KR20190056957A - COMPUTING SYSTEM AND METHOD FOR INTELLIGENT IoE INFORMATION FRAMEWORK - Google Patents

Abstract

According to the present invention, a computer system for an intelligent Internet of everything (IoE) information framework comprises: a plurality of edge computing platforms individually including an edge networking node for supporting connectivity with a terminal, an intelligent computing node that analyzes data of the terminal through the edge networking node, and an edge gateway node for information-centric networking within the edge computing platform and Internet protocol networking outside the edge computing platform; and an edge identifier management node for generating mapping information mapping the plurality of edge computing platforms on the terminal and an identifier of data.

Description

TECHNICAL FIELD The present invention relates to a computing system and method for intelligent IoE information framework,

The present invention relates to a computing system and method for an intelligent IoE information framework.

Recently, a variety of techniques for analyzing and processing the collected data have been developed and provided to provide connectivity for collecting a large amount of data from a large number of IoT objects.

However, there is a limitation in handling data of numerous IoT objects or supporting scalable connectivity only by using technologies such as IPv6, CoAP, and a central cloud server, which are IP technologies such as the Internet.

In addition, existing IoT platforms are also developed specifically for different technology and service domains, and it is also difficult to develop a solution that integrates them.

Therefore, there is a need for an integrated framework for effectively supporting connectivity and mobility in a resource constrained environment and a networking constrained environment of many IoT objects, and collecting and processing a large amount of data generated from these objects through a distributed structure.

The embodiment of the present invention is based on an intelligent networking platform based on an information-centric networking for providing connectivity to support a wide range of objects with scalability, mobility and low latency, and a distributed Edge / Fog intelligent networking Provides a computing system and method for an intelligent IoE information framework that includes the ability to manage and control the platform.

It should be understood, however, that the technical scope of the present invention is not limited to the above-described technical problems, and other technical problems may exist.

According to a first aspect of the present invention, there is provided a computing system for an intelligent IoE information framework, comprising: an edge networking node for supporting connectivity with a terminal; A plurality of edge computing platforms each including an intelligent computing node for analyzing data and an edge gateway node for information-centric networking within the edge computing platform and for Internet protocol networking outside the edge computing platform; And an edge identifier management node for generating mapping information mapping terminal and data identifiers.

The data of the terminal is stored in a temporary storage and categorized according to types of data and services, and the classified data can be stored in a data storage in the edge computing platform and analyzed through the intelligent computing node.

The edge networking node can be connected to the terminal through bootstrapping for transmitting the response message after registering the identifier information upon receiving the identifier information from the terminal.

Wherein the edge networking node sends and registers identifier information of the terminal to the intelligent computing node and the intelligent computing node transmits and registers its own identifier information and the identifier information of the terminal to the edge identifier management node, Can be maintained.

Wherein the edge networking node of the edge computing platform of the plurality of edge computing platforms receives the message requesting the data of the IoT terminal from the user terminal and transmits the message to the IoT terminal from the edge identifier management node via the intelligent computing node, The intelligent computing node of the IoT terminal side edge computing platform receives the identifier information of the intelligent computing node of the IoT terminal side edge computing platform in which the data of the terminal is stored, May be received and provided to the user terminal.

Wherein the edge gateway node receives the message requesting the data of the IoT terminal from the cloud server and converts the message into the message based on the information centering network and transmits the message to the intelligent computing node, And converting the received data into a message based on the Internet protocol networking upon receiving at least one of the analyzed data and providing the message to the cloud server.

Wherein the edge gateway node receives a data processing model corresponding to the service profile from the cloud server as it transmits the service profile to the cloud server through the Internet protocol networking and the intelligent computing node is based on the received data processing model To analyze the data.

Data analyzed through the intelligent computing node may be provided for operation of the terminal or may be transmitted to the cloud server through the edge gateway node.

Wherein when the user terminal connected to the first edge networking node in the edge computing platform changes to a connection to the second edge networking node, the second edge networking node transmits, from the edge identifier management node to the intelligent computing node, And transmits the message requesting the data of the IoT terminal to the intelligent computing node based on the received identifier information so as to forward one or more of the data of the IoT terminal and the analyzed data have.

The second edge networking node registers the IoT terminal with the intelligent computing node when the IoT terminal connected to the first edge networking node in the edge computing platform changes to the connection to the second edge networking node, The intelligent computing node may register the IoT terminal with the edge identifier management node.

The intelligent computing node can acquire data of the IoT terminal by transmitting a message requesting data of the IoT terminal to the second edge networking node.

The edge identifier management node may generate the mapping information by mapping the identifier information of the IoT terminal and the identifier information of the intelligent computing node to process the data request for the IoT terminal of the user terminal.

Wherein when the IoT terminal of the plurality of edge computing platforms moves from the first edge computing platform to the second edge computing platform, the edge networking node of the second edge computing platform registers the IoT terminal with the intelligent computing node, The intelligent computing node of the two edge computing platform may register the IoT terminal with the edge identifier management node.

Wherein the intelligent computing node of the second edge computing platform provides the identifier of the IoT terminal and its identifier to the edge identifier management node and the edge identifier management node maps the identifier of the IoT terminal to the identifier of the intelligent computing node And generate the mapping information.

The intelligent computing node of the second edge computing platform transmits a message for requesting the data of the IoT terminal through the edge networking node of the second edge computing platform so that the IoT terminal moving to the second edge computing platform And provide the data to an intelligent computing node of the second edge computing platform via an edge networking node of the second edge computing platform.

As the IoT terminal completes the move to the second edge computing platform, the intelligent computing node at the user terminal receives the identifier of the intelligent computing node of the second edge computing platform as an edge identifier management node, Lt; RTI ID = 0.0 > IoT < / RTI >

Wherein the edge networking node of the second edge computing platform is configured to request, via a request to an edge networking node of the first edge computing platform, data of the IoT terminal stored in the intelligent computing node of the first edge computing platform One or more of which may be forwarded to the intelligent computing node of the second edge computing platform.

According to a second aspect of the present invention, there is provided a computing method for an intelligent IoE information framework, comprising: receiving data of the terminal through an edge networking node supporting connectivity with the terminal; Collecting and analyzing the received data of the terminal by the intelligent computing node and transmitting the analyzed data to the cloud server through the edge gateway node or providing the analyzed data as operation data of the terminal through the intelligent computing node, Wherein the intelligent IoE information framework comprises a plurality of edge computing platforms each including the edge networking node, the intelligent computing node and the edge gateway node, and a plurality of edge computing platforms, mappings mapping identifiers of the terminal and data And an edge identifier management node for generating information.

According to a third aspect of the present invention, an edge computing platform apparatus in an intelligent IoE information framework includes an edge networking node for supporting connectivity with a terminal, an intelligent computing node for analyzing data of the terminal through the edge networking node, And an edge gateway node for information-centric networking and Internet protocol networking with the outside world.

According to any one of the above-mentioned objects of the present invention, since an information-based networking (ICN) is included to directly provide connectivity to IoT terminals in an IoT environment, which is a resource limitation, existing TCP / IP, TLS, DTLS To reduce the number of message exchanges and code overhead, to check the integrity of IoT data, and to provide connectivity to direct Layer 2 networks.

In addition, URI types used in existing application services can be efficiently used by using a technique designed to be directly available in networking and exchange of interest / content object messages.

In addition, intelligent IoE information framework based on edge computing platform that collects, analyzes, and refines data generated by IoT terminals based on connectivity enables networking technology to deliver data as well as various IoE service domains Data can be processed, refined and processed, and an operation command of the actual IoT terminal can be provided.

