Method and apparatus for s-ai synchronization in communication system using s-ai

Abstract

The present disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate than a 4G communication system, such as LTE. In addition, the present disclosure relates to a method performed by a terminal in a wireless communication system, the method comprising the steps of: receiving a system information block (SIB) from a base station, wherein the SIB includes scheduling information related to system information block-artificial intelligence (SIB-AI); identifying a change in synchronized-AI (S-AI) being used on the basis of information included in the SIB; receiving the SIB-AI from the base station on the basis of the scheduling information, wherein the SIB-AI includes information on S-AI to be changed; and changing the S-AI on the basis of the S-AI information to be changed, which is included in the SIB-AI.

Description

Method and device for S-AI synchronization in a communication system using S-AI

[0001] The present disclosure relates to the operation of a terminal and a base station in a wireless communication system. More specifically, the present disclosure relates to a method and device for maintaining synchronization of S-AI in a wireless communication system using artificial intelligence (AI).

[0002] Looking back at the evolution of wireless communication over successive generations, technologies have primarily been developed for human-facing services such as voice, multimedia, and data. With the commercialization of 5G (5th-generation) communication systems, an explosive increase in connected devices is expected to be connected to communication networks. Examples of networked objects include vehicles, robots, drones, home appliances, displays, smart sensors installed in various infrastructures, construction equipment, and factory equipment. Mobile devices are expected to evolve into diverse form factors, including augmented reality glasses, virtual reality headsets, and holographic devices. In the 6th-generation (6G) era, efforts are being made to develop improved 6G communication systems to connect hundreds of billions of devices and objects and provide diverse services. For this reason, 6G communication systems are often referred to as "Beyond 5G" systems.

[0003] The 6G communication system, expected to be realized around 2030, will have a maximum transmission speed of terabytes per second (i.e., 1,000 gigabits per second) and a wireless latency of 100 microseconds (μsec). In other words, compared to 5G, the transmission speed in a 6G communication system will be 50 times faster, while the wireless latency will be reduced to one-tenth.

[0004] To achieve these high data rates and ultra-low latency, 6G communication systems are being considered for implementation in the terahertz band (e.g., from 95 gigahertz (GHz) to 3 terahertz (THz)). Compared to the millimeter wave (mmWave) band introduced in 5G, the terahertz band is expected to experience more severe path loss and atmospheric absorption, making it more crucial to ensure signal reach, or coverage, in this band. Key technologies to ensure coverage include radio frequency (RF) components, antennas, new waveforms that offer better coverage than orthogonal frequency division multiplexing (OFDM), beamforming, and multiple antenna transmission technologies such as massive multiple-input and multiple-output (MIMO), full-dimensional MIMO (FD-MIMO), array antennas, and large-scale antennas. In addition, new technologies such as metamaterial-based lenses and antennas, high-dimensional spatial multiplexing using orbital angular momentum (OAM), and reconfigurable intelligent surfaces (RIS) are being discussed to improve the coverage of terahertz band signals.

[0005] In addition, in order to improve frequency efficiency and system network, 6G communication systems are developing full duplex technology that utilizes the same frequency resources at the same time for uplink and downlink; network technology that integrates satellites and high-altitude platform stations (HAPS); network structure innovation technology that supports mobile base stations and enables optimization and automation of network operation; dynamic spectrum sharing technology through collision avoidance based on spectrum usage prediction; AI-based communication technology that utilizes artificial intelligence (AI) from the design stage and internalizes end-to-end AI support functions to realize system optimization; and next-generation distributed computing technology that realizes services with complexity that exceeds the limits of terminal computing capabilities by utilizing ultra-high-performance communication and computing resources (mobile edge computing (MEC), cloud, etc.). In addition, efforts are being made to further strengthen connectivity between devices, further optimize networks, promote softwareization of network entities, and increase the openness of wireless communications through the design of new protocols to be used in 6G communication systems, the implementation of hardware-based security environments, the development of mechanisms for the safe use of data, and the development of technologies for maintaining privacy.

[0006] Research and development of these 6G communication systems are expected to enable a new level of hyper-connected experience through the hyper-connectivity of 6G communication systems, which encompass not only connections between things but also connections between people and things. Specifically, 6G communication systems are expected to enable services such as truly immersive extended reality (XR), high-fidelity mobile holograms, and digital replicas. Furthermore, services such as remote surgery, industrial automation, and emergency response, which are provided through 6G communication systems through enhanced security and reliability, will be applied in diverse fields such as industry, medicine, automobiles, and home appliances.

[0007] The present disclosure provides a method and device for effectively providing services in a wireless communication system. The present disclosure relates to a method for maintaining synchronization of Synchronized Artificial Intelligence (S-AI) between a base station and a terminal connected to the base station, and updating the S-AI when the network environment changes.

[0008] In a method performed by a terminal in a wireless communication system according to one embodiment of the present disclosure, the method may include the steps of: receiving a System Information Block (SIB) from a base station, the SIB including scheduling information related to System Information Block-Artificial Intelligence (SIB-AI), and confirming a change of a synchronized-AI (S-AI) being used based on information included in the SIB, receiving a SIB-AI from the base station based on the scheduling information, the SIB-AI including information on an S-AI to be changed, and performing a change of the S-AI based on the changed S-AI information included in the SIB-AI.

[0009] In a method performed by a base station in a wireless communication system according to one embodiment of the present disclosure, the method may include the steps of: determining to perform a synchronized-Artificial Intelligence (S-AI) change procedure; transmitting a System Information Block (SIB) to a terminal, the SIB including scheduling information related to the System Information Block-AI (SIB-AI); and transmitting a SIB-AI to the terminal based on the scheduling information, the SIB-AI including information on an S-AI to be changed, and performing a change of the S-AI based on the S-AI information to be changed.

[0010] In accordance with one embodiment of the present disclosure, a terminal of a wireless communication system may include a transceiver and a control unit connected to the transceiver, wherein the control unit receives a System Information Block (SIB) from a base station, the SIB includes scheduling information related to System Information Block-Artificial Intelligence (SIB-AI), and determines a change in a synchronized-AI (S-AI) being used based on information included in the SIB, receives an SIB-AI from the base station based on the scheduling information, the SIB-AI includes information on an S-AI to be changed, and performs a change of the S-AI based on the S-AI information to be changed included in the SIB-AI.

[0011] In accordance with one embodiment of the present disclosure, a base station of a wireless communication system may include a transceiver and a control unit connected to the transceiver, wherein the control unit determines to perform a synchronized-Artificial Intelligence (S-AI) change procedure, transmits a System Information Block (SIB) to a terminal, the SIB includes scheduling information related to the System Information Block-AI (SIB-AI), and transmits the SIB-AI to the terminal based on the scheduling information, wherein the SIB-AI includes information on an S-AI to be changed, and a change of the S-AI is performed based on the S-AI information to be changed.

[0012] The present disclosure provides a method and device for effectively performing wireless communication system services to better adapt to changing environments by updating AI. Furthermore, the present disclosure can maintain the S-AI held by the base station and the S-AI set in the terminal identically when the base station decides to change the S-AI based on the network environment.

[0013] The effects that can be obtained from the present disclosure are not limited to the effects mentioned in the various embodiments, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the present disclosure belongs from the description below.

[0014] Figure 1 is a diagram illustrating a wireless communication system including S-AI.

[0015] Figure 2 is a diagram illustrating the process of updating S-AI.

[0016] FIG. 3a is a diagram illustrating a procedure for maintaining synchronization for S-AI according to one embodiment of the present disclosure.

[0017] FIG. 3b is a diagram illustrating a procedure for maintaining synchronization with S-AI by defining a new SIB according to one embodiment of the present disclosure.

[0018] FIG. 4a is a diagram illustrating a procedure for changing (or updating) S-AI according to one embodiment of the present disclosure.

[0019] FIG. 4b is a diagram illustrating a procedure for changing (or updating) S-AI using an S-AI deploy result report message and an S-AI inference message according to one embodiment of the present disclosure.

[0020] FIG. 5a is a diagram illustrating an S-AI synchronization procedure according to one embodiment of the present disclosure.

[0021] FIG. 5b is a diagram illustrating an S-AI synchronization procedure according to one embodiment of the present disclosure.

[0022] FIG. 6 is a diagram illustrating a procedure when acquisition of system information for a new S-AI fails according to an embodiment of the present disclosure.

[0023] FIG. 7a is a diagram illustrating a procedure for a terminal to change (or update) S-AI according to one embodiment of the present disclosure.

[0024] FIG. 7b is a diagram illustrating a procedure for a terminal to change (or update) an S-AI by sending an S-AI deploy result report message according to one embodiment of the present disclosure.

[0025] FIG. 8A is a diagram illustrating an S-AI synchronization procedure of a terminal according to one embodiment of the present disclosure.

[0026] FIG. 8b is a diagram illustrating an S-AI synchronization procedure of a terminal according to an embodiment of the present disclosure.

[0027] FIG. 9 is a diagram illustrating a procedure of a terminal when acquisition of system information for a new S-AI fails according to an embodiment of the present disclosure.

[0028] FIG. 10a is a diagram illustrating a procedure for a base station to change (or update) S-AI according to one embodiment of the present disclosure.

[0029] FIG. 10b is a diagram illustrating a procedure for a base station to change (or update) S-AI by transmitting an S-AI inference message according to one embodiment of the present disclosure.

[0030] FIG. 11a is a diagram illustrating an S-AI synchronization procedure of a base station according to one embodiment of the present disclosure.

[0031] FIG. 11b is a diagram illustrating an S-AI synchronization procedure of a base station according to one embodiment of the present disclosure.

[0032] FIG. 12 is a diagram illustrating a base station procedure when acquisition of system information for a new S-AI fails according to one embodiment of the present disclosure.

[0033] FIG. 13 is a diagram illustrating a configuration of a terminal according to an embodiment of the present disclosure.

[0034] FIG. 14 is a diagram illustrating a configuration of a base station according to an embodiment of the present disclosure.

[0035] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0036] In describing the embodiments, descriptions of technical details that are well known in the technical field to which the present disclosure pertains and are not directly related to the present disclosure will be omitted. This is to more clearly convey the gist of the present disclosure without obscuring it by omitting unnecessary explanations.

