating modes. This deterministic triggering mechanism enables strict synchronous timing control for various radar system modules (such as the excitation source, ADC module, and signal processing), avoiding timing deviations associated with traditional asynchronous control and ensuring timing consistency for pulse transmission, echo sampling, and signal processing. The token ring's fixed-order transmission mechanism avoids conflicts caused by multi-node competition and reduces the risk of data loss or control failure due to timing disruptions, making it particularly suitable for weather radar scenarios with high real-time requirements. A programmable token allocation strategy dynamically adjusts the switching timing of radar operating modes to meet varying detection range and resolution requirements while maintaining mode transition stability. Distributed timing triggering using a standardized Ethernet protocol replaces traditional dedicated synchronization lines (such as cables or FPGA hardwiring), simplifying system architecture and reducing wiring costs. The token ring's deterministic transmission characteristics suppress external electromagnetic interference (EMI) that perturbs control signals, ensuring stable radar operation in complex electromagnetic environments.

[0028] The software-defined radio (SDR)-based weather radar signal and data processing method of this embodiment utilizes the aforementioned software-defined radio (SDR)-based weather radar signal and data processing system, including the following steps: S100, detection mode design: generating control instructions based on meteorological observation requirements and dynamically configuring the weather radar's transmission parameters and scanning strategy; S200, processing system response: executing radio frequency signal transmission and echo acquisition via the SDR hardware platform based on the control instructions, and providing the hardware resources required for real-time processing; S300, signal processing: receiving raw signal data transmitted by the processing system and performing pulse compression, moving target detection, and Doppler spectrum analysis based on the parameters configured in the detection mode design to generate preprocessed data; S400, data processing: performing wind field inversion and quality control on the preprocessed data, outputting three-dimensional wind field information, optimizing the hardware resource allocation of the system structure based on quality feedback, and adjusting the detection mode design parameters according to performance requirements. Steps S100 to S400 form a closed-loop processing flow. Quality feedback includes signal-to-noise ratio, velocity ambiguity, and data consistency indicators, and performance requirements include detection accuracy, real-time performance, and anti-interference capability. This method, based on software-defined radio (SDR) weather radar signal and data processing, dynamically adjusts transmission parameters and scanning strategies (such as PRF and pulse width) through real-time interaction between quality feedback and performance requirements. It automatically increases the sampling rate in heavy precipitation scenarios and optimizes energy consumption in clear-sky mode, achieving an adaptive balance between detection accuracy and resource consumption. The SDR hardware platform dynamically reconfigures the processing pipeline based on control instructions, ensuring strict synchronization between hardware acceleration units for algorithms like pulse compression and real-time acquisition, addressing data pipeline blockage issues caused by the disconnect between software and hardware in traditional systems. The method automatically selects the optimal filtering algorithm combination based on the current environmental noise characteristics, suppressing ground clutter while retaining weak meteorological signals, improving the signal-to-noise ratio by more than 5dB compared to fixed processing links. Three-dimensional wind field results are cross-validated using data consistency indicators, automatically identifying and eliminating anomalous data points caused by velocity ambiguity or beam obstruction, reducing inversion errors compared to traditional methods. A system health model is established based on long-term accumulated signal-to-noise ratio (SNR) and ambiguity metrics. Machine learning is used to optimize hardware resource allocation strategies, mitigating performance degradation caused by aging of key components such as ADC modules. Through the deep coupling of software-defined radar architecture and closed-loop control mechanism, the technical bottleneck of traditional weather radar's fixed parameters and rigid processing procedures has been broken through.