Accordingly, the present invention can process a large number of IoT terminals and data in a scalable manner, support a low-latency service capable of supporting a time critical IoE service such as a plant safety service domain, User mobility, IoT device mobility, and data mobility.

In addition, the intelligent IoE information framework according to the present invention can provide modules and components for analyzing and processing IoE data generated in real time by mounting an artificial intelligence engine such as machine learning and reinforcement learning.

1 is a block diagram of a computing system for an intelligent IoE information framework in accordance with an embodiment of the present invention.
2 is a diagram showing a data processing flowchart through the intelligent IoE information framework.
3 is a diagram illustrating nodes and components for data and networking processing on an intelligent IoE information framework.
4 is a diagram for explaining an intelligent IoE information processing process.
5 is a diagram for explaining a process of utilizing an IoE data analysis model provided through a cloud server.
6 is a diagram for explaining an intelligent IoT / IoE networking support process.
7 is a diagram for explaining a boot spreading process between an IoT terminal and an edge networking node.
8 is a diagram for explaining a boot spreading process between an edge networking node and an intelligent computing node.
9 is a diagram for explaining a process of acquiring data of an IoT terminal by a user terminal.
10 is a diagram for explaining a process of acquiring data of an IoT terminal by an external cloud server.
11 is a diagram for explaining a process of supporting mobility and seamless handover of a user terminal.
12 is a diagram for explaining a mobility support procedure of an IoT terminal on an edge computing platform.
13 is a diagram for explaining a data mobility support procedure and a seamless handover procedure between a plurality of edge computing platforms.
14 is a flowchart of a computing method for an intelligent IoE information framework.
15 is a diagram showing a data processing scenario in two domains including the same data.
FIG. 16 is a view for explaining contents supporting a privacy guarantee service in a smart healthcare system. FIG.
17 is a diagram for explaining a process of supporting urgent message processing in the smart factory system.
18 is a diagram for explaining a data path bypass processing support process through a mobile terminal.
FIG. 19 is a diagram for explaining a process of supporting creation of a machine learning model by data classification and combination. FIG.
20 is a diagram for explaining a virtualization-based platform and framework.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly explain the present invention in the drawings, parts not related to the description are omitted.

Whenever a component is referred to as " including " an element throughout the specification, it is to be understood that the element may include other elements, not the exclusion of any other element, unless the context clearly dictates otherwise.

The present invention relates to a computing system (1) and method for an intelligent IoE information framework.

The current resource-constrained IoT environment includes a limited environment of the device itself and a limited environment of networking technology. In the case of devices, it means a limited environment of CPU performance and memory capacity. In the case of networking technology, a limited environment of wireless networking technology such as LPWA (Low Power Wide Area Network) and LLN (Low Power Lossy Network) .

Therefore, in order to provide connectivity in the resource-limited IoT environment, the conventional technology supports networking and data transmission services by mounting a TCP / IP or UDP / IP stack, which is an existing Internet technology.

In addition, the prior art uses IPv6 instead of IPv4, which is an existing Internet protocol, to assign IP addresses to a large number of IoT devices, and a convergence layer between the Ipv6 layer and the MAC layer is newly provided so as to mount a new wireless technology such as LPWA Standardization and development, and CoAP and other technologies that make HTTP lightweight for the application layer are provided.

However, this conventional IPv6-based connectivity support still embraces the shortcomings of IP technology. In particular, in the case of automobiles, in order to store internal sensors and image data in a central server, mobility support is required. However, in existing Internet technologies, a separate mobility support protocol has to be additionally installed and the infrastructure still needs to be additionally constructed .

In addition, existing Internet technology is a protocol technology designed on the assumption of fixed terminals and servers, and overhead of message transmission occurs like a multicast protocol for connection establishment. In particular, the existing Internet technology has caused the overhead of sending and receiving additional messages for TLS, DTLS, etc. for the communication channel security of the application service.

In addition to this, it is expected that more than 10 ¹⁰ IoT devices will be added in the future, and the need for scalability, effective networking technology, and computing platform technology to handle a large amount of data generated through these terminals is emerging. Accordingly, an edge / fog computing technique for processing the user and the IoT device close to each other has been popularized.

In the case of Edge Computing, Mobile Edge Computing (MEC) technology for mobile communication corresponds to this, and devices, access gateways and base stations directly perform functions necessary for services. In the case of Fog Computing, there is a recent technical discussion as a distributed computing platform located between devices (especially IoT devices) and a central cloud computing architecture provided over the Internet.

However, in many cases, edge computing is often used together with technical rather than technical difference between edge computing and fog computing. Therefore, the edge computing described in the present invention may be a concept including fog computing technology.

In the context of the various edge / fog computing technologies and the numerous IoT platforms developed for the various existing IoT services, we are constantly engaged in additional R & D and standardization activities that support integration and interoperability. Is in progress.

For example, many IoT services such as Smart City, Smart Factory, and Smart Healthcare are developing services and technologies centered on separate IoT platforms. However, as a result, numerous IoT platforms have become obsolete, which in turn does not share a common technology, which is expected to continue to increase development and maintenance costs.

In particular, IoT devices have been developed to operate with gateways or actuators, and have the disadvantage that they are fixed with proprietary services together with specific company products, which is a major obstacle to industrialization.

On the other hand, there are activities to support the integration of IoT platform through international standardization, and platforms such as oneM2M and OCF IoTivity are being developed, but it is not yet an integrated IoT platform.

In conclusion, we can not solve the problem of processing data with a large number of IoT terminals or supporting scalable connectivity with only IPv6 technology such as Internet, IPv6, CoAP, and central cloud server technology. In addition, existing IoT platforms are also developed for different technologies and service domains, and therefore, it is still difficult to develop integrated solutions.

Therefore, it is possible to provide a method and apparatus for supporting efficient connectivity and mobility in a resource constrained environment and networking constrained environment of numerous IoT devices, and an edge / fog computing technique for collecting and processing a large amount of data generated in these devices through a distributed structure Data collection and processing techniques in areas close to and in data generation areas are needed.

In addition, for compatibility with existing IoT platforms or cloud servers, it provides networking and security technologies such as RESTful, which is used in application services such as the Web, and can be applied to various service domains. There is a need for an integrated framework.

In order to solve such a problem, a computing system and method for an intelligent IoE information framework according to an embodiment of the present invention provides scalability, mobility, and low latency to ¹⁰ or more things Based on an information-centric networking-based edge computing platform for delivering connectivity, it can collect, store, process, and control information from objects or send analysis results to a cloud server, You can manage and control the platform.

Hereinafter, a computing system 1 for an intelligent IoE information framework according to an embodiment of the present invention will be described with reference to FIG. 1 to FIG.

1 is a block diagram of a computing system 1 for an intelligent IoE information framework in accordance with an embodiment of the present invention. 2 is a diagram showing a data processing flowchart through the intelligent IoE information framework.

The computing system 1 for the intelligent IoE information framework according to one embodiment of the present invention includes a plurality of edge computing platforms 100 and an edge for interworking between the other edge computing platforms 100 developed and deployed on the edge network, And an identifier management node (200).

At this time, the intelligent IoE information framework includes the Internet (b) in which the IoT resource-restricted terminals 10 and the proprietary thing networks (a) composed of the wireless network section and the general cloud server 20 are located, Respectively.

