Method and apparatus for ai-based communication in wireless communication system supporting open-radio access network

Abstract

The present disclosure relates to a method and apparatus for efficient AI-based communication in a wireless communication system supporting O-RAN, wherein, in the wireless communication system supporting O-RAN, according to embodiments of the present disclosure, an O-RU apparatus of a base station including an O-DU apparatus and the O-RU apparatus, comprises: a transceiver; one or more processors including processing circuitry; and a memory storing commands, wherein the commands, when executed individually or collectively by the one or more processors, may cause the O-RU apparatus to: receive, from the O-DU apparatus through the transceiver, a control plane message including at least one AI command for processing uplink data passing through the O-RU apparatus; and transmit, to the O-DU apparatus through the transceiver, a user plane message including uplink data to which the at least one AI command is applied.

Description

Method and apparatus for AI-based communication in a wireless communication system supporting an open wireless access network

[0001] The present disclosure relates to a communication method and apparatus of a base station in a wireless communication system supporting an O-RAN (open-radio access network).

[0002] To meet the increasing demand for wireless data traffic following the 4G system (i.e., the LTE (long-term evolution) system), 5G systems have been developed and commercialized. 5G systems can be implemented in the mmWave band. To mitigate path loss and increase the transmission distance of radio waves in the mmWave band, beamforming, massive array multiple input / output (massive MIMO), full-dimensional multiple input / output (Full Dimensional MIMO: FD-MIMO), array antenna, analog beamforming, and large-scale antenna technologies are being discussed for 5G systems.

[0003] With the commercialization of 5G systems, connected devices, which are increasing explosively, will be connected to communication networks. Consequently, it is expected that there will be a need for enhanced functionality and performance, as well as integrated operation of connected devices, in next-generation communication systems, such as 5G systems or 6G systems, which are referred to as post-5G communication systems. To this end, new research is underway regarding 5G performance improvement and complexity reduction using Artificial Intelligence (AI) and / or Machine Learning (ML), support for AI services, support for metaverse services, and drone communication.

[0004] In addition, O-RAN technology proposed by the O-RAN Alliance is being researched for the construction of open intelligent wireless access networks for next-generation mobile communication systems. While existing RANs utilized baseband units (BBUs) and radio units (RUs) provided by a single vendor, the O-RAN Alliance divides the wireless access network into the following three units / modules, enabling various vendors to build open intelligent wireless access networks.

[0005] - O-RU (O-RAN radio unit): A device (logical node) that hosts processing of Low-PHY (lower physical layer) based on RF (radio frequency) and lower layer partitioning.

[0006] - O-DU (O-RAN distributed unit): A device (logical node) that hosts High-PHY (upper physical layer), MAC (Media Access Control), and RLC (Radio Link Control) layer processing based on lower layer partitioning.

[0007] - O-CU (O-RAN central unit) - A device (logical node) hosting PDCP (Packet Data Translation Protocol), SDAP (Service Data Adaptation Protocol), RRC (Radio Resource Control), and other control functions

[0008] To establish the above-mentioned open intelligent wireless access network, a method for efficient communication between the O-RU and O-DU is required.

[0009] The present disclosure provides a method and apparatus for efficient AI-based communication in a wireless communication system supporting O-RAN, and a storage medium for the same.

[0010] The present disclosure provides a method and apparatus for AI-based communication between an O-DU (O-RAN distributed unit) and an O-RU (O-RAN radio unit) in a wireless communication system supporting an O-RAN.

[0011] The present disclosure provides a method and apparatus in which an O-DU and an O-RU in a wireless communication system supporting an O-RAN distribute uplink data or downlink data based on AI.

[0012] The present disclosure provides a method and apparatus for efficiently transmitting and receiving control plane messages containing AI command(s) between an O-DU and an O-RU for AI-based communication in a wireless communication system supporting an O-RAN.

[0013] According to an embodiment of the present disclosure, in a wireless communication system supporting O-RAN, the O-RU device of a base station comprising an O-DU device and an O-RU device comprises a transceiver, one or more processors including processing circuitry, and a memory for storing instructions. When the instructions are executed individually or collectively by the one or more processors, the O-RU device may cause the O-DU device to receive a control plane message containing at least one AI instruction for processing uplink data passing through the O-RU device via the transceiver. In one embodiment, when the instructions are executed individually or collectively by the one or more processors, the O-RU device may cause the O-DU device to transmit a user plane message containing uplink data to which the at least one AI instruction is applied via the transceiver.

[0014] In one embodiment, the at least one AI command is learned based on at least one of the location and channel status of the O-RU device, and the uplink data to which the at least one AI command is applied may be uplink data with quality compensated based on the at least one AI command or uplink data scheduled based on the at least one AI command.

[0015] In one embodiment, the at least one AI command may include at least one AI weight value to be applied to the uplink data and at least one AI parameter identifier that identifies the at least one AI weight.

[0016] In one embodiment, the at least one AI command includes at least one consecutive AI parameter identifier, and at least one weight value corresponding to the at least one consecutive AI parameter identifier can be divided into at least one group based on the amount of data of the control plane message.

[0017] In one embodiment, when the commands are executed individually or collectively by the one or more processors, the O-RU device may cause the at least one AI command to perform AI inference for applying the processing of the uplink data.

[0018] In addition, according to an embodiment of the present disclosure, in a wireless communication system supporting O-RAN, the O-RU device of a base station comprising an O-DU device and an O-RU device comprises a transceiver, one or more processors including a processing circuit, and a memory for storing instructions. When the instructions are executed individually or collectively by the one or more processors, the O-RU device may cause the O-RU device to transmit a control plane message containing information inferred from the O-RU device for downlink data processing through the transceiver. In one embodiment, when the instructions are executed individually or collectively by the one or more processors, the O-RU device may cause the O-DU device to receive a user plane message containing downlink data processed based on the inferred information through the transceiver.

[0019] In one embodiment, when the commands are executed individually or collectively by the one or more processors, the O-RU device may further cause the O-RU device to infer the information included in the control plane message based on at least one of the location and channel state of the O-RU device for downlink data processing.

