Microwave-laser composite detection method and system based on heterogeneous data spatio-temporal matching

Abstract

This invention discloses a microwave-laser composite detection method and system based on spatiotemporal matching of heterogeneous data, specifically relating to the field of microwave and laser atmospheric remote sensing radar data fusion technology. It addresses the problem of insufficient physical consistency in data fusion caused by existing composite detection methods neglecting the intrinsic differences in the observation processes of the two devices. First, observation data from both devices are acquired, and their corresponding spatiotemporal representation ranges are determined. Then, the overlapping region of the spatiotemporal representation ranges is calculated, and the potential interference of the laser radar scanning process on the microwave radiometer observations is analyzed within this region to obtain interference assessment results. Next, the data fusion reliability is evaluated by combining the spatiotemporal coverage relationship and the interference assessment results. Finally, based on this reliability, strategies such as joint inversion, weighted averaging, or independent output are adaptively selected to fuse the two types of observation data, thereby achieving physically consistent and reliable composite detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of microwave and laser atmospheric remote sensing radar detection data fusion, more particularly, the present application relates to a microwave and laser composite detection method and system based on heterogeneous data space-time matching. BACKGROUND

[0002] Using a microwave radiometer to obtain atmospheric temperature and humidity profiles and using a laser radar to obtain wind field profiles are important means of atmospheric remote sensing in the near space. Integrating the two to obtain a synergistically enhanced three-dimensional atmospheric parameter field through data fusion has become an important development direction for improving detection capability. In the prior art, to achieve such composite detection, two devices are usually controlled independently for observation, and the data obtained by each device is subsequently time-aligned and spatially matched to obtain temperature, humidity and wind parameters at the same time and at the same place, which are then used for joint analysis and application.

[0003] However, due to the inherent differences in the core working principles and observation modes of the microwave radiometer and the laser wind measurement radar, the existing data fusion method based on simple time-space alignment has limitations. The microwave radiometer continuously receives the integrated signals of atmospheric radiation within its field of view through a wide-beam antenna, and the single observation result is the combined effect of spatial continuous integration and time accumulation. The laser wind measurement radar usually uses mechanical scanning to discretely sample at different pointing directions through a narrow-beam, and the single wind field profile is the combined result of sequentially measuring multiple discrete spatial points within a period of time. These two completely different observation processes essentially impose different time-space filtering on the real and continuously changing atmospheric field. The existing fusion method only time-stamps and spatially matches the filtered data, but ignores the time-space representation differences introduced by the observation operators, resulting in the fused data not being strictly derived from the same atmospheric state in physics, thereby affecting the accuracy and physical consistency of the final fused data product. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides a microwave and laser composite detection method and system based on heterogeneous data space-time matching to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solution:

[0006] The microwave and laser composite detection method based on heterogeneous data space-time matching comprises the following steps:

[0007] S1: obtaining observation data of a microwave radiometer and observation data of a laser wind measurement radar, the observation data containing corresponding time information;

[0008] S2: Based on the beam parameters of the microwave radiometer and its observation process within the time information, determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data;

[0009] S3: Determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction;

[0010] S4: Determine the overlapping spatiotemporal region based on the first and second spatiotemporal characterization ranges, analyze the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period within the overlapping spatiotemporal region, and obtain the interference assessment results.

[0011] S5: Based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results, evaluate the reliability of data fusion between microwave radiometer observation data and laser wind radar observation data;

[0012] S6: Based on the reliability of data fusion, the microwave radiometer observation data and the laser wind radar observation data are fused.

[0013] Furthermore, S1 includes:

[0014] A set of radiation brightness temperature data with continuous time stamps was collected from a microwave radiometer;

[0015] A set of radial wind speed data with corresponding scan time markers was collected from the laser wind radar;

[0016] Extract unified time reference information from the shared clock source or timing device of the microwave radiometer and the laser wind radar;

[0017] Based on a unified time reference, the time information of microwave radiometer observation data is extracted from radiation brightness temperature data, and the time information of laser wind radar observation data is extracted from radial wind speed data.

[0018] Furthermore, S2 includes:

[0019] Based on the beamwidth and beam pointing of the microwave radiometer, its instantaneous observation area in the spatial dimension is determined;

[0020] Based on the time information and data integration period of microwave radiometer observation data, the effective observation period in the time dimension is determined.

[0021] The instantaneous observation area is spatiotemporally extended along the effective observation period to form a joint representation of a spatial cone with the beam pointing as the central axis and a time interval, and this joint representation is defined as the first spatiotemporal representation range.

[0022] Furthermore, S3 includes:

[0023] Acquire the multiple spatial directions visited sequentially by the laser wind radar during the scanning process and the corresponding dwell time for each spatial direction;

[0024] For each spatial orientation, the instantaneous observation area corresponding to the spatial orientation is determined based on the beamwidth and spatial orientation of the laser wind measuring radar;

[0025] For each spatial direction, the effective observation period corresponding to the spatial direction is determined based on the dwell time corresponding to the spatial direction and the time information of the laser wind radar observation data;

[0026] Each spatial pointer is spatiotemporally extended along the effective observation period corresponding to the spatial pointer to obtain the spatiotemporal unit corresponding to the spatial pointer.

[0027] All spatial points are merged into corresponding spatiotemporal units in the spatiotemporal domain to form a second spatiotemporal representation range.

[0028] Furthermore, S4 includes:

[0029] Calculate the intersection of the first spatiotemporal representation range and the second spatiotemporal representation range in the spatiotemporal domain, and define the intersection as the overlapping spatiotemporal region;

[0030] Within the overlapping spatiotemporal region, identify the time periods during which the laser wind-measuring radar is in scanning motion and the time periods during which it is in stationary pointing state;

[0031] During the period when the laser wind radar is in scanning motion, the short-term fluctuation characteristics of the microwave radiometer observation data are extracted.

[0032] During the period when the laser wind measuring radar is in a stationary pointing state, the short-term fluctuation characteristics of the microwave radiometer observation data are extracted.

[0033] By comparing the short-term fluctuation characteristics under scanning motion and under stationary pointing conditions, interference assessment results are generated based on the comparison differences.

[0034] Furthermore, the scanning process parameters include the scanning angular velocity of the laser wind radar, and the short-term fluctuation characteristics are the standard deviation of the microwave radiometer observation data within the corresponding time period.

[0035] Furthermore, S5 includes:

[0036] The spatiotemporal coverage matching degree is determined based on the proportion of the overlapping spatiotemporal regions of the first and second spatiotemporal representation ranges within their respective representation ranges.