[0037] For the same reason, some components in the attached drawings are exaggerated, omitted, or schematically depicted. Furthermore, the dimensions of each component do not entirely reflect the actual size. Identical or corresponding components in each drawing are assigned the same reference numbers.

[0038] The advantages and features of the present disclosure, and methods for achieving them, will become clearer with reference to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided only to complete the configuration of the present disclosure and to fully inform those skilled in the art of the present disclosure of the scope of the disclosure, and the present disclosure is defined only by the scope of the claims. Like reference numerals refer to like elements throughout the specification. In addition, when describing the present disclosure, if a specific description of a related function or configuration is determined to unnecessarily obscure the gist of the present disclosure, the detailed description thereof will be omitted. In addition, the terms described below are terms defined in consideration of the functions of the present disclosure, and these may vary depending on the intention or custom of the user or operator. Therefore, the definitions should be made based on the contents throughout the specification.

[0039] Hereinafter, the base station is an entity that performs resource allocation of the terminal, and may be at least one of a gNode B, an eNode B, a Node B, a BS (Base Station), a wireless access unit, a base station controller, or a node on a network. The terminal may include a UE (User Equipment), an MS (Mobile Station), a cellular phone, a smartphone, a computer, or a multimedia system capable of performing a communication function. In the present disclosure, downlink (DL) refers to a wireless transmission path of a signal transmitted from a base station to a terminal, and uplink (UL) refers to a wireless transmission path of a signal transmitted from a terminal to a base station. In addition, although the LTE or LTE-A system may be described below as an example, the embodiments of the present disclosure may also be applied to other communication systems having a similar technical background or channel type. For example, the 5th generation mobile communication technology (5G, new radio, NR) developed after LTE-A may be included here, and the 5G below may also be a concept that includes existing LTE, LTE-A, and other similar services. In addition, the present disclosure may be applied to other communication systems with some modifications within a scope that does not significantly deviate from the scope of the present disclosure, as judged by a person having skilled technical knowledge.

[0040] At this time, it will be understood that each block of the processing flow diagrams and combinations of the flow diagrams can be performed by computer program instructions. These computer program instructions can be installed in a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing equipment, so that the instructions executed by the processor of the computer or other programmable data processing equipment create a means for performing the functions described in the flow diagram block(s). These computer program instructions can also be stored in a computer-available or computer-readable memory that can be directed to a computer or other programmable data processing equipment for implementation in a specific manner, so that the instructions stored in the computer-available or computer-readable memory can also produce a manufactured item that includes an instruction means for performing the functions described in the flow diagram block(s). Since the computer program instructions can also be installed on a computer or other programmable data processing device, a series of operational steps can be performed on the computer or other programmable data processing device to create a computer-executable process, and the instructions that cause the computer or other programmable data processing device to perform the steps for performing the functions described in the flowchart block(s) can also provide steps for performing the functions described in the flowchart block(s).

[0041] Additionally, each block may represent a module, segment, or portion of code that contains one or more executable instructions for performing a specific logical function(s). It should also be noted that in some alternative implementation examples, the functions described in the blocks may occur out of order. For example, two blocks depicted in succession may actually be executed substantially concurrently, or the blocks may sometimes be executed in reverse order, depending on their respective functions.

[0042] Here, the term '~ unit' used in the present embodiment means software or hardware components such as FPGA (field programmable gate array) or ASIC (application-specific integrated circuit), and the '~ unit' performs certain roles. However, the '~ unit' is not limited to software or hardware. The '~ unit' may be configured to be on an addressable storage medium, and may be configured to play one or more processors. Accordingly, as an example, the '~ unit' includes components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, 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 '~ units' may be combined into a smaller number of components and '~ units', or further separated into additional components and '~ units'. In addition, the components and '~parts' may be implemented to play one or more central processing units (CPUs) within the device or secure multimedia card. In addition, in the embodiment, the '~part' may include one or more processors. Hereinafter, a / b may be understood as at least one of a or b.

[0043] The term 'AI function' used in this embodiment may be understood as a function that performs a specific function in each layer of a wireless communication system using AI technology. For example, the AI ​​function may be a function used in the channel encoding / scrambling process in the physical layer (PHY) below the media access control (MAC) layer. In addition, the AI ​​function may be applied to layers such as low-PHY, high-PHY, low-MAC, and high-MAC. For example, the AI ​​function may mean functions such as channel state information compression (CSI compression), CSI estimation (prediction), AI noise canceling (AI-NC), and AI multiple-input and multiple-output (AI-MIMO).

[0044] The term 'state of AI function' used in the present embodiment may indicate a state of the AI ​​modeling process when performing the AI ​​function. For example, the AI ​​modeling process may include a data collection process for collecting data from a terminal or a RU (radio unit), an AI model training process for learning the collected data, an AI model inference process for inputting the collected data or new data into the learned AI model to derive a result, an AI model performance feedback process for feeding back the performance of the AI ​​model using the result value obtained through the AI ​​model inference process, an AI model upgrade process for upgrading (updating) the AI ​​through feedback on the AI ​​model performance or additional learning, and an operation process for performing a specific action by receiving a result produced through AI model inference. The state of the AI ​​function may mean the state of any one of the processes above.

[0045] Furthermore, while various embodiments of the present disclosure are described below using LTE, LTE-A, NR, or 6G-based systems as examples, the various embodiments of the present disclosure may also be applied to other communication systems with similar technical backgrounds or channel types. Furthermore, the various embodiments of the present disclosure may be applied to other communication systems with some modifications within the scope of the present disclosure, as determined by a person skilled in the art.

[0046] For the convenience of the following description, some terms and names defined in the 3rd generation partnership project (3GPP) standards (standards for 5G, NR, LTE, or similar systems) may be used. In addition, terms and names newly defined in next-generation communication systems (e.g., 6G, Beyond 5G systems) to which the present disclosure may be applied, or terms and names used in existing communication systems may be used. The use of such terms is not limited to the terms and names of the present disclosure, and may be equally applied to systems conforming to other standards, and may be modified into other forms without departing from the technical spirit of the present disclosure. Embodiments of the present disclosure may be easily modified and applied to other communication systems.

[0047] Wireless communication technology has made significant advancements in recent years, offering faster data rates, improved coverage, and more stable connections. However, as demand for wireless communication services continues to grow, many challenges remain, including network congestion, signal interference, and dynamic changes in network conditions.

[0048] Prior art solutions have attempted to address some of these issues through a variety of techniques, including adaptive modulation, channel estimation, and power control, which use mathematical models and algorithms to adjust wireless communication parameters based on network conditions and user requirements.

[0049] Recently, artificial intelligence (AI) technology has been applied to wireless communication systems to improve network performance and user experience. Some prior art solutions use machine learning algorithms to predict network traffic and optimize resource allocation, while others use reinforcement learning to dynamically adjust communication parameters.

[0050] Artificial intelligence technology enables efficient operation in wireless communication environments without complex signal transmission and reception. S-AI communication is one such method. It involves base stations and terminals using the same S-AI to derive the same output from the same input, eliminating the overhead associated with signal transmission.

[0051] Figure 1 is a diagram illustrating a wireless communication system including S-AI.

[0052] Referring to Fig. 1, at 110, S-AI collects input values ​​from each terminal and performs S-AI Training.

[0053] In 120, the S-AI can derive an output value for each terminal from the base station using the input value for each terminal, and the terminal and the base station can obtain the same output value for each terminal.

[0054] In this case, S-AI only shares the same input values, and the output values ​​can be derived individually by the base station. Since the base station and terminals can each obtain their own input values, they can share the same output values, eliminating the need for a separate signaling process for output value sharing. This reduces overhead.

[0055] Figure 2 is a diagram illustrating the process of updating S-AI.

[0056] Referring to FIG. 2, at step 210, terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0057] Afterwards, if a change (or update) of the S-AI is required, the base station can change (or update) the S-AI to a new S-AI (e.g., S-AI-2) through the System Information Block (SIB).

[0058] At step 220, the base station can transmit the changed S-AI to the terminals within the cell via SIB. At this time, the SIB may be an existing SIB or a newly defined SIB-AI.

[0059] Here, a situation requiring an update of the S-AI may refer to a situation in which a network environment has changed. Changes in the network environment may include changes in time (e.g., day to night, night to day, weekday to weekend, weekend to weekday, etc.), changes in the number of terminals connected to the cell, changes in the location of the terminals (e.g., indoor / outdoor), and changes in the amount of traffic requested by the terminals. The changes in time, number of terminals, and amount of traffic may be based on thresholds set by the network or predetermined thresholds.

[0060] Therefore, at step 230, the terminal can change (or update) the S-AI to the received S-AI (e.g., S-AI-2), and the base station and the terminal can communicate using the changed S-AI.

[0061] Here, since it is more advantageous to update all terminals within a base station simultaneously via SIB rather than updating S-AI on a per-terminal basis, this disclosure proposes a method utilizing SIB. However, the scope of the present disclosure is not limited to this, and other messages may be used in addition to SIB.

[0062] Since the cell-specific / site-specific S-AI is shared by the base station and all terminals connected to the base station, S-AI updates must be performed throughout the cell when the network environment changes. In other words, even if the environment learned by the S-AI and the current network environment change, it is difficult to secure network performance and Quality of Experience (QoE) by continuing to use the existing S-AI. Therefore, the existing S-AI needs to be changed to an S-AI suitable for the changed environment. However, delivering the S-AI to each individual terminal connected to the cell through the AI ​​delivery and change procedure can cause signaling overhead.

[0063] The present disclosure comprises a procedure for maintaining synchronization of S-AI when a base station and a terminal connected to the base station use S-AI, and a procedure for updating the S-AI of terminals connected to the base station when a network environment change occurs. The present disclosure assumes that the base station and the terminals connected to the base station all share the same S-AI, and it is assumed that the terminals connected to the base station share a specific S-AI (e.g., S-AI-1). However, the embodiments of the present disclosure are not limited thereto, and a process in which the terminal receives S-AI settings from the base station may be included.

[0064] Accordingly, the present disclosure includes a procedure for determining whether an S-AI is changed in a situation where a specific S-AI is shared, and a procedure for instructing terminals to change to a new S-AI (e.g., S-AI-2) when a base station decides to change from an existing S-AI (due to a change in network environment, etc.).

[0065] FIG. 3a is a diagram illustrating a procedure for maintaining synchronization for S-AI according to one embodiment of the present disclosure.

