Apparatus and method of processing sensor data in sensor network based on open radio access network

Abstract

The present invention relates to a sensor data processing device and method in an O-RAN-based sensor network. A sensor data processing device according to one embodiment of the present invention may include a service management organization unit that collects sensor data of a plurality of sensors as user plane (UP) data through an open interface from a sensor network composed of a plurality of sensors, and distributes an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors based on a non-real-time intelligent control unit, and a real-time intelligent control unit that checks whether an emergency event has occurred based on the output of the anomaly management model through a monitoring interface and generates a control command when the emergency event occurs.

Description

Technology Field

[0001] The present invention relates to a sensor data processing device and method in an O-RAN-based sensor network, and more specifically, to a technology for collecting and processing sensor data from sensor devices, such as IoT devices that constitute a sensor network, in preparation for using a specialized network instead of a short-range wireless network such as Wi-Fi, by applying a standardized O-RAN-based solution for the specialized network to facilitate the expansion and establishment of smart factories. Background Technology

[0002] With the advent of the 5G era, the scope of communication has expanded from people to objects, and mMTC (massive machine type communications), which was previously defined only by theory and standards, is now on the verge of commercialization, led by Industrial IoT (IIoT) technology.

[0003] In particular, as IIoT technology is integrated into smart factories where numerous sensor networks and robots are used in industrial settings, a massive amount of data is being generated.

[0004] Ultra-dense mmtc networks (UDMNs) are formed in which more than one million objects are densely distributed within one square kilometer, and to process this industrial data requiring a high level of security from IoT devices with low mobility, technologies that are differentiated from currently used human-oriented networks may be required.

[0005] Unlike human-oriented networks intended to provide services, UDMNs that use 5G communication for the purpose of collecting data primarily collect short packet data, such as from sensor networks, on a large scale to form big data.

[0006] Beyond data collection, a network is needed to enable efficient industry management based on this data.

[0007] Therefore, a 5G specialized network (private 5G) has emerged to manage and process such a massive amount of data generated in industrial sites with a high level of security.

[0008] In smart factories, there are numerous robots that move along fixed paths and statically deployed sensor networks.

[0009] High performance can be achieved by deploying a 5G specialized network targeting areas with high communication demand caused by these factors, and security can be enhanced by minimizing sharing with the public network.

[0010] Meanwhile, O-RAN, which can build network equipment cost-effectively and intelligently process what happens in the network, was proposed by the Open Radio Access Network (O-RAN) Alliance, established in February 2018 by five global telecommunications operators, and open source platforms to expand the ecosystem are still being developed.

[0011] O-RAN achieves cost efficiency by enabling equipment compatibility between different vendors based on open interfaces.

[0012] A new element responsible for intelligent functions called a RAN intelligent controller is added to enable efficient operation.

[0013] It enables monitoring and collecting data generated from multiple RANs, allowing for an intelligent data processing approach (data-driven approach).

[0014] Therefore, if a 5G specialized network is built to be compatible with O-RAN standards, the network can be built cost-effectively, and at the same time, the RAN intelligent control unit function, which intelligently processes data generated in the network, can be utilized.

[0015] However, in the original O-RAN standard, RAN intelligent control units are defined for the purpose of improving network performance by collecting control plane signals.

[0016] Existing studies also collect these control plane signals and utilize them to focus on the content of the control plane signals, such as load balancing, handover, and KPI monitoring.

[0017] On the other hand, sensor data generated in an IIoT environment or information about the industrial environment corresponds to user plane data.

[0018] Therefore, it is difficult to collect this user plane data using only the RAN intelligent control unit defined in the current standard.

[0019] Therefore, in conventional technology, user plane data has been collected using an application server.

[0020] In conventional technology, traffic information (congestion / availability, user applications, etc.) or GPS information is exemplified for user plane data.

[0021] However, unlike sensor data, there is a significant difference in that this data is not collected from the network but is brought in from an external source.