[0037] Based on the interference intensity characterized by the interference assessment results, the interference impact factor is determined;

[0038] Using the spatiotemporal coverage matching degree as the basic credibility, the basic credibility degree is adjusted down according to the interference impact factor to obtain the data fusion credibility degree.

[0039] Furthermore, S6 includes:

[0040] When the data fusion credibility is higher than the preset first credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are jointly inverted and fused.

[0041] When the data fusion credibility is lower than or equal to the preset first credibility threshold and higher than the preset second credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are subjected to weighted average fusion processing, where the weighting coefficient is positively correlated with the data fusion credibility.

[0042] When the data fusion credibility is lower than or equal to the preset second credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are retained as independent data products.

[0043] Furthermore, the joint inversion and fusion processing includes inputting microwave radiometer observation data and laser wind radar observation data as common constraints into the atmospheric parameter inversion algorithm. The weighting coefficient in the weighted average fusion processing is equal to the product of the data fusion confidence and the fixed scaling factor.

[0044] On the other hand, the present invention provides a microwave-laser composite detection system based on spatiotemporal matching of heterogeneous data, comprising the following modules:

[0045] The data acquisition module is used to acquire observation data from the microwave radiometer and the laser wind radar. The observation data includes corresponding time information.

[0046] The first characterization module is used to determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data based on the beam parameters of the microwave radiometer and its observation process within the time information.

[0047] The second characterization module is used to determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction.

[0048] The interference analysis module is used to determine the overlapping spatiotemporal region based on the first spatiotemporal characterization range and the second spatiotemporal characterization range. Within the overlapping spatiotemporal region, it analyzes the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period, and obtains the interference assessment results.

[0049] The credibility assessment module is used to assess the credibility of data fusion between microwave radiometer observation data and laser wind radar observation data based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results.

[0050] The fusion processing module is used to fuse microwave radiometer observation data and laser wind radar observation data according to the data fusion reliability.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. By establishing and analyzing the spatiotemporal characterization models of the observation processes of microwave radiometers and lidar wind radars, the mismatch problem of observation operators caused by the inherent differences in their working principles and observation modes is quantified and addressed. Existing technologies only perform apparent temporal and spatial coordinate alignment, while this scheme further constructs a first spatiotemporal characterization range reflecting the continuous integration characteristics of microwave radiometers and a second spatiotemporal characterization range reflecting the discrete scanning characteristics of lidar. By calculating the overlapping spatiotemporal regions of the two, a refined matching and comparison of the spatiotemporal footprints of heterogeneous data is achieved before data fusion. This process transforms the abstract spatiotemporal filtering differences into specific, calculable geometric and temporal relationships, ensuring that the data on which subsequent fusion processing is based has higher consistency in its spatiotemporal origin, thereby improving the physical authenticity and accuracy of composite detection data products from the source.

[0053] 2. By comparing the fluctuation characteristics of microwave data under scanning motion and stationary states to assess interference intensity, the decision-making basis for data fusion is expanded from simple spatiotemporal coverage to comprehensive reliability, including the reliability of equipment collaboration. Based on this reliability grading, different fusion strategies are adopted. When the reliability is high, deep joint inversion is performed to fully explore data complementarity; when the reliability is insufficient, a conservative weighted or independent output mode is adopted to control risks. This adaptive fusion mechanism based on quantitative evaluation significantly enhances the robustness of the composite detection system and the reliability of its output products under various actual observation environments. Attached Figure Description

[0054] Figure 1 This is a flowchart of the microwave-laser composite detection method based on heterogeneous data spatiotemporal matching of the present invention;

[0055] Figure 2 This is a schematic diagram of the structure of the microwave-laser composite detection system based on heterogeneous data spatiotemporal matching of the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0057] Example 1: Figure 1 The present invention provides a microwave-laser composite detection method based on spatiotemporal matching of heterogeneous data, which includes the following steps:

[0058] S1: Acquire observation data from microwave radiometer and laser wind radar, including corresponding time information;

[0059] S2: Based on the beam parameters of the microwave radiometer and its observation process within the time information, determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data;

[0060] S3: Determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction;

[0061] S4: Determine the overlapping spatiotemporal region based on the first and second spatiotemporal characterization ranges, analyze the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period within the overlapping spatiotemporal region, and obtain the interference assessment results.

[0062] S5: Based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results, evaluate the reliability of data fusion between microwave radiometer observation data and laser wind radar observation data;

[0063] S6: Based on the reliability of data fusion, the microwave radiometer observation data and the laser wind radar observation data are fused.

[0064] S1: Acquire observation data from the microwave radiometer and the laser wind radar. The observation data includes corresponding time information. The specific implementation is as follows:

[0065] During this step, the microwave radiometer continuously receives microwave-frequency electromagnetic radiation signals from the atmosphere. These signals are amplified, frequency-converted, and detected by the internal receiver components of the radiometer, converting them into a set of electrical quantity values. These electrical quantity values ​​characterize the atmospheric radiation brightness temperature at different detection frequencies and are referred to as radiation brightness temperature data. The microwave radiometer's acquisition system samples at a fixed period, such as 50 milliseconds or 100 milliseconds. Each sampling generates a corresponding timestamp for each sample, using Coordinated Universal Time (UTC) format and including year, month, day, hour, minute, second, and millisecond information. This results in a set of radiation brightness temperature data with continuous timestamps. This continuously timestamped radiation brightness temperature data is transmitted to the data processing unit via the radiometer's data output interface, which may include an Ethernet interface or a serial communication interface.

[0066] The laser wind radar operates according to a preset scanning mode, which includes a spatial pointing sequence and a corresponding dwell time for each spatial pointing. During the dwell time corresponding to a spatial pointing, the laser emitting unit of the laser wind radar emits a sequence of pulsed lasers in that spatial pointing direction. The optical receiving unit of the laser wind radar receives the backscattered light signal from atmospheric aerosols. The signal processing unit of the laser wind radar analyzes the Doppler frequency shift in the backscattered light signal using coherent detection technology, calculates the radial wind speed in that spatial pointing direction, and generates a radial wind speed data point. The control system of the laser wind radar records the precise moment when each new spatial pointing scan is executed. The control system uses this precise moment as a scan time marker and associates the scan time marker with all radial wind speed data obtained during the dwell time of that spatial pointing, forming a set of radial wind speed data with corresponding scan time markers. The radial wind speed data with corresponding scan time markers is transmitted to the data processing unit through the data output interface of the laser wind radar.

