WO2022045458A1 - Method and device for transmitting content from core node controlling plurality of edge nodes - Google Patents

Abstract

A device for transmitting content from a core node controlling a plurality of edge nodes according to an embodiment may comprise: a communication module which receives a user's content request information and the user's location and route information by means of edge nodes connected to a short-range communication AP of a transportation means that the user is riding; a content division module which determines a plurality of content transmission edge nodes with which to transmit content to the short-range communication AP from among the plurality of edge nodes on the basis of the user's location and route information, and then divides the content into a plurality of pieces according to the number of content transmission edge nodes to generate divided pieces of content, and transmits the divided pieces of content to the respective content transmission edge nodes according to the user's route information; and a control module which controls the content transmission edge nodes so that the divided pieces of content can be transmitted to a user terminal by means of the short-range communication AP connected to the user terminal.

Description

Method and apparatus for content transmission in a core node that controls a plurality of edge nodes

The present invention relates to a method and apparatus for content transmission in a core node that controls a plurality of edge nodes, and more specifically, more quickly by utilizing a short-range communication AP of the transportation means in which a user is riding and a CDN server in the edge node. It is an invention about a technology that can provide content to users.

Recently, with the continuous development of mobile terminals and mobile communication technologies, the purpose and field of use of the mobile terminals are becoming more and more diversified. That is, in the past, mobile terminal users mainly used the mobile terminal for calls or message transmission, but in recent years, it is not only for calls and messages, but also for movies or music, video lectures, Internet surfing, chatting, social network service (SNS), etc. 2. Description of the Related Art [0002] Portable terminals are being used for various purposes. In particular, as high-performance portable terminals including smart phones become common and data transmission speed in mobile communication systems is greatly improved, the use of contents through portable terminals such as movies, music, TV programs, school lectures, UCC, and celebrity pictorials has become very difficult. is being activated.

These contents provided to the portable terminal are transmitted from the contents provider (CP) to the portable terminal at the request of the portable terminal user. However, in general, there is one or a few content providers who provide content, whereas there are many mobile terminals that receive content. Therefore, when content is provided on a one-to-one basis to each of these multiple clients from a server existing on the content provider side, traffic congestion may occur, which may cause delay or failure of content transmission.

In order to solve such a problem, a CDN (Contents Delivery Network) service is being used in recent years.

CDN service is a service for stably delivering various types of content such as photos, movies, and music videos to user terminals. After sending frequent content to Edge Nodes distributed across multiple Internet Service Provider (ISP) Interconnection Points (PoPs), when there is a user's request for content, the edge node sends the content to users It means a technology for transmitting to a terminal.

In the case of CND servers, part or all of the content is copied and stored in advance in edge nodes distributed around the user's terminal, and then, when a content request occurs, the edge node located closest to the user's terminal acts on behalf of the user on behalf of the content provider. Since the content is provided to the terminal, there is an advantage of being able to respond to a user's content request more quickly. Accordingly, in general, a CDN service includes a plurality of CDN servers, and each CDN server may store the same content or different content, and may store one content separately for each section.

On the other hand, due to the portability of the user terminal, users often use the user terminal while on the move. There is a disadvantage in that it is expensive compared to the Internet line.

Accordingly, a public Wi-Fi service is provided in public transportation such as a bus or subway. In the public Wi-Fi service, the user can access the Internet through the Wi-Fi AP after accessing the Wi-Fi AP mounted in the vehicle.

However, due to the nature of free public Wi-Fi, it is difficult to secure sufficient Internet bandwidth from a budgetary point of view. As a result, the usage rate of the service is not high.

Accordingly, the method and apparatus for transmitting content in a core node for controlling a plurality of edge nodes according to an embodiment is an invention devised to solve the above-described problems, and includes a short-range wireless AP mounted on a vehicle in which a user is riding and The purpose of the present invention is to transmit content to user terminals more efficiently by using communication between CDN servers installed in edge nodes.

Specifically, the content requested by the user is divided into a plurality of contents based on the user's current location and movement path, and the divided plurality of contents are transmitted to a plurality of edge nodes according to the user's movement path, and then at the edge node, the user By transmitting the divided content to the short-distance communication AP of the vehicle in which the user is boarding, the object is to enable more efficient use of the short-range communication AP mounted in the vehicle.

The apparatus for transmitting content in a core node for controlling a plurality of edge nodes according to an embodiment includes the user's content request information, the user's content request information, and the A communication module for receiving location information and route information, and after determining a plurality of content transmission edge nodes to transmit the content to the short-range communication AP from among the plurality of edge nodes based on the location information and route information of the user, the content A content division module that divides the content into a plurality of pieces corresponding to the number of transmission edge nodes to generate divided contents, and transmits the divided contents to the content transmission edge nodes according to the user's route information, respectively; and the user terminal and a control module for controlling the content transmission edge node so that the divided content can be transmitted to the user terminal through the short-range communication AP.

The edge node may include a Contents Delivery Network (CDN) server capable of transmitting data to the short-range communication AP, and the CDN server may be disposed at at least one of a bus stop and a subway stop.

The content division module is configured to calculate the estimated movement times of the transportation means between the content transmission edge nodes, respectively, and then divide the contents into divided contents having different sizes based on the number of the content transmission edge nodes and the expected movement time. can be created with

The content division module is configured to calculate the expected stop times for the transportation means to stop at the content transmission edge module, and then divide the contents based on the number of the content transmission edge nodes and the expected stop time. It can be created as divided content.

The control module may transmit the divided content to the short-range communication AP while the transport means respectively stop in the content transmission edge module.

The content division module includes a pre-learned getting off point prediction artificial neural network module using the user's location information and route information as input information, and using the user's expected get off point as output information, and the getting off point predicted artificial neural network The module may perform learning based on a difference between the user's expected getting off point and the user's actual getting off point.

The content transmission apparatus in the core node controlling the plurality of edge nodes determines, for each of the plurality of edge nodes, the preferred content to be preferred by the users for each specific time period based on past content request information of users who have requested the content, The method may further include a content determination module for transmitting preferred content to the plurality of edge nodes before the specific time period.

The content determination module may include a pre-learned preferred content prediction artificial neural network module using the user's past content request information and time information of the past content request as input information, and using the preferred content information as output information. .

The loss function of the artificial neural network module for predicting preferred content includes a distance between the preferred content for each time period and the content information requested by the users, and the preferred content predicting artificial neural network module is configured to minimize the value of the loss function. It is possible to learn the weight of the preferred content prediction artificial neural network module.

In a method for controlling a content transmission apparatus in a core node for controlling a plurality of edge nodes according to an embodiment, the user's content request information is provided through the edge node connected to the short-range communication AP of the transportation means in which the user is boarding. , receiving the user's location information and route information through a communication module; determining a plurality of content transmission edge nodes to transmit the content to the short-range communication AP based on the user's location information and route information; After dividing the content into a plurality of pieces corresponding to the number of the content transmitting edge nodes to generate divided content, the divided content is transmitted to the content transmitting edge node according to the user's path information using the communication module, respectively and controlling the content transmission edge node so that the divided content can be transmitted to the user terminal through the short-range communication AP connected to the user terminal.