[0029] In this embodiment, the parameters of the detection mode design in step S100 are optimized, specifically including: height resolution: using a combination of short pulses and long pulses, balancing the height resolution and detection altitude through pulse compression technology, and ensuring the overlap of detection altitudes between modes; maximum unambiguous Doppler velocity: set to 15m / s below 5km and 20m / s above 5km to reduce the probability of velocity ambiguity; velocity resolution: ≤0.2m / s, guaranteed by the number of FFT points, which is ≥2×(maximum unambiguous Doppler velocity / 0.2); time resolution: combining the number of beams, the number of modes, the number of time domain accumulations, the number of FFT points, and the spectral average to achieve parallel data processing and transmission. The emission strategy combining short pulses and long pulses, combined with pulse compression technology, ensures high resolution in the near area (such as 0-5km) and sufficient detection sensitivity in the far area (such as 5-20km). The vertical detection blind spot caused by the traditional single pulse mode is eliminated through the high overlap design between modes, and the continuity of the vertical profile of the wind field is improved. The maximum unambiguous speed is dynamically set according to the altitude layer (15m / s at low altitude and 20m / s at high altitude). Combined with the typical vertical distribution characteristics of atmospheric motion speed, the speed ambiguity probability is controlled below 5%. At the same time, the adaptive PRF adjustment technology is used to automatically switch modes in strong wind shear scenarios to avoid the spectrum aliasing problem caused by the traditional fixed PRF. Based on the strict matching relationship between the number of FFT points and the velocity resolution (number of points ≥ 2×V max / 0.2), ensuring a speed resolution of 0.2m / s, significantly enhancing the detection capability of weak wind fields (such as boundary layer breeze circulation) and reducing wind speed errors; through the joint modeling of beam dwell time, FFT calculation amount and time domain accumulation number, a hardware-level parallel mechanism is established between the data processing pipeline and the radar transmission timing, so that the system can complete the scanning of 128 beam positions while maintaining a 10-second refresh rate, which is significantly more efficient than the traditional serial processing solution; through the deep coupling of physical constraints and algorithm requirements, the difficult problem of "high resolution-large coverage-high precision-real-time" in meteorological radar is solved.

[0030] In this embodiment, the detection mode design in step S100 specifically includes: S101, determining the pulse width according to the height resolution; S102, determining the pulse repetition period based on the detection height and the detection range ≥ 1.5 times the target height; S103, determining the number of coherent averages based on the maximum unambiguous speed and the pulse repetition period; S104, determining the number of FFT points; S105, determining the spectral average based on the beam dwell time of 20-30s. By dynamically selecting pulse width and repetition period based on target altitude, the system avoids the near-field blind spot and far-field insensitivity issues associated with traditional fixed pulse parameters, ensuring a detection range of 1.5 times the target altitude, enabling continuous and stable detection of meteorological targets at all altitude levels. Dynamically setting the number of coherent averaging times based on the constraint relationship between the maximum unambiguous velocity and the pulse repetition period suppresses spectral contamination from random noise and prevents loss of information about rapidly changing wind fields due to over-averaging, significantly reducing the ambiguity in velocity inversion. By scientifically matching the number of FFT points with the system's velocity resolution requirements, the system avoids spectrum leakage or computational resource waste caused by traditional empirical values, ensuring effective extraction of weak wind signals while maintaining reasonable real-time performance. By controlling the beam dwell time to 20-30 seconds and dynamically allocating spectral averages, the system meets the requirements for monitoring wind field temporal evolution while improving the signal-to-noise ratio through sufficient signal accumulation, resolving the conflict between data quality and refresh rate in traditional fixed-dwell time modes. By modeling the physical relationships between parameters, a complete mapping chain from underlying hardware constraints to upper-level meteorological requirements is constructed, enabling the radar system to automatically generate the optimal detection mode under various observation scenarios.

[0031] In this embodiment, the signal processing in step S300 is as follows: saturation detection is performed on the digitized signal, and the gain of the receiving channel is dynamically adjusted through AGC to ensure the dynamic range of the signal; a linear phase filter is used to suppress interference; in-phase accumulation is performed in the time domain according to the phase relationship to enhance the signal-to-noise ratio; FFT spectrum analysis is performed on the accumulated signal after windowing, with the number of FFT points ≥ effective bandwidth / 1.6 Hz, and a frequency domain filter with an optimal rectangular coefficient and a small notch width is used to suppress ground clutter; and corresponding frequency points of P power spectral density functions are averaged to further reduce noise. By detecting signal saturation in real time and dynamically adjusting the receiving gain, the system avoids nonlinear distortion caused by strong echoes and quantization errors caused by weak signals, expanding the system's dynamic range and ensuring the simultaneous and accurate acquisition of meteorological targets of varying intensities, from heavy precipitation to weak turbulence. A digital filter with strictly linear phase characteristics eliminates radio frequency interference and ground clutter while preserving the phase integrity of the meteorological echo. Combined with a time-domain coherent accumulation technique based on phase consistency, the signal-to-noise ratio is improved, significantly enhancing the detection probability of weak meteorological targets. Spectral resolution is ensured to be better than 1 Hz by adaptively selecting a transform length that satisfies the requirement that the number of FFT points ≥ the effective bandwidth / 1.6 Hz. A frequency-domain filter with a steep transition band introduces less than 0.2 dB of signal amplitude distortion when suppressing ground clutter. Frequency-point averaging of P independent power spectra is performed, suppressing random noise fluctuations through statistical properties, reducing the standard deviation of the final wind speed estimate and, in particular, improving the reliability of the boundary layer wind field inversion under clear-sky mode.