[0020] In one embodiment, when the commands are executed individually or collectively by the one or more processors, the O-RU device may further cause to receive, through the transceiver from the O-DU device, a user plane message containing downlink data, or a control plane message containing channel information related to a PMI (precoding matrix indicator) or SRS (sounding reference signal).

[0021] In one embodiment, the downlink data processed based on the inferred information may include downlink data scheduled based on the inferred information or downlink data to which at least one AI command is applied based on the inferred information.

[0022] In one embodiment, the at least one AI command includes at least one AI weight value applied to the downlink data and at least one consecutive AI parameter identifier identifying the at least one AI weight, and the at least one weight value corresponding to the consecutive at least one AI parameter identifier may be divided into at least one group.

[0023] In addition, according to an embodiment of the present disclosure, in a wireless communication system supporting O-RAN, the O-RU device of a base station comprising an O-DU device and an O-RU device may include a transceiver and a processor. Here, the processor may be configured to perform AI inference related to at least one of the location and channel state of the O-RU device for processing uplink data or downlink data passing through the O-RU device. Here, the processor may be configured to transmit a user plane message or a control plane message containing data based on the AI ​​inference to the O-DU device of the base station via the transceiver.

[0024] In addition, according to an embodiment of the present disclosure, a method performed at the O-RU device of a base station comprising an O-DU device and an O-RU device in a wireless communication system supporting an O-RAN may include a process of performing AI inference related to at least one of the location and channel state of the O-RU device for processing uplink data or downlink data passing through the O-RU device. In addition, the method may include a process of transmitting a user plane message or a control plane message containing data obtained by the AI ​​inference to the O-DU device of the base station.

[0025] In addition, according to an embodiment of the present disclosure, a storage medium storing at least one computer-readable command, wherein the at least one command causes the O-RU device to perform at least one operation when executed individually or collectively by one or more processors including a processing circuit of the O-RU device in a base station comprising an O-DU device and an O-RU device in a wireless communication system supporting an O-RAN, and the at least one operation may include an operation of performing AI inference related to at least one of the location and channel state of the O-RU device for processing uplink data or downlink data passing through the O-RU device. In one embodiment, the at least one operation may include an operation of transmitting a user plane message or a control plane message containing data by the AI ​​inference to the O-DU device of the base station.

[0026] In one embodiment, the at least one operation further includes receiving a control plane message from the O-DU device that includes at least one AI command learned for processing the uplink data when the O-RU device processes the uplink data, and the transmitting operation may include transmitting the user plane message that includes the uplink data to which the at least one AI command is applied to the O-DU device.

[0027] In one embodiment, the at least one AI command includes at least one consecutive AI parameter identifier, and at least one weight value corresponding to the at least one consecutive AI parameter identifier can be divided into at least one group based on the amount of data of the control plane message.

[0028] In one embodiment, the at least one operation further includes receiving a user plane message or a control plane message containing the downlink data from the O-DU device when the O-RU device processes the downlink data, and the transmitting operation may include transmitting the control plane message containing information inferred based on at least one of the location and channel state of the O-RU device to the O-DU device.

[0029] In one embodiment, the AI ​​inference can be performed based on at least one AI command using a neural processing unit (NPU).

[0030] FIG. 1 is a drawing showing an example of the structure of a base station in a wireless communication system supporting O-RAN to which the present disclosure applies.

[0031] FIG. 2 is a diagram showing an example of data flow between an O-DU and an O-RU in a wireless communication system supporting an O-RAN to which the present disclosure applies.

[0032] FIG. 3 is a diagram showing an example of a configuration for AI-based communication between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure.

[0033] FIG. 4 is a flowchart illustrating an example of an AI-based communication method between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure.

[0034] FIG. 5 is a flowchart illustrating an example of an AI-based communication method for uplink transmission between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure.

[0035] FIG. 6 is a flowchart illustrating an example of an AI-based communication method for downlink transmission between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure, and

[0036] FIG. 7 is a drawing showing an example of a configuration of a network entity in a wireless communication system according to an embodiment of the present disclosure.

[0037] The operating principles of the present disclosure will be described in detail below with reference to the attached drawings. In describing the present disclosure below, specific descriptions of related known functions or configurations will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Furthermore, the terms described below are defined in consideration of their functions in the present disclosure, 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.

[0038] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure is complete and to fully inform those skilled in the art of the scope of the invention, and the present disclosure is defined only by the scope of the claims. Throughout the specification, like reference numerals refer to like components.

[0039] At this point, it will be understood that each block of the process flow diagrams and combinations of the flow diagrams can be executed by computer program instructions.

[0040] Additionally, each block may represent a module, segment, or part of code containing one or more executable instructions for executing a specified logical function(s). It should also be noted that in some alternative execution examples, the functions mentioned in the blocks may occur out of order. For instance, two blocks described in succession may actually be executed substantially simultaneously, or the blocks may be executed in reverse order according to their corresponding functions.

[0041] In this embodiment, the term "part" refers to a software or hardware component such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit), and the "part" performs certain roles. However, the meaning of "part" is not limited to software or hardware. The "part" may be configured to reside in an addressable storage medium or configured to run one or more processors. Thus, as an example, the "part" includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and "parts" may be combined into a smaller number of components and "parts" or further separated into additional components and "parts." In addition, the components and 'parts' may be implemented to utilize one or more CPUs within the device or secure multimedia card. Also, in the embodiments, 'parts' may include one or more processors.

[0042] In the present disclosure, each of the phrases such as “A / B”, “A or B”, “A and / or B”, “at least one of A and B”, “at least one of A or B”, “A, B or C”, “at least one of A, B and C”, and “at least one of A, B, or C” may include any one of the items listed together in the corresponding phrase, or all possible combinations thereof. Terms such as “first”, “second”, or “first” or “second” may be used simply to distinguish a component from another component and do not limit the components in any other aspect (e.g., importance or order).

[0043] Terms used in the following description to identify connection nodes, terms referring to network entities, terms referring to messages, terms referring to interfaces between network entities, terms referring to various identification information, etc., are examples provided for the convenience of explanation. Accordingly, the present disclosure is not limited to the terms described below, and other terms referring to objects having equivalent technical meanings may be used.