The edge node may include a Contents Delivery Network (CDN) server capable of transmitting data to the short-range communication AP, and the CDN server may be disposed at at least one of a bus stop and a subway stop.

A content transmission system that transmits content by controlling a plurality of edge nodes uses a CDN server installed at a bus stop connected to a short-distance communication AP of the bus the user is riding on, the user's content request information, and the location of the user. A plurality of edge nodes for receiving information and route information, a communication module for receiving the user's content request information and the user's location information and route information from the plurality of edge nodes, based on the user's location information and route information After determining a plurality of content transmission edge nodes to transmit the content to the short-range communication AP from among the plurality of edge nodes, and dividing the content into a plurality of pieces corresponding to the number of the content transmission edge nodes to generate divided content, so that the divided content can be transmitted to the user terminal through a content determination module that respectively transmits the divided content to the content transmission edge node according to the user's route information and the short-range communication AP connected to the user terminal It may include a core node including; a control module for controlling the content transmission edge node.

A content transmission method and apparatus in a core node for controlling a plurality of edge nodes according to an embodiment connects a CDN server installed in the edge node and a short-range wireless AP mounted in a vehicle to provide various contents to a user at high speed Since it can transmit, the user's satisfaction with the content can be increased, and at the same time, the utility of the short-range wireless AP installed in the vehicle can be increased.

In addition, since the content is divided according to the user's movement path and then transmitted to the edge node, problems such as interruption and insufficient reception time that occur when one edge node receives all one content can be prevented.

In addition, after learning the content preferred by users at each edge node by using an artificial neural network, the preferred content is transmitted to the CDN server at the edge node in advance to provide content tailored to users’ preferences. At the same time, since preferred content is stored in the CDN server in advance, overload problems caused by content transmission can be prevented.

1 is an exemplary system diagram for explaining a content transmission control system according to an embodiment of the present invention.

2 is an exemplary diagram for explaining MMT content according to an embodiment of the present invention.

3 is a block diagram illustrating some components of a content transmission apparatus according to an embodiment of the present invention.

4 is a diagram illustrating a state in which content is divided and transmitted from a content transmission apparatus to a plurality of edge nodes according to an embodiment of the present invention.

5 is a diagram illustrating a state in which content is transmitted from an edge node to a user terminal through a transportation means according to an embodiment of the present invention.

6 is a diagram illustrating a state in which content is divided and transmitted from a content transmission apparatus to a plurality of edge nodes according to an embodiment of the present invention.

7 is a diagram illustrating input information input to an artificial neural network module for predicting a departure point and output information output according to an embodiment of the present invention.

8 is a diagram illustrating a learning process of an artificial neural network module for predicting an exit point according to an embodiment of the present invention.

9 is a schematic diagram showing the configuration of a feature extraction module according to another embodiment of the present invention.

10 is a schematic diagram illustrating an example of a feature extraction module according to another embodiment of the present invention.

11 is a schematic diagram illustrating an attention module according to another embodiment of the present invention.

12 is a schematic diagram illustrating an expected getting off point output module according to another embodiment of the present invention.

13 is a diagram illustrating input information input to a content determination artificial neural network module and output information outputted to the content determination artificial neural network module according to an embodiment of the present invention.

14 is a diagram illustrating a learning process of a content determination artificial neural network module according to an embodiment of the present invention.

15 is a schematic diagram showing the structure of a content determination reinforcement learning module of the content determination module according to another embodiment of the present invention.

16 and 17 are diagrams illustrating a state in which the content determined by the content determination artificial neural network module is transmitted to a plurality of edge nodes, according to an embodiment of the present invention.

18 is a flowchart illustrating an operation sequence of a content transmission control system according to an embodiment of the present invention.

Hereinafter, embodiments according to the present invention will be described with reference to the accompanying drawings. In adding reference numerals to the components of each drawing, it should be noted that the same components are given the same reference numerals as much as possible even though they are indicated on different drawings. In addition, in describing the embodiment of the present invention, if it is determined that a detailed description of a related known configuration or function interferes with the understanding of the embodiment of the present invention, the detailed description thereof will be omitted. In addition, although embodiments of the present invention will be described below, the technical spirit of the present invention is not limited thereto and may be modified by those skilled in the art and variously implemented.

In addition, the terms used herein are used to describe the embodiments, and are not intended to limit and/or limit the disclosed invention. The singular expression includes the plural expression unless the context clearly dictates otherwise.

In this specification, terms such as "comprises", "comprises" or "have" are intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, but one It does not preclude in advance the possibility of the presence or addition of other features or numbers, steps, operations, components, parts, or combinations thereof, or other features.

In addition, throughout the specification, when a certain part is "connected" with another part, it is not only "directly connected" but also "indirectly connected" with another element interposed therebetween. Including, terms including an ordinal number, such as "first", "second", etc. used herein may be used to describe various elements, but the elements are not limited by the terms.

Hereinafter, with reference to the accompanying drawings, the embodiments of the present invention will be described in detail so that those of ordinary skill in the art can easily implement them. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description will be omitted.

1 is an exemplary system diagram for explaining a content transmission control system according to an embodiment of the present invention, and FIG. 2 is an exemplary diagram for explaining MMT content according to an embodiment of the present invention.

Referring to FIG. 1 , a content transmission control system 1 according to an embodiment of the present invention includes a user terminal 60, a plurality of edge nodes (Edge MANE, 40), and a core node (Core MANE, 30) and a content server. (Original Contents Server, 10) may be included. Here, MANE means MMT Aware Network Entity.

The user terminal 60 may transmit a content request message for requesting specific content to the content server 10 through a network according to a user's request. Here, the content request message of the user terminal 60 may be generated according to a request input of a specific content from the user. For example, when a user input such as a Uniform Resource Locator (URL) for a specific content or a user input for executing an application capable of reproducing content is confirmed, the user terminal 60 is a name server such as a DNS (Domain Name Server) can be accessed to obtain address information (IP address) of the core node 30 that manages the corresponding content. Then, the user terminal 60 transmits a content request message for requesting content to the core node 30 , and receives and uses the content under the control of the core node 30 .

In particular, the user terminal 60 of the present invention is a user's device capable of supporting the MPEG Media Transport (MMT) protocol, and is a short-distance communication ap 55 mounted on the content server 10 or the transportation means 50 for MPEG media. It is also possible to request transport streaming data (hereinafter referred to as 'MMT content'), and receive and reproduce the MMT content transmitted by the content server 10 from the short-range communication ap 55 mounted on the transportation means 50 . there is.

In addition, the user terminal 60 according to an embodiment of the present invention transmits a content request message to the short-range communication ap 55 mounted on the core node 30 or the transportation means 50, and transmits a mobile descriptor. can be transmitted including Here, the mobile descriptor may include information about the user terminal 60 (identification information, IP address, etc.), connected cell identification information (current_cell_id), and the like.