[0022] Therefore, efforts may be required to reduce overhead caused by the application server in accordance with the characteristics of sensor data collected directly from the network. Prior art literature

[0023] Korean Published Patent No. 10-2021-0117960, "Method and apparatus for O-RAN-based performance optimization and configuration" Korean Published Patent No. 10-2023-0132434, "System and method for enabling an auto-configured network in an Open RAN" Korean Published Patent No. 10-2024-00006851, "Open RAN unit for Eeum 5G in unlicensed millimeter wave band" Korean Registered Patent No. 10-2024-0011481, "O-RAN Cell control device and base station device, and Cell control method performed in the device" The problem to be solved

[0024] The present invention aims to provide a sensor data processing device and method in which a service management organization unit hosts a non-real-time intelligent control unit and is also connected to a near-real-time intelligent control unit and a base station through an open interface to collect and process sensor data corresponding to user plane data.

[0025] The present invention aims to collect and process sensor data from sensor devices, such as IoT devices constituting a sensor network, in preparation for the use of specialized networks instead of short-range wireless networks such as Wi-Fi, by applying a standardized O-RAN-based solution for specialized networks, thereby facilitating the expansion and establishment of smart factories.

[0026] The present invention aims to provide a sensor data processing device and method that manages network priority and QoS (Quality of Service) by utilizing some of the management services (MnS) provided by a service management organization unit, and supports sensor data processing of a sensor network by setting a slice type when a network slice is applied.

[0027] The present invention aims to increase the universality of sensor data collection and management by collecting sensor data generated in a sensor network without going through an application server for sensor data collection and processing, as the service management organization unit collects sensor data, which is user plane data, through an open interface. means of solving the problem

[0028] A sensor data processing device according to one embodiment of the present invention may include a service management organization unit that collects sensor data of a plurality of sensors as user plane (UP) data through an open interface from a sensor network composed of a plurality of sensors, and distributes an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors based on a non-real-time intelligent control unit, and a real-time intelligent control unit that checks whether an emergency event has occurred based on the output of the anomaly management model through a monitoring interface and generates a control command when the emergency event has occurred.

[0029] The above service management organization unit can collect the user plane data through the open interface by using at least one of the provisioning management service (MnS), fault supervision management service, performance assurance management service, and file management service.

[0030] The above service management organization unit manages the Quality of Service (QoS) related to the priority of the sensor network based on the above provisioning management service, and can set the priority of the sensor data by setting the slice type when a network slice is applied.

[0031] The above service management organization unit can collect reporting data in real time regarding the detection of abnormal signals, such as the occurrence of errors in the sensor network, based on the above defect monitoring management service.

[0032] The above service management organization unit can collect data among file data and streaming data from the plurality of sensors based on the above performance assurance management service, analyze the collected data, and report inference performance.

[0033] The above service management organization unit can facilitate data transmission to the plurality of sensors based on the above file management service.

[0034] The above service management organization unit can be connected to the base station connected to the sensor network and the real-time intelligent control unit through the open interface to operate complementarily.

[0035] The above plurality of sensors can be applied to a plurality of IoT (Internet of Things) equipment and a plurality of work robots operating within a smart factory.

[0036] A sensor data processing method according to one embodiment of the present invention may include the steps of: collecting sensor data of a plurality of sensors as user plane (UP) data from a sensor network composed of a plurality of sensors through an open interface in a service management organization unit, and distributing an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors based on a non-real-time intelligent control unit; and in a real-time intelligent control unit, checking whether an emergency event has occurred based on the output of the anomaly management model through a monitoring interface, and generating a control command when the emergency event has occurred.

[0037] The step of collecting sensor data of a plurality of sensors as user plane (UP) data through an open interface from a sensor network composed of a plurality of sensors, and deploying an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors based on a non-real-time intelligent control unit, may include the step of collecting the user plane data through the open interface using at least one management service among a provisioning management service (MnS), a fault supervision management service, a performance assurance management service, and a file management service. Effects of the invention

[0038] The present invention can provide a sensor data processing device and method in which a service management organization unit hosts a non-real-time intelligent control unit and is also connected to a near-real-time intelligent control unit and a base station through an open interface to collect and process sensor data corresponding to user plane data.

