Data processing method and communication system

Abstract

Provided is a data processing method executed by a communication system which comprises: an information processing board having a signal processing unit, a user plane unit, an AI management section that receives and manages AI training data, and an AI storage section that stores the training data; and a network management section that manages the information processing board. The data processing method comprises: a request transmission step in which the AI management section, in response to receiving, from the communication terminal, an upload request for uploading the training data, transmits the training data upload request by the communication terminal to the network management section; a determination step in which the network management section, in response to receiving the upload request, determines whether or not the training data can be uploaded by the communication terminal; a setting step in which the network management section, in response to an affirmative determination in the determination step, sets a QoS policy for the communication terminal; and an acceptance step in which the AI management section authenticates an authentication request in accordance with the QoS policy by the communication terminal and accepts the uploading of the training data to the AI storage section.

Description

Data processing method and communication system

[0001] This invention relates to a data processing method and a communication system.

[0002] Patent Document 1 describes a wireless access network control device comprising at least one processor that performs the following actions: obtaining the operational status of a plurality of wireless access network nodes from a virtualization infrastructure that virtually manages a set of a plurality of wireless access network nodes; obtaining measured operational data from each of the wireless access network nodes; and issuing operational guidelines regarding the operation of each wireless access network node to at least one of the virtualization infrastructure and each of the wireless access network nodes based on the operational status and the operational data. [Prior Art Documents] [Patent Documents] [Patent Document 1] International Publication No. 2023 / 100385

[0003] In so-called AI (Artificial Intelligence)-RAN (Radio Access Network), GPUs (Graphics Processing Units) not used for RAN processing can be utilized for training. The advantages of this are data security and locality, but to realize this, data must be uploaded via a mobile network, and the traffic size can reach several terabytes, which is likely to cause congestion if not properly controlled.

[0004] According to an embodiment of the present invention, a data processing method executed by a communication system is provided. The communication system may include an information processing infrastructure and a network management unit that manages the information processing infrastructure. The information processing infrastructure may include a signal processing unit, a user plane unit, an AI management unit that receives and manages AI learning data, and an AI storage unit that stores the learning data. The data processing method may include a request transmission step in which the AI management unit transmits a request to upload learning data from a communication terminal to the network management unit in response to receiving an upload request for the learning data from the communication terminal. The data processing method may include a determination step in which the network management unit determines whether it is possible to upload the learning data by the communication terminal in response to receiving the upload request. The data processing method may include a setting step in which the network management unit sets a QoS policy for the communication terminal in response to determining that it is possible in the determination step. The data processing method may include an acceptance step in which the AI management unit authenticates an authentication request according to the QoS policy by the communication terminal and accepts the upload of the learning data to the AI storage unit.

[0005] In the data processing method, in the setting step, the network management unit may set a QoS policy for uploading learning data for the communication terminal in response to determining that it is possible to upload the learning data in the determination step.

[0006] In any of the data processing methods, in the setting step, the network management unit may set a QoS policy having a larger default maximum data burst amount than the QoS policy for uploading data other than learning data for the communication terminal in response to determining that it is possible to upload the learning data in the determination step.

[0007] In any of the above data processing methods, the network management unit may, in the setting stage, determine in the determination stage that it is possible to upload the learning data, and set a QoS policy for the communication terminal in which the packet delay budget is longer than the QoS policy for uploading data other than the learning data.

[0008] In any of the above data processing methods, the network management unit may, in the setting stage, determine in the determination stage that it is possible to upload the learning data, and set a QoS policy for the communication terminal that does not specify a packet delay budget.

[0009] In any of the above data processing methods, the network management unit may, in the setting stage, determine in the determination stage that it is possible to upload the learning data, and set a QoS policy for the communication terminal that has a packet error rate higher than the QoS policy for uploading data other than the learning data.

[0010] In any of the above data processing methods, the network management unit may, in the setting stage, determine in the determination stage that it is possible to upload the learning data, and set a QoS policy for the communication terminal that does not specify a packet error rate.

[0011] In any of the above data processing methods, the upload request transmitted by the communication terminal may include the use of the learning data, and the network management unit may, in the setting stage, set the QoS policy corresponding to the use, depending on whether it determined in the determination stage that uploading the learning data is possible. The network management unit may set different QoS policies depending on whether the use is learning that requires real-time performance or learning that does not require real-time performance.

