Devices and methods for communication
Devices in communication networks manage AI-related states through message-based transitions, improving communication performance by coordinating AI-related resource management across multiple entities.
Patent Information
- Application Number
- PCT/CN2025/093457
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-05-08
- Publication Date
- 2026-11-12
AI Technical Summary
As communication networks grow in size, complexity, and user base, managing AI-related states and resources across multiple devices becomes increasingly complex, necessitating improved methods for efficient transition and management of AI-related states to enhance communication performance.
A system where devices can receive messages to transition between AI-related states, with other devices informed of the transition status, enabling coordinated management of AI-related resources and operations.
Facilitates efficient and coordinated management of AI-related states and resources across devices, enhancing communication performance and alignment among network entities.
Smart Images

Figure CN2025093457_12112026_PF_FP_ABST
Abstract
Description
DEVICES AND METHODS FOR COMMUNICATIONFIELDS
[0001] Example embodiments of the present disclosure generally relate to the field of communication techniques and in particular, to devices and methods for communication.BACKGROUND
[0002] As communication networks and services increase in size, complexity, and number of users, operations in the communication networks may become increasingly more complicated. In order to improve communication performance, machine learning (ML) / artificial intelligence (AI) technology is proposed to be used in the wireless communication network, such as beam prediction, channel state information (CSI) prediction, CSI compression and so on.SUMMARY
[0003] In general, embodiments of the present disclosure provide a solution for communication.
[0004] In a first aspect, there is provided a first device. The first device comprises: a processor configured to cause the first device to: receive, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; and transit to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0005] In a second aspect, there is provided a second device. The second device comprises: a processor configured to cause the second device to: determine whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; and in accordance with a determination that the first transit condition is met, transmit, to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0006] In a third aspect, there is provided a third device. The third device comprises: a processor configured to cause the third device to: establish a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; and receive, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.
[0007] In a fourth aspect, there is provided a communication method performed by a first device. The method comprises: receiving, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; and transiting to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0008] In a fifth aspect, there is provided a communication method performed by a second device. The method comprises: determining whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; and in accordance with a determination that the first transit condition is met, transmitting to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0009] In a sixth aspect, there is provided a communication method performed by a third device. The method comprises: establishing a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; and receiving, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.
[0010] In a seventh aspect, there is provided a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to carry out the method according to the fourth, fifth, or sixth aspect.
[0011] Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Through the more detailed description of some example embodiments of the present disclosure in the accompanying drawings, the above and other objects, features and advantages of the present disclosure will become more apparent, wherein:
[0013] FIG. 1A illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;
[0014] FIG. 1B illustrates an example diagram of AI-related state transiting in accordance with some embodiments of the present disclosure;
[0015] FIG. 2 illustrates a signaling flow for communication in accordance with some embodiments of the present disclosure;
[0016] FIG. 3A to FIG. 9C illustrate signaling flows for communication in accordance with some embodiments of the present disclosure;
[0017] FIG. 10 illustrates a flowchart of a communication method implemented at a first device according to some example embodiments of the present disclosure;
[0018] FIG. 11 illustrates a flowchart of a communication method implemented at a second device according to some example embodiments of the present disclosure;
[0019] FIG. 12 illustrates a flowchart of a communication method implemented at a third device according to some example embodiments of the present disclosure;
[0020] FIG. 13 illustrates a simplified block diagram of an apparatus that is suitable for implementing example embodiments of the present disclosure.
[0021] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION
[0022] Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. Embodiments described herein can be implemented in various manners other than the ones described below.
[0023] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
[0024] As used herein, the term ‘terminal device’ refers to any device having wireless or wired communication capabilities. Examples of the terminal device include, but not limited to, user equipment (UE) , personal computers, desktops, mobile phones, cellular phones, smart phones, personal digital assistants (PDAs) , portable computers, tablets, wearable devices, internet of things (IoT) devices, Ultra-reliable and Low Latency Communications (URLLC) devices, Internet of Everything (IoE) devices, machine type communication (MTC) devices, devices on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure / network, devices for Integrated Access and Backhaul (IAB) , Space borne vehicles or Air borne vehicles in Non-terrestrial networks (NTN) including Satellites and High Altitude Platforms (HAPs) encompassing Unmanned Aircraft Systems (UAS) , eXtended Reality (XR) devices including different types of realities such as Augmented Reality (AR) , Mixed Reality (MR) and Virtual Reality (VR) , the unmanned aerial vehicle (UAV) commonly known as a drone which is an aircraft without any human pilot, devices on high speed train (HST) , or image capture devices such as digital cameras, sensors, gaming devices, music storage and playback appliances, or Internet appliances enabling wireless or wired Internet access and browsing and the like. The ‘terminal device’ can further have ‘multicast / broadcast’ feature, to support public safety and mission critical, V2X applications, transparent IPv4 / IPv6 multicast delivery, IPTV, smart TV, radio services, software delivery over wireless, group communications and IoT applications. It may also incorporate one or multiple Subscriber Identity Module (SIM) as known as Multi-SIM. The term “terminal device” can be used interchangeably with a UE, a mobile station, a subscriber station, a mobile terminal, a user terminal or a wireless device.
[0025] The term “network device” refers to a device which is capable of providing or hosting a cell or coverage where terminal devices can communicate. Examples of a network device include, but not limited to, a Node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a next generation NodeB (gNB) , a transmission reception point (TRP) , a remote radio unit (RRU) , a radio head (RH) , a remote radio head (RRH) , an IAB node, a low power node such as a femto node, a pico node, a reconfigurable intelligent surface (RIS) , and the like. Further, the network device also may be an operation administration and maintenance (OAM) , a server, location management function (LMF) , access and mobility management function (AMF) , and other core network node.
[0026] The terminal device or the network device may have Artificial intelligence (AI) or Machine learning capability. It generally includes a model which has been trained from numerous collected data for a specific function, and can be used to predict some information.
[0027] The terminal device or the network device may work on several frequency ranges, e.g., FR1 (410 MHz -7125 MHz) , FR2 (24.25GHz -71GHz) , FR3 (7.125 GHz -24.25 GHz) , frequency band larger than 100A GHz as well as Tera Hertz (THz) . It can further work on licensed / unlicensed / shared spectrum. The terminal device may have more than one connection with the network devices under Multi-Radio Dual Connectivity (MR-DC) application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.
[0028] The embodiments of the present disclosure may be performed in test equipment, e.g., signal generator, signal analyzer, spectrum analyzer, network analyzer, test terminal device, test network device, channel emulator. In some embodiments, the terminal device may be connected with a first network device and a second network device. One of the first network device and the second network device may be a master node and the other one may be a secondary node. The first network device and the second network device may use different radio access technologies (RATs) . In some embodiments, the first network device may be a first RAT device and the second network device may be a second RAT device. In some embodiments, the first RAT device is eNB and the second RAT device is gNB. Information related with different RATs may be transmitted to the terminal device from at least one of the first network device or the second network device. In some embodiments, first information may be transmitted to the terminal device from the first network device and second information may be transmitted to the terminal device from the second network device directly or via the first network device. In some embodiments, information related with configuration for the terminal device configured by the second network device may be transmitted from the second network device via the first network device. Information related with reconfiguration for the terminal device configured by the second network device may be transmitted to the terminal device from the second network device directly or via the first network device.
[0029] As used herein, the singular forms ‘a’ , ‘an’ and ‘the’ are intended to include the plural forms as well, unless the context clearly indicates otherwise. The term ‘includes’ and its variants are to be read as open terms that mean ‘includes, but is not limited to. ’ The term ‘based on’ is to be read as ‘at least in part based on. ’ The term ‘one embodiment’ and ‘an embodiment’ are to be read as ‘at least one embodiment. ’ The term ‘another embodiment’ is to be read as ‘at least one other embodiment. ’ The terms ‘first, ’ ‘second, ’ and the like may refer to different or same objects. Other definitions, explicit and implicit, may be included below.
[0030] In some examples, values, procedures, or apparatus are referred to as ‘best, ’ ‘lowest, ’ ‘highest, ’ ‘minimum, ’ ‘maximum, ’ or the like. It will be appreciated that such descriptions are intended to indicate that a selection among many used functional alternatives can be made, and such selections need not be better, smaller, higher, or otherwise preferable to other selections.
[0031] As used herein, the term “resource, ” “transmission resource, ” “uplink resource, ” or “downlink resource” may refer to any resource for performing a communication, such as a resource in time domain, a resource in frequency domain, a resource in space domain, a resource in code domain, or any other resource enabling a communication, and the like. In the following, unless explicitly stated, a resource in both frequency domain and time domain will be used as an example of a transmission resource for describing some example embodiments of the present disclosure. It is noted that example embodiments of the present disclosure are equally applicable to other resources in other domains.
[0032] In the context of the present disclosure, AI / ML-enabled feature: refers to a Feature where AI / ML may be used. AI / ML Model: refers to a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs.
