Multi-mode cooperative compact X-waveband weather radar system

Abstract

The invention relates to the technical field of meteorological radar detection and data processing, and discloses a multi-mode collaborative compact X-waveband weather radar system, which comprises a plurality of radar nodes, and adopts a time division collaborative mechanism to divide a work period into active detection and passive collaborative time slots. The node integrates a main observation and omnidirectional cooperative antenna, and the cooperative sequential control unit is switched between a precise calibration mode and a background maintenance mode according to beam pointing. And the baseband signal processing subsystem executes a self-adaptive frequency spectrum avoidance strategy, and enables the reference signal to avoid a meteorological echo main energy region by injecting the optimal frequency offset. The system utilizes a radar equation reverse solution logic to decouple and separate system gain drift and radome water film attenuation from a received signal, and performs networking data fusion in combination with a calibration confidence coefficient weight. On the premise that meteorological services are not interrupted, online mutual calibration and physical parameter decoupling of the whole network are achieved, and the observation consistency of the X-band radar network under the complex weather condition is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of meteorological radar detection and data processing, in particular to a multi-mode cooperative compact X-band weather radar system. BACKGROUND

[0002] X-band weather radar network observation has become the core means for monitoring mesoscale and small-scale severe convective weather due to its extremely high space-time resolution and flexible deployment. In order to ensure the quantitative accuracy of network observation data, it is necessary to maintain the consistency of the receiving and transmitting link gain of each radar node, and accurately correct the attenuation caused by electromagnetic wave propagation path and antenna cover water. Especially in the face of compact low-cost radar nodes, the hardware is greatly affected by environmental temperature, and the realization of all-weather uninterrupted business high-precision online calibration and data correction is the key to guarantee the quality of meteorological products.

[0003] At present, the performance parameter calibration of meteorological radar mainly depends on periodic offline field calibration, or internal self-calibration using built-in noise source and test loop. For network application, the existing data processing center usually adopts a mosaic algorithm based on geometric distance or height weight to fuse the echo data of multiple radar overlapping coverage areas to form a regional mosaic reflectivity factor map. Some advanced schemes also try to use the overlapping observation area between radars to test each other, and evaluate the consistency deviation of the radar by comparing the echo intensity difference of the same precipitation target.

[0004] Although the existing technology provides basic calibration and networking means, there are still obvious deficiencies in actual application. First, the built-in test loop usually does not include the antenna and antenna cover part, and cannot sense the gain loss caused by antenna cover aging or surface water film. The method based on the sun method or the field calibration ball needs to interrupt the normal observation business, and the calibration frequency is low, which cannot track the rapid gain drift of the compact transmitter caused by temperature change. Second, in the precipitation weather, X-band electromagnetic waves are easily affected by rain attenuation and antenna cover water film. The existing technology lacks effective physical layer decoupling mechanism, and it is difficult to distinguish the gain drift of the hardware itself and the signal attenuation caused by the external environment, resulting in distortion of the calibration parameters. In addition, when trying to calibrate the mutual transmission frequency signal in the precipitation process, the wideband meteorological background echo often overwhelms the narrowband reference signal, resulting in deterioration of the signal-to-noise ratio, making it difficult for the system to obtain reliable calibration data in complex weather background, and ultimately affecting the accuracy and consistency of the network fusion data. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a multi-mode cooperative compact X-band weather radar system, which solves the problem that the existing X-band radar network is difficult to effectively resist meteorological echo interference under complex weather conditions without interrupting meteorological observation services, and cannot real-time decouple and distinguish hardware gain drift and antenna cover water film attenuation to cause poor consistency of observation data.

[0006] To achieve the above object, the present application is implemented by the following technical solutions: The present application provides a multi-mode cooperative compact X-band weather radar system, which comprises a plurality of radar nodes distributed in different geographical locations, and the radar nodes divide the working period into active detection time slots and passive cooperative time slots in time sequence; Each radar node comprises an antenna subsystem, a radio frequency transceiver subsystem, a cooperative sequential control unit and a baseband signal processing subsystem; The antenna subsystem comprises a main observation antenna and an omnidirectional cooperative antenna; The radio frequency transceiver subsystem comprises a frequency offset injection transmitting unit and a sector synchronous pre-attenuation receiving unit; The cooperative sequential control unit is configured to real-time solve the relative geometric position relationship between the observation node and the cooperative node according to the beam pointing, and control the system to switch between the precision calibration mode and the background maintenance mode in the passive cooperative time slot; The baseband signal processing subsystem is configured to use radar equation inverse solving logic to decouple and separate the system gain drift and the antenna cover water film attenuation from the received reference signal, and use the system gain drift to calibrate the meteorological echo data obtained in the active detection time slot.

[0007] Preferably, the antenna subsystem adopts a zero-shielding composite structure layout; the omnidirectional cooperative antenna is installed on the central axis at the top of the antenna cover or the top point of the independent mast higher than the main observation antenna rotation envelope circle, so that the omnidirectional cooperative antenna has unobstructed line of sight in the horizontal plane when the main observation antenna performs 0-90 degree elevation angle and 0-360 degree azimuth scanning; The precision calibration mode is defined as a state in which the included angle error of the main beam pointing of the observation node and the cooperative node is less than half of the 3dB beam width of the main beam, at which time the system is configured to call the main observation antenna to transmit a reference signal; The background maintenance mode is defined as a state in which the main beam of the observation node points to the cooperative node but the cooperative node is not aligned with the observation node, at which time the system is configured to switch to the omnidirectional cooperative antenna to transmit a reference signal.