[0039] In preparation for the use of a specialized network instead of a short-range wireless network such as Wi-Fi, the present invention applies a standardized O-RAN-based solution for the specialized network to facilitate the expansion and establishment of smart factories, thereby enabling the collection and processing of sensor data from sensor devices, such as IoT devices that constitute a sensor network.

[0040] The present invention can provide a sensor data processing device and method that manages network priority and QoS (Quality of Service) by utilizing some of the management services (MnS) provided by a service management organization unit, and supports sensor data processing of a sensor network by setting a slice type when a network slice is applied.

[0041] The present invention can increase the universality of sensor data collection and management by collecting sensor data generated from a sensor network without going through an application server for sensor data collection and processing, as the service management organization unit collects sensor data, which is user plane data, through an open interface. Brief explanation of the drawing

[0042] FIGS. 1 to 3 are drawings illustrating a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention. FIG. 4 is a diagram illustrating a sensor data collection environment in an O-RAN-based sensor network according to an embodiment of the present invention. FIGS. 5A and 5B are drawings illustrating a management simulation of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention. FIG. 6 is a diagram illustrating a method for processing sensor data in an O-RAN-based sensor network according to an embodiment of the present invention. Specific details for implementing the invention

[0043] Hereinafter, various embodiments of this document are described with reference to the attached drawings.

[0044] The embodiments and terms used therein are not intended to limit the technology described in this document to specific embodiments and should be understood to include various modifications, equivalents, and / or substitutions of said embodiments.

[0045] In describing various embodiments below, if it is determined that a detailed description of related known functions or configurations could unnecessarily obscure the essence of the invention, such detailed description will be omitted.

[0046] Furthermore, the terms described below are defined considering their functions in various embodiments, and these may vary depending on the intentions or practices of the user or operator. Therefore, their definitions should be based on the content throughout this specification.

[0047] In relation to the description of the drawings, similar reference numerals may be used for similar components.

[0048] A singular expression may include a plural expression unless the context clearly indicates otherwise.

[0049] In this document, expressions such as "A or B" or "at least one of A and / or B" may include all possible combinations of the items listed together.

[0050] Expressions such as "first," "second," "first," or "second" may modify the corresponding components regardless of order or importance, and are used merely to distinguish one component from another without limiting the components.

[0051] Where it is stated that a certain (e.g., first) component is "(functionally or telecommunicationally) connected" or "connected" to another (e.g., second) component, said certain component may be directly connected to said other component or connected through another component (e.g., third component).

[0052] In this specification, "configured to" may be used interchangeably with, depending on the context, for example, in hardware or software, "suitable for," "capable of," "modified to," "made to," "capable of," or "designed to."

[0053] In some situations, the expression "device configured to..." may mean that the device is "able to..." together with other devices or parts.

[0054] For example, the phrase “a processor configured (or set) to perform A, B, and C” may mean a dedicated processor for performing said operations (e.g., an embedded processor), or a general-purpose processor capable of performing said operations by executing one or more software programs stored in a memory device (e.g., a CPU or an application processor).

[0055] Also, the term 'or' means an inclusive or rather an exclusive or.

[0056] That is, unless otherwise noted or is not clear from the context, the expression 'x uses a or b' means any one of the natural inclusive permutations.

[0057] Terms such as '..bu', '..gi' used below refer to a unit that processes at least one function or operation, and this may be implemented in hardware or software, or a combination of hardware and software.

[0059] FIGS. 1 to 3 are drawings illustrating a sensor data processing device in an O-RAN (Open Radio Access Network) based sensor network according to an embodiment of the present invention.

[0060] FIG. 1 illustrates the components of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention.

[0061] Referring to FIG. 1, in an O-RAN-based sensor network according to an embodiment of the present invention, a sensor data processing device (100) may include a service management organization unit (110) and a real-time intelligent control unit (120).

[0062] For example, the service management organization (110) collects sensor data from multiple sensors as user plane (UP) data through an open interface from a sensor network composed of multiple sensors.