The content server 10 is a server for providing various content, such as photos, videos, audio, applications, etc., to the user terminal 60 , and provides at least one content from a content provider (CP). It receives and manages, and provides content directly to a plurality of user terminals 60 through a network, or through a short-distance communication ap 55 mounted on the core node 30 , the edge node 40 , and the transportation means 50 . Content may be transmitted to the user terminal 60 .

The content server 10 of the present invention cooperates with a streaming server (not shown) to provide content to the streaming server (not shown), and the streaming server (not shown) converts the content into MMT content to the core node 30 . , and if the content server 10 of the present invention supports the streaming function, the content may be provided to the core node 30 in a direct streaming form.

Accordingly, the content server described in the specification of the present invention is for providing various types of content to the user terminal 60, for example, content such as photos, videos, audio, applications, etc., from a content provider (CP). It is possible to receive and manage at least one content, and transmit the corresponding content according to the request of the user terminal 60, and preferably provides the content agreed with the communication service provider that provides the CDN service. Such a content server may be exemplified by a web server or a web application server (WAS; Web Application Server).

In the following description, it is assumed that the content server 10 directly provides content for convenience, but it should be noted that as described above, it may operate in conjunction with a separate streaming server (not shown).

In addition, as long as the above-described content can be expressed by a specific address (eg, URL) on a network, any content may be included in the content of the present invention. For example, when the entire web site is also expressed as one URL, it may be distributed as one content through the CDN service. In addition, the content of the present invention may be defined according to the MMT (MPEG Media Transport) protocol, but is not necessarily limited thereto.

A plurality of edge nodes 40 and core nodes 30 distributed over the network play a role of controlling the content flow between the user terminal 60 and the content server 10 . At this time, the plurality of edge nodes 40 are located between the access network 90 and the core network 80, and the core node 30 is preferably located between the core network 80 and the Internet network 20, The present invention is not limited thereto, and if the content flow between the user terminal 60 and the content server 10 can be controlled, a plurality of edge nodes 40 and core nodes 30 may be disposed on the network in various forms. there is.

In addition, the core node 30 of the present invention may serve to manage a plurality of edge nodes 40 distributed over a network. In particular, the core node 30 of the present invention may periodically receive a coverage message (ECC; Edge Cdn Coverage) from a plurality of edge nodes 40 . In addition, information on the cell environment in which the edge node 40 is located can be periodically updated and managed using the base station information included in the coverage message.

In addition, the core node 30 of the present invention may control the overall content transmission process. In particular, the core node 30 of the present invention may receive the content request message transmitted from the user terminal 60 through the short-range communication ap 55 and the edge node 40 mounted on the transportation means 50 . .

The edge node 40 of the present invention may refer to a place in which a high-speed Internet line is installed and a content delivery network (CDN) server at the same time. In FIG. 1 , a bus stop 41 and a subway stop 42 are illustrated as an example of the edge node 40 . However, the location of the edge node 40 is not limited to the bus stop 41 and the subway stop 42 , and a place where many users stay for a certain time to move to another place, such as the bus stop 41 or the subway stop 42 . At the same time, it may correspond to a place where a high-speed Internet line and a CDN server are installed.

The edge node 40 of the present invention is provided with a high-speed Internet line linked to the CDN server, so that a large amount of data can be quickly transmitted to the short-distance communication AP 55 mounted in the transportation means 50 .

In addition, the edge node 40 according to an embodiment of the present invention can store and manage the divided MMT content by dividing the MMT content into media processing unit (MPU; Media Processing Unit, hereinafter referred to as 'MPU') unit. , may be transmitted to the user terminal 60 by streaming the divided MMT content in units of MPU.

If described in more detail with reference to FIG. 2 for such MMT content, the asset 70 of the present invention may be composed of at least one MPU 71 .

Here, the asset 70 refers to an element of a package that encapsulates encoded media data such as audio, video, or web page data having a timed or non-timed property, and means the largest logical data unit. These assets 70 refer to data objects including data with the same transfer characteristics. For example, if the MMT package means one independent content such as a movie as the largest entity among the data models defined in the MMT functional area, the asset 70 includes audio data, video data, and It means an element such as script data, and an MMT package that is a set of the assets 70 may be one content.

The asset 70 includes a plurality of MPUs 71 , and the plurality of MPUs 71 mean logical data objects including the same asset ID. The MPU 71 is encoded media data that can be completely and independently processed, and the processing of the MPU may mean packetization for encapsulation or delivery to a package.

Further, the MPU 71 may include one or more access units for timed data. Here, an access unit (AU) is the smallest data object that can have time information. MPU 71 may also be configured to include a portion of data that has no access unit boundaries for non-timed data. This MPU 71 is a coded media data unit that can be completely and independently processed, and may be composed of a header 72 and coded data 73 . In addition, the MPU 71 of the present invention may be further configured as a media fragment unit (MFU; Media Fragment Unit).

The MMT content of this invention may be transmitted directly to the user terminal 60 in streaming form or transmitted to the user terminal 60 through the short-distance communication ap 55 mounted on the transportation means 50. In this case, based on the MPU can be transmitted. That is, in this way, the content server 10 may sequentially deliver the MMT content in units of MPU to the user terminal 60, and the user terminal 60 may receive the MMT content in units of MPUs that are sequentially received and play them. there will be However, the MMT content of the present invention is not necessarily limited to the MPU unit, and the processing of the present invention may be performed in a media fragment unit (MFU) unit according to a system implementation method.

In addition, it should be noted that the content according to an embodiment of the present invention will be described as an example that is defined according to the MMT protocol, but is not necessarily limited thereto, and may be defined according to a protocol of the same or similar nature to the MMT protocol. do.

Returning to FIG. 1 and describing the content providing system 1 , the network in which a plurality of edge nodes 40 of the present invention are distributed and arranged at main points is a network of information between the user terminal 60 and the content server 10 . It means a communication network for transmission and reception, and in particular, rather than a single communication network, various types of communication networks, for example, intranet networks, mobile communication networks, satellite networks, etc., may be a mixed type of communication networks implemented using various types of wired and wireless communication technologies.

Here, the network may be configured to include an external network connected to a mobile network including a plurality of access networks 32 and a core network 80 , for example, the Internet network 20 . Here, the wired/wireless access network constituting the mobile network is an access network for performing wired/wireless communication with the user terminal 60, for example, with a plurality of base stations such as BS (Base Station), BTS (Base Transceiver Station), NodeB, eNodeB, and the like. , may be implemented as a base station controller such as a base station controller (BSC) or a radio network controller (RNC).

In another method, the digital signal processing unit and the radio signal processing unit, which were integrally implemented in the base station, are divided into digital units (Digital Unit, hereinafter referred to as DU and radio unit, hereinafter referred to as RU), respectively, and a plurality of A plurality of RUs (not shown) may be installed in each area of , and a plurality of RUs (not shown) may be connected to a centralized DU (not shown) to be configured.