[0066] Referring to FIG. 3a, at step 330a, the terminal (310) can receive a SIB (e.g., SIB 1) from the base station (320). The terminal can perform decoding on the received SIB.

[0067] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0068] Scheduling information for a new SIB may include at least one of periodicity and window size (SI-window size).

[0069] At step 340a, the terminal can use the information contained in the SIB to determine the S-AI's functionality, whether it is supported within the base station (or network), and the ID of the activated S-AI. Furthermore, the terminal can determine whether the ID of the S-AI set in the terminal is identical to the ID of the S-AI received from the base station.

[0070] As a result of the above verification, if the synchronization of S-AI matches (i.e., if the ID of S-AI set in the terminal and the ID of S-AI received from the base station are the same), the terminal can maintain synchronization in step 350a.

[0071] To determine whether a subsequent change (or update) procedure is required, information on whether the S-AI has been changed (or updated) may be included in the SIB. If the information on the S-AI change (or update) included in the SIB indicates a change (or update) of the S-AI (e.g., if the information is set to 1), the terminal may decode the new SIB (SIB-AI) to perform the S-AI change (or update).

[0072] Specifically, the S-AI information to be changed (or updated) may be included in the SIB, and the terminal may change (or update) the S-AI based on the received S-AI information.

[0073] Additionally, when performing an S-AI change (or update) via SIB-AI, the terminal performs SIB-AI decoding based on the scheduling information for the SIB-AI contained in the SIB. Therefore, the terminal can change (or update) the S-AI based on the S-AI information contained in the SIB-AI. Specific details will be described later.

[0074] FIG. 3b is a diagram illustrating a procedure for maintaining synchronization with S-AI by defining a new SIB according to one embodiment of the present disclosure.

[0075] Referring to FIG. 3b, at step 330b, the terminal (310) can receive a SIB (e.g., SIB 1) from the base station (320).

[0076] Here, a separate SIB (SIB-AI) for S-AI can be defined. The terminal and base station can maintain synchronization of the S-AI's functions and ID based on the SIB-AI, and can perform updates if synchronization is out of sync. When performing synchronization and updates of S-AI using the SIB-AI, scheduling information for the SIB-AI can be included within the SIB.

[0077] In step 340b, the terminal may receive a SIB-AI from the base station. The terminal may receive the SIB-AI based on scheduling information for the SIB-AI included in the SIB.

[0078] Here, the SIB-AI may include at least one of the following information: the function of the S-AI, whether it is supported within the base station (or network), the ID of the active S-AI, and whether it has been changed (or updated).

[0079] At step 350b, the terminal can use the information contained in the SIB-AI to determine the S-AI's functionality, whether it is supported within the base station (or network), and the ID of the activated S-AI. Furthermore, the terminal can determine whether the S-AI ID set in the terminal matches the S-AI ID received from the base station.

[0080] If the above verification result shows that the S-AI synchronization matches (i.e., the S-AI ID set in the terminal and the ID of the S-AI received from the base station are the same), the terminal can maintain synchronization in step 360b.

[0081] Here, in order to determine whether a change (or update) procedure of the S-AI is required in the future, information on whether the S-AI has been changed (or updated) may be included in the SIB-AI. If the information on the S-AI change (or update) included in the SIB-AI indicates a change (or update) of the S-AI (e.g., if the information is set to 1 or if the information is included in the SIB-AI), the terminal may decode the new SIB (SIB-AI) to perform the S-AI change (or update).

[0082] In an environment where S-AI is shared by a base station and terminals connected to the base station, the procedure when the base station decides to change to a new S-AI is described below.

[0083] FIG. 4a is a diagram illustrating a procedure for changing (or updating) S-AI according to one embodiment of the present disclosure.

[0084] According to the present embodiment, a procedure for changing an S-AI using only an SIB is disclosed. According to the present embodiment, it is difficult for a base station to check whether an SIB has been received from a terminal, whether the configuration for an S-AI has been completed, and the minimum requirements for installing and inferring an S-AI (e.g., available hardware requirements of the terminal, time taken for inference, whether input values ​​are collected, and whether the output times between the terminal and the base station match (time constraint)). S-AI inference information may be included in the SIB, and the terminal may operate the changed S-AI when a specified time has elapsed.

[0085] In Fig. 4a, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0086] Referring to FIG. 4a, at step 430a, if a change occurs in the network environment in which S-AI (e.g., S-AI-1) is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure.

[0087] At step 440a, the terminal (410) can receive a SIB (e.g., SIB 1) from the base station (420). The terminal can perform decoding on the received SIB.

[0088] The SIB may include at least one of scheduling information for a new SIB (SIB-AI), including information that there is an S-AI to be changed (or updated) for terminals connected to the cell (or information indicating toggle of S-AI change (or update)), confirmation information for the S-AI to be changed (or updated) (e.g., information indicating the function and ID of the S-AI to be changed (or updated)). The terminal may confirm synchronization based on the information included in the SIB. The scheduling information for the new SIB may include at least one of periodicity information and window size (SI-window size).

[0089] At step 450a, the terminal that decoded the SIB can confirm that the S-AI it was using has been changed (or updated). Accordingly, the terminal can check the scheduling information of the SIB-AI to receive the new S-AI.

[0090] In step 460a, the terminal may receive a SIB-AI for S-AI change (or update) from the base station. The terminal may receive the SIB-AI based on the scheduling information in the SIB-AI. The terminal may decode the received SIB-AI.

[0091] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0092] When the S-AI is pre-configured, the base station can include in the SIB-AI information indicating the function and ID of the SIB-AI to be changed (or updated). The SIB-AI can include information for applying the S-AI after setting a new S-AI (hereinafter, information for S-AI inference), and the information for the S-AI inference can include at least one of information related to the start time of the S-AI inference and information on an initial input value for starting a new S-AI. Here, the information related to the start time of the S-AI inference can include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a start point for applying the offset can be further included.

[0093] If the S-AI is not pre-configured, the base station can decide whether to change (or update) the entire S-AI or change (or update) a part of the S-AI. If the entire S-AI is changed (or updated), the base station can transmit to the terminal an S-AI information message including at least one of a layer, a dimension, and a detailed parameter value. If the part of the S-AI is changed (or updated), the base station can transmit to the terminal an S-AI information message including information about a part that is different from the previously used S-AI (e.g., at least one of a parameter value, a function, and an ID).

[0094] Here, SIB-AI may include information related to S-AI inference. The information related to S-AI inference may include at least one of information related to the start time of S-AI inference and an initial input value for starting a new S-AI. Here, the information related to the start time of S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding a starting point for applying the offset may be further included.

[0095] In step 470a, the terminal can perform an update by performing deployment with a new S-AI (e.g., S-AI-2). The terminal that has completed S-AI deployment can change the ID of the S-AI and then start S-AI inference according to information related to S-AI inference. In the present disclosure, the term “deployment” may be used to mean that the base station has transmitted the S-AI to the terminal, that the terminal has downloaded the S-AI from the base station, or that a new (or changed) S-AI has been applied or set to the terminal.

[0096] According to this embodiment, a procedure for changing S-AI using SIB-AI is initiated. According to this embodiment, it is possible to check whether the configuration for S-AI is complete from the terminal, and the minimum requirements (e.g., available hardware requirements of the terminal, time taken for inference, whether input values ​​are collected, whether the output times between the terminal and the base station match (time constraint)), etc. As a result, the terminal can more dynamically adjust the application time of the changed (or updated) S-AI.

[0097] FIG. 4b is a diagram illustrating a procedure for updating S-AI according to one embodiment of the present disclosure.

[0098] FIG. 4b may include a result report message for changes in S-AI as well as SIB-AI (hereinafter, S-AI deploy result report message) and an information message for applying the changed S-AI (hereinafter, S-AI inference information message).

[0099] In FIG. 4b, the base station (420) and the terminals (410) connected to the base station can use the same S-AI (e.g., S-AI-1).

[0100] Referring to FIG. 4b, if a change occurs in the network environment in which S-AI is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure at step 425b.

[0101] At step 430b, the terminal can receive a SIB (e.g., SIB1) from the base station. The terminal can perform decoding on the received SIB.

[0102] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0103] Scheduling information for a new SIB may include at least one of periodicity information, SI-window size, and resource information (e.g., slot information, frequency information) to report the results of testing (e.g., whether the S-AI model learned at the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured at the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether the S-AI inference is completed within a specified time).

[0104] Using the information contained in the SIB, the terminal can verify the S-AI's functionality, whether it is supported within the base station (or network), and the ID of the activated S-AI. Furthermore, the terminal can verify whether the ID of the S-AI it was using is identical to the ID of the S-AI received from the base station.

[0105] As a result of the above verification, if the synchronization of the S-AI does not match (i.e., if the ID of the S-AI set in the terminal is different from the ID of the S-AI received from the base station), in step 435b, the terminal that decoded the SIB can determine that the S-AI being used must be changed (or updated). Accordingly, the terminal can check the scheduling information of the SIB-AI to receive a new S-AI.

[0106] At step 440b, the terminal can receive a SIB-AI from the base station. The terminal can receive the SIB-AI based on the scheduling information in the SIB-AI. The terminal can decode the received SIB-AI.

[0107] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0108] When the S-AI is pre-configured, the base station can include information indicating the function and ID of the SIB-AI to be changed (or updated) in the SIB-AI. The SIB-AI can include information for applying the S-AI after setting a new S-AI (hereinafter, information for S-AI inference). The information for the S-AI inference can include at least one of information related to the start time of the S-AI inference and information on the initial input value for starting the new S-AI. Here, the information related to the start time of the S-AI inference can include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a start point for applying the offset can be further included.

[0109] Additionally, the SIB-AI may include information for setting up a new S-AI and then receiving the results of the S-AI's setting. Information for receiving the result for the above S-AI setting may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test result (e.g., whether the S-AI model learned by the base station was fully downloaded to the terminal, whether the performance instructed by the base station was secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirement (e.g., available hardware requirement of the terminal, time taken for inference, whether input value is collected, whether the output time between the terminal and the base station matches (time constraint)), whether the minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output value according to a specified input value)), input value used for the test, output value to be output after the test, resource information or channel information for reporting the S-AI test result, and window information (timer 2) for receiving S-AI inference information.