[0032] In this embodiment, the signal processing flow of the SDR hardware platform specifically includes: high-speed A / D sampling and saturation detection, first performing saturation detection on the intermediate frequency direct sampling signal, and adjusting the receiving channel through the AGC control loop to ensure that the input data is not saturated before sampling; time domain filtering, designing a linear phase filter to suppress interference such as secondary range folding echoes and rainfall echoes, thereby improving the signal-to-noise ratio and reducing the FFT calculation amount; time domain averaging, adding the echoes of adjacent cycles in phase according to the phase relationship, thereby improving the echo signal-to-noise ratio and target detection capability; spectrum analysis, windowing the time domain average signal, combining the I and Q signals into a complex signal and then performing FFT analysis, with the number of FFT points determined according to the effective bandwidth and Doppler spectrum interval; frequency domain filtering, using a filter with a good rectangular coefficient and a small notch width to suppress ground clutter and retain meteorological echoes; spectrum averaging, averaging the values ​​at the corresponding frequencies of P power spectral density functions to further improve the signal-to-noise ratio. Pre-saturation detection and dynamic AGC adjustment ensure that the intermediate frequency signal is digitally sampled within the optimal dynamic range, avoiding nonlinear distortion caused by strong echoes and quantization errors of weak signals, thereby improving the system's effective dynamic range and eliminating spectrum leakage caused by signal saturation. A linear phase filter is used to precisely remove interference components such as secondary folded echoes, reducing the amount of ineffective FFT operations while preserving the phase characteristics of the meteorological echo. A time-domain coherent accumulation technique based on phase consistency is combined to improve the signal-to-noise ratio and significantly enhance the detection probability of low-observable targets such as weak turbulence. Spectral leakage energy is suppressed through adaptive windowing and a scientifically set FFT count (matching the effective bandwidth and Doppler resolution requirements). A frequency-domain filter with a steep transition band is used to simultaneously optimize the ground clutter suppression ratio and meteorological signal amplitude distortion. Frequency-point statistical averaging of P independent power spectra is performed, suppressing random noise fluctuations through the law of large numbers, ensuring that the standard deviation of the final wind speed estimate approaches the theoretical Cramer-Rao lower bound, particularly for controlling velocity measurement errors in clear-air turbulence observations.

[0033] In this embodiment, the data processing in step S400 specifically includes: S401, target detection and spectral moment calculation: identifying the echo signal of the meteorological target through the power spectrum peak detection algorithm, and calculating the noise level, Doppler velocity, spectral width and signal-to-noise ratio based on the power spectrum peak; S402, consistency averaging processing: performing time domain averaging on multiple Doppler velocity measurements with the same beam pointing and within the same range gate to suppress random measurement errors; S403, wind vector inversion: adopting a three-beam or five-beam scanning strategy, and solving the horizontal wind vector component based on the Doppler velocity measurements with different beam pointings through the least squares method; S404, data quality control: marking the data credibility through wind shear check, continuity check and two-dimensional median check; S405, anti-interference processing: using half-plane cancellation, data smoothing, pattern recognition, multi-peak extraction, median filtering and cluster analysis to process the power spectrum. Through adaptive power spectrum peak detection technology, meteorological echoes and interference signals can be effectively distinguished in complex noise backgrounds, solving the problem of missed detection in weak signal scenarios caused by traditional fixed threshold methods, and significantly improving the system's ability to identify low-observable targets such as weak turbulence; a time-domain coherent averaging algorithm is used to fuse multiple observation data, and a phase consistency verification mechanism is used to suppress random measurement errors, effectively improving the reliability and repeatability of Doppler velocity measurements; based on a multi-beam scanning strategy and a least-squares optimization algorithm, the inherent geometric error limitations of single-beam measurements are overcome, and the horizontal wind field components are accurately calculated, significantly improving the ability to analyze complex wind field structures; through multi-dimensional wind field feature verification (including spatial continuity, temporal consistency, and physical rationality), an automatic data credibility labeling mechanism is established to provide a reliability basis for subsequent data applications; by comprehensively using intelligent algorithms such as spectral domain cancellation, pattern recognition, and cluster analysis, meteorological echoes can be accurately extracted in scenarios where multiple interferences coexist, effectively solving the problem of misjudgment of traditional methods in interference environments such as strong ground clutter and biological targets. By establishing a closed-loop quality control chain from raw signal analysis to final product generation, the system is able to adapt to complex meteorological conditions and interference environments, providing highly reliable wind field observation data for meteorological monitoring.