The terminal described in the embodiment of the present invention includes both the IoT terminal 10 and the user terminal 30 and hereinafter the IoT terminal 10 and the user terminal 30 are separated .

Each of the plurality of edge computing platforms 100-1, 100-2, ..., 100-n includes an edge networking node 110, an intelligent computing node 120, and an edge gateway node 130, respectively.

In this case, each node 110, 120, and 130 in the embodiment of the present invention may be configured as independent physical hardware, as well as virtualized software in one or more physical hardware . That is, each node 110, 120, 130 in the edge computing platform 100 configured with one physical server can be configured as a separate virtualized program. In addition, the components 140, 150, and 160 connected to the nodes 110, 120, and 130 may be configured as independent hardware or may be included in each node and configured as virtualized software.

In this case, when the server is constituted by physical hardware, the server may include a memory (not shown) storing a program for performing each function and a processor (not shown) executing a program stored in the memory.

At this time, the memory is collectively referred to as a non-volatile storage device and a volatile storage device which keep the stored information even when power is not supplied. For example, the memory may be a memory card such as a compact flash (CF) card, a secure digital (SD) card, a memory stick, a solid-state drive (SSD) A magnetic computer storage device such as a NAND flash memory, a hard disk drive (HDD), and the like, and an optical disc drive such as a CD-ROM, a DVD-ROM, and the like.

Information-centric networking (ICN) between the functional nodes 110, 120, and 130 and the components 140, 150, and 160 in accordance with the actual node development type, Technology, or can be connected through IPC (Inter-Process Communication) using IP technology.

Through this connection, an embodiment of the present invention connects each object network 110, 120, 130 and the components 140, 150, 160 with each object network to collect IoT data through it, And the like.

It is also possible to transmit data or analyzed data of the IOT terminal 10 to the cloud server 20 via the Internet b or directly control the operation of the terminals 10, It is applicable to IoT / IoE service domain.

First, the edge networking node 110 performs the function of enabling the information-centric networking node to support the connectivity to the IOT terminals 10. That is, the edge networking node 110 according to an exemplary embodiment of the present invention is a networking node that supports the existing ICN as a base and supports direct connectivity through the Layer 2 MAC of the OSI communication layer without the IP protocol .

In addition, TCP, UDP, and CoAP can be supported together with the IP protocol for connection with various IoT terminals 10 and interoperability with the network.

The edge networking node 110 is connected to the edge identifier management node 200 via the control channel as a networking node for the object network of the edge computing platform 100. The edge networking node 110 does not manage the statuses of all the terminals 10 every time the terminals 10 move through the channel but the identifier and data information of the terminal 10 are transmitted to the edge identifier management node 200 Thereby managing the non-stateless terminal 10 and the mobility of the data.

Meanwhile, a plurality of edge networking nodes 110 may be provided on one edge computing platform 100.

The intelligent computing node 120 collects the data of the IoT terminal 10 through the edge networking node 110 and analyzes the collected data. That is, the intelligent computing node 120 directly collects data of the IOT terminals 10, analyzes and processes the data, and then determines a processing procedure.

For this, the intelligent computing node 120 may mount a platform such as a machine learning. In addition, the intelligent computing node 120 dynamically creates an intelligent computing node 120 in the form of a virtual machine so that the physical nodes can be linked through a clustering method for scalability and can perform various data processing functions as needed It is also possible to support it.

Meanwhile, data generated in the IoT terminal 10 such as an actual sensor may include not only sensor information but also media data, real time data, and the like. For example, when the data generated in the IOT terminal 10 is real-time data sequentially generated, the intelligent computing node 120 can support functions such as online traffic analysis (real-time message processing) have.

For this, an embodiment of the present invention may include an intelligent information processing management unit 140 as a component. The intelligent information processing management unit 140 intervenes in the overall process of transmitting, analyzing and processing the IoT data to support real-time message processing and analysis. And performs connection, management, and control functions between the nodes 110, 120, and 130 in the edge computing platform 100.

Meanwhile, the intelligent information processing management unit 140 may be configured in a form independent of the intelligent computing node 120, and the functions may be performed in the intelligent computing node 120.

Although the data analyzed by the intelligent computing node 120 is meaningful to some extent, it is necessary to transmit and collect the analyzed data to the external cloud server 20 and analyze and process the data in a large scale.

For this purpose, the edge gateway node 130 supports an object network, an information-centric networking technology (ICN) commonly used in the edge computing platform, and Internet protocol networking outside the edge computing platform.

That is, the edge gateway node 130 interprets a message for data processing of the cloud server 20 located outside the edge computing platform 100, and performs a function of transmitting the analyzed result quickly.

In the edge computing platform 100, the intelligent information processing manager 140 described above as components other than the respective nodes 110, 120, and 130, a data storage unit 150 for storing analysis data, And a reward / operation unit 160 for determining a policy through a reward.

It is possible to collect and analyze the data received from the IOT terminals 10 and transmit the result to the cloud server 20 immediately. However, the result may be stored in the data storage unit 150 according to the status of the related service.

Alternatively, the reward / operation unit 160 may provide a reward for taking an actual action using a service-specific policy, and may transmit an operation command to the IoT terminal 10 via the edge networking node 110. [

Finally, the edge identifier management node 200 generates mapping information mapping a plurality of edge computing platforms 100, the terminal 10, and an identifier of data. That is, the edge identifier management node 200 is a node for supporting the mobility of the IoT terminals 10 and the mobility of data between the plurality of edge computing platforms 100, and identifies the IoT terminal 10 and the data through an identifier can do.

It also generates mapping information for dynamically mapping and managing identifiers identifying the IOT terminal 10 and data and the edge computing platform 100, and in particular the edge networking node 110 within the edge computing platform 100 So as to support the mobility of the terminal 10 and the data.

The processing flow of data applied to the IoT / IoE service domain ultimately through the computing system 1 for the intelligent IoE information framework having the configuration as shown in FIG. 1 is as shown in FIG.

IoT terminals 10 such as sensors generate single unit data D1 such as temperature and humidity and this single unit data D1 has raw data D1 ).

The computing system 1 for the intelligent IoE information framework according to an embodiment of the present invention collects (P1) the row data D1 for different types, meanings, and environments generated from the respective IoT domains And analyze data (P2) for providing IoT / IoE service through analysis and refinement of the collected raw data D1.

The computing system 1 for the intelligent IoE information framework is configured to generate an 'action command P3' such as 'transfer P2' of the analysis data D2 to the cloud server 20 or directly apply it to the IoT domain, For example. Therefore, the intelligent IoE information framework according to an embodiment of the present invention may be a framework for intelligently processing and applying data differently from the conventional IoT data processing method.

Hereinafter, the specific configuration and functions performed in the computing system 1 for the intelligent IoE information framework will be described for each item.

[Intelligent IoE  Information Framework  Configuration of data and networking components on the network]

The intelligent IoE information framework according to an embodiment of the present invention can provide not only the function of providing IoE information processing as shown in FIG. 1, but also the connectivity for providing data from the object network. It is also contemplated that networking between each of the configuration nodes 110, 120, and 130 and the components 140, 150, and 160 within the intelligent IoE information framework and networking with the cloud servers 20, As shown in FIG.

3 is a diagram illustrating nodes and components for data and networking processing on an intelligent IoE information framework.

The edge networking node 110 is responsible for data collection of the IOT terminal 10 and supports the connectivity of the IOT terminals 10.

The edge gateway node 130 transmits data analyzed by the intelligent computing node 120 and supports connectivity with the cloud server 20 via the external Internet b.