In addition, the core network 80 constituting the mobile network together with the wired/wireless access network may serve to connect the wired/wireless access network 90 and an external network, for example, the Internet network 20 . As described above, the core network 80 is a network system that performs a main function for a mobile communication service such as mobility control and switching between the wired and wireless access networks 90 , and is circuit switching or packet switching. ), and can manage and control the packet flow within the mobile network.

In addition, the core network 80 manages inter-frequency mobility, and may play a role for interworking with traffic in the wired/wireless access network 90 and the core network 80 and other networks, for example, the Internet network 20 . The core network 80 may be configured to further include a Mobile Switching Center (MSC), a Home Location Register (HLR), a Mobile Mobility Entity (MME), and a Home Subscriber Server (HSS) in addition to the above-described configuration.

In addition, the Internet network 20 refers to a general public communication network, that is, a public network, through which information is exchanged according to the TCP/IP protocol, and is connected to the content server 10 to provide content provided by the content server 10 . may be provided to the user terminal 60 via the core network 80 and the wired/wireless access network 90 , or may be provided to a plurality of edge nodes 40 according to a preset distribution policy.

The network structure of the present invention has been briefly described above. This network is combined with the edge node 40 to store computing resources such as hardware and software, and provides the computing resources required by the user terminal 60 to the corresponding terminal. It may include a cloud computing network that can Here, cloud computing refers to a computer environment in which information is permanently stored in a server on the Internet, and temporarily stored in the user terminal 60 such as a desktop, tablet computer, notebook, netbook, or smartphone, and cloud computing is the It refers to a computer environment access network that stores all information on a server on the Internet and allows this information to be used anytime, anywhere through various IT devices. Such a network is a closed network such as a local area network (LAN), a wide area network (WAN), etc., and an open network such as the Internet, as well as CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access) , GSM (Global System for Mobile Communications), LTE (Long Term Evolution), which are recently attracting attention, EPC (Evolved Packet Core) networks, etc., as well as next-generation networks and cloud computing networks to be implemented in the future.

The transportation means 50 of the present invention is a public transportation means that users board to move, and means a transportation means in which the short-distance communication ap 55 capable of performing short-range wireless communication with the user terminal 60 is mounted. . 1 shows an example of the bus 51 passing through the bus stop 41 and the subway 52 passing through the subway stop 42 as an example, but an example of the transportation means 50 is the bus 51 It is not limited to and subway 52, and may be included as long as it is a transportation means 50 that can perform a similar function.

In addition, in the case of the short-range communication ap (55), the present invention will be described based on Wi-Fi for convenience of description, but is not limited thereto and is included in the short-distance communication ap (55) as long as it is a communication means capable of performing a similar function. can

The user terminal 60 of the present invention may be implemented as a portable terminal capable of performing short-range wireless communication with a short-range wireless (AP) mounted in the transportation means 50 through a wired/wireless network. Here, the portable terminal may be a mobile communication device that guarantees portability and mobility. For example, the portable terminal includes a Personal Communication System (PCS), a Global System for Mobile communication (GSM), a Personal Digital Cellular (PDC), a Personal Handyphone System (PHS), a Personal Digital Assistant (PDA), and an International Mobile Telecommunication (IMT). -2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminal, Smartphone, SmartPad, Tablet PC, Smart Watch (smart watch), smart glass (smart glass), wearable device (wearable device), such as all kinds of handheld (Handheld) based wireless communication device may be included.

3 is a block diagram illustrating some components of the content transmission apparatus 100 according to an embodiment of the present invention.

Referring to FIG. 3 , the content transmission apparatus 100 in the core node for controlling a plurality of edge nodes according to an embodiment includes a communication module 110 , a content division module 120 , a content determination module 130 , and a memory It may include a module 140 and a control module 150 , and the like.

The communication module 110 may transmit and receive data while performing communication with the content server 10 , the plurality of edge nodes 40 , and the user terminal 60 .

Specifically, the communication module 110 receives the user's content request information and the user's location information from the edge node 40 that communicates with the short-range communication AP 55 mounted on the transportation means 50 on which the user is boarding. and path information, and transmit the received information to the content division module 120 , the content determination module 130 , and the memory module 150 .

The content division module 120 may determine a content transmission edge node to transmit the content requested by the user to the short-range communication AP 55 from among the plurality of edge nodes 40 based on the user's location information and path information.

The content division module 120 divides the content requested by the user into a plurality of pieces according to the determined number of content transmission edge nodes to generate divided contents, and transmits the generated divided contents to the content transmission edge nodes according to the user's path information, respectively. can

In general, when the content requested by the user is a video, the size of the file for playing the video is relatively large, whereas the time the transportation means 50 is stopped at one edge node 40 is relatively short. It is difficult to transmit all the videos requested by the user at once while the transportation means 50 is stopped at a specific edge node 40 . However, according to the present invention, if the content requested by the user is divided into a plurality of pieces according to the user's path and then the content is transmitted to the user terminal 60 through a plurality of edge nodes 40, the content can be transmitted more efficiently. . It will be described in detail with reference to the drawings below.

4 is a diagram illustrating a state in which content is divided and transmitted from a content transmission device to a plurality of edge nodes according to an embodiment of the present invention, and FIG. 5 is a transportation means at the edge node according to an embodiment of the present invention It is a diagram illustrating a state in which content is transmitted to a user terminal through

Referring to FIG. 4 , when the content to be transmitted to the user is determined as content X by the content transmission device 100 , the content division module 120 performs the short-distance communication ap 55 mounted on the bus 51 on which the user is boarding. ), the current position of the bus 51 and the future movement path can be predicted, and the content X can be divided based on this.

For example, the bus 51 is stopped at the first bus stop 41a, and as a result of determining the user's moving path, the bus 51 goes through the second bus stop 41b to the third bus stop 41c. When it is predicted to move, since there are a total of three bus stops 41 passing through, the content division module 120 sends the content transmission edge node to transmit the content requested by the user to the short-distance communication ap 55 to the first bus stop. (41a), the second bus stop (42b) and the third bus stop (42c), a total of three can be determined.

Thereafter, the content division module 120 divides the content X into the first content 81a, the second content 81b, and the third content 81c according to the number of content transmission edge nodes, as shown in FIG. 4 . After dividing into three, the first content 81a is transmitted to the CDN server installed at the first bus stop 41a, and the second content 81b is transmitted to the CDN server installed at the second bus stop 41b. , and the third content 81c may be transmitted to a CDN server installed at the third bus stop 41c.

When the divided content X is transmitted to the CDN server installed at each bus stop 41, the bus 51 the user is riding on is transmitted to the first bus stop 41a, the second bus 51, respectively, as shown in FIG. The CDN server 45 installed in the edge node during the time it is stopped at the second bus stop 41b and the third stop 41c is mounted on the bus 51 so that the user terminal 60 can receive the divided content. The divided content can be transmitted to the short-distance communication ap (55).