[0033] AI / ML model delivery: refers to a generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. Note: An entity could mean a network node / function (e.g., gNB, location management function (LMF) , and the likes) , UE, proprietary server, and the likes.
[0034] AI / ML model inference: refers to a process of using a trained AI / ML model to produce a set of outputs based on a set of inputs. AI / ML model testing: refers to a subprocess of training, to evaluate the performance of a final AI / ML model using a dataset different from one used for model training and validation. Differently from AI / ML model validation, testing does not assume subsequent tuning of the model.
[0035] AI / ML model training: refers to a process to train an AI / ML Model [by learning the input / output relationship] in a data driven manner and obtain the trained AI / ML Model for inference. AI / ML model transfer: refers to a delivery of an AI / ML model over the air interface in a manner that is not transparent to 3GPP signalling, either parameters of a model structure known at the receiving end or a new model with parameters. Delivery may contain a full model or a partial model.
[0036] AI / ML model validation: refers to a subprocess of training, to evaluate the quality of an AI / ML model using a dataset different from one used for model training, that helps selecting model parameters that generalize beyond the dataset used for model training.
[0037] Data collection: refers to a process of collecting data by the network nodes, management entity, or UE for the purpose of AI / ML model training, data analytics and inference.
[0038] Federated learning / federated training: refers to a machine learning technique that trains an AI / ML model across multiple decentralized edge nodes (e.g., UEs, gNBs) each performing local model training using local data samples. The technique requires multiple interactions of the model, but no exchange of local data samples.
[0039] Functionality identification: refers to a process / method of identifying an AI / ML functionality for the common understanding between the NW and the UE. Note: Information regarding the AI / ML functionality may be shared during functionality identification. Where AI / ML functionality resides depends on the specific use cases and sub use cases.
[0040] Model activation refers to enable an AI / ML model for a specific function. Model deactivation refers to disable an AI / ML model for a specific function. Model download: refers to a model transfer from the network to UE. Model identification: refers to a process / method of identifying an AI / ML model for the common understanding between the NW and the UE. Note: The process / method of model identification may or may not be applicable. Note: Information regarding the AI / ML model may be shared during model identification.
[0041] Model monitoring: refers to a procedure that monitors the inference performance of the AI / ML model. Model parameter update: refers to a process of updating the model parameters of a model. Model selection: refers to a process of selecting an AI / ML model for activation among multiple models for the same AI / ML enabled feature. Note: Model selection may or may not be carried out simultaneously with model activation. Model switching: refers to a deactivating a currently active AI / ML model and activating a different AI / ML model for a specific function.
[0042] Model update: refers to a process of updating the model parameters and / or model structure of a model. Model upload: refers to a model transfer from UE to the network. Network-side (AI / ML) model: refers to an AI / ML Model whose inference is performed entirely at the network. Offline field data: refers to the data collected from field and used for offline training of the AI / ML model. Offline training: refers to an AI / ML training process where the model is trained based on collected dataset, and where the trained model is later used or delivered for inference. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as offline training by commonly accepted conventions.
[0043] Online field data: refers to the data collected from field and used for online training of the AI / ML model. Online training: refers to an AI / ML training process where the model being used for inference) is (typically continuously) trained in (near) real-time with the arrival of new training samples. Note: the notion of (near) real-time vs. non real-time is context-dependent and is relative to the inference time-scale. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as online training by commonly accepted conventions. Note: Fine-tuning / re-training may be done via online or offline training.
[0044] Reinforcement Learning (RL) : refers to a process of training an AI / ML model from input (also known as state) and a feedback signal (also known as reward) resulting from the model’s output (a.k.a. action) in an environment the model is interacting with.
[0045] Semi-supervised learning: refers to a process of training a model with a mix of labelled data and unlabelled data. Supervised learning: refers to a process of training a model from input and its corresponding labels. Two-sided (AI / ML) model: refers to a paired AI / ML Model (s) over which joint inference is performed, where joint inference comprises AI / ML Inference whose inference is performed jointly across the UE and the network, i.e., the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa. UE-sided (AI / ML) model: An AI / ML Model whose inference is performed entirely at the UE.
[0046] Unsupervised learning: refers to a process of training a model without labelled data. Proprietary-format models: refers to ML models of vendor- / device-specific proprietary format, from 3GPP perspective. They are not mutually recognizable across vendors and hide model design information from other vendors when shared. Note: An example is a device-specific binary executable format. Open-format models: refers to ML models of specified format that are mutually recognizable across vendors and allow interoperability, from 3GPP perspective. They are mutually recognizable between vendors and do not hide model design information from other vendors when shared.
[0047] For better descriptions, some terms used herein are listed as below: - Conditions: configurations supported indicated via UE capability reporting (e.g., component field) , related to model training, model inference, performance monitoring, validation procedure, fallback, of an AI / ML model / functionality or a group of AI / ML models / functionalities. - Additional conditions (e.g., scenarios, sites, and datasets) : including but not limited to, application conditions, scenarios, datasets, cell ID, timestamp and SNR, and so on; For an AI / ML-enabled feature / FG, additional conditions refer to any aspects that are assumed for the training of the model but are not a part of UE capability for the AI / ML-enabled feature / FG. It does not imply that additional conditions are necessarily specified. Additional conditions can be divided into two categories: NW-side additional conditions and UE-sided additional conditions. Note: whether specification impact is needed is a separate discussion. - Applicable condition: signal to noise ratio (SNR) , line of sight (LOS) / Non-line of sight (NLOS) , channel condition, and so on. - UE internal conditions: including but not limited to, memory, battery, computation resource, overheating and other hardware limitations on functionality / model operations. - Complexity / processing capability: TOPs, floating point operations (FLOPs) , MACs, the number of parameters and / or size. - Functionality: in case of AI / ML functionality identification and functionality-based LCM of UE-sided models and / or UE-part of two-sided models, functionality is referred to an AI / ML-enabled Feature / feature group enabled by configuration (s) , where configuration (s) is (are) supported based on conditions indicated by UE capability. Correspondingly, functionality-based LCM may operate based on, at least, one configuration of AI / ML-enabled Feature / feature group, or specific configurations of an AI / ML-enabled Feature / Feature group.
[0048] In the context of the present disclosure, terms of “model” , “functionality” , “model / functionality” , ‘configurations’ , ‘inference configuration’ may be used interchangeably. Terms of “model” , “model set” and “model group” may be used interchangeably. Terms of “functionality” , “functionality group” , “functionality set” may be used interchangeably. Terms of “ID” , “index” , “indicator” and “identifier” may be used interchangeably.
[0049] In the context of the present disclosure, terms of “transit (s) ” , “transition” , “transiting” , “transitioning” and the similar wordings may be used interchangeably.
[0050] In the context of the present disclosure, terms of “status” , “state” , “mode” and the similar wordings may be used interchangeably.
[0051] In the context of the present disclosure, terms of “AI-related state” , “AI / ML state” , “AIML state” , “AI state” , “ML state” and the similar wordings may be used interchangeably.
[0052] In the context of the present disclosure, terms of “AIML” , “AI / ML” , “AI” , “ML” , “data” and the similar wordings may be used interchangeably.
[0053] In some examples, “current AI-related state” is used as an example for describing the notifying / informing example processes. It should be noted that all the discussions about the “current AI-related state” are applicable to “future / preferred / predicted / prediction AI-related state” . Merely for brevity, the same or similar contents are omitted herein.
[0054] AI for 6G refers to using AI technologies to enhance and optimize 6G network operations, performance, and management. This involves integrating AI algorithms and machine learning models into various aspects of the 6G infrastructure to enable intelligent automation, efficient resource allocation, enhanced security, and improved service delivery. AI is a tool to make 6G better. Examples include using AI for network management and orchestration, including resource allocation, predictive maintenance, and anomaly detection; security threat detection and prevention; QoS optimization and service personalization; energy efficiency and sustainability; and radio interface optimizations such as RS overhead reduction, channel estimation, modulation, AI-based receiver, protocol parameter adaptation, etc.
[0055] 6G for AI refers to designing 6G networks to natively support and facilitate AI-driven applications, services, and use cases. This involves creating a network architecture that can efficiently handle the unique requirements of AI workloads, such as large data volumes, low latency, and high computing power. 6G is a platform to enable AI. Examples include providing AI-as-a-Service (AIaaS) to various stakeholders, enabling AI agent communication and collaboration, supporting AI model training, inference, and management, facilitating data collection, transmission, storage, and processing for AI applications, and enabling new AI-driven services such as personalized experiences, intelligent connectivity, and autonomous systems.
[0056] Principles and implementations of the present disclosure will be described in detail below with reference to the figures.
[0057] FIG. 1 illustrates schematic diagrams of an example communication environment 100 in which example embodiments of the present disclosure can be implemented.
[0058] As illustrated, for different application scenarios, the number of devices comprised in the communication environment 100 and the connections among the devices may be different. As illustrated in the examples of (A) to (G) , except for the first device 110, one or more second devices 120 (including the second devices 120-1, 120-2) and one or more third devices 130 (including the third devices 130-1, 130-2, 130-3) may be included. For purposes of discussion, the second devices 120-1, 120-2 are collectively or individually referred to as the second device 120, and the third devices 130-1, 130-2, 130-3 are collectively or individually referred to as the third device 130.
