KR20240016466A - Server, method and computer program for modeling wireless network equipment - Google Patents

Abstract

The server that models wireless network equipment is a collection unit that collects sensor data including installation environment information about the location where the wireless network equipment is installed and shape information about the wireless network equipment, and models the shape of the wireless network equipment using sensor data. A shape modeling unit that performs comparative verification between modeling data on the shape of the wireless network equipment and image data on the wireless network equipment, and derives labeling information for the wireless network equipment through a learning model based on the comparative verification results. It may include an information derivation unit and a transmission unit that transmits labeling information to the user terminal.

Description

Server, method, and computer program for modeling wireless network equipment {SERVER, METHOD AND COMPUTER PROGRAM FOR MODELING WIRELESS NETWORK EQUIPMENT}

The present invention relates to servers, methods, and computer programs for modeling wireless network equipment.

Digital Twin technology refers to a technology that creates twins of objects in reality on a computer and predicts the results in advance by simulating situations that may occur in reality with a computer. Digital twins are digital objects that can be used to optimize the physical world, and can significantly improve operational performance and business processes, attracting attention as a technology that can solve not only manufacturing but also various industrial/social problems.

Meanwhile, in the case of object recognition and modeling for digital twins, previously, simple labeled data was stored based on data measured by images or sensors, and services were provided only for object modeling. Because it only provides modeling and storage functions for objects, it has the limitation that it is difficult to reflect various environmental information in the real world in modeling.

Korean Patent Publication No. 10-1726677 (registered on April 7, 2017)

The present invention is intended to solve the problems of the prior art described above, and aims to model the shape of wireless network equipment using sensor data including installation environment information about the place where the wireless network equipment is installed and shape information about the wireless network equipment. do.

In addition, the present invention seeks to provide labeling information about wireless network equipment to a user terminal through comparative verification between modeling data about the shape of the wireless network equipment and image data about the network equipment.

However, the technical challenges that this embodiment aims to achieve are not limited to the technical challenges described above, and other technical challenges may exist.

As a technical means for achieving the above-described technical problem, a server modeling wireless network equipment according to the first aspect of the present invention includes installation environment information about the place where the wireless network equipment is installed and shape information about the wireless network equipment. A collection unit that collects sensor data including; a shape modeling unit that models the shape of the wireless network equipment using the sensor data; Derive labeling information by performing comparative verification between modeling data about the shape of the wireless network equipment and image data about the wireless network equipment, and deriving labeling information about the wireless network equipment through a learning model based on the comparative verification results. wealth; And it may include a transmission unit that transmits the labeling information to the user terminal.

A method of modeling wireless network equipment performed by a network equipment modeling server according to a second aspect of the present invention includes sensor data including installation environment information about a place where the wireless network equipment is installed and shape information about the wireless network equipment. collecting; modeling the shape of the wireless network equipment using the sensor data; performing comparative verification between modeling data for the shape of the wireless network equipment and image data for the wireless network equipment, and deriving labeling information for the wireless network equipment through a learning model based on the comparative verification results; And it may include transmitting the labeling information to the user terminal.

A computer program stored in a computer-readable recording medium containing a sequence of instructions for modeling wireless network equipment according to the third aspect of the present invention, when executed by a computing device, provides installation environment information about the place where the wireless network equipment is installed, and Collect sensor data including shape information about the wireless network equipment, model the shape of the wireless network equipment using the sensor data, model data about the shape of the wireless network equipment, and model the shape of the wireless network equipment. It may include a sequence of commands for performing comparative verification between image data, deriving labeling information for the wireless network equipment through a learning model based on the comparative verification results, and transmitting the labeling information to a user terminal.

The above-described means for solving the problem are merely illustrative and should not be construed as limiting the present invention. In addition to the exemplary embodiments described above, there may be additional embodiments described in the drawings and detailed description of the invention.

According to one of the means for solving the problems of the present invention described above, the present invention determines the shape of the wireless network equipment using sensor data including installation environment information about the place where the wireless network equipment is installed and shape information about the wireless network equipment. Labeling information about the wireless network device can be provided to the user terminal through modeling and comparison verification between modeling data about the shape of the wireless network device and image data about the network device.

Through this, the present invention can increase the convenience of facility management and minimize the risk of equipment inspection by securing modeling data reflecting various installation environment information about the shape of wireless network equipment installed outdoors.

In addition, the present invention can build a digital twin site capable of remote recognition and management of wireless network equipment according to changes in the outdoor environment where wireless network equipment is installed. Through the digital twin site, users can check the shape of wireless network equipment installed in dangerous locations through modeling data on the wireless network equipment, which can help prevent industrial accidents.