The intelligent computing node 120 responsible for intelligent data processing includes functions for analyzing and processing most of the data. The intelligent computing node 120 can include a component of the intelligent information processing management unit 140 that performs management functions because it can allocate and provide computing resources by clustering the edge terminals 10 as needed. The intelligent information processing management unit 140 may be operated as a dedicated server device such as a real bare metal server.

In addition, intelligent IoE information framework adopts networking technology based on ICN technology basically for connection between nodes, and an edge identifier management node which dynamically maps an identifier required for networking and an identifier according to various purposes 200).

Hereinafter, functions performed by each node and component will be described in detail.

[Intelligent About IoE  Processing procedure]

 4 is a diagram for explaining an intelligent IoE information processing process.

The computing system 1 in the intelligent IoE information framework according to an embodiment of the present invention processes data (D1, raw data or single unit data) generated from the IoT terminals 10 through the procedure shown in FIG. 4 do.

First, the data D1 provided from the IoT terminal 10 needs to be analyzed and refined in real time. Accordingly, the intelligent computing node 120 stores the data D1 generated in real time in a temporary storage such as a memory queue by utilizing information such as a service protocol, a data profile, and identifier information (UID type data identifier). Then, the data stored in the temporary storage are subjected to machine learning and artificial intelligence platform-based analysis. The intelligent computing node 120 can analyze the type of data and the type of service in this process (Classification process, S11).

The filtered data is filtered together with the service model using the existing IoT terminal 10 or the cloud server 20 by filtering the data that can be analyzed only by cumulatively accumulating the data continuously (S11) ) (S12, S16).

Thereafter, the real-time data that has been pre-processed is stored in the data storage unit 150 according to the analyzed type (S13), and the stored data is accumulated and can be analyzed once again through the machine learning and artificial intelligence framework (S14). This analysis process may be simply extracting changes in sensor data per second, minute, or hour depending on the service domain. Alternatively, using the deep learning method using a cloud server, the user may be provided with a learned model (model serving) and analyze the sensor data generated in real time. At this time, the learning model provided is periodically updated with a new model .

However, in one embodiment of the present invention, the intelligent computing node 120 is not limited to this functionality alone, and can be extended to processing data on the edge computing platform 100 according to service requirements utilizing service domain information.

Unlike the initial raw data D1, the analyzed data D2 generated through the analysis process may be generated by combining additional metadata. The analyzed data D2 may be transmitted to the cloud server 20 through the classification process S15 or may be temporarily stored at S17 or provided as an operation command of the IoT device 10 S18).

The data processing process shown in FIG. 4 occurs in the intelligent IoE information framework in the computing system 100 according to an embodiment of the present invention, and can be applied to smart health care, smart building, smart factory, etc. Do.

[ cloud  Dynamic using models IoE  Apply Data Processing Model]

5 is a diagram for explaining a process of utilizing the IoE data analysis model provided through the cloud server 20.

The computing system 100 in the intelligent IoE information framework according to an embodiment of the present invention uses the intelligent artificial intelligence platform to generate the raw data D1 generated in real time in the intelligent computing node 120 as shown in FIG. Analysis and processing. In addition, the intelligent computing node 120 may dynamically receive and execute the IoE data analysis model M specialized in the service domain provided by the cloud server 20 or the like.

Specifically, in one embodiment of the present invention, the intelligent computing node 120 may analyze the service profile or send the service profile directly to the cloud server 20 via the edge gateway node 130. At this time, the service profile may include input data and output data, and description contents for the operation of the IO terminal 10. [

That is, the intelligent computing node 120 may send a request / response message exchange of the IoE data processing model M corresponding to the service profile to the cloud server 20 via the edge gateway 130 (S21) The cloud server 20 can find and provide an optimal IoE data processing model M based on the service profile (S22).

These request / response messages can be implemented in the form of a RESTful message exchange.

Meanwhile, the process of transmitting / receiving the IoE data processing model M can be implemented using an existing communication process, and the received IoE data processing model M can be executed through the physical or virtualized intelligent computing node 120 have. Or implemented as a separate virtualized intelligent computing node 120 and executed.

The process of analyzing the low data D1 after the IoE data processing model M is applied and applied (S23) is the same as that of FIG.

As described above, in one embodiment of the present invention, the intelligent computing node 120 can be configured using virtualization technology for continuous execution by providing a dynamic IoE data processing model M as shown in FIG. 5 And exchange between the existing IoE data processing model and the new IoE data processing model.

That is, one embodiment of the present invention not only utilizes the IoE data processing model M already provided through the interworking of the cloud server 20, but also uses the IoE data processing model M periodically through the reward / The generated data can be analyzed to provide a cyclic model that allows the IoE data processing model (M) to be changed according to the situation.

[Intelligent Information Networking Procedures]

6 is a diagram for explaining an intelligent IoT / IoE networking support process. 7 is a diagram for explaining a boot spreading process between the IOT terminal 10 and the edge networking node 110. FIG. FIG. 8 is a diagram for explaining a boot spreading process between the edge networking node 110 and the intelligent computing node 120. FIG.

First, in order to support ICN-based connectivity between the IoT terminal 10 and the edge networking node 110, the IoT terminal 10 transmits its profile to the edge networking node 110 (S31). At this time, the profile transmitted by the IoT terminal 10 may include information such as an identifier, a data identifier, and a service type of the IoT terminal 10 itself configured in URI form.

Prior to this, when the Layer 2 technology is directly used, the connection between the IoT terminal 10 and the edge computing node 110 can be established when the node is booted for the first time. This process will be described with reference to FIG. 7 which will be described later.

At the same time, between the edge computing node 110, the intelligent computing node 120, and the edge identifier management node 200, a bootstrapping process as shown in FIG. 8 is performed for each configuration. Through this process, ICN-based connectivity between the IoT terminal 10 and the edge networking node 110 and the intelligent computing node 120 can be ensured.

The IoT data may then be collected by the intelligent computing node 120 via a simple one-time exchange of the Interest / Content Object, which is the basic communication procedure of the ICN. In this case, unlike the existing Internet, an embodiment of the present invention generates signature information for access rights of the IOT terminal 10 and the data itself, instead of exchanging security messages for communication channels such as TLS and DTLS, In this way, data can be freely exchanged and validity can be checked at the same time (S32, S33).

Also, the edge networking node 110 generates signature information of the IoT terminal 10 connected thereto based on the public-key, and an access control function may be performed based on the signature information (S34).

At this time, since the IoT terminal 10 has a restriction on the security for direct security, the edge computing node 110 acts as a proxy in the case of access control. Accordingly, the edge computing node 110 may use the identifier of each IoT terminal 10 as a seed value, and separately generate and manage the signature information of the IoT terminal 10 using the public-key.

The data collected through the intelligent computing node 120 is transmitted to the cloud server 20 through the data processing process described above or transmits an operation command to the IoT terminal 10 again. In this case, the edge gateway node 130 plays the role of transmitting to the cloud server 20 on the Internet (S35), and the edge networking node 110 transmits again an identifier-based operation command directed to the IoT terminal 10 (S36).

The bootstrapping process for connection between the identifier based IoT terminal 10 and the edge networking node 110 and the bootstrapping process between the edge networking node 110 and the intelligent computing node 120 are shown in FIGS. Respectively.

Since both the IoT terminal 10 and the edge computing node 120 are based on identifier-based communication (ICN technology), an ICN stack must be installed in each node. In addition, identifiers must be allocated among the nodes through the function of setting and generating identifiers.