[0059] In the example of FIG. 1, the second device 120 may function as an AI / ML node which provides an AI-related resource / service / management for the first device 110. Additionally, different AI / ML nodes may be responsible for different use cases. Additionally, the second device 120 may function as an AI / ML node which provides an AI-related resource / service / management for the third device 130.
[0060] Additionally, the second device 120 may function as a data node which provides a data-related resource / service / management for the first device 110 and / or for the third device 130.
[0061] In the example of FIG. 1, the first device 110 / the second device 120 and the third device 130 may include (or may be at least part of) a terminal device, a network device, an over the top (OTT) (server) , an operation administration and maintenance (OAM) (server) , an edge cloud (server) , a neutral site, core network (location management function (LMF) , Access and Mobility Management Function (AMF) , Session Management Function (SMF) ) , transmission reception point (TRP) and so on.
[0062] As one specific example, the first device 110 may be a UE, the second device 110 may be an AI / ML node and the third device 130 may be a RAN node. In this specific example, from the UE perspective, UE may be connected to the AI / ML node directly or indirectly via the RAN node forwarding. Specifically, in the example (A) of FIG. 1, UE is connected to the AI / ML node via the RAN node connection. In the example (B) of FIG. 1, RAN node has multiple AI / ML nodes connections. In the example (C) of FIG. 1, UE is connected to multiple RAN nodes sharing the same AI / ML node. In the example (D) of FIG. 1, UE is connected to the AI / ML node directly and UE is connected to the RAN node directly. In some examples, UE and RAN node are connected to the same AI / ML node. In the example (E) of FIG. 1, UE is connected to the AI / ML node directly, and UE is connected to the RAN node directly, but UE and RAN node are connected to different AI / ML nodes. In the example (F) of FIG. 1, UE has multiple AI / ML nodes connections.
[0063] It is to be understood that the number of devices and their connections shown in FIG. 1 are only for the purpose of illustration without suggesting any limitation. The communication environment 100A may include any suitable number of devices configured to implement example embodiments of the present disclosure.
[0064] In some embodiments, the first device 110, the second device 120 and the third device 130 may communicate with each other via a channel such as a wireless communication channel on an air interface (e.g., Uu interface, N1, N2) , a PC5 interface, or a new interface dedicated for AI / ML data or service. The wireless communication channel may comprise a sidelink, a physical uplink control channel (PUCCH) , a physical uplink shared channel (PUSCH) , a physical random-access channel (PRACH) , a physical downlink control channel (PDCCH) , a physical downlink shared channel (PDSCH) and a physical broadcast channel (PBCH) . Of course, any other suitable channels are also feasible.
[0065] The communications in the communication environment 100 may conform to any suitable standards including, but not limited to, Global System for Mobile Communications (GSM) , Long Term Evolution (LTE) , LTE-Evolution, LTE-Advanced (LTE-A) , New Radio (NR) , Wideband Code Division Multiple Access (WCDMA) , Code Division Multiple Access (CDMA) , GSM EDGE Radio Access Network (GERAN) , Machine Type Communication (MTC) and the like. The embodiments of the present disclosure may be performed according to any generation communication protocols either currently known or to be developed in the future. Examples of the communication protocols include, but not limited to, the first generation (1G) , the second generation (2G) , 2.5G, 2.75G, the third generation (3G) , the fourth generation (4G) , 4.5G, the fifth generation (5G) communication protocols, 5.5G, 5G-Advanced networks, or the sixth generation (6G) networks.
[0066] As discussed above, in order to improve communication performance, ML / AI technology is proposed to be used in the wireless communication network. According to current discussions, it is suggested to build an extensible AI / ML framework on the currently-proposed AI / ML framework, with native support for AI / ML lifecycle management, such as configuration, performance monitoring, deactivation, and seamless transition to conventional algorithms.
[0067] The AI / ML node which manages the AI / ML related operations may be implemented as a separated AI / ML network node, and UE may be connected to AI / ML node directly or indirectly. As a result, at least 3 parties will be involved, such as, UE, gNB and AI / ML node. It is required to align the common understanding on AI / ML related status among all involved parties.
[0068] According to some embodiments of the present disclosure, it is proposed a solution for the case where the AI-related state (s) are introduced. In some embodiments, the AI / ML node (i.e., the second device 120) may be assumed to be an entity separated from the third device 130 (such as, the RAN node, e.g., NW, gNB, TRP) and the first device 110 (UE) . Further, the AI / ML node may be with 3GPP access or without 3GPP access. The AI / ML node may be a UE-side server or an NW-side server. The AI / ML node may be seen as a UE device or an NW device.
[0069] The AI / ML node may provide AI / ML-related model management, data management, and resource management. Model management may include lifecycle management, specifically model training, performance monitoring, optionally may include model inference and so on. Further, the model management may also include model transfer, pre-processing of the model and post-processing of the model, retraining, fine-tuning and so on. Data management may include data collection, data storage, pre-processing of data and post-processing of data, data transfer, configurations to collected data, and so on. Resource management processing of data the association of computation resources, processing units, time, frequency, storage, power to different NW / UE / model / use cases, and so on.
[0070] As one implementation, the AI / ML node may be responsible for the UE-side model, the UE part of the two-sided model. As another implementation, the AI / ML node also be responsible for the NW-sided model, the NW part of the two-sided model. In some examples, the AI / ML node is a NW function or a part of the RAN node. In some examples, the first device 110 may be a UE and the UE AI / ML state needs to be known at NW for activating / deactivating the AI-related operation at the UE.
[0071] In other examples, the first device may be an NW, and the NW AI / ML state needs to be known at UE for UE to collect data for the NW-sided model or the NW part of the two-sided model. In other examples, the first device 110 may be an NW, and the NW AI / ML state needs to be known at other NW devices to collect data for NW-sided model or the NW part of the two-sided model.
[0072] More example processes will be discussed with reference to FIG. 2, which illustrates a signaling flow 200 for communication in accordance with some embodiments of the present disclosure. The signaling flow 200 will be discussed with reference to FIG. 1, for example, by using the first device 110, the second device 120 and the third device 130.
[0073] It should be noted that the first device 110 may comprise a terminal device or an access network device, the second device 120 may comprise a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, and the third device 130 may comprise a terminal device, an access network device or a core network entity. Merely for a better understanding, in some embodiments, a UE is used as an example of the first device 110, an NW is used as an example of the second device 120 and AI / ML node is used as an example of the third device 130.
[0074] It is possible that the first device 110 and the second device 120 have a direct connection or an indirect connection. For the case that the first device 110 and AI / ML node have an indirect connection, the connection may be established via RAN mode, e.g., the first device 110 to RAN to AI / ML node, or AI / ML node to RAN to the first device 110.
[0075] According to the example embodiments of the present disclosure, the first device 110 may be operated in different AI-related states. One example AI-related state is a first AI-related state, wherein an AI-related resource (which is provided by the second device 120) is not available to the first device 110 if the first device 110 is in the first AI-related state. In other words, no connection has been established between the first device 110 and the second device 120 if the first device 110 is in the first AI-related state. In some embodiments, in response to transiting to the first AI-related state, the first device 110 (or the third device 130) may deactivate at least one AI-related operation at the first device 110. Further, the first AI-related state may be also called AI / ML_idle state, AI / ML_unavailable state.
[0076] Another example AI-related state is a second AI-related state, wherein an AI-related resource is available to the first device 110 if the first device 110 is in the second AI-related state. In other words, there is a connection between the first device 110 and the second device 120 if the first device 110 is in the second AI-related state. In some embodiments, in response to transiting to a second AI-related state, the first device 110 (or the third device 130) may activate or resume at least one AI-related operation at the first device 110. Further, the second AI-related state may be also called AI / ML_connected state, AI / ML_available state.
[0077] According to some embodiments of the present disclosure, an AI-related operation at the first device 110 may be only allowed to be activated if the first device 110 is operated in the second AI-related state. In view of this, in some embodiments, the first device 110 may receive a message for deactivating an AI-related operation at the first device 110 in response to the first device 110 transiting to the first AI-related state from the third device 130. In other words, the third device 130 may deactivate an AI-related operation at the first device 110 in response to the first device 110 transiting to the first AI-related state.
[0078] Accordingly, in some embodiments, the first device 110 may receive a message for activating an AI-related operation at the first device 110 if the first device 110 is in a second AI-related state from the third device 130. In other words, the third device 130 may active an AI-related operation at the first device 110 if the first device 110 is in a second AI-related state.
[0079] A further example AI-related state is a third AI-related state, wherein a first set of AI-related operations is supported by the first device 110 if the first device 110 is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device 110 if the first device 110 is in the second AI-related state. In some embodiments, the connection between the first device 110 and the second device 120 is suspended if the first device 110 is in the third AI-related state Further, the third AI-related state may be also called AI / ML_inactive state.