[0008] Preferably, the sector synchronous pre-attenuation receiving unit in the radio frequency transceiver subsystem is configured with a hierarchical protection circuit; The hierarchical protection circuit is connected in series with a high-power passive limiter in the front-end input circuit of the low-noise amplifier, for clamping the direct wave strong signal; The hierarchical protection circuit is connected in series with a high-power passive limiter in the front-end input circuit of the low-noise amplifier, for clamping the direct wave strong signal; The cooperative sequential control unit is configured to trigger the digital step attenuator to switch to a high attenuation state within the protection time before entering the passive cooperative time slot, so that the amplitude of the received reference signal falls back to the linear dynamic range of the analog-to-digital converter.

[0009] Preferably, the frequency offset injection transmitting unit is configured to perform an adaptive frequency spectrum avoidance strategy; The baseband signal processing subsystem is configured to analyze the weather echo Doppler spectrum of the last scanning period, extract the spectrum width and center speed, and calculate the optimal injection frequency offset amount; The optimal injection frequency offset amount is constrained to simultaneously satisfy the flat band bandwidth limitation of the receiver front-end filter and the spectrum isolation requirement relative to the weather echo center speed; The frequency offset injection transmitting unit controls the transmitter to generate a reference signal with a dynamic frequency offset in the passive cooperative time slot according to the optimal injection frequency offset amount, so that the reference signal avoids the main energy area of the weather echo in the frequency domain.

[0010] Preferably, the baseband signal processing subsystem includes an anti-aliasing spectrum separation module, which is configured to calculate a spectrum domain overlap coefficient of the reference signal; The spectrum domain overlap coefficient represents the ratio of the weather spectrum background energy within the reference signal integration bandwidth to the peak power of the reference signal; When the spectrum domain overlap coefficient is less than a preset overlap threshold, the module is configured to extract the reference signal using frequency domain notch filtering, and perform interpolation repair on the filtered frequency band to restore the weather spectrum; When the spectrum domain overlap coefficient is greater than or equal to the preset overlap threshold, the module is configured to trigger a calibration fuse mechanism, suspend gain inversion calculation, maintain the system gain drift amount at the last time, and output the original weather spectrum without repair.

[0011] Preferably, the baseband signal processing subsystem is configured to perform a mode-aware parameter update strategy: In the precise calibration mode, the system simultaneously updates the system gain drift amount and the radome water film attenuation amount using the received signal, and stores them as reference values; In the background maintenance mode, the system updates the system gain drift amount using the received signal, and performs a parameter retention strategy for the radome water film attenuation amount, directly calling the effective value obtained in the last precise calibration mode in the memory to participate in the calculation.

[0012] Preferably, the baseband signal processing subsystem is configured to calculate the system gain drift in the following manner: Acquire differential propagation phase shift slope distribution data along the baseline direction in the active detection time slot, and calculate the one-way total path attenuation of the baseline path through path integration; Calculate the ideal received power based on the transmitter's nominal power, the transmit antenna gain corresponding to the current operating mode, the observation antenna gain, the baseline free space loss, and the receiver filter correction loss. The actual received power is obtained by summing the equivalent power level output by the analog-to-digital converter and the compensation value of the digital step attenuator. The residual obtained by subtracting the ideal received power from the actual received power, and compensating for the total path attenuation and the attenuation of the radome water film, is the system gain drift.

[0013] Preferably, the baseband signal processing subsystem is configured to invert the attenuation of the radome water film by measuring the linear depolarization ratio increment of the reference signal; The inversion process is based on a pre-calibrated empirical polynomial model, which establishes a mapping relationship between the linear depolarization ratio increment and the signal attenuation value caused by the water film on the radome. The baseband signal processing subsystem is configured to calculate the linear depolarization ratio increment only through the high polarization purity signal received by the main observation antenna in the precision calibration mode.

[0014] Preferably, the system further includes a data fusion processing module for generating calibration confidence weights and performing gridded fusion; The calibration confidence weight is configured to be negatively exponentially correlated with the square of the system gain drift, negatively exponentially correlated with the time interval since the last precision calibration, and controlled by the water film timeliness factor. The water film timeliness factor is defined as follows: when the local reflectivity factor of the radar site is lower than the precipitation determination threshold, the value is 1; when the local reflectivity factor of the radar site is higher than or equal to the precipitation determination threshold, it exhibits a negative exponential relationship of rapid decay with time interval. The gridded fusion is configured to perform a weighted average of the observation data from multiple radar nodes using the calibration confidence weight and geometric distance weight.

[0015] Preferably, the baseband signal processing subsystem embeds sidelobe suppression logic, which is configured as follows: Set a minimum effective received power threshold, which is equal to the sum of the nominal transmit power of the cooperating node, the gain of the omnidirectional cooperating antenna, and the gain of the first sidelobe of the observation node, minus the free space transmission loss and the system decision margin. When the received reference signal power is lower than the minimum effective received power threshold, it is determined that the current link is in a sidelobe coupling or multipath interference state, and the inversion calculation of the system gain drift amount using the signal is prohibited.

[0016] The present application provides a multi-mode cooperative compact X-band weather radar system. 1、The present application divides the working period into active detection time slots and passive cooperative time slots through the cooperative sequential control unit, and automatically switches the precise calibration and background maintenance modes in the passive cooperative time slots according to the relative position of the beam. This design utilizes the idle time resources in the radar beam scanning process, establishes an online mutual calibration link without stopping meteorological observation business, solves the problem that traditional radars need to interrupt business for calibration, and realizes real-time monitoring and compensation of the gain drift of the radio frequency transceiver link.