[0012] In any of the above data processing methods, the network management unit may, at the decision stage, prioritize the upload of data other than learning data by other communication terminals over the upload of the learning data by the communication terminal.

[0013] In any of the above data processing methods, the signal processing unit may include a DU and a CU, and the user plane unit may include a UPF. The network management unit may include a RIC.

[0014] In any of the above data processing methods, the signal processing unit may include a BBU, and the user plane unit may include a PGW and an SGW.

[0015] According to one embodiment of the present invention, a communication system is provided. The communication system may include an information processing infrastructure and a network management unit that manages the information processing infrastructure. The information processing infrastructure may include a signal processing unit, a user plane unit, an AI management unit that receives and manages AI learning data, and an AI storage unit that stores the learning data. The AI ​​management unit may, upon receiving an upload request from a communication terminal to upload learning data, transmit the communication terminal's request to upload learning data to the network management unit. Upon receiving the upload request, the network management unit may determine whether or not the communication terminal can upload the learning data, and, if it determines that it can, may set a QoS policy for the communication terminal. The AI ​​management unit may authenticate the communication terminal's authentication request in accordance with the QoS policy and accept the upload of the learning data to the AI ​​storage unit.

[0016] The AI ​​management unit executes AI processing. The AI ​​processing may include AI processing related to RAN control (sometimes referred to as RAN control AI processing) and AI processing not related to RAN control (sometimes referred to as non-RAN control AI processing).

[0017] An example of AI-based RAN control processing is RIC (Rapid Interconnection). RIC is a technology that uses AI to optimize RAN wireless resources and automate RAN operations. RIC includes Non-RT RIC (Non-Real Time RIC) and Near-RT RIC (Near-Real Time RIC). Non-RT RIC is sometimes called Centralized RIC. Non-RT RIC is located within the SMO (Site Management Organization) that manages and orchestrates the RAN. Non-RT RIC generates and notifies policies related to RAN control and transmits information to Near-RT RIC. For example, Non-RT RIC generates a learning model for RAN control by performing machine learning using data collected from the RAN and transmits it to Near-RT RIC. Near-RT RICs are sometimes called Distributed RICs. Compared to Non-RT RICs, Near-RT RICs are located closer to RAN nodes and perform tasks such as controlling RAN nodes and resources. Near-RT RICs perform processing that is more real-time than Non-RT RICs. For example, Near-RT RICs perform inference processing related to RAN control using learning models acquired from Non-RT RICs. RAN control AI processing is not limited to RICs.

[0018] The AI ​​management unit may execute only Non-RT RIC, only Near-RT RIC, or both Non-RT RIC and Near-RT RIC. The AI ​​management unit may also choose not to execute either Non-RT RIC or Near-RT RIC.

[0019] Non-RAN controlled AI processing may be a so-called MEC AI application. Non-RAN controlled AI processing includes learning and inference processing of any AI that is not related to RAN control.

[0020] It should be noted that the above summary of the invention does not enumerate all the necessary features of the present invention. Furthermore, subcombinations of these features may also constitute an invention.

[0021] An example of the communication system 10 is shown in general terms. An example of the functional configuration of the distributed infrastructure 200 is shown in general terms. An example of the processing flow by the communication system 10 is shown in general terms. The 5QI table 500 is shown. An example of the functional configuration of the distributed infrastructure 200 is shown in general terms. An example of the hardware configuration of the computer 1200 that functions as the distributed infrastructure 200 or the management infrastructure 100 is shown in general terms.

[0022] The present invention will be described below through embodiments, but these embodiments are not intended to limit the scope of the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0023] Figure 1 schematically shows an example of a communication system 10. The communication system 10 includes a distributed infrastructure 200. The communication system 10 may include multiple distributed infrastructures 200. The communication system 10 may include a management infrastructure 100 that manages multiple distributed infrastructures 200. The communication system 10 may include multiple wireless base stations 300.

[0024] The communication system 10 according to this embodiment may constitute an AI-RAN. For example, in the communication system 10, the management infrastructure 100 and a plurality of distributed infrastructures 200 cooperate to control the RAN 310 and execute AI processing. The RAN 310 is a network that provides wireless communication services to the communication terminal 30. The communication terminal 30 may be a smartphone. The communication terminal 30 may be a tablet terminal. The communication terminal 30 may be a PC (Personal Computer). The communication terminal 30 may be a wearable terminal. The communication terminal 30 may be an IoT (Internet of Things) terminal. The communication terminal 30 may include any terminal that falls under IoE (Internet of Everything). RAN310 may be a virtualized vRAN (Virtual RAN), and the communication system 10 may perform control of the vRAN. RAN310 may also be a physical RAN, and the communication system 10 may perform control of the physical RAN.