[0110] If the S-AI is not pre-configured, the base station can decide whether to change (or update) the entire S-AI or change (or update) a part of the S-AI. If the entire S-AI is changed (or updated), the base station can transmit to the terminal an S-AI information message including at least one of a layer, a dimension, and a detailed parameter value. If the part of the S-AI is changed (or updated), the base station can transmit to the terminal an S-AI information message including information about a part that is different from the previously used S-AI (e.g., at least one of a parameter value, a function, and an ID).

[0111] Here, the SIB-AI may include information related to S-AI inference. The information related to the S-AI inference may include at least one of information related to the start time of the S-AI inference and an initial input value for starting a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a starting point for applying the offset may be further included. In addition, the SIB-AI may include information for receiving a result for setting the S-AI after setting a new S-AI.

[0112] In addition, information for receiving the result for the above S-AI setting may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test results (e.g., whether the S-AI model learned by the base station was fully downloaded to the terminal, whether the performance instructed by the base station was secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirement (e.g., at least one of a predefined set (e.g., an output value according to a specified input value) such as the available hardware requirement of the terminal, the time taken for inference, whether input values ​​are collected, and whether the output time between the terminal and the base station matches (time constraint)), whether the minimum requirement is satisfied, computing time report, etc.), input values ​​used for the test, output values ​​to be output after the test, resource information or channel information for reporting the S-AI test result, and window information (timer 2) for receiving S-AI inference information.

[0113] At step 450b, the terminal can perform a change (or update) by deploying a new S-AI (e.g., S-AI-2). A terminal that has completed S-AI deployment can change (or update) the S-AI ID.

[0114] In step 455b, the terminal can set up a new S-AI and perform testing on the newly set S-AI (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after changing (or updating) the S-AI of the terminal, and whether the S-AI inference is completed within a specified time).

[0115] Results for testing can be reported using either the resource information (slot information) included in the SIB or separate channel information.

[0116] At step 460b, the terminal can send an S-AI deploy result report message to the base station.

[0117] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through the PUCCH (physical uplink control channel).

[0118] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0119] Afterwards, the terminal can start timer 2 (470b). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0120] Meanwhile, the base station can transmit SIB-AI to the terminal and start timer 1 (445b).

[0121] The base station can receive the S-AI deploy result report message during timer 1.

[0122] The base station, which has received the S-AI deploy result report message, checks the S-AI testing result transmitted by the terminal in step 465b and determines the S-AI inference based on the result.

[0123] The base station, which has received the S-AI deployment result report message, transmits an S-AI inference information message to the terminal that transmitted the S-AI deployment result report message in step 475b. The terminal that has received the S-AI inference message can perform S-AI inference based on information related to S-AI inference.

[0124] At this time, the value of timer 1 may be preset. The base station receives an S-AI deploy result report message while timer 1 is running, and can know the number of terminals that transmitted the S-AI deploy result report message during timer 1 and the ratio of the number of terminals to the total number of terminals connected to the base station.

[0125] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0126] A terminal that receives an S-AI inference information message within Timer 2 can change the ID of the S-AI and perform S-AI inference according to the S-AI inference information.

[0127] If the S-AI synchronization between the base station and the terminal is out of sync, the base station can compensate by re-synchronizing with the terminal. Cases where the S-AI synchronization is out of sync include the terminal not receiving the SIB itself for change (or update), the base station not receiving the S-AI deployment result report message for a specific terminal during timer 1, and the terminal not receiving the S-AI inference information message from the base station during timer 2.

[0128] FIG. 5A is a diagram illustrating an S-AI synchronization procedure according to an embodiment of the present disclosure. In FIG. 5A, a terminal may perform a procedure for maintaining synchronization between the terminal and a base station for S-AI according to the procedure disclosed in FIG. 3A or 3B.

[0129] In this embodiment, if the ID of the S-AI used by the terminal and the ID of the S-AI received via the SIB or SIB-AI are different, it is determined that the S-AI synchronization between the base station and the terminal is broken. In this embodiment, if synchronization is broken, a synchronization supplementation procedure is initiated to re-establish synchronization.

[0130] Synchronization may be broken, for example, when all SIB-AIs for S-AI change (or update) fail to be received, when the terminal is idle or inactive, or when a new connection is made through handover from another cell.

[0131] In FIG. 5a, the base station (520) and the terminals (510) connected to the base station can use the same S-AI (e.g., S-AI-1).

[0132] Referring to FIG. 5a, at step 525a, if there is a change in the network environment in which the S-AI was being used or at the discretion of the base station, the base station may decide to perform a change (or update) procedure to a new S-AI (e.g., S-AI-2).

[0133] At step 530a, the terminal may receive a SIB (e.g., SIB 1) from the base station. The terminal may perform decoding on the received SIB.

[0134] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0135] Specifically, at step 540a, the terminal that has decoded the SIB can check the function and ID of the S-AI being used and confirm that the ID of the S-AI being used is different from the ID of the S-AI to be changed (or updated) included in the SIB received from the base station.

[0136] At step 550a, the terminal deploys the S-AI to be changed (or updated).

[0137] At this time, if the S-AI to be changed (or updated) is pre-configured, the terminal requires information to apply the S-AI.

[0138] In step 560a, the terminal transmits to the base station an S-AI inference information request message (hereinafter, S-AI inference information request message) related to the S-AI to be changed.

[0139] The S-AI inference information request message may include an S-AI deploy result report.

[0140] The S-AI inference information request message may include the function and ID of the S-AI for which inference information is required.

[0141] In step 570a, the base station can transmit an S-AI inference information message to the terminal that transmitted the S-AI inference information request message.

[0142] The terminal can apply a changed S-AI (e.g., S-AI-2) based on the above S-AI inference information.

[0143] FIG. 5b is a diagram illustrating an S-AI synchronization procedure according to one embodiment of the present disclosure.

[0144] In FIG. 5b, the terminal can perform a procedure for maintaining synchronization for S-AI between the terminal and the base station according to the procedure disclosed in FIG. 3a or 3b.

[0145] In this embodiment, if the ID of the S-AI used by the terminal and the ID of the S-AI received via the SIB or SIB-AI are different, it is determined that the S-AI synchronization between the base station and the terminal is broken. In this embodiment, if synchronization is broken, a synchronization supplementation procedure is initiated to re-establish synchronization.

[0146] Synchronization may be broken in one of the following cases: failure to receive all SIB-AIs for S-AI change (or update), the UE being idle, the UE being inactive, or a new connection being made through handover from another cell.

[0147] In FIG. 5b, the base station (520) and the terminals (510) connected to the base station can use the same S-AI (e.g., S-AI-1).

[0148] Referring to FIG. 5b, at step 525b, if there is a change in the network environment in which the S-AI was being used or at the discretion of the base station, the base station may decide to perform a change (or update) procedure to a new S-AI (e.g., S-AI-2).

[0149] At step 530b, the terminal may receive a SIB (e.g., SIB 1) from the base station. The terminal may perform decoding on the received SIB.

[0150] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0151] Specifically, at step 540b, the terminal that decoded the SIB can check the function and ID of the S-AI being used and confirm that the ID of the S-AI being used is different from the ID of the S-AI to be changed (or updated) included in the received SIB.

[0152] At this time, if the S-AI to be changed (or updated) is not pre-configured, in step 545b, the terminal may transmit an S-AI information request message to the base station.

[0153] The S-AI information request message may include at least one of the ID of the S-AI received via SIB, the ID of the S-AI currently in use, and the AI ​​capability of the terminal.

[0154] In step 550b, the base station may transmit an S-AI information message to the terminal that transmitted the S-AI information request message. The S-AI information message may include at least one of the layer, dimension, and detailed parameter values ​​of the changed S-AI model.

[0155] The S-AI information message may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test results (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance instructed by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirements (e.g., available hardware requirements of the terminal, time taken for inference, whether input values ​​are collected, whether the output times between the terminal and the base station match (time constraint)), whether the minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output values ​​according to specified input values)), input values ​​used in the test, output values ​​to be output after the test, resource information or channel information to report the S-AI test results, and window information (Timer 2) for receiving S-AI inference information.

[0156] The base station can decide whether to change (or update) the entire S-AI or change (or update) a part of the S-AI. If changing (or updating) the entire S-AI, the base station can transmit an S-AI information message including at least one of a layer, dimension, and detailed parameter value. If changing (or updating) a part of the S-AI, the base station can transmit an S-AI information message including information about a part that is different from the previously used S-AI (e.g., at least one of a parameter value, a function, and an ID).

[0157] At step 555b, the terminal can perform a change (or update) by performing a deployment with a new S-AI (e.g., S-AI-2).

[0158] Specifically, a terminal that has received an S-AI information message can deploy a new S-AI and perform testing (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after the S-AI of the terminal is changed (or updated), and whether S-AI inference is completed within a specified time) as in the procedure in 4B above. The result of testing can be reported using either resource information (slot information) included in the SIB for reporting the result of testing or separate channel information.

[0159] At step 560b, the terminal can send an S-AI deploy result report message to the base station.

[0160] Here, the S-AI deploy result report message may include at least one of whether the S-AI deploy is successful, the test result (the S-AI deploy result report message may include at least one of whether the deploy is successful and the test result (for example, whether the S-AI deploy is successful, the test output value according to the input value, whether the test output value is satisfied, the computing time during the test, the minimum requirement such as the computing time (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through a PUCC (physical uplink control channel).

[0161] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0162] Afterwards, the terminal can start timer 2 (565b). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0163] Meanwhile, the base station transmits an S-AI information message to the terminal in step 550b. The base station that transmitted the S-AI information message to the terminal can start timer 3 (535b).

[0164] The base station can receive the S-AI deploy result report message during timer 3. The base station that received the S-AI deploy result report message transmits an S-AI inference information message to the terminal that transmitted the S-AI deploy result report message in step 570b.

[0165] At this time, the value of timer 3 may be preset. The base station can receive an S-AI deploy result report message while timer 3 is running.

[0166] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0167] A terminal that receives an S-AI inference information message within Timer 2 can change the ID of the S-AI and perform S-AI inference according to the S-AI inference information.

[0168] FIG. 6 is a diagram illustrating a procedure when acquisition of system information (SI) for a new S-AI fails according to an embodiment of the present disclosure.

