CN112131989A - Millimeter wave rain measurement model parameter obtaining method based on space rainfall data - Google Patents
Millimeter wave rain measurement model parameter obtaining method based on space rainfall data Download PDFInfo
- Publication number
- CN112131989A CN112131989A CN202010964548.3A CN202010964548A CN112131989A CN 112131989 A CN112131989 A CN 112131989A CN 202010964548 A CN202010964548 A CN 202010964548A CN 112131989 A CN112131989 A CN 112131989A
- Authority
- CN
- China
- Prior art keywords
- data
- link
- millimeter wave
- rainfall
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/14—Rainfall or precipitation gauges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/20—Monitoring; Testing of receivers
- H04B17/21—Monitoring; Testing of receivers for calibration; for correcting measurements
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Environmental & Geological Engineering (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computer Networks & Wireless Communication (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Electromagnetism (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Hydrology & Water Resources (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a millimeter wave rain measurement model parameter obtaining method based on space rainfall data, which comprises the following steps of obtaining the corrected value of the attenuation value of each link signal in a millimeter wave networkA C (ii) a Secondly, acquiring theoretical line average rainfall data of each link by using spatial grid rainfall dataR i (ii) a Step three, uniformly and synchronously extracting 75% of time period data according to each rainfall field to form a calibration data set, and combining the rest data into a verification data set; and fourthly, establishing a data-driven inversion model of the millimeter wave link. Fifthly, utilizing the calibration data set to calibrate or train the data-driven inversion model constructed in the fifth step; sixth aspect of the inventionStep (d), verifying the accuracy of the validated model using the verification dataset. The method effectively solves the problem that the data set is obtained by the data-driven millimeter wave rain measurement inversion model, improves the reliability of the training data set, and can meet the requirement of constructing the complex millimeter wave rain measurement network data-driven inversion model.
Description
Technical Field
The invention relates to a millimeter wave rain measurement model parameter acquisition method based on space rainfall data, and belongs to the technical field of meteorological element monitoring.
Background
The inversion model driven by data is a key means for improving the accuracy of the millimeter wave network rain measurement in a complex environment. However, the existing historical rainfall data and the millimeter wave link have the problem of mismatch of spatial scales. Taking a rainfall station as an example, rainfall data of the rainfall station is in a form of point data, rainfall provided by the millimeter wave link is in a linear average form, and the rainfall data and the millimeter wave link have difference in spatial scale, so that even if the rainfall station is installed right below the link, the link monitoring area is larger than that of the rainfall station, and the rainfall station data cannot be completely matched with the spatial scale of the millimeter wave attenuation data; spatial grid rainfall data in the form of radar, satellite, INCA and the like can provide regional rainfall spatial distribution, however, matching grid data with spatial features with millimeter wave data is a pending problem, and the problem of scale difference causes that historical data used for constructing data-driven model parameters in the prior art is not reasonable. On the other hand, because the millimeter wave rain measuring network is seriously influenced by the surrounding meteorological environment, and the number of the historical data matched with the millimeter wave rain measuring network is limited, how to divide the historical data set into a calibration data set and a verification data set under the condition of less data and enable the calibration data set and the verification data set to contain attenuation data under more environmental conditions can be realized, so that the data set is more representative, the key problem of obtaining reasonable parameters and ensuring the model effect can be realized, the data volume can be reduced, and the memory and the calculated amount can be saved.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects in the prior art, the invention provides a millimeter wave rain measurement model parameter acquisition method based on spatial rainfall data, provides a rainfall data spatial scale conversion method suitable for a millimeter wave link, effectively solves the problem that a data-driven millimeter wave rain measurement inversion model acquires a data set, improves the reliability of a training data set, and simultaneously provides a data set construction method for calibration and verification of model parameters, so that more reasonable model parameters can be acquired under the condition that the amount of historical data is as small as possible.