[0080] A further example AI-related state is a fourth AI-related state, wherein the first device 110 is known to the second device 120 if the first device 110 is in the fourth AI-related state. A further example AI-related state is a fifth AI-related state, wherein the first device 110 is unknown to the second device 120 if the first device 110 is in the fifth AI-related state. Further, the fourth AI-related state may be also called as AI / ML_registered state, and the fifth AI-related state may be also called as AI / ML_unregistered state or AI / ML_deregistered state.
[0081] Merely for better understanding the AI-related state introduced herein, some example embodiments about the AI-related state will be further discussed.
[0082] In some embodiments, when a connection has been established between the first device 110 and the second device 120 (the AI / ML node) , the first device 110 may be considered in an AI / ML_connected state. If no connection has been established, the first device 110 may be considered in an AI / ML_idle state.
[0083] For the first device 110 in the AI / ML_connected state, it supports at least one of the following: the transfer of AI / ML model to / from the entity; the transfer of AI / ML-related data to / from the entity; and AI / ML data collection, including RS transmission and / or measurement and / or report to obtain data, log (or store) data. The first device 110 in AI / ML_connected state also performs at least one of the following: model inference, performance monitoring, model training and so on.
[0084] For the first device 110 in AI / ML_idle state, the first device 110 may not be able to perform at least one of the following: model inference and model training, and the first device 110 may not be able to support the transfer of AI / ML model to / from the entity. Further, for the first device 110 in AI / ML_idle state, the first device 110 may still be able to perform data collection, including RS transmission and / or measurement to obtain data, log (or store) data.
[0085] In some embodiments, if the first device 110 is UE, the NW may only schedule / activate AI / ML operation if UE is in the AI / ML_connected state. In some embodiments, if the AI / ML node is not introduced, for example, it is implicit as part of UE, then the connection is not an accurate description of the states.
[0086] In some embodiments, the AI / ML_connected and AI / ML_idle may be replaced by AI / ML_available and AI / ML_unavailable. In some examples, a finer granularity may be defined for the state to reflect UE AI / ML status, for example, for a UE in AI / ML_connected state, the following states can be further decided for UE in different phases: AI / ML_tranining, AI / ML_inference, AI / ML_monitor, AI / ML_datapreparation, AI / ML_update, AI / ML_maintenance and so on.
[0087] In some embodiments, the AI / ML node provides resources for AI / ML operations (e.g., computation resources, storage, power, model management, and so on) . For the UE which needs UE-sided model or UE part of two-sided models, the AI / ML node needs to schedule and assign the AI / ML-related resources to UE and / or UE needs to request AI / ML resources. In some examples, when UE obtains the AI / ML resources, UE is in AI / ML_connected state, and when AI / ML resources are released, UE is in AI / ML_idle state.
[0088] In some embodiments, the AI / ML_inactive state may be considered optionally. The first device 110 in the AI / ML_inactive may be still connected to the AI / ML node, but it may perform fewer functions (or part of functions) than in AI / ML_connected state, for example, it can only perform data collection but not model inference and model training.
[0089] In some examples, when the first device 110 has obtained the required AI / ML resources sufficiently, the first device 110 may be in AI / ML_connected state, and when the first device 110does not obtain the required AI / ML resources sufficiently, the first device 110 may be in AI / ML_inactive state.
[0090] In some examples, AI / ML_registered and AI / ML_deregistered / unregistered may be considered. The first device 110 may be known or unknown to the AI / ML node (or another entity responsible for user subscription for AI / ML functionality or service or capability) respectively.
[0091] In some examples, when the first device 110 has obtained an identity from the AI / ML node (such as, UE ID for AI / ML) , the first device 110 may be is in AI / ML_registered state, and when the first device 110 does not obtain or releases the identity obtain from the AI / ML node (such as, UE ID for AI / ML) , the first device 110 may be in AI / ML_deregistered / unregistered state.
[0092] For the first device 110 which is in AI / ML_deregistered / unregistered, it may need to register first device 110 to the AI / ML node (or another entity) before it may get connected to the AI / ML node. In some embodiments, after the first device 110 registered, the default AI / ML state may be in AI / ML_idle state.
[0093] In some examples, the AI / ML node can be a 3GPP node or a non-3GPP node. Therefore, one or more of the following states may be used: AI / ML_connected with 3GPP state, AI / ML_connected with non-3GPP state, AI / ML_idle with 3GPP state, AI / ML_idle with non-3GPP state, AI / ML_inactive with 3GPP state, AI / ML_inactive with non-3GPP state, AI / ML_regestered with 3GPP state, AI / ML_regestered with non-3GPP state, AI / ML_deregestered / underegestered with 3GPP state, AI / ML_deregestered / underegestered with non-3GPP state.
[0094] In some examples, the first device 110 may not be assigned with a state. In some examples, coarser granularity or more general description is needed to define the state to reflect UE AI / ML status, including active / inactive or configured / not configured. The first device 110 and the second device 120 (AI / ML node) may determine the state of the first device 110 by using the stateless approach or Representational State Transfer (RESTful architecture) , e.g., the first device 110 includes all necessary information (current status including power, storage, computation resources required, use cases, data required) with each request to the second device 120 (AI / ML node) .
[0095] In case of stateless (or Representational State Transfer, RESTful architecture) scenario, the server (s) (second device 120) does not maintain any state about clients / UE. Clients / UE may include all pieces of necessary information with each request. This allows for load balancing, scalability, and resilience in a distributed environment. UDSF (Unstructured Data Storage Function) offers a centralized storage and thereby facilitates stateless implementations. If a network function (NF) instance goes down, another instance can obtain context data (such as service profile used at session management functions (SMF) ) from UDSF. The unified data repository (UDR) and UDSF cannot be stateless. In the stateless scenario, the server may be the AI / ML node, and the client may be the first device 110.
[0096] In addition to the AI-related states, the first device 110 also may be operation in different RRC states (or connection management (CM) states) . The RRC states and / or CM states indicates the connection state between the first device 110 and the third device 130. For example, the RRC states and / or CM states may be: RRC idle state, CM idle state, RRC connected state, CM connected state, RRC inactive state, CM inactive state and so on.
[0097] According to some embodiments of the present disclosure, the AI / ML state may be dependent on a second connection state (the AI / ML state may be jointly considered with the RRC state / CM state) , which may be the RRC state or CM state. In other words, the current AI-related state may be dependent on the RRC state or CM state, which is suitable for the case that UE is connected to the AI / ML node via gNB.
[0098] In one example, the second AI-related state may be conditioned on an RRC or CM connected state. In another example, the first device 110 may be in the first AI-related state if the first device 110 is not in an RRC or CM connected state. In a further example, the first device 110 may transit to the first AI-related state in response to the first device 110 transiting to an RRC or CM idle state, and / or transit to the first AI-related state in response to the first device 110 transiting to in RRC or CM inactive state. In a further example, the first device 110 may transit out of the second AI-related state in response to the first device 110 transiting to an RRC or CM idle state, or / or transit out of the second AI-related state in response to the first device 110 transiting to in RRC or CM inactive state.
[0099] In some embodiments, AI / ML_connected state may be conditioned on RRC_connected. When UE is not in RRC_connected, AI / ML may be in the AI-related idle state. If the RRC state changes to RRC_idle, the AI / ML state changes to AI-related idle state. This may prevent RRC_idle UE from using AI / ML.
[0100] In some embodiments, AI / ML_connected state is conditioned on CM_connected. When UE is not in CM_connected, the AI / ML state is in AI / ML_idle; or if the CM state changes to CM_idle, the AI / ML state changes to AI / ML_idle.
[0101] Alternatively, according to some embodiments of the present disclosure, the current AI-related state may be separated from (may be independent from) an RRC or CM state, which is suitable for the case that UE is connected to the AI / ML node and gNB independently. In one example, the first device 110 may be allowed to be either in the first AI-related state or in the second AI-related state if the first device 110 is in an RRC or CM connected state. Alternatively, or in addition, in another example, the first device 110 may be allowed to be either in the first AI-related state or in the second AI-related state if the first device 110 is in an RRC or CM idle state, and / or may be allowed to be either in the first AI-related state or in the second AI-related state if the first device 110 is in RRC or CM inactive state.
[0102] In some embodiments, when the first device 110is in RRC_connected state, the first device 110may be in either AI / ML_connected state or AI / ML_idle state. When the first device 110is in RRC_idle state, the first device 110may be in either AI / ML_connected state or AI / ML_idle state.
[0103] In some examples, the AI / ML state is independent of the CM state, more specifically, when the first device 110 is in CM_connected state, the first device 110 may be in either AI / ML_connected state or AI / ML_idle state; when UE is in CM_idle state, the first device 110 may be in either AI / ML_connected state or AI / ML_idle state.
[0104] According to the example embodiments of the present disclosure, the first device 110 may transit among the AI-related states. Refer to FIG. 1B, which illustrates an example diagram of AI-related state transiting procedure in accordance with some embodiments of the present disclosure.
