CN112686483B - Early warning area identification method, early warning area identification device, computing equipment and computer storage medium - Google Patents

Abstract

The embodiment of the invention relates to the technical field of big data, and discloses a method, a device, computing equipment and a computer storage medium for identifying an early warning area, wherein the method comprises the following steps: collecting current network data and historical data, including: connecting the number of users, the relation of adjacent cells and longitude and latitude information; calculating the predicted value user number and the reference value user number of the cell according to the current network data and the connection user number of the cell in the historical data; inputting the predicted value user number and the reference value user number of the cell into a linear regression model to obtain an early warning cell; and converging the early warning cells and identifying the coverage areas by using a Thiessen polygon according to the neighbor relation and the longitude and latitude information, and outputting the early warning areas. Through the mode, the embodiment of the invention can accurately find the early warning cell in advance, provide effective support for high-load cell analysis, improve the running quality of the whole network and optimize the use perception of a user to the communication network.

Description

Warning area identification method, device, computing equipment and computer storage medium

Technical Field

The embodiments of the present invention relate to the field of big data technology, and specifically to a warning area identification method, device, computing equipment and computer storage medium.

Background Art

At present, the identification of areas where sudden crowds gather mainly relies on the monitoring of background indicators of the entire network. By discovering high-load cells within a time period, relying on the latitude and longitude information of base stations, and manually combining electronic maps, we output warning areas and carry out optimization work.

There are two main aspects in the identification of sudden crowd gathering areas: one is the identification of early warning cells, and the other is the identification of activity areas. In the identification of early warning cells in sudden major events, it is necessary to first monitor the indicators of the entire network manually, and then manually select the cells in the entire network according to the high-load rules to determine the high-load cells. The whole process takes a long time, which is both labor-intensive and time-effective. When identifying sudden activity areas, the base station location is first determined based on the latitude and longitude, engineering parameters, and electronic maps of the high-load cells, and then the activity area is determined based on the site category (macro base station, indoor distribution) and optimization experience. The entire process is manually handled, and efficiency and accuracy are difficult to guarantee.

It can be seen that at present, the identification of warning cells and activity areas needs to be done manually, which is time-consuming and laborious. In addition, the timeliness of warning cells and the accuracy of activity areas are not high. If the activity information is missed, it will seriously affect the customer's perception of the use of the communication network, causing unnecessary network complaints and negative public opinion. Therefore, how to accurately and timely predict sudden activities and areas has become a technical problem that needs to be solved urgently.

Summary of the invention

In view of the above problems, embodiments of the present invention provide a warning area identification method, apparatus, computing device and computer storage medium, which overcome the above problems or at least partially solve the above problems.

According to one aspect of an embodiment of the present invention, a method for identifying an early warning area is provided, the method comprising: collecting existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculating the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputting the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain an early warning cell; applying Thiessen polygons to converge and identify the coverage area of the early warning cell according to the neighboring area relations and the longitude and latitude information, and outputting an early warning area.

In an optional manner, the calculating of the predicted number of users and the benchmark number of users based on the existing network data and the number of connected users of the cell in the historical data includes: obtaining the predicted number of users of the cell using a long short-term memory network algorithm based on the existing network data and the number of connected users of the cell in the historical data; and calculating the benchmark number of users of the cell using a K-MEANS clustering algorithm and Apollonius theorem based on the number of connected users of the cell in the historical data.

In an optional manner, the method of obtaining the predicted number of users of the cell based on the number of connected users of the cell in the existing network data and the historical data using a long short-term memory network algorithm also includes: using a long short-term memory network algorithm to slice the existing network data and the historical data at a granularity of 15 minutes to establish a model, and correcting and optimizing the model based on the historical data; and applying the model to predict the predicted number of users for multiple time periods in the future at a granularity of 15 minutes based on the number of connected users of the cell in the existing network data and the historical data.

In an optional manner, the calculating the reference value user number of the cell according to the number of connected users of the cell in the historical data using a K-MEANS clustering algorithm and Apollonius theorem includes:

According to the historical data of 15-minute granularity, 3 points are randomly selected, and the number of connected users of each 15-minute granularity is set as clusters C1, C2, and C3. The historical data is aggregated into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

Connect the three centroids to form a triangle; calculate the number of connected users corresponding to the intersection points of the three medians of the triangle according to the Apollonius theorem, and determine it as the benchmark number of users of the cell.