[0044] In the present disclosure, a base station (BS) is a network entity capable of performing resource allocation for terminals and communicating with terminals through a wireless network, and may be at least one of an eNode B, Node B, gNB, RAN (Radio Access Network), AN (Access Network), RAN node, IAB (Integrated Access / Backhaul) node, a wireless access unit, a base station controller, a node on a network, or a TRP (transmission reception point). A terminal (user equipment: UE) may be at least one of a terminal, MS (Mobile Station), cellular phone, smartphone, computer, or a multimedia system capable of performing communication functions.

[0045] For convenience of explanation, some terms and names defined in the ORAN standard may be used below. However, the present disclosure is not limited by the said terms and names and may be equally applied to systems conforming to other standards.

[0046] FIG. 1 is a diagram showing an example of the structure of a base station in a wireless communication system supporting O-RAN to which the present disclosure applies.

[0047] Referring to FIG. 1, a base station (100) (e.g., eNB / gNB, etc.) may include an O-DU (110) and at least one O-RU (120-1, …120-n: 120). The O-DU (110) controls the operation of at least one O-RU (120-1, …120-n: 120) and is a network node capable of hosting at least one layer processing of High-PHY (upper physical layer), MAC (media access control), and RLC (radio link control) based on lower layer split. And the above at least one O-RU (120-1, …120-n : 120) is a network node capable of transmitting and receiving at least one of user plane (UP) data / messages and control plane (CP) data / messages with the O-DU (110) under the control of the O-DU (110), and hosting processing of Low-PHY (lower physical layer) based on RF (radio frequency) and lower layer partitioning.

[0048] When using lower layer splitting (High-PHY, Low-PHY) in the above O-DU (110) and O-RU (120), the logical interface between the O-DU (110) and O-RU (120) can use the LLS (Lower Layer Split) defined in the ORAN standard. In FIG. 1, LLS-U (Lower Layer Split-User Plane) is a logical interface used for transmitting and receiving UP data / messages between the O-DU (110) and O-RU (120), and LLS-C (Lower Layer Split-Control Plane) is a logical interface used for transmitting and receiving CP data / messages between the O-DU (110) and O-RU (120).

[0049] FIG. 2 is a diagram showing an example of data flow between an O-DU (210) and an O-RU (220) in a wireless communication system supporting an O-RAN to which the present disclosure applies.

[0050] Referring to FIG. 2, the data flow of the user plane (UP) may include at least one of reference numerals 1a, 1b, and 1c, and the data flow of the control plane (CP) may include at least one of reference numerals 2a to 2f. And reference numeral S is a data flow for timing and synchronization.

[0051] An example of the data flow of the user plane (UP) in Fig. 2 is as follows.

[0052] - Data Flow 1a: Flow of IQ data in the frequency domain of the FFT (Fast Fourier Transform) of the downlink

[0053] - Data Flow 1b: Flow of IQ data in the frequency domain of the uplink's FFT (Fast Fourier Transform).

[0054] - Data Flow 1c: Flow of PRACH (Physical Random Access Channel) IQ data in the FFT frequency domain

[0055] The above IQ data refers to the modulation signals of the I channel (In-phase) and Q channel (Quadrature-phase) in communication.

[0056] An example of the data flow of the control plane (CP) in Fig. 2 is as follows.

[0057] - Data Flow 2a: Scheduling commands and beamforming commands on the downlink and / or uplink

[0058] - Data Flow 2b: LAA (Licensed-Assisted Access) LBT (listen-before-talk) configuration commands and requests for unlicensed band communication

[0059] - Data Flow 2c: LAA LBT status and response messages

[0060] - Data Flow 2d : Terminal Channel Information (UE Channel Information)

[0061] - Data Flow 2e: ACK / NACK message

[0062] - Data Flow 2f: Wake-up Ready Indication

[0063] In the example of FIG. 2, the data flow of the user plane (UP) and the data flow of the control plane (CP) may each include a downlink data flow and / or an uplink data flow. In the embodiments of the present disclosure described below, the user plane (U-plane) data / messages may include at least one of the data flows of the user plane (UP) in the example of FIG. 2, and the control plane (C-plane) data / messages may include at least one of the data flows of the control plane (CP) in the example of FIG. 2.

[0064] In the present disclosure, user plane (U-plane) data / messages and control plane (C-plane) data / messages are not limited to the example of FIG. 2, and various types of data may be transmitted. The terms U-plane data and U-plane message(s) have the same meaning and may be used interchangeably. The terms C-plane data and C-plane message(s) have the same meaning and may be used interchangeably.

[0065] Conventional O-RAN specifications did not include separate technology for supporting on-radio AI, but the present disclosure enables the provision of AI service(s) based on the installation location and / or channel environment of the O-RU through on-radio AI. In the present disclosure, at least one of U-plane data and C-plane data for O-RAN-based AI services may be transmitted and received between the O-DU and the O-RU. In the present disclosure, a predefined AI model may be installed, stored, or included in the O-DU and the O-RU to process the U-plane data and / or C-plane data. The same AI model may be shared between the O-DU and the O-RU.

[0066] In one embodiment, AI training for learning parameter(s) that constitute AI command(s) based on at least one of the (installation) location and channel status of the O-RU may be performed in the AI ​​model. Additionally, for processing uplink data or downlink data passing through the O-RU, AI inference related to at least one of the location and channel status of the O-RU may be performed in the AI ​​model.

[0067] In one embodiment, for processing uplink data, the AI ​​model of the O-DU may transmit C-plane data to the O-RU containing AI command(s) learned based on at least one of the O-RU's (installation) location and channel status. Upon receiving the C-plane data, the AI ​​model of the O-RU may transmit U-plane data to the O-DU to which the AI ​​command(s) have been applied through AI inference. The U-plane data to which the AI ​​command(s) have been applied may be understood, for example, as uplink data with improved / compensated QoS (quality of service) based on the AI ​​command(s) or uplink data scheduled based on the AI ​​command(s).

[0068] In one embodiment, for processing downlink data, the AI ​​model of the O-DU provides U-plane or C-plane data to the O-RU, and can receive inferred information (e.g., channel status information, etc.) based on at least one of the (installation) location and channel status of the O-RU from the AI ​​model of the O-RU that received the U-plane or C-plane data. Based on the inferred information, the O-DU can perform scheduling of U-plane data to be transmitted through the O-RU by considering at least one of the (installation) location and channel status of the O-RU.