The technical scheme is as follows: in order to solve the technical problem, the invention provides a millimeter wave rain measurement model parameter acquisition method based on space rainfall data, which comprises the following steps:
the method comprises the steps of firstly, obtaining attenuation values A of signals of each link in a millimeter wave network, and carrying out missing value interpolation and singular value correction preprocessing on the attenuation values A;
secondly, further correcting the attenuation value A of each link, removing the attenuation caused by the environmental noise and obtaining a corrected signal AC；
Thirdly, theoretical line average rainfall data of each link is obtained by utilizing spatial grid rainfall data, and for the ith link, the condition that the link passes through a grid rainfall data monitoring area is counted, wherein the condition comprises the length l that the ith millimeter wave link passes through the jth gridijAnd rainfall data r of the gridjMean rainfall data R of theoretical line of the linkiThe calculation method comprises the following steps:
wherein, biIs the power law value of the ith link, LiThe total length of the ith link is, and n is the total number of grids crossed by the ith link;
and fourthly, uniformly and synchronously extracting 75% of time period data from the historical attenuation data obtained in the second step and the historical rainfall data set obtained in the third step according to each rainfall field to form a rating data set, and combining the rest data into a verification data set.
Fifthly, establishing a data-driven inversion model of the millimeter wave link;
and sixthly, utilizing the calibration data set divided in the fourth step to calibrate or train the data-driven inversion model constructed in the fifth step, wherein the used data also comprises the self attributes of each link of the millimeter wave network, such as frequency, length, polarization mode and the like.
And seventhly, verifying the accuracy of the model confirmed in the sixth step by using the verification data set divided in the fourth step, receiving the model if the model accuracy meets the user requirement, and readjusting the parameter range to perform the sixth step again or expanding the data time length if the model accuracy does not meet the user requirement, and returning to the beginning of the first step.
Preferably, in the fifth step, a BP neural network is selected as a data-driven inversion model.
In the invention, the millimeter wave rainfall inversion model for the ith link can also be
Wherein R isiThe rainfall intensity of the ith link; si、di、mi、ci、τiFor the ith link model parameter, si、diDepending on the link frequency, polarization, mi、ciRelated to regional environment, tau is a sliding window parameter; a'CCorrecting the ith link obtained in the fourth step by resampling to obtain a basic attenuation amount; l isiIs the ith link length. For each parameter, measuring attenuation value A 'according to history'CAnd constructing a historical data set by historical rainfall data R of the rainfall station in the same period, and determining model parameters s, d, m, c and tau of each link in the fifth step in a parameter calibration mode. The influence of regional environment elements on millimeter wave attenuation is considered in the model structure, corresponding inversion models can be provided for different regions and different link attributes, the requirement of millimeter wave rainfall inversion under a complex scene can be met, and the model structure has stronger applicability compared with other technologies.
Has the advantages that: according to the millimeter wave rain measurement model parameter acquisition method based on spatial rainfall data, the grid rainfall data with spatial information can be subjected to scale conversion to obtain millimeter wave link theoretical rainfall data in a line average form, the problem that a data set is acquired by a data-driven millimeter wave rain measurement inversion model is effectively solved, the reliability of a training data set is improved, and the requirement of constructing the complex millimeter wave rain measurement network data-driven inversion model can be met; the method for uniformly and synchronously extracting the data according to the rainfall fields provided by the invention is an optimal division mode suitable for the historical data set of the millimeter wave network according to the characteristics of the millimeter wave attenuation data, ensures the representativeness of the data set to the greatest extent, improves the representativeness of the training data set, can reduce the data volume, and saves the memory and the calculated amount.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
Fig. 2 is a schematic diagram of millimeter wave links traversing spatial grid data in the present invention.
Fig. 3 shows the effect of the data set partitioning method in the present invention on the representative retention of the data set (only the rainfall period is listed, and 6 rainfall events are selected, and each box represents one rainfall event).