In an optional manner, the predicted number of users and the benchmark number of users of the cell are input into a linear regression model to obtain a warning cell, including: establishing a linear regression model based on the number of daily connected users during major historical events and the number of connected users during the events; inputting the predicted number of users and the benchmark number of users into the linear regression model to determine that the cell that meets the continuous increase is the warning cell.

In an optional manner, the application of Thiessen polygons to converge the warning cell and identify the coverage area according to the neighboring area relationship and the longitude and latitude information, and output the warning area, includes: obtaining information of the warning cell, at least including longitude and latitude information and neighboring area relationship; finding the base station of the warning cell according to the longitude and latitude information, and converging the base stations according to the longitude and latitude information and the neighboring area relationship of the base station; constructing a Delaunay triangulation network of all base stations according to the longitude and latitude information of the base station, connecting the centers of the circumscribed circles of adjacent triangles of each base station, and obtaining the coverage areas of all base stations; outputting the warning area according to the neighboring area relationship of the warning cell and the coverage area of the base station.

In an optional manner, the warning area is output based on the neighboring cell relationship of the base station of the warning cell and the coverage area of the base station, including: if it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell and the surrounding base stations with neighboring cell relationships are converged, and the warning area is output based on the coverage area of the base station; if it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined externally, the coverage areas of all warning base stations are crawled by a network crawler, and the warning area is output.

According to another aspect of an embodiment of the present invention, a warning area identification device is provided, the device comprising: a data acquisition unit, used to collect existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; a calculation unit, used to calculate the predicted number of users and the benchmark number of users of the cell according to the existing network data and the number of connected users of the cell in the historical data; a warning cell acquisition unit, used to input the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; a warning area identification unit, used to apply Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and output a warning area.

According to another aspect of an embodiment of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the steps of the above-mentioned warning area identification method.

According to another aspect of an embodiment of the present invention, a computer storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables the processor to execute the steps of the above-mentioned warning area identification method.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below.

BRIEF DESCRIPTION OF THE DRAWINGS

Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

FIG1 is a schematic diagram showing a flow chart of a warning area identification method provided by an embodiment of the present invention;

FIG2 is a schematic flow chart of step S12 of the warning area identification method provided by an embodiment of the present invention;

FIG3 shows a schematic diagram of a model for calculating the number of users of a predicted value in the warning area identification method provided by an embodiment of the present invention;

FIG4 is a schematic diagram showing a method for calculating the number of users of a reference value in the warning area identification method provided by an embodiment of the present invention;

FIG5 is a schematic diagram showing a method for determining a base station coverage area in a warning area identification method provided by an embodiment of the present invention;

FIG6 shows a schematic diagram of the structure of a warning area identification device provided by an embodiment of the present invention;

FIG. 7 shows a schematic diagram of the structure of a computing device provided by an embodiment of the present invention.

DETAILED DESCRIPTION

The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

FIG1 is a schematic diagram showing a flow chart of a warning area identification method provided by an embodiment of the present invention. As shown in FIG1 , the warning area identification method includes:

Step S11: Collect current network data and historical data, including: number of connected users, neighboring area relations, and longitude and latitude information.

In the embodiment of the present invention, the current network data collection can be completed by automatic docking with the whole network background indicator monitoring platform, and the main information of the collected data includes: time granularity, base station number (eNodeBID), cell number (CellId), number of connected users, longitude, latitude, and neighboring area relationship information. The historical data collection can be completed by automatic docking with the whole network background indicator monitoring platform, and the main information of the collected data includes: time granularity, eNodeBID, CellId, number of connected users, longitude, latitude, neighboring area relationship information, etc.

Then, the data of major events are cleaned from the collected live network data and historical data. The cleaned live network data and historical data include: time granularity during historical daily and event periods, eNodeBID, CellId, number of connected users, etc. The cleaned live network data and historical data are further cleaned in different time periods of the same cell to remove abnormal data, and the daily and event data of major events are collated and aggregated, and output in a format that can be used for analysis and modeling applications, and stored.

Step S12: Calculate the predicted number of users and the reference number of users of the cell according to the existing network data and the number of connected users of the cell in the historical data.