[0069] The present disclosure proposes a method of performing AI training in either the O-DU or the O-RU and performing AI inference in the other. The present disclosure proposes a method of performing AI training in the O-DU and performing AI inference in the O-RU. If, unlike the present disclosure, it is assumed that both AI training and AI inference are performed in the O-RU, data processing using the AI ​​model in the O-RU increases rapidly, which can increase the complexity of the O-RU. Therefore, to reduce the complexity of the O-RU, the present disclosure proposes a method of performing AI inference in the O-RU and performing AI training in the O-DU. In this case, if a processor for AI services (e.g., a neural processing unit (NPU), which is an accelerator designed for AI computation) is included in the O-RU and utilized for AI inference, AI services can be provided with lower complexity. Additionally, AI-related information can be transmitted and received through a communication interface between the O-DU and the O-RU (e.g., the LLS interface of FIG. 1 as a fronthaul interface). The logical link for communication between the above O-DU and O-RU can utilize the fronthaul defined in the ORAN standard.

[0070] FIG. 3 is a diagram showing an example of a configuration for AI-based communication between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure.

[0071] For the basic structure and data flow of the O-DU (310) and O-RU (320) in the example of FIG. 3, refer to the descriptions in FIG. 1 and FIG. 2.

[0072] In the example of FIG. 3, it is assumed that O-DU (310) and O-RU (320) each have all or at least part of a predefined AI model installed, stored, or included. The O-DU (310) and O-RU (320) may use the same AI model. The predefined AI model may provide / perform at least one of an AI training function and an AI inference function. In the example of FIG. 3, O-DU (310) includes an AI training unit (311) in the predefined AI model, and may transmit AI-related information(s) trained / learned by the AI ​​training unit (311) to O-RU (320) through an O-RAN interface (e.g., fronthaul interface).

[0073] In the present disclosure, the learned AI-related information(s) may include AI command(s) that include AI parameter(s) related to the processing of uplink data or downlink data passing through the O-RU (320). The processing of the uplink data or downlink data may be for improving the Quality of Service (QoS) of the uplink data or downlink data passing through the O-RU (320) based on at least one of the (installation) location and channel status of the O-RU (320) (i.e., considering the communication environment at the installation location of the O-RU (320)) or for improving communication performance considering Line of Sight (LOS), None Line of Sight (NLOS), etc. at the installation location of the O-RU (320).

[0074] In FIG. 3, the O-DU (310) can transmit C-plane data / messages containing AI command(s) composed of AI parameter(s) to the O-RU (320).

[0075] In one embodiment, the O-RU (320) can transmit to the O-DU (310) O-RAN capability information, which includes information indicating that the O-RU (320) supports AI inference capabilities, using, for example, an O-RAN management plane (M-Plane) message defined in the O-RAN specification.

[0076] An example of the AI ​​parameter(s) in the present disclosure is as shown in [Table 1] below. Each parameter of the above AI parameter(s) (e.g., an AI weight value ("aiWeight") used in an AI model) can be identified using an AI parameter identifier ("aiParamId"). That is, the above AI command(s) may include a plurality of AI weight values ​​learned in the AI ​​training unit (311) based, for example, at least one of the (installation) location and channel state of the O-RU (320), and a plurality of AI parameter identifiers (aiParamIds) that identify said plurality of AI weight values. The plurality of AI weight values ​​may be learned in the O-DU (310) periodically or non-periodically, or provided to the O-RU (320) upon the occurrence of a predetermined event for updating the AI ​​weight values, based on at least one of the location of the O-RU (320) and the changing channel environment around the O-RU (320). The plurality of AI parameter identifiers (aiParamIds) may be set as consecutive values, for example, if they are parameters related to the same AI service.

[0077] [Table 1]

[0078]

[0079] In the example of FIG. 3, the O-RU (320) includes an AI inference unit (321) that provides AI inference functions in the predefined AI model, and the AI ​​inference unit (321) can apply / reflect learned AI command(s) provided from the AI ​​training unit (311) of the O-DU (310) to user plane data (e.g., uplink data) and transmit it to the O-DU (310). In other words, the O-RU (320) can transmit U-plane data / messages to which AI command(s) have been applied / reflected to the O-DU (310).

[0080] In the present disclosure, AI commands may include a series of AI parameter identifiers (aiParamIds), and a number of AI weight values ​​mapped to / corresponding to the series of AI parameter identifiers may be divided into one or more groups according to the type of parameter. A detailed description will be provided later.

[0081] In one embodiment, if the user plane data to be processed at the O-RU (320) is uplink data passing through the O-RU (320), the O-DU (310) transmits C-plane data containing AI command(s) learned based on at least one of the (installation) location and channel status of the O-RU (320) to the O-RU (320), and the O-RU (320) transmits U-plane data containing uplink data to which the AI ​​command(s) have been applied / reflected to the O-DU (310).

[0082] In one embodiment, if the user plane data to be processed at the O-RU (320) is downlink data passing through the O-RU (320), the O-DU (310) transmits to the O-RU (320) U-plane data containing downlink data or C-plane data containing channel information related to the O-RU, configured based on the PMI (precoding matrix indicator) or SRS (sounding reference signal) received by the O-DU from the UE, and the O-RU (320) that receives the U-plane data or C-plane data can transmit to the O-DU (310) C-plane data containing information inferred by the AI ​​inference unit (321) of the O-RU (320) based on at least one of the (installation) location and channel status of the O-RU (320). O-DU (310) can perform scheduling of U-plane data to be transmitted through O-RU (320) based on the above-mentioned information, taking into account at least one of the (installation) location and channel status of O-RU (320).

[0083] FIG. 4 is a flowchart illustrating an example of an AI-based communication method between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure. In the example of FIG. 4, the O-DU and O-RU may have the same configuration as the O-DU (310) and O-RU (320) in the embodiment of FIG. 3.

[0084] Referring to Fig. 4, it is assumed that in process 401, the O-DU and O-RU have the same predefined AI model installed / stored / included.

