CN113177360B - Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relation - Google Patents
Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relationDownload PDFInfo
- Publication number
- CN113177360B CN113177360BCN202110489466.2ACN202110489466ACN113177360BCN 113177360 BCN113177360 BCN 113177360BCN 202110489466 ACN202110489466 ACN 202110489466ACN 113177360 BCN113177360 BCN 113177360B
- Authority
- CN
- China
- Prior art keywords
- rain
- rainfall
- attenuation
- link
- clustering
- 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.)
- Active
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及雨强自适应估计技术领域，特别是涉及一种基于聚类雨衰关系的雨强自适应估计系统及方法。The invention relates to the technical field of rain intensity self-adaptive estimation, in particular to a rain intensity self-adaptive estimation system and method based on a clustered rain attenuation relationship.
背景技术Background technique
城市水文信息的监测需要获取高时空分辨率的降雨资料。虽然天气雷达能够满足这一要求，但是其需要标定和经常维护，且目前只有稀疏部署的雨量计可以用于标定天气雷达。现有商用通信网络微波信号在传输过程中会受到路径上降雨的衰减作用，根据这一基本原理，可以实现基于商用通信网络的降雨信息监测。由于该网络覆盖了地球大部分陆地表面，且密度很高(尤其在城市范围内)，因此可以满足全球范围内近地面、低成本、高时空分辨率的降雨信息获取需求。The monitoring of urban hydrological information requires the acquisition of rainfall data with high spatial and temporal resolution. While weather radars can meet this requirement, they require calibration and frequent maintenance, and currently only sparsely deployed rain gauges can be used to calibrate weather radars. The microwave signal of the existing commercial communication network will be attenuated by the rainfall on the path during the transmission process. According to this basic principle, the monitoring of rainfall information based on the commercial communication network can be realized. Because the network covers most of the earth's land surface and has a high density (especially in urban areas), it can meet the needs of near-ground, low-cost, high-spatial-temporal rainfall information acquisition on a global scale.
基于微波链路衰减信息反演降雨强度一般利用的是国际电信联盟推荐的雨衰-雨强经验幂律关系(ITU-R P.838-3)，但是该关系的准确性受雨滴谱分布的影响。当微波运行频率在40GHz以下(传统商用通信网络所运行频段范围)时，该关系受雨滴谱分布影响有限，因此对雨强及累积降雨量估计误差较小。但是当频率大于40GHz后，这种雨滴谱分布影响将不可忽略。随着5G通信技术的不断普及，更高频段如50GHz、60GHz、E波段(71-76GHz和81-86GHz)和92-95GHz的微波回程链路将在全世界广泛部署。因此，为了高频微波链路也能应用于精确的定量降雨估计，需要对现有的雨衰关系进行改进。Rain intensity inversion based on microwave link attenuation information generally uses the empirical power-law relationship between rain attenuation and rain intensity recommended by the International Telecommunication Union (ITU-R P.838-3), but the accuracy of this relationship is affected by the spectral distribution of raindrops. influences. When the microwave operating frequency is below 40 GHz (the frequency range used by traditional commercial communication networks), this relationship is limited by the spectral distribution of raindrops, so the estimation error of rain intensity and cumulative rainfall is small. But when the frequency is greater than 40GHz, the influence of this raindrop spectral distribution cannot be ignored. With the increasing popularity of 5G communication technology, microwave backhaul links in higher frequency bands such as 50GHz, 60GHz, E-band (71-76GHz and 81-86GHz) and 92-95GHz will be widely deployed around the world. Therefore, in order that the high-frequency microwave link can also be applied to accurate quantitative rainfall estimation, the existing rain attenuation relationship needs to be improved.
发明内容SUMMARY OF THE INVENTION
本发明的目的是提供一种基于聚类雨衰关系的雨强自适应估计系统及方法，以解决现有技术中存在的技术问题，能够适用于各类降雨样本的雨衰关系，进而实现降雨的精细化定量估计，可广泛应用于降水信息监测等气象信息遥感领域。The purpose of the present invention is to provide a rain intensity self-adaptive estimation system and method based on the clustered rain attenuation relationship, so as to solve the technical problems existing in the prior art, which can be applied to the rain attenuation relationship of various types of rainfall samples, and then realize rainfall The refined quantitative estimation can be widely used in the field of meteorological information remote sensing such as precipitation information monitoring.