[0017] 2、The present application adopts an adaptive spectrum avoidance strategy based on Doppler spectrum characteristics, calculates the best injection frequency offset amount that meets the dual constraints of hardware bandwidth and spectrum isolation, and controls the transmitter to generate a reference signal offset from the center frequency. This method can dynamically move the reference signal to a frequency band with weak meteorological echo energy, and cooperate with the spectral domain overlap coefficient detection and notch filtering processing to effectively avoid the covering of the weak reference signal by strong meteorological clutter, ensuring that high signal-to-noise ratio calibration data can be extracted from the mixed signal in precipitation weather.

[0018] 3、The present application constructs a closed-loop link budget model containing path attenuation, water film loss and system gain drift, uses dual polarization parameter inversion to invert the antenna cover water film attenuation, realizes the physical layer decoupling of external environmental loss and internal hardware drift. On this basis, the system further generates a calibration confidence weight according to the gain drift amount and the water film state, automatically reduces the weight of nodes seriously disturbed by the water film or unstable hardware state during network data fusion, thereby improving the observation consistency of the X-band radar network in complex weather conditions. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The present application is a radar node internal architecture and system composition schematic diagram; Figure 2 The present application is a mode determination and antenna switching logic schematic diagram based on beam pointing state awareness; Figure 3 The present application is a mode determination and antenna switching logic schematic diagram based on beam pointing state awareness; Figure 4 The present application is a sector synchronous radio frequency front end control and adaptive frequency offset injection logic timing diagram; Figure 5 The present application is a Doppler spectrum domain signal separation and purification process diagram; Figure 6 Flow chart for decoupling multi-dimensional physical parameter inversion and gain drift of the present application; Figure 7 Flow chart for networked data fusion processing based on dynamic confidence of the present application; Figure 8 Schematic diagram of zero-shielding physical layout and spatial structure of the antenna subsystem of the present application.

[0020] In the formula; 10, antenna subsystem; 20, cooperative sequential control unit; 30, radio frequency transceiver subsystem; 40, baseband signal processing subsystem. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the specification of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0022] Referring to the drawings Figure 1 , Figure 1 is a schematic diagram of the internal architecture and system composition of a radar node according to an embodiment of the present application. The present embodiment discloses a weather observation and online calibration system with high robustness, which adopts a time-division cooperative architecture and is mainly composed of multiple radar nodes distributed in different geographical locations at the hardware level. Each radar node includes a precise antenna subsystem 10, a cooperative sequential control unit 20 as the core scheduling hub, a radio frequency transceiver subsystem 30 with a protection mechanism, and a baseband signal processing subsystem 40 for executing complex algorithms in physical structure.

[0023] Referring to the drawings Figure 8 , Figure 8Fig. 1 is a schematic diagram of a zero-blockage physical layout and spatial structure of an antenna subsystem according to an embodiment of the present application. The antenna subsystem is configured as a zero-blockage dual-channel compound structure, aiming to achieve all-weather and all-directional signal interaction. The subjective antenna is configured as a high-gain parabolic antenna or a phased array antenna, used to perform regular weather sounding scanning and high-precision mutual calibration. A high-precision absolute azimuth encoder is integrated inside the mechanical turntable or electronic scanning controller, capable of real-time feedback of beam pointing with a resolution better than 0.5 degrees. The omnidirectional cooperative antenna is specifically installed on the central axis of the antenna cover or the top of the independent mast above the main antenna rotation envelope circle. This specific physical layout is to overcome the blind area problem of traditional radars when scanning backward, ensuring that the omnidirectional cooperative antenna has absolute unblocked line of sight in the horizontal plane of 360 degrees when the main antenna performs arbitrary azimuth scanning including zero to ninety-degree elevation angle change and zero to three hundred and sixty-degree horizontal rotation. In order to serve as a reliable radiation reference, the gain parameters of the omnidirectional cooperative antenna include the electromagnetic coupling effect of the installation support and are accurately calibrated in the microwave anechoic chamber, and the gain fluctuation is strictly controlled within the target of better than ±0.5 decibels in the horizontal omnidirectional range.

[0024] The cooperative sequential control unit is configured as a timing scheduler based on a field programmable gate array or a high-performance processor, strictly dividing the working time slice of the radar into an active detection time slot for weather echo acquisition and a passive cooperative time slot for reference signal interaction. The unit maintains a dynamically updated network topology mapping table inside, configured to combine satellite timing information to solve the relative geometric position relationship between the observation node and the cooperative node in real time, and to identify the precise calibration state and background maintenance state according to the beam pointing. The judgment standard of the precise calibration state is that the included angle error of the two main beam pointings is less than half of the three decibel beam width of the main beam, in which state the system is in the maximum gain link with the highest signal-to-noise ratio, suitable for full-parameter inversion; the judgment standard of the background maintenance state is that the observation radar main beam points to the cooperative radar, but the cooperative radar main beam does not point to the observation radar, at this time there is only one-way high-gain pointing, mainly used for maintaining the heart beat detection and rough calibration of the link.

[0025] The radio frequency transceiver subsystem performs physical isolation and overload protection of the transceiving link, which is the key to realize short baseline mutual shooting. The frequency offset injection transmitting unit is integrated with a direct digital frequency synthesizer and is configured to be triggered to work only in the passive cooperative time slot. It controls the cooperative radar transmitter to load a frequency offset at the baseband or intermediate frequency band, generating a continuous wave or long pulse reference signal with dynamic frequency offset, so that the signal is naturally separated from zero speed clutter and low speed meteorological echo in the frequency domain. The sector synchronization pre-attenuation receiving unit is located at the receiving front end of the observation radar. In order to prevent the strong direct wave emitted by the cooperative radar from burning out the receiver, the circuit design is extremely critical: a high-power passive limiter is connected in series at the front end of the low-noise amplifier input circuit, and the triggering action threshold of the limiter is set in the range of ten milliwatts decibel to twenty milliwatts decibel, as the last hard defense line to prevent physical damage to the device. A nanosecond response digital step attenuator is configured after the low-noise amplifier and before the analog-to-digital converter. The attenuator is configured to receive the trigger instruction of the cooperative sequential control unit, and instantaneously insert a preset attenuation value in the microsecond window of receiving the cooperative signal, and the preset attenuation value is set in the range of twenty decibels to forty decibels, so as to ensure that the amplitude of the strong signal falls back to the optimal linear dynamic range of the analog-to-digital converter.