Generally, in the bootstrapping process, a hierarchical identifier is used together with a URI, and the IoT terminal 10 generating the data by exchanging identifier information (advertisement / response) transmits its identifier and data identifier to the edge networking node (110). Thereby sending an Interest message to the IoT terminal 10 which generates data at the internal network node, such as the edge networking node 110 or the intelligent computing node 120.

7 shows a connection and an identifier allocation procedure between the IOT terminal 10 and the edge networking node 110 that generate data through the IEEE 802.11 WiFi network.

The IOT terminal 10 automatically generates an identifier of data to be generated along with a value obtained by hashing the EUI-48 data based on the Ethernet MAC address of the IOT terminal 10 and transmits an identifier advertisement message to the edge networking node 110 (S41).

The edge networking node 110 (corresponding to the AP in the example of FIG. 7) receives the identifier, registers the identifier of the IoT terminal 10, updates the routing path for data transmission and reception, and transmits a response message to the IoT terminal 10 (S42).

The edge networking node 110 may then request data using the identifier of the IoT terminal 10 (S43).

FIG. 8 illustrates a procedure for supporting connection between an edge networking node 110 and an intelligent computing node 120 on an intelligent IoE information framework according to an embodiment of the present invention.

The edge networking node 110 can receive the identifier of the IOT terminal 10 through the bootstrapping process between the IOT terminal 10 and the edge networking node 110 as shown in FIG. .

The edge networking node 110 then registers the identifier of the IOT terminal 10 with the intelligent computing node 120 on its edge computing platform 100 (S52, Registration process). Through the registration process, the intelligent computing node 120 can acquire the information of the IoT terminal 10 and the data of the IoT terminal 10 existing in the domain.

Then, the intelligent computing node 120 registers the identifier of the IoT terminal 10 and its own identifier in the edge identifier management node 200 to maintain the mapping information (S53). This information can be used to support seamless handover of data when the IOT terminal 10 moves later.

As the registration is completed, the edge identifier management node 200 transmits the response message to the intelligent computing node 120 and the intelligent computing node 120 transmits the response message to the edge networking node 110 respectively (S54, S55).

The subsequent steps will be described later in the section on mobility.

[ IoT  Data collection and use of terminals]

Hereinafter, a process in which the user terminal 30 in the edge computing platform 100 and the cloud server 20 located in the outside obtain data of the IOT terminal 10 will be described with reference to FIG. 9 and FIG.

FIG. 9 is a diagram for explaining a process of acquiring data of the IoT terminal 10 by the user terminal 30. FIG.

Unlike the general IoT terminal-cloud server model or the typical ICN model, an embodiment of the present invention can provide data through the edge computing platform 100. Meanwhile, the user terminal 30 may refer to a general user, but the edge computing platform 100 located in another domain may correspond to the user.

Accordingly, one embodiment of the present invention is an IoT terminal side edge computing platform 100-T for generating data around the IoT terminal 10, a user terminal 30 consuming data around the user terminal 30 The edge computing platform 100-C may be located as shown in FIG.

At this time, the intelligent computing node 120-C in the edge computing platform 100-C on the user terminal side can be operated in an asymmetric manner in which the intelligent computing node 120-C does not operate largely. The intelligent computing node 120-C in the user terminal side edge computing platform 100-C can be used to support seamless handover among the mobility of the user terminal 30, .

The data of the IoT terminals 10 that generate data by the networking settings of Figs. 7 and 8 are accumulated in the intelligent computing node 120-T on the IoT terminal side edge computing platform 100-T (P4) .

The user terminal 30 sends a request message (Interest) for data to the edge networking node 110-C in the edge computing platform 100-C of the user terminal side to obtain the data of the IoT terminal 10 S61), and the edge networking node 110-C receives the corresponding request message (Interest).

Next, the edge networking node 110-C determines a corresponding request message (Interest) and transmits a control message for the identifier query to the edge identifier management node 200 (S62). The edge networking node 110-C is connected to the edge computing node 100-C via the intelligent computing node 120-C, which stores data generated by the IoT terminal 10 from the edge identifier management node 200 -T) < / RTI > of the intelligent computing node 120-T.

The user terminal side edge computing node 110-C then transmits the data of the IoT terminal 10 and the analyzed data 120-T to the intelligent computing node 120-T of the IoT terminal side edge computing platform 100-T based on the identifier information (Interest) (S63). The intelligent computing node 120-T transmits the data to the edge networking node 110-C (Content Object, S64). Finally, the edge networking node 110- (Content Object, S64).

Although not shown in FIG. 9, messages and data transmission / reception between the edge computing platform 100-T on the IoT terminal side and the edge computing platform 100-C on the user terminal side are performed by the respective edge computing platforms 100-T and 100- C through the edge gateway nodes 130-T and 130-C included in the network. In contrast, between the node 110-C, T, 120-C, T, and 130-C, T in each edge computing platform 100-T and 100-C and the edge identifier management node 200, Lt; / RTI >

FIG. 10 is a diagram for explaining a process of acquiring data of the IoT terminal 10 by the external cloud server 20. In FIG. 10, an edge computing FIG. 4 is a diagram for explaining an interoperability support procedure for acquiring data of the IoT terminal 10 on the platform 100; FIG.

First, the cloud server 20 on the Internet (b) transmits a request message for data acquisition based on an HTTP standard message to the edge gateway node 130 through a RESTful API such as the web (S71). This message can acquire the IP address on the edge computing platform 100 in which the IoT terminal exists through the DNS service as in the general Internet scenario.

Since the IP address to be acquired at this time is the IP address of the edge gateway node 130 responsible for interworking with the IP Internet on the edge computing platform 100, the corresponding request message is transmitted to the edge gateway node 130 do.

The edge gateway node 130 receiving the request message analyzes the message, converts it into an ICN-based message (Interest), and transmits the message to the intelligent computing node 120 (S72).

Next, the edge gateway node 130 receives at least one of the data of the IoT terminal 10 and the analyzed data from the intelligent computing node 120 (S73, Content Object) And transmits the data to the cloud server 20 (S74, Response).

Meanwhile, an embodiment of the present invention may include a header format conversion module to interoperate with the ICN and the Internet for the above process, and can act as a proxy to the HTTP to ICN.

[Various mobility support]

The computing system 1 for the intelligent IoE information framework according to an embodiment of the present invention supports ICN-type mobility unlike the commonly known IP-based mobility support technology on the Internet.

That is, one embodiment of the present invention is an end-to-end state preservation technique for directly managing and controlling location information of a terminal such as the IOT terminal 10, the user terminal 30, It proposes a structure for acquiring data when mobile terminal is moved around data itself rather than existing Mobility IP and Proxy Mobile IP.

Accordingly, an embodiment of the present invention can support stateless mobility, which manages data each time a request is made, without having to manage an end-to-end state. The mobility of the user terminal 30 in the ICN position, the mobility of the IoT terminal 10, and the mobility of data in which the data storage is changed, respectively, will be described in detail.

- Procedure for supporting mobility of user terminal

11 is a diagram for explaining a process of supporting mobility of the user terminal 30 and seamless handover.

In order to continuously acquire data when the user terminal 30 moves, an embodiment of the present invention can simply support mobility in the PULL type form through an existing ICN Request / Response (Interest / Content Object) message.

That is, when the user terminal 30 moves, the user terminal 30 can perform mobility support by first performing a connection setup process, and then immediately sending a request message (Interest) to the user terminal 30 again.