为实现上述目的，本发明提供了如下方案：本发明提供一种基于聚类雨衰关系的雨强自适应估计系统，包括依次连接的：微波通信网、数据采集处理终端、聚类单元、雨强反演单元；In order to achieve the above purpose, the present invention provides the following scheme: the present invention provides a rain intensity adaptive estimation system based on clustered rain attenuation relationship, including: microwave communication network, data acquisition and processing terminal, clustering unit, rain Strong inversion unit;
其中，所述微波通信网包括若干条链路，每条所述链路包括接收端和发射端；Wherein, the microwave communication network includes several links, and each of the links includes a receiving end and a transmitting end;
所述数据采集处理终端基于微波通信网中各链路发射端和接收端的电平信号，分别获取待测区域的雨致衰减率以及若干组历史降雨时段中的雨致衰减率，并将若干组历史降雨时段中的雨致衰减率构建为降雨样本集；The data acquisition and processing terminal respectively obtains the rain-induced attenuation rate of the area to be measured and the rain-induced attenuation rate in several groups of historical rainfall periods based on the level signals of the transmitter and the receiver of each link in the microwave communication network, and calculates the rate of rain-induced attenuation in several groups. The rain-induced attenuation rate in the historical rainfall period is constructed as a rainfall sample set;
所述聚类单元将各链路的所述雨致衰减率作为特征量，采用聚类方法将所述降雨样本集中的降雨样本划分为k类；所述聚类单元还基于各类的聚类中心，根据距离最小原则对待测区域的雨致衰减率进行自动归类；The clustering unit uses the rain-induced attenuation rate of each link as a feature quantity, and adopts a clustering method to divide the rainfall samples in the rainfall sample set into k categories; the clustering unit is also based on the clustering of various types. Center, according to the principle of minimum distance, the rain attenuation rate of the area to be measured is automatically classified;
所述雨强反演单元利用最小二乘法对降雨样本集中的各类降雨样本分别建立雨衰关系；所述雨强反演单元还基于待测区域雨致衰减率的自动归类结果，采用相应的雨衰关系对待测区域进行降雨强度反演。The rain intensity inversion unit uses the least squares method to establish the rain attenuation relationship for various types of rainfall samples in the rainfall sample set respectively; Rainfall intensity inversion is carried out in the area to be measured.
优选地，所述数据采集处理终端包括依次连接的数据获取单元、链路衰减获取单元、雨致衰减率获取单元；所述数据获取单元分别与所述微波通信网中各链路的接收端和发射端连接，所述雨致衰减单元与所述聚类单元连接；Preferably, the data acquisition and processing terminal includes a data acquisition unit, a link attenuation acquisition unit, and a rain attenuation rate acquisition unit connected in sequence; the data acquisition unit is respectively connected to the receiving end and the receiving end of each link in the microwave communication network. The transmitting end is connected, and the rain-induced attenuation unit is connected with the clustering unit;
其中，所述数据获取单元用于获取微波通信网中各链路的发射电平、接收电平；Wherein, the data acquisition unit is used to acquire the transmit level and receive level of each link in the microwave communication network;
所述链路衰减获取单元基于各链路的发射电平和接收电平，计算各链路的总衰减；The link attenuation obtaining unit calculates the total attenuation of each link based on the transmit level and the receive level of each link;
所述雨致衰减率获取单元基于各链路的总衰减，提取各链路的平均雨致衰减率。The rain-induced attenuation rate obtaining unit extracts the average rain-induced attenuation rate of each link based on the total attenuation of each link.
优选地，微波通信网中链路的条数包括但不限于1条。Preferably, the number of links in the microwave communication network includes but is not limited to one.
本发明还提供一种基于聚类雨衰关系的雨强自适应估计方法，其特征在于，包括如下步骤：The present invention also provides a rain intensity adaptive estimation method based on the clustered rain attenuation relationship, which is characterized in that it includes the following steps:
S1、利用数据采集处理终端获取若干组历史降雨时段内微波通信网的链路接收端和发射端电平信号，基于微波通信网的链路接收端和发射端电平信号提取雨致衰减率；S1. Use the data acquisition and processing terminal to obtain the level signals of the link receiving end and the transmitting end of the microwave communication network in several groups of historical rainfall periods, and extract the rain-induced attenuation rate based on the level signals of the link receiving end and the transmitting end of the microwave communication network;
S2、将各链路的雨致衰减率作为特征量，采用聚类方法将所述降雨样本集中的降雨样本划分为k类；S2, using the rain-induced attenuation rate of each link as a feature quantity, and using a clustering method to divide the rainfall samples in the rainfall sample set into k categories;
S3、利用最小二乘法，对降雨样本集中的各类降雨样本分别建立雨衰关系；S3. Use the least squares method to establish a rain attenuation relationship for various rainfall samples in the rainfall sample set respectively;
S4、获取待测区域的雨致衰减率，基于步骤S2中各类的聚类中心，根据距离最小原则对待测区域的雨致衰减率进行自动归类，基于归类结果，采用相应的雨衰关系对待测区域进行降雨强度反演。S4. Obtain the rain-induced attenuation rate of the area to be measured, and automatically classify the rain-induced attenuation rate of the area to be measured based on the cluster centers of various types in step S2 according to the principle of minimum distance. Based on the classification result, use the corresponding rain attenuation rate The relationship between rainfall intensity inversion in the area to be measured.