[0067] The microwave radiometer and the laser wind radar are connected to a shared clock source, which provides a unified time reference. This shared clock source is a high-precision external clock source, such as a satellite navigation timing module. The satellite navigation timing module receives satellite signals and outputs a standard time synchronization signal, which includes pulses per second and a serial time code. Both the microwave radiometer and the laser wind radar have clock synchronization interfaces for receiving this signal. The internal clock circuits of both the microwave radiometer and the laser wind radar are calibrated according to the standard time synchronization signal. The internal clocks of both systems are synchronized with the shared clock source, and both extract unified time reference information from it. This unified time reference information is represented by Coordinated Universal Time (UTC).

[0068] Based on a unified time reference, time information is extracted from radiation brightness temperature data with continuous time stamps and radial wind speed data with corresponding scan time stamps. For radiation brightness temperature data with continuous time stamps, the continuous time stamps are initially generated based on the local clock of the microwave radiometer. The data processing unit reads the continuous time stamps and calls the clock synchronization parameters stored internally by the microwave radiometer. The clock synchronization parameters include a fixed offset between the local clock and Coordinated Universal Time (UTC). The data processing unit uses the clock synchronization parameters to convert the local time stamp attached to each radiation brightness temperature data point into UTC time based on a unified time reference. The converted UTC time is recorded as the time information of the microwave radiometer observation data. For radial wind speed data with corresponding scan time stamps, the scan time stamps are generated based on the local clock of the laser wind radar. The data processing unit reads the scan time stamps and calls the clock synchronization parameters stored internally by the laser wind radar. The data processing unit uses the clock synchronization parameters to convert each scan time stamp into UTC time based on the same unified time reference. The converted UTC time is recorded as the time information of the laser wind radar observation data. After this step, the radiation brightness temperature data from the microwave radiometer and the radial wind speed data from the laser wind radar both have time information based on the exact same time reference.

[0069] S2: Based on the beam parameters of the microwave radiometer and its observation process within the time information, determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data. Specifically, this is implemented as follows:

[0070] In determining the first spatiotemporal characterization range corresponding to the microwave radiometer observation data, the beam parameters of the microwave radiometer first need to be obtained. These parameters include beamwidth and beam pointing. Beamwidth is an angular value representing the angular span of the microwave radiometer antenna's main lobe between two specific directions. For example, a beamwidth of 3 degrees or 5 degrees can be obtained directly from the microwave radiometer's equipment technical specifications document, provided by the equipment manufacturer and stored in the equipment's configuration file. Beam pointing represents the spatial orientation of the microwave radiometer antenna's main lobe axis. It is defined by both azimuth and elevation angles, with geographic north and the horizontal plane as reference points. The beam pointing is obtained by reading the real-time feedback signal from the microwave radiometer's servo mechanism. This real-time feedback signal includes the real-time angle encoder readings from the antenna turntable, which are transmitted to the data processing unit via a communication interface.

[0071] Based on the obtained beamwidth and beam pointing, the instantaneous observation area of ​​the microwave radiometer in spatial dimensions is determined. The instantaneous observation area is a three-dimensional spatial range, defined as the intersection region of a cone with the beam pointing as its central axis and the beamwidth as its cone angle at a specific distance with the atmospheric space. The specific process for determining the instantaneous observation area is to use the beam pointing as the central axis and the beamwidth as the cone angle; the spatial volume covered by this cone is the theoretical instantaneous observation area. Considering the propagation characteristics of microwave radiation, the spatial distribution differs between the near-field and far-field regions. Therefore, in practical calculations, a more accurate spatial volume calculation based on a beam propagation model can be used, which considers factors such as antenna radiation patterns and distance attenuation. To simplify the process and ensure feasibility, one implementation method approximates the instantaneous observation area as a spatial cone, with the apex of the spatial cone located at the antenna phase center of the microwave radiometer, the central axis of the spatial cone along the beam pointing direction, and the cone angle equal to the beamwidth. The atmospheric space pointed to by this spatial cone is defined as the instantaneous observation area of ​​the microwave radiometer in the spatial dimension.

[0072] Next, based on the time information and data integration period of the microwave radiometer observation data, the effective observation period of the microwave radiometer in the time dimension is determined. The time information of the microwave radiometer observation data has been extracted in step S1. The time information is a Coordinated Universal Time (UTC) timestamp, corresponding to the moment when a radiation brightness temperature data point is recorded. The data integration period refers to the length of time the microwave radiometer accumulates and averages the raw signal to generate an effective radiation brightness temperature data point. The data integration period is a configurable parameter of the device and can be read from the configuration parameters of the microwave radiometer's control software. For example, the data integration period can be 1 second, 5 seconds, or 10 seconds. The effective observation period represents the acquisition time window of the atmospheric radiation signal that actually corresponds to the generation of an observation data point. The method for determining the effective observation period is to take the moment indicated by the time information of the microwave radiometer observation data as the end moment of the effective observation period, and then trace back a certain duration from the end moment. The traceback duration is equal to the data integration period, and the start moment of the traceback is defined as the start moment of the effective observation period. Therefore, the effective observation period is a continuous time interval. The start time of the effective observation period is the beginning time, and the end time is the end time. The length of the effective observation period is equal to the data integration period. For example, if the time information of a radiation brightness temperature data is 12:00:05 and the data integration period is 5 seconds, then the corresponding effective observation period is the time interval from 12:00:00 to 12:00:05.

[0073] Finally, the instantaneous observation area is spatiotemporally extended along the effective observation period, forming a joint representation of a spatial cone centered on the beam direction and a time interval. This joint representation is defined as the first spatiotemporal representation range. Spatiotemporal extension is a logical operation; its purpose is to merge the simple concepts of space and time into a four-dimensional spatiotemporal volume. Specifically, the spatial geometric parameters of the instantaneous observation area (the spatial cone) remain unchanged, allowing the spatial cone to exist continuously in the time dimension. The duration of this continuous existence equals the length of the effective observation period, and the duration of this continuous existence equals the effective observation period. Thus, within the entire time interval from the start to the end of the effective observation period, the instantaneous observation area in space is considered a possible source of microwave radiometer signals. The resulting joint representation is a four-dimensional spatiotemporal volume, which spatially is a spatial cone centered on the beam direction, and temporally is a continuous interval from the start to the end of the effective observation period. This four-dimensional spacetime volume is defined as the first spacetime representation range, which fully characterizes the possible spacetime location of the signal source corresponding to a single microwave radiometer observation data point. In the actual data structure representation, the first spacetime representation range can be described by a set of parameters, including the azimuth angle of the central axis, the elevation angle of the central axis, the cone angle of the spatial cone, the start time of the effective observation period, and the end time of the effective observation period. Using these parameters, a computer program can reconstruct or determine whether any given spatial point and time point lies within the first spacetime representation range.