[0085] In the 402 process, the O-RU may receive a control plane (C-plane) message from the O-DU containing at least one AI command learned in the O-DU based on at least one of the O-RU's (installation) location and channel state for processing uplink data passing through the O-RU. The at least one AI command may include, for example, at least one AI weight value learned in the O-DU based on at least one of the O-RU's (installation) location and channel state and at least one AI parameter identifier identifying the at least one AI weight.

[0086] In the 403 process, the O-RU that receives a control plane (C-plane) message containing at least one AI command may apply / reflect the at least one AI weight value included in the at least one AI command to the processing of the uplink data using an AI inference function in the predefined AI model. In this process, the O-RU may perform AI inference related to at least one of the O-RU's (installation) location and channel status. In the 403 process, the O-RU may transmit a user plane (U-plane) message to the O-DU containing uplink data to which the at least one AI command has been applied / reflected (e.g., uplink data corrected by AI inference or uplink data scheduled by the AI ​​command).

[0087] Additionally, as an optional embodiment, in the above 402 process, the O-RU may receive a user plane (U-plane) message containing downlink data from the O-DU for processing downlink data passing through the O-RU, or a control plane (C-plane) message containing channel information related to the O-RU, configured based on the PMI or SRS received by the O-DU from the UE. In the above 403 process, the O-RU that has received the user plane (U-plane) or control plane (C-plane) message may transmit a control plane (C-plane) message to the O-DU containing information inferred based on at least one of the (installation) location and channel status of the O-RU. The O-DU may adjust / control scheduling, etc. for transmitting downlink data passing through the O-RU based on the inferred information.

[0088] FIG. 5 is a flowchart illustrating an example of an AI-based communication method for uplink transmission between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure. In the example of FIG. 5, the O-DU and O-RU may have the same configuration as the O-DU (310) and O-RU (320) in the embodiment of FIG. 3.

[0089] Referring to Fig. 5, it is assumed that in process 501, the O-DU and O-RU have the same predefined AI model installed / stored / included.

[0090] In the 502 process, the O-RU can transmit a user plane (U-plane) message containing uplink data to the O-DU.

[0091] In the 503 process, the O-DU can perform AI learning on the received uplink data based on at least one of the (installation) location and channel status of the O-RU to generate / configure at least one AI command including at least one AI weight value. The at least one AI weight value may be applied / reflected, for example, in compensation processing to improve the quality of I-channel and Q-channel data of uplink data transmitted through the O-RU, power control for the O-RU in the uplink, beamforming control for the O-RU in the uplink, or scheduling for receiving uplink data from the UE via the O-RU.

[0092] In the above 503 process, the O-RU may receive a control plane (C-plane) message from the O-DU containing at least one AI command learned in the O-DU based on at least one of the O-RU's (installation) location and channel status for processing uplink data passing through the O-RU. Here, the processing of the uplink data may, for example, be related to at least one of various examples of the at least one AI weight value. Additionally, the processing of the uplink data may be related to at least one of the flow of user plane data in the uplink in the example of FIG. 2. The at least one AI command may include the at least one AI weight value and at least one AI parameter identifier that identifies the at least one AI weight.

[0093] In the above 503 process, the O-RU, having received a control plane (C-plane) message containing at least one AI command from the O-DU, can apply / reflect the at least one AI weight value included in the at least one AI command to the processing of the uplink data using an AI inference function in the predefined AI model. In this process, the O-RU can perform AI inference related to at least one of the O-RU's (installation) location and channel status. Accordingly, in the present disclosure, the O-RU can perform AI inference based on the AI ​​command received from the O-DU to perform uplink data processing with low complexity, which can improve QoS or adaptively respond to a rapidly changing communication environment by considering the O-RU's installation location (e.g., dead zone, non-dead zone, etc.) and / or the O-RU's channel status. Additionally, the AI ​​command can be applied / reflected in scheduling for receiving uplink data from the O-DU.

[0094] In the above 504 process, the O-RU may transmit to the O-DU a user plane (U-plane) message containing uplink data to which at least one AI command has been applied / reflected (e.g., uplink data compensated by AI inference). The user plane (U-plane) message may be transmitted to an external network, such as an IP (internet protocol) network, via a core network not shown.

[0095] FIG. 6 is a flowchart illustrating an example of an AI-based communication method for downlink transmission between an O-DU and an O-RU in a wireless communication system supporting an O-RAN according to an embodiment of the present disclosure. In the example of FIG. 6, the O-DU and O-RU may have the same configuration as the O-DU (310) and O-RU (320) in the embodiment of FIG. 3.

[0096] Referring to Fig. 6, it is assumed that in process 601, the O-DU and O-RU have the same predefined AI model installed / stored / included.

[0097] In the 602 process, the O-DU can transmit to the O-RU a user plane (U-plane) message containing downlink data or a control plane (C-plane) message containing channel information related to the O-RU, configured based on the PMI or SRS received by the O-DU from the UE.

[0098] In the 603 process, the O-RU may transmit a control plane (C-plane) message to the O-DU, which includes information inferred by an AI model based on at least one of the O-RU's (installation) location and channel state, for processing downlink data passing through the O-RU. The inferred information may include, for example, O-RU specific channel information that can be used in scheduling for the transmission of downlink data passing through the O-RU from the O-DU. Generally, the UE may transmit channel state information (CSI) indicating the UE's channel state to the network periodically or non-periodically. As an example, in a 3GPP-based communication system, the CSI may include at least one of a Channel Quality Indicator (CQI), a Precoding Matrix Indicator (PMI), and a Rank Indication (RI) related to the channel state in the downlink. The O-RU receives the CSI from the UE(s) within the service area and can generate / configure O-RU specific channel information indicating a more accurate channel state at the installation location of the O-RU based on the received CSI. In one embodiment, the O-RU specific channel information may be configured by modifying / supplementing the received CSI to reflect a communication environment such as LOS or NLOS at the location of the O-RU.

[0099] In the 604 process, the O-DU can apply / reflect AI command(s) based on the O-DU’s AI learning to downlink data to be transmitted via the O-RU, or adjust / control scheduling for the transmission of downlink data via the O-RU, based on the inferred information received from the O-RU via the control plane (C-plane) message. For example, the O-DU can perform AI learning based on the inferred information to generate / configure at least one AI command including at least one AI weight value that can be applied to or for scheduling downlink data. Additionally, as an example, the at least one AI weight value may include, for instance, a weight value for compensation processing to improve the quality of I-channel and Q-channel data of downlink data to be transmitted via the O-RU.