1 is a block diagram of a network equipment modeling server according to an embodiment of the present invention.
Figure 2 is a diagram for explaining a method of training a learning model according to an embodiment of the present invention.
Figure 3 is a diagram showing simulation results for labeling information for wireless network equipment, according to an embodiment of the present invention.
Figure 4 is a flowchart showing a method for modeling wireless network equipment, according to an embodiment of the present invention.

Below, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected," but also the case where it is "electrically connected" with another element in between. . Additionally, when a part "includes" a certain component, this means that it may further include other components rather than excluding other components, unless specifically stated to the contrary.

In this specification, 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware.

In this specification, some of the operations or functions described as being performed by a terminal or device may instead be performed on a server connected to the terminal or device. Likewise, some of the operations or functions described as being performed by the server may also be performed on a terminal or device connected to the server.

Hereinafter, specific details for implementing the present invention will be described with reference to the attached configuration diagram or processing flow diagram.

Figure 1 is a block diagram of a network equipment modeling server 10 according to an embodiment of the present invention.

Referring to FIG. 1, the network equipment modeling server 10 includes a collection unit 100, a shape modeling unit 110, a labeling information derivation unit 120, a transmission unit 130, a pattern data extraction unit 140, and a learning unit. It may include a unit 150 and a storage unit 160. However, the network equipment modeling server 10 shown in FIG. 1 is only one implementation example of the present invention, and various modifications are possible based on the components shown in FIG. 1.

Hereinafter, FIG. 1 will be described with reference to FIGS. 2 and 3 together.

The collection unit 100 may collect sensor data including installation environment information about the location where the wireless network equipment is installed and shape information about the wireless network equipment.

The storage unit 160 may store the collected sensor data in a sensor database.

Here, the sensor data may be information measured through a sensor device (eg, a lidar sensor module) installed in a location where a site twin is to be built.

The collection unit 100 may collect image data about the shape of wireless network equipment captured by the image photographing device from the image photographing device.

The storage unit 160 may store collected image data about wireless network equipment in an image database.

The shape modeling unit 110 can model the shape of wireless network equipment using sensor data.

The shape modeling unit 110 may generate modeling data about the shape of the wireless network equipment from shape information about the wireless network equipment using an object detection algorithm.

For example, the shape modeling unit 110 may detect the shape of wireless network equipment, such as RU equipment, a rectifier, and a distribution board, from shape information about the wireless network equipment in sensor data stored in a sensor database. At this time, the shape modeling 110 may detect the shape of the wireless network equipment using an object detection algorithm to which a cloud point threshold (a threshold set by an administrator or a default threshold) is applied.

Additionally, the shape modeling unit 110 may perform modeling on the shape of at least one detected wireless network device. The storage unit 160 may organize modeling data about the shape of wireless network equipment into separate layers and store them in a modeling database.

The shape modeling unit 110 generates panoramic data about the place where the wireless network equipment is installed from installation environment information about the place where the wireless network equipment is installed, and uses the generated panoramic data to create modeling data about the installation environment of the wireless network equipment. can be created.

When performing modeling, if there is a point area where the wireless network equipment cannot be recognized in the installation environment information about the place where the wireless network equipment is installed, the shape modeling unit 110 simplifies the installation environment of the place where the wireless network equipment is installed and improves the recognition rate. For this purpose, point areas where wireless network equipment cannot be recognized can be judged as noise and removed. Additionally, the shape modeling unit 110 may generate modeling data for the installation environment after removing the area of the layer containing the shape of the wireless network device detected in the point area where the wireless network device cannot be recognized.

The storage unit 160 may organize modeling data about the installation environment of wireless network equipment by layer and store it in a modeling database.

The pattern data extractor 140 may extract pattern data from modeling data about the shape of the wireless network equipment and image data about the wireless network equipment.

The pattern data extractor 140 may extract first feature data from modeling data about the shape of the wireless network device and extract second feature data extracted from image data about the wireless network device.

The pattern data extraction unit 140 may extract pattern data common to each of the first feature data and the second feature data.

For example, the pattern data extraction unit 140 extracts each of modeling data about the shape of the first wireless network device searched from the modeling database and image data about the first wireless network device searched from the image database in the form of a Key-Value. After reconstruction, feature data for each can be extracted.

The pattern data extractor 140 may assume that the number of each Key-Value is T and allocate as many Key-Values as a separate F dimension. Here, in the F dimension, feature weights may be stored for each piece of data common to each feature data.