[0025] The distributed infrastructure 200 may have one or more layers. The number of distributed infrastructures 200 placed in each layer may be one or more. Multiple distributed infrastructures 200 of one layer may be placed under the management infrastructure 100, or multiple layers of distributed infrastructures 200 may be placed under the management infrastructure 100. Figure 1 illustrates a case where the management infrastructure 100 has one layer of distributed infrastructures 200.

[0026] The distributed infrastructure 200 may be data centers located in various locations. The distributed infrastructure 200 may be composed of multiple devices. The distributed infrastructure 200 may be implemented on a virtualization infrastructure composed of multiple devices. The distributed infrastructure 200 may be implemented by a single device. In other words, the distributed infrastructure 200 may be a distributed device. The unit of the region in which the distributed infrastructure 200 is deployed may be any unit. For example, the region in which the distributed infrastructure 200 is deployed may be a region of roughly the same size as each prefecture in Japan. The region in which the distributed infrastructure 200 is deployed may be a region of finer scale than each prefecture in Japan. The region in which the distributed infrastructure 200 is deployed may be a region of broader scale than each prefecture in Japan.

[0027] The management infrastructure 100 may be a data center that manages multiple distributed infrastructures 200. The management infrastructure 100 may be composed of multiple devices. The management infrastructure 100 may be implemented on a virtualization infrastructure composed of multiple devices. The management infrastructure 100 may be implemented by a single device. In other words, the management infrastructure 100 may be a management device.

[0028] The management infrastructure 100 may be called the Core Brain, and the distributed infrastructure 200 may be called the Regional Brain. If two layers of distributed infrastructure 200 are arranged below the management infrastructure 100, the management infrastructure 100 may be called the Core Brain, the lower layer of distributed infrastructure 200 may be called the Regional Brain, and the further lower layer of distributed infrastructure 200 may be called the Sub-Regional Brain.

[0029] The distributed infrastructure 200 has a RAN control function for controlling the RAN 310 and an AI processing function for executing AI processing. The distributed infrastructure 200 may be an example of an information processing infrastructure. The distributed infrastructure 200 may have one or more CPUs (Central Processing Units). The distributed infrastructure 200 may have one or more GPUs. The distributed infrastructure 200 may have multiple superchips in which CPUs and GPUs are connected by an interconnect. The interconnect may be a memory-consistent interconnect capable of achieving high bandwidth and low latency. Thus, the distributed infrastructure 200 may have CPU resources and GPU resources as computing resources.

[0030] The management infrastructure 100 may have a RAN control function for controlling the RAN 310 and an AI processing function for executing AI processing. The management infrastructure 100 may have one or more CPUs. The management infrastructure 100 may have one or more GPUs. The management infrastructure 100 may have multiple superchips in which the CPUs and GPUs are connected by an interconnect. The interconnect may be a memory-consistent interconnect capable of achieving high bandwidth and low latency. Thus, the management infrastructure 100 may have CPU resources and GPU resources as computing resources.

[0031] The distributed infrastructure 200, deployed in each region, collects learning data from multiple communication terminals 30 located within its corresponding range, performs learning, and provides the collected learning data to the management infrastructure 100, where learning is performed. This enhances security by making it less likely for regional data to leak, and improves accuracy by enabling region-specific learning. However, with a large number of smartphones and IoT devices uploading learning data, the traffic size can reach several terabytes, and without proper control, congestion is likely to occur. The communication system 10 according to this embodiment has a configuration that contributes to solving such problems.

[0032] Figure 2 schematically shows an example of the functional configuration of the distributed infrastructure 200. Figure 2 illustrates the case where the communication system 10 conforms to the 5G (5th Generation) communication method.

[0033] The distributed infrastructure 200 comprises a DU 222, a CU 224, a UPF 226, an AI management unit 230, and an AI storage unit 212. The DU 222 and CU 224 may be examples of signal processing units. Multiple wireless base stations 300 are connected to the DU 222. The wireless base stations 300 comprise an RU 320 and multiple antennas 330. The UPF 226 may be an example of a user plane unit. The AI ​​management unit 230 performs processing related to AI. For example, the AI ​​management unit 230 receives and manages AI learning data. The AI ​​storage unit 212 stores the learning data received by the AI ​​management unit 230.