However, when the user terminal 30 moves out of the corresponding edge computing platform 100-COld to the edge computing platform 100-Cnew corresponding to the new domain, the user terminal 30 newly updates its identifier as shown in FIG. 7 and FIG. It is necessary to perform a procedure of registering with the edge networking node 110-Cnew and the edge identifier management node 200 on the computing platform 100-Cnew. An embodiment of the present invention can receive data through the intelligent computing node 120-Cnew in the edge computing platform 100-Cnew where the user terminal 30 is located, together with the validity check of the user terminal 30 .

Meanwhile, one embodiment of the present invention can perform the operation of the intelligent computing node as shown in FIG. 11 to support seamless handover.

As shown in FIG. 9, in the case of a general user terminal 30, the intelligent computing node 120-C of the edge computing platform 100-C of the user terminal side is directly connected to the edge networking node 110-C ). ≪ / RTI >

However, for a seamless handover, a resolution message (Interest-get message) to the edge identifier management node 200 is set to bypass the intelligent computing node 120-C and then the edge networking node 110- (Interest) for intelligent computing node 120-C is also forwarded via intelligent computing node 120-C. Accordingly, the data in the intelligent computing node 100-T on the IoT terminal side that generates the data is continuously accumulated in the intelligent computing node 100-C on the user terminal side.

Thus, in the edge computing platform 100-C, the user terminal 30 connected to the first edge networking node 110-Cold moves within the edge computing platform 100-C to the new second edge networking node 110- The second edge networking node 110-Cnew transmits a resolution message I-get (th.) To the edge-identifier management node 200 after the existing registration procedure in step S81 And receives the identifier information of the intelligent computing node 120-C (S82).

Then, the second edge networking node 110-Cnew transmits a request message (Interesting) for acquiring the data of the IoT terminal 10 or the analyzed data based on the identifier information of the intelligent computing node 120- The data of the IoT terminal 10 which has been continuously accumulated can be directly returned to the computing node 120-C through the intelligent computing node 120-C (S84, Forwarding) user terminal 30 (S85).

As described above, according to the embodiment of the present invention, even when the user terminal 30 moves and is connected to the new edge networking node 110-Cnew and data can be lost during the movement process, Seamless handover can be supported.

- IoT  Mobility support procedure of terminal

12 is a diagram for explaining the mobility support procedure of the IoT terminal 10 on the edge computing platform 100. As shown in FIG.

In general, mobility support of IoT terminals is not provided naturally in case of ICN. Meanwhile, the computing system 1 for the intelligent IoE information framework according to an embodiment of the present invention can support the mobility of the IoT terminal 10 through the intelligent computing node 120. [

Referring to FIG. 12, as the IoT terminal 10 connected to the first edge networking node 110-1 moves in the edge computing platform 100, the connection to the second edge networking node 110-2 is changed (S91), the second edge networking node 110-2 performs the registration procedure for the new IOT terminal 10. [

That is, the second edge networking node 110-2 registers the IoT terminal 10 with the intelligent computing node 120, and then the intelligent computing node 120 transmits the IoT terminal 10 (S92 to S5).

Upon completion of the registration procedure, the intelligent computing node 120 transmits a message (Interest) requesting data of the IoT terminal 10 to the second edge networking node 110-2 connected to the current IoT terminal 10 And obtain the data of the requested IoT terminal 10 (S96 to S99).

When the user terminal 10 requests data for the IoT terminal 10, the edge identifier management node 200 transmits the identifier information of the IoT terminal 10 to the intelligent computing node 120 The mapping information can be generated. Accordingly, the user terminal 30 can perform name-aware routing to the intelligent computing node 120 to naturally acquire data generated by the IOT terminal 10 through the edge identifier management node 200 .

- Data mobility support procedures

FIG. 13 is a diagram for explaining a data mobility support procedure and a seamless handover process between a plurality of edge computing platforms 100-1 and 100-2.

When moving between domains by the edge computing platforms 100-1 and 100-2, the intelligent computing node 120 storing data simultaneously with the mobility of the IoT terminal 10 generating the data changes the data Mobility occurs.

In this way, even when the IOT terminal 10 moves from the first edge computing platform 100-1 to the second edge computing platform 100-2, an embodiment of the present invention can support this mobility.

First, the edge networking nodes 110-1 and 110-2 are changed as the IOT terminal 10 moves from the first edge computing platform 100-1 to the second edge computing platform 100-2, The intelligent computing node 120-1 is also changed to the node 120-2 in the second edge computing platform 100-2 (S101).

When the IOT terminal 10 moves to the second edge computing platform 100-2, the edge networking node 110-2 of the second edge computing platform 100-2 transmits the IoT The registration process informs that the terminal 10 has moved to a new domain (S102).

Then, the intelligent computing node 120-2 registers the IoT terminal 10 with the edge identifier management node 200 (S103, S104). The intelligent computing node 120-2 provides the identifier information of the IOT terminal 10 and its own identifier information to the edge identifier management node 200 and the edge identifier management node 200 provides the identifier information of the IoT terminal 10 to the IoT terminal 10, And generates and manages the mapping information.

When the registration procedure is completed (S105), the intelligent computing node 120-2 of the second edge computing platform 100-2 receives the data of the edge networking node 110-2 to collect data of the IoT terminal 10, (S106). ≪ / RTI >

When the request message is transmitted to the IoT terminal 10 through the edge networking node 110-2 in step S107, the IoT terminal 10 transmits the request message to the edge networking node 110-2 in the second edge computing platform 100-2 (S108), the data requested by the intelligent computing node 120-2 is provided (S109), and the data of the IOT terminal 10 moved to the new domain is stored in the new intelligent computing node 120-2 .

When the user terminal 30 desires to acquire the data of the IoT terminal 10 moved to the new domain, the identifier of the newly changed intelligent computing node 120-2 is provided through the edge identifier management node 200 And the corresponding data can be acquired.

Meanwhile, in the case of the inter-domain mobility as shown in FIG. 13, there is a disadvantage that the data of the IoT terminal 10 stored in the intelligent computing node 120-1 in the existing first edge computing platform 100-1 can not be utilized .

To support seamless handover in such situations, an embodiment of the present invention may be implemented at the same time as or separately from the registration process described above, in the edge networking node 110-2 of the second edge computing platform 100-2 ) Requests forwarding of the first edge computing platform 100-1 to the edge networking node 110-1 (S110).

By this request, one or more of the data and the analyzed data of the IoT terminal 10 stored in the intelligent computing node 120-1 of the first edge computing platform 100-1 is transmitted to the second edge computing platform 100-2 (S111, S112), so that it is possible to provide effective seamless handover even when data is moved.

1 may be implemented in hardware such as software or an FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit), and may perform predetermined roles can do.

However, 'components' are not meant to be limited to software or hardware, and each component may be configured to reside on an addressable storage medium and configured to play one or more processors.

Thus, by way of example, an element may comprise components such as software components, object-oriented software components, class components and task components, processes, functions, attributes, procedures, Routines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables.

The components and functions provided within those components may be combined into a smaller number of components or further separated into additional components.

Hereinafter, a computing method for the intelligent IoE information framework according to an embodiment of the present invention will be described with reference to FIG.

14 is a flowchart of a computing method for an intelligent IoE information framework.

First, data of a terminal is received through an edge networking node supporting connection with the terminal (S1).

Next, the intelligent computing node collects and analyzes the received data of the terminal (S2), transfers the analyzed data to the cloud server through the edge gateway node, or provides it as operational data of the terminal through the intelligent computing node (S3 ).