优选地，所述步骤S1具体包括如下步骤：Preferably, the step S1 specifically includes the following steps:
S1.1、在区域内的微波通信网中选择n条链路，n条链路所发射的微波频率分别为f1，f2，…，fn，单位GHz，极化方式分别为α1，α2，…，αn；同一时刻，n条链路的发射电平分别为tx1，tx2，…，txn，单位dB，经过降雨空间后，接收电平分别为rx1，rx2，…，rxn，单位dB；S1.1. Select n links in the microwave communication network in the area. The microwave frequencies emitted by then links are f1 ,f2 , . , α2 , ..., αn ; at the same time, the transmit levels of n links are tx1 , tx2 , ..., txn , respectively, in dB, and after the rainfall space, the receive levels are rx1 , rx respectively2 , …, rxn , in dB;
S1.2、基于各链路的发射电平和接收电平，计算各链路的总衰减Ai,单位dB，如下式所示：S1.2. Based on the transmit level and receive level of each link, calculate the total attenuation Ai of each link, in dB, as shown in the following formula:
Ai＝txi-rxi(i＝1,…,n)；Ai =txi -rxi (i=1,...,n);
S1.3、基于各链路的总衰减，提取各链路的平均雨致衰减率γi，如下式所示：S1.3. Based on the total attenuation of each link, extract the average rain-induced attenuation rate γi of each link, as shown in the following formula:
γi＝(Ai-Aref,i)/Li(i＝1,...,n)γi =(Ai -Aref,i )/Li (i=1,...,n)
其中，Aref,i表示第i条链路的参考衰减值，单位dB，Li表示第i条链路的长度。Among them, Aref,irepresents the reference attenuation value of the ith link, in dB, and Li represents the length of the ith link.
优选地，所述极化方式包括水平极化、垂直极化。Preferably, the polarization modes include horizontal polarization and vertical polarization.
优选地，所述步骤S2,所述聚类方法包括但不限于模糊C-均值聚类方法。Preferably, in the step S2, the clustering method includes but is not limited to the fuzzy C-means clustering method.
优选地，所述步骤S3,所述雨衰关系包括但不限于单链路幂律关系。Preferably, in step S3, the rain attenuation relationship includes but is not limited to a single-link power-law relationship.
优选地，所述步骤S3,利用最小二乘法，对降雨样本集中的各类降雨样本分别建立雨衰关系的方法如下式所示：Preferably, in the step S3, using the least squares method, the method for establishing the rain attenuation relationship for each type of rainfall samples in the rainfall sample set is shown in the following formula:
其中，α和β为雨衰系数，Rinverse,t和Rreal,t(mm h-1)分别为某类降雨样本中第t个降雨样本的雨强估计值和真值，s为某类降雨样本的数量。Among them, α and β are the rain attenuation coefficients, Rinverse,t and Rreal,t (mm h-1 ) are the estimated value and true value of the rain intensity of the t-th rainfall sample in a certain type of rainfall sample, respectively, and s is a certain type of rainfall sample. Number of rainfall samples.
本发明公开了以下技术效果：The present invention discloses the following technical effects:
本发明提供一种基于聚类雨衰关系的雨强自适应估计系统及方法，通过采集微波通信网的链路接收端和发射端电平信号，提取出平均雨致衰减率作为特征量；根据“物以类聚”的基本思想，采用聚类方法对降雨样本聚类，并利用最小二乘法，拟合聚类雨衰关系；本发明通过聚类的方式，建立了适用于各类降雨样本的雨衰关系，进而实现降雨的精细化定量估计，可广泛应用于降水信息监测等气象信息遥感领域。同时，本发明所提供的方法在应用过程中，只需输入各链路雨致衰减特征量即可将降雨样本自动归类，进而采用对应的雨衰关系反演雨强，极大地降低了传统单链路幂律雨衰关系的不确定性。该方法可以作为微波链路定量降水反演新方法应用到实际业务当中。The invention provides a rain intensity self-adaptive estimation system and method based on a clustered rain attenuation relationship. By collecting the level signals of the link receiving end and the transmitting end of a microwave communication network, the average rain-induced attenuation rate is extracted as a characteristic quantity; The basic idea of "clustering things together" adopts the clustering method to cluster the rainfall samples, and uses the least squares method to fit the clustered rain attenuation relationship; the present invention establishes a rain attenuation suitable for various types of rainfall samples by means of clustering. It can be widely used in meteorological information remote sensing fields such as precipitation information monitoring and so on. At the same time, in the application process of the method provided by the present invention, the rain samples can be automatically classified only by inputting the characteristic quantities of rain-induced attenuation of each link, and then the corresponding rain-attenuation relationship is used to invert the rain intensity, which greatly reduces the traditional Uncertainty in the power-law rain attenuation relationship for a single link. This method can be applied to practical business as a new method for quantitative precipitation retrieval in microwave links.
附图说明Description of drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the present invention. In the embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.