[0074] S3: Based on the scanning parameters of the laser wind measuring radar and the dwell time in each spatial direction, determine the second spatiotemporal characterization range corresponding to the laser wind measuring radar observation data. The specific implementation is as follows:

[0075] In determining the second spatiotemporal characterization range corresponding to the observation data of the laser wind radar, it is first necessary to obtain the multiple spatial directions accessed sequentially by the laser wind radar during the scanning process and the corresponding dwell time for each spatial direction. Scanning parameters refer to a series of control commands and equipment status parameters followed by the laser wind radar when executing a preset scanning mode. These parameters can be read from the laser wind radar's control system configuration file or real-time control log. Multiple spatial directions constitute a scanning sequence, and each spatial direction in the scanning sequence is defined by a set of azimuth and elevation coordinates, with the equipment's installation coordinate system as a reference. For example, a typical volume velocity processing scanning sequence includes four spatial directions with azimuth angles of 0 degrees, 90 degrees, 180 degrees, and 270 degrees, and elevation angles of 60 degrees for all four. The dwell time for each spatial direction refers to the duration for which the laser wind radar's servo mechanism stably points the beam towards the spatial direction and effectively acquires data. The dwell time is a preset duration, such as 3 seconds or 5 seconds, and is specified in the scanning mode configuration in association with each spatial direction. By analyzing the scanning control commands of the laser wind radar or reading the scanning status feedback, the azimuth angle, elevation angle, and corresponding dwell time of each spatial orientation in the scanning sequence can be obtained.

[0076] For each acquired spatial direction, the corresponding instantaneous observation area needs to be determined based on the beamwidth and spatial direction of the laser wind measuring radar. The beamwidth of the laser wind measuring radar is an angular value, representing the divergence angle of the laser beam emitted by the radar. The beamwidth is, for example, 100 microradians or 200 microradians, and can be obtained from the radar's technical specifications. The spatial direction provides the direction of the central axis of the instantaneous observation area. The instantaneous observation area corresponding to the spatial direction is defined as the range covered by a cone extending through the atmosphere with the spatial direction as its central axis and the beamwidth as its cone angle. The specific method for determining the instantaneous observation area corresponding to the spatial direction is to construct a spatial cone by using the direction determined by the azimuth and elevation angles of the spatial direction as the central axis of the cone and the beamwidth as the cone angle. The apex of the spatial cone is located at the center of the optical transmitting antenna of the laser wind measuring radar, the central axis of the spatial cone is along the direction of the spatial direction, and the cone angle of the spatial cone is equal to the beamwidth. The spatial volume defined by the spatial cone is defined as the instantaneous observation area corresponding to the spatial direction.

[0077] For each spatial direction, the effective observation period needs to be determined based on the dwell time corresponding to the spatial direction and the time information of the laser wind radar observation data. The time information of the laser wind radar observation data has been extracted in step S1, and the time information is a Coordinated Universal Time (UTC) timestamp. For a specific spatial direction, the time information of the laser wind radar observation data usually marks the moment when a complete data acquisition is completed within the dwell time of the spatial direction, such as the moment when a radial wind speed profile inversion is completed. The calculation of the effective observation period corresponding to the spatial direction needs to be combined with the dwell time. The method for determining the effective observation period is to use the moment indicated by the time information of the laser wind radar observation data related to the spatial direction as the reference moment, and use the reference moment as the end moment of the effective observation period. Since the data acquisition occurs within the entire dwell time interval, the start moment of the effective observation period needs to be calculated by subtracting the dwell time corresponding to the spatial direction from the end moment. Specifically, the start moment of the effective observation period is equal to the end moment minus the dwell time corresponding to the spatial direction. Therefore, the effective observation period corresponding to spatial pointing is a continuous time interval. The starting point of the effective observation period is the calculated start time, and the ending point is the reference end time. The length of the effective observation period is equal to the dwell time corresponding to the spatial pointing. For example, if the dwell time of a laser wind-measuring radar on a spatial pointing with an azimuth angle of 0 degrees and an elevation angle of 60 degrees is 5 seconds, and the time information of the observation data related to this pointing is 12:00:10, then the effective observation period corresponding to the spatial pointing is the time interval from 12:00:05 to 12:00:10.

[0078] The instantaneous observation area corresponding to each spatial direction is spatiotemporally extended along the effective observation period corresponding to that spatial direction, resulting in a spatiotemporal unit corresponding to the spatial direction. Spatiotemporal extension is a logical construction process; its purpose is to create a four-dimensional spatiotemporal description for each discrete spatial direction observation. Specifically, for a given spatial direction, the geometric shape of the spatial cone corresponding to the instantaneous observation area remains unchanged, allowing the spatial cone to persist in the time dimension. The duration of this persistence is exactly equal to the effective observation period corresponding to the spatial direction. In other words, throughout the entire time interval from the start to the end of the effective observation period corresponding to the spatial direction, the spatial cone described by the instantaneous observation area corresponding to the spatial direction is considered a possible source region for the laser wind radar signal. Through this extension, each spatial direction observation is characterized as a four-dimensional spatiotemporal unit. Spatially, this unit is a spatial cone with the spatial direction as its central axis; temporally, it is a specific time interval corresponding to the dwell time of the spatial direction.

[0079] Finally, all spatial pointers and their corresponding spatiotemporal units are merged in the spatiotemporal domain to form the second spatiotemporal representation range. The merging operation refers to aggregating multiple discrete spatiotemporal units into a continuous or composite spatiotemporal region to comprehensively describe the spatiotemporal range covered by all observations from the laser wind radar within a complete scan cycle. Merging in the spatiotemporal domain means taking the union of the spatiotemporal volumes occupied by all spatiotemporal units in a four-dimensional spatiotemporal coordinate system. In practice, all spatial pointers in the scan sequence can be traversed, and the parameters of the spatiotemporal unit corresponding to each spatial pointer can be added to a set or list. The parameters of the spatiotemporal unit include the central axis azimuth angle, central axis elevation angle, spatial cone angle, start time of the effective observation period, and end time of the effective observation period. The second spatiotemporal representation range is thus defined by all spatiotemporal units in the set or list. When it is necessary to determine the inclusion of a spatial point or time point, it is only necessary to check whether the spatial point or time point is located within any spatiotemporal unit in the set. The second spatiotemporal representation range formed by this merging method is a set composed of multiple discrete four-dimensional spatiotemporal volumes. This set completely represents the possible spatiotemporal locations of the signal sources corresponding to all observation data points of the lidar during a single scan. Through this second spatiotemporal representation range, it can be understood that the observations of the lidar are spatiotemporally discontinuous and discretely sampled, which contrasts sharply with the continuous integration observations of the microwave radiometer. This difference forms the basis for subsequent spatiotemporal matching and interference analysis.