In step S12, as shown in FIG2 , it includes:

Step S121: according to the number of connected users of the cell in the existing network data and the historical data, a long short-term memory network algorithm is used to obtain the predicted number of users of the cell.

Specifically, the real-time continuous number of users in the cell at a 15-minute granularity in the existing network data and the number of connected users in the cell at a 15-minute granularity in the same period in the historical data are obtained. The Long Short-Term Memory (LSTM) algorithm in machine learning is used to slice the existing network data and the historical data in 15-minute time periods according to the granularity of one week to establish a model, and the model is modified and optimized according to the historical data; then, the model is applied to predict the predicted number of users in multiple future time periods at a 15-minute granularity according to the number of connected users in the cell in the existing network data and the historical data.

Figure 3 is a schematic diagram of the model used to calculate the number of users with a predicted value. First, the current network data and some historical data are time-sliced and encoded with a 15-minute time interval according to a weekly granularity, and embedded into the input end of the LSTM algorithm model. Some historical data are normalized, and batch normalized (Batchnorm) is performed, then output through the first linear layer, activated, and merged with the time-slice encoding result as the input layer x_in of the LSTM algorithm model. After the LSTM algorithm operation, it is output through the output layer n_out and the second linear layer output Y_out.

Based on the historical data and current network data combined with the model, the predicted number of users in the cell in multiple time periods in the future is predicted at a 15-minute granularity.

Step S122: Calculate the reference value user number of the cell using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data.

According to the historical data of 15-minute granularity, 3 points are randomly selected, and the number of connected users of each 15-minute granularity is set as clusters C1, C2, and C3. The historical data is aggregated into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

The three centroids are calculated and connected to form a triangle.

As shown in FIG4 , the number of connected users corresponding to the intersection points of the three medians of the triangle is calculated according to the Apollonius theorem and is determined as the reference value number of users of the cell.

Step S13: input the predicted value user number and the reference value user number of the cell into a linear regression model to obtain a warning cell.

Specifically, a linear regression model is established based on the number of daily connected users during major historical events and the number of connected users during the events; the predicted value number of users and the benchmark value number of users are input into the linear regression model, and the cell that meets the continuous increase is determined as the warning cell.

In the embodiment of the present invention, the number of daily connected users of historical major events and the number of connected users during the event are selected, and a linear regression model is established through the number of daily connected users of historical major events and the number of connected users during the event. In the linear regression model, the number of daily connected users is set to x, and the number of active connected users is set to y. The established linear regression model is:

y＝h _θ (x)＝x·θ ^T

The data is corrected and calculated based on the number of connected users during multiple activity periods as follows:

Where m is the number of samples, θ is the relationship matrix, L is the difference between the calculated and actual number of active connected users, and i and j are positive integers.

according to Perform regression, with a as the coefficient, and finally determine the most accurate linear regression model.

The embodiment of the present invention compares the data increase of multiple periods with the linear regression model and sets a correction value in each increase, preferably plus or minus 3%. The information to be input is shown in Table 1:

Table 1 Information input into the linear regression model

By comparing the results, if the cell meets the continuous increase condition, it will be a warning cell. The continuous increase condition means that the difference between the predicted value of users and the reference value of users continues to increase.

Step S14: applying Thiessen polygons to converge the warning cells and identify the coverage area according to the neighboring cell relationship and the longitude and latitude information, and outputting the warning area.

In step S14, the information of the warning cell is obtained, including at least the longitude and latitude information and the neighboring cell relationship; the base station of the warning cell is found according to the longitude and latitude information, and the base station is converged according to the longitude and latitude information and the neighboring cell relationship of the base station; the Delaunay triangulation of all base stations is constructed according to the longitude and latitude information of the base station by applying the algorithm in the Thiessen polygon, and the circumscribed circle centers of adjacent triangles of each base station are connected to obtain the coverage area of all base stations; the warning area is output according to the neighboring cell relationship of the warning cell and the coverage area of the base station.