[0063] Additionally, the service management organization unit (110) can deploy an anomaly management model for detecting anomalies in collected sensor data and predictive maintenance of multiple sensors based on a non-real-time intelligent control unit.

[0064] For example, multiple sensors can be applied to multiple IoT (Internet of Things) devices and multiple work robots operating within a smart factory.

[0065] The service management organization unit (110) can collect the user plane data through an open interface using at least one of the provisioning management service (MnS), fault supervision management service, performance assurance management service and file management service.

[0066] The service management organization unit (110) manages the Quality of Service (QoS) related to the priority of the sensor network based on the provisioning management service, and can set the priority of the sensor data by setting the slice type when a network slice is applied.

[0067] The service management organization unit (110) can collect reporting data in real time regarding abnormal signal detection, such as the occurrence of an error in the sensor network, based on the fault monitoring management service.

[0068] The service management organization unit (110) can collect data from either file data or streaming data from multiple sensors based on the performance assurance management service, analyze the collected data, and report the inference performance.

[0069] The service management organization unit (110) can facilitate data transmission to multiple sensors based on a file management service.

[0070] The service management organization unit (110) can be connected to a base station connected to a sensor network and a real-time intelligent control unit through an open interface and operate complementarily.

[0071] User plane data includes traffic information (congestion / availability, user applications, etc.) and GPS information.

[0072] However, unlike sensor data, user plane data differs in that it is not collected from the sensor network but is obtained from an external source.

[0073] On the other hand, since sensor data is collected directly from the network and the size of individual packets is also very small, the service management organization unit (110) operates to reduce network resource overhead by utilizing the O-RAN structure as much as possible rather than having a separate application server.

[0074] For example, the interface between the application server and the RAN may not be a standardized interface, but rather a proprietary black box interface.

[0075] Therefore, there is a major problem in that it cannot be universally implemented in various industrial sites.

[0076] However, sensor data can be collected through a standardized interface by utilizing the O1 interface, an open interface defined in the O-RAN standard.

[0077] The service management and orchestration department (110) can be referred to as SMO (service management and orchestration).

[0078] SMO and open interfaces can collect sensor data because, according to the standards, functions for collecting and processing user plane data are defined.

[0079] Therefore, when collecting sensor data using SMO, it is possible to collect sensor data generated from the network without using a separate application server, and since it utilizes an open interface, it has the advantage of being universally applicable.

[0080] A real-time intelligent control unit (120) according to one embodiment of the present invention can check whether an emergency event has occurred based on the output of an abnormal management model through a monitoring interface, and can generate a control command when an emergency event has occurred.

[0081] The real-time intelligent control unit (120) may be referred to as a near-time RAN intelligent controller (RAN intelligent controller, RIC).

[0082] Since the sensor network is where the most significant data is generated in UDMNs at industrial sites, the SMO is defined in the standard to allow for the collection of user plane data, so the items among them that can be utilized for sensor data collection can be surveyed and data collection can be implemented.

[0083] Accordingly, the present invention can provide a sensor data processing device and method in which a service management organization unit hosts a non-real-time intelligent control unit and is also connected to a near-real-time intelligent control unit and a base station through an open interface to collect and process sensor data corresponding to user plane data.

[0084] FIG. 2 more specifically illustrates the components of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention.

[0085] Referring to FIG. 2, in an O-RAN-based sensor network according to an embodiment of the present invention, the sensor data processing device (200) may consist of a service management organization unit (210), a real-time intelligent control unit (220), and a base station (230).

[0086] The service management organization unit (210) includes a non-real-time intelligent control unit (211), a first management service processing unit (212), a second management service processing unit (213), a third management service processing unit (214), and a fourth management service processing unit (215).

[0087] The first management service processing unit (212) can facilitate data transmission to multiple sensors based on a file management service.

[0088] For example, the first management service processing unit (212) may be the entity performing the file management service among at least one management service.

[0089] The second management service processing unit (213) manages the Quality of Service (QoS) related to the priority of the sensor network based on the provisioning management service, and can set the priority of the sensor data by setting the slice type when a network slice is applied.