[0080] S4: Determine the overlapping spatiotemporal region based on the first and second spatiotemporal characterization ranges. Within the overlapping spatiotemporal region, analyze the potential interference of the laser wind radar scanning process parameters on the observation data obtained by the microwave radiometer during its integration observation period, and obtain the interference assessment results. The specific implementation is as follows:

[0081] In step S4, the intersection of the first and second spatiotemporal representation ranges in the spatiotemporal domain is first calculated, and this intersection is defined as the overlapping spatiotemporal region. The first spatiotemporal representation range is a four-dimensional spatiotemporal volume describing the source of a single observation signal from a microwave radiometer, and the second spatiotemporal representation range is a set of four-dimensional spatiotemporal volumes describing the source of all observation signals within a complete scan cycle of a laser wind radar. Calculating the intersection requires identifying those spatiotemporal points simultaneously covered by both the first and second spatiotemporal representation ranges. In practice, since the first spatiotemporal representation range is typically a continuous four-dimensional spatiotemporal volume, and the second spatiotemporal representation range is a set of multiple discrete four-dimensional spatiotemporal units, calculating the intersection requires checking each spatiotemporal unit in the second spatiotemporal representation range for overlap with the first spatiotemporal representation range. Determining whether two four-dimensional spatiotemporal volumes overlap requires comparing whether their intervals in the spatial and temporal dimensions intersect. For the spatial dimension, determining whether two spatial cones have a common part in three-dimensional space is accomplished by calculating the angle between the central axes of the two cones and their distances to the common spatial region. Specifically, the angle between the central axis of the spatial cone in the first spatiotemporal representation range and the central axis of the spatial cone in a certain spatiotemporal unit within the second spatiotemporal representation range is calculated. If the calculated angle is less than the sum of the half-cone angle of the spatial cone in the first spatiotemporal representation range and the half-cone angle of the spatial cone in the second spatiotemporal representation range, then the two spatial cones are determined to overlap spatially. For the time dimension, it is determined whether the two time intervals intersect, i.e., whether the start time of the effective observation period in the first spatiotemporal representation range is earlier than the end time of the effective observation period of the spatial unit in the second spatiotemporal representation range, and whether the end time of the effective observation period in the first spatiotemporal representation range is later than the start time of the effective observation period of the spatial unit in the second spatiotemporal representation range. When a spatiotemporal unit in the second spatiotemporal representation range and the first spatiotemporal representation range simultaneously satisfy the overlap conditions in both spatial and time dimensions, their overlapping portion constitutes a sub-intersection. All spatiotemporal units within the second spatiotemporal representation range are traversed, and all sub-intersections are merged to obtain the complete overlapping spatiotemporal region. The overlapping spatiotemporal region itself is also a collection of four-dimensional spatiotemporal regions. The overlapping spatiotemporal region represents the area where the observations of microwave radiometer and laser wind radar may affect each other in spatiotemporal space.

[0082] Within the overlapping spatiotemporal region, the time periods during which the laser wind measuring radar is in scanning motion and stationary pointing state are identified. The scanning process of the laser wind measuring radar includes a scanning motion state of moving from one spatial direction to the next, and a stationary pointing state of stabilizing and measuring in a certain spatial direction. Identifying these time periods requires the scanning process parameters of the laser wind measuring radar, including the scanning angular velocity and the dwell time for each spatial direction obtained in step S3. The scanning angular velocity refers to the angular velocity of the servo mechanism rotating the beam; the scanning angular velocity is a preset or measured parameter, such as 5 degrees per second or 10 degrees per second. Based on the start and end times of the effective observation period corresponding to each spatial direction obtained in step S3, the time intervals corresponding to these effective observation periods are the time periods during which the laser wind measuring radar is in stationary pointing state. The time gaps between the effective observation periods of each spatial direction arranged in chronological order are extracted; these time gaps are the time periods during which the laser wind measuring radar is in scanning motion. Specifically, for the i-th spatial pointing direction arranged chronologically, the end time of its effective observation period is Tiend, and the start time of the effective observation period of the (i+1)-th spatial pointing direction is Ti+1start. The time interval [Tiend, Ti+1start] represents the time period during which the laser wind measuring radar is in a scanning motion from the i-th pointing direction to the (i+1)-th pointing direction. It is necessary to compare the identified stationary pointing state time periods and scanning motion state time periods with the time range covered by the overlapping spatiotemporal region. Only those state time periods that fall completely or partially within the time range of the overlapping spatiotemporal region are retained. These retained time periods are the time periods during which the laser wind measuring radar is in a scanning motion state and a stationary pointing state identified within the overlapping spatiotemporal region.

[0083] During the scanning motion period of the laser wind measuring radar, short-term fluctuation characteristics of microwave radiometer observation data are extracted. Microwave radiometer observation data refers to radiation brightness temperature data. Specifically, the short-term fluctuation characteristic is the standard deviation of the microwave radiometer observation data within the corresponding time period. The specific process for extracting the standard deviation is as follows: First, from the radiation brightness temperature data sequence of the microwave radiometer, all data points whose time markers fall within the time period when the laser wind measuring radar is in scanning motion are selected. These data points constitute a subset of data, called the scanning motion state data subset. The standard deviation of the scanning motion state data subset is calculated using the following formula: First, calculate the arithmetic mean of all data values ​​in the scanning motion state data subset; then, calculate the difference between each data value and the arithmetic mean; square each difference; sum all the squared values; divide the sum by the number of data points in the scanning motion state data subset minus one; finally, take the square root of the division result. The calculated square root value is the short-term fluctuation characteristic value of the microwave radiometer observation data during the scanning motion period of the laser wind measuring radar. This characteristic value is a non-negative real number that quantifies the degree of data dispersion.

[0084] During the period when the laser wind measuring radar is in a stationary pointing state, short-term fluctuation characteristics of the microwave radiometer observation data are extracted. The extraction method is the same as that used during the scanning motion state. From the microwave radiometer's radiation brightness temperature data sequence, all data points whose time markers fall within the period when the laser wind measuring radar is in a stationary pointing state are selected, forming a subset of the stationary pointing state data. The standard deviation of the stationary pointing state data subset is calculated, and the calculation process is exactly the same as that for the scanning motion state data subset, but applied to the stationary pointing state data subset. The calculated value is the short-term fluctuation characteristic value of the microwave radiometer observation data during the period when the laser wind measuring radar is in a stationary pointing state.