[0034] In this embodiment, the data quality control in step S404 specifically includes: S4041, spatial consistency check based on wind field shear threshold; S4042, outlier removal based on time continuity constraint; S4043, smoothing wind field data using a two-dimensional median filter. By setting the wind field shear threshold, the physical rationality of the wind speed gradient of adjacent range gates or altitude layers is verified, effectively identifying and eliminating discontinuous wind field data caused by beam obstruction or velocity ambiguity, ensuring that the inversion results conform to the fluid continuity law of atmospheric motion; dynamic constraints are established based on the time scale characteristics of atmospheric evolution, automatically filtering out sudden change data points that do not conform to the meteorological evolution law within adjacent scanning cycles, and solving the problem of false wind field jumps caused by transient interference or system noise; using a two-dimensional median filter algorithm to suppress the impact of isolated noise points on the overall wind field while retaining the small- and medium-scale characteristics of the real wind field, especially effectively eliminating local data anomalies caused by radar sidelobe interference, and improving the spatial analysis usability of wind field products; by integrating atmospheric dynamics constraints with digital signal processing technology, the final output three-dimensional wind field data meets both physical rationality and measurement reliability requirements.

[0035] In this embodiment, the anti-interference processing in step S405 specifically includes: S4051, suppressing ground clutter through a half-plane cancellation algorithm; S4052, using a Savitzky-Golay filter for spectral smoothing; S4053, identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S4054, combining morphological median filtering to eliminate pulse interference noise. By establishing a dynamic clutter reference spectrum model and performing adaptive cancellation operations, the system achieves directional suppression of ground clutter while preserving the phase information of meteorological echoes, thus overcoming the technical challenge of weak precipitation signal loss caused by traditional static filtering. A Savitzky-Golay filter is used to perform joint differential smoothing in the time and frequency domains, effectively suppressing random noise while maintaining the integrity of the shape parameters of the meteorological spectrum peaks and avoiding the spectrum broadening effect caused by conventional moving average methods. A K-means clustering algorithm is used to perform multimodal decomposition of complex power spectra, automatically identifying spectral peak subspaces to achieve radial velocity decoupling of adjacent meteorological targets (such as stratus and convective clouds), thus overcoming the application limitations of traditional single-peak detection models. A morphological opening operation is combined with adaptive median filtering to simultaneously detect and eliminate anomalous spectral lines generated by transient interference such as lightning in both the time and frequency domains, preserving the temporal continuity of meteorological echoes. By establishing a closed-loop technical framework of "feature extraction-interference modeling-intelligent filtering," the system maintains an effective data acquisition rate even in harsh observation environments.

[0036] During implementation, a simplified weather radar signal and data processing system and method based on software-defined radio (SDR) technology was provided to improve radar detection accuracy, anti-interference capabilities, and data processing efficiency, enabling accurate wind field inversion and quality control. The system directly digitizes the 60MHz intermediate frequency signal using the SDR hardware platform, implementing signal processing steps such as time-domain filtering, frequency-domain filtering, and spectral averaging. It then combines three-beam or five-beam pointing to invert the wind field. Data processing utilizes algorithms such as consistency averaging, quality control, and multi-peak processing. The detection mode design balances height resolution and detection range, addressing the issues of traditional radar's limited accuracy and weak anti-interference capabilities. The system is suitable for high-precision detection of meteorological wind profiles and is suitable for signal acquisition, processing, and wind field synthesis by wind measurement radars.