图1为本发明基于聚类雨衰关系的雨强自适应估计系统结构示意图；1 is a schematic structural diagram of the rain intensity adaptive estimation system based on the clustered rain attenuation relationship of the present invention;
图2为本发明基于聚类方法的精细化雨衰关系确定流程图；Fig. 2 is a flow chart for determining the refined rain attenuation relationship based on the clustering method of the present invention;
图3为本发明基于聚类雨衰关系的雨强自适应估计流程图；Fig. 3 is the rain intensity self-adaptive estimation flow chart of the present invention based on clustered rain attenuation relationship;
图4为本发明基于聚类雨衰关系的雨强自适应估计示意图(当k＝3时)；Fig. 4 is the rain intensity adaptive estimation schematic diagram (when k=3) based on the cluster rain attenuation relation of the present invention;
图5(a)为本发明实例中的基于聚类雨衰关系的雨强自适应评估结果；图5(b)为本发明实例中的基于聚类雨衰关系的累积降雨量评估结果。FIG. 5( a ) is the self-adaptive evaluation result of rain intensity based on the clustered rain attenuation relationship in the example of the present invention; FIG. 5( b ) is the cumulative rainfall evaluation result based on the clustered rain attenuation relationship in the example of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合附图和具体实施方式对本发明作进一步详细的说明。In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.
参照图1所示，本实施例提供一种基于聚类雨衰关系的雨强自适应估计系统，包括依次连接的：微波通信网、数据采集处理终端、聚类单元、雨强反演单元；1 , the present embodiment provides a rain intensity adaptive estimation system based on a clustered rain attenuation relationship, including: a microwave communication network, a data acquisition and processing terminal, a clustering unit, and a rain intensity inversion unit connected in sequence;
其中，所述微波通信网包括若干条链路，每条所述链路包括接收端和发射端；微波通信网中链路的条数包括但不限于1条。Wherein, the microwave communication network includes several links, and each of the links includes a receiving end and a transmitting end; the number of links in the microwave communication network includes but is not limited to one.
所述数据采集处理终端基于微波通信网中各链路发射端和接收端的电平信号，分别获取待测区域的雨致衰减率以及若干组历史降雨时段中的雨致衰减率，并将若干组历史降雨时段中的雨致衰减率构建为降雨样本集。The data acquisition and processing terminal respectively obtains the rain-induced attenuation rate of the area to be measured and the rain-induced attenuation rate in several groups of historical rainfall periods based on the level signals of the transmitter and the receiver of each link in the microwave communication network, and calculates the rate of rain-induced attenuation in several groups. Rain-induced attenuation rates in historical rainfall periods are constructed as rainfall sample sets.
所述数据采集处理终端包括依次连接的数据获取单元、链路衰减获取单元、雨致衰减率获取单元；所述数据获取单元分别与所述微波通信网中各链路的接收端和发射端连接，所述雨致衰减单元与所述聚类单元连接；The data acquisition and processing terminal includes a data acquisition unit, a link attenuation acquisition unit, and a rain attenuation rate acquisition unit that are connected in sequence; the data acquisition unit is respectively connected to the receiving end and the transmitting end of each link in the microwave communication network , the rain attenuation unit is connected with the clustering unit;
其中，所述数据获取单元用于获取微波通信网中各链路的发射电平、接收电平；在微波通信网中选择n条链路，n条链路所发射的微波频率分别为f1，f2，…，fn(单位：GHz)，极化方式分别为α1，α2，…，αn(一般为水平极化或垂直极化)。同一时刻，n条链路的发射电平分别为tx1，tx2，…，txn(单位：dB)，经过降雨空间后，接收电平分别为rx1，rx2，…，rxn(单位：dB)。Wherein, the data acquisition unit is used to acquire the transmission level and reception level of each link in the microwave communication network; n links are selected in the microwave communication network, and the microwave frequencies emitted by the n links are respectively f1 , f2 , ..., fn (unit: GHz), and the polarization modes are α1 , α2 , ..., αn (generally horizontal polarization or vertical polarization). At the same time, the transmit levels of the n links are tx1 , tx2 ,..., txn (unit: dB), and after the rain space, the receive levels are rx1 , rx2 ,..., rxn ( Unit: dB).
所述链路衰减获取单元基于各链路的发射电平和接收电平，计算各链路的总衰减；各链路的总衰减Ai如下式所示：The link attenuation acquisition unit calculates the total attenuation of each link based on the transmit level and receive level of each link; the total attenuation Ai of each link is shown in the following formula:
Ai＝txi-rxi(i＝1,…,n)；Ai =txi -rxi (i=1,...,n);
所述雨致衰减率获取单元基于各链路的总衰减，提取各链路的平均雨致衰减率。各链路的平均雨致衰减率γi，如下式所示：The rain-induced attenuation rate obtaining unit extracts the average rain-induced attenuation rate of each link based on the total attenuation of each link. The average rain-induced attenuation rate γi of each link is as follows:
γi＝(Ai-Aref,i)/Li(i＝1,...,n)γi =(Ai -Aref,i )/Li (i=1,...,n)
其中，Aref,i表示第i条链路的参考衰减值，单位dB，Li表示第i条链路的长度。Among them, Aref,irepresents the reference attenuation value of the ith link, in dB, and Li represents the length of the ith link.