[0100] Hereinafter, an example of the configuration of AI command(s) included in a C-plane message according to an embodiment of the present disclosure will be described with reference to [Table 2] to [Table 5] below.

[0101] [Table 2]

[0102]

[0103] In the present disclosure, the O-DU may transmit AI parameter(s) learned using an AI model to the O-RU using AI command(s) such as the example in [Table 2]. The AI ​​command(s) may include at least one learned AI weight value ("aiWeight") and at least one AI parameter identifier ("aiParamId") that identifies the at least one AI weight value.

[0104] Referring to [Table 2] above, the AI ​​command(s) may include a general transport header field and a common header field as defined in the existing O-RAN standard. The transport header field provides basic data routing functions, such as a description of the data flow type. The common header field may include dataDirection (Tx or Rx), payloadVersion, filterIndex, frameId, subframeId, slotId, startSymbolId, numberOfSections, and sectionType fields at the base station (RAN). The transport header field and the common header field can ensure backward compatibility by maintaining the same format as the existing O-RAN standard. For a description of the fields announced above in the existing O-RAN specifications, refer to, for example, O-RAN.WG4.CUS.0-R003 (O-RAN Working Group 4 (Open Fronthaul Interfaces WG), Control, User and Synchronization Plane Specification).

[0105] In [Table 2] above, the numberOfParamGroups field may indicate the number of groups for consecutive AI parameter identifiers (aiParamIds), as exemplified in [Table 1] above. In the present disclosure, the amount of data that can be included in a single C-plane message may be limited. For example, if the maximum amount of data that can be included in a single C-plane message is 9,000 bytes, and the amount of data of the AI ​​command(s) exceeds 9,000 bytes, the AI ​​parameters of the AI ​​command(s) related to the same AI service may be divided into two groups and transmitted. In this case, the numberOfParamGroups may be set to 2. Additionally, in [Table 2], the AI ​​parameter header (aiParamHdr) field may indicate information related to the AI ​​weight value (aiWeight) (e.g., number of bits, data type, compression status, etc.).

[0106] In the example of [Table 2] above, the section fields distinguished by the extension flag (ef) field may represent the values ​​of consecutive AI weights (AI weight) starting from the start AI parameter identifier (startAiParamId) to the number of AI parameters (numAiParams). Accordingly, in the present disclosure, the consecutive AI weight values ​​may be included in at least one C-plane message based on the number of parameter groups (numberOfParamGroups) and transmitted in segments from the O-DU to the O-RU. If the C-plane message is a non-delay managed C-plane message, multiple C-plane messages may be transmitted in segments over a relatively long period of time.

[0107] In addition, in the present disclosure, the O-RU can notify the O-DU that the O-RU supports AI inference functions using an O-RAN management plane (M-Plane) message, and the O-DU that has confirmed that the O-RU supports AI inference functions can construct and transmit a C-plane message containing AI command(s) such as the example in [Table 2] to the O-RU based on the capability information of the O-RU.

[0108] [Table 3]

[0109]

[0110] [Table 3] above shows an example of AI weight values ​​(aiWeights) learned in O-DU and a series of AI parameter identifiers (aiParamIds) that identify the AI ​​weight values ​​(aiWeights), when the AI ​​model used in O-RU and O-DU is a self-attention method. The self-attention method is a method in which each element of an input sequence interacts with all other elements to learn information using Query, Key, and Value parameters in the weight matrix of the AI ​​model. [Table 3] above assumes, as an example, that the feature dimension (d_model) used in self-attention is 64 and the parameter used is an 8-bit integer (int8). In addition, the example in [Table 3] above illustrates a case where a total of 16,634 parameters are trained in the AI ​​model of O-DU, and aiParamIds 0 to 4159 are weights for key parameters, aiParamIds 4160 to 8319 are weights for query parameters, aiParamIds 8320 to 12479 are weights for value parameters, and aiParamIds 12480 to 16639 are weights for projection parameters. For convenience of explanation, the present disclosure assumes that the AI ​​model uses a self-attention method, but it should be noted that the available AI models are not limited to the self-attention method. That is, various AI models capable of classifying AI command(s) into at least one group of at least one consecutive AI parameter identifier and at least one weight value corresponding to said consecutive AI parameter identifier can be used in O-RU and O-DU.

[0111] The O-DU of the present disclosure can perform learning using an AI model and, as a result of the learning, transmit AI command(s) including learned AI weight values ​​(aiWeights) and consecutive AI parameter identifiers (aiParamIds), such as the example in [Table 3], to the O-RU using a C-plane message.

[0112] The examples in [Table 4] and [Table 5] below illustrate how to split AI weight values ​​(aiWeights) into two C-plane messages so that the total amount of data that can be included in one C-plane message does not exceed, for example, 9000 bytes, according to the data amount limit of C-plane messages.

[0113] [Table 4]

[0114]

[0115] [Table 5]

[0116]

[0117] [Table 4] above shows an example in which AI weights for key parameters from aiParamId 0 to 4159 and AI weights for query parameters from aiParamId 4160 to 8319 are included in the first C-plane message, and [Table 5] shows an example in which AI weights for value parameters from aiParamId 8320 to 12479 and AI weights for projection parameters from aiParamId 12480 to 16639 are included in the second C-plane message.

[0118] According to the embodiments of the present disclosure described above, for AI-based communication between the O-RU and O-DU, AI learning considering the installation location and channel environment of the O-RU, which is installed regionally and has environmental characteristics, is performed by the O-DU, and AI inference is performed by the O-RU, thereby enabling on-radio AI to be realized in the O-RU and O-DU with low complexity.

[0119] According to an embodiment of the present disclosure, in a wireless communication system supporting O-RAN, the O-RU device (320) of a base station comprising an O-DU device (310) and an O-RU device (320) comprises a transceiver (710), one or more processors (730) comprising processing circuitry, and a memory (720) for storing instructions, wherein when the instructions are executed individually or collectively by the one or more processors (730), the O-RU device (320) receives a control plane message (402, 503) from the O-DU device (310) via the transceiver (710) containing at least one Artificial Intelligence (AI) command for processing uplink data passing through the O-RU device (320), and a user including uplink data to which the at least one AI command is applied, to the O-DU device (310) via the transceiver (710). It can cause the transmission of user plane messages (403, 504).