[0037] A simple weather radar signal and data processing system and method based on software defined radio (SDR) technology, specifically including: 1. System structure: 1) Hardware Architecture: This includes the antenna, microwave front end, ADC (analog-to-digital converter), SDR hardware platform, and display console. The excitation source signal is amplified and converted by a T / R switch before being radiated by the antenna. The echo signal is converted to a 60MHz intermediate frequency signal by the microwave front end. The ADC digitizes the signal at a sampling rate of no less than 3μs and sends it to the SDR platform for processing. Finally, it is sent to the display console via the network.

[0038] 2) Timing system: It uses Ethernet token ring trigger mechanism with timing accuracy better than 10ns to ensure the control timing of radar working mode.

[0039] 2. Signal processing method: 1) High-speed A / D sampling and saturation detection: Perform saturation detection on the intermediate frequency direct sampling signal first, and adjust the receiving channel through the AGC control loop to ensure that the input data is not saturated before sampling.

[0040] 2) Time domain filtering: Design a linear phase filter to suppress interference such as secondary range folding echo and rainfall echo, improve the signal-to-noise ratio and reduce the amount of FFT calculation.

[0041] 3) Time domain averaging (coherent accumulation): Add the echoes of adjacent cycles in phase according to their phase relationship to improve the echo signal-to-noise ratio and target detection capability.

[0042] 4) Spectral analysis: Window the time-domain average signal, combine the I and Q signals into a complex signal, and then perform FFT analysis. The number of FFT points is determined according to the effective bandwidth and Doppler spectrum interval (≥1.6Hz).

[0043] 5) Frequency domain filtering: Use a filter with a good rectangular coefficient and a small notch width to suppress ground clutter and retain weather echoes.

[0044] 6) Spectral averaging (accumulation after inspection): Average the values ​​of P power spectrum density functions at corresponding frequencies to further improve the signal-to-noise ratio.

[0045] 3. Data processing method: 1) Target detection and spectral moment calculation: Identify signal spectrum peaks, calculate noise level, first-order moment (Doppler velocity), second-order moment (spectral width) and signal-to-noise ratio.

[0046] 2) Consistency Averaging: Average multiple Doppler velocity measurements of the same beam and range gate to eliminate isolated interference such as airplanes and cars.

[0047] 3) Wind profile calculation: Use the Doppler velocity of three-beam or five-beam pointing to invert the u, v, and w components of the wind.

[0048] 4) Quality control: Mark suspicious or invalid data through wind shear check, continuity check, two-dimensional median check, etc.

[0049] 5) Anti-interference algorithm: Use half-plane cancellation (using the symmetry of ground clutter), data smoothing, pattern recognition, multi-peak extraction, median filtering and cluster analysis to process the "contaminated" power spectrum.

[0050] 4. Detection mode design: 1) Parameter optimization: Height resolution: Combining short pulses (low mode) with long pulses (high mode), pulse compression technology is used to balance height resolution and detection height, ensuring high overlap between detection modes.

[0051] Maximum unambiguous Doppler velocity: Set to 15 m / s below 5 km and 20 m / s above 5 km to reduce the probability of velocity ambiguity.

[0052] Velocity resolution: ≤0.2 m / s, guaranteed by the number of FFT points (≥2×(maximum unambiguous Doppler velocity / 0.2)).

[0053] Time resolution: Combined with the number of beams, number of modes, time domain accumulation, number of FFT points and spectral averaging calculations, data processing and transmission can be carried out in parallel.

[0054] 2) Design steps: Determine the pulse width (based on the height resolution) → determine the pulse repetition period (based on the detection altitude, the detection range ≥ 1.5 times the target altitude) → determine the number of coherent averages (based on the maximum unambiguous velocity and the pulse repetition period) → determine the number of FFT points → determine the spectral average (based on the beam dwell time of 20-30s).