所述聚类单元将各链路的所述雨致衰减率作为特征量，采用聚类方法将所述降雨样本集中的降雨样本划分为k类；所述聚类单元还基于各类的聚类中心，根据距离最小原则对待测区域的雨致衰减率进行自动归类；其中，所述聚类方法包括但不限于模糊C-均值聚类方法。The clustering unit uses the rain-induced attenuation rate of each link as a feature quantity, and adopts a clustering method to divide the rainfall samples in the rainfall sample set into k categories; the clustering unit is also based on the clustering of various types. According to the principle of minimum distance, the rain attenuation rate of the area to be measured is automatically classified; wherein, the clustering method includes but is not limited to the fuzzy C-means clustering method.
所述雨强反演单元利用最小二乘法对降雨样本集中的各类降雨样本分别建立雨衰关系；所述雨强反演单元还基于待测区域雨致衰减率的自动归类结果，采用相应的雨衰关系对待测区域进行降雨强度反演；其中，所述雨衰关系包括但不限于单链路幂律关系。The rain intensity inversion unit uses the least squares method to establish the rain attenuation relationship for various types of rainfall samples in the rainfall sample set respectively; Perform rainfall intensity inversion for the area to be measured according to the rain attenuation relationship; wherein, the rain attenuation relationship includes but is not limited to a single-link power-law relationship.
参照图2-4所示本实施例还提供一种基于聚类雨衰关系的雨强自适应估计方法，包括如下步骤：Referring to Figures 2-4, the present embodiment also provides a rain intensity adaptive estimation method based on a clustered rain attenuation relationship, comprising the following steps:
S1、利用数据采集处理终端获取若干组历史降雨时段内微波通信网的链路接收端和发射端电平信号，基于微波通信网的链路接收端和发射端电平信号提取雨致衰减率，构建为降雨样本集；具体包括：S1. Use the data acquisition and processing terminal to obtain the level signals of the link receiving end and the transmitting end of the microwave communication network in several groups of historical rainfall periods, and extract the rain-induced attenuation rate based on the level signals of the link receiving end and the transmitting end of the microwave communication network, Constructed as a rainfall sample set; specifically includes:
S1.1、在区域内的微波通信网中选择n条链路，n条链路所发射的微波频率分别为f1，f2，…，fn(GHz)，极化方式分别为α1，α2，…，αn(一般为水平极化或垂直极化)。同一时刻，n条链路的发射电平分别为tx1，tx2，…，txn(dB)，经过降雨空间后，接收电平分别为rx1，rx2，…，rxn(dB)；S1.1. Select n links in the microwave communication network in the area, the microwave frequencies emitted by the n links are f1 , f2 , ..., fn (GHz) respectively, and the polarization modes are α1 respectively , α2 , ..., αn (generally horizontal polarization or vertical polarization). At the same time, the transmit levels of n links are tx1 , tx2 ,..., txn (dB), respectively, and after the rain space, the receive levels are rx1 , rx2 ,..., rxn (dB) ;
S1.2、基于各链路的发射电平和接收电平，计算各链路的总衰减Ai,单位dB，如下式所示：S1.2. Based on the transmit level and receive level of each link, calculate the total attenuation Ai of each link, in dB, as shown in the following formula:
Ai＝txi-rxi(i＝1,…,n)；Ai =txi -rxi (i=1,...,n);
S1.3、基于各链路的总衰减，提取各链路的平均雨致衰减率γi，如下式所示：S1.3. Based on the total attenuation of each link, extract the average rain-induced attenuation rate γi of each link, as shown in the following formula:
γi＝(Ai-Aref,i)/Li(i＝1,...,n)γi =(Ai -Aref,i )/Li (i=1,...,n)
其中，Aref,i表示第i条链路的参考衰减值，单位dB，Li表示第i条链路的长度。Among them, Aref,irepresents the reference attenuation value of the ith link, in dB, and Li represents the length of the ith link.