Detailed Description
The present invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, a millimeter wave rain measurement model parameter obtaining method based on spatial rainfall data includes the following steps:
the method comprises the steps of firstly, obtaining signal strength TSL of a signal transmitting end of each link and signal strength RSL of a receiving end in a millimeter wave network, subtracting the signal strength RSL of the receiving end from the signal strength TSL of the transmitting end of each link to obtain a signal attenuation value A, preprocessing data, interpolating lost signal strength, removing singular values and interpolating;
secondly, judging the dry period and the wet period by using the sliding standard deviation of the attenuation A, wherein the size of a sliding window is consistent with the required rainfall data time; determining a base attenuation value A using the dry and wet period resultsBLTaking the average value of the data of the first 24-hour dry period of the basic attenuation; the effect of the wet antenna is corrected to a correction value of 0.2dBm to obtain a corrected attenuation value AC＝A-ABL-0.2；
Thirdly, acquiring theoretical line average rainfall data R by using regional radar rainfall dataiCounting the condition that each millimeter wave link passes through the radar monitoring area, wherein the condition comprises the length l that the ith millimeter wave link passes through the jth radar pixelijAnd the pixel radar data rjThe link theoretical line average rainfall data calculation method comprises the following steps:
wherein, biIs the power law value of the ith link, LiThe total length of the ith link is shown, and n is the number of radar pixels crossed by the ith link;
and fourthly, uniformly and synchronously extracting 75% of time period data from the historical attenuation data obtained in the second step and the historical rainfall data set obtained in the third step according to the rainfall times to form a calibration data set, and combining the rest data into a verification data set.
And fifthly, selecting the BP neural network as a data-driven inversion model.
And sixthly, training the BP neural network constructed in the fifth step by using the calibration data set divided in the fourth step and the frequency, length and polarization mode of each link of the millimeter wave network, wherein the structural expression of each layer of neural network is as follows:
wherein x isiRepresenting the input items of attenuation, link frequency, length, polarization mode and the like after correction; m is the total number of the neural nodes; θ is the connection weight of the ith input; f is an excitation function, optionally a Sigmoid function; b is a correction term; h is the output;
preprocessing training data, including unified conversion of training sample size and magnitude; assuming that the number of the hidden layer nodes is 4-10, and comparing the convergence conditions of the models of the nodes with the corresponding number; setting learning efficiency and training times, wherein the learning efficiency is set to be 0.2, and the maximum training time is set to be 1000000; the calculated expected error is set to 0.1.
And seventhly, verifying the accuracy of the BP neural network trained in the sixth step by using the verification data set divided in the fourth step, receiving the model if the model accuracy meets the user requirement, and readjusting the number of nodes and the learning efficiency to perform the sixth step again or expanding the data time length if the model accuracy does not meet the user requirement, and returning to the first step to start.
In the invention, the model of the fifth step is obtained through the sixth step and training, and in practice, the attenuation value A is only needed to be correctedCMillipore and riceThe average rainfall data R can be obtained according to the self attributes of the wave network such as the frequency, the length, the polarization mode and the like of each link. According to the millimeter wave rain measurement model parameter acquisition method for the spatial rainfall data, provided by the invention, the grid rainfall data with spatial information can be subjected to scale conversion to obtain millimeter wave link theoretical rainfall data in a line average form, the problem that a data set is acquired by a data-driven millimeter wave rain measurement inversion model is effectively solved, the reliability of a training data set is improved, and the requirement for constructing the complex millimeter wave rain measurement network data-driven inversion model can be met. The millimeter wave attenuation caused by