[0105] In the example of FIG. 1B, in response to establishing a connection with the second device 120, the first device 110 may transit from the first AI-related state to the second AI-related state; in response to releasing a connection with the second device 120, the first device 110 may transit from the second AI-related state to the first AI-related state.
[0106] Further, in the example of FIG. 1B, in response to resuming a connection with the second device 120, the first device 110 may transit from the third AI-related state to the second AI-related state; in response to releasing a connection with the second device 120, the first device 110 may transit from the third AI-related state to the first AI-related state; in response to suspending a connection with the second device 120, the first device 110 may transit from the second AI-related state to the third AI-related state.
[0107] In the example of FIG. 1B, the first device 110 may trigger the state transition, the second device 120 / AI / ML node may trigger the state transition, and / or the third device 130 / NW may trigger the state transition. Alternatively, the state transition may be performed periodically, with a pre-defined (such as, by the service provider / wireless standards and so on) , configured (such as, by the AI / ML node) , or reported periodicity (such as, reported by the first device 110) .
[0108] Additionally, the state transition may be performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window. For example, when the first device 110 transits to a first AI / ML state, a timer or a time window starts. When the timer expires or the time window ends, the first device 110 transits to a second AI / ML state. The length of the timer or the time window may be pre-defined, configured, or reported together with AI / ML connection establishment request / response, or AI / ML connection release request / response, or AI / ML state transition notification, or AI / ML transition notification. Additionally, different timers or time windows may be associated with different AI / ML states.
[0109] In some embodiments, the first AI-related state may be AI / ML_idle state, the second AI-related state may be AI / ML_connected state, and the third AI-related state may be AI / ML_inactive state.
[0110] In some embodiments, AI / ML state transition at least includes AI / ML_idle to AI / ML_connected and AI / ML_connected to AI / ML_idle.
[0111] In some embodiments, upon transiting from AI / ML_connected to AI / ML_idle, AI / ML-related operations at the first device 110 may be stopped / dropped / deactivated / released, e.g., AI / ML model inference configuration / monitoring configuration is deactivated, model transfer is stopped, dataset is dropped, model training is suspended. In some embodiments, upon transiting from AI / ML_idle to AI / ML_connected, AI / ML-related operations at the first device 110 may be started / activated / resumed.
[0112] Optionally, AI / ML state transition may include AI / ML_idle to AI / ML_inactive, AI / ML_connected to AI / ML_inactive, AI / ML_inactive to AI / ML_idle, AI / ML_inactive to AI / ML_connected. Other states introduced before may also be controlled by state transition, the details are omitted here. That is because based on the discussion about transiting from AI / ML_idle to AI / ML_connected and from AI / ML_connected to AI / ML_idle, other state transitions may be performed similarly.
[0113] In some embodiments, AI / ML state transition notification is needed to inform the party not involved in the AI / ML state transition triggering. In one example, the AI / ML state transition needs to be aware at the second device 120 if it is triggered by the first device 110 or the third device 130. In another example, AI / ML state transition needs to be aware at the third device 130 if it is triggered by the first device 110 or the second device 120. In a further example, AI / ML state transition needs to be aware at the first device 110 if it is triggered by the second device 120 or the third device 130.
[0114] The above discussions are brief descriptions about the scenario where the AI-related states are introduced. In the following, more example processes when the AI-related states are introduced will be discussed with reference to FIG. 2, which illustrates a signaling flow 200 for communication in accordance with some embodiments of the present disclosure.
[0115] In operation, a connection between the first device 110 and the third device 130 may be established, and the first device 110 may obtain an AI-related resource from the second device 120.
[0116] In some cases, the first device 110 needs to register to both the second device 120 and the third device 130, and thus may be allocated with different user identities. Specifically, illustrated in FIG. 2, the first device 110 may transmit (210-1) a first register request to the second device 120, and the second device 130 may receive (220-2) the first register request accordingly. In response to the first register request, the second device 120 may transmit (220-1) a first register response comprising a first identity allocated to the first device 110, and the first device 110 may receive (220-2) the first register response accordingly.
[0117] Similarly, as illustrated in FIG. 2, the first device 110 may transmit (230-1) a second register request to the third device 130, and the third device 130 may receive (230-2) the second register request accordingly. In response to the second register request, the third device 130 may transmit (240-1) a second register response comprising a second identity allocated to the first device 110, and the first device 110 may receive (240-2) the second register response accordingly.
[0118] However, the third device 130 does not know the first identity and the second device 120 does not know the second identity. In view of this, in some cases, the first device 110 may transmit (250-1) an indication indicative of the first identity to the third device 130, and the third device 130 may receive the indication accordingly. Similarly, in some cases, the first device 110 may transmit an indication indicative of the second identity to the second device 120, and the second device 120 may receive the indication accordingly. In this way, the different user identities may be known by all of the devices.
[0119] According to some embodiments of the present disclosure, the current AI-related state should be known by the first device 110, the second device 120 and the third device 130. Specifically, in some cases, the first device 110 may receive an indication indicative of a current AI-related state of the first device 110 from the second device 120 or the third device 130. In some other cases, the second device 120 may receive an indication indicative of a current AI-related state of the first device 110 from the first device 110 or the third device 130. In some further cases, the third device 130 may receive an indication indicative of a current AI-related state of the first device 110 from the first device 110 or the second device 120. More example embodiments about the unification procedure will be discussed in the following.
[0120] In some embodiments, the third device 130 may request for the current AI-related state from the first device 110 or the second device 120. Specifically, the third device 130 may transmit an indication for requesting the current AI-related state of the first device 110 to the first device 110 or the second device 120. In response to the request, the first device 110 may transmit (290-1) / the second device 120 may transmit (295-1) an indication indicative of the current AI-related state of the first device 110 to the third device 130, and the third device 130 may receive (290-2, 295-2) the indication accordingly.
[0121] Some example processes about the third device 130 requesting for the current AI-related state will be discussed by referring to FIG. 3A and FIG. 3B, which is useful for the third device 130 (such as, an NW) to obtain information about whether the first device 110 (such as, a UE) can be scheduled or configured with specific AI / ML operations.
[0122] In the example of FIG. 3A, the third device 130 may acquire the AI / ML state from the first device 110 directly, for example, via UE capability enquiry, or other configurations to enable UAI report, RRC reconfiguration message, and so on. The first device 110 may feedback its current AI / ML state and / or its preferred AI / ML state to the third device 130. The feedback information may be via the first device 110 capability reporting, UAI reporting, RRC reconfiguration completion message, applicability report, and so on. The third device 130 may confirm the successful reception of the feedback.
[0123] For the first device 110 in AI / ML_idle, the first device 110 cannot perform any corresponding AI / ML operation, and the third device 130 may request the first device 110 to transit to the second AI-related state / AI / ML_connected state. The request can be carried in the confirmation message or via a separated message different from the confirmation message.
[0124] In the example of FIG. 3B, the third device (NW) may acquire the AI / ML state from the second device 120 (AI / ML node) , for example, via an enquiry message that includes the first device 110 information like the UE ID.
[0125] In FIG. 3B, the first device 110 may be assigned with a different UE ID (e.g., the second UE ID) by the second device 120 from the UE ID (e.g., the first UE ID) assigned by the third device 130. Therefore, the third device 130 needs to obtain the second UE ID to check the AI / ML state in the second device 120.
[0126] The enquiry message is between the second device 120 and the third device 130, and the enquiry message may be via a dedicated interface if introduced. The second device 120 responds to the enquiry with the AI / ML state of the specific first device 110. The third device 130 may confirm the successful reception of the feedback to the second device 120 / AI / ML node.
[0127] In some embodiments, for the first device 110 in AI / ML_idle, the first device 110 cannot perform any corresponding AI / ML operation, and the third device 130 may request the second device 120 to trigger the first device 110 to transit to the second AI-related state / AI / ML_connected state. The request may be carried in the confirmation message or via a separated message. In some examples, the third device 130 may directly request the first device 110 to transit to AI / ML_connected. In some other embodiments, the third device 130 may request the second device 120 to trigger the first device 110 to transit to AI / ML_connected. In some examples, according to the actual network deployment, the trigger message may be transmitted to the first device 110 via the third device 130.
[0128] In some embodiments, rather than requesting by the third device 130, the first device 110 also may inform its current AI-related state to the second device 120 / the third device 130 as discussed below.
[0129] In some embodiments, the first device 110 may transmit an indication indicative of the current AI-related state to the second device 120. As one example, the first device 110 may transmit the indication indicative of the current AI-related state to the second device 120 in response to a completion of the transiting. In another example, the first device 110 may transmit the indication indicative of the current AI-related state to the second device 120 in response to receiving an indication from the second device 120 for requesting the current AI-related state of the first device 110.