In an embodiment of the present invention, when obtaining the coverage area of all base stations, as shown in FIG5 , the algorithm in the Thiessen polygon is applied according to the latitude and longitude information of the base station to sort the triangles adjacent to a certain base station in a clockwise or counterclockwise direction, and a certain base station is set as o. Find a triangle with base station o as the vertex, set as triangle A; take another vertex of triangle A except base station o, set as base station a, then another vertex can also be found, that is, base station f; then the next triangle must be of as the edge, that is, triangle F; the other vertex of triangle F is base station e, then the next triangle is oe as the edge; repeat this until returning to the oa edge, and calculate the center of the circumscribed circle of each triangle, and record it. According to the adjacent triangles of each base station, connect the centers of the circumscribed circles of these adjacent triangles to obtain polygon a'b'c'd'e'f', and the area defined by the polygon a'b'c'd'e'f' is the coverage area of base station o.

In calculating the coverage area of the base station, assume that there is a set of discrete points (x _j ,y _j ) on the plane area B, i = 1, 2, 3, ..., k, k is the number of discrete points. If the area B is divided into k adjacent polygons by a set of straight line segments, it must satisfy:

(1) Each polygon contains only one discrete point.

(2) If any point (x ¹ ,y ¹ ) on region B is inside the polygon containing the discrete point (x _j ,y _j ), the following inequality always holds when (i≠j):

(3) If a point (x ¹ ,y ¹ ) is located on the common edge of two polygons containing discrete points (x _j ,y _j ), then the following equation holds:

In an embodiment of the present invention, when the warning area is identified, after determining the coverage areas of all base stations, if it is determined based on the longitude and latitude of the base stations that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell and the surrounding base stations with neighboring cell relationships are converged, and the warning area is output based on the coverage area of the base station.

If it is determined based on the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined by expansion, and the coverage area of all warning base stations is crawled by a network crawler to output the warning area. Specifically, the final warning base station is determined according to the rule of continued expansion. That is, the base stations of the warning cell and the base stations that have neighboring area relationships around the base station of the warning cell are gathered until there is no warning cell around the base station of any warning cell, thereby determining the final warning base station. Then, the network crawler technology is applied to crawl the base station boundary according to the warning base station list and the algorithm in the aforementioned Thiessen polygon, and the final warning area is output.

The embodiment of the present invention utilizes a machine learning method for intelligent early warning, and can perform intelligent monitoring and early warning of communities throughout the province, with a wide warning range; compared with the traditional method of manually circling communities based on maps, the embodiment of the present invention can automatically discover early warning communities in advance, and automatically aggregate them, to achieve early prediction with high efficiency and low cost; the embodiment of the present invention can be further promoted based on the intelligent early warning method for major events, and can associate information such as changes in the number of users and traffic in the community, identify high-value areas, and provide effective support for high-load community analysis, thereby improving the overall network operation quality.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

FIG6 shows a schematic diagram of the structure of a warning area identification device according to an embodiment of the present invention. As shown in FIG6 , the warning area identification device comprises: a data acquisition unit 601, a calculation unit 602, a warning cell acquisition unit 603 and a warning area identification unit 604. Among them:

The data collection unit 601 is used to collect current network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; the calculation unit 602 is used to calculate the predicted number of users and the benchmark number of users of the cell according to the current network data and the number of connected users of the cell in the historical data; the early warning cell acquisition unit 603 is used to input the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain the early warning cell; the early warning area identification unit 604 is used to apply Thiessen polygons to converge and identify the coverage area of the early warning cell according to the neighboring area relations and the longitude and latitude information, and output the early warning area.

In an optional manner, the calculation unit 602 is used to: obtain the predicted number of users of the cell by using a long short-term memory network algorithm based on the number of connected users of the cell in the existing network data and the historical data; and calculate the baseline number of users of the cell by using a K-MEANS clustering algorithm and Apollonius theorem based on the number of connected users of the cell in the historical data.

In an optional manner, the computing unit 602 is also used to: use a long short-term memory network algorithm to slice the existing network data and the historical data at a granularity of 15 minutes to establish a model, and modify and optimize the model based on the historical data; apply the model to predict the predicted number of users in multiple time periods in the future at a granularity of 15 minutes based on the number of connected users in the cell in the existing network data and the historical data.