[0090] For example, the second management service processing unit (213) may be the entity performing the provisioning management service among at least one management service.

[0091] The third management service processing unit (214) can collect reporting data in real time regarding the detection of abnormal signals, such as the occurrence of an error in the sensor network, based on the fault monitoring management service.

[0092] For example, the third management service processing unit (214) may be the entity performing the defect monitoring management service among at least one management service.

[0093] The fourth management service processing unit (215) can collect data from either file data or streaming data from multiple sensors based on the performance assurance management service, analyze the collected data, and report the inference performance.

[0094] For example, the fourth management service processing unit (215) may be the entity performing the performance assurance management service among at least one management service.

[0095] According to one embodiment of the present invention, the service management organization unit (210) can be connected to a base station (230) connected to a sensor network and a real-time intelligent control unit (220) through an open interface (O1 interface) and operate complementarily.

[0096] The real-time intelligent control unit (220) can monitor through the A1 interface and, if it is determined that an urgent situation has occurred based on the prediction results, issue necessary commands through the E2 interface.

[0097] For example, the A1 interface can be referred to as a monitoring interface.

[0098] In an O-RAN-based sensor network according to one embodiment of the present invention, the sensor data processing device (200) has a service management organization unit (210), a non-real-time intelligent control unit (211), and a real-time intelligent control unit (220) all operating complementarily, and can also process urgent events at industrial sites that require a delay time of less than 1 second.

[0099] The base station (230) may consist of an open RAN distributed unit (O-DU) and an antenna (open RAN central unit, O-CU).

[0100] The base station (230) can collect sensor network and sensor data in conjunction with the open communication unit (240) and transmit control commands based on the collected sensor data.

[0101] The base station (230) and the open communication unit (240) can be connected through an open front home interface.

[0102] The management service processing units of the service management organization unit (210) can be linked with the non-real-time intelligent control unit (211) to implement various functions.

[0103] The first management service processing unit (212) can collect large volume data (Bulk data) collected from the sensor network.

[0104] The second management service processing unit (213) can perform functions such as assigning priorities to multiple sensor networks and notifying the non-real-time intelligent control unit (211) and the real-time intelligent control unit (220).

[0105] The third management service processing unit (214) can perform error checking defined in the standard, such as in a situation where data collection is suddenly stopped.

[0106] The fourth management service processing unit (215) can train an AI / ML model using collected sensor data and then install it in the non-real-time intelligent control unit (211).

[0107] The service management organization unit (210) can process requests by assigning priority when the non-real-time intelligent control unit (211) and the real-time intelligent control unit (220) receive requests simultaneously from multiple sensor networks.

[0108] In an O-RAN-based sensor network according to one embodiment of the present invention, the sensor data processing device (200) can be utilized with high compatibility in various communication environments using standard O-RAN by selectively using management functions among the management functions defined in SMO.

[0109] Accordingly, the present invention can provide a sensor data processing device and method that manages network priority and QoS (Quality of Service) by utilizing some of the management services (MnS) provided by the service management organization unit, and supports sensor data processing of a sensor network by setting a slice type when a network slice is applied.

[0110] FIG. 3 illustrates a case in which components of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention are applied to a sensor network environment such as a smart factory.

[0111] Referring to FIG. 3, a configuration is illustrated in which sensor data is collected and monitored through an IoT device (380) to which a sensor is attached or sensor data is collected through a sensing board (370) that is linked to an actuator (360) via a wireless communication unit (350) through a core network (330) that connects a remote monitoring unit (350) and a service management organization unit (310) that constitutes a sensor data processing unit in an O-RAN-based sensor network according to an embodiment of the present invention.

[0112] For example, regarding a UDMN built at high density in a smart factory based on a sensor data processing device according to one embodiment of the present invention, data of a three-phase motor can be collected using an integrated 5G frequency by utilizing a sensing board capable of modeling the UDMN.

[0113] Currently, a function to monitor this using a web server has been implemented, and work can be carried out to develop O-RAN elements to ensure compatibility with integrated 5G equipment.