[0130] In some embodiments, the first device 110 may transmit an indication indicative of the current AI-related state to the third device 130. In one example, the first device 110 may transmit the indication indicative of the current AI-related state to the third device 130 in response to a completion of the transiting. In another example, the first device 110 may transmit the indication indicative of the current AI-related state to the third device 130 in response to receiving an indication from the third device 130 for requesting the current AI-related state. In a further example, the first device 110 may transmit the indication indicative of the current AI-related state to the third device 130 in response to receiving an indication from the second device 120 indicating the first device 110 to transmit the current AI-related state of the first device 110 to the third device 130.
[0131] Some example processes about the first device 110 notifies the current AI-related state will be discussed by referring to FIG. 4A and FIG. 4B, which is useful for the first device 110 to notify the third device 130 about its AI / ML state if the first device 110 has completed AI / ML state transition.
[0132] In the example of FIG. 4A, the first device 110 notifies the third device 130 about its AI / ML state directly, for example, via the first device 110 capability reporting, UAI reporting, RRC reconfiguration completion message and the likes. The third device 130 may confirm the successful reception of the notification.
[0133] In the example of FIG. 4B, the first device 110 notifies the second device 120 about the AI / ML state and the second device 120 delivers the AI / ML state to the third device 130.
[0134] In some cases, the first device 110 may be assigned with a different UE ID (e.g., the second UE ID) by the second device 120 from the UE ID (the first UE ID) assigned by the third device 130. Therefore, the second device 120 needs to obtain the first UE ID to notify the third device 130 about the AI / ML state of this specific the first device 110. The first and / or second the UE ID can be included in the AI / ML state notification. In some examples, according to the actual network deployment, the notification message may be transmitted from the second device 120 the first device 110 via the third device 130.
[0135] The third device 130 confirms the successful reception of the feedback to the second device 120. In some examples, the third device 130 confirms directly to the first device 110 that the notification is received.
[0136] According to some embodiments of the present disclosure, except for the first device 110, the second device 120 also may inform the current AI-related state to the third device 130 / the first device 110, or indicate the first device 110 to inform the current AI-related state to the third device 130, as discussed below.
[0137] In some embodiments, the second device 120 may transmit an indication indicative of a current AI-related state of the first device 110 to the first device 110 and / or the third device 130. In one example, the second device 120 may transmit indication indicative of the current AI-related state to the third device 130 in response to receiving an indication indicative of the current AI-related state of the first device 110 from the first device 110. In another example, the second device 120 may transmit indication indicative of a current AI-related state of the first device 110 to the third device 130 in response to receiving an indication indicative of a completion of an AI-related state transiting. In a further example, the second device 120 may transmit indication indicative of a current AI-related state of the first device 110 to the third device 130 in response to receiving an indication for requesting the current AI-related state of the first device 110 from the third device 130.
[0138] In some embodiments, the second device 120 may transmit an indication indicating the first device 110 to transmit the current AI-related state to the third device 130 to the first device 110. In one example, the second device 120 may transmit the indication indicating the first device 110 to transmit the current AI-related state to the third device 130 in response to receiving an indication indicative of the current AI-related state of the first device 110 from the first device 110. In another example, the second device 120 may transmit the indication indicating the first device 110 to transmit the current AI-related state to the third device 130 in response to receiving an indication indicative of a completion of an AI-related state transiting from the first device 110. In a further example, the second device 120 may transmit the indication indicating the first device 110 to transmit the current AI-related state to the third device 130 in response to receiving an indication for requesting the current AI-related state of the first device 110 from the third device 130.
[0139] Example processes about the second device 120 notifies the current AI-related state will be discussed by referring to FIG. 5A and FIG. 5B, which is useful for the second device 120 to notify the third device 130 about the AI / ML state report of the first device 110 connected to the third device 130 when the connected first device 110 has changed its AI / ML state.
[0140] In the example of FIG. 5A, the AI / ML node notifies the third device 130 about the AI / ML state. In FIG. 5A, the first device 110 may be assigned with a different UE ID (e.g., the second UE ID) by the second device 120 from the UE ID (e.g., the first UE ID) assigned by the third device 130. Therefore, the second device 120 needs to obtain the first UE ID to notify the third device 130 about the AI / ML state of this specific the first device 110. The notification message is between the second device 120 and the third device 130. The third device 130 may confirm the successful reception of the feedback to the second device 120.
[0141] In the example of FIG. 5B, the second device 120 may command the first device 110 to notify the third device 130 about the AI / ML state. The second device 120 may send a command to the first device 110, which may be via a dedicated interface if introduced. The command may be used to trigger the first device 110 to report its current AI / ML state to the third device 130. The notification message is between the first device 110 and the third device 130. The third device 130 may confirm the successful reception of the feedback to the first device 110.
[0142] After discussing the AI-related state notification procedure, the AI-related state transiting procedure will be discussed. According to some embodiments of the present disclosure, the first device 110 may transit among different AI-related states under the control of the second device 120.
[0143] In some embodiments, each AI-related state may correspond to a corresponding transit condition. When a transit condition is met, the second device 120 may trigger the first device 110 to transit a corresponding AI-related state. Details will be discussed by taking the first transit condition as an example. The other transit conditions are similar, and the same or similar contents are omitted here.
[0144] As illustrated in FIG. 2, the second device 120 determine (260) whether a first transit condition associated with a first AI-related state is met for a first device 110. In accordance with a determination that the first transit condition is met, the second device 120 transmits (270-1) a first message to the first device 110, wherein the first message is used for transiting the first device 110 to the first AI-related state. As for the first device 110, the first device 110 transits (280) to the first AI-related state in response to receiving the first message.
[0145] In some embodiments, the first message may comprise a command indicating a transit to the first AI-related state. Alternatively, in some embodiments, the first message may comprise a confirmation confirming a request to transit to the first AI-related state transmitted by the first device 110.
[0146] According to some embodiments of the present disclosure, although the AI-related state transiting may be controlled by the second device 120, any of the first device 110, the second device 120 or the third device 130 may trigger the AI-related state transiting.
[0147] According to some embodiments of the present disclosure, the first device 110 may trigger the AI-related state transiting. Such example embodiments will be discussed.
[0148] In some embodiments, the first device 110 may request to transiting to a second AI-related state by transmitting a connection establishment request. In one example, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if the first device 110 receives a message for paging the first device 110 from the second device 120. In another example, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if there is need to perform an AI-related operation. In a further example, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if receives a message for activating an AI-related operation at the first device 110, from the third device 130. In a further example, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if the first device 110 is indicated to perform an AI-related operation.
[0149] In some embodiments, in a case where the state transition is performed periodically, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if the periodicity is satisfied.
[0150] In some embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the first device 110 may transmit the connection establishment request for transiting to the second AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0151] In some embodiments, the connection establishment request may comprise at least one of the following: amount of AI-related resources requested by the first device 110, an identity of the first device 110, at least one use cases for which the first device 110 requests an AI-related resource, or a cause of the connection establishment request. It should be noted that other related parameters may be included in the connection establishment request.
[0152] In some other embodiments, the first device 110 may request to transiting to the first AI-related state by transmitting a connection release request. In one example, the first device 110 may transmit the connection release request if there is no need to perform an AI-related operation. In one example, the first device 110 may transmit the connection release request if the first device 110 receives a message for deactivating an AI-related operation at the first device 110 from the third device 130. In a further example, the first device 110 may transmit the connection release request if the first device 110 is unable to perform an AI-related operation.
[0153] In some other embodiments, in a case where the state transition is performed periodically, the first device 110 may transmit the connection release request for transiting to the first AI-related state if the periodicity is satisfied.
[0154] In some other embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the first device 110 may transmit the connection release request for transiting to the first AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0155] In some embodiments, the connection release request may comprise at least one of the following: a releasing cause, an identity of the first device 110, amount of AI-related resources to be released, or at least one use case for which the first device 110 requests to release an AI-related resource. It should be noted that other related parameters may be included in the connection release request.
[0156] Some example processes about the AI / ML state transition triggered by the first device 110 (UE) will be discussed by referring to FIG. 6A to FIG. 6C, which is useful for the first device 110 to trigger AI / ML state transition to better adapt to its own internal conditions such as memory, power and so on.
[0157] In the example of 6A (from AI / ML_idle to AI / ML_connected) , the first device 110 in AI / ML_idle state initiates the AI / ML Connection Request (or AI / ML service request, i.e., the connection establishment request) procedure in order to be able to perform at least one of the following: model transfer (or start / resume model transfer) , data transfer / collection (or start / resume data transfer / collection) , model training, model updating, model fine-tuning, performance monitoring, performance validation of model, model inference and so on.
[0158] In some embodiments, the first device 110 initiates the AI / ML Connection Request procedure in response to a request / command from other entities, e.g., the second device 120 (AI / ML node) or the third device 130 if the second device 120 (AI / ML node) or the third device 130 wants to activate the first device 110 to perform AI / ML-based operations.