In an optional manner, the calculation unit 602 is further used to: randomly select 3 points according to the historical data at a 15-minute granularity, set the number of connected users at each 15-minute granularity to clusters C1, C2, and C3, and aggregate the historical data into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

Connect the three centroids to form a triangle; calculate the number of connected users corresponding to the intersection points of the three medians of the triangle according to the Apollonius theorem, and determine it as the benchmark number of users of the cell.

In an optional manner, the warning cell acquisition unit 603 is used to: establish a linear regression model based on the number of daily connected users during historical major events and the number of connected users during the event; input the predicted value number of users and the benchmark value number of users into the linear regression model to determine that the cell that meets the continuous increase is the warning cell.

In an optional manner, the warning area identification unit 604 is used to: obtain information on the warning cell, including at least longitude and latitude information and neighboring cell relationships; find the base station of the warning cell based on the longitude and latitude information, and perform base station aggregation based on the longitude and latitude information and neighboring cell relationships of the base station; construct a Delaunay triangulation network of all base stations based on the longitude and latitude information of the base station using the algorithm in the Thiessen polygons, connect the centers of the circumscribed circles of adjacent triangles of each base station, and obtain the coverage areas of all base stations; output the warning area based on the neighboring cell relationships of the warning cell and the coverage areas of the base stations.

In an optional manner, the warning area identification unit 604 is used to: if it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, then the base station of the warning cell and the surrounding base stations with neighboring cell relationships are converged, and the warning area is output according to the coverage area of the base station; if it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, then the warning base station is determined externally, the coverage areas of all warning base stations are crawled by a network crawler, and the warning area is output.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the warning area identification method in any of the above method embodiments.

The executable instructions may be specifically used to cause the processor to perform the following operations:

Collect current network data and historical data, including: number of connected users, neighboring area relationships, and longitude and latitude information;

Calculate the predicted number of users and the reference number of users of the cell according to the existing network data and the number of connected users of the cell in the historical data;

Input the predicted value user number and the reference value user number of the cell into a linear regression model to obtain a warning cell;

Thiessen polygons are used to converge the warning cells and identify the coverage areas according to the neighboring area relationships and the longitude and latitude information, and the warning areas are output.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Acquire the predicted value of the number of users of the cell using a long short-term memory network algorithm according to the number of connected users of the cell in the existing network data and the historical data;

The reference value user number of the cell is calculated using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Using a long short-term memory network algorithm to slice the existing network data and the historical data at a weekly granularity of 15 minutes to establish a model, and modifying and optimizing the model based on the historical data;

The model is applied to predict the predicted number of users in multiple future time periods at a granularity of 15 minutes based on the number of connected users in the cell in the existing network data and the historical data.

In an optional manner, the executable instruction causes the processor to perform the following operations:

According to the historical data of 15-minute granularity, 3 points are randomly selected, and the number of connected users of each 15-minute granularity is set as clusters C1, C2, and C3. The historical data is aggregated into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

Connect the 3 centroids to form a triangle;

The number of connected users corresponding to the intersection points of the three medians of the triangle is calculated according to the Apollonius theorem and determined as the reference value number of users of the cell.

In an optional manner, the executable instruction causes the processor to perform the following operations:

A linear regression model was established based on the number of daily connected users during major historical events and the number of connected users during the event.

The predicted value number of users and the reference value number of users are input into the linear regression model to determine the cell that meets the continuous increase requirement as the warning cell.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Acquire information of the warning cell, including at least longitude and latitude information and neighboring cell relationship;

Finding the base station of the warning cell according to the longitude and latitude information, and performing base station aggregation according to the longitude and latitude information of the base station and the neighboring cell relationship;

According to the latitude and longitude information of the base stations, the Delaunay triangulation network of all base stations is constructed by applying the algorithm in the Thiessen polygon, and the centers of the circumscribed circles of the adjacent triangles of each base station are connected to obtain the coverage areas of all base stations;

The warning area is output according to the neighboring cell relationship of the warning cell and the coverage area of the base station.

In an optional manner, the executable instruction causes the processor to perform the following operations:

If it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell is aggregated with base stations having neighboring cell relationships therearound, and the warning area is output according to the coverage area of the base station;

If it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined externally, and the coverage areas of all the warning base stations are crawled by a network crawler to output the warning area.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

An embodiment of the present invention provides a computer program product, which includes a computer program stored on a computer storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the warning area identification method in any of the above method embodiments.