[0085] The interference assessment result is generated by comparing the short-term fluctuation characteristics under scanning motion and stationary pointing states. The comparison difference refers to the difference between the short-term fluctuation characteristic values ​​under scanning motion and stationary pointing states. The specific logic for generating the interference assessment result is to divide the short-term fluctuation characteristic value under scanning motion by the short-term fluctuation characteristic value under stationary pointing states to obtain a ratio. The interference assessment result can be a judgment based on this ratio. For example, a preset interference judgment threshold, which is a real number greater than 1, can be used. When the calculated ratio is greater than the interference judgment threshold, an assessment result indicating significant interference is generated; when the calculated ratio is less than or equal to the interference judgment threshold, an assessment result indicating no significant interference is generated. The interference judgment threshold can be set based on historical observation data analysis or experimental calibration. For example, long-term observation under a known interference-free environment can be used to statistically calculate the ratio distribution, and a high percentage of the ratio distribution, such as the 95th percentile, can be set as the interference judgment threshold. The interference assessment result can also be a continuous interference intensity index, such as directly using the calculated ratio as the interference intensity value. The interference assessment results are ultimately quantified into a numerical value or level that can be used for subsequent steps.

[0086] S5: Based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results, evaluate the reliability of data fusion between microwave radiometer observation data and laser wind radar observation data. The specific implementation is as follows:

[0087] In step S5, the spatiotemporal coverage matching degree is determined based on the proportion of the overlapping spatiotemporal region of the first and second spatiotemporal representation ranges within their respective representation ranges. The first spatiotemporal representation range describes a four-dimensional spatiotemporal volume of a single observation by a microwave radiometer, while the second spatiotemporal representation range describes a set of four-dimensional spatiotemporal volumes of a complete scan observation by a laser wind radar. The overlapping spatiotemporal region is the intersection of these two representation ranges calculated in step S4. Determining the proportion requires quantifying the measurement of the overlapping part relative to its respective whole. The measurement adopts the concept of a four-dimensional hypervolume. The volume of the first spatiotemporal representation range is calculated. The first spatiotemporal representation range is a cone in space and an interval in time. Its four-dimensional volume is calculated as the three-dimensional volume of the spatial cone multiplied by the length of the time interval. The three-dimensional volume of the spatial cone is calculated using the cone volume formula: the cone volume equals the cone's base area multiplied by its height and then divided by 3. The cone's base radius equals the cone's height multiplied by the tangent of the half-cone angle. The height of the cone is taken as a representative detection distance, which is set according to the typical detection range of the microwave radiometer, such as 1000 meters or 3000 meters. The time interval length of the first spatiotemporal representation range is equal to the end time of its effective observation period minus the start time of the effective observation period. The volume of the second spatiotemporal representation range is calculated. The second spatiotemporal representation range consists of multiple spatiotemporal units, and the total volume of the second spatiotemporal representation range is the sum of the volumes of each spatiotemporal unit. The volume of each spatiotemporal unit is calculated similarly to that of the first spatiotemporal representation range. The three-dimensional volume of the spatial cone of each spatiotemporal unit is calculated, multiplied by the length of the effective observation period of that spatiotemporal unit, and then the product of all spatiotemporal units is summed. The volume of the overlapping spatiotemporal region is calculated. The overlapping spatiotemporal region may consist of multiple discrete four-dimensional sub-regions, and the total volume of the overlapping spatiotemporal region is the sum of the volumes of each sub-region. The volume of each sub-region is calculated based on its specific spatial shape and time span. When the geometry is complex, numerical integration methods are used for approximate calculation, such as the trapezoidal rule or Simpson's rule. The proportion of overlapping spatiotemporal regions within the first spatiotemporal representation range is equal to the volume of the overlapping spatiotemporal regions divided by the volume of the first spatiotemporal representation range. Similarly, the proportion of overlapping spatiotemporal regions within the second spatiotemporal representation range is equal to the volume of the overlapping spatiotemporal regions divided by the volume of the second spatiotemporal representation range. The spatiotemporal coverage matching degree is jointly determined by the proportions of overlapping spatiotemporal regions within the first and second spatiotemporal representation ranges. A specific method for determining the spatiotemporal coverage matching degree is to average these two proportions, calculating the arithmetic or geometric mean, and using this average as the spatiotemporal coverage matching degree. The spatiotemporal coverage matching degree is a value between 0 and 1, where 1 indicates a perfect match and 0 indicates no overlap.

[0088] Based on the interference intensity characterized by the interference assessment results, the interference impact factor is determined. The interference assessment results, derived from step S4, may be a Boolean flag indicating significant interference or a continuous interference intensity value, such as the ratio of the standard deviation of microwave radiometer observation data in a scanning motion state to that in a stationary pointing state. The interference impact factor is a coefficient used to quantify the negative impact of interference on the reliability of data fusion. It is typically a value between 0 and 1, where 1 indicates a complete negative impact and 0 indicates no impact. The specific logic for determining the interference impact factor needs to be designed according to the type of interference assessment results. If the interference assessment result is a Boolean flag, when the flag indicates significant interference, the interference impact factor is set to a fixed decrement, such as 0.3 or 0.5; when the flag indicates no significant interference, the interference impact factor is set to 0. If the interference assessment result is a continuous interference intensity value, a mapping relationship from the interference intensity value to the interference impact factor needs to be established. One mapping method involves setting an interference intensity threshold. When the interference intensity value is below the threshold, the interference impact factor is 0. When the interference intensity value is above the threshold, the interference impact factor linearly increases from 0 to 1 as the interference intensity value increases. The slope of this linear increase is determined by a preset parameter. The setting of the interference intensity threshold and the linear increase slope is based on prior knowledge or experimental calibration. For example, different intensities of mechanical vibration are artificially introduced through controlled experiments, and the corresponding interference intensity observations and the degree of degradation in microwave radiometer data quality are recorded. An empirical relationship curve between the interference intensity observations and the degree of data quality degradation is established, and the interference intensity threshold and the linear increase slope are calibrated based on this empirical relationship curve.

[0089] Using spatiotemporal coverage matching as the base confidence level, the data fusion confidence level is obtained by downgrading the base confidence level based on the interference impact factor. The base confidence level, i.e., the spatiotemporal coverage matching level, reflects the data fusionability assessed purely from the perspective of spatiotemporal geometric coverage. The downgrading operation is to incorporate the negative impact of physical interference between devices. The specific mathematical operation for downgrading is a multiplicative downgrading, i.e., the data fusion confidence level equals the base confidence level multiplied by a discount factor determined by the interference impact factor. The discount factor equals 1 minus the interference impact factor. Therefore, the data fusion confidence level is calculated as: Data fusion confidence level = Spatiotemporal coverage matching level × (1 - interference impact factor). This multiplicative model means that the interference impact factor directly reduces the confidence level proportionally. The final data fusion confidence level is an overall assessment value that integrates spatiotemporal coverage completeness and inter-device interference factors. The data fusion confidence level will be directly used to guide the fusion processing decision in step S6. The closer the data fusion confidence level is to 1, the more suitable the two sets of observation data are for deep fusion; the closer the data fusion confidence level is to 0, the greater the risk or the lower the value of fusion.