[0120] Here, the at least one AI command is learned based on at least one of the location and channel state of the O-RU device (320), and the uplink data to which the at least one AI command is applied may be uplink data with quality compensated based on the at least one AI command or uplink data scheduled based on the at least one AI command.

[0121] Here, the at least one AI command may include at least one AI weight value to be applied to the uplink data and at least one AI parameter identifier that identifies the at least one AI weight.

[0122] Here, the at least one AI command includes at least one consecutive AI parameter identifier, and at least one weight value corresponding to the at least one consecutive AI parameter identifier can be divided into at least one group based on the amount of data of the control plane message.

[0123] Here, when the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320) may cause the at least one AI command to perform AI inference for applying to the processing of the uplink data.

[0124] According to an embodiment of the present disclosure, in a wireless communication system supporting O-RAN, the O-RU device (320) of a base station comprising an O-DU device (310) and an O-RU device (320) comprises a transceiver (710), one or more processors (730) comprising a processing circuit, and a memory (720) for storing commands, and when the commands are executed individually or collectively by the one or more processors (730), the O-RU device (320) may cause the O-DU device (310) to transmit a control plane message containing information inferred from the O-RU device (320) for downlink data processing through the transceiver (710) (403, 603), and to receive a user plane message (604) from the O-DU device (310) through the transceiver (710) containing downlink data processed based on the inferred information.

[0125] Here, when the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320) may further cause the information included in the control plane message to be inferred based on at least one of the location and channel state of the O-RU device (320) for downlink data processing.

[0126] Here, when the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320) may further cause to receive a user plane message containing downlink data, or a control plane message containing channel information related to PMI or SRS, from the O-DU device (310) through the transceiver (710).

[0127] Here, the downlink data processed based on the inferred information may include downlink data scheduled based on the inferred information or downlink data to which at least one AI command is applied based on the inferred information.

[0128] Here, the at least one AI command includes at least one AI weight value applied to the downlink data and at least one consecutive AI parameter identifier identifying the at least one AI weight, and at least one weight value corresponding to the consecutive at least one AI parameter identifier can be divided into at least one group.

[0129] According to an embodiment of the present disclosure, in a wireless communication system supporting an O-RAN, the O-RU device of a base station comprising an O-DU device and an O-RU device may include a transceiver and a processor configured to perform AI inference related to at least one of the location and channel state of the O-RU device for processing uplink data or downlink data passing through the O-RU device, and to transmit a user plane message or a control plane message containing data based on the AI ​​inference to the O-DU device of the base station through the transceiver.

[0130] Here, when the processor processes the uplink data, the processor may be further configured to receive a control plane message containing at least one AI command learned for processing the uplink data from the O-DU device via the transceiver, and to transmit a user plane message containing the uplink data to which the at least one AI command is applied to the O-DU via the transceiver.

[0131] Here, the at least one AI command includes at least one consecutive AI parameter identifier, and at least one weight value corresponding to the at least one consecutive AI parameter identifier can be divided into at least one group based on the amount of data of the control plane message.

[0132] Here, when the processor processes the downlink data, the processor may be further configured to receive a user plane message or a control plane message containing the downlink data from the O-DU device through the transceiver, and to transmit the control plane message containing information inferred based on at least one of the location and channel status of the O-RU device to the O-DU device through the transceiver.

[0133] Here, the processor may include a neural processing unit that performs operations based on at least one AI instruction.

[0134] FIG. 7 is a diagram showing an example of a configuration of a network entity in a wireless communication system according to an embodiment of the present disclosure. The configuration of FIG. 7 corresponds to the O-RU or O-DU described in the embodiments of FIG. 1 to 6.

[0135] The network entity of FIG. 7 may include a processor (701), a transceiver (703), and a memory (705). The processor (701), transceiver (703), and memory (705) of the network entity of FIG. 7 may operate according to at least one of the embodiments of FIG. 1 to 6. However, the components of the network entity are not limited to the examples described above. For example, the network entity may include more components or fewer components than the components described above. In addition, the processor (701), transceiver (703), and memory (705) may be implemented in the form of a single chip. The transceiver (703) is a collective term for the receiver and the transmitter of the network entity and can transmit and receive signals with a terminal or another network entity. At this time, the transmitted and received signal may include at least one of control information and data (e.g., control plane data / message and user plane data / message). To this end, the transceiver (703) may include various configurations for performing communication via wired and / or wireless means. The transceiver (703) may receive a signal and output it to the processor (701), and transmit the signal output from the processor (701). Additionally, the transceiver (703) may receive a communication signal and output it to the processor (701), and transmit the signal output from the processor (701) to another network entity via a network. The memory (705) may store programs and data necessary for the operation of a network entity according to at least one of the embodiments of FIGS. 1 to 6. Additionally, the memory (705) may store control information or data included in a signal obtained from the network entity. The memory (705) may be composed of a storage medium or a combination of storage media such as ROM, RAM, hard disk, CD-ROM and DVD.Additionally, the processor (701) can control a series of processes to enable a network entity to operate according to at least one of the embodiments of FIGS. 1 to 6. For example, the processor (701) may include at least one processor and can control operations according to the present disclosure.

[0136] Methods according to the embodiments described in the claims or specification of the present disclosure may be implemented in the form of hardware, software, or a combination of hardware and software.

[0137] When implemented in software, a computer-readable storage medium may be provided for storing one or more programs (software modules). One or more programs stored in the computer-readable storage medium are configured for execution by one or more processors within an electronic device. One or more programs include instructions that cause the electronic device to execute methods according to the embodiments described in the claims or specification of this disclosure.

[0138] Such programs (software modules, software) may be stored in random access memory, non-volatile memory including flash memory, ROM (Read Only Memory), Electrically Erasable Programmable Read Only Memory (EEPROM), magnetic disc storage devices, Compact Disc-ROM (CD-ROM), Digital Versatile Discs (DVDs), or other forms of optical storage devices, magnetic cassettes. Alternatively, they may be stored in memory composed of some or all of these. Additionally, each constituent memory may include multiple units.