[0034] The communication terminal 30 uploads training data to the distributed infrastructure 200. The communication system 10 performs processing to appropriately accept the training data uploaded by the communication terminal 30.

[0035] Figure 3 schematically shows an example of the processing flow in the communication system 10. Here, we will explain the processing content when one communication terminal 30 uploads training data.

[0036] In step 102 (sometimes abbreviated as S), the communication terminal 30 requests the UPF 226 to create a session, and the session is established.

[0037] In S104, the communication terminal 30 sends an upload request to the AI ​​management unit 230 to upload the learning data. In S106, the AI ​​management unit 230, upon receiving the upload request from the communication terminal 30, sends an upload request for the learning data from the communication terminal 30 to the RIC 400. The RIC 400 may be an example of a network management unit.

[0038] In S108, RIC 400 determines whether or not the communication terminal 30 can upload the learning data, in response to receiving the upload request in S106. RIC 400 may determine whether or not the upload is possible according to known criteria. For example, RIC 400 may determine that the upload of learning data by the communication terminal 30 is permissible based on the usage rate of the PRB (Physical Resource Block), and determine that the upload is possible if it is not permissible. For example, RIC 400 may determine that the upload of learning data by the communication terminal 30 is permissible based on the traffic of the backbone network, and determine that the upload is possible if it is not permissible. RIC 400 may also determine whether or not the upload is possible according to criteria other than those mentioned above.

[0039] RIC400 may prioritize the uploading of data other than training data by other communication terminals 30 over the uploading of training data by communication terminal 30. For example, RIC400 may prioritize allocating resources to the uploading of data other than training data by other communication terminals 30 over the uploading of training data by communication terminal 30, and then determine that it is possible for communication terminal 30 to upload training data if there are still resources available for it to be allocated. If there are no resources available, it determines that uploading training data by communication terminal 30 is not possible.

[0040] If uploading is deemed possible, the process proceeds to S110; otherwise, it proceeds to S116.

[0041] In S110, RIC400 sets a QoS policy for the communication terminal 30. In this example, RIC400 sets a QoS policy for uploading training data.

[0042] In S112, the communication terminal 30 transmits an authentication request regarding the upload of learning data to the AI storage unit 212 in accordance with the QoS policy set in S110 to the AI management unit 230, and the AI management unit 230 performs authentication. In S114, the communication terminal 30 uploads the learning data to the AI storage unit 212.

[0043] In S116, the RIC 400 rejects the upload of learning data by the communication terminal 30 as a response to S106. In S116, the AI management unit 230 notifies the communication terminal 30 that the upload of learning data has been rejected.

[0044] Conventionally, learning data transmitted from a smartphone or an IoT terminal has been uploaded to a database on the Internet via, for example, a mobile network and used for learning. On the other hand, in the communication system 10 according to the present embodiment, since learning data can be collected in the AI storage unit 212 within the mobile network instead of the Internet, the security of the learning data can be improved, and the communication distance of the learning data can be shortened.

[0045] Conventionally, collecting learning data in the AI storage unit 212 within the mobile network has not been widely assumed, and the protocol therefor has not been established. On the other hand, in the communication system 10 according to the present embodiment, by implementing a protocol as shown in FIG. 3, it is possible to appropriately collect learning data.

[0046] Basically, real-time performance is not required for the upload of learning data by the communication terminal 30, and in many cases, a QoS of uploading when upload is possible is sufficient. In addition, the upload of learning data by the communication terminal 30 is characterized by a large number of traffic.

[0047] The RIC 400 according to this embodiment sets, for example, a QoS policy for learning data that has a larger default maximum data burst amount than the QoS policy for uploading data other than learning data. The RIC 400 sets, for example, a QoS policy for learning data that has a longer packet delay budget than the QoS policy for uploading data other than learning data. The RIC 400 sets, for example, a QoS policy for learning data that has no specified packet delay budget. The RIC 400 sets, for example, a QoS policy for learning data that has a higher packet error rate than the QoS policy for uploading data other than learning data. The RIC 400 sets, for example, a QoS policy for learning data that has no specified packet error rate. By doing so, a QoS policy suitable for the characteristics of uploading learning data by the communication terminal 30 can be set, and control can be performed so that the uploading of learning data is appropriately executed.