[0090] S6: Based on the reliability of data fusion, the microwave radiometer observation data and the laser wind radar observation data are fused. The specific implementation is as follows:

[0091] In step S6, the microwave radiometer observation data and the laser wind radar observation data are fused based on the data fusion reliability. The data fusion reliability is a comprehensive evaluation value calculated in step S5. Before performing the fusion process, a first reliability threshold and a second reliability threshold need to be preset. The first reliability threshold and the second reliability threshold are two critical values ​​used to classify the processing levels, with the first reliability threshold being greater than the second reliability threshold. The specific values ​​of the first reliability threshold and the second reliability threshold are determined by analyzing the data fusion reliability distribution of successful and unsuccessful fusion cases in historical observation data. Specifically, a large number of historical observation instances are collected, and the data fusion reliability of each historical observation instance is calculated according to the method in step S5. Based on the results of post-event manual verification or independent high-precision data verification, the historical observation instances are divided into three categories: high-quality fusion, acceptable fusion, and unsuitable fusion. The minimum data fusion reliability of the high-quality fusion historical observation instances is statistically analyzed, and this minimum value is used as the reference value for the first reliability threshold. The minimum data fusion reliability of the acceptable fusion historical observation instances is statistically analyzed, and this minimum value is used as the reference value for the second reliability threshold. In a practical system, the first confidence threshold can be set to, for example, 0.8, and the second confidence threshold can be set to, for example, 0.5. The first and second confidence thresholds are stored as configuration parameters in the system configuration file, and can be adjusted during the system calibration phase based on new calibration data.

[0092] When the data fusion reliability exceeds a preset first reliability threshold, joint inversion and fusion processing is performed on the microwave radiometer observation data and the lidar wind measurement radar observation data. The joint inversion and fusion processing uses both types of heterogeneous observation data as constraints, inputting them into a unified atmospheric parameter inversion algorithm to solve for the optimal atmospheric state parameter estimates. Specifically, an atmospheric state parameter and forward observation model are first established. The atmospheric state parameters include the vertical profiles of temperature, humidity, horizontal wind speed, and vertical wind speed. The forward observation model includes the microwave radiative transfer equation and the lidar wind measurement equation. The microwave radiative transfer equation describes the theoretical radiative brightness temperature that the microwave radiometer should observe in each frequency channel under given atmospheric conditions, while the lidar wind measurement equation describes the theoretical radial wind speed that the lidar wind measurement radar should observe in each spatial direction under given wind field conditions. The core of joint inversion is constructing a joint cost function, which consists of two parts. The first part is the sum of squares of the differences between microwave radiometer observation data and the theoretical radiative brightness temperature calculated using the microwave radiative transfer equation. The second part is the sum of squares of the differences between lidar observation data and the theoretical radial wind speed calculated using the lidar wind measurement equation. The joint cost function is equal to the sum of the first and second parts multiplied by their respective weights. The joint inversion fusion process minimizes the joint cost function using a numerical optimization algorithm to find a set of atmospheric state parameters that best fits both the microwave radiometer and lidar observation data. The minimization process uses an iterative optimization algorithm, such as the Levenberg-Marquardt algorithm. The atmospheric state parameter profile obtained after the optimization algorithm converges is the product of the joint inversion fusion process.

[0093] When the data fusion confidence level is lower than or equal to a preset first confidence threshold and higher than a preset second confidence threshold, a weighted average fusion process is performed on the microwave radiometer observation data and the laser wind radar observation data. This weighted average fusion process is a data-level fusion; it does not perform joint inversion on the original observation data, but rather weights and combines the primary products obtained independently by the two devices. The microwave radiometer observation data obtains temperature and humidity profiles through independent inversion algorithms, while the laser wind radar observation data obtains horizontal wind field profiles through independent inversion algorithms. The weighted average fusion process is performed separately for each atmospheric parameter at each altitude level. For the same atmospheric parameter at the same altitude level, such as temperature, given the temperature value Tmwr obtained independently from the microwave radiometer and the temperature reference value Tlidar obtained with the assistance of the laser wind radar data, the weighted average fused temperature value Tfused is calculated as: Tfused = w × Tmwr + (1-w) × Tlidar, where w is the weighting coefficient assigned to the microwave radiometer data. The weighting coefficient is positively correlated with the data fusion confidence level. One specific implementation involves making the weighting coefficient equal to the product of the data fusion reliability and a fixed scaling factor, i.e., w = data fusion reliability × k, where k is the fixed scaling factor. The fixed scaling factor k maps the data fusion reliability to a reasonable weight range. The value of the fixed scaling factor k is determined experimentally based on the relative accuracy of the two independently retrieved products. By comparing the weighted average fusion result with high-precision reference data, the value of k is adjusted to minimize the fusion error. The weighting coefficient w must be limited to between 0 and 1. If the calculated value is less than 0, the weighting coefficient w is set to 0; if the calculated value is greater than 1, the weighting coefficient w is set to 1. This fusion strategy reduces dependence on low-reliability data sources when the data fusion reliability is moderate, achieving robust fusion through weighting.

[0094] When the data fusion confidence level is lower than or equal to a preset second confidence level threshold, the microwave radiometer observation data and the laser wind radar observation data are retained as independent data products. This means that no substantive data fusion operation is performed. The processing system packages and outputs the temperature and humidity profiles independently generated by the microwave radiometer and the wind field profiles independently generated by the laser wind radar separately. Simultaneously, metadata flags are added to the temperature, humidity, and wind field profiles, indicating that the data fusion confidence level for this observation is low and joint application analysis is not recommended. Retaining independent data products is a conservative strategy, avoiding the potentially larger errors or misleading results that might be introduced by forced fusion when data quality is questionable or spatiotemporal matching is extremely poor. The independent microwave radiometer and laser wind radar products can still be used by users individually or by experts for subsequent offline analysis and judgment.

[0095] Example 2: Figure 2 A schematic diagram of the microwave-laser composite detection system based on heterogeneous data spatiotemporal matching of the present invention is given. The microwave-laser composite detection system based on heterogeneous data spatiotemporal matching includes the following modules:

[0096] The data acquisition module is used to acquire observation data from the microwave radiometer and the laser wind radar. The observation data includes corresponding time information.