本实施例中，在待测区域内的微波通信网中选择2条链路，发射的微波频率分别为15GHz、81GHz，极化方式均为水平极化，链路长度均为1Km，同一时刻，2条链路的发射电平分别为tx1和tx2(dB)，经过降雨空间后，接收电平分别为rx1和rx2(dB)；In this embodiment, two links are selected in the microwave communication network in the area to be tested, the transmitted microwave frequencies are 15GHz and 81GHz respectively, the polarization modes are horizontal polarization, and the link lengths are both 1Km. The transmit levels of the two links are tx1 and tx2 (dB), respectively, and after the rain space, the receive levels are rx1 and rx2 (dB), respectively;
基于2条链路的发射电平和接收电平，个别计算各链路的总衰减(dB)：Based on the transmit and receive levels of the 2 links, calculate the total attenuation (dB) for each link individually:
Ai＝txi-rxi(i＝1,2)；Ai =txi -rxi (i=1,2);
提取2条链路的平均雨致衰减率(dB/Km)：Extract the average rain-induced attenuation (dB/Km) of the 2 links:
γi＝(Ai-Aref,i)/Li(i＝1,2)γi =(Ai -Aref,i )/Li (i=1,2)
其中，Aref,i可由一周内小时平均链路总衰减的中值确定；Among them, Aref,i can be determined by the median of the hourly average link total attenuation in a week;
S2、将各链路的雨致衰减率作为特征量，采用聚类方法将所述降雨样本集中的降雨样本划分为k类；S2, using the rain-induced attenuation rate of each link as a feature quantity, and using a clustering method to divide the rainfall samples in the rainfall sample set into k categories;
其中，特征量Xj＝[γ1,γ2,…,γn]，j∈[1,m]，m为降雨样本集中降雨样本的数量，所述聚类方法包括但不限于模糊C-均值聚类方法。Among them, the feature quantity Xj =[γ1 ,γ2 ,...,γn ], j∈[1,m], m is the number of rainfall samples in the rainfall sample set, and the clustering method includes but is not limited to fuzzy C- Means clustering method.
模糊c均值聚类融合了模糊理论的精髓。相较于k-means的硬聚类，模糊c均值聚类提供了更加灵活的聚类结果。因为大部分情况下，数据集中的对象不能划分成为明显分离的簇，指派一个对象到一个特定的簇有些生硬，也可能会出错。故，对每个对象和每个簇赋予一个权值，指明对象属于该簇的程度。当然，基于概率的方法也可以给出这样的权值，但是有时候我们很难确定一个合适的统计模型，因此使用具有自然地、非概率特性的模糊c均值就是一个比较好的选择。Fuzzy c-means clustering incorporates the essence of fuzzy theory. Compared with the hard clustering of k-means, fuzzy c-means clustering provides more flexible clustering results. Because in most cases objects in a dataset cannot be divided into distinct clusters, assigning an object to a particular cluster can be blunt and error-prone. Therefore, a weight is assigned to each object and each cluster, indicating the degree to which the object belongs to the cluster. Of course, probability-based methods can also give such weights, but sometimes it is difficult to determine a suitable statistical model, so using fuzzy c-means with natural, non-probabilistic properties is a better choice.
模糊C-均值聚类方法具体实现首先需要定义目标函数J(u,c)：The specific implementation of the fuzzy C-means clustering method first needs to define the objective function J(u,c):
ujp＝||Xj-cp||ujp =||Xj -cp ||
其中djp＝||Xj-cp||，cp为第p个簇的聚类中心，ujp为第j个降雨样本对第p个簇的隶属度
其中，λ为常数。在此基础上，根据下式计算并替换各聚类中心：where λ is a constant. On this basis, calculate and replace each cluster center according to the following formula:
通过上述公式交替更新样本隶属度和聚类中心值，直至到达最大迭代次数或算法收敛(判别条件连续两次目标函数值降低不足10-6)，即完成对降雨样本的聚类。The sample membership and the cluster center value are alternately updated by the above formula until the maximum number of iterations is reached or the algorithm converges (the discriminant condition is reduced by less than 10-6 twice in a row), that is, the clustering of rainfall samples is completed.
本实施例中，2条链路的雨致衰减率作为特征向量Xj＝[γ1,γ2]，根据模糊C-均值聚类方法，设定聚类簇数k＝10，将m个历史降雨样本分为10类。In this embodiment, the rain-induced attenuation rates of the two links are taken as the feature vector Xj =[γ1 ,γ2 ], and according to the fuzzy C-means clustering method, the number of clusters k=10 is set, and m The historical rainfall samples are divided into 10 categories.
S3、利用最小二乘法，对降雨样本集中的各类降雨样本分别建立雨衰关系；其中，所述雨衰关系包括但不限于单链路幂律关系。S3. Use the least squares method to establish a rain attenuation relationship for various types of rainfall samples in the rainfall sample set, respectively; wherein, the rain attenuation relationship includes but is not limited to a single-link power-law relationship.
本实施例中，利用最小二乘法分别确定每类降雨样本对应雨衰关系Rinverse＝aγβ(此处选取81GHz雨致衰减)的系数：In this embodiment, the least squares method is used to determine the coefficient of the rain attenuation relationship Rinverse = aγβ (here, 81GHz rain-induced attenuation is selected) corresponding to each type of rainfall sample:
其中，α和β为雨衰系数，与频率、极化方式、雨滴谱相关，Rinverse,t和Rreal,t(mm h-1)分别为某类降雨样本中第t个降雨样本的雨强估计值和真值，s为某类降雨样本的数量。Among them, α and β are rain attenuation coefficients, which are related to frequency, polarization mode, and raindrop spectrum, and Rinverse,t and Rreal,t (mm h-1 ) are the rain of the t-th rainfall sample in a certain type of rainfall sample, respectively. Strong estimates and true values, s is the number of rainfall samples of a certain type.