The executable instructions may be specifically used to cause the processor to perform the following operations:

Collect current network data and historical data, including: number of connected users, neighboring area relationships, and longitude and latitude information;

Calculate the predicted number of users and the reference number of users of the cell according to the existing network data and the number of connected users of the cell in the historical data;

Input the predicted value user number and the reference value user number of the cell into a linear regression model to obtain a warning cell;

Thiessen polygons are used to converge the warning cells and identify the coverage areas according to the neighboring area relationships and the longitude and latitude information, and the warning areas are output.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Acquire the predicted value of the number of users of the cell using a long short-term memory network algorithm according to the number of connected users of the cell in the existing network data and the historical data;

The reference value user number of the cell is calculated using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Using a long short-term memory network algorithm to slice the existing network data and the historical data at a weekly granularity of 15 minutes to establish a model, and modifying and optimizing the model based on the historical data;

The model is applied to predict the predicted number of users in multiple future time periods at a granularity of 15 minutes based on the number of connected users in the cell in the existing network data and the historical data.

In an optional manner, the executable instruction causes the processor to perform the following operations:

According to the historical data of 15-minute granularity, 3 points are randomly selected, and the number of connected users of each 15-minute granularity is set as clusters C1, C2, and C3. The historical data is aggregated into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

Connect the 3 centroids to form a triangle;

The number of connected users corresponding to the intersection points of the three medians of the triangle is calculated according to the Apollonius theorem and determined as the reference value number of users of the cell.

In an optional manner, the executable instruction causes the processor to perform the following operations:

A linear regression model was established based on the number of daily connected users during major historical events and the number of connected users during the event.

The predicted value number of users and the reference value number of users are input into the linear regression model to determine the cell that meets the continuous increase requirement as the warning cell.

In an optional manner, the executable instruction causes the processor to perform the following operations:

Acquire information of the warning cell, including at least longitude and latitude information and neighboring cell relationship;

Finding the base station of the warning cell according to the longitude and latitude information, and performing base station aggregation according to the longitude and latitude information of the base station and the neighboring cell relationship;

According to the latitude and longitude information of the base stations, the Delaunay triangulation network of all base stations is constructed by applying the algorithm in the Thiessen polygon, and the centers of the circumscribed circles of the adjacent triangles of each base station are connected to obtain the coverage areas of all base stations;

The warning area is output according to the neighboring cell relationship of the warning cell and the coverage area of the base station.

In an optional manner, the executable instruction causes the processor to perform the following operations:

If it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell is aggregated with base stations having neighboring cell relationships therearound, and the warning area is output according to the coverage area of the base station;

If it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined externally, and the coverage areas of all the warning base stations are crawled by a network crawler to output the warning area.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

FIG. 7 shows a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. The specific embodiment of the present invention does not limit the specific implementation of the device.

As shown in FIG. 7 , the computing device may include: a processor (processor) 702 , a communications interface (Communications Interface) 704 , a memory (memory) 706 , and a communication bus 708 .

The processor 702, the communication interface 704, and the memory 706 communicate with each other via the communication bus 708. The communication interface 704 is used to communicate with other devices such as a client or other server network elements. The processor 702 is used to execute the program 710, which can specifically execute the relevant steps in the above-mentioned warning area identification method embodiment.

Specifically, the program 710 may include program codes, which include computer operation instructions.

The processor 702 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the device may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

The memory 706 is used to store the program 710. The memory 706 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

The program 710 may be specifically configured to enable the processor 702 to perform the following operations:

Collect current network data and historical data, including: number of connected users, neighboring area relationships, and longitude and latitude information;

Calculate the predicted number of users and the reference number of users of the cell according to the existing network data and the number of connected users of the cell in the historical data;

Input the predicted value user number and the reference value user number of the cell into a linear regression model to obtain a warning cell;

Thiessen polygons are used to converge the warning cells and identify the coverage areas according to the neighboring area relationships and the longitude and latitude information, and the warning areas are output.

In an optional manner, the program 710 enables the processor to perform the following operations:

Acquire the predicted value of the number of users of the cell using a long short-term memory network algorithm according to the number of connected users of the cell in the existing network data and the historical data;

The reference value user number of the cell is calculated using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data.