[0048] In the communication system 10, a QoS policy for learning data that is not included in the existing 5QI policy may be newly defined, and the QoS policy may be set for the communication terminal 30. Although the existing 5QI policy does not include a QoS policy assuming the uploading of learning data, by newly defining and using a QoS policy for learning data, it may be possible to handle the uploading of learning data more appropriately.

[0049] Note that the communication system 10 may select a QoS policy for learning data from among the existing 5QI policies.

[0050] Figure 4 shows an existing 5QI table 500. RIC400 may select a QoS policy with a larger default maximum data burst amount from among the multiple QoS policies included in the 5QI table 500 as the QoS policy for training data. RIC400 may select a QoS policy with a longer packet delay budget from among the multiple QoS policies included in the 5QI table 500 as the QoS policy for training data. RIC400 may select a QoS policy with no specified packet delay budget from among the multiple QoS policies included in the 5QI table 500 as the QoS policy for training data. RIC400 may select a QoS policy with a higher packet error rate from among the multiple QoS policies included in the 5QI table 500 as the QoS policy for training data. RIC400 may select a QoS policy with no specified packet error rate from among the multiple QoS policies included in the 5QI table 500 as the QoS policy for training data.

[0051] A 5QI value of "1" indicates an application for Conversational Voice. A 5QI value of "2" indicates an application for Conversational Voice (Live Streaming). A 5QI value of "3" indicates an application for Real Time Gaming and V2X messages. A 5QI value of "4" indicates an application for Non-Conversational Video (Buffered Streaming). A 5QI value of "65" indicates an application for Mission Critical user plane Push To Talk voice (e.g., MCPTT). A 5QI value of "66" is used for Non-Mission-Critical user plane Push To Talk voice. A 5QI value of "67" is used for Mission Critical Video user plane. A 5QI value of "71" is used for "Live" Uplink Streaming (e.g., TS 26.238). A 5QI value of "72" is used for "Live" Uplink Streaming (e.g., TS 26.238). A 5QI value of "73" indicates an application for "Live" Uplink Streaming (e.g., TS 26.238). A 5QI value of "74" indicates an application for "Live" Uplink Streaming (e.g., TS 26.238). A 5QI value of "76" indicates an application for "Live" Uplink Streaming (e.g., TS 26.238).

[0052] A 5QI value of "5" indicates an application for IMS Signaling. A 5QI value of "6" indicates an application for Video (Buffered Streaming) TCP-based. A 5QI value of "7" indicates an application for Voice, Video (Live Streaming) Interactive Gaming. A 5QI value of "8" indicates an application for Video (Buffered Streaming) TCP-based. A 5QI value of "9" indicates an application for Video (Buffered Streaming) TCP-based. A 5QI value of "69" is used for Mission Critical delay sensitive signaling. A 5QI value of "70" is used for Mission Critical Data. A 5QI value of "79" is used for V2X messages. A 5QI value of "80" is used for Low Latency eMBB applications Augmented Reality.

[0053] The 5QI value "82" is used for Discrete Automation. The 5QI value "83" is used for Discrete Automation and V2X messages. The 5QI value "84" is used for Intelligent transport systems. The 5QI value "85" is used for Electricity Distribution - high voltage and V2X messages. The 5QI value "86" is used for V2X messages. The 5QI value "87" is used for Interactive Service - Motion tracking data. The 5QI value "88" is used for Interactive Service - Motion tracking data. The 5QI value "89" is used for Visual content for cloud / edge / split rendering. The 5QI value "90" is used for Visual content for cloud / edge / split rendering.

[0054] Figure 5 schematically shows an example of the functional configuration of the distributed infrastructure 200. The distributed infrastructure 200 comprises a storage unit 210, a RAN function unit 220, and an AI management unit 230. The storage unit 210 includes an AI storage unit 212. The RAN function unit 220 may implement DU222. The RAN function unit 220 may implement CU224. The RAN function unit 220 may implement UPF226.

[0055] When the AI ​​management unit 230 receives an upload request from the communication terminal 30 to upload training data, it transmits the training data upload request from the communication terminal 30 to the network management unit. The network management unit determines whether or not the communication terminal 30 can upload the training data in response to receiving the upload request from the AI ​​management unit 230. If it determines that it can upload the training data, the network management unit sets a QoS policy for the communication terminal 30. The AI ​​management unit 230 may authenticate the authentication request from the communication terminal 30 in accordance with the QoS policy set by the network management unit and accept the upload of training data to the AI ​​storage unit 212.