[0055] System workflow: The excitation source signal is amplified, T / R switched, and then radiated by the antenna. The echo is converted to a 60MHz intermediate frequency signal by the microwave front end. The ADC module digitizes the echo through high-speed sampling and transmits it to the SDR platform. The SDR performs time-domain filtering (linear phase filter to suppress interference), time-domain averaging (in-phase addition), windowing, and FFT spectrum analysis on the four parallel data channels (the number of FFT points is determined by the 1.6Hz spectral line spacing). Frequency-domain filtering (filter with a notch width ≤ 1Hz to suppress ground clutter) and spectral averaging (P = 16 averages) are then applied. Finally, combined with the azimuth and pitch data provided by the display console, direction finding and wind field synthesis are completed, and the data is sent to the display console via the network for display.

[0056] Data processing example: Taking wind measurement data from a specific region as an example, after receiving power spectrum data, interference signals are first removed through noise level calculation. Half-plane cancellation is used to remove ground clutter, and the remaining signal is averaged (averaged over 10 measurements at the same range gate). Horizontal wind speed and direction are inverted using Doppler velocity (slant beam projection ≤ 30 m / s) with three-beam pointing (zenith angle 15°). Invalid data is marked through vertical wind shear checks (wind shear thresholds for adjacent altitude layers are set at 5 m / s / 100 m) and temporal continuity checks (the difference threshold between previous and subsequent moments is set at 2 m / s). The resulting wind profile product has an altitude resolution of 50 m in low-mode and a velocity resolution of 0.15 m / s, meeting meteorological detection requirements.

[0057] Detection mode configuration: Low mode (short pulse): pulse width 1μs, pulse repetition period 500μs, coherent averaging times 8, FFT points 1024, spectral averaging times 12, achieving 50m altitude resolution and maximum detection altitude 3km.

[0058] High mode (long pulse): pulse width 5μs, pulse repetition period 1000μs, coherent averaging times 16, FFT points 2048, spectral averaging 8, altitude resolution 250m, maximum detection altitude 10km, and the two modes have an overlap of 1km in detection altitude, ensuring continuous inversion of wind profiles.

[0059] Matters not covered by the present invention are known technologies.

[0060] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above-described embodiments merely illustrate several implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that variations and improvements are possible without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

[0062] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A weather radar signal and data processing system based on software defined radio, characterized in that: Including antenna, microwave front end, ADC module, SDR hardware platform and display console; The excitation source signal is radiated by the antenna, and the echo signal is converted into a 60MHz intermediate frequency signal by the microwave front end. The ADC module digitizes the echo through high-speed sampling and sends it to the SDR hardware platform for processing. Finally, it is transmitted to the display console through the network to display the wind field synthetic data.

2. The system for processing weather radar signals and data based on software defined radio according to claim 1, wherein: The processing system also includes a timing system, which uses an Ethernet token ring trigger mechanism to ensure the control timing of the radar working mode.

3. A method for processing weather radar signals and data based on software defined radio, characterized in that: The system for processing weather radar signals and data based on software defined radio according to claim 1 or 2 comprises the following steps: S100, Detection Mode Design: Generate control instructions based on meteorological observation requirements and dynamically configure the transmission parameters and scanning strategy of the weather radar; S200, processing system response: Based on the control instruction, the SDR hardware platform performs radio frequency signal transmission and echo acquisition, and provides hardware resources required for real-time processing; S300, signal processing: Receives and processes the raw signal data transmitted by the processing system, performs pulse compression, moving target detection, and Doppler spectrum analysis based on the parameter configuration of the detection mode design, and generates pre-processed data; S400, data processing: performing wind field inversion and quality control on the pre-processed data, outputting three-dimensional wind field information, optimizing hardware resource allocation of the system structure based on quality feedback, and adjusting parameters of the detection mode design according to performance requirements; Steps S100 to S400 form a closed-loop processing flow. Quality feedback includes signal-to-noise ratio, velocity ambiguity, and data consistency indicators. Performance requirements include detection accuracy, real-time performance, and anti-interference capability.