S4、获取待测区域的雨致衰减率，基于步骤S2中各类的聚类中心，根据距离最小原则对待测区域的雨致衰减率进行自动归类，基于归类结果，采用相应的雨衰关系对待测区域进行降雨强度反演。S4. Obtain the rain-induced attenuation rate of the area to be measured, and automatically classify the rain-induced attenuation rate of the area to be measured based on the cluster centers of various types in step S2 according to the principle of minimum distance. Based on the classification result, use the corresponding rain attenuation rate The relationship between rainfall intensity inversion in the area to be measured.
在实际应用过程中，通过提取新降雨样本的雨致衰减特征，在10个聚类中心里寻找与其距离接近的一个，并以此类作为降雨样本的类型。然后使用该类对应的81GHz雨衰关系反演出雨强值。图5(a)及图5(b)为基于聚类雨衰关系的雨强及累积降雨量自适应评估结果。In the actual application process, by extracting the rain attenuation characteristics of the new rainfall samples, one of the 10 cluster centers with a close distance to it is found, and this type is used as the type of rainfall samples. Then use the corresponding 81GHz rain attenuation relationship to invert the rain intensity value. Fig. 5(a) and Fig. 5(b) are the self-adaptive evaluation results of rain intensity and accumulated rainfall based on the clustered rain attenuation relationship.
本发明具有以下技术效果：The present invention has the following technical effects:
本发明提供一种基于聚类雨衰关系的雨强自适应估计方法，通过采集微波通信网的链路接收端和发射端电平信号，提取出平均雨致衰减率作为特征量；根据“物以类聚”的基本思想，采用聚类方法对降雨样本聚类，并利用最小二乘法，拟合聚类雨衰关系；本发明通过聚类的方式，建立了适用于各类降雨样本的雨衰关系，进而实现降雨的精细化定量估计，可广泛应用于降水信息监测等气象信息遥感领域。同时，本发明所提供的方法在应用过程中，只需输入各链路雨致衰减特征量即可将降雨样本自动归类，进而采用对应的雨衰关系反演雨强，极大地降低了传统单链路幂律雨衰关系的不确定性。该方法可以作为微波链路定量降水反演新方法应用到实际业务当中。The invention provides a rain intensity self-adaptive estimation method based on the clustering rain attenuation relationship. By collecting the level signals of the link receiving end and the transmitting end of the microwave communication network, the average rain-induced attenuation rate is extracted as a feature quantity; The basic idea of using the clustering method to cluster the rainfall samples, and using the least squares method to fit the clustered rain attenuation relationship; the present invention establishes the rain attenuation relationship suitable for various types of rainfall samples through the method of clustering. Then, the refined quantitative estimation of rainfall can be realized, which can be widely used in the field of meteorological information remote sensing such as rainfall information monitoring. At the same time, in the application process of the method provided by the present invention, the rain samples can be automatically classified only by inputting the rain attenuation characteristic quantities of each link, and then the rain intensity can be inverted by using the corresponding rain attenuation relationship, which greatly reduces the traditional Uncertainty of single link power-law rain attenuation relationship. This method can be applied to practical business as a new method for quantitative precipitation retrieval in microwave links.
以上所述的实施例仅是对本发明的优选方式进行描述，并非对本发明的范围进行限定，在不脱离本发明设计精神的前提下，本领域普通技术人员对本发明的技术方案做出的各种变形和改进，均应落入本发明权利要求书确定的保护范围内。The above-mentioned embodiments are only to describe the preferred mode of the present invention, but not to limit the scope of the present invention. Without departing from the design spirit of the present invention, those of ordinary skill in the art can Variations and improvements should fall within the protection scope determined by the claims of the present invention.