In the above description, steps S1 to S3 may be further divided into further steps or combined into fewer steps, according to an embodiment of the present invention. Also, some of the steps may be omitted as necessary, and the order between the steps may be changed. In addition, the contents already described with respect to the computing system 1 in Figs. 1 to 13 apply to the computing method in Fig. 14, even if other contents are omitted.

Hereinafter, an embodiment in which the computing system 1 and the method for the intelligent IoE information framework are applied will be described with reference to FIG. 15 to FIG.

[ Example ]

1. Information replication services between intelligent information platforms

15 is a diagram showing a data processing scenario in two edge computing domains 100-1 and 100-2 including the same data.

Among the IoE information, there are cases where it is necessary to respond and provide the information quickly, such as media generated in real time such as CCTV or media generated privately.

In this case, as shown in FIG. 15, the storage media between the edge computing domains 100-1 and 100-2 are held together in advance, and the edge networking nodes 110-1 and 110-2 ) May exist in some cases. On the other hand, the description of the information copying procedure and method for maintaining the same information together will be omitted in the specification of the present invention.

In this situation, the intelligent computing nodes 120-1 and 120-2 can obtain the same data through the registration procedure for the identifier mapping with the edge identifier management node 200, Can be registered in the edge computing platforms 100-1 and 100-2 corresponding to the two domains.

In this situation, in order for the user terminal 30 to acquire the closest information, the user terminal 30 first transmits a request message (Interest) to the network as a method for acquiring general data (S201).

Then, the edge networking node 110-1 on the edge computing platform 100-1 obtains two edge computing platform information (intelligent computing node information-URL) having the target data through the edge identifier management node 200 (S202).

In one embodiment of the present invention, since the identifier portion includes networking part information, the edge networking node 110-1 that has acquired the request message can select the intelligent computing node 120-1 closest to itself So that data can be acquired quickly through the nearest intelligent computing node 120-1.

2. Smart Healthcare  Privacy support services in the system

FIG. 16 is a view for explaining contents supporting a privacy guarantee service in a smart healthcare system. FIG.

In FIG. 16, the user wears a smart health device 10 capable of measuring his / her biometric information. However, since data generated in such a smartheath device 10 includes personal information, a privacy invasion problem may occur when uploading to a public cloud server or a service user's cloud server.

Therefore, the edge computing platform 100 in the embodiment of the present invention can be mounted on a limited computer such as a Raspberry bi or the like and configured by a CPU and a memory.

Accordingly, a user can use a smart phone to store data in a personal store installed and managed by the user, at home or at work or on the move.

In addition, by uploading data to the cloud server 20 that performs a function of eliminating non-identification and privacy problems, daily smart health care services can be made possible.

For example, in a state in which the registration with the intelligent computing node 120-1 in the company is completed in the edge identifier management node 200 (S211), the intelligent computing node 120-1 transmits the smart phone of the user to the edge networking node 110 - 1 to receive data of the smart health appliance 10 (S 212 to S 214). If the user moves from home to home (S215), the registration process between the intelligent computing node 120-2 and the edge identifier management node 200 in the home is performed (S216) (S217 to S219) through the mobile communication terminal 110-2.

In this case, the smart health service provider can confirm that the smart health appliance 10 has been moved from the company to the home by performing the registration procedure with the edge identifier management node 200 (S220), and the intelligent computing node 120-2 And obtains the data of the smart health appliance 10 (S221).

As described above, the computing system (1) and the method for the intelligent IoE information framework according to an embodiment of the present invention can create a platform usable in the domain, regardless of where the user is, and perform edge identifier management By using the service through the node 200, privacy can be easily supported and existing services can be supported.

3. Smart plant (piping information processing)

FIG. 17 is a diagram for explaining an urgent message processing support process P7 in the smart factory system. FIG.

As shown in FIG. 17, in order to provide safety service in the automation factory or the nuclear power plant piping, many sensors are used to monitor the safety (P5).

However, in the case of such monitoring, since the data continuously generated is stored in the cloud server 20 and analyzed, when there is a problem, there occurs a transmission delay and analysis delay through the cloud server 20 and a delay to notify the user do. Therefore, this model is not suitable for time-critical systems that require real-time processing.

Alternatively, in one embodiment of the present invention, the real-time data received from the piping of each plant and the plant can be analyzed and monitored in real time through the intelligent computing node 120 (P6).

Also, when a pattern different from the existing pattern is generated, a notification is sent directly to all the user terminals 30-1 and 30-2 and the specific user terminal 30-2 without passing through the cloud server 20 It has a temporal advantage in that it can be provided.

In addition, through the Reward system that supports actual operation commands, it is possible to immediately diagnose piping safety and support on-off control of the process immediately (P7).

Particularly, in an embodiment of the present invention, since it is possible to acquire information through a simple message, it can be more useful in a time-critical situation.

4. Bypass routing for data processing through multiple mobile terminals

18 is a diagram for explaining a data path bypass processing support process through the mobile terminal 110-2.

In order to provide not only the processing of a large amount of data provided by the 10 ¹⁰ IOT terminals 10 but also connectivity in a situation where the volume of data generated by the IOT terminal 10 is gradually increasing, 110) may still suffer from problems such as scalability.

Particularly, in case of data generated in real time, when data is lost depending on the situation, it causes a big problem in analyzing it.

18, an edge networking node 110-1 of the edge computing platform 100 may be developed to be extended to a smartphone 110-2 that many people have, have.

Accordingly, an embodiment of the present invention can diversify the path to the storage of the edge computing platform 100. In addition, since the smartphone 110-2 provided by a plurality of users participates and can collect data generated in the IoT terminal 10, problems such as scalable connection and data loss can be solved.

In addition, an embodiment of the present invention may also resolve scalability issues for processing of the intelligent computing node 120 by distributing the data and diverting the path to each store.

In addition, an embodiment of the present invention includes an intelligent information processing management unit 140 and an edge networking node (hereinafter, referred to as " intelligent information processing ") 140 capable of generating a new intelligent computing node dynamically according to the load of the intelligent computing node 120, 110-2 and the edge identifier management node 200 to assist in bypassing the transmission path.

5. Support to create machine learning model by data classification and combination

FIG. 19 is a diagram for explaining a process of supporting creation of a machine learning model by data classification and combination. FIG.

One embodiment of the present invention classifies and combines the data P8 to P10 provided from the various IoT terminals 10 to various IoT services P11 to P13, for example, traffic signal processing or use of a nuclear power plant It is possible to provide data that can generate the necessary machine learning models.

First, patterns and flows of data P8 to P10 generated from the IoT terminal 10 are learned through the computing system 1 for the intelligent IoE information framework according to an embodiment of the present invention.

Through the learning process, classified data can be combined and transmitted to servers P11 to P13 that are located in an actual central cloud server or the like and generate a machine learning model for providing IoT specific services.

For example, it can be used as a base data for creating a machine learning model for recognizing a dangerous situation at a factory by providing not only simple sensor data but also other data sensed by the surrounding smartphone at the time according to need, in time synchronization .

This procedure supports the ability to directly generate data used to create a service model through supervised learning by providing data that is already labeled with data rather than real-time data.

6. Virtualization  Intelligent based IoE  Information Framework  Configuration

20 is a diagram for explaining a virtualization-based platform 100 and a framework.

The intelligent IoE information framework in one embodiment of the present invention can be mounted on a smart phone as an IoT gateway or actuator directly through a container, a handler, or other type of virtualization technology.