[0056] The network management unit may set a QoS policy for the communication terminal 30 for uploading training data. The network management unit may set a QoS policy specifically for uploading training data for the communication terminal 30.

[0057] The network management unit may, for example, set a QoS policy for training data with a higher default maximum data burst than the QoS policy for uploading data other than training data. The network management unit may, for example, set a QoS policy for training data with a longer packet delay budget than the QoS policy for uploading data other than training data. The network management unit may, for example, set a QoS policy for training data with no specified packet delay budget. The network management unit may, for example, set a QoS policy for training data with a higher packet error rate than the QoS policy for uploading data other than training data. The network management unit may, for example, set a QoS policy for training data with no specified packet error rate.

[0058] The upload request transmitted by the communication terminal 30 may include the intended use of the training data. The network management unit may set a QoS policy for the communication terminal 30 that corresponds to the intended use of the training data. For example, the network management unit may set different QoS policies depending on whether the intended use of the training data requires real-time processing or not.

[0059] The network management unit, for example, sets a QoS policy with a higher default maximum data burst volume when the training data is not used in an application that requires real-time processing, compared to when real-time processing is required.

[0060] The network management unit, for example, sets a QoS policy with a longer packet delay budget when the training data is not used in an application where real-time performance is required, compared to when real-time performance is required. The network management unit sets a QoS policy without a specified packet delay budget when the training data is not used in an application where real-time performance is required, and sets a QoS policy with a specified packet delay budget when real-time performance is required.

[0061] The network management unit, for example, sets a QoS policy with a higher packet error rate when the training data is not used in an application that requires real-time processing, compared to when real-time processing is required. The network management unit sets a QoS policy without specifying a packet error rate when the training data is not used in an application that requires real-time processing, and sets a QoS policy with a specified packet error rate when the application requires real-time processing.

[0062] The network management unit may prioritize the uploading of data other than learning data by other communication terminals over the uploading of learning data by communication terminal 30.

[0063] In the above embodiment, the case in which the communication system 10 conforms to the 5G communication method is illustrated, but the communication system 10 is not limited to this, and may conform to the LTE communication method or the 6G communication method or later. When conforming to the LTE communication method, the wireless base station 300 may be equipped with an RRH instead of the RU320, the distributed infrastructure 200 may be equipped with a BBU instead of the DU222 and CU224, and an SGW, or an SGW and PGW, instead of the UPF226, and the network management unit may be an OSS (Operation Support System), etc. The BBU may be an example of a signal processing unit, and the SGW, or an SGW and PGW, may be an example of a user plane unit.

[0064] Figure 6 schematically shows an example of the hardware configuration of a computer 1200 that functions as a distributed infrastructure 200, a RIC 400, or a management infrastructure 100. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of the apparatus according to this embodiment, or to cause the computer 1200 to execute operations associated with the apparatus according to this embodiment or such one or more "parts", and / or to cause the computer 1200 to execute a process or a stage of such process according to this embodiment. Such a program may be executed by the CPU 1212 to cause the computer 1200 to execute specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0065] The computer 1200 according to this embodiment includes a CPU 1212, a GPU 1213, a RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 may be a DVD-ROM drive and a DVD-RAM drive, etc. The storage device 1224 may be a hard disk drive and a solid-state drive, etc. The computer 1200 also includes legacy input / output units such as a ROM 1230 and a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0066] The CPU 1212 operates according to the programs stored in the ROM 1230 and RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires the image data generated by the CPU 1212 and stores it in the frame buffer provided in the RAM 1214 or within itself, so that the image data is displayed on the display device 1218.

[0067] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from the DVD-ROM 1227, etc., and provides them to the storage device 1224. The IC card drive reads programs and data from the IC card and / or writes programs and data to the IC card.

[0068] The ROM 1230 stores boot programs and / or hardware-dependent programs of the computer 1200, which are executed by the computer 1200 when activated. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via USB ports, parallel ports, serial ports, keyboard ports, mouse ports, etc.