In an optional manner, the program 710 enables the processor to perform the following operations:

Using a long short-term memory network algorithm to slice the existing network data and the historical data at a weekly granularity of 15 minutes to establish a model, and modifying and optimizing the model based on the historical data;

The model is applied to predict the predicted number of users in multiple future time periods at a granularity of 15 minutes based on the number of connected users in the cell in the existing network data and the historical data.

In an optional manner, the program 710 enables the processor to perform the following operations:

According to the historical data of 15-minute granularity, 3 points are randomly selected, and the number of connected users of each 15-minute granularity is set as clusters C1, C2, and C3. The historical data is aggregated into 3 regions E according to the K-Means algorithm:

Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3,

Connect the 3 centroids to form a triangle;

The number of connected users corresponding to the intersection points of the three medians of the triangle is calculated according to the Apollonius theorem and determined as the reference value number of users of the cell.

In an optional manner, the program 710 enables the processor to perform the following operations:

A linear regression model was established based on the number of daily connected users during major historical events and the number of connected users during the event.

The predicted value number of users and the reference value number of users are input into the linear regression model to determine the cell that meets the continuous increase requirement as the warning cell.

In an optional manner, the program 710 enables the processor to perform the following operations:

Acquire information of the warning cell, including at least longitude and latitude information and neighboring cell relationship;

Finding the base station of the warning cell according to the longitude and latitude information, and performing base station aggregation according to the longitude and latitude information of the base station and the neighboring cell relationship;

According to the latitude and longitude information of the base stations, the Delaunay triangulation network of all base stations is constructed by applying the algorithm in the Thiessen polygon, and the centers of the circumscribed circles of the adjacent triangles of each base station are connected to obtain the coverage areas of all base stations;

The warning area is output according to the neighboring cell relationship of the warning cell and the coverage area of the base station.

In an optional manner, the program 710 enables the processor to perform the following operations:

If it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell is aggregated with base stations having neighboring cell relationships therearound, and the warning area is output according to the coverage area of the base station;

If it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined externally, and the coverage areas of all the warning base stations are crawled by a network crawler to output the warning area.

The embodiment of the present invention collects existing network data and historical data, including: the number of connected users, neighboring area relations, and longitude and latitude information; calculates the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; inputs the predicted number of users and the benchmark number of users of the cell into a linear regression model to obtain a warning cell; applies Thiessen polygons to converge and identify the coverage area of the warning cell according to the neighboring area relations and the longitude and latitude information, and outputs the warning area, so as to accurately discover the warning cell in advance, provide effective support for high-load cell analysis, improve the overall network operation quality, and optimize the user's perception of the use of the communication network.

The algorithm or display provided here is not inherently related to any specific computer, virtual system or other equipment. Various general systems can also be used together with the teaching based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the embodiment of the present invention is not directed to any specific programming language yet. It should be understood that various programming languages can be utilized to realize the content of the present invention described here, and the description of the above specific language is for disclosing the best mode of the present invention.

In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and/or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the following claims, any one of the claimed embodiments may be used in any combination.

It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim that lists a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims (7)