[0159] In some embodiments, AI / ML node selection / reselection may need to be considered. In operation, if more than one AI / ML node is available, the first device 110 needs to perform AI / ML node selection / reselection. The selection criteria include stored AI / ML nodes information, such as, (remaining) capacities of the second device 120, including the computation resource, storage, connection quality, power, latency and so on. The AI / ML node may provide such information. Specifically, information about available AI / ML nodes and their (remaining) capacities may be broadcast / multicast to the first device 110s within the same region / area / domain by the second device 120s or by the associated RAN nodes. The AI / ML node ID may be PLMN unique. In some examples, it may be cell-specific.
[0160] In some examples, the first device 110 may only select the AI / ML node connected to the same third device 130 (e.g., RAN node, gNB) that the first device 110 is associated with. If an AI / ML node is selected, the first device 110 may send the AI / ML Connection Request to the selected AI / ML node.
[0161] FIG. 6B illustrates an example transiting procedure. In the example of FIG. 6B, any AI / ML node state may be introduced. In the any AI / ML node state, the first device 110 is in idle mode and has completed AI / ML node selection / reselection process and has chosen a node irrespective of PLMN identity. In the any AI / ML node state, the first device 110 may perform the second device 120 selection process to find a suitable AI / ML node. If the selection process fails, the first device 110 shall attempt to find an acceptable AI / ML node.
[0162] In some embodiments, the first device 110 may initiate the transit to the AI / ML_connected by transmitting an AI / ML connection request. In operation, if the first device 110 has a direct connection with the second device 120, the connection request message may include the first device 110 information such as the UE ID, and the amount of AI / ML resources that the first device 110 requests in terms of time, frequency, computation resource, storage, power, data and so on. In addition, the connection request may include the information about the use cases that the first device 110 requests AI / ML resources for, or the cause of the AI / ML connection request.
[0163] If the first device 110 has no direct link with the second device 120, the connection request may be forwarded by the third device 130. The connection request may also include information of the second device 120 (AI / ML node, such as, AI / ML node ID) and so on, so that the third device 130 can understand which second device 120 (AI / ML node) it should forward this message to.
[0164] In response to the AI / ML Connection Request, The second device 120 (AI / ML node) could send a confirmation / rejection message considering its own (remaining) capacity and number of subscribed devices. The first device 110 transits to AI / ML_connected state after receiving the confirmation message from the second device 120, and the first device 110 stays in AI / ML_idle state after receiving the rejection message from the second device 120.
[0165] Alternatively, in one example, the first device 110 transits to AI / ML_connected state after sending the AI / ML connection release request. In another example, the first device 110 transits to AI / ML_connected state after receiving AI / ML related resources from the second device 120, for example, model transfer, data set transfer from the second device 120.
[0166] In some embodiments, after completion of the AI / ML state transition, the first device 110 may send a completion message to the second device 120. In some examples, the completion message is the same as the AI / ML state notification message. In some examples, the first device 110 also may send a completion message to the third device 130.
[0167] In the example of 6C (from AI / ML_connected to AI / ML_idle) , the first device 110 in AI / ML_connected state initiates the AI / ML Connection Release Request procedure if the first device 110 is not able to perform at least one of the following: model transfer (or start / resume model transfer) , data transfer / collection (or start / resume data transfer / collection) model training, updating, fine-tuning, performance monitoring, performance validation of model, model inference.
[0168] In the example of 6C, the first device 110 in AI / ML_connected state may initiate the AI / ML Connection Release Request procedure in response to a request from the third device 130 if the third device 130 wants to deactivate the first device 110 to perform AI / ML-based operations, or in response to an AI / ML resource release request from the second device 120.
[0169] In some examples, if the first device 110 has a direct connection with the second device 120, the AI / ML connection request release message may include the first device 110 information such as the UE ID, and the cause of the AI / ML connection release, e.g., low power, low storage, request by AI / ML mode and so on.
[0170] In other examples, if the first device 110 has no direct link with the second device 120, the connection release request may be forwarded by the third device 130. The connection release request may also include information of the second device 120 / AI / ML node such as AI / ML node ID and so on, so that the third device 130 can understand which AI / ML node it should forward this message to.
[0171] In one example, the first device 110 transits to AI / ML_idle state after sending the AI / ML connection release request. In another example, the second device 120 could send a confirmation / rejection message and the first device 110 transitions to AI / ML_idle state after receiving the confirmation, and the first device 110 transits to AI / ML_idle state after receiving the confirmation.
[0172] In some examples, in response to transiting to AI / ML_idle state, AI / ML-related resources at the first device 110 side may be released, e.g., the stored model, the stored dataset.
[0173] According to some embodiments, the second device 120 also may trigger the AI-related state transiting. Such example embodiments will be discussed.
[0174] In some embodiments, the second device 120 may determine the first transit condition is met in response to: receiving a connection release request for transiting to a first AI-related state of the plurality of AI-related state, from the first device 110. In this event, the second device 120 may confirm or reject the connection release request.
[0175] In some embodiments, the second device 120 may determine the first transit condition is met in response to a remaining AI-related resource at the second device being below a threshold. In this event, the second device 120 may transmit a command for transiting the first device to the first AI-related state.
[0176] In some embodiments, the second device 120 may determine the first transit condition is met in response to a determination that there is no need to trigger the first device to perform an AI-related operation. In this event, the second device 120 may transmit a command for transiting the first device to the first AI-related state.
[0177] In some embodiments, the second device 120 may determine the first transit condition is met in response to: receiving a message for deactivating an AI-related operation at the first device 110 from the third device 130. In this event, the second device 120 may transmit a command for transiting the first device to the first AI-related state.
[0178] In some embodiments, the second device 120 may determine to transit the first device 110 to the second AI-related state in response to receiving a connection establishment request for transiting to a second AI-related state of the plurality of AI-related state from the first device 110. In this event, the second device 120 may confirm or reject the connection establishment request.
[0179] In some embodiments, the second device 120 may determine to transit the first device 110 to the second AI-related state in response to a remaining AI-related resource at the second device 120 being above a threshold. In this event, the second device 120 may transmit a command for transiting the first device 110 to the second AI-related state.
[0180] In some embodiments, the second device 120 may determine to transit the first device 110 to the second AI-related state in response to a determination that there is a need to trigger the first device 110 to perform an AI-related operation. In this event, the second device 120 may transmit a command for transiting the first device 110 to the second AI-related state.
[0181] In some embodiments, the second device 120 may determine to transit the first device 110 to the second AI-related state in response to receiving a message for activating an AI-related operation at the first device 110 from the third device 130. In this event, the second device 120 may transmit a command for transiting the first device 110 to the second AI-related state.
[0182] In some embodiments, in a case where the state transition is performed periodically, the second device 120 may transmit the connection establishment command for transiting to the second AI-related state if the periodicity is satisfied.
[0183] In some embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the second device 120 may transmit the connection establishment command for transiting to the second AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0184] Additionally, in a case where the state transition is performed periodically, the second device 120 may transmit the connection release command for transiting to the first AI-related state if the periodicity is satisfied.
[0185] In some embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the second device 120 may transmit the connection release command for transiting to the first AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0186] Some example processes about the AI / ML state transition triggered by the second device 120 (AI / ML node) will be discussed by referring to FIG. 7A to FIG. 7C, which is useful for the second device 120 to trigger the AI / ML state transition so that the second device 120 can assign its AI / ML resources efficiently and update model / data timely.
[0187] In the example of FIG. 7A (from the AI / ML_idle to AI / ML_connected) , the second device 120 initiates the AI / ML Connection procedure for the first device 110 in AI / ML_idle state in order to let the first device 110 be able to perform at least one of the following: model transfer (or start / resume model transfer) , data transfer / collection (or start / resume data transfer / collection) , model training, model updating, model fine-tuning, performance monitoring, performance validation of model, model inference and so on.
[0188] In some embodiments, the second device 120 initiates the AI / ML Connection procedure in response to a request from a third device 130 if the third device 130 wants to activate the first device 110 to perform AI / ML-based operations.
[0189] In the example of FIG. 7A, the second device 120 / AI / ML node uses the AI / ML connection Command to trigger a specific the first device 110 or the first device 110 group to transit from AI / ML_idle to AI / ML_connected.
[0190] If the first device 110 has a direct connection with the second device 120, the AI / ML connection command may include information of the first device 110, such as, the UE ID, and the amount of AI / ML resources that the first device 110 requests in terms of time, frequency, computation resource, storage, power and so on. In addition, the AI / ML connection command may include information of the use cases that the first device 110 requests AI / ML resources for.
[0191] If the first device 110 has no direct link with the second device 120, the connection command may be forwarded by the third device 130. The connection command may also include information of the second device 120 / AI / ML node, such as, AI / ML node ID and so on.
[0192] The first device 110 sends a response message such as a confirmation / rejection message considering its own condition. In some examples, the rejection message may include the cause of rejection such as low power, low memory and so on.
[0193] The first device 110 transitions to AI / ML_connected state after sending confirmation and the first device 110 stays in AI / ML_idle state after sending rejection. In some examples, the first device 110 transits to AI / ML_connected state after receiving AI / ML related resources from the second device 120, for example, model transfer, data set transfer from the second device 120.