The pattern data extraction unit 140 may generate an attention value for the T*F dimension as each feature list. For example, referring to Figure 2, each feature list may be transmitted to the attention branch and attention mechanism included in the learning model 200, as shown in Figure 2.

Returning to FIG. 1, the learning unit 150 trains a learning model to derive label information for image data of wireless network equipment based on pattern data common to the first feature data and the second feature data. You can.

The learning model 200 can distinguish types of wireless network equipment by applying an attention mechanism between modeling data about the shape of the wireless network equipment and image data about the labeled wireless network equipment.

The learning model 200 can improve performance for classification of wireless network devices by relearning image data about labeled wireless network devices.

For example, referring to FIG. 2, the learning model 200 may be an attention CNN model composed of an attention convolution neural network.

The learning model 200 may include an attention branch and a perception branch.

The attention branch is composed of individual attention CNN layers, and the individual attention CNN layers can measure the score corresponding to the hidden state value of each layer using the attention value in the feature list for the input image. The attention branch can measure the attention score that reflects the attention value and score.

Recognition branches may be constructed based on data stored in the labeled image database and the training database.

The learning model 200 can perform learning and labeling tasks on integrated images through a Softmax process and store the results in a training database.

Returning to FIG. 1, the labeling information deriving unit 120 performs comparative verification between modeling data for the shape of the wireless network equipment and image data for the wireless network equipment, and based on the comparison and verification results, wireless network information is generated through a learning model. Labeling information about network equipment can be derived.

The transmitter 130 may transmit labeling information about wireless network equipment to the user terminal.

For example, referring to FIG. 3, the transmission unit 130 visualizes modeling data about the shape of the wireless network equipment and modeling data about the installation environment of the wireless network equipment and transmits the visualized result 300 to the user terminal. You can.

The user of the user terminal can monitor the modeling results of the shape of the wireless network equipment, the modeling results of the installation environment of the wireless network equipment, and the labeling information about the wireless network equipment through the screen of the user terminal.

If the modeled result is inaccurate, after changing the angle and position of the sensor device that measures the shape information and installation environment information for the wireless network equipment, the collection unit 100 determines that the wireless network equipment measured from the corresponding sensor device is Sensor data including installation environment information about the installed location and shape information about the wireless network equipment can be recollected. Additionally, the collection unit 100 may re-collect image data about the shape of wireless network equipment newly captured from an image capturing device whose shooting angle and position have changed depending on resolution.

The shape modeling unit 110 generates modeling data about the shape of the wireless network equipment using the re-collected sensor data, and the labeling information derivation unit 120 generates modeling data about the shape of the wireless network equipment and the re-collected wireless network equipment. Based on the comparative verification results between image data for network equipment, labeling information for wireless network equipment can be derived through a learning model. The transmission unit 130 may transmit labeling information about the derived wireless network equipment to the user terminal.

If the label information is incorrect or a new labeling task is required, the label work department (not shown) can perform a new labeling task by labeling the new label information input from the user to the image data for wireless network equipment extracted from the modeling database. there is.

The learning model can retrain labeled image data based on an image database.

Meanwhile, those skilled in the art will recognize the collection unit 100, the shape modeling unit 110, the labeling information derivation unit 120, the transmission unit 130, the pattern data extraction unit 140, the learning unit 150, and the storage unit 160. ) It will be fully understood that each may be implemented separately, or one or more of them may be integrated and implemented.

Figure 4 is a flowchart showing a method for modeling wireless network equipment, according to an embodiment of the present invention.

Referring to FIG. 4, in step S401, the network equipment modeling server 10 may collect sensor data including installation environment information about the place where the wireless network equipment is installed and shape information about the wireless network equipment.

In step S403, the network equipment modeling server 10 may model the shape of the wireless network equipment using sensor data.

In step S405, the network equipment modeling server 10 performs comparative verification between modeling data about the shape of the wireless network equipment and image data about the wireless network equipment, and based on the comparative verification results, the network equipment modeling server 10 performs comparative verification of the wireless network equipment through a learning model. Labeling information can be derived.

In step S407, the network equipment modeling server 10 may transmit labeling information to the user terminal.

In the above description, steps S401 to S407 may be further divided into additional steps or combined into fewer steps, depending on the implementation of the present invention. Additionally, some steps may be omitted or the order between steps may be changed as needed.

One embodiment of the present invention may also be implemented in the form of a recording medium containing instructions executable by a computer, such as program modules executed by a computer. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable media may include all computer storage media. Computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data.