In particular, as in the case of FIG. 20, it is possible to extend to a deep learning platform located in the central cloud server 20 or a framework capable of dynamically requesting and executing a microservice for each data processing.

That is, when various micro services required according to the IoE service are located on the central cloud server 20, an embodiment of the present invention requests an associated micro service on an identifier basis, acquires it, ) So as to process data generated in the IOT terminal 10 directly.

To this end, the micro-services supported on the framework implemented in the cloud server 20 are registered in the edge identifier management node 200 in detail in the form of identifiers. Accordingly, it is possible to acquire and combine the necessary micro-services on the edge computing platform 100 through the information networking technique using the existing edge identifier management node 200 to perform necessary operations.

For example, in a smart healthcare domain, data originating from a user's sensor is privacy-sensitive data. Therefore, the required data is required to be non-discriminated and preprocessed.

If such a service profile is defined, it is possible to dynamically request the models required on the central cloud server 20 in a microservice form, such as a preprocessing service, a non-identifying service, etc. according to the service profile, To process the health data. Accordingly, the processed health data can be directly reported to the user terminal 30 or directly forwarded to the cloud server 20, thereby enabling further processing.

On the other hand, an embodiment of the present invention may also be embodied in the form of a computer program stored in a medium executed by a computer or a recording medium including instructions executable by the computer. Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes any information delivery media, including computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transport mechanism.

While the methods and systems of the present invention have been described in connection with specific embodiments, some or all of those elements or operations may be implemented using a computer system having a general purpose hardware architecture.

It will be understood by those skilled in the art that the foregoing description of the present invention is for illustrative purposes only and that those of ordinary skill in the art can readily understand that various changes and modifications may be made without departing from the spirit or essential characteristics of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

1: Computing System for Intelligent IoE Information Framework
10: IoT terminal
20: Cloud server
30: User terminal
100: Edge Computing Platform
110: Edge networking node
120: Intelligent Computing Node
130: edge gateway node
140: intelligent information processing management unit
150: Data storage unit
160: Reward /

Claims (19)

A computing system for an intelligent IoE information framework,
An intelligent computing node for analyzing data of the terminal through the edge networking node, an edge gateway for edge-based networking within the edge computing platform, and an edge gateway for internet protocol networking outside the edge computing platform, A plurality of edge computing platforms each including a plurality of nodes,
And an edge identifier management node for generating mapping information mapping the identifiers of the terminal and data to the plurality of edge computing platforms. The method according to claim 1,
Wherein the data of the terminal is stored in a temporary storage and categorized according to types of data and services, and the classified data is stored in a data storage in the edge computing platform and analyzed through the intelligent computing node. The method according to claim 1,
Wherein the edge networking node is connected to the terminal through bootstrapping for transmitting the response message after registering the identifier information upon receiving the identifier information from the terminal. The method of claim 3,
Wherein the edge networking node sends and registers identifier information of the terminal to the intelligent computing node and the intelligent computing node transmits and registers its own identifier information and the identifier information of the terminal to the edge identifier management node, Is maintained. The method according to claim 1,
The edge networking node of the edge computing platform of the user terminal of the plurality of edge computing platforms,
Receiving identifier of the intelligent computing node of the IoT terminal side edge computing platform in which the data of the IoT terminal is stored from the edge identifier management node through the intelligent computing node by receiving a message requesting data of the IoT terminal from the user terminal Receiving,
And receives and provides to the user terminal at least one of the data of the IoT terminal and the analyzed data from the intelligent computing node of the IoT terminal side edge computing platform based on the identifier information. The method according to claim 1,
Wherein the edge gateway node receives the message requesting the data of the IoT terminal from the cloud server and converts the message into the message based on the information centering network and transmits the message to the intelligent computing node, And converting at least one of the analyzed data into a message based on Internet Protocol networking upon receiving the message, and providing the converted message to the cloud server. The method according to claim 1,
Wherein the edge gateway node receives a data processing model corresponding to the service profile from the cloud server as it transmits the service profile to the cloud server through the Internet protocol networking and the intelligent computing node is based on the received data processing model And analyzing the data. The method according to claim 1,
Wherein data analyzed through the intelligent computing node is provided for operation of the terminal or is delivered to the cloud server via the edge gateway node. The method according to claim 1,
When a user terminal connected to the first edge networking node in the edge computing platform changes to a connection to the second edge networking node as the user terminal moves,
The second edge networking node receives the identifier information of the intelligent computing node from the edge identifier management node and transmits a message requesting the data of the IoT terminal to the intelligent computing node based on the received identifier information, To forward one or more of the data of the IoT terminal and the analyzed data. The method according to claim 1,
If the IoT terminal connected to the first edge networking node in the edge computing platform changes to the connection to the second edge networking node as the IoT terminal moves,
Wherein the second edge networking node registers the IoT terminal with an intelligent computing node and the intelligent computing node registers the IoT terminal with the edge identifier management node. 11. The method of claim 10,
Wherein the intelligent computing node transmits a message requesting data of the IoT terminal to the second edge networking node to obtain data of the IoT terminal. 11. The method of claim 10,
Wherein the edge identifier management node maps the identifier information of the IoT terminal and the identifier information of the intelligent computing node to the mapping information to process the data request for the IoT terminal of the user terminal. The method according to claim 1,
When the IoT terminal of the plurality of edge computing platforms moves from the first edge computing platform to the second edge computing platform,
Wherein the edge networking node of the second edge computing platform registers the IoT terminal with an intelligent computing node and the intelligent computing node of the second edge computing platform registers the IoT terminal with the edge identifier management node. 14. The method of claim 13,
Wherein the intelligent computing node of the second edge computing platform provides the identifier of the IoT terminal and its identifier to the edge identifier management node and the edge identifier management node maps the identifier of the IoT terminal to the identifier of the intelligent computing node And generates the mapping information as the mapping information. 14. The method of claim 13,
The intelligent computing node of the second edge computing platform transmits a message for requesting the data of the IoT terminal through the edge networking node of the second edge computing platform so that the IoT terminal moving to the second edge computing platform And provide the data to an intelligent computing node of the second edge computing platform via an edge networking node of the second edge computing platform. 14. The method of claim 13,
As the IoT terminal completes the move to the second edge computing platform, the intelligent computing node at the user terminal receives the identifier of the intelligent computing node of the second edge computing platform as an edge identifier management node, The intelligent computing node acquiring the data of the IoT terminal. 14. The method of claim 13,
Wherein the edge networking node of the second edge computing platform is configured to request, via a request to an edge networking node of the first edge computing platform, data of the IoT terminal stored in the intelligent computing node of the first edge computing platform To forward one or more to an intelligent computing node of the second edge computing platform. A computing method for an intelligent IoE information framework,
Receiving data of the terminal through an edge networking node supporting connectivity with the terminal;
Collecting and analyzing the data of the received terminal by the intelligent computing node;
Transmitting the analyzed data to a cloud server through an edge gateway node or providing the analyzed data as operation data of the terminal through the intelligent computing node,
Wherein the intelligent IoE information framework comprises a plurality of edge computing platforms each including the edge networking node, the intelligent computing node and the edge gateway node, and mapping information mapping the identifier of the terminal and data to the edge computing platform And an edge identifier management node for generating an edge identifier management node. In an edge computing platform device in an intelligent IoE information framework,
An edge networking node supporting connectivity with the terminal,
An intelligent computing node that analyzes data of the terminal through the edge networking node;
An edge computing platform device comprising an edge gateway node for information-centric networking and Internet protocol networking with the outside.

Images