[0069] The program is provided on a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The program is read from the computer-readable storage medium and installed on a storage device 1224, RAM 1214, or ROM 1230, which are examples of computer-readable storage media, and executed by the CPU 1212. The information processing described within these programs is read by the computer 1200, resulting in coordination between the program and the various types of hardware resources described above. The apparatus or method may be configured to realize the operation or processing of information in accordance with the use of the computer 1200.

[0070] For example, when communication is performed between a computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into the RAM 1214 and, based on the processing described in the communication program, instruct the communication interface 1222 to perform communication processing. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer area provided in a recording medium such as the RAM 1214, storage device 1224, DVD-ROM 1227, or IC card, transmits the read transmission data to the network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

[0071] Furthermore, the CPU 1212 may read all or necessary parts of a file or database stored on an external recording medium such as a storage device 1224, a DVD drive 1226 (DVD-ROM 1227), or an IC card into the RAM 1214, and perform various types of processing on the data in the RAM 1214. The CPU 1212 may then write the processed data back to the external recording medium.

[0072] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and subjected to information processing. The CPU 1212 may perform various types of processing on the data read from the RAM 1214, including various types of operations, information processing, conditional judgments, conditional branching, unconditional branching, information retrieval / replacement, etc., as described throughout this disclosure and specified by the program instruction sequence, and write the results back to the RAM 1214. The CPU 1212 may also retrieve information in files, databases, etc., within the recording medium. For example, if a plurality of entries are stored in the recording medium, each having an attribute value of a first attribute associated with an attribute value of a second attribute, the CPU 1212 may search among the plurality of entries for an entry that matches the specified condition for the attribute value of the first attribute, read the attribute value of the second attribute stored in that entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0073] The program or software module described above may be stored on or near the computer 1200 in a computer-readable storage medium. Alternatively, a recording medium such as a hard disk or RAM provided within a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the program to the computer 1200 via the network.

[0074] In this embodiment, blocks in the flowchart and block diagram may represent a stage in a process in which an operation is performed or a "part" of a device that has the role of performing an operation. A particular stage and "part" may be implemented by a dedicated circuit, a programmable circuit supplied with computer-readable instructions stored on a computer-readable storage medium, and / or a processor supplied with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuit may include digital and / or analog hardware circuits, and may include integrated circuits (ICs) and / or discrete circuits. The programmable circuit may include reconfigurable hardware circuits, such as field-programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), which include logical AND, logical OR, exclusive OR, negated AND, negated OR, and other logical operations, flip-flops, registers, and memory elements.

[0075] A computer-readable storage medium may include any tangible device capable of storing instructions to be executed by a suitable device, and as a result, a computer-readable storage medium having instructions stored therein will comprise a product that includes instructions that can be executed to create means for performing operations specified in a flowchart or block diagram. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks (registered trademark), diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disk read-only memory (CD-ROM), digital versatile disk (DVD), Blu-ray (registered trademark) disk, memory stick, integrated circuit card, etc.

[0076] Computer-readable instructions may include assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, Java®, C++, and conventional procedural programming languages ​​such as the C programming language or similar programming languages.

[0077] Computer-readable instructions may be provided locally or via a wide area network (WAN) such as a local area network (LAN) or the internet to a processor or programmable circuit of a general-purpose computer, a special-purpose computer, or another programmable data processing device, so that the processor or programmable circuit of the programmable data processing device, such as a computer, may execute the computer-readable instructions to generate means for performing operations specified in a flowchart or block diagram. Here, the computer may be a PC (personal computer), a tablet computer, a smartphone, a workstation, a server computer, a general-purpose computer, or a special-purpose computer, and may also be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system and is a computer in a broad sense. In a distributed computing system, multiple computers execute a program collectively by each computer executing a part of the program and passing data during program execution between computers as needed.

[0078] Examples of processors include computer processors, central processing units (CPUs), processing units, microprocessors, digital signal processors, controllers, and microcontrollers. A computer may have one or more processors. In a multiprocessor system with multiple processors, each processor executes a portion of the program, and the processors collectively execute the program by passing program execution data between them as needed. For example, in the execution of multitasks, each of the multiple processors may execute a portion of each task in small chunks by switching tasks at each time slice. In this case, which part of a program each processor executes changes dynamically. Which part of a program each of the multiple processors executes may also be statically determined by multiprocessor-aware programming.

[0079] By using the invention according to this embodiment, it is possible to provide an environment in which learning data can be uploaded appropriately, thereby contributing to the achievement of Sustainable Development Goal (SDG) 9, "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation."