[0169] In FIG. 6, the base station can receive an S-AI deploy result report message during timer 1 according to the procedure disclosed in FIG. 4a or FIG. 4b.

[0170] In this embodiment, the base station calculates the reception ratio of the terminals that received the S-AI deploy result report message during timer 1 and the total terminals, and initiates additional procedures based on the calculation result.

[0171] In FIG. 6, the terminal can receive information including information about S-AI via SIB or SIB-AI. In addition, it can include an S-AI deployment result report message and an S-AI inference information message in addition to the SIB.

[0172] In FIG. 6, the base station (620) and the terminals (610) connected to the base station can use the same S-AI (e.g., S-AI-1).

[0173] Referring to FIG. 6, if a change occurs in the network environment in which S-AI is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure at step 625.

[0174] At step 630, the terminal can receive a SIB-AI from the base station. The terminal can receive the SIB-AI based on the scheduling information in the SIB-AI. The terminal can decode the received SIB-AI.

[0175] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0176] At step 640, the terminal can perform a change (or update) by performing deployment to a new S-AI using the information included in the received SIB-AI.

[0177] In step 645, the terminal sets up a new S-AI and can perform testing on the newly set S-AI (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after the S-AI of the terminal is changed (or updated), and whether the S-AI inference is completed within a specified time).

[0178] Results for testing can be reported using either the resource information (slot information) included in the SIB or separate channel information.

[0179] At step 650, the terminal can send an S-AI deploy result report message to the base station.

[0180] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through the PUCCH (physical uplink control channel).

[0181] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0182] Afterwards, the terminal can start timer 2 (655). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0183] Meanwhile, the base station can transmit SIB-AI to the terminal and start timer 1 (635).

[0184] At this time, the value of timer 1 may be preset. The base station receives an S-AI deploy result report message while timer 1 is running, and can know the number of terminals that transmitted the S-AI deploy result report message during timer 1 and the ratio of the number of terminals to the total number of terminals connected to the base station.

[0185] If the base station does not receive a certain percentage of S-AI deployment result report messages from the terminal, in step 660, the base station may determine that sufficient new S-AIs have not been deployed, or that it is difficult to secure sufficient network performance even if inference of the new S-AI is performed.

[0186] At step 665, the base station may transmit a timer 2 extension message to the terminal that transmitted the S-AI deploy result report message to extend Timer 2.

[0187] The base station may modify scheduling information within the SIB or extend timer 1 to allow longer transmission of the SIB-AI for update while transmitting the above timer 2 extension message.

[0188] A terminal that receives a Timer 2 extension message can extend or renew Timer 2.

[0189] The Timer 2 extension message may be sent as a separate message or included in the S-AI inference message.

[0190] If the base station receives a certain percentage of S-AI deploy result report messages from the terminal during timer 1 (or extended timer 1), then at step 670, the base station can determine that the new S-AI has been sufficiently deployed or that there will be no problem with the inference of the new S-AI.

[0191] As above, if the base station receives a certain percentage or more of S-AI deploy result report messages from the terminal during timer 1, the base station can transmit an S-AI inference information message to the terminal that transmitted the S-AI deploy result report in step 675 to synchronize the start of the next S-AI inference.

[0192] At this time, the base station can perform a supplementary procedure for S-AI through the procedure according to Drawing 5A or 5B for a terminal that has not transmitted an S-AI deploy result report message.

[0193] In addition, if the terminal does not receive a timer 2 extension message or an S-AI inference information message from the base station during timer 2 (or during extended timer 2) even though the terminal has transmitted an S-AI deploy result report message to the base station, the base station may perform a supplementary procedure for S-AI for the terminal according to the procedure of FIG. 5a or 5b.

[0194] FIG. 7a is a diagram illustrating a procedure for a terminal to update S-AI according to one embodiment of the present disclosure.

[0195] According to the present embodiment, a procedure for changing the S-AI using only the SIB is initiated. According to the present embodiment, it is difficult for the base station to check whether the SIB has been received from the terminal, whether the configuration for the S-AI has been completed, and whether the minimum requirements (e.g., the available hardware requirements of the terminal, the time taken for inference, whether input values ​​are collected, and whether the output times between the terminal and the base station match (time constraint)) are the same. S-AI inference information may be included in the SIB, and the terminal may operate the changed S-AI when a specified time has passed.

[0196] In Fig. 7a, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0197] Referring to FIG. 7a, at step 710a, the terminal may receive a SIB (e.g., SIB 1) from the base station. The terminal may perform decoding on the received SIB.

[0198] The information included in the SIB may include at least one of scheduling information for a new SIB (SIB-AI), including information that there is an S-AI to be changed (or updated), identification information for the S-AI to be changed (or updated), and information indicating the function and ID of the S-AI to be changed (or updated).

[0199] Scheduling information for a new SIB may include at least one of periodicity and window size (SI-window size).

[0200] At step 720a, the terminal that decoded the SIB can determine that the S-AI it was using has changed (or been updated). Therefore, the terminal can check the scheduling information of the SIB-AI to receive the new S-AI.

[0201] In step 730a, the terminal may receive a SIB-AI from the base station. The terminal may receive the SIB-AI based on the scheduling information in the SIB-AI. The terminal may decode the received SIB-AI.

[0202] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0203] At step 740a, the terminal can perform an update by deploying a new S-AI (e.g., S-AI-2). After completing S-AI deployment, the terminal can change the S-AI ID and then initiate S-AI inference based on information related to S-AI inference.

[0204] According to this embodiment, a procedure for changing S-AI using SIB is disclosed. According to this embodiment, it is possible to check whether the configuration for S-AI is complete from the terminal, and the minimum requirements (e.g., the terminal's available hardware requirements, the time required for inference, whether input values ​​are collected, whether the output times between the terminal and the base station match (time constraint)), etc. As a result, the terminal can more dynamically adjust the application time of the changed (or updated) S-AI.

[0205] FIG. 7b is a diagram illustrating a procedure for updating S-AI according to one embodiment of the present disclosure.

[0206] Figure 7b may include not only SIB-AI but also S-AI deploy result report message and S-AI inference information message.

[0207] In Fig. 7b, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0208] Referring to FIG. 7b, at step 710b, the terminal may receive a SIB (e.g., SIB1) from the base station. The terminal may perform decoding on the received SIB.

[0209] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0210] Scheduling information for a new SIB may include at least one of periodicity and window size (SI-window size).

[0211] Using the information contained in the SIB, the terminal can verify the S-AI's functionality, whether it is supported within the base station (or network), and the ID of the activated S-AI. Furthermore, the terminal can verify whether the ID of the S-AI it was using is identical to the ID of the S-AI received from the base station.

[0212] As a result of the above verification, if the synchronization of the S-AI does not match (i.e., the ID of the S-AI set in the terminal and the ID of the S-AI received from the base station are different), in step 720b, the terminal that decoded the SIB can confirm that the S-AI it was using has been changed (or updated). Accordingly, the terminal can check the scheduling information of the SIB-AI to receive a new S-AI.

[0213] At step 730b, the terminal can receive a SIB-AI from the base station. The terminal can receive the SIB-AI based on the scheduling information in the SIB-AI. The terminal can decode the received SIB-AI.

[0214] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0215] At step 740b, the terminal can perform a change (or update) by deploying a new S-AI (e.g., S-AI-2). After completing S-AI deployment, the terminal can change the S-AI ID and then perform S-AI inference based on information related to S-AI inference.

[0216] At step 750b, the terminal can set up a new S-AI and perform testing on the newly set S-AI (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after changing (or updating) the S-AI of the terminal, and whether S-AI inference is completed within a specified time).

[0217] The results for testing are reported using either the resource information (slot information) included in the SIB or separate channel information.

[0218] At step 760b, the terminal can send an S-AI deploy result report message to the base station.

[0219] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through the PUCCH (physical uplink control channel).

[0220] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0221] Afterwards, the terminal can start timer 2 (770b). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0222] In step 780b, the terminal may receive an S-AI inference information message from the base station. The terminal that receives the S-AI inference information message may change the S-AI ID and perform S-AI inference based on the S-AI inference information.

[0223] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0224] FIG. 8A is a diagram illustrating an S-AI synchronization procedure of a terminal according to one embodiment of the present disclosure.

[0225] In FIG. 8a, the terminal can perform a procedure for maintaining synchronization between the S-AI between the terminal and the base station according to the procedure disclosed in FIG. 3a or 3b. In this embodiment, if the ID of the S-AI used by the terminal and the ID of the S-AI received via SIB or SIB-AI are different, it is determined that the synchronization of the S-AI between the base station and the terminal is broken. In this embodiment, if synchronization is broken, a synchronization supplementation procedure is initiated to maintain synchronization again.

[0226] Synchronization may be broken, for example, when all SIB-AIs for S-AI change (or update) fail to be received, when the terminal is idle or inactive, or when a new connection is made through handover from another cell.

[0227] In Fig. 8a, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0228] Referring to FIG. 8a, at step 810a, the terminal may receive a SIB (e.g., SIB 1) from the base station. The terminal may perform decoding on the received SIB.

[0229] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0230] Specifically, at step 820a, the terminal that decoded the SIB can check the function and ID of the S-AI being used and confirm that the ID of the S-AI being used is different from the ID of the S-AI to be changed (or updated) included in the received SIB.

[0231] If the ID is the same, in step 830a, the terminal maintains synchronization with the S-AI it was using.

[0232] However, if the IDs are different, the terminal determines that the synchronization of the S-AI being used is broken. If synchronization is broken, the terminal deploys the S-AI to be changed (or updated) in step 840a.

[0233] At this time, if the S-AI to be changed (or updated) is pre-configured, the terminal requires information for inference of the S-AI.

[0234] In step 850a, the terminal transmits an S-AI inference information request message to the base station.

[0235] The S-AI inference information request message may include an S-AI deploy result report.

[0236] The S-AI inference information request message may include the function and ID of the S-AI for which inference information is required.

[0237] At step 860a, the terminal can receive an S-AI inference information message from the base station.

[0238] The terminal can apply a changed S-AI (e.g., S-AI-2) based on the above S-AI inference information.

[0239] FIG. 8b is a diagram illustrating an S-AI synchronization procedure of a terminal according to an embodiment of the present disclosure.