1. A method for identifying an early warning area, characterized in that the method comprises: Collect current network data and historical data, including: number of connected users, neighboring area relationships, and longitude and latitude information; The method comprises: calculating the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data by using a long short-term memory network algorithm to obtain the predicted number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; calculating the benchmark number of users of the cell according to the number of connected users of the cell in the historical data by using a K-MEANS clustering algorithm and Apollonius theorem; wherein, obtaining the predicted number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data by using a long short-term memory network algorithm comprises: using a long short-term memory network algorithm to perform a 15-minute clustering of the connected users of the cell according to a weekly granularity; The existing network data and the historical data are sliced to establish a model, and the model is modified and optimized according to the historical data; the model is applied to predict the predicted value user number in multiple time periods in the future with a granularity of 15 minutes according to the number of connected users of the cell in the existing network data and the historical data; the benchmark value user number of the cell is calculated using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data, including: randomly selecting 3 points according to the historical data with a granularity of 15 minutes, setting the number of connected users of each granularity of 15 minutes to clusters C1, C2, and C3, and aggregating the historical data into 3 regions E according to the K-Means algorithm: Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3, Connect the three centroids to form a triangle; calculate the number of connected users corresponding to the intersection points of the three midlines of the triangle according to the Apollonius theorem, and determine it as the benchmark number of users of the cell; Input the predicted value user number and the reference value user number of the cell into a linear regression model to obtain a warning cell; Thiessen polygons are used to converge the warning cells and identify the coverage areas according to the neighboring area relationships and the longitude and latitude information, and the warning areas are output. 2. The method according to claim 1, characterized in that the step of inputting the predicted value user number and the reference value user number of the cell into a linear regression model to obtain the warning cell comprises: A linear regression model was established based on the number of daily connected users during major historical events and the number of connected users during the event. The predicted value number of users and the reference value number of users are input into the linear regression model to determine the cell that meets the continuous increase requirement as the warning cell. 3. The method according to claim 1 is characterized in that the application of Thiessen polygons to converge the warning cells and identify the coverage area according to the neighboring area relationship and the longitude and latitude information, and output the warning area, comprises: Acquire information of the warning cell, including at least longitude and latitude information and neighboring cell relationship; Finding the base station of the warning cell according to the longitude and latitude information, and performing base station aggregation according to the longitude and latitude information of the base station and the neighboring cell relationship; According to the latitude and longitude information of the base stations, the Delaunay triangulation network of all base stations is constructed by applying the algorithm in the Thiessen polygon, and the centers of the circumscribed circles of the adjacent triangles of each base station are connected to obtain the coverage areas of all base stations; The warning area is output according to the neighboring cell relationship of the warning cell and the coverage area of the base station. 4. The method according to claim 3, characterized in that the step of outputting the early warning area according to the neighboring cell relationship of the base station of the early warning cell and the coverage area of the base station comprises: If it is determined according to the longitude and latitude of the base station that there is no warning cell around the base station of any of the warning cells, the base station of the warning cell is aggregated with base stations having neighboring cell relationships therearound, and the warning area is output according to the coverage area of the base station; If it is determined according to the longitude and latitude of the base station that there is a warning cell around the base station of any of the warning cells, the warning base station is determined externally, and the coverage areas of all the warning base stations are crawled by a network crawler to output the warning area. 5. A warning area identification device, characterized in that the device comprises: Data collection unit, used to collect current network data and historical data, including: number of connected users, neighboring area relations, and longitude and latitude information; A calculation unit is used to calculate the predicted number of users and the benchmark number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data, including: using a long short-term memory network algorithm to obtain the predicted number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data; using a K-MEANS clustering algorithm and Apollonius theorem to calculate the benchmark number of users of the cell according to the number of connected users of the cell in the historical data; wherein, using a long short-term memory network algorithm to obtain the predicted number of users of the cell according to the number of connected users of the cell in the existing network data and the historical data includes: using a long short-term memory network algorithm to calculate the benchmark number of users of the cell according to the weekly granularity of 15 minutes The present network data and the historical data are sliced to establish a model, and the model is modified and optimized according to the historical data; the model is applied to predict the predicted value user number in multiple time periods in the future with a granularity of 15 minutes according to the number of connected users of the cell in the present network data and the historical data; the benchmark value user number of the cell is calculated by using the K-MEANS clustering algorithm and Apollonius theorem according to the number of connected users of the cell in the historical data, including: randomly selecting 3 points according to the historical data with a granularity of 15 minutes, setting the number of connected users of each granularity of 15 minutes as clusters C1, C2, and C3, and aggregating the historical data into 3 regions E according to the K-Means algorithm: Among them, the centroid μi is the mean vector of cluster Ci, k,i = 1, 2, 3, Connect the three centroids to form a triangle; calculate the number of connected users corresponding to the intersections of the three midlines of the triangle according to the Apollonius theorem, and determine it as the benchmark number of users of the cell; An early warning cell acquisition unit, used for inputting the predicted value user number and the reference value user number of the cell into a linear regression model to acquire an early warning cell; The warning area identification unit is used to apply Thiessen polygons to converge and identify the coverage area of the warning cells according to the neighboring area relationship and the longitude and latitude information, and output the warning area. 6. A computing device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the steps of the warning area identification method according to any one of claims 1-4. 7. A computer storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute the steps of the warning area identification method according to any one of claims 1 to 4.