[0080] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention.

[0081] It should be noted that the execution order of operations, procedures, steps, and stages in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not explicitly stated as "before" or "prior to," and that these can be performed in any order unless the output of a previous operation is used in a later operation. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," and "next," for convenience, this does not mean that it is mandatory to perform the operations in that order.

[0082] 10 Communication system, 30 Communication terminal, 100 Management infrastructure, 200 Distributed infrastructure, 210 Memory unit, 212 AI memory unit, 220 RAN function unit, 222 DU, 224 CU, 226 UPF, 230 AI management unit, 300 Wireless base station, 310 RAN, 320 RU, 330 Antenna, 400 RIC, 500 5QI table, 1200 Computer, 1210 Host controller, 1212 CPU, 1213 GPU, 1214 RAM, 1216 Graphics controller, 1218 Display device, 1220 Input / Output controller, 1222 Communication interface, 1224 Storage device, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 Input / Output chip

Claims

1. A data processing method performed by a communication system comprising: a signal processing unit, a user plane unit, an AI management unit for receiving and managing AI learning data, an information processing infrastructure having an AI storage unit for storing the learning data, and a network management unit for managing the information processing infrastructure, the method comprising: a request transmission step in which the AI ​​management unit transmits an upload request for learning data from a communication terminal to the network management unit in response to the AI ​​management unit receiving an upload request for learning data from the communication terminal; a determination step in which the network management unit determines whether or not the communication terminal can upload the learning data in response to the upload request; a setting step in which the network management unit sets a QoS policy for the communication terminal in response to the determination step in response to the network management unit determining that it can upload the learning data; and a reception step in which the AI ​​management unit authenticates an authentication request from the communication terminal in accordance with the QoS policy and accepts the upload of the learning data to the AI ​​storage unit.

2. The data processing method according to claim 1, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for uploading learning data to the communication terminal.

3. The data processing method according to claim 1 or 2, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for the communication terminal that has a higher default maximum data burst amount than the QoS policy for uploading data other than learning data.

4. The data processing method according to any one of claims 1 to 3, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for the communication terminal in which the packet delay budget is longer than the QoS policy for uploading data other than the learning data.

5. The data processing method according to any one of claims 1 to 3, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for the communication terminal that does not specify a packet delay budget.

6. The data processing method according to any one of claims 1 to 5, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for the communication terminal in which the packet error rate is higher than that of the QoS policy for uploading data other than learning data.

7. The data processing method according to any one of claims 1 to 5, wherein the network management unit, in the setting stage, determines in the determination stage that it is possible to upload the learning data, sets a QoS policy for the communication terminal that does not specify a packet error rate.

8. The data processing method according to any one of claims 1 to 7, wherein the upload request transmitted by the communication terminal includes the use of the learning data, and the network management unit, in the setting stage, determines in the determination stage that the upload of the learning data is possible, to set the QoS policy corresponding to the use for the communication terminal.

9. The data processing method according to claim 8, wherein the network management unit sets different QoS policies depending on whether the learning data is used for an application that requires real-time performance or an application that does not require real-time performance.

10. The data processing method according to any one of claims 1 to 9, wherein the network management unit, in the determination stage, prioritizes the upload of data other than learning data by other communication terminals over the upload of the learning data by the communication terminal.

11. The data processing method according to any one of claims 1 to 10, wherein the signal processing unit includes a DU and a CU, and the user plane unit includes a UPF.

12. The data processing method according to claim 11, wherein the network management unit includes a RIC.

13. The data processing method according to any one of claims 1 to 10, wherein the signal processing unit includes a BBU, and the user plane unit includes a PGW and an SGW.

14. A communication system comprising: an information processing infrastructure having a signal processing unit, a user plane unit, an AI management unit for receiving and managing AI learning data, and an AI storage unit for storing the learning data; and a network management unit for managing the information processing infrastructure, wherein the AI ​​management unit, upon receiving an upload request for uploading learning data from a communication terminal, transmits the communication terminal's upload request for learning data to the network management unit; the network management unit, upon receiving the upload request, determines whether or not the communication terminal can upload the learning data, and, if it determines that it can, sets a QoS policy for the communication terminal; and the AI ​​management unit authenticates the communication terminal's authentication request in accordance with the QoS policy and accepts the upload of the learning data to the AI ​​storage unit.

Images