[0240] In FIG. 8b, the terminal can perform a procedure for maintaining synchronization between the terminal and the base station for the S-AI according to the procedure disclosed in FIG. 3a or 3b. In this embodiment, if the ID of the S-AI used by the terminal and the ID of the S-AI received via the SIB or SIB-AI are different, it is determined that the synchronization of the S-AI between the base station and the terminal is broken. In this embodiment, if synchronization is broken, a synchronization supplementation procedure is initiated to maintain synchronization again.

[0241] Synchronization may be broken, for example, when all SIB-AIs for S-AI change (or update) fail to be received, when the terminal is idle or inactive, or when a new connection is made through handover from another cell.

[0242] In Fig. 8b, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0243] Referring to FIG. 8b, at step 810b, the terminal may receive a SIB (e.g., SIB 1) from the base station. The terminal may perform decoding on the received SIB.

[0244] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0245] Specifically, at step 820b, the terminal that decoded the SIB can check the function and ID of the S-AI being used and confirm that the ID of the S-AI being used is different from the ID of the S-AI to be changed (or updated) included in the received SIB.

[0246] If the ID is the same, at step 830b, the terminal maintains synchronization with the S-AI it was using.

[0247] However, if the IDs are different, the terminal determines that the synchronization of the S-AI being used is broken. If the synchronization is broken and the S-AI to be changed (or updated) is not pre-configured, the terminal may send an S-AI information request message to the base station in step 840b.

[0248] The S-AI information request message may include one of the IDs of the S-AI received via SIB, the ID of the S-AI currently in use, and the AI ​​capability of the terminal.

[0249] At step 850b, the terminal can receive an S-AI inference information message from the base station.

[0250] The S-AI information message may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test results (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance instructed by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirements (e.g., available hardware requirements of the terminal, time taken for inference, whether input values ​​are collected, whether the output times between the terminal and the base station match (time constraint)), whether the minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output values ​​according to specified input values)), input values ​​used in the test, output values ​​to be output after the test, resource information or channel information to report the S-AI test results, and window information (Timer 2) for receiving S-AI inference information.

[0251] At step 860b, the terminal can perform a change (or update) by performing a deployment with a new S-AI (e.g., S-AI-2).

[0252] A terminal that receives an S-AI information message can deploy a new S-AI and perform testing (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after the S-AI of the terminal is changed (or updated), and whether S-AI inference is completed within a specified time) as in the procedure in 7B above. The result of testing uses either the resource information (slot information) included in the SIB for reporting the result of testing or separate channel information.

[0253] At step 870b, the terminal can send an S-AI deploy result report message to the base station.

[0254] Here, the S-AI deploy result report message may include at least one of whether the S-AI deploy is successful, the test result (the S-AI deploy result report message may include at least one of whether the S-AI deploy is successful and the test result (for example, whether the S-AI deploy is successful, the test output value according to the input value, whether the test output value is satisfied, the computing time during the test, the minimum requirement such as the computing time (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through a PUCCH (physical uplink control channel).

[0255] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0256] Afterwards, the terminal can start timer 2 (880b). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0257] In step 890b, the terminal can receive an S-AI inference information message. The terminal that receives the S-AI inference information message can change the S-AI ID and perform S-AI inference according to the S-AI inference information.

[0258] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0259] FIG. 9 is a diagram illustrating a procedure of a terminal when acquisition of system information for a new S-AI fails according to an embodiment of the present disclosure.

[0260] In Fig. 9, the terminal initiates an additional procedure based on the reception ratio of the terminal and the entire terminal that received the S-AI deploy result report message during timer 1 calculated by the base station.

[0261] In FIG. 9, the terminal can receive information including information about S-AI via SIB or SIB-AI. In addition, it can include an S-AI deployment result report message and an S-AI inference information message in addition to the SIB.

[0262] Referring to FIG. 9, at step 910, the terminal may receive a SIB-AI from the base station. The terminal may receive the SIB-AI based on scheduling information in the SIB-AI. The terminal may perform decoding on the received SIB-AI.

[0263] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0264] At step 920, the terminal can perform a change (or update) by performing deployment to a new S-AI using the information included in the received SIB-AI.

[0265] In step 930, the terminal sets up a new S-AI and can perform testing on the newly set S-AI (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance instructed by the base station has been secured in the terminal, whether the output value is output as is after the S-AI of the terminal is changed (or updated), and whether the S-AI inference is completed within a specified time).

[0266] The results for testing are reported using either the resource information (slot information) included in the SIB or separate channel information.

[0267] In step 940, the terminal can send an S-AI deploy result report message to the base station.

[0268] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through the PUCCH (physical uplink control channel).

[0269] The terminal can transmit the result of the S-AI deployment success or failure to the base station using an uplink signal (e.g., Uplink Control Information (UCI)). At this time, the terminal can indicate the success or failure of the S-AI deployment by adding 1 bit to the UCI, or if the S-AI deployment is successful, can transmit the result to the base station by adding 1 bit to the UCI.

[0270] Afterwards, the terminal can start timer 2 (950). Then, the terminal waits to receive an S-AI inference information message while timer 2 is running.

[0271] Meanwhile, if the base station does not receive a certain percentage of S-AI deployment result report messages from the terminal during timer1, the terminal may receive a timer2 extension message from the base station at step 960. The terminal that receives the timer2 extension message extends or updates timer2.

[0272] The Timer 2 extension message may be sent as a separate message or included in the S-AI inference message.

[0273] Meanwhile, if the base station receives a certain percentage or more of S-AI deploy result report messages from the terminal during timer 1 (or extended timer 1), the terminal may receive an S-AI inference information message from the base station at step 970.

[0274] At this time, a terminal that has not sent an S-AI deploy result report message can proceed with a supplementary procedure for S-AI through the procedure according to FIG. 8a or 8b.

[0275] In addition, if the terminal does not receive a timer 2 extension message or an S-AI inference information message from the base station during timer 2 (or during extended timer 2) even though the terminal has transmitted an S-AI deploy result report message to the base station, the base station may perform a supplementary procedure for S-AI for the terminal according to the procedure of FIG. 8a or 8b.

[0276] FIG. 10a is a diagram illustrating a procedure for a base station to update S-AI according to one embodiment of the present disclosure.

[0277] According to this embodiment, a procedure for changing the S-AI using only the SIB is initiated. According to this embodiment, it is difficult for the base station to confirm whether the SIB has been received from the terminal, whether the S-AI configuration has been completed, and whether the minimum requirements (e.g., the available hardware requirements of the terminal, the time required for inference, whether input values ​​are collected, and whether the output times between the terminal and the base station match (time constraint)) are met. S-AI inference information may be included in the SIB.

[0278] In Fig. 10a, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0279] Referring to FIG. 10a, at step 1010a, if a change occurs in the network environment in which S-AI (e.g., S-AI-1) is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure.

[0280] At step 1020, the base station may transmit a SIB (e.g., SIB 1) to the terminal.

[0281] Information included in a SIB may include at least one of: information that there is an S-AI to be changed (or updated); and scheduling information for a new SIB (SIB-AI) including the function and ID of the S-AI to be changed (or updated).

[0282] Scheduling information for a new SIB may include at least one of periodicity and window size (SI-window size).

[0283] At step 1030a, the base station can transmit SIB-AI to the terminal.

[0284] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0285] When the S-AI is pre-configured, the base station can include in the SIB-AI information indicating the function and ID of the SIB-AI to be changed (or updated). The SIB-AI can include information for applying the S-AI after setting a new S-AI (hereinafter, information for S-AI inference), and the information for the S-AI inference can include at least one of information related to the start time of the S-AI inference and information on the initial input value for starting the new S-AI. Here, the information related to the start time of the S-AI inference can include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a start point for applying the offset can be further included.

[0286] If the S-AI is not pre-configured, the base station can decide whether to change (or update) the entire S-AI or change (or update) a part of the S-AI. If changing (or updating) the entire S-AI, the base station can transmit an S-AI information message including at least one of the values ​​of the layer, dimension, and detailed parameters. If changing (or updating) a part of the S-AI, the base station can transmit an S-AI information message including information about a part that is different from the previously used S-AI (e.g., at least one of the parameter values, function, and ID).

[0287] Here, SIB-AI may include information related to S-AI inference. The information related to S-AI inference may include at least one of information related to the start time of S-AI inference and an initial input value for starting a new S-AI. Here, the information related to the start time of S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding a starting point for applying the offset may be further included.

[0288] According to the present embodiment, a procedure for changing an S-AI using SIB-AI is disclosed. According to the present embodiment, it is possible to check whether the configuration for the S-AI has been completed from the terminal and at least one of the minimum requirements (e.g., the available hardware requirement of the terminal, the time taken for inference, whether input values ​​are collected, and whether the output times between the terminal and the base station match (time constraint)). As a result, the terminal can more dynamically adjust the application time of the changed (or updated) S-AI.

[0289] FIG. 10b is a diagram illustrating a procedure for a base station to update S-AI by transmitting an S-AI inference message according to one embodiment of the present disclosure.

[0290] Figure 10b may include not only SIB-AI but also S-AI deploy result report message and S-AI inference information message.

[0291] In Fig. 10b, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0292] Referring to FIG. 10b, if there is a change in the network environment in which S-AI is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure in step 1010b.

[0293] At step 1020b, the base station may transmit a SIB (e.g., SIB 1) to the terminal.

[0294] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0295] Scheduling information for a new SIB may include at least one of periodicity and window size (SI-window size).

[0296] At step 1030b, the base station can transmit SIB-AI to the terminal.

[0297] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about the parameters to be modified.

[0298] When the S-AI is pre-configured, the base station can include information indicating the function and ID of the SIB-AI to be changed (or updated) in the SIB-AI. The SIB-AI can include information for applying the S-AI after setting a new S-AI (hereinafter, information for S-AI inference). The information for the S-AI inference can include at least one of information related to the start time of the S-AI inference and information on the initial input value for starting the new S-AI. Here, the information related to the start time of the S-AI inference can include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a start point for applying the offset can be further included.

[0299] Additionally, the SIB-AI may include information for setting up a new S-AI and then receiving the results of the S-AI's setting. Information for receiving the result for the above S-AI setting may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test result (e.g., whether the S-AI model learned by the base station has been completely downloaded to the terminal, whether the performance instructed by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirement (e.g., available hardware requirement of the terminal, time taken for inference, whether input value is collected, whether the output time between the terminal and the base station matches (time constraint)), whether minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output value according to a specified input value)), input value used for test, output value to be output after test, resource information or channel information for reporting the S-AI test result, and window information (timer 2) for receiving S-AI inference information.