[0114] Accordingly, the present invention can collect and process sensor data from sensor devices, such as IoT devices constituting a sensor network, in preparation for the use of a specialized network instead of a short-range wireless network such as Wi-Fi, by applying a standardized O-RAN-based solution for the specialized network, thereby facilitating the expansion and establishment of smart factories.

[0116] FIG. 4 is a diagram illustrating a sensor data collection environment in an O-RAN-based sensor network according to an embodiment of the present invention.

[0117] FIG. 4 illustrates a sensor data collection environment in an O-RAN-based sensor network according to an embodiment of the present invention.

[0118] Referring to FIG. 4, a sensor data collection environment (400) according to one embodiment of the present invention may be composed of a base station (401), a sensor network (402), and a robot (403).

[0119] In an O-RAN-based sensor network according to an embodiment of the present invention, a sensor data processing device can collect and process sensor data from a plurality of sensors and a robot (403) constituting the sensor network (402) through a base station (401).

[0120] In an O-RAN-based sensor network according to one embodiment of the present invention, the sensor data processing device may be located at a base station (401).

[0122] FIGS. 5A and 5B are drawings illustrating a management simulation of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention.

[0123] FIG. 5a illustrates a model that utilizes an LSTM to predict current data in relation to the management simulation of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention, and detects anomalies by comparing the current data with actual data to prepare for failure.

[0124] Referring to FIG. 5a, graphs (500) and (501) illustrate results in which a predictive maintenance algorithm based on AI / ML and utilizing the statistical characteristics of data is installed in the developed RIC.

[0125] According to graph (500) and graph (501), the predictive maintenance algorithm performed by the sensor data processing device according to one embodiment of the present invention can use a model that first predicts current data by utilizing an LSTM that shows the best performance in predicting time series data, and detects anomalies by comparing this with actual data to prepare for failure.

[0126] FIG. 5b illustrates the distribution estimation of vibration data in relation to the management simulation of a sensor data processing device in an O-RAN-based sensor network according to an embodiment of the present invention.

[0127] Referring to FIG. 5b, graphs (510) and (511) show that a model can be developed to predict failures in advance by estimating the distribution of vibration data and utilizing statistical characteristics.

[0128] Through predictive maintenance, stable operation is possible, and large-scale data generated by UDMN can be intelligently processed, allowing for efficient management of the IoT network.

[0130] FIG. 6 is a diagram illustrating a method for processing sensor data in an O-RAN-based sensor network according to an embodiment of the present invention.

[0131] FIG. 6 illustrates a procedure in which a sensor data processing method in an O-RAN-based sensor network according to an embodiment of the present invention is performed by connecting a service management organization unit through an open interface to collect and process sensor data corresponding to user plane data and to perform processing of abnormal data in real time.

[0132] Referring to FIG. 6, in step (S601), a sensor data processing method according to an embodiment of the present invention collects sensor data as user plane data through an open interface and distributes an anomaly management model.

[0133] That is, a sensor data processing method according to one embodiment of the present invention collects sensor data from a plurality of sensors as user plane (UP) data through an open interface from a sensor network composed of a plurality of sensors, and can deploy an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors based on a non-real-time intelligent control unit.

[0134] In step (S602), the sensor data processing method according to an embodiment of the present invention generates and transmits a control command based on whether an emergency event occurs, based on the output of an abnormality management model.

[0135] That is, the sensor data processing method according to one embodiment of the present invention can check whether an emergency event has occurred based on the output of an anomaly management model through a monitoring interface, and can generate and transmit a control command if an emergency event has occurred.

[0136] Accordingly, the present invention can increase the universality of sensor data collection and management by collecting sensor data generated from a sensor network without going through an application server for sensor data collection and processing, as the service management organization unit collects sensor data, which is user plane data, through an open interface.

[0137] In addition, by not going through an application server, the overhead that could be increased for the application server can be reduced.

[0139] The device described above may be implemented as a hardware component, a software component, and / or a combination of a hardware component and a software component. For example, the device and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. In addition, other processing configurations, such as parallel processors, are also possible.