[0026] The baseband signal processing subsystem internally runs enhanced logic, including an anti-aliasing spectrum separation module and a multi-source parameter decoupling inversion module, for realizing lossless separation of meteorological data and calibration data in the digital domain.

[0027] Referring to the drawings Figure 2 , Figure 2 is a schematic diagram of mode determination and antenna switching logic based on beam pointing state awareness according to an embodiment of the application.

[0028] The embodiment constructs an adaptive cooperative mechanism based on beam pointing state awareness, aiming to solve the link instability caused by non-synchronous scanning strategy in the process of dual-radar or multi-radar networking. The cooperative sequential control unit dynamically divides the node interaction logic in the network into a precise calibration mode and a background maintenance mode according to the real-time calculated radar beam pointing data.

[0029] In the precision calibration mode, the observation radar and the cooperative radar satisfy the main lobe to main lobe line-of-sight alignment condition. The specific physical determination logic of the alignment condition is that the included angle between the main beam axis of the observation radar and the main beam axis of the cooperative radar is less than half of the 3dB beam width of the main beam. In this state, the cooperative sequential control unit sends an instruction to the radio frequency link controller of the cooperative radar to switch the transmission channel to the high-gain main observation antenna. The reference signal transmitted by the main observation antenna has extremely high signal-to-noise ratio and excellent cross-polarization isolation, and the baseband signal processing subsystem can perform full-parameter decoupling operation based on the signal to accurately separate the system gain drift and the radome water film attenuation from the received power, and store the two parameters as the current time reference value in the non-volatile memory. For X-band or higher frequency radar, the signal attenuation caused by the radome water film has a strong correlation with the depolarization effect, and the high polarization purity of the main lobe signal is the key to the inversion of the water film attenuation.

[0030] In the background maintenance mode, the system detects that the main beam of the observation radar points to the cooperative radar, and the main beam of the cooperative radar points to other directions to perform the conventional weather scanning task. In order to avoid interrupting the normal detection business of the cooperative radar, the cooperative sequential control unit controls the radio frequency switch matrix to route the transmission signal of the cooperative radar to the omnidirectional cooperative antenna installed on the top. At this time, the system uses the omnidirectional radiation characteristics of the omnidirectional antenna in the horizontal dimension to maintain the link connection. Since the omnidirectional antenna is difficult to achieve high-index cross-polarization discrimination in physical structure, it cannot provide accurate depolarization information for water film inversion, so in this mode, the data processing logic performs parameter maintenance strategy. The system calls the effective radome water film attenuation obtained in the precision calibration mode in the last time from the memory as the estimated value in the current time slice, and only updates the system gain drift.

[0031] Referring to the drawings Figure 3 , Figure 3 is a schematic diagram of the link validity filtering and hierarchical processing strategy according to an embodiment of the application.

[0032] In order to ensure that the data input into the inversion equation has physical validity, the system is embedded with sidelobe suppression and multipath filtering logic. When the radar beam sweeps through the cooperative node, if it is not main lobe alignment, but sidelobe sweeps or multipath signals formed by ground reflection enter the receiver, the amplitude fluctuation will deviate from the true link loss. For this reason, the system sets a minimum effective received power threshold, and for signals with received power lower than the threshold, the logic unit determines that it is in a weak connection state. Signals in a weak connection state are only allowed to be used to extract timestamp information to correct network clock synchronization deviation, and are strictly prohibited from being used for gain inversion calculation.

[0033] Minimum effective received power threshold According to the link budget equation, the calculation formula is as follows: ; In the formula, represents the nominal transmitting power of the cooperative radar transmitter; represents the nominal gain of the omnidirectional cooperative antenna, and the parameter is obtained by microwave darkroom calibration; represents the first side lobe gain approximation of the observation radar antenna, and the value is usually taken as the main lobe gain minus 25 dB to 30 dB; represents the free space transmission loss corresponding to the geometric distance between the observation radar and the cooperative radar; represents the decision margin set by the system, and in the embodiment, the value of the decision margin ranges from 3 dB to 6 dB.

[0034] For the calculation of the free space transmission loss , the person skilled in the art can calculate it by using the classical Friis transmission equation or its logarithmic form according to the radar operating frequency and the straight line distance between two points, which belongs to the known technology in the art and will not be repeated here.

[0035] Through the above logical judgment, the system can automatically exclude non-ideal link data such as side lobe to side lobe and side lobe to omnidirectional, to ensure that the hardware parameter calibration is only performed in the interval where the link budget meets the physical expectation. The mechanism enables the radar network to complete high-precision self-checking and mutual checking by utilizing the intersection time of fragmented beams while performing regular weather scanning, without the need for manual intervention to maintain the consistency of the entire network.

[0036] Referring to the accompanying Figure 4 , Figure 4 is a sector synchronization radio frequency front-end control and adaptive frequency offset injection logical timing diagram according to an embodiment of the application.

[0037] The embodiment adopts a time division multiplexing cooperative working mechanism, and a cooperative sequential control unit accurately divides a scanning beam residence period of a radar into an active detection time slot and a passive cooperative time slot in the time dimension. In the active detection time slot, the observation radar transmitter is in a working state, transmits a wideband weather detection pulse to a target airspace, and the receiver synchronously collects weather echo signals. When entering the passive cooperative time slot, the observation radar transmitter is silent, the cooperative radar transmitter is in a working state, and the system executes an adaptive spectrum avoidance strategy, aiming to solve the aliasing problem of strong reference signals and weather echoes in the frequency domain.