environmental elements has locality, namely the millimeter wave attenuation can be regarded as the same ambient environment in a short time, and for different rainfall fields, due to the fact that the rainfall environment and the rainfall type are different, the fields have great difference, and a better model can be trained by using attenuation data under more environmental conditions; on the other hand, the relationship between rainfall and attenuation is in one-to-one correspondence, and only the correspondence is considered in inversion, but the sequencing relationship is not considered. Therefore, the uniform extraction of the data set according to each rainfall field is the best mode for ensuring that the data set still has the corresponding relation of the original data set after extraction by the millimeter wave rain measurement technology. According to the method, the representativeness of the data set is ensured to the greatest extent according to the dividing mode of the historical data set provided by the data characteristics of rainfall data, the representativeness of the training data set is improved, the data volume can be reduced, and the memory and the calculated amount are saved.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (2)
1. A millimeter wave rain measurement model parameter obtaining method based on space rainfall data is characterized by comprising the following steps:
the method comprises the steps of firstly, obtaining attenuation values A of signals of each link in a millimeter wave network, and carrying out missing value interpolation and singular value correction preprocessing on the attenuation values A;
secondly, further correcting the attenuation value A of each link, removing the attenuation caused by the environmental noise and obtaining a corrected signal ACAs a historical millimeter wave link attenuation data set;
thirdly, theoretical line rainfall data of each link is obtained by utilizing spatial grid rainfall data, and for the ith link, the condition that the link passes through a grid rainfall data monitoring area is counted, wherein the condition comprises the length l that the ith millimeter wave link passes through the jth gridijAnd rainfall data r of the gridjThe link theoretical line rainfall data RiThe calculation method comprises the following steps:
wherein, biIs the power law value of the ith link, LiThe total length of the ith link is, and n is the total number of grids crossed by the ith link;
step four, uniformly and synchronously extracting 75% of time period data from the historical millimeter wave link attenuation data set obtained in the step two and the historical rainfall data set obtained in the step three according to each rainfall field to form a calibration data set, and combining the rest data into a verification data set;
fifthly, establishing a data-driven inversion model of the millimeter wave link;
sixthly, calibrating or training the data-driven inversion model constructed in the fifth step by using the calibration data set divided in the fourth step, wherein the used data further comprises the frequency, the length and the polarization mode of each link of the millimeter wave network;
and seventhly, verifying the accuracy of the model confirmed in the sixth step by using the verification data set divided in the fourth step, receiving the model if the model accuracy meets the user requirement, and readjusting the parameter range to perform the sixth step again or expanding the data time length if the model accuracy does not meet the user requirement, and returning to the beginning of the first step.
2. The millimeter wave rain measurement model parameter acquisition method based on spatial rainfall data according to claim 1, characterized in that: in the fifth step, a BP neural network is selected as a data-driven inversion model, and the structural expression of each layer of neural network is as follows:
wherein x isiRepresenting the input items of attenuation, link frequency, length, polarization mode and the like after correction; m is the total number of the neural nodes; θ is the connection weight of the ith input; f is an excitation function which is a Sigmoid function; b is a correction term; h is the output.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010964548.3A CN112131989B (en) | 2020-09-15 | 2020-09-15 | Millimeter wave rain measurement model parameter obtaining method based on space rainfall data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010964548.3A CN112131989B (en) | 2020-09-15 | 2020-09-15 | Millimeter wave rain measurement model parameter obtaining method based on space rainfall data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112131989A true CN112131989A (en) | 2020-12-25 |