[0140] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0141] Although the embodiments have been described above with reference to the limited drawings, those skilled in the art can make various modifications and variations from the description above. For example, suitable results can be achieved even if the described techniques are performed in a different order than described, and / or the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0142] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below. Explanation of the symbols

[0143] 100: Sensor data processing unit 110: Service Management Planning Unit 120: Real-time Intelligent Control Unit

Claims

Claim 1 A sensor data processing device comprising: a service management organization unit that collects sensor data from a plurality of sensors as user plane (UP) data through an O1 interface from a sensor network composed of a plurality of sensors, trains an AI / ML model using the collected sensor data and loads it onto a non-real-time intelligent control unit to generate an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors, and distributes the anomaly management model to a real-time intelligent control unit; and a real-time intelligent control unit that receives the anomaly management model through an A1 interface, performs monitoring, checks whether an emergency event has occurred based on the output of the anomaly management model, and transmits the generated control command to a base station connected to the sensor network through an E2 interface when the emergency event has occurred, wherein the service management organization unit processes the request according to an assigned priority when the non-real-time intelligent control unit and the real-time intelligent control unit simultaneously receive requests from the plurality of sensor networks. Claim 2 A sensor data processing device according to claim 1, characterized in that the service management organization unit collects the user plane data through the O1 interface using at least one management service among a provisioning management service, a fault supervision management service, a performance assurance management service, and a file management service corresponding to the management service (MnS) of SMO defined in the O-RAN standard. Claim 3 A sensor data processing device characterized in that, in paragraph 2, the service management organization unit manages the Quality of Service (QoS) related to the priority of the sensor network based on the provisioning management service, and sets the priority of the sensor data by setting a slice type when a network slice is applied. Claim 4 A sensor data processing device characterized in that, in paragraph 2, the service management organization unit collects reporting data regarding the detection of abnormal signals, such as the occurrence of an error in the sensor network, in real time through the O1 interface based on the fault monitoring management service. Claim 5 A sensor data processing device according to paragraph 2, wherein the service management organization unit collects data among file data and streaming data from the plurality of sensors through the O1 interface based on the performance assurance management service, analyzes the collected data, and reports inference performance. Claim 6 A sensor data processing device characterized in that, in paragraph 2, the service management organization unit promotes data transmission through the O1 interface for the plurality of sensors based on the file management service. Claim 7 A sensor data processing device according to claim 1, characterized in that the service management organization unit is connected to a base station connected to the sensor network and the real-time intelligent control unit through the O1 interface and operates complementarily. Claim 8 A sensor data processing device according to claim 1, characterized in that the plurality of sensors are applied to a plurality of IoT (Internet of Things) equipment and a plurality of work robots operating within a smart factory. Claim 9 A sensor data processing method characterized by comprising: a step of collecting sensor data from a plurality of sensors as user plane (UP) data through an O1 interface from a sensor network composed of a plurality of sensors in a service management organization unit, training an AI / ML model using the collected sensor data and then installing it in a non-real-time intelligent control unit to generate an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors, and distributing the anomaly management model to a real-time intelligent control unit; a step of receiving the anomaly management model through an A1 interface to perform monitoring, checking whether an emergency event has occurred based on the output of the anomaly management model, and, if the emergency event has occurred, transmitting the generated control command to a base station connected to the sensor network through an E2 interface; and a step of processing the request according to an assigned priority when the non-real-time intelligent control unit and the real-time intelligent control unit simultaneously receive requests from the plurality of sensor networks in the service management organization unit. Claim 10 In claim 9, the step of collecting sensor data of a plurality of sensors as user plane (UP) data through an O1 interface from a sensor network composed of a plurality of sensors, training an AI / ML model using the collected sensor data and then installing it in a non-real-time intelligent control unit to generate an anomaly management model for detecting anomalies in the collected sensor data and predictive maintenance of the plurality of sensors, and distributing the anomaly management model to a real-time intelligent control unit is characterized by including the step of collecting the user plane data through the O1 interface using at least one management service among a provisioning management service (MnS), a fault supervision management service, a performance assurance management service, and a file management service.

Images