[0038] The core logic of the adaptive spectrum avoidance strategy is to utilize the Doppler spectrum characteristics of weather echoes to dynamically plan the transmission frequency of the reference signal. The baseband signal processing subsystem reads the weather echo data of the previous scan period or adjacent azimuth, extracts the center velocity and spectral width parameters of the Doppler velocity spectrum, and identifies the energy distribution interval of the weather echo in the frequency domain. Based on this distribution interval, the system calculates the optimal injection frequency offset, controls the direct digital frequency synthesizer of the collaborative radar to generate a carrier signal offset from the center frequency, and moves the narrowband reference signal to the low-power baseband signal processing subsystem of the weather spectrum.

[0039] In order to ensure measurement accuracy and prevent signal distortion, the calculation of the optimal injection frequency offset is subject to the dual constraints of hardware bandwidth limitation and spectral separation requirement.

[0040] Firstly, the frequency offset must be limited within the effective linear bandwidth of the receiver analog front-end. If the frequency offset is too large, the signal will fall into the transition band or stop band of the filter, causing unknown attenuation in amplitude measurement. Therefore, the optimal injection frequency offset needs to satisfy the following hardware bandwidth constraint condition: ; In the formula, represents the calculated optimal injection frequency offset; represents the flat band bandwidth of the receiver front-end low-noise amplifier and intermediate frequency filter, within which the amplitude-frequency response fluctuation is less than the system set tolerance value.

[0041] Secondly, in order to achieve physical separation of the reference signal and weather echo in the frequency domain, the Doppler velocity corresponding to the frequency offset must avoid the main energy area of the weather echo. The equivalent Doppler velocity corresponding to the optimal injection frequency offset should satisfy the following spectral isolation constraint condition: ; In the formula, represents the equivalent velocity value of the optimal injection frequency offset converted by the Doppler formula; represents the center velocity of the detected weather echo; represents the Doppler spectral width (usually the second moment of the spectrum) of the weather echo; represents the system preset protection interval velocity. In this embodiment, the value range of the protection interval velocity is set to 2 meters per second to 5 meters per second. The setting of the protection interval can effectively prevent signal crosstalk caused by weather spectrum tailing or spectral leakage effect, ensuring that the subsequent signal processing can be separated without loss through simple frequency domain filtering.

[0042] In cooperation with the frequency domain control, the system performs strict synchronization pre-attenuation control in time domain to protect the receiver hardware. Since the reference signal transmitted by the cooperative radar is line-of-sight transmission, the power level of the reference signal reaching the observation radar receiving port is usually much higher than the meteorological echo. In order to prevent the low noise amplifier from being saturated or the back-end analog-to-digital converter from being clipped, the cooperative sequential control unit sends a control instruction to the digital step attenuator in advance before the start of the passive cooperative time slot.

[0043] The intervention timing of the digital step attenuator is determined by a protection time parameter. The system sets the protection time, which covers the switching establishment time of the radio frequency switch and the transmission time delay jitter of the electromagnetic wave. Within the protection time before the start of the passive cooperative time slot, the digital step attenuator completes the switching from the low attenuation state to the high attenuation state, and the increase amount of the attenuation value is usually set to 20dB to 40dB. The value range of the protection time is set to 5 microseconds to 20 microseconds. This timing design ensures that when the strong reference signal reaches the receiver antenna port, the receiving link has already entered the high attenuation state of the baseband signal processing subsystem, thereby eliminating the impact of the transient overshoot caused by the gain switching lag on the circuit, while ensuring the integrity of the reference signal sampling. For the specific circuit topology and driving mode of the digital step attenuator, those skilled in the art can select a PIN diode network or a FET switch network according to the specific radio frequency band and power requirement, which belongs to the known technology in the art and will not be described here.

[0044] Referring to the accompanying drawings Figure 5 , Figure 5 is a flow chart of Doppler spectral domain signal separation and purification processing according to an embodiment of the present application.

[0045] The embodiment performs fine separation of meteorological echoes and reference signals in the digital signal processing domain, aiming to solve the feature extraction and repair problem of time domain overlapping signals in the frequency domain. The baseband signal processing subsystem receives the time domain digital sequence from the analog-to-digital converter, maps it to the Doppler frequency domain through windowed fast Fourier transform, and thus obtains high-resolution power spectral density data. In this process, in order to suppress the masking effect of the spectral sidelobe leakage of the strong reference signal on the weak meteorological echo, the system uses a window function with low sidelobe characteristics, such as the Blackman window or the Kaiser window, to weight and truncate the time domain data. For the specific implementation algorithm of the fast Fourier transform and the window function, those skilled in the art can design it according to the conventional process of signal processing, which belongs to the known technology in the art and will not be described here.

[0046] After obtaining the power spectrum of the mixed signal, the system core logic enters the signal purity evaluation stage, and performs overlap detection and calibration fuse mechanism. This mechanism is used to identify whether the reference signal is severely contaminated by strong weather echoes (such as wide spectrum wide tornado or high turbulence echoes). The system calculates the spectral domain overlap coefficient, which quantitatively represents the degree of intrusion of weather background noise or weather signal energy in the frequency band occupied by the reference signal.