| CN112131989B CN112131989B (en) | 2021-07-23 |
Family
ID=73845230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010964548.3A Active CN112131989B (en) | 2020-09-15 | 2020-09-15 | Millimeter wave rain measurement model parameter obtaining method based on space rainfall data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112131989B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023066186A1 (en) * | 2021-10-22 | 2023-04-27 | 维沃移动通信有限公司 | Weather sensing method, and device |
| CN116383701A (en) * | 2023-04-13 | 2023-07-04 | 中国人民解放军国防科技大学 | Microwave link rainy period detection method based on learning reconstruction |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102540148A (en) * | 2012-02-15 | 2012-07-04 | 上海大学 | Inversion algorithm of rainfall three-dimensional distribution parameters |
| CN103616736A (en) * | 2013-09-06 | 2014-03-05 | 中国人民解放军理工大学 | Rainfall foundation monitoring method based on GNSS signal depolarization effect |
| CN106324580A (en) * | 2016-11-02 | 2017-01-11 | 中国人民解放军理工大学 | Method for correcting radar echo attenuation based on microwave link network |
| CN107656279A (en) * | 2017-09-19 | 2018-02-02 | 中国电子科技集团公司第三十八研究所 | A kind of rain detection radar radiation transmission system based on two-parameter volume size distribution |
| CN109255100A (en) * | 2018-09-10 | 2019-01-22 | 河海大学 | A kind of Urban Rain inversion algorithm based on microwave attenuation characteristic response fingerprint recognition |
| CN109325206A (en) * | 2018-09-10 | 2019-02-12 | 柳创新 | A kind of Rainfall Runoff Model parameter optimization method |
| CN109697323A (en) * | 2018-12-27 | 2019-04-30 | 武汉大学 | A kind of rainfall observation method merging satellite remote sensing and mobile communication base station signal |
| CN109840873A (en) * | 2019-02-02 | 2019-06-04 | 中国水利水电科学研究院 | A kind of Cross Some Region Without Data Hydro-Model Parameter Calibration Technology fields method based on machine learning |
| CN110361742A (en) * | 2019-06-21 | 2019-10-22 | 中国人民解放军国防科技大学 | Cloud rain micro physical parameter inversion method based on satellite-borne three-frequency millimeter wave radar |
| CN110533233A (en) * | 2019-08-20 | 2019-12-03 | 河海大学 | Wireless microwave based on fitness optimization surveys rain link planing method |
| CN110543893A (en) * | 2019-08-07 | 2019-12-06 | 河海大学 | Microwave attenuation precipitation particle type identification method based on BP neural network |
| CN110850510A (en) * | 2019-11-05 | 2020-02-28 | 中国人民解放军国防科技大学 | Method for reconstructing vertical rainfall field based on satellite-ground link scanning |
| CN110909485A (en) * | 2019-12-05 | 2020-03-24 | 重庆交通大学 | SWMM model parameter self-calibration method based on BP neural network |
| CN111553394A (en) * | 2020-04-20 | 2020-08-18 | 中国长江三峡集团有限公司 | Reservoir water level prediction method based on cyclic neural network and attention mechanism |
| CN107316095B (en) * | 2016-09-23 | 2020-09-11 | 武汉大学 | Regional weather drought level prediction method coupled with multi-source data |
-
2020
- 2020-09-15 CN CN202010964548.3A patent/CN112131989B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102540148A (en) * | 2012-02-15 | 2012-07-04 | 上海大学 | Inversion algorithm of rainfall three-dimensional distribution parameters |
| CN103616736A (en) * | 2013-09-06 | 2014-03-05 | 中国人民解放军理工大学 | Rainfall foundation monitoring method based on GNSS signal depolarization effect |
| CN107316095B (en) * | 2016-09-23 | 2020-09-11 | 武汉大学 | Regional weather drought level prediction method coupled with multi-source data |
| CN106324580A (en) * | 2016-11-02 | 2017-01-11 | 中国人民解放军理工大学 | Method for correcting radar echo attenuation based on microwave link network |
| CN107656279A (en) * | 2017-09-19 | 2018-02-02 | 中国电子科技集团公司第三十八研究所 | A kind of rain detection radar radiation transmission system based on two-parameter volume size distribution |
| CN109255100A (en) * | 2018-09-10 | 2019-01-22 | 河海大学 | A kind of Urban Rain inversion algorithm based on microwave attenuation characteristic response fingerprint recognition |
| CN109325206A (en) * | 2018-09-10 | 2019-02-12 | 柳创新 | A kind of Rainfall Runoff Model parameter optimization method |
| CN109697323A (en) * | 2018-12-27 | 2019-04-30 | 武汉大学 | A kind of rainfall observation method merging satellite remote sensing and mobile communication base station signal |
| CN109840873A (en) * | 2019-02-02 | 2019-06-04 | 中国水利水电科学研究院 | A kind of Cross Some Region Without Data Hydro-Model Parameter Calibration Technology fields method based on machine learning |
| CN110361742A (en) * | 2019-06-21 | 2019-10-22 | 中国人民解放军国防科技大学 | Cloud rain micro physical parameter inversion method based on satellite-borne three-frequency millimeter wave radar |