[0097] The first characterization module is used to determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data based on the beam parameters of the microwave radiometer and its observation process within the time information.

[0098] The second characterization module is used to determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction.

[0099] The interference analysis module is used to determine the overlapping spatiotemporal region based on the first spatiotemporal characterization range and the second spatiotemporal characterization range. Within the overlapping spatiotemporal region, it analyzes the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period, and obtains the interference assessment results.

[0100] The credibility assessment module is used to assess the credibility of data fusion between microwave radiometer observation data and laser wind radar observation data based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results.

[0101] The fusion processing module is used to fuse microwave radiometer observation data and laser wind radar observation data according to the data fusion reliability.

[0102] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0103] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0104] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0105] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0107] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0108] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A microwave-laser composite detection method based on spatiotemporal matching of heterogeneous data, characterized in that, Includes the following steps: S1: Acquire observation data from microwave radiometer and laser wind radar, including corresponding time information; S2: Based on the beam parameters of the microwave radiometer and its observation process within the time information, determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data; S3: Determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction; S4: Determine the overlapping spatiotemporal region based on the first and second spatiotemporal characterization ranges, analyze the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period within the overlapping spatiotemporal region, and obtain the interference assessment results. S5: Based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results, evaluate the reliability of data fusion between microwave radiometer observation data and laser wind radar observation data; S6: Based on the reliability of data fusion, the microwave radiometer observation data and the laser wind radar observation data are fused.

2. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S1 includes: A set of radiation brightness temperature data with continuous time stamps was collected from a microwave radiometer; A set of radial wind speed data with corresponding scan time markers was collected from the laser wind radar; Extract unified time reference information from the shared clock source or timing device of the microwave radiometer and the laser wind radar; Based on a unified time reference, the time information of microwave radiometer observation data is extracted from radiation brightness temperature data, and the time information of laser wind radar observation data is extracted from radial wind speed data.

3. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S2 include: Based on the beamwidth and beam pointing of the microwave radiometer, its instantaneous observation area in the spatial dimension is determined; Based on the time information and data integration period of microwave radiometer observation data, the effective observation period in the time dimension is determined. The instantaneous observation area is spatiotemporally extended along the effective observation period to form a joint representation of a spatial cone with the beam pointing as the central axis and a time interval, and this joint representation is defined as the first spatiotemporal representation range.

4. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S3 include: Acquire the multiple spatial directions visited sequentially by the laser wind radar during the scanning process and the corresponding dwell time for each spatial direction; For each spatial orientation, the instantaneous observation area corresponding to the spatial orientation is determined based on the beamwidth and spatial orientation of the laser wind measuring radar; For each spatial direction, the effective observation period corresponding to the spatial direction is determined based on the dwell time corresponding to the spatial direction and the time information of the laser wind radar observation data; Each spatial pointer is spatiotemporally extended along the effective observation period corresponding to the spatial pointer to obtain the spatiotemporal unit corresponding to the spatial pointer. All spatial points are merged into corresponding spatiotemporal units in the spatiotemporal domain to form a second spatiotemporal representation range.

5. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S4 include: Calculate the intersection of the first spatiotemporal representation range and the second spatiotemporal representation range in the spatiotemporal domain, and define the intersection as the overlapping spatiotemporal region; Within the overlapping spatiotemporal region, identify the time periods during which the laser wind-measuring radar is in scanning motion and the time periods during which it is in stationary pointing state; During the period when the laser wind radar is in scanning motion, the short-term fluctuation characteristics of the microwave radiometer observation data are extracted. During the period when the laser wind measuring radar is in a stationary pointing state, the short-term fluctuation characteristics of the microwave radiometer observation data are extracted. By comparing the short-term fluctuation characteristics under scanning motion and under stationary pointing conditions, interference assessment results are generated based on the comparison differences.

6. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 5, characterized in that, The scanning process parameters include the scanning angular velocity of the laser wind radar, and the short-term fluctuation characteristics are the standard deviation of the microwave radiometer observation data within the corresponding time period.

7. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S5 include: The spatiotemporal coverage matching degree is determined based on the proportion of the overlapping spatiotemporal regions of the first and second spatiotemporal representation ranges within their respective representation ranges. Based on the interference intensity characterized by the interference assessment results, the interference impact factor is determined; Using the spatiotemporal coverage matching degree as the basic credibility, the basic credibility degree is adjusted down according to the interference impact factor to obtain the data fusion credibility degree.

8. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 1, characterized in that, S6 include: When the data fusion credibility is higher than the preset first credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are jointly inverted and fused. When the data fusion credibility is lower than or equal to the preset first credibility threshold and higher than the preset second credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are subjected to weighted average fusion processing, where the weighting coefficient is positively correlated with the data fusion credibility. When the data fusion credibility is lower than or equal to the preset second credibility threshold, the microwave radiometer observation data and the laser wind radar observation data are retained as independent data products.

9. The microwave-laser composite detection method based on heterogeneous data spatiotemporal matching according to claim 8, characterized in that, The joint inversion and fusion processing includes using microwave radiometer observation data and laser wind radar observation data as common constraints to input atmospheric parameter inversion algorithms. The weighting coefficient in the weighted average fusion processing is equal to the product of the data fusion confidence and the fixed scaling factor.

10. A microwave-laser composite detection system based on heterogeneous data spatiotemporal matching, used to implement the microwave-laser composite detection method based on heterogeneous data spatiotemporal matching as described in any one of claims 1-9, characterized in that, Includes the following modules: The data acquisition module is used to acquire observation data from the microwave radiometer and the laser wind radar. The observation data includes corresponding time information. The first characterization module is used to determine the first spatiotemporal characterization range corresponding to the microwave radiometer observation data based on the beam parameters of the microwave radiometer and its observation process within the time information. The second characterization module is used to determine the second spatiotemporal characterization range corresponding to the laser wind measurement radar observation data based on the scanning parameters of the laser wind measurement radar and the dwell time in each spatial direction. The interference analysis module is used to determine the overlapping spatiotemporal region based on the first spatiotemporal characterization range and the second spatiotemporal characterization range. Within the overlapping spatiotemporal region, it analyzes the potential interference of the scanning process parameters of the laser wind radar on the observation data obtained by the microwave radiometer during its integration observation period, and obtains the interference assessment results. The credibility assessment module is used to assess the credibility of data fusion between microwave radiometer observation data and laser wind radar observation data based on the coverage relationship between the first and second spatiotemporal characterization ranges and the interference assessment results. The fusion processing module is used to fuse microwave radiometer observation data and laser wind radar observation data according to the data fusion reliability.

Images