4. The method for processing weather radar signals and data based on software defined radio according to claim 3, wherein: The parameter optimization of the detection mode design in step S100 specifically includes: High resolution: Combining short and long pulses, and using pulse compression technology to balance high resolution and detection height, ensuring high overlap between detection modes; Maximum unambiguous Doppler velocity: 15 m / s below 5 km, 20 m / s above 5 km, to reduce the probability of velocity ambiguity; Velocity resolution: ≤0.2m / s, guaranteed by FFT points, FFT points ≥2×(maximum unambiguous Doppler velocity / 0.2); Time resolution: Combined with the number of beams, number of modes, time domain accumulation, number of FFT points and spectral averaging calculations, data processing and transmission can be carried out in parallel.

5. The method for processing weather radar signals and data based on software defined radio according to claim 4, wherein: The detection mode design in step S100 specifically includes: S101, determining pulse width according to height resolution; S102, based on the detection height, the detection range is ≥ 1.5 times the target height, and the pulse repetition period is determined; S103, determining the number of coherent averaging times based on the maximum unambiguous speed and the pulse repetition period; S104, determining the number of FFT points; S105 : Determine the spectrum average based on the beam dwell time of 20-30 seconds.

6. The method for processing weather radar signals and data based on software defined radio according to claim 3, wherein: The signal processing in step S300 is as follows: Perform saturation detection on the digitized signal, and dynamically adjust the receiving channel gain through AGC to ensure the signal dynamic range; use a linear phase filter to suppress interference; perform in-phase accumulation in the time domain according to the phase relationship to enhance the signal-to-noise ratio; perform FFT spectrum analysis on the accumulated signal after windowing, with the number of FFT points ≥ effective bandwidth / 1.6Hz, and use a frequency domain filter with an excellent rectangular coefficient and a small notch width to suppress ground clutter; average the corresponding frequency points of P power spectral density functions to further reduce noise.

7. The method for processing weather radar signals and data based on software defined radio according to claim 6, wherein: The signal processing flow of the SDR hardware platform specifically includes: High-speed A / D sampling and saturation detection: saturation detection is performed on the intermediate frequency direct sampling signal first, and the receiving channel is adjusted through the AGC control loop to ensure that the input data is not saturated before sampling; Time domain filtering: design a linear phase filter to suppress the interference of secondary range folding echo and rainfall echo, improve the signal-to-noise ratio and reduce the amount of FFT calculation; Time domain averaging, adding echoes in phase according to the phase relationship of adjacent cycles, improving the echo signal-to-noise ratio and target detection capability; Spectral analysis: windowing the time-domain average signal, combining the I and Q signals into a complex signal, and then performing FFT analysis. The number of FFT points is determined by the effective bandwidth and Doppler spectrum interval; Frequency domain filtering uses filters with good rectangular coefficients and small notch widths to suppress ground clutter and retain weather echoes; Spectral averaging averages the values ​​of P power spectral density functions at corresponding frequencies to further improve the signal-to-noise ratio.

8. The method for processing weather radar signals and data based on software defined radio according to claim 3, wherein: The data processing in step S400 specifically includes: S401, target detection and spectral moment calculation: Identify the echo signal of the meteorological target through the power spectrum peak detection algorithm, and calculate the noise level, Doppler velocity, spectrum width and signal-to-noise ratio based on the power spectrum peak; S402, consistency averaging processing: performing time domain averaging on multiple Doppler velocity measurements with the same beam pointing direction and within the same range gate to suppress random measurement errors; S403, Wind Vector Inversion: Using a three-beam or five-beam scanning strategy, the horizontal wind vector components are calculated using the least squares method based on Doppler velocity measurements at different beam orientations. S404, Data Quality Control: Mark the data credibility through wind shear check, continuity check and 2D median check; S405, anti-interference processing: using half-plane cancellation, data smoothing, pattern recognition, multi-peak extraction, median filtering and cluster analysis to process the power spectrum.

9. The method for processing weather radar signals and data based on software defined radio according to claim 8, wherein: The data quality control in step S404 specifically includes: S4041. Spatial consistency test based on wind field shear threshold; S4042. Outlier elimination based on time continuity constraints; S4043. Use a two-dimensional median filter to smooth the wind field data.

10. The method for processing weather radar signals and data based on software defined radio according to claim 8, wherein: The anti-interference processing in step S405 specifically includes: S4051, suppressing ground clutter through half-plane cancellation algorithm; S4052, spectral smoothing using Savitzky-Golay filter; S4053, Identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S4054. Combine morphological median filtering to eliminate pulse interference noise.

Images