Claims (7)
Translated fromChinese
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110489466.2ACN113177360B (en) | 2021-05-06 | 2021-05-06 | Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110489466.2ACN113177360B (en) | 2021-05-06 | 2021-05-06 | Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113177360A CN113177360A (en) | 2021-07-27 |
| CN113177360Btrue CN113177360B (en) | 2022-04-29 |
Family
ID=76928176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110489466.2AActiveCN113177360B (en) | 2021-05-06 | 2021-05-06 | Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113177360B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113923618B (en)* | 2021-10-11 | 2022-05-31 | 中国人民解放军国防科技大学 | Three-dimensional microwave link networking detection near-ground three-dimensional rainfall experimental field system |
| CN113962083B (en)* | 2021-10-21 | 2022-11-11 | 中国人民解放军国防科技大学 | A method for constructing rainfall field with poor microwave information based on sparse representation |
| CN120129846A (en)* | 2022-12-13 | 2025-06-10 | 华为技术有限公司 | Method and device for measuring precipitation intensity and atmospheric humidity |
| CN116068559B (en)* | 2022-12-14 | 2023-09-12 | 广东省气象台(南海海洋气象预报中心、珠江流域气象台) | Combined raindrop spectrum and networked dual-polarization radar quantitative precipitation estimation method and device |
| CN119322955B (en)* | 2024-12-19 | 2025-04-08 | 中国电子科技集团公司第五十四研究所 | A method for calculating electromagnetic wave rain attenuation based on situation fusion |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106547037A (en)* | 2016-10-21 | 2017-03-29 | 中国人民解放军理工大学 | The united raindrop size distribution inverting of double frequency dual-polarization microwave link and type of precipitation recognition methodss |
| CN106650784A (en)* | 2016-11-04 | 2017-05-10 | 许继集团有限公司 | Feature clustering comparison-based power prediction method and device for photovoltaic power station |
| 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 |
| CN111666656A (en)* | 2020-05-09 | 2020-09-15 | 江苏微之润智能技术有限公司 | Rainfall estimation method and rainfall monitoring system based on microwave rainfall attenuation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10367677B2 (en)* | 2016-05-13 | 2019-07-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Network architecture, methods, and devices for a wireless communications network |
- 2021
- 2021-05-06CNCN202110489466.2Apatent/CN113177360B/enactiveActive
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106547037A (en)* | 2016-10-21 | 2017-03-29 | 中国人民解放军理工大学 | The united raindrop size distribution inverting of double frequency dual-polarization microwave link and type of precipitation recognition methodss |
| CN106650784A (en)* | 2016-11-04 | 2017-05-10 | 许继集团有限公司 | Feature clustering comparison-based power prediction method and device for photovoltaic power station |
| 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 |
| CN111666656A (en)* | 2020-05-09 | 2020-09-15 | 江苏微之润智能技术有限公司 | Rainfall estimation method and rainfall monitoring system based on microwave rainfall attenuation |
Non-Patent Citations (2)
| Title |
|---|
| Machine Learning Classification of Rainfall Types Based on the Differential Attenuation of Multiple Frequency Microwave Links;Kang Pu 等;《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》;20200103;第1-12页* |
| Rainfall estimation using a microwave link based on an improved rain-induced attenuation model;Kun Song 等;《Remote Sensing Letters》;20191231;第1057-1066页* |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113177360A (en) | 2021-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113177360B (en) | Rain intensity self-adaptive estimation system and method based on clustering rain attenuation relation | |
| CN109581546B (en) | Rainfall type identification method based on microwave link attenuation and polarization information | |
| Messer et al. | A new approach to precipitation monitoring: A critical survey of existing technologies and challenges | |
| US5440483A (en) | Process and device for evaluating the precipitations over an area of terrain | |
| CN118011358B (en) | A dual-band radar raindrop spectrum inversion method and device | |
| CN109255100B (en) | Urban rainfall inversion method based on microwave attenuation characteristic response fingerprint identification | |
| CN112906310A (en) | Method for optimizing BP neural network microwave remote sensing soil moisture inversion by considering firefly algorithm | |
| Kumar et al. | Truncated gamma drop size distribution models for rain attenuation in Singapore | |
| CN109936855B (en) | Method, device, equipment and medium for retrieving precipitation distribution using base station signals | |
| CN113923594B (en) | A method of distinguishing between sunny and rainy days based on time-division long-term evolution network | |
| CN116449331B (en) | Dust particle number concentration estimation method based on W-band radar and meteorological satellite | |
| CN116068559A (en) | Quantitative precipitation estimation method and device of joint raindrop spectrum and networked dual polarization radar | |
| Adewumi et al. | Characterization and modeling of vegetation effects on UHF propagation through a long forested channel | |
| Gharanjik et al. | Learning-based rainfall estimation via communication satellite links | |
| Bhat et al. | Correlating the ambient conditions and performance indicators of the LoRaWAN via surrogate Gaussian process-based bidirectional LSTM stacked autoencoder | |
| Nguyen et al. | Deep learning for path loss prediction at 7 GHz in urban environment | |
| Zhang et al. | An artificial intelligence radio propagation model based on geographical information | |
| Waheed et al. | Deepchannel: Robust multimodal outdoor channel model prediction in lte networks using deep learning | |
| Scognamiglio et al. | Deep learning for opportunistic rain estimation via satellite microwave links | |
| CN114675277B (en) | Near-ground atmosphere refractive index profile monitoring method based on commercial microwave return link | |
| CN113219465B (en) | A method for identification of aquatic products and automatic microwave frequency optimization based on polarization attenuation information | |
| CN111860562B (en) | Self-adaptive dry-wet distinguishing method based on multiple statistics of microwave link | |
| CN113111593B (en) | A Rainfall Fuzzy Clustering Method Based on Dual-Link Attenuation | |
| Asen et al. | A novel method for predicting site dependent specific rain attenuation of millimeter radio waves | |
| Ometan et al. | Six Years Result of Rainfall Rate Measurement at Covenant University, Southwest, Nigeria |
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 |