[0047] Spectral domain overlap coefficient The calculation formula is as follows: ; In the formula, represents the spectral domain overlap coefficient, which is a dimensionless ratio; represents the center frequency of the reference signal; represents the integral bandwidth radius, which defines the main energy concentration area of the reference signal. In this embodiment, the value range of this radius is set to 3 to 5 times the frequency resolution of the fast Fourier transform; represents the weather spectrum bottom background function estimated by noise level or adjacent spectral line fitting; represents the peak power density measured at the reference signal frequency point.

[0048] Based on the calculated spectral domain overlap coefficient, the system performs bifurcation processing logic. The system presets an overlap threshold, and the value range of the threshold is set to 0.1 to 0.3.

[0049] When the calculated overlap coefficient is greater than or equal to the overlap threshold, it indicates that a large amount of weather echo energy has been mixed into the frequency band of the reference signal, and at this time it is impossible to accurately strip out the true power of the reference signal from the mixed power. The system immediately triggers the calibration fuse logic to forcibly terminate the current gain inversion calculation process. In order to prevent false calibration parameters from polluting the historical record, the system maintains the system gain drift as the stable value of the last time, and sets the calibration pause flag. At the same time, the system directly outputs the original weather spectrum data without filtering processing, to ensure that the weather observation integrity under strong wind shear or extreme weather is better than the equipment calibration requirement.

[0050] When the calculated overlap coefficient is less than the overlap threshold, it indicates that the reference signal and the weather echo have good separation degree in the frequency domain. The system determines that the current data is valid, and performs signal separation and spectrum repair operations. In the separation stage, the system uses a frequency domain notch filter to extract the total energy within the bandwidth of the reference signal for subsequent link calculation. In the spectrum repair stage, for the spectral hole of the baseband signal processing subsystem left by filtering out the reference signal, the system uses the weather spectrum data on both sides of the hole to generate filling data by linear interpolation or Gaussian model fitting method, and repairs the weather spectrum to a continuous smooth curve. The repaired weather spectrum is sent to the subsequent weather parameter estimation module for calculating standard weather products such as reflectivity factor, mean radial velocity and spectrum width. This processing flow realizes the acquisition of high-precision equipment calibration data and the continuity and availability of weather observation data in the same time slice and the same receiving channel.

[0051] Referring to the drawings Figure 6 , Figure 6 is a multi-dimensional physical parameter inversion and gain drift decoupling calculation flowchart according to an embodiment of the application.

[0052] The embodiment discloses a parameter decoupling architecture based on physical model constraints, aiming to separate three types of attenuation components with different properties, i.e. atmospheric transmission loss, radome water film loss and hardware gain drift, from a single received power observation. The baseband signal processing subsystem uses the inverse solving logic of the radar equation to construct a closed-loop link budget balancing system combined with multi-source perception data.

[0053] The system first reconstructs the medium attenuation field along the propagation path based on weather observation data. Using the differential propagation phase shift slope distribution data along the baseline direction obtained in the immediately adjacent active detection time slot, the system performs path integral operation to quantify the absorption and scattering effects of the precipitation medium on electromagnetic waves. Those skilled in the art use conventional weather radar attenuation correction algorithms to establish a power function relationship between the differential propagation phase shift slope and the specific attenuation rate according to the attenuation law of X-band electromagnetic waves in the precipitation medium, integrate the specific attenuation rate on the baseline path, and superimpose the gas absorption attenuation under the standard atmospheric model, thereby obtaining the one-way total path attenuation of the baseline path. In this calculation process, in order to adapt to the physical properties of X-band radar under different precipitation types, the value range of the medium attenuation coefficient used to convert the differential propagation phase shift to the attenuation rate is set to 0.2 to 0.3, and the value range of the exponential coefficient is set to 1.0 to 1.1.

[0054] While acquiring the environmental path attenuation, the system calculates the theoretical ideal received power, which represents the expected signal level in an ideal vacuum environment with normal transmitter output, no loss of antenna gain, and no atmospheric attenuation and water film loss. The calculation process dynamically adapts to the current cooperative mode and selects the corresponding gain parameter according to the pointing state of the antenna beam.

[0055] Ideal received power The calculation formula is as follows: ; In the formula, represents the ideal received power; represents the nominal transmission power calibrated at the factory or measured by a power meter; represents the transmission antenna gain, which has a state correlation. When the system is determined to be in the precise calibration mode, the calibrated gain of the subjective observation antenna is taken. When the system is determined to be in the background maintenance mode, the calibrated gain of the omnidirectional cooperative antenna is taken; represents the calibrated gain of the observation radar receiving antenna; represents the free space propagation loss calculated according to the baseline distance ; represents the in-band unevenness correction loss of the receiver analog filter at the best injection frequency offset point. The loss value is obtained by consulting the factory test data table of the receiver amplitude-frequency response.

[0056] For the high dynamic interference term of the water film on the radome surface, the system performs independent inversion based on polarization characteristics. When in the precise calibration mode, the system measures the linear depolarization ratio increment of the reference signal by using the high cross-polarization discrimination rate characteristics of the main lobe to main lobe link. Since the uneven distribution of the water film can significantly enhance the depolarization effect of electromagnetic waves, the system inputs the measured linear depolarization ratio increment into a pre-constructed empirical polynomial model to directly map the radome water film attenuation. The empirical model is calibrated through artificial rain test in the ground test field. When in the background maintenance mode, since the omnidirectional antenna cannot provide high-precision polarization purity, the system performs parameter maintenance logic to directly read the water film attenuation value recorded at the last effective precise calibration time in the storage.

[0057] After completing the independent quantification of each component, the system constructs a link power balance equation and calculates the unique unknown quantity, i.e., the system gain drift, through residual calculation. The drift represents the comprehensive gain deviation of all active devices in the radar transmission and receiving links relative to the nominal state.