[0300] Here, the SIB-AI may include information related to S-AI inference. The information related to the S-AI inference may include at least one of information related to the start time of the S-AI inference and an initial input value for starting a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information on a starting point for applying the offset may be further included. In addition, the SIB-AI may include information for receiving a result for setting the S-AI after setting a new S-AI.

[0301] In addition, information for receiving the result for the above S-AI setting may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test result (e.g., whether the S-AI model learned by the base station was completely downloaded to the terminal, whether the performance instructed by the base station was secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirement (e.g., available hardware requirement of the terminal, time taken for inference, whether input value is collected, whether the output time between the terminal and the base station matches (time constraint)), whether minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output value according to a specified input value)), input value used for test, output value to be output after test, resource information or channel information for reporting the S-AI test result, window information (timer 2) for receiving S-AI inference information.

[0302] Meanwhile, the base station can transmit SIB-AI to the terminal and start timer 1 (1040b).

[0303] At step 1050b, the base station can receive an S-AI deploy result report message from the terminal.

[0304] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be received through the PUCCH (physical uplink control channel).

[0305] The base station can receive the result of the success or failure of S-AI deployment from the terminal using an uplink signal (e.g., Uplink Control Information (UCI)).

[0306] The base station, which has received the S-AI deploy result report message, checks the S-AI testing result transmitted by the terminal at step 1060b (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether the S-AI inference is completed within a specified time), and determines the S-AI inference based on the result.

[0307] The base station can receive the S-AI deploy result report message during timer 1. The base station that received the S-AI deploy result report message transmits an S-AI inference information message to the terminal that transmitted the S-AI deploy result report message at step 1070b.

[0308] At this time, the value of timer 1 may be preset. The base station receives an S-AI deploy result report message while timer 1 is running, and can know the number of terminals that transmitted the S-AI deploy result report message during timer 1 and the ratio of the number of terminals to the total number of terminals connected to the base station.

[0309] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0310] If the S-AI synchronization between the base station and the terminal is out of sync, the base station can compensate by re-synchronizing with the terminal. Cases where the S-AI synchronization is out of sync include the terminal not receiving the SIB itself for change (or update), the base station not receiving the S-AI deployment result report message for a specific terminal during timer 1, and the terminal not receiving the S-AI inference information message from the base station during timer 2.

[0311] FIG. 11a is a diagram illustrating an S-AI synchronization procedure of a base station according to one embodiment of the present disclosure.

[0312] In Figure 11a, the base station can perform a procedure to maintain synchronization between the terminal and the base station for the S-AI. In this embodiment, if synchronization is lost, a synchronization recovery procedure is initiated to reestablish synchronization.

[0313] Synchronization may be broken, for example, when all SIB-AIs for S-AI change (or update) fail to be received, when the terminal is idle or inactive, or when a new connection is made through handover from another cell.

[0314] In Fig. 11a, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0315] Referring to FIG. 11a, at step 1110a, if there is a change in the network environment in which the S-AI was being used or at the discretion of the base station, the base station may decide to perform a change (or update) procedure to a new S-AI (e.g., S-AI-2).

[0316] At step 1120a, the base station may transmit a SIB (e.g., SIB 1) to the terminal.

[0317] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0318] At step 1130a, the base station can receive an S-AI inference information request message from the terminal.

[0319] The S-AI inference information request message may include an S-AI deploy result report.

[0320] The S-AI inference information request message may include the function and ID of the S-AI for which inference information is required.

[0321] At step 1140a, the base station can transmit an S-AI inference information message to the terminal that transmitted the S-AI inference information request message.

[0322] The terminal can apply a changed S-AI (e.g., S-AI-2) based on the above S-AI inference information.

[0323] FIG. 11b is a diagram illustrating an S-AI synchronization procedure of a base station according to one embodiment of the present disclosure.

[0324] In Figure 11b, a procedure for maintaining synchronization between the terminal and the base station for S-AI can be performed. In this embodiment, if synchronization is lost, a synchronization recovery procedure is initiated to reestablish synchronization.

[0325] Synchronization may be broken, for example, when all SIB-AIs for S-AI change (or update) fail to be received, when the terminal is idle or inactive, or when a new connection is made through handover from another cell.

[0326] In Fig. 11b, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0327] Referring to FIG. 11b, at step 1110b, if there is a change in the network environment in which the S-AI was being used or at the discretion of the base station, the base station may decide to perform a change (or update) procedure to a new S-AI (e.g., S-AI-2).

[0328] At step 1120b, the base station may transmit a SIB (e.g., SIB 1) to the terminal.

[0329] The SIB may include at least one of information regarding the existence of an S-AI that changes (or updates) the S-AI for terminals connected to the cell (e.g., the function and ID of the S-AI), information regarding whether the S-AI has been changed (or updated), and scheduling information for a new SIB (SIB-AI) when the S-AI is changed (or updated) (e.g., resource information for receiving the SIB-AI). The terminal may confirm synchronization based on the information included in the SIB.

[0330] At this time, if the S-AI to be changed (or updated) is not pre-configured, at step 1140b, the base station can receive an S-AI information request message from the terminal.

[0331] The S-AI information request message may include one of the IDs of the S-AI received via SIB, the ID of the S-AI currently in use, and the AI ​​capability of the terminal.

[0332] At step 1150b, the base station can transmit an S-AI inference information message to the terminal that transmitted the S-AI inference information request message.

[0333] The S-AI information message may include at least one of information to be reported after S-AI deployment (e.g., whether S-AI deployment was successful, test results (e.g., whether the S-AI model learned by the base station has been fully downloaded to the terminal, whether the performance instructed by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time), minimum requirements (e.g., available hardware requirements of the terminal, time taken for inference, whether input values ​​are collected, whether the output times between the terminal and the base station match (time constraint)), whether the minimum requirement is satisfied, computing time report, at least one of a predefined set (e.g., output values ​​according to specified input values)), input values ​​used in the test, output values ​​to be output after the test, resource information or channel information to report the S-AI test results, and window information (Timer 2) for receiving S-AI inference information.

[0334] The base station can decide whether to change (or update) the entire S-AI or change (or update) a portion of the S-AI. When changing (or updating) the entire S-AI, an S-AI information message including at least one of the layer, dimension, and detailed parameter values ​​can be transmitted. When changing (or updating) a portion of the S-AI, only the portions that are different from the previously used S-AI (e.g., at least one of the parameter values, function, and ID) can be transmitted.

[0335] At step 1160b, the terminal can send an S-AI deploy result report message to the base station.

[0336] Here, the S-AI deploy result report message may include at least one of whether the S-AI deploy is successful, the test result (the S-AI deploy result report message may include at least one of whether the S-AI deploy is successful and the test result (for example, whether the S-AI deploy is successful, the test output value according to the input value, whether the test output value is satisfied, the computing time during the test, the minimum requirement such as the computing time (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be received through a PUCCH (physical uplink control channel).

[0337] The base station can receive the result of the success or failure of S-AI deployment from the terminal using an uplink signal (e.g., Uplink Control Information (UCI)).

[0338] Meanwhile, the base station can transmit SIB to the terminal and start timer 1 (1130b).

[0339] The base station can receive the S-AI deploy result report message during timer 1. In step 1170b, the base station that received the S-AI deploy result report message transmits an S-AI inference information message to the terminal that transmitted the S-AI deploy result report message.

[0340] At this time, the value of timer 1 may be preset. The base station receives an S-AI deploy result report message while timer 1 is running, and can know the number of terminals that transmitted the S-AI deploy result report message during timer 1 and the ratio of the number of terminals to the total number of terminals connected to the base station.

[0341] The S-AI inference information message may include at least one of information related to the start time of a new S-AI inference and information on the initial input value for the start of a new S-AI. Here, the information related to the start time of the S-AI inference may include at least one of information directly indicating the start time or offset information for indicating the start time. In this case, when an offset is used, information regarding the starting point for applying the offset may be further included.

[0342] A terminal that receives an S-AI inference information message within Timer 2 can change the ID of the S-AI and perform S-AI inference according to the S-AI inference information.

[0343] FIG. 12 is a diagram illustrating a base station procedure when acquisition of system information for a new S-AI fails according to one embodiment of the present disclosure.

[0344] In FIG. 12, the base station can receive an S-AI deploy result report message during timer 1 according to the procedure disclosed in FIG. 10a or FIG. 10b. In this embodiment, the base station calculates the reception ratio of terminals that received the S-AI deploy result report message during timer 1 and all terminals, and initiates an additional procedure based on the calculation result.

[0345] In FIG. 12, the base station can transmit information including information about S-AI via SIB or SIB-AI. In addition, it can include an S-AI deployment result report message and an S-AI inference information message in addition to the SIB.

[0346] In Fig. 12, the base station and the terminals connected to the base station can use the same S-AI (e.g., S-AI-1).

[0347] Referring to FIG. 12, if there is a change in the network environment in which S-AI is being used, or at the discretion of the base station, the base station may decide to perform an S-AI change (or update) procedure at step 1210.

[0348] At step 1220, the base station transmits SIB-AI to the terminal.

[0349] The SIB-AI may include information about an S-AI (e.g., S-AI-2) to be changed (or updated). Here, the information about the S-AI to be changed (or updated) may include information about a parameter to be modified.

[0350] At step 1240, the terminal can send an S-AI deploy result report message to the base station.

[0351] The S-AI deploy result report message may include at least one of the following: whether the S-AI deploy was successful or not, and the test result (for example, whether the S-AI deploy was successful or not, the test output value according to the input value, whether the test output value is satisfied, whether the minimum requirement such as the computing time and computing timer during the test (for example, the available hardware requirement of the terminal, the time taken for inference, whether the input value is collected, and whether the output time between the terminal and the base station matches (time constraint)) is satisfied). The S-AI deploy result report message may be transmitted through the PUCCH (physical uplink control channel).

[0352] The base station can receive the result of the success or failure of S-AI deployment from the terminal using an uplink signal (e.g., Uplink Control Information (UCI)).