In FIG. 5, it is assumed that the plurality of user terminals 60a, 60b, and 60c have requested the same content 81a, and it is only illustrated that the CDN server 45 has received only the first divided content 81a. However, if the plurality of user terminals 60a , 60b , and 60c want different content, different types of content may also be received.

In addition, the content division module 120 does not simply divide the content to be transmitted based on the number of content transmission edge nodes 40 through which the transportation means 50 the user is boarding passes, but transports between the content transmission edge modules. After calculating the estimated movement times of the means 50 , the content may be generated into divided content having different sizes based on the number of content transmission edge modules and the estimated movement times.

4, the distance d1 between the first bus stop 41a and the second bus stop 41b as shown in FIG. 4 is the second bus stop 41b and the third bus stop. Since the distance between 41c is longer than d2, the travel time of the transportation means 50 from the first bus stop 41a to the second bus stop 41b is from the second bus stop 41b to the third bus. It can be expected that it will take longer than the travel time of the transportation means 50 to the stop 41c. Accordingly, in dividing the content X, the content division module 120 transmits the first divided content 81a sent to the CDN server of the first bus stop 41a to the second bus stop 41b, and the second divided content After dividing to have a longer playback time than (81b), the divided content can be transmitted. In this case, there is an effect that the content can be transmitted more efficiently according to the user's situation than when the content is equally divided and transmitted to a plurality of content transmission edge modules.

In addition, the content division module 120 further calculates the expected stop time at which the transportation means 50 stops at the content transmission edge module, and then the content requested by the user based on the number and the expected stop time of the content transmission edge module. can be created as divided content having different sizes.

For example, when the time that is generally stopped at the first bus stop 41a due to congestion is longer than the time that is stopped at the second bus stop 41b, the content division module 120 divides the content X, After dividing the first divided content 81a sent to the CDN server of the first bus stop 41a into content having a playback time longer than the second divided content 81b sent to the CDN server of the second bus stop 41b, can send In this case, there is an effect that the content can be transmitted more efficiently according to the user's situation than when the content is equally distributed and transmitted to the content transmission edge module.

6 is a diagram illustrating a state in which content is transmitted from an edge node to a user terminal through a transportation means according to another embodiment of the present invention.

In the case of FIG. 6 , it is assumed that the transportation means 50 is not the bus 51 but the subway 52 , unlike FIGS. 4 and 5 . Accordingly, each subway stop in FIG. 6 may serve as a plurality of edge nodes.

That is, when the content transmission device 100 determines that the content to be transmitted to the user is content X, the content division module 120 receives it through the short-distance communication ap 55 mounted in the subway 52 on which the user is boarding. Based on the received information, the current location of the subway 52 and the future moving route may be predicted, and the content X may be divided based on this.

For example, as a result of determining the movement path of a user who is boarding the subway 52 that is stopped at the first subway stop 42a, it is predicted that the user will move through the second subway stop 42b to the third subway stop 42c. In this case, since there are a total of three subway stops 42 passing through, the content division module 120 sets the content transmission edge node to transmit the content requested by the user to the short-distance communication ap 55 to the first subway stop 42a. , the second subway stop 42b and the third subway stop 42c, a total of three can be determined.

After that, the content division module 120 divides the content X 80 into the first content 81a, the second content 81b, and the third content 81c as shown in FIG. The first content 81a is transmitted to the CDN server installed at the first subway stop 42a, the second content 81b is transmitted to the CDN server installed at the second subway stop 42b, and the third content ( 81c) can be transmitted to the CDN server installed at the third subway stop 41c.

According to the prior art, when a user terminal receives all content from a CDN server installed in one edge node, if the stopping time is short or the size of the content is large, the user receives all the content from one edge node There was an unacceptable problem.

However, as shown in FIGS. 4 to 6 , when the user terminal 60 divides and receives the requested content based on the moving path, the content is divided into small sizes and received, so that the content can be received more stably. this exists

In addition, the content division module 120 according to the present invention predicts the user's drop off point by utilizing the artificial neural network module in dividing the user's requested content, and then divides the content based on this. Hereinafter, it will be described in detail with reference to FIGS. 7 and 8 .

7 is a diagram illustrating input information input to an artificial neural network module for predicting a departure point and output information output according to an embodiment of the present invention, and FIG. It is a diagram showing the learning process of the neural network module.

If clear information about the user's current location and getting off point is received in the information received through the communication module 110, it is sufficient to determine the content transmission edge node based on this, but otherwise, the content division module 120 is the user In dividing the requested content of , an artificial neural network module can be used to predict the user's drop off point, and a content transmission edge node can be determined based on this.

Referring to FIGS. 7 and 8 , the content division module 120 may include a disembarkation point prediction artificial neural network module 121 , and the disembarkation point prediction artificial neural network module 131 provides location information 15 and Using the route information 16 as input information and the output information as expected getting off point information 17 , learning may be performed through the learning session 122 , and inference may be performed through the inference session 133 .

In addition, in performing learning using the artificial neural network module, the drop-off point prediction artificial neural network module 121 performs learning based on the difference between the output expected get-off point information 17 and the actual users' get-off point information 18 can be done Specifically, after calculating the distance difference between the predicted defective point and the actual defective point, the weight of the drop-off point prediction artificial neural network module 121 may be learned so that the value of the loss function is minimized.

Therefore, in the case of the present invention, even if information on the user's exact get off point is not received from the user terminal 60, the get off point prediction artificial neural network module 121 based on the past data infers the expected get off point of the user relatively accurately. Therefore, there is an advantage that the content division module 120 can determine the content transmission edge node according to the user's moving path rather than when determining the content transmission edge node.

According to another embodiment of the present invention, the getting off point prediction artificial neural network module 121 may include a feature extraction module 122 , an attention module 123 , and an expected getting off point output module 124 .

The feature extraction module 122 is a module that uses a thumbnail image or a specific frame of content as input information 21 requested by a user as input information, and uses a feature vector 22 as output information. In relation to the feature vector 22 output of the feature extraction module 122, FIG. 9 is a schematic diagram showing the configuration of the feature extraction module 122 according to another embodiment of the present invention.

As shown in FIG. 9, the feature extraction module 122 is a fully connected layer (23) and a layer before the output layer (22) in the pre-learned image classification artificial neural network including ConvNet. This module outputs the output vector as the input thumbnail image or the feature vector 31 for a specific frame.

According to another embodiment of the present invention, the feature extraction module 122 is an n-dimensional vector to which average value pooling (avg_pooling) is applied to the result of extracting the feature in the last stage 22 of the convolution layer of the pre-learned image classification artificial neural network. may be configured to output as a feature vector 31 for an input thumbnail image or a specific frame.

10 is a schematic diagram illustrating an example of the feature extraction module 122 according to another embodiment of the present invention.

As shown in Fig. 10, for example, when the pre-trained image classification artificial neural network uses the structure of ResNet-50, a 7x7 Average Pooling layer is performed on the vector of 7x7x2,048 output from Conv5, the last convolution layer. The output 2,048-dimensional vector may be configured as the feature vector 31 .

In relation to the output of the attention vector of the attention module 123, FIG. 11 is a schematic diagram illustrating the attention module 123 according to another embodiment of the present invention.

The attention module 123 includes a feature vector 31 for a thumbnail image or a specific frame, a context vector 32 for a title or content content included in the content corresponding to the content request information, and location information 33 for a content request. , a module in which route information 34 at the time of content request, time information 35 at the time of content request as input information, and an attention vector 61 as output information.

The context vector 32 may be generated by embedding the content title or text information constituting the content content included in the content corresponding to the content request information into a text embedding model of Word2Vec or Sent2Vec.

The location information 33 when requesting content may mean GPS coordinate information or information on the nearest stop.

The route information 34 at the time of content request is a transportation means that is output based on information on the short-range communication AP 55 of the transportation means 50 connected to the user terminal 60 when the content is requested from the user terminal 60 . (50) may mean information about.

The time information 35 at the time of content request may refer to time information when the content is requested from the user terminal 60 .

As shown in FIG. 11 , the attention module 123 includes five pieces of input information (a feature vector 31 , a context vector 32 , location information 33 , path information 34 , and time each having different dimensions). Information 35) is reduced to one dimension by applying ConvNet to each input information, and an activation function (Softmax, ReLU, etc.) is applied to the reduced vector to output a 5x1 dimension attention vector. The attention module 123 is configured to output the attention input information 62 that is input information to which the attention vector is applied by multiplying the output attention vector 61 as a weight by the input information.

Regarding the configuration of the artificial neural network of the attention module, as shown in FIG. 11, the artificial neural network of the attention module is, for example, "Conv1-BN, ReLU-Conv2-BN, ReLU-Conv3-BN, ReLU-Conv4-BN. -Softmax" structure, and Dense Connectivity that channel-wise concatenation (xl=Hl([x0,x1,...,xl])) of the output feature map can be applied. BN stands for Batch Normalization, and ReLU stands for Rectified Linear Unit. Compared to the existing ResNet structure, this structure alleviates the vanishing gradient problem in which features disappear as it passes through layers, the effect of intensifying the propagation of features, the effect of reducing complexity by reducing parameters, and performing with a small number of training data. possible and the effect of reducing overfitting occurs.

The expected getting off point output module 124 is a module that receives attention input information and outputs expected get off point information.

The expected getting off point output module 124 may use a structure such as a support vector machine (SVM), a decision tree, or an artificial neural network as a classifier. The attention input information 62 is a 5x1-dimensional vector, and refers to a weight vector to which the attention vector 61 output from the attention module is applied to five input information (feature vector, context vector, location information, path information, time information). do.

The expected getting off point information 64 may mean information on how many stops are used after the content is requested from the user terminal 60 . For example, when the user gets off after passing 5 additional stops after requesting the content A, the expected disembarkation point information may be '5'. In comparison with the method of using the coordinates of the disembarkation point or the stop serial number of the disembarkation point as expected disembarkation point information, information on how many stops are used after the content is requested from the user terminal 60 is expected disembarkation point information When configured as , it becomes easy to construct comparison data (ground truth) for parameter update of the expected drop off point output module in the learning session of the expected drop off point output module 124 .

With respect to the structure of the expected disembarkation point output module 124, FIG. 12 is a schematic diagram illustrating the expected disembarkation point output module 124 according to another embodiment of the present invention.

As shown in FIG. 12 , when an artificial neural network is used as the expected getting off point output module 124 , for example, the full coupling layer (Fully) like the structure of “FC-ReLU-FC-ReLU-FC-Softmax” It can be composed of an artificial neural network including a Connected Layer, 63). FC stands for Fully Connected Layer. The structure of the expected get-off point output module 124 is configured to output the expected drop-off point information 64 by stacking n pre-coupling layers (FC Layers), thereby improving inference performance.

Referring back to FIG. 3 , another component of the content transmission apparatus 100 will be described. For each of the plurality of edge nodes 40 , the content determination module 130 provides a user based on past content request information of users who have requested content. may determine preferred content to be preferred for each specific time period, and transmit the determined preferred content to the plurality of edge nodes 40 .

The content requested by users does not always request the same content, and the types of requested content vary according to time and place, but the kinds of content that users intensively request in a specific time period are limited. Accordingly, as in the content determination module 130 of the present invention, for each of the plurality of edge nodes 40 that exist in different locations, based on the past content request information of users who requested content based on the edge node 40, users If preferred content is determined for each specific time period and the determined preferred content is transmitted to the plurality of edge nodes 40 before the specific time period, there is an effect of more stably providing content to users.

In addition, the content determination module 130 according to the present invention may determine preferred content by using an artificial neural network module in determining users' preferred content. Hereinafter, it will be described in detail with reference to FIGS. 13 to 17 .

13 is a diagram illustrating input information input to and output information to a content determination artificial neural network module according to an embodiment of the present invention, and FIG. 14 is a content determination artificial neural network module according to an embodiment of the present invention. It is a diagram showing the learning process. 15 is a schematic diagram showing the structure of a content determination reinforcement learning module of a content determination module according to another embodiment of the present invention, and FIGS. 16 and 17 are, according to various embodiments of the present invention, determined by the content determination artificial neural network module It is a diagram illustrating a state in which content is transmitted to a plurality of edge nodes.

13 and 14 , the content determination module 130 may include a content determination artificial neural network module 131, and the content determination artificial neural network module 131 may include content request information 11 and content of past users. Using the information 12 of the requested time as input information, and the output information as the preferred content information 13 preferred by users for each time period, learning is performed through the learning session 132, and the inference session 133 is used for inference. can be done

Specifically, as shown in FIG. 9 , the content determination artificial neural network module 131 uses the content request information 11 and the content request time information 12 as input information, and then infers this to determine the user by time zone. It is possible to output the preferred content information 13 preferred by the users. Furthermore, in outputting the preferred content information 13 , the content determination artificial neural network module 131 may divide the preferred content information 13 and output the preferred content information 13 preferred by users for each specific time period.

For example, the content determination artificial neural network module 131 may divide the output layer by time period (eg, one hour unit) to learn and infer content preferred by users for each time period. Therefore, when the preferred content information 13 is output in units of one hour, the output layer of the content determination artificial neural network module 131 may be a total of 24 sheets, and when the preferred content information 13 is output in units of two hours, the output layer There can be a total of 12 layers. Such a time unit may be variously set according to a usage environment.

In addition, the content determination artificial neural network module 131 may perform learning based on the difference between the output preferred content information 13 and the actual requested content information 14 when learning is performed using the artificial neural network module. .

Specifically, it is possible to learn the weight of the content determination artificial neural network module 131 so that the value of the loss function is minimized by comparing the output preferred content for each time period with the content information requested by users in the actual time period and the distance.

For example, the content determination artificial neural network module 131 determines the preferred content of a specific time period (1:00 pm to 2:00 pm) as X with a reliability of 0.8. As a result of the analysis, if the content that occupies the highest proportion as requested content is content Y, learning can be performed by adjusting the reliability of content X.

In addition, such learning may be performed by integrating information received from the plurality of edge nodes 40 , but may be independently performed for each of the plurality of edge nodes 40 . The information requested by users may vary according to time, but depending on the region where users are located, different contents may be requested even in the same time zone. Accordingly, when the preferred content information 13 is output for each time period by learning for each of a plurality of edge nodes like the content determination artificial neural network module 131 of the present invention, the preferred content that users in a specific area can prefer is more precisely edged in advance. There is an advantage of being able to transmit to the node 40 .

In relation to another embodiment of the content determination module 130 , the content determination module 130 according to another embodiment of the present invention may include a content determination reinforcement learning module. 15 is a schematic diagram showing the structure of the content determination reinforcement learning module 134 of the content determination module according to another embodiment of the present invention.

15, the content determination reinforcement learning module 134 of the content determination module uses the Agent as the content determination reinforcement learning module, and the Environment as a thumbnail image corresponding to the existing content request information or a feature for a specific frame A vector 31, a context vector 32 of a content title or content content included in the content corresponding to the content request information, location information 33 at the time of a content request, route information at the time of a content request 34, when requesting a content of the time information 35, State as edge node information, which is information about which edge node the Action is, and prediction time information, which is a time period in which preferred content is to be predicted, and Action as the determination of preferred content information, Content is determined in the direction of increasing the reward by setting the reward as the preferred content share (the number of preferred content requests in the corresponding time period/the total number of content requests in the corresponding time period), which is the share of the number of times that content corresponding to the preferred content information is actually requested in the corresponding time period. The parameters of the reinforcement learning module 134 may be configured to be learned. Therefore, in the present invention, as the learning progresses in the content determination reinforcement learning module 134, the reward is learned in a direction that increases, so it is possible to provide the user with content more tailored to the user's preference, and such content is preferred by a large number of users. Since there is a high probability that it corresponds to content, if it is stored on the CDN server before a specific time period, there is an effect of increasing the utilization of the CDN server and preventing traffic overload due to content requests.

Referring to FIG. 16 , as shown in FIG. 16 , the content determination artificial neural network module 131 of the content transmission device 110 performs the first bus corresponding to the first edge node between 1:00 PM and 2:00 PM. At the stop 41a, it is determined that the preference of the X content is the highest, and at the second bus stop 41b corresponding to the second edge node, it is determined that the preference of the Y content is the highest, and the preference of the Y content is determined as the highest at the second bus stop 41b corresponding to the second edge node. When it is determined that the preference of the Z content is the highest at the 3 bus stop 41c, the X content is transferred to the CDN server of the first bus stop 41a before 1 o'clock, which is a specific time, and the content of the second bus stop 41b. The Y content may be transmitted to the CDN server, and the Z content may be transmitted to the CDN server of the third bus stop 41c corresponding to the third edge node. In this way, if the content is transmitted to the CDN server in advance, there is an effect that the content can be more easily transmitted to the user terminal 60 without overloading the user's request for the specific content in a specific time period.

In the case of FIG. 17, it is assumed that the transportation means 50 is a subway 52 instead of a bus 51, unlike in FIG. 16, and the content transmission apparatus 100 transmits a plurality of contents to several edge nodes. it is one drawing Accordingly, each subway stop in FIG. 12 may serve as a plurality of edge nodes.

Referring to FIG. 17 , the content determination artificial neural network module 131 determines that the preference of X content is the highest at the first subway stop 42a corresponding to the first edge node between 6:00 PM and 8:00 PM. In the second subway stop 42b corresponding to the second edge node, it is determined that the preference of the Y content is the highest, and the preference of the Z content is the highest at the third subway stop 42c corresponding to the third edge node. If it is determined that there is, the content transmission apparatus 100 may transmit the plurality of content appropriately to the plurality of edge nodes before 6 o'clock, which is a specific time period. 12, the CDN server of the first subway stop 42a contains the X content, the CDN server of the second subway stop 42b receives the Y content, and the third subway stop corresponding to the third edge node ( 42c) of the CDN server can transmit the content. In this way, if the content is transmitted to the CDN server in advance, there is an effect that the content can be transmitted more easily without overloading the user's request for the specific content in a specific time period.

Referring back to FIG. 3 , the remaining components of the content transmission device 100 will be described. The control module 150 controls the overall operation of the content transmission device 100 while simultaneously controlling the plurality of edge nodes 40 . operation can be controlled. Specifically, the control module 150 transmits a plurality of contents divided according to the movement path of the user from the plurality of edge nodes 40 to the short-range communication AP 55 of the transportation means 50 in which the user is riding in sequence. You can control the operation of the content sending edge node so that it can be

The memory module 150 may store various kinds of information about users transmitted by the communication module 110, and the stored information is the get-off point prediction artificial neural network module 121 and the content determining artificial neural network module 131 to learn. and can be used as necessary data for making inferences.

18 is a flowchart illustrating an operation sequence of a content transmission control system according to an embodiment of the present invention.

Referring to FIG. 18 , the user terminal 60 may transmit content transmission request information to the edge node 40 ( S10 ).

Specifically, although not shown in the drawing, information requested by the user terminal 60 is transmitted between the user terminal 60 and the edge node 40 through the short-distance communication ap 55 mounted on the transportation means 50 . It may be transmitted to the CDN server of the edge node 40, and when transmitted, the current user's location information and route information may also be transmitted. (S20)

Thereafter, the edge node 40 transmits all information received to the core node 30 , and the core node 30 may receive the content requested by the user from the content server 10 . (S30 ~S50)

Thereafter, the core node 30 determines a content transmission edge node to transmit content based on the user's location information and path information, and receives the content received from the content server 10 according to the determined number of content transmission edge nodes. Split into multiple pieces. (S60, S70)

When the content is divided into a plurality of pieces, the core node 30 transmits the divided content to each edge node 40 according to the movement path of the transportation means 50 , and receives the divided content from the core node 30 . The node 40 transmits the received content to the short-range communication AP 55 of the transportation means 50 while the transportation means 50 is stopped at each edge node 40, so that the user terminal 60 can transmit the content. to be able to receive (S80, S90)

So far, the features of the present invention have been described in detail with reference to the drawings.

A content transmission method and apparatus in a core node for controlling a plurality of edge nodes according to an embodiment connects a CDN server installed in the edge node and a short-range wireless AP mounted in a vehicle to provide various contents to a user at high speed Since it can transmit, the user's satisfaction with the content can be increased, and at the same time, the utility of the short-range wireless AP installed in the vehicle can be increased.

In addition, since the content is divided according to the user's movement path and then transmitted to the edge node, problems such as interruption and insufficient reception time that occur when one edge node receives all one content can be prevented.

In addition, after learning the content preferred by users at each edge node by using an artificial neural network, the preferred content is transmitted to the CDN server at the edge node in advance to provide content tailored to users’ preferences. At the same time, since preferred content is stored in the CDN server in advance, overload problems caused by content transmission can be prevented.

On the other hand, components, units, modules, components, etc. described as "~" described in this specification may be implemented together or individually as interoperable logic devices. Depictions of different features of modules, units, etc. are intended to emphasize different functional embodiments, and do not necessarily imply that they must be realized by separate hardware or software components. Rather, functionality associated with one or more modules or units may be performed by separate hardware or software components or integrated within common or separate hardware or software components.

Although acts are shown in the drawings in a specific order, it should not be understood that such acts need to be performed in the specific order shown, or sequential order, or all shown acts to achieve a desired result. . In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the division of various components in the above-described embodiments should not be construed as requiring such division in all embodiments, and that the described components will generally be integrated together into a single software product or packaged into multiple software products. It should be understood that there can be

A computer program (also known as a program, software, software application, script or code) may be written in any form of any programming language, including compiled or interpreted language or a priori or procedural language, as a stand-alone program or module; It can be deployed in any form, including components, subroutines, or other units suitable for use in a computer environment.

Additionally, the logic flows and structural block diagrams described in this patent document describe corresponding acts and/or specific methods supported by corresponding functions and steps supported by the disclosed structural means, and corresponding It can also be used to build software structures and algorithms and their equivalents.

The processes and logic flows described herein may be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output.

The present description sets forth the best mode of the invention, and provides examples to illustrate the invention, and to enable any person skilled in the art to make or use the invention. The specification thus prepared does not limit the present invention to the specific terms presented.

Although the above has been described with reference to the preferred embodiment of the present invention, those skilled in the art or those with ordinary skill in the art will not depart from the spirit and scope of the present invention described in the claims to be described later. It will be understood that various modifications and variations of the present invention can be made without departing from the scope of the present invention. Accordingly, the technical scope of the present invention should be defined by the claims rather than being limited to the contents described in the detailed description of the specification.

Claims (12)

1. A content transmission apparatus in a core node that controls a plurality of edge nodes, the content transmission apparatus comprising:

   a communication module for receiving the user's content request information, the user's location information, and route information through the edge node connected to the short-range communication AP of the transportation means in which the user is riding;

   After determining a plurality of content transmission edge nodes from among the plurality of edge nodes to transmit the content to the short-range communication AP based on the user's location information and path information, the content is transmitted in response to the number of the content transmission edge nodes a content division module dividing the divided contents into a plurality of pieces to generate divided contents, and transmitting the divided contents to the contents transmission edge node according to the user's path information; and

   A control module for controlling the content transmission edge node so that the divided content can be transmitted to the user terminal through the short-range communication AP connected to the user terminal;

   A content transmission device in a core node that controls a plurality of edge nodes.
2. The method of claim 1,

   The edge node is

   and a CDN (Contents Delivery Network) server capable of transmitting data to the short-range communication AP, wherein the CDN server is disposed at at least one of a bus stop and a subway stop,

   A content transmission device in a core node that controls a plurality of edge nodes.
3. The method of claim 1,

   The content division module,

   After calculating the estimated travel times of the transportation means between the content transmitting edge nodes, respectively, generating the content into divided content having different sizes based on the number of the content transmitting edge nodes and the expected moving time,

   A content transmission device in a core node that controls a plurality of edge nodes.
4. The method of claim 1,

   The content division module,

   After calculating the expected stop times at which the transportation means stops at the content transmission edge module, respectively, based on the number of the content transmission edge nodes and the expected stop time, the content is generated into divided contents having different sizes,

   A content transmission device in a core node that controls a plurality of edge nodes.
5. The method of claim 1,

   The control module is

   transmitting the divided content to the short-range communication AP while the transport means are respectively stopped in the content transmission edge module;

   A content transmission device in a core node that controls a plurality of edge nodes.
6. According to claim 1,

   The content division module,

   Pre-learned getting off point prediction artificial neural network module using the user's location information and route information as input information and the user's expected getting off point as output information;

   The get-off point prediction artificial neural network module,

   Learning based on the difference between the user's expected getting off point and the user's actual getting off point,

   A content transmission device in a core node that controls a plurality of edge nodes.
7. According to claim 1,

   For each of the plurality of edge nodes, based on past content request information of users who requested the content, the users determine the preferred content to be preferred for each specific time period, and transmit the preferred content to the plurality of edge nodes before the specific time period Content determination module; further comprising,

   A content transmission device in a core node that controls a plurality of edge nodes.
8. 8. The method of claim 7,

   The content determination module,

   Pre-learned preferred content prediction artificial neural network module using the user's past content request information and time information of the past content request as input information, and using the preferred content information as output information;

   A content transmission device in a core node that controls a plurality of edge nodes.
9. According to claim 1,

   The loss function of the preferred content prediction artificial neural network module is,

   including the distance between the preferred content for each time period and the content information requested by the actual users,

   The preferred content prediction artificial neural network module learns the weight of the preferred content prediction artificial neural network module so that the value of the loss function is minimized.

   A content transmission device in a core node that controls a plurality of edge nodes.
10. A method for controlling a content transmission device in a core node for controlling a plurality of edge nodes, the method comprising:

    receiving content request information of the user, location information of the user, and route information through a communication module through the edge node connected to the short-range communication AP of the transportation means in which the user is riding;

    determining a plurality of content transmission edge nodes to transmit the content to the short-range communication AP based on the user's location information and path information;

    After dividing the content into a plurality of pieces corresponding to the number of the content transmitting edge nodes to generate divided content, the divided content is transmitted to the content transmitting edge node according to the user's path information using the communication module, respectively step; and

    controlling the content transmission edge node so that the divided content can be transmitted to the user terminal through the short-range communication AP connected to the user terminal;

    A method for controlling content transmission in a core node that controls a plurality of edge nodes.
11. 11. The method of claim 10,

    The edge node is

    and a CDN (Contents Delivery Network) server capable of transmitting data to the short-range communication AP, wherein the CDN server is disposed at at least one of a bus stop and a subway stop,

    A method for controlling content transmission in a core node that controls a plurality of edge nodes.
12. a plurality of edge nodes for receiving the user's content request information, the user's location information, and the route information using a CDN server installed at a bus stop connected to the short-distance communication AP of the bus the user is boarding; and

    a communication module configured to receive the user's content request information and the user's location information and route information from the plurality of edge nodes;

    Based on the location information and path information of the user, a plurality of content transmission edge nodes to transmit the content to the short-range communication AP are determined from among the plurality of edge nodes, and the content is transmitted according to the number of the content transmission edge nodes. a content determination module for generating divided content by dividing it into a plurality of pieces, and then transmitting the divided content to the content transmission edge node according to the user's path information; and

    A core node comprising a; a control module for controlling the content transmission edge node so that the divided content can be transmitted to the user terminal through the short-range communication AP connected to the user terminal

    A content delivery system that transmits content by controlling a plurality of edge nodes.

Images