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

10: Network equipment modeling server
100: Collection Department
110: Shape modeling unit
120: Labeling information derivation unit
130: Transmission unit
140: Pattern data extraction unit
150: Learning Department
160: storage unit

Claims (17)

In a server that models wireless network equipment,
a collection unit that collects sensor data including installation environment information about a place where the wireless network equipment is installed and shape information about the wireless network equipment;
a shape modeling unit that models the shape of the wireless network equipment using the sensor data;
Derive labeling information by performing comparative verification between modeling data about the shape of the wireless network equipment and image data about the wireless network equipment, and deriving labeling information about the wireless network equipment through a learning model based on the comparative verification results. wealth; and
A transmission unit that transmits the labeling information to the user terminal
A network equipment modeling server that includes a.
According to claim 1,
The sensor data is information measured through a sensor device installed in a location where a site twin is to be built. A network equipment modeling server.
According to claim 1,
A network equipment modeling server further comprising a pattern data extraction unit that extracts pattern data from modeling data for the shape of the wireless network equipment and image data for the wireless network equipment.
According to claim 3,
The pattern data extractor extracts first feature data from modeling data about the shape of the wireless network equipment, extracts second feature data extracted from image data about the wireless network device, and extracts the first feature data and the A network equipment modeling server that extracts the pattern data common to each second feature data.
According to claim 3,
A network equipment modeling server further comprising a learning unit that trains the learning model to derive label information for the image data based on the pattern data.
According to claim 1,
The shape modeling unit
A network equipment modeling server that generates modeling data about the shape of the wireless network equipment from shape information about the wireless network equipment using an object detection algorithm.
According to claim 6,
The shape modeling unit
A network equipment modeling server that generates panoramic data about a place where the wireless network equipment is installed from the installation environment information, and generates modeling data about the installation environment of the wireless network equipment using the generated panoramic data.
According to claim 7,
The transmission unit
A network equipment modeling server that visualizes modeling data about the shape of the wireless network equipment and modeling data about the installation environment of the wireless network equipment and transmits them to the user terminal.
In a method of modeling wireless network equipment performed by a network equipment modeling server,
collecting sensor data including installation environment information about a place where the wireless network equipment is installed and shape information about the wireless network equipment;
modeling the shape of the wireless network equipment using the sensor data;
performing comparative verification between modeling data for the shape of the wireless network equipment and image data for the wireless network equipment, and deriving labeling information for the wireless network equipment through a learning model based on the comparative verification results; and
Transmitting the labeling information to the user terminal
A network equipment modeling method comprising:
According to clause 9,
A network equipment modeling method in which the sensor data is information measured through a sensor device installed in a location where a site twin is to be built.
According to clause 9,
Network equipment modeling method further comprising extracting pattern data from modeling data for the shape of the wireless network equipment and image data for the wireless network equipment.
According to claim 11,
The step of extracting the pattern data is
extracting first feature data from modeling data about the shape of the wireless network equipment;
extracting second feature data extracted from image data for the wireless network equipment; and
A network equipment modeling method comprising extracting the pattern data common to each of the first feature data and the second feature data.
According to claim 11,
Network equipment modeling method further comprising training the learning model to derive label information for the image data based on the pattern data.
According to clause 9,
The step of modeling the shape of the wireless network equipment using the sensor data is
A network equipment modeling method comprising generating modeling data about the shape of the wireless network equipment from shape information about the wireless network equipment using an object detection algorithm.
According to claim 14,
generating panoramic data about the location where the wireless network equipment is installed from the installation environment information; and
A network equipment modeling method further comprising generating modeling data for an installation environment of the wireless network equipment using the generated panorama data.
According to claim 15,
Network equipment modeling method further comprising visualizing modeling data about the shape of the wireless network equipment and modeling data about the installation environment of the wireless network equipment and transmitting them to the user terminal.
A computer program stored on a computer-readable recording medium containing a sequence of instructions for modeling wireless network equipment, comprising:
When the computer program is executed by a computing device,
Collecting sensor data including installation environment information about the location where the wireless network equipment is installed and shape information about the wireless network equipment,
Model the shape of the wireless network equipment using the sensor data,
Perform comparative verification between modeling data for the shape of the wireless network equipment and image data for the wireless network equipment, and derive labeling information for the wireless network equipment through a learning model based on the comparative verification results,
A computer program stored on a computer-readable recording medium, comprising a sequence of instructions for transmitting the labeling information to a user terminal.

Images