[0139] Additionally, the above program may be stored on an attachable storage device that can be accessed via a communication network such as the Internet, Intranet, Local Area Network (LAN), Wide LAN (WLAN), or Storage Area Network (SAN), or a combination thereof. Such a storage device may be connected to a device performing an embodiment of the present disclosure through an external port. Additionally, a separate storage device on a communication network may be connected to a device performing an embodiment of the present disclosure.

[0140] In the specific embodiments of the present disclosure described above, the components included in the invention are expressed in a singular or plural form according to the specific embodiments presented. However, the singular or plural expression is selected to suit the situation presented for convenience of explanation, and the present disclosure is not limited to singular or plural components; even if a component is expressed in the plural form, it may be composed of a singular form, or even if a component is expressed in the singular form, it may be composed of a plural form.

[0141] Meanwhile, although specific embodiments have been described in the detailed description of the present disclosure, it is understood that various modifications are possible within the scope of the present disclosure. Therefore, the scope of the present disclosure should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof.

Claims

1. In a wireless communication system supporting an O-RAN (open-radio access network), the O-RU (O-RAN distributed unit) device (310) and the O-RU (O-RAN radio unit) device (320) of a base station, the O-RU device (320) Transmitter / receiver (710); One or more processors (730) including processing circuitry; and The O-RU device (320) includes a memory (720) for storing instructions, and when the instructions are executed individually or collectively by one or more processors (730), the O-RU device (320), A control plane message including at least one AI (Artificial Intelligence) command for processing uplink data passing through the O-RU device (320) via the transceiver (710) from the O-DU device (310) (402, 503), and An O-RU device that causes the O-DU device (310) to transmit (403, 504) a user plane message containing uplink data to which the at least one AI command is applied, through the transceiver (710).

2. In Paragraph 1, The above at least one AI command is learned based on at least one of the position and channel state of the O-RU device (320), and The O-RU device to which the uplink data to which the at least one AI command is applied is uplink data whose quality is compensated based on the at least one AI command or uplink data scheduled based on the at least one AI command.

3. In Paragraph 1, The above at least one AI command is an O-RU device comprising at least one AI weight value to be applied to the uplink data and at least one AI parameter identifier identifying the at least one AI weight.

4. In Paragraph 1, The above at least one AI command includes at least one consecutive AI parameter identifier, and An O-RU device in which at least one weight value corresponding to at least one consecutive AI parameter identifier is divided into at least one group based on the amount of data of the control plane message.

5. In Paragraph 1, When the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320), An O-RU device that causes AI inference to be performed for applying at least one AI command to the processing of the uplink data.

6. In a wireless communication system supporting an O-RAN (open-radio access network), the O-RU (O-RAN distributed unit) device (310) and the O-RU (O-RAN radio unit) device (320) of a base station, the O-RU device (320) Transmitter / receiver (710); One or more processors (730) including processing circuitry; and The O-RU device (320) includes a memory (720) for storing instructions, and when the instructions are executed individually or collectively by one or more processors (730), the O-RU device (320), A control plane message containing information inferred from the O-RU device (320) for downlink data processing is transmitted to the O-DU device (310) through the transceiver (710) (403, 603), An O-RU device that causes to receive (604) a user plane message containing downlink data processed based on the inferred information from the above O-DU device (310) through the above transceiver (710).

7. In Paragraph 6, When the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320), An O-RU device that further causes to infer the information included in the control plane message based on at least one of the position and channel state of the O-RU device (320) for processing the downlink data.

8. In Paragraph 6, When the above commands are executed individually or collectively by the one or more processors (730), the O-RU device (320), From the above O-DU device (310) through the above transceiver (710), A user flat message containing down link data, or An O-RU device that further causes to receive control plane messages containing channel information related to PMI (precoding matrix indicator) or SRS (sounding reference signal).

9. In Paragraph 6, The downlink data processed based on the above-mentioned inferred information is, An O-RU device comprising downlink data scheduled based on the above-mentioned inferred information or downlink data to which at least one artificial intelligence (AI) command is applied based on the above-mentioned inferred information.

10. In Paragraph 9, The above at least one AI command is, It includes at least one AI weight value applied to the above downlink data and at least one consecutive AI parameter identifier identifying the at least one AI weight, An O-RU device in which at least one weight value corresponding to at least one consecutive AI parameter identifier is divided into at least one group.

11. In a storage medium storing at least one instruction readable by a computer, The above at least one command causes the O-RU device to perform at least one operation when executed individually or collectively by one or more processors including the processing circuitry of the O-RU device in a base station comprising an O-DU (O-RAN distributed unit) device and an O-RU (O-RAN radio unit) device in a wireless communication system supporting an O-RAN (open-radio access network). The above at least one operation is: An operation to perform artificial intelligence (AI) inference related to at least one of the location and channel state of the O-RU device for processing uplink data or downlink data passing through the O-RU device; and A storage medium comprising the operation of transmitting a user plane message or a control plane message containing data based on AI inference to the O-DU device of the base station.

12. In Paragraph 11, The above at least one operation is: When processing the uplink data in the O-RU device, the operation further includes receiving a control plane message from the O-DU device that includes at least one AI command learned for processing the uplink data. A storage medium comprising the above transmitting operation, wherein the above transmitting operation includes the operation of transmitting the user plane message, which includes uplink data to which the at least one AI command is applied, to the O-DU device.

13. In Paragraph 12, The above at least one AI command includes at least one consecutive AI parameter identifier, and A storage medium in which at least one weight value corresponding to at least one consecutive AI parameter identifier is divided into at least one group based on the amount of data of the control plane message.

14. In Paragraph 13, The above at least one operation is: When processing the downlink data in the O-RU device, the operation of receiving a user plane message or a control plane message containing the downlink data from the O-DU device is further included. A storage medium comprising the above transmitting operation, wherein the above transmitting operation includes the operation of transmitting the control plane message to the O-DU device, the control plane message including information inferred based on at least one of the location and channel state of the O-RU device.

15. In Paragraph 11, A storage medium in which the above AI inference is performed based on at least one AI command using a neural processing unit (NPU).

Images