[0058] System gain drift The calculation formula is as follows: ; In the formula, representing the calculated system gain drift; representing the equivalent power level at the physical input port of the analog-to-digital converter after the digital signal output by the analog-to-digital converter is subjected to fast Fourier transform and frequency domain extraction processing; representing the hardware setting attenuation value of the digital step attenuator at the current receiving time, used to restore the signal to the entrance of the radio frequency front end; representing the one-way total path attenuation calculated above; representing the radome water film attenuation obtained by the inversion above. Through this calculation logic, the system successfully decouples the signal fluctuations caused by external environment (path rain attenuation, water film) and the signal fluctuations caused by internal equipment (aging, temperature drift) at the physical layer, ensuring that the calibration result truly reflects the hardware state.

[0059] Referring to the accompanying drawings Figure 7 , Figure 7 is a networked data fusion processing flowchart based on dynamic confidence according to an embodiment of the present application.

[0060] This embodiment constructs a heterogeneous data quality evaluation and weighted fusion mechanism, aiming to solve the observation consistency problem caused by differences in hardware state, calibration timeliness and radome environment of each node in radar networking. The data processing center not only performs traditional interpolation and mapping according to geometric distance, but also introduces a dynamic variable, calibration confidence, to score the observation data of each radar node in the network in real time, and decides the weight contribution of the node in the final weather product according to the score result.

[0061] The data processing logic first performs hardware bias correction of single-station data. The system calls the system gain drift calculated by the previous sequence, and performs linear compensation on the raw reflectivity factor data output by each radar node. This process directly eliminates the system error introduced by transmitter power fluctuations, receiver gain drift and other hardware factors, and unifies the observation data of each node to the nominal reference surface.

[0062] After completing the basic correction, the system core logic enters the confidence weight generation stage. In order to quantify the reliability of the current calibration parameters, the system constructs a multi-dimensional evaluation model including drift level, time decay and environmental interference, and calculates the calibration confidence weight. The weight value reflects the degree of trust of the system in the accuracy of the current radar node data. If the gain drift of a node is too large, it implies that its hardware may be in the nonlinear region or on the edge of failure; if the time interval since the last precise calibration is too long, it means that the calibration parameters may be outdated; if the radar site is experiencing heavy precipitation, it means that the radome water film state is in a state of rapid change, and the historically inverted water film parameters are no longer applicable.

[0063] Based on the above physical logic, the calibration confidence weight The calculation formula of the calibration confidence weight is as follows: ; In the formula, represents the calibration confidence weight; represents the calculated system gain drift; represents the allowable error tolerance for defining the tolerance of the system to the gain drift, which is set to be 0.5-1.5 decibels, and the weight of the node will rapidly decrease if the drift exceeds the tolerance; represents the time elapsed in seconds since the last successful execution of the precision calibration mode; represents the time decay constant for characterizing the failure risk of the hardware state over time, which is set to be 0.001-0.01; represents the water film timeliness factor.

[0064] For the water film timeliness factor , the system introduces a local weather feedback mechanism. The system reads the minimum elevation reflectivity data of the radar site in real time to determine whether there is an active water accumulation process on the surface of the antenna cover.

[0065] The calculation logic of the water film timeliness factor is as follows: ; In the formula, represents the reflectivity factor observation value of the radar site locally; represents the reflectivity threshold for determining local precipitation, which is set to be 20 dBZ in the present embodiment; represents the strong attenuation factor, which is set to be 0.01-0.05. This logic shows that if no precipitation occurs at the radar site, the antenna cover is in a dry or stable state, and the calibration parameter is highly reliable; if precipitation occurs, the water film thickness of the antenna cover changes rapidly with the rain intensity, causing the previous calibration parameter to rapidly fail, thereby forcing the system to rapidly reduce the weight of the node.

[0066] Finally, the system performs a gridding fusion operation. For multiple radar echo data covering the same geographical grid point, the system combines the above calibration confidence weight and the traditional geometric distance weight for weighted averaging to generate the final grid point data. For the calculation of the geometric distance weight, those skilled in the art can use the distance inverse function in the Kressman interpolation or Barnes interpolation algorithm, which belongs to the known technology in the art, and will not be described here.

[0067] The calculation formula of the value of the final grid point is as follows: ; In the formula, ​a final grid point reflectivity factor after fusion; a number of valid radar nodes covering the grid point; a drift-corrected observation data of the a calibration confidence weight of the a geometric distance weight of the

[0068] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.​​​

Claims

1. A multi-mode cooperative compact X-band weather radar system, characterized in that, It includes multiple radar nodes distributed in different geographical locations, and the radar nodes divide their working cycle into active detection time slots and passive coordination time slots in terms of timing; Each of the radar nodes includes an antenna subsystem, a radio frequency transceiver subsystem, a cooperative sequential control unit, and a baseband signal processing subsystem; The antenna subsystem includes a main observation antenna and an omnidirectional cooperative antenna; The radio frequency transceiver subsystem includes a frequency offset injection transmitting unit and a sector synchronization pre-attenuation receiving unit; The cooperative sequential control unit is configured to calculate the relative geometric position relationship between the observation node and the cooperative node in real time based on the beam pointing, and to control the system to switch between precision calibration mode and background maintenance mode within the passive cooperative time slot. The baseband signal processing subsystem is configured to use the inverse solution logic of the radar equation to decouple and separate the system gain drift and the radome water film attenuation from the received reference signal, and to use the system gain drift to calibrate the meteorological echo data acquired by the active detection time slot.

2. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The antenna subsystem adopts a zero-obstruction composite structure layout; the omnidirectional cooperative antenna is installed on the central axis of the top of the radome or at the top of an independent mast above the rotation envelope of the main observation antenna, so that when the main observation antenna performs elevation scanning from 0 to 90 degrees and azimuth scanning from 0 to 360 degrees, the omnidirectional cooperative antenna has an unobstructed line of sight on the horizontal plane. The precision calibration mode is defined as the state when the angle error between the main beam pointing of the observation node and the cooperative node is less than half of the 3dB beamwidth of the main beam. At this time, the system is configured to call the main observation antenna to transmit a reference signal. The background maintenance mode is defined as the state when the main beam of the observation node points to the cooperating node but the cooperating node is not aligned with the observation node. At this time, the system is configured to switch to the omnidirectional cooperating antenna to transmit a reference signal.

3. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The sector synchronization pre-attenuation receiving unit in the radio frequency transceiver subsystem is equipped with a graded protection circuit. The graded protection circuit has a high-power passive limiter connected in series in the front-end input circuit of the low-noise amplifier to clamp the direct-beam intensity signal. The graded protection circuit is configured with a digital step attenuator after the low-noise amplifier and before the analog-to-digital converter. The cooperative sequential control unit is configured to trigger the digital step attenuator to switch to a high attenuation state during the protection period before entering the passive cooperative time slot, so that the amplitude of the received reference signal falls back to the linear dynamic range of the analog-to-digital converter.

4. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The frequency offset injection transmitter unit is configured to execute an adaptive spectrum avoidance strategy; The baseband signal processing subsystem is configured to analyze the meteorological echo Doppler spectrum of the previous scan cycle, extract the spectral width and center velocity, and calculate the optimal injection frequency offset. The optimal injection frequency offset is constrained to simultaneously satisfy the flat bandwidth limit of the receiver front-end filter and the spectral isolation requirement relative to the center velocity of the weather echo. The frequency offset injection transmission unit controls the transmitter to generate a reference signal with dynamic frequency offset in the passive coordination time slot according to the optimal injection frequency offset amount, so that the reference signal avoids the main energy region of the meteorological echo in the frequency domain.

5. The multi-mode cooperative compact X-band weather radar system according to claim 4, characterized in that, The baseband signal processing subsystem includes an anti-aliasing spectral separation module, which is configured to calculate the spectral overlap coefficient of the reference signal. The spectral overlap coefficient represents the ratio of the meteorological spectral background energy to the peak power of the reference signal within the integral bandwidth of the reference signal. When the spectral overlap coefficient is less than a preset overlap threshold, the module is configured to extract the reference signal using frequency domain notch filtering and perform interpolation repair on the filtered frequency band to restore the meteorological spectrum. When the spectral overlap coefficient is greater than or equal to the preset overlap threshold, the module is configured to trigger the calibration circuit breaker mechanism, suspend the gain inversion calculation, maintain the system gain drift amount of the previous moment, and output the unpatched original meteorological spectrum.

6. The multi-mode cooperative compact X-band weather radar system according to claim 2, characterized in that, The baseband signal processing subsystem is configured to execute a pattern-aware parameter update strategy: In the precise calibration mode, the system simultaneously uses the received signal to invert and update the system gain drift and the attenuation of the radome water film, and stores both as reference values. In the background maintenance mode, the system updates the system gain drift using the received signal, and executes a parameter preservation strategy for the attenuation of the radome water film, directly calling the valid value obtained in the most recent precise calibration mode in the memory to participate in the calculation.

7. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The baseband signal processing subsystem is configured to calculate the system gain drift in the following manner: Acquire differential propagation phase shift slope distribution data along the baseline direction in the active detection time slot, and calculate the one-way total path attenuation of the baseline path through path integration; Calculate the ideal received power based on the transmitter's nominal power, the transmit antenna gain corresponding to the current operating mode, the observation antenna gain, the baseline free space loss, and the receiver filter correction loss. The actual received power is obtained by summing the equivalent power level output by the analog-to-digital converter and the compensation value of the digital step attenuator. The residual obtained by subtracting the ideal received power from the actual received power, and compensating for the total path attenuation and the attenuation of the radome water film, is the system gain drift.

8. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The baseband signal processing subsystem is configured to invert the attenuation of the radome water film by measuring the linear depolarization ratio increment of the reference signal; The inversion process is based on a pre-calibrated empirical polynomial model, which establishes a mapping relationship between the linear depolarization ratio increment and the signal attenuation value caused by the water film on the radome. The baseband signal processing subsystem is configured to calculate the linear depolarization ratio increment only through the high polarization purity signal received by the main observation antenna in the precision calibration mode.

9. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The system also includes a data fusion processing module for generating calibration confidence weights and performing gridded fusion; The calibration confidence weight is configured to be negatively exponentially correlated with the square of the system gain drift, negatively exponentially correlated with the time interval since the last precision calibration, and controlled by the water film timeliness factor. The water film timeliness factor is defined as follows: when the local reflectivity factor of the radar site is lower than the precipitation determination threshold, the value is 1; when the local reflectivity factor of the radar site is higher than or equal to the precipitation determination threshold, it exhibits a negative exponential relationship of rapid decay with time interval. The gridded fusion is configured to perform a weighted average of the observation data from multiple radar nodes using the calibration confidence weight and geometric distance weight.

10. The multi-mode cooperative compact X-band weather radar system according to claim 1, characterized in that, The baseband signal processing subsystem has embedded sidelobe suppression logic, which is configured as follows: Set a minimum effective received power threshold, which is equal to the sum of the nominal transmit power of the cooperating node, the gain of the omnidirectional cooperative antenna, and the gain of the first sidelobe of the observation node, minus the free space transmission loss and the system decision margin. When the power of the received reference signal is lower than the minimum effective received power threshold, the current link is determined to be in a state of sidelobe coupling or multipath interference, and the use of this signal to perform the inversion calculation of the system gain drift is prohibited.

Images