| CN110543893A (en) * | 2019-08-07 | 2019-12-06 | 河海大学 | Microwave attenuation precipitation particle type identification method based on BP neural network |
| CN110533233A (en) * | 2019-08-20 | 2019-12-03 | 河海大学 | Wireless microwave based on fitness optimization surveys rain link planing method |
| CN110850510A (en) * | 2019-11-05 | 2020-02-28 | 中国人民解放军国防科技大学 | Method for reconstructing vertical rainfall field based on satellite-ground link scanning |
| CN110909485A (en) * | 2019-12-05 | 2020-03-24 | 重庆交通大学 | SWMM model parameter self-calibration method based on BP neural network |
| CN111553394A (en) * | 2020-04-20 | 2020-08-18 | 中国长江三峡集团有限公司 | Reservoir water level prediction method based on cyclic neural network and attention mechanism |
Non-Patent Citations (6)
| Title |
|---|
| JIAXING DU等: "An Application on the Immune Evolutionary Algorithm Based on Back Propagation in the Rainfall Prediction", 《2012 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND ELECTRONICS ENGINEERING》 * |
| MINGHAO XIAN等: "Rainfall Monitoring Based on Machine Learning by Earth-Space Link in the Ku Band", 《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》 * |
| 任佳等: "基于BP网络的卫星通信雨衰预测模型", 《通信系统与网络技术》 * |
| 印敏等: "微波链路测量降水研究综述", 《气象》 * |
| 印敏等: "微波雨衰特性在降雨测量中的应用", 《气象科技》 * |
| 姜世泰等: "基于微波链路的降雨场反演方法研究", 《物理学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023066186A1 (en) * | 2021-10-22 | 2023-04-27 | 维沃移动通信有限公司 | Weather sensing method, and device |
| CN116383701A (en) * | 2023-04-13 | 2023-07-04 | 中国人民解放军国防科技大学 | Microwave link rainy period detection method based on learning reconstruction |
| CN116383701B (en) * | 2023-04-13 | 2024-01-26 | 中国人民解放军国防科技大学 | Microwave link rainy period detection method based on learning reconstruction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112131989B (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363327B (en) | ConvLSTM and 3D-CNN-based short rainfall prediction method | |
| CN111078678B (en) | Satellite precipitation data correction method based on multi-source information fusion and scale reduction | |
| CN112131989B (en) | Millimeter wave rain measurement model parameter obtaining method based on space rainfall data | |
| CN111666656A (en) | Rainfall estimation method and rainfall monitoring system based on microwave rainfall attenuation | |
| CN111027752A (en) | Crop yield estimation method based on deep spatiotemporal feature joint learning | |
| CN113534159B (en) | Sea surface wind speed inversion method and system based on SAR satellite remote sensing data | |
| Rios Gaona et al. | Measurement and interpolation uncertainties in rainfall maps from cellular communication networks | |
| CN101482929B (en) | Remote-sensing image processing method and system | |
| CN113255972B (en) | Short-term rainfall prediction method based on Attention mechanism | |
| CN114049570B (en) | Satellite-borne remote sensing water vapor space inversion method and system based on neural network | |
| CN110263838B (en) | Multi-sensor rainfall estimation fusion method | |
| Schepen et al. | Calibration, bridging, and merging to improve GCM seasonal temperature forecasts in Australia | |
| CN116297068B (en) | Aerosol optical thickness inversion method and system based on earth surface reflectivity optimization | |
| CN112906310A (en) | Method for optimizing BP neural network microwave remote sensing soil moisture inversion by considering firefly algorithm | |
| CN110516552B (en) | Multi-polarization radar image classification method and system based on time sequence curve | |
| CN115526413A (en) | Prediction method based on daily maximum temperature of fully-connected neural network | |
| CN114490905A (en) | Clear sky surface net long wave radiation integrated inversion method and system | |
| CN112131990B (en) | Millimeter wave network rainfall inversion model construction method suitable for complex scene | |
| CN115936238B (en) | Method, system, equipment and medium for predicting medium-long-term output of global wind power | |
| CN111915570A (en) | Atmospheric delay estimation method based on back propagation neural network | |
| CN115980755A (en) | Storm approach extrapolation method and device based on ConvLSTM | |
| CN113836731B (en) | Construction method and device of land surface stable target atmosphere top reflectivity model | |
| CN109905190A (en) | A kind of modeling method of low frequency propagation of ground wave time delay time-varying characteristics | |
| CN112836449B (en) | Method for calibrating hydrologic model | |
| CN114065931A (en) | Land water vapor inversion method and system based on physical model and neural network fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |