KR102310212B1 - Method for enabling information centric networking-based edge computing in 5G networks - Google Patents

Abstract

According to the present invention, a method for enabling information centric networking-based edge computing includes the steps of: transmitting, by a mobile user, an interest packet requesting content to a mobile device performing local device to device (D2D) communication; returning the content to the requesting mobile user, when the content is found on the local mobile device which has received the interest packet, and if not found, delivering the interest packet to a base station (BS) which performs edge computing; searching the CS (Contents Store) owned by the BS and transmitting the corresponding content to the mobile user if found, and forwarding the interest packet to an ICN router if not found; and retrieving cache memory from the ICN router, if the corresponding content is found, transmitting the content to the BS, and if not found, requesting a provider of a cloud environment through a core network and receiving the content. Accordingly, an overall waiting time can be shortened.

Description

{Method for enabling information centric networking-based edge computing in 5G networks}

The present invention relates to a method for activating edge computing, and more particularly, to a method for activating edge computing based on information-centric networking in a 5G network that can reduce the delay time caused by dynamic content of the 5G network.

Information-Centric Networking (ICN) and Edge Computing are emerging technologies to more rapidly distribute content to end-users, and are emerging as a new paradigm for the future Internet.

Edge computing is a technology that processes data not at the central server but at the edge where data is generated. The limitations of cloud computing have been revealed, and edge computing is attracting attention. By using the edge computing method, it is possible to reduce the time for data delivery, analysis, and result processing to the central server by distributing data around the IoT terminal. In addition, ICN is a communication based on an information identifier rather than an address of a communication target host, and is a 5G wireless network such as consumer mobility support, name-based forwarding, multi-homing, content security, and caching functions. It provides essential functions not covered in this study.

Meanwhile, in the upcoming 5G network, content distribution and latency will be very important issues, challenging the existing network infrastructure. Therefore, such ICN and edge computing are considered promising technologies that can reduce network latency by bringing content closer to the end user. In other words, if ICN is activated with such edge computing in the 5G Radio Access Network (RAN), there is an advantage in that the distance between users and services can be reduced to distribute content and increase communication performance efficiency.

However, ICN and edge caching methods have difficulties in handling dynamic content, and because dynamic content is always provided to the existing server through the core network, the edge server's caching method is not suitable for dynamic content. exist. That is, ICN and edge caching methods work well for static content that does not change over time, such as files or images, but there is a problem in that it is not useful in dynamic situations such as searching a search result page or real-time content.

Republic of Korea Patent Publication No. 10-1873238 (Registered on June 26, 2018)

The present invention has been proposed to solve the problems of the conventional ICN and edge caching method, and an object of the present invention is to provide an information-centric networking-based edge computing activation method to reduce the delay time of dynamic content in a 5G network. there is

The information-centric networking-based edge computing activation method according to the present invention for achieving the above object is a method for activating edge computing (Edge Computing) based on information-centric networking (ICN) in a 5G network, and a mobile user transmitting, by a mobile user, an interest packet requesting content to a mobile device performing local device to device (D2D) communication; If the local mobile device that has received the Interest packet finds the corresponding content, returning the content to the mobile user who requested the content, and if not found, delivering the Interest packet to a BS (Base Station) performing edge computing; Searching the CS (Contents Store) owned by the BS and transmitting the corresponding content to the mobile user if it is found, and forwarding the Interest packet to the ICN router if not found; The ICN router searches the cache memory and, if it finds the corresponding content, transmits it to the BS, and if it does not find it, requests it from the Provider through the Core Network and receives it.

Here, the BS tracks the request frequency according to the number of dynamic content requests of the mobile user, and when the number of requests for a specific content exceeds a set frequency threshold, prefetching the corresponding content to the provider through the ICN router ) to receive the corresponding content, store it in the CS, and provide it when a request for the corresponding content comes from a mobile user.

In addition, the ICN router tracks the frequency of content utilization according to the number of dynamic content requests from a plurality of BSs, and when the number of requests for a specific content exceeds a set frequency threshold, the content is provided to the provider through the core network. After requesting fetching, receiving the corresponding content and storing it in the memory, when a request for the corresponding content comes from the BS, it is provided.

Meanwhile, it is preferable that the mobile device performing the D2D communication performs ad-hoc communication with another mobile device based on ICN, and broadcasts and transmits content request information transmitted from the mobile user.

In addition, the BS builds an ICN application layer so that the content can be verified through the ICN naming system, and when a content request is received from a mobile device, the content availability is checked using a hierarchical naming scheme.

Here, the ICN responsive layer of the BS includes an API layer that receives and processes content requests from mobile users, a CS implementation layer that registers and manages content in the CS, and a PIT (Pending Interest) table. Table), event-based timer setting, PIT input and timer connection, PIT removal, PIT entries layer that performs PIT item search, and information on the node (Forwarder) delivered to the next for a specific name It includes a forwarder layer (Forwarder layer) and a data access layer (Data access layer) that manages the forwarder information storage and CS stored data.

According to the present invention, by providing the ICN edge support 5G network structure as D2D communication and by providing the ICN application layer for the BS in the 5G RAN and the content prefetching strategy based on the ICN hierarchical naming structure, access to the core network is minimized and cache hits occur locally, and most of the content is connected to the 5G RAN, affecting the overall latency of the network. Therefore, it is possible to minimize the traffic to the cloud, and there is an effect of reducing the overall waiting time.

1 is a conceptual diagram showing an information-oriented networking-based edge computing activation process according to the present invention;
2 is a conceptual diagram illustrating a D2D communication setting process according to the present invention;
3 is an example of a prefetching naming system according to the present invention;
4 is an overall structural diagram of an ICN-supported 5G system according to the present invention;
5 shows the application layer structure of the ICN for the BS according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The present invention proposes an ICN-capable RAN (Radio Access Network) structure that supports D2D (Device to Device) or ICN application layer in BS (Base Station) for activating information-centric networking-based edge computing in 5G network do. In addition, in order to apply the ICN and edge caching methods, which were conventionally useful only for static content, to dynamic content, we propose a content prefetching schema based on the ICN Naming technique.

The content prefetching strategy applied to the present invention solves the problem of dynamic content by requesting content from the provider in advance based on the frequency of content request messages, thereby minimizing access to the core network. In addition, cache concentration occurs locally, and as most of the content accesses the 5G RAN, the overall latency of the network can be reduced.

1 is a conceptual diagram illustrating a process of activating edge computing based on information-oriented networking according to an embodiment of the present invention, and shows a process processed when a mobile user 100 as an end consumer sends a message requesting content. .

1, in the present invention, a response to a message is made through two procedures: name-based D2D communication and a named-based prefetching mechanism for dynamic content. In the present invention, in order to activate the 5G network with ICN and edge computing and access the core network 500, first, ICN is activated locally at the device level and devices can communicate with each other. Next, ICN communication is set to be possible in the application layer of the BS (Base Station) 300, and a main data structure of ICN such as PIT, CS, and FIB is constructed. In addition, hierarchical name-based content prefetching strategy is implemented to handle 5G network dynamic content.

In more detail, as shown in FIG. 1 , the mobile user 100 searching for content transmits an Interest packet, which is a content request message, to a nearby mobile device 200 belonging to Local D2D communication. The D2D local mobile device 200 receiving this Interest packet searches for the requested content in its cache memory. When content is found, the content is transmitted to the mobile user 100 who sent the request, and when content is not found in the cache memory, content request information is broadcasted to other mobile devices 200 belonging to Local D2D communication. . If the corresponding content is found in another local mobile device 200 belonging to the Local D2D communication, the corresponding content is transmitted to the mobile user 100 who requested the content. On the other hand, when content cannot be found in all local mobile devices 200 , the content request information is transmitted to the BS 300 .

In the present invention, the application layer is implemented in the BS 300 to achieve the purpose of the ICN-based edge system. When the BS 300 receives a content request through an interest packet, it searches for content in its own content store (Contents Store, CS). If the corresponding content is found, the content is transmitted to the mobile user 100 . However, if the content is not found, the BS 300 transmits the corresponding Interest packet to the ICN routers (ICN enabled Content Routers) 400 located between the BS 300 and the core network 500 .

When the ICN router 400 receives the Interest packet, it checks the corresponding content in its cache memory. If the content is found, the content is transmitted to the BS 300 . If not found, the content request is transmitted to the provider 600 through the core network (Core Network) (500). The Provider 600 refers to an environment that provides content, and the Provider 600 may be an existing server that stores content or a cloud that provides content through the Internet. The Provider 600 finds the corresponding content and transmits it to the ICN router 400 . Through the above process, the quality of experience (QoE) of the mobile user 100 can be improved by reducing the delay time for searching the content and reducing the frequency of accessing the core network 500 .

On the other hand, dynamic content changes with time, and a request for dynamic content must be transmitted to the original server through the core network 500 . Accordingly, in the present invention, the name-based content prefetching strategy in the ICN router 400 known as the RAN Contents Router (RCR) connected to the BS 300 and the core network 500 is a name-based content prefetching strategy. made it possible The BS 300 determines the request frequency while continuously tracking the number of requests of the mobile user 100 and maintains a constant frequency threshold. When the number of requests for a specific content exceeds this threshold, a prefetching request is sent to the Provider 600 as shown next to the BS 300 of FIG. 1 to download the latest content in advance. As described above, in the present invention, since the existing server simultaneously prefetches and caches the BS 300 , the content can be quickly provided to the mobile user 100 without forwarding the request to the existing server.

Moreover, in the 5G network, there may be several BSs 300 according to the trend of the region. Therefore, in the present invention, all traffic from the BS 300 moves to the server through the ICN router 400 and the core network 500, so that the content is delivered before the core network 500 of the RCR directly connected to the core network 500. A strategy for prefetching was implemented. The RCR tracks the requests of all BSs 300 and checks the frequency of use of the content. However, since the RCR receives more traffic such as receiving a request from many BSs 300 , the RCR threshold was measured to be higher than that of the BS 300 . When the request for specific content exceeds the RCR threshold, a prefetching request is transmitted to the provider 600 as shown in front of the core network 500 of FIG. 1 to download the content in advance. Through this prefetching process, the latest content is preset in RCR. Such a mechanism may reduce the need for access to the core network 500 and, in turn, reduce the latency of the 5G network.

Hereinafter, such D2D communication and hierarchical name-based content prefetching strategy will be described in detail. First, limitations and assumptions in D2D communication will be described, respectively, and then a name-based prefetching mechanism for dynamic content will be described.

2 is a conceptual diagram illustrating a D2D communication setting process according to an embodiment of the present invention.

As shown in FIG. 2 , in the present invention, the D2D communication function takes an ad-hoc method so that two nearby devices can communicate with each other. However, D2D communication in a cellular network has several technical problems as follows.

The first is that when cellular communication devices communicate with each other via the BS 300 in a permitted bandwidth, Ad-hoc D2D communication must consider how the devices will be billed for local connections. In the initial 4G cellular network, the D2D communication function was not considered because communication could be made only through the BS 300 . Therefore, in the present invention, when the device acts as a relay for other devices, the focus is on how to protect data and to ensure data privacy. In addition, when the BS 300 is not involved, how to perform connection establishment, interference management, and resource allocation was also considered. As such, the existing limitations will be described below.

In the present invention, the following assumptions were made to realize the ICN-based D2D communication goal in the 5G network.

Hypothesis: To date, mobile devices in cellular networks only operate within the permitted bandwidth. The first assumption is an assumption related to D2D communication through an ad-hoc wireless network in a local environment. Since all devices are authenticated, there are no privacy concerns when passing data to another device. It is also assumed that the devices involved in this D2D communication do not charge the battery and bandwidth usage as data passes while transferring data. The communication environment is entirely based on the familiar device, which allows the free use of resources on other devices. This assumption supports the D2D communication element of the proposed structure.

When the mobile device/user needs content, the procedure proceeds as follows.

Step 1: The mobile user 100 transmits a content request message (Interest packet) to the nearest mobile device 200 .

Step 2: Upon receiving the content request, the nearest mobile device 200 returns the content according to the usual ICN method. If the requested content is found, the request is satisfied and data is sent to the requested device.

Step 3: If the content cannot be found in the D2D local mobile device 200, the Consumer node (mobile device) sends a request to the MEC server in the BS 300, which is 1 hop away.

Step 4: In this step, the Consumer starts communication with the BS 300 .

The Consumer starts communication with the BS 300, and the BS 300 checks the content availability using a hierarchical naming scheme when a request arrives. However, since the edge node BS 300 is an actual device, the ICN mechanism will not work smoothly, and the contents of the edge node cannot be verified through the ICN naming method.

Since the architecture of the present invention is ICN-based, ICN is implemented in the BS 300, and the ICN data structure is implemented in the application layer of the edge node to check the contents of the actual device through the ICN naming system. The reason for using ICN in the application layer is to realize ICN at the edge so that the edge node can follow the mechanism.

Hereinafter, a name-based prefetching strategy for dynamic content will be described.

1) Contents prefetching

The content prefetching strategy proposed in the present invention is based on the hierarchical naming scheme of ICN, and it can be applied to many use-case cases beyond the specific case of dynamic content mentioned in the present invention. . For example, dynamic content can be used on Twitter and YouTube.

In the present invention, in order to describe a use-case scenario, an International Sports Network (ESPN) is selected and described. In real-time games, team scores are dynamically updated on existing servers. All request messages for scores are forwarded to this server. In the present invention, a problem situation in which points are conceded in three games of soccer, cricket, and baseball is considered. A similar problem applies to the smart campus scenario in which IoT devices are connected to the cloud to transmit and receive related content. In the existing Multiple-access edge computing (MEC) system, when a user requests a score, the request message is transmitted to the existing server through the edge server and the core network 500 each time, and millions of users around the world are the same. Because it is possible to request real-time scores, there is a problem of high latency and overhead. In addition, the mobile user 100 may request static content that can be searched from the MEC server. In this case, the request messages of each mobile user 100 must reach the BS 300 . The device's caching capabilities allowed such static content to be retrieved from nearby devices. Finally, the dynamic content may be set to limit requests for access to the core network 500 in advance according to the prefetching strategy.

Hereinafter, the naming scheme will be described first and then the content prefetching strategy will be described.

1-1) Prefetching Naming Scheme

3 shows an example of the prefetching naming system proposed by the present invention. As shown in FIG. 3 , the prefetching naming scheme consists of several elements, and the first element shows a unique Provider Name called an ESPN server. The sub-element is the name of the content to be requested from the Provider 600, and the first sub-element is a game type such as soccer, baseball, or cricket. The second sub-component is the nature of the data, such as real-time scores and live streaming. The third subcomponent shows Teams (Pakistan, Korea). The final sub-component shows the time and time zone the Interest packet was sent. This naming scheme is flexible and can be changed according to user requirements.

For example, in a smart building, each node is typically placed to control various parameters such as temperature, pressure, humidity and air quality. The name of the provider 600 may be a smart campus using IoT, and may be a sub-element such as Hongik/building/Room425/Temperature/GMT showing the indoor temperature of Hongik University Building D. The system can also be used to perform simple tasks such as turning on lights and turning on and off air conditioners.

Next, we describe how the proposed name-based prefetching mechanism for dynamic content works.

1-2) Content Prefetching Strategy

4 is a view showing the overall structure of the ICN support 5G system including the mobile device 200, the MEC server, the ICN router 400, the core network 500, and the Provider 600 proposed in the present invention. Here, the MEC server is a server that processes content at the edge of the BS 300 and implements distributed edge computing, and the Provider 600 is a cloud that provides content through the Internet.

The left side of FIG. 4 shows the D2D communication setup as described above. The mobile device 200 checks the availability of content when sending a request to the MEC server. When the content is found, it is transmitted to the mobile user 100 who requested the content. However, the request for dynamic content must reach the cloud that is the provider 600 through the core network 500 . Therefore, the present invention proposes a content prefetching strategy for limiting access to the core network 500 . In order to rank the content according to the frequency of use and download the content from the server at the same time, the content prefetching strategy is applied in the ICN router 400 , which is an RCR node connected to the BS 300 and the core network 500 . The reasons for prefetching in two places and the details of each strategy are as follows.

1-3) Prefetching Strategy at the BS

First, in order to obtain the content with the highest frequency of use, the present invention enables a content prefetching strategy in the BS 300 . To this end, the number of requests from the mobile user 100 to the BS 300 was tracked using a table of edge nodes that calculates the frequency of all requests. In addition, a limit based on the popularity of content in each BS 300 was maintained. For example, if the request for cricket-related content exceeds or equals a certain threshold, the content is classified as popular. In the strategy proposed in the present invention, a threshold of 1000 requests for a specific request in each BS 300 is assumed. This threshold can be changed according to the use case requirements and an optimized value can also be used.

After receiving the interest packet from the mobile user 100 , the BS 300 checks the request frequency. When the mobile user 100 requests content whose frequency is equal to or exceeds the threshold, the BS 300 transmits a request including the name of the content and Greenwich Mean Time (GMT). For example, if the GMT of the edge node is 13:20:48, the latest content download request is transmitted to the original server after that time. When the content arrives at the BS 300 , the content storage CS of the BS 300 checks the available space. Content is stored to respond to future requests when there is enough space available, and unsolicited content is removed from the CS when there is not enough space.

For example, the frequency of requesting soccer scores from users around the world during the World Cup is quite high. The edge node transmits a request for the latest content to the server and starts downloading all the latest content from the original server in advance along with the timing information used to deliver the required content after a certain period of time. All these contents are delivered to the mobile user 100 through the mobile user 100 and are cached along the path, so the need to access the core network 500 is reduced, ultimately reducing the latency and traffic of the existing server. This will result in a higher cache hit ratio in the RAN.

However, in a 5G network, it is possible to have multiple BSs 300 . As shown in FIG. 4 , the MEC server is connected to the core network 500 through the ICN router 400 , so that all requests from the BS 300 may pass through the ICN router 400 connected to the core network 500 . . Since each BS 300 keeps a record of user requests, the request trends in different BSs 300 may be different. In this case, the RCR node, which is the ICN router 400, checks the accumulated popular content and determines the content to be fetched in advance from the existing server.

Next, we will refer to a simple scenario describing RCR nodes and traffic trends in various BSs (300).

1-4) Prefetching strategy at RAN content router (RCR)

It is assumed that three BSs 300 transmit data to the ICN router 400 disposed between the core network 500 and the BS 300 . In the case of the ICN router 400 directly connected to the core network 500 , it is assumed that the frequency of use of the end users is different in the three BSs 300 . For example, in BS-1, the requested frequency is 1000 points for cricket points, 500 points for soccer, and 700 points for baseball. If we assume that the BS threshold is 1000, requests greater than or equal to that threshold will appear as popular content. BS-1 sends a prefetch request for popular content (cricket in this example) to the existing server along with the corresponding GMT time.

Assume that requests arrive differently depending on the trend of BS-2 (eg 1000 soccer, 800 cricket, 600 baseball). BS-2 prepares the latest content for the soccer request in advance and stores it in the content store (CS) (depending on the amount of space available in the content store). In BS-3, the trend changes again (eg 1000 baseball, 400 cricket, 900 soccer). This means that BS-3 obtains the latest baseball content from existing servers in advance. When all such requests pass through the RCR node that is the ICN router 400 , the RCR node can check the popularity of the content because it is receiving a request from another BS 300 . In addition, the threshold value of the RCR node is different from the threshold value of the BS 300 because the RCR node may be connected to several BS servers in different trends.

4 shows a detailed structure in which several MEC servers are arranged, and all of them indicate that a request is received from the mobile user 100 . Then, the MEC server transmits such a request to the ICN router 400 disposed between the core network 500 and the BS 300 . As the traffic of all MEC servers goes through the RCR node, the traffic amount of the RCR node will be much higher than that of the BS 300 . Therefore, the RCR node checks the frequency count of all request messages requested from all BSs 300 and prepares in advance the latest popular content for storage in the CS.

Here, Least recently used (LRU) is adopted as the replacement policy, but other replacement policies such as least recently used (LFU) may also be adopted. In the example above, based on the score, the RCR node will receive 2200 baseballs 2300 soccer 2400 requests for cricket.

Assuming that the threshold is 2000 in the RCR node (which is higher than the threshold of the BS), the most frequent request (in this case, a request for soccer) will be preset to the existing server. This means that access to the core network 500 can be minimized and the content is pre-filled in the RCR node and the BS 300 . By restricting access to the core network 500 , it is possible to minimize latency and support dynamic content closer to the mobile user 100 .

Hereinafter, ICN application layer implementation for the BS 300 will be described.

2) ICN application layer construction for BS (Application layer ICN implementation for BS)

5 shows the application layer structure of the ICN for the BS according to the present invention, the ICN application layer structure for the BS 300 proposed in the present invention is an API layer, a CS implementation layer, a PIT Entries layer, a Forwarder layer and Data Access It consists of an application-layer ICN implementation that includes several sub-layers, such as layers. The implementation details of all lower layers are as follows.

2-1) API layer

The BS 300 device provides an API (Application Programming Interface) layer, and this API layer has two main roles: processing a request of the mobile user 100 and implementing the basic data structure of the ICN. All requests of the mobile user 100 (eg, from ndnSIM) are provided in the form of API, and reach the RequestHandler Controller of the Web API. Such a request is first converted into ICN Interest format within the request processing operation, and then transferred to the ICN Implementation class. ICN Implementation is a separate class inside the API layer that implements the basic algorithm of ICN. The ICN Implementation class processes Intereset based on the mechanism of ICN, then returns data to RequestHandler/RequestProcessing Action and delivers the result to the mobile user 100 .

2-2) CS implementation layer

The CS implementation layer plays a role in providing all necessary C# methods, such as checking CS data, listing the names of all content available in CS (SQL database), adding new content to CS, and deleting content to CS. All of these methods are written in the CSImplementation class that calls the Data Access Layer (DAL) to communicate with the SQL-based storage.

2-3) PIT Entries layer (PIT entries layer)

The PIT Entries layer contains all the necessary things, such as adding a new Pending Interest Table (PIT) to the table, setting event-based timers, associating PIT inputs with timers, removing PIT inputs after timer expires, removing PIT entries after receiving data packets, and retrieving all PIT entries. It serves to provide a function. In the present invention, a cache is used to store PIT items in the form of a C# list.

2-4) Forwarder layer

The responsibility of the Forwarder layer is to provide information on the node to be forwarded next for a given name. This layer was implemented as a Forwarder class with various C# methods such as adding new Forwarder items in JSON collection, retrieving forwarder items by name of Interest, removing items, updating and modifying existing items, etc. JSON file to storage system for storing forwarder items I used LINQ-To JSON C# library for communication between JSON file store and application.

2-5) Data access layer (DAL; Data access layer)

DAL is a layer that manages data. Typically, you use a database such as SQL, MySQL, Oracle MongoDB, JSON files, etc. In the implementation, a JSON file was used to store forwarder information and a SQL database was used as the content store (CS). A code module in the DAL is triggered when it needs to communicate with a SQL database or when it communicates with a JSON file. In both cases, each layer (CSImplementation or Forwarder) is derived. We used the Microsoft Entity framework to communicate with the SQL database. The Entity framework provides related libraries and executes related queries in the database to generate results for that application. In the present invention, a Database First approach is used. In addition, I used the LINQ-JSON library for JSON file communication, and used the Repository pattern to create a separate data access class for the SQL database and JSON file.

As described above, in the present invention, the BS 300 supports the D2D or ICN application layer, and by providing a content prefetching schema based on the ICN Naming technique, ICN and edge caching methods are applied to dynamic content to focus on information in 5G networks. It will be possible to enable networking-based edge computing.

The present invention is not limited to the above-described embodiments, and various modifications and variations can be made by those of ordinary skill in the art to which the present invention pertains within the scope of equivalents of the technical spirit of the present invention and the claims to be described below. Of course, this can be done.

100: Mobile User
200: D2D local mobile device) (Mobile Device)
300 : BS (Base Station)
400 : ICN routers (ICN enabled Content Routers)
500: Core Network
600 : Provider

Claims (9)

A step (a) of transmitting an interest packet requesting content by a mobile user (100) to a mobile device (200) performing local device to device (D2D) communication in a 5G network; If the local mobile device 200 that has received the interest packet finds the corresponding content, it returns the content to the mobile user 100 who requested the content, and if not found, the interest packet is delivered to the BS (Base Station) 300 performing edge computing Step (b) and; a step (c) of searching the CS (Contents Store) possessed by the BS 300 and transmitting the corresponding content to the mobile user 100 if it is found, and forwarding the Interest packet to the ICN router 400 if not found; The ICN router 400 searches the cache memory, and if the corresponding content is found, it is transmitted to the BS 300, and if it is not found, it is provided by requesting the Provider 600 through the Core Network 500 (d) ); as an information-centric networking-based edge computing activation method comprising:
The BS 300 tracks the request frequency according to the number of dynamic content requests of the mobile user 100, and when the number of requests for specific content exceeds a set frequency threshold, the Provider ( 600) to request prefetching of the content, receive the content, store it in the CS, and provide it when a request for the content comes from the mobile user 100,
The ICN router 400 tracks the frequency of content utilization according to the number of dynamic content requests from a plurality of BSs 300, and when the number of requests for specific content exceeds a set frequency threshold, the core network 500 After requesting the Provider 600 for prefetching of the corresponding content through the Provider 600, receiving the corresponding content and storing it in the memory, when a request for the corresponding content comes from the BS 300, it is provided,
The information-centric networking-based edge computing activation method, characterized in that the frequency threshold set in the ICN router (400) is set higher than the frequency threshold set in the BS (300).
delete The method of claim 1,
When the BS 300 requests dynamic content through the ICN router 400, it transmits Greenwich Mean Time (GMT) information indicating the request time together with the requested content name,
When the requested content is received from the provider (600), the information-centric networking-based edge computing activation method, characterized in that the content is stored or discarded according to the available space of the CS.
delete The method of claim 1,
The ICN router 400 receives the most recently used (LRU; Least recently used) content among the requested content from the Provider 600 and stores it in a memory.
delete The method of claim 1,
The mobile device 200 performing the D2D communication performs ad-hoc communication with another mobile device based on ICN, and broadcasts and delivers the content request information transmitted from the mobile user 100. A method of activating edge computing based on information-centric networking characterized by.
The method of claim 1,
The BS 300 builds an ICN application layer so that the content can be verified through the ICN naming system, and when a content request is received from the mobile device 200, the content availability is checked using a hierarchical naming scheme. Information-centric networking-based edge computing activation method, characterized in that to do so.
8. The method of claim 7,
The ICN-adaptive layer of the BS 300 is
An API layer that receives and processes content requests from the mobile user 100, and
A CS implementation layer that registers and manages content in CS, and
PIT entries layer that adds PIT (Pending Interest Table) to table, sets event-based timer, connects PIT input and timer, removes PIT, and performs PIT item search;
A Forwarder layer that provides information on the next forwarded node for a specific name (Forwarder layer);
Information-centric networking-based edge computing activation method, characterized in that it includes a data access layer that manages forwarder information storage and CS stored data.

Images