[0353] Meanwhile, the base station can transmit SIB-AI to the terminal and start timer 1 (1230).

[0354] At this time, the value of timer 1 may be preset. The base station receives an S-AI deploy result report message while timer 1 is running, and can know the number of terminals that transmitted the S-AI deploy result report message during timer 1 and the ratio of the number of terminals to the total number of terminals connected to the base station.

[0355] If the base station does not receive a certain percentage of S-AI deployment result report messages from the terminal, in step 1250, the base station may determine that the new S-AI has not been sufficiently deployed, or that it is difficult to secure sufficient network performance even if inference of the new S-AI is performed.

[0356] At step 1260, the base station may send a timer 2 extension message to the terminal that sent the S-AI deploy result report message to extend Timer 2.

[0357] The base station may modify scheduling information within the SIB or extend timer 1 to allow longer transmission of the SIB-AI for update while transmitting the above timer 2 extension message.

[0358] The Timer 2 extension message may be sent as a separate message or included in the S-AI inference message.

[0359] If the base station receives a certain percentage of S-AI deploy result report messages from the terminal during timer 1 (or extended timer 1), then at step 1270, the base station can determine that the new S-AI has been sufficiently deployed or that there will be no problem with the inference of the new S-AI.

[0360] As above, if the base station receives a certain percentage or more of S-AI deploy result report messages from the terminal during timer 1, in step 1280, the base station can transmit an S-AI inference information message to the terminal that transmitted the S-AI deploy result report in order to synchronize the start of the next S-AI inference.

[0361] At this time, the base station can perform a supplementary procedure for S-AI through the procedure according to Drawing 11A or 11B for a terminal that has not transmitted an S-AI deploy result report message.

[0362] In addition, if the terminal does not receive a timer 2 extension message or an S-AI inference information message from the base station during timer 2 (or during extended timer 2) even though the terminal has transmitted an S-AI deploy result report message to the base station, the base station may perform a supplementary procedure for S-AI for the terminal according to the procedure of FIG. 5a or 5b.

[0363] FIG. 13 is a diagram illustrating a configuration of a terminal according to an embodiment of the present disclosure.

[0364] Referring to FIG. 13, the terminal may include a transceiver (1310), a control unit (1320), and a storage unit (1330). In the present disclosure, the control unit may be defined as a circuit or an application-specific integrated circuit or at least one processor.

[0365] The transceiver can transmit and receive signals with other network entities. For example, the transceiver can receive system information from a base station and transmit and receive S-AI deployment result report messages, S-AI inference messages, S-AI information messages, and the like with the base station.

[0366] The control unit can control the overall operation of the terminal according to the embodiment proposed in the present disclosure. For example, the control unit can control the signal flow between each block to perform operations according to the flowchart described above. Specifically, the control unit can determine whether to change (or update) the S-AI information received according to the embodiment of the present disclosure, and if a change is determined, can apply the S-AI to be changed based on the information in the received S-AI.

[0367] The storage unit can store at least one of the information transmitted and received through the above-described transceiver unit and the information generated through the control unit. For example, the storage unit can store information related to S-AI deployment result reporting, such as S-AI testing (e.g., whether the S-AI model learned by the base station has been completely downloaded to the terminal, whether the performance indicated by the base station has been secured in the terminal, whether the output value is output as is after the S-AI change (or update) of the terminal, and whether S-AI inference is completed within a specified time).

[0368] FIG. 14 is a diagram illustrating a configuration of a base station according to an embodiment of the present disclosure.

[0369] Referring to FIG. 14, the base station may include a transceiver (1410), a control unit (1420), and a storage unit (1430). In the present disclosure, the control unit may be defined as a circuit or an application-specific integrated circuit or at least one processor.

[0370] The transceiver unit can transmit and receive signals with other network entities. For example, the transceiver unit (1410) can transmit system information to a terminal and determine whether to change (or update) the S-AI.

[0371] The control unit can control the overall operation of the base station according to the embodiments proposed in this disclosure. For example, the control unit (1420) can control the signal flow between each block to perform operations according to the flowchart described above. Specifically, the control unit can control the operations proposed in this disclosure to transmit information necessary for S-AI modification (or update) according to the embodiments of this disclosure.

[0372] The storage unit may store at least one of information transmitted and received through the transceiver unit and information generated through the control unit. For example, the storage unit may store at least one of SIB-AI scheduling information and S-AI inference information related to S-AI change (or update).

Claims

1. A method performed by a terminal in a wireless communication system, A step of receiving a System Information Block (SIB) from a base station, the SIB including scheduling information related to SIB-Artificial Intelligence (SIB-AI); A step for checking changes to the synchronized-AI (S-AI) being used based on the information included in the above SIB; A step of receiving SIB-AI based on the scheduling information from the base station, wherein the SIB-AI includes information about the S-AI to be changed; and A method characterized by comprising: a step of performing a change of the S-AI based on the S-AI information to be changed included in the SIB-AI.

2. In paragraph 1, The above SIB contains information that instructs to toggle the above S-AI change, A method characterized in that the SIB-AI further includes at least one of a function of the S-AI to be changed, or an identifier (ID) of the S-AI to be changed.

3. In paragraph 1, A step of conducting testing on the above-mentioned modified S-AI; A step of transmitting a result report message regarding a change of the S-AI to the base station, wherein the result report message regarding a change of the S-AI includes at least one of the success or failure of the change of the S-AI or the result of the testing; and Further comprising a step of receiving a message (S-AI inference information message) including information for applying the S-AI to be changed from the base station; A method characterized in that the S-AI inference information message includes at least one of information on the start time of application of the S-AI to be changed and information on the initial input value for start.

4. In paragraph 3, The step of receiving a message including information for applying the S-AI to be changed is as follows: A step of transmitting an inference information request message related to the S-AI to be changed to the base station; and A method further comprising: a step of receiving the S-AI inference information message from the base station in response to the inference information request message related to the S-AI.

5. A method performed by a base station in a wireless communication system, A step of deciding to perform a synchronized-Artificial Intelligence (S-AI) change procedure; A step of transmitting a System Information Block (SIB) to a terminal, wherein the SIB includes scheduling information related to SIB-AI; and A step of transmitting SIB-AI to the terminal based on the scheduling information, the SIB-AI including information on the S-AI to be changed; A method characterized in that the change of the S-AI is performed based on the S-AI information to be changed.

6. In paragraph 5, The above SIB includes information that instructs to toggle the change of the above S-AI, A method characterized in that the SIB-AI further includes at least one of a function of the S-AI to be changed, or an identifier (ID) of the S-AI to be changed.

7. In paragraph 5, A step of receiving a result report message regarding a change of the S-AI from the terminal, wherein the result report message regarding the change of the S-AI includes at least one of the success or failure of the change of the S-AI or the result of testing performed by the terminal; A step for determining information for applying the S-AI to be changed based on a result report message regarding the change of the S-AI received from the terminal; and It further includes a step of transmitting a message (S-AI inference information message) including information for applying the S-AI to be changed to the terminal; A method characterized in that the S-AI inference information message includes at least one of information on the start time of application of the S-AI to be changed and information on the initial input value for start.

8. In paragraph 7, The step of transmitting a message including information for applying the S-AI to be changed is as follows: A step of receiving an inference information request message related to the S-AI to be changed from the terminal; and A method characterized by further comprising the step of transmitting the S-AI inference information message to the terminal in response to the inference information request message related to the S-AI.

9. In the terminal of a wireless communication system, Transmitter and receiver; and Including a control unit connected to the above transmitter and receiver, The above control unit; Receive a System Information Block (SIB) from a base station, wherein the SIB includes scheduling information related to SIB-Artificial Intelligence (SIB-AI), Check for changes to the synchronized-AI (S-AI) used based on the information contained in the above SIB, Receive SIB-AI based on the scheduling information from the base station, and the SIB-AI includes information about the S-AI to be changed, and A terminal characterized in that it performs a change of the S-AI based on the S-AI information to be changed included in the SIB-AI.

10. In paragraph 9, The above SIB contains information that instructs to toggle the above S-AI change, A terminal characterized in that the SIB-AI further includes at least one of the function of the S-AI to be changed, or the identifier (ID) of the S-AI to be changed.

11. In paragraph 9, the control unit, We are conducting testing on the above-mentioned modified S-AI, Transmitting a result report message for the change of the S-AI to the base station, wherein the result report message for the change of the S-AI includes at least one of the success or failure of the change of the S-AI or the result of the testing, and Receive a message (S-AI inference information message) including information for applying the S-AI to be changed from the base station, A terminal characterized in that the above S-AI inference information message includes at least one of information on the start time of application of the S-AI to be changed and information on the initial input value for start.

12. In paragraph 11, the control unit, Transmitting an inference information request message related to the S-AI to be changed to the above base station, and A terminal characterized by receiving the S-AI inference information message from the base station in response to the inference information request message related to the S-AI.

13. In a base station of a wireless communication system, Transmitter and receiver; and Including a control unit connected to the above transmitter and receiver, The above control unit; Decide to perform synchronized-Artificial Intelligence (S-AI) change procedure, Transmitting a System Information Block (SIB) to the terminal, wherein the SIB includes scheduling information related to SIB-AI, and A control unit for transmitting SIB-AI to the terminal based on the scheduling information, wherein the SIB-AI includes information about the S-AI to be changed, A base station characterized in that the change of the S-AI is performed based on the S-AI information to be changed.

14. In paragraph 13, The above SIB includes information that instructs to toggle the change of the above S-AI, A base station, characterized in that the SIB-AI further includes at least one of the function of the S-AI to be changed, or the identifier (ID) of the S-AI to be changed.

15. In paragraph 13, the control unit, Receive a result report message regarding a change in the S-AI from the terminal, and the result report message regarding a change in the S-AI includes at least one of the success or failure of the change in the S-AI or the result of testing performed by the terminal. Determine information for applying the S-AI to be changed based on a result report message received from the terminal regarding the change of the S-AI, Transmitting a message (S-AI inference information message) containing information for applying the S-AI to be changed to the terminal, Receive an inference information request message related to the S-AI to be changed from the terminal, and In response to the above S-AI and related inference information request message, the S-AI inference information message is transmitted to the terminal, A base station, characterized in that the S-AI inference information message includes at least one of information on the start time of application of the S-AI to be changed and information on the initial input value for start.

Images