[0194] In the example of FIG. 7B, the second device 120 / AI / ML node uses an AI / ML paging message to trigger a specific the first device 110 or the first device 110 group to transit from AI / ML_idle to AI / ML_connected. In the example of FIG. 7B, the first device 110 monitors the AI / ML paging message for AI / ML node (s) , and if needed, the first device 110 starts the Connection request procedure as the first device 110 initiated / triggered AI / ML state transition. This is useful to find target the first device 110s to update the model / dataset stored at the first device 110 side.
[0195] In some embodiments, the paging message may include information of the first device 110 and information about the model to be transferred / updated, dataset to be transferred / updated.
[0196] In some embodiments, AI / ML node selection / reselection may be needed to select one AI / ML node if one or more paging messages are received.
[0197] In some embodiments, the second device 120 could send a confirmation / rejection message considering its own (remaining) capacity and number of subscribed devices. The first device 110 transits to AI / ML_connected state after receiving a confirmation message from the second device 120, and the first device 110 stays in AI / ML_idle state after receiving a rejection message form the second device 120.
[0198] In one example, the first device 110 transits to AI / ML_connected state after sending the AI / ML connection release request. In another example, the first device 110 transits to AI / ML_connected state after receiving AI / ML related resources from the second device 120, for example, model transfer, data set transfer from the second device 120.
[0199] In the example of FIG. 7C (from AI / ML_connected to AI / ML_idle) , the second device 120 initiates the AI / ML Connection Release procedure for the first device 110 in AI / ML_connected state, for example, the second device 120 does not have sufficient AI / ML resources and cannot handle the number of the first device 110s connected.
[0200] In some embodiments, the first device 110 transits to AI / ML_idle state after sending a release response. Alternatively, the first device 110 transits to AI / ML_idle state after receiving the release command. In some embodiments, upon transiting to the AI / ML_idle state, AI / ML-related resources at the first device 110 side may be released, e.g., the stored model, the stored dataset.
[0201] According to some embodiments, the third device 130 also may trigger the AI-related state transiting. Such example embodiments will be discussed.
[0202] In some embodiments, the third device 130 may determine whether a corresponding transit condition associated with an AI-related state of a plurality of AI-related states is met for a first device 110. In accordance with a determination that the first transit condition is met, the third device 130 may transmit a second message for transiting the first device 110 to the first AI-related state to the first device 110.
[0203] In some embodiments, in response to the corresponding transit condition being met, the third may transmit a message for activating / deactivating an AI-related operation at the first device 110 (to the first device 110 or the second device 120) .
[0204] In above processes are mainly about how to trigger the AI-related state transiting. In the following, some details about the operations for performing the AI-related state transiting will be discussed.
[0205] In some embodiments, in order to enable the first device 110 may transit to the second AI_related state, the second device (s) may broadcast or multicast at least one of the following: an identity of the second device 120, or information about remaining AI-related resources of the second device 120.
[0206] As for the first device 110, in response to determining to transit to a second AI-related state, the first device 110 may select the second device 120 from more than one available second device 120 based on at least one of the following: information about remaining AI-related resources of the more than one available second device 120, or whether both an available second device 120 and the first device 110 are connected to the third device 130.
[0207] In some embodiments, in a case where the state transition is performed periodically, the third device 130 may transmit activation command for transiting to the second AI-related state if the periodicity is satisfied.
[0208] In some embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the third device 130 may transmit activation command for transiting to the second AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0209] In some embodiments, in a case where the state transition is performed periodically, the third device 130 may transmit the connection release command for transiting to the first AI-related state if the periodicity is satisfied.
[0210] In some embodiments, in a case where the state transition is performed based on a timer or time window based approach which is controlled by timer expiry or the ending of the time window, the third device 130 may transmit the connection release command for transiting to the first AI-related state if a timer for the first AI-related state expires or a time window for the first AI-related state ends.
[0211] Some example processes about the AI / ML state transition triggered by the third device 130 (NW) will be discussed by referring to FIG. 8A to FIG. 8C, which is useful for the third device 130 to trigger the AI / ML state transition if the third device 130 wants to let the first device 110 apply the UE sided model to get some predictions from the first device 110 or the third device 130 wants to let the first device 110 apply the UE part of two-sided model so that the third device 130 can start using the two-sided model.
[0212] The third device 130 initiates the AI / ML Connection procedure for the first device 110 in AI / ML_idle state in order to let the first device 110 be able to perform at least one of the following: model transfer (or start / resume model transfer) , data transfer / collection (or start / resume data transfer / collection) , model training, model updating, model fine-tuning, performance monitoring, performance validation of model, model inference and so on. In some examples, the third device 130 initiates the AI / ML Connection procedure in response to a request from the second device 120.
[0213] In the example of FIG. 8A, the third device 130 sends a message (e.g., AI / ML activation command) to the first device 110, and the first device 110 initiates the AI / ML connection request procedure as discussed with reference to FIG. 6A to FIG. 6C. The AI / ML Activation command may include information about the amount of AI / ML resources that the first device 110 needs to request from the second device 120 in terms of time, frequency, computation resource, storage, power and so on. The AI / ML Activation command also may include information about the use cases that the first device 110 needs to request AI / ML resources for, or the cause of the connection request.
[0214] In the example of FIG. 8B, the third device 130 sends a message (e.g., AI / ML activation request for a specific the first device 110) to the second device 120, the second device 120 triggers the AI / ML Connection procedure as discussed with reference to FIG. 7A to FIG. 7C. In some cases, the first device 110 may be assigned with a different UE ID (e.g., a second UE ID) by the second device 120 from the UE ID (e.g., a first UE ID) assigned by the third device 130. Therefore, the third device 130 needs to include at least the second UE ID in the AI / ML activation for the specific the first device 110.
[0215] In one example, the first device 110 transits to AI / ML_connected state after sending confirmation and the first device 110 stays in AI / ML_idle state after sending rejection. In another example, the first device 110 transits to AI / ML_connected state after receiving AI / ML related resources from the second device 120, for example, model transfer, data set transfer from the second device 120.
[0216] In the example of FIG. 8C, the third device 130 initiates the AI / ML Connection Release procedure for the first device 110 in AI / ML_connected state. In one example, the third device 130 sends a message (e.g., AI / ML connection release command) to the first device 110, the first device 110 transitions to AI / ML_idle. In another example, the third device 130 sends a message (e.g., AI / ML connection release command) to the first device 110, the first device 110 initiates the AI / ML connection release procedure. In a further example, the third device 130 sends a message (e.g., AI / ML connection release request for a specific the first device 110) to the second device 120, the second device 120 triggers the AI / ML Connection procedure.
[0217] In some cases, the first device 110 may be assigned with a different UE ID (e.g., a second UE ID) by the second device 120 from the UE ID (e.g., a first UE ID) assigned by the third device 130. Therefore, the third device 130 needs to include at least the second UE ID in the AI / ML connection release request for the specific the first device 110.
[0218] In some embodiments, the first device 110 transits to AI / ML_idle state after sending a response. Alternatively, the first device 110 transits to AI / ML_idle state after receiving the release command.
[0219] In some examples, upon transiting to the AI / ML_idle state, the AI / ML-related resources at the first device 110 side are released, e.g., the stored model, the stored dataset.
[0220] According to some embodiments of the present disclosure, the AI / ML state transition notification may be sent to the entity which is not involved in the AI / ML state transition.
[0221] In the example of FIG. 9A, the AI / ML state transition needs to be aware at the third device 130 (NW) , if it is triggered by the first device 110 or AI / ML node. Specifically, after the AI / ML state transition, AI / ML state transition notification can be sent to the third device 130 by the first device 110 or by the second device 120.
[0222] In the example of FIG. 9B, the AI / ML state transition needs to be aware at the second device 120, if it is triggered by the first device 110 or AI / ML node. Specifically, after the AI / ML state transition, AI / ML state transition notification can be sent to the second device 120 by the first device 110 or by the NW. In some examples, the transition is known during the AI / ML connection request procedure or AI / ML connection release procedure discussed in previous cases.
[0223] In the example of FIG. 9C, the AI / ML state transition needs to be aware at the first device 110, if it is triggered by AI / ML node or NW. Before the AI / ML state transition, AI / ML state transition notification can be sent to the first device 110 by the second device 120 or by the third device 130. In some examples, the transition notification is the same as the trigger message used during the AI / ML connection request procedure or AI / ML connection release procedure discussed in previous cases.
[0224] FIG. 10 illustrates a flowchart of a communication method 1000 implemented at a first device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1000 will be described from the perspective of the first device 110 in FIG. 1.
[0225] At block 1010, the first device receives, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states.
[0226] At block 1020, the first device transits to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0227] In some example embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0228] In some example embodiments, the first device may receive, from the third device, a message for deactivating an AI-related operation at the first device in response to the first device transiting to the first AI-related state; or receive, from the third device, a message for activating an AI-related operation at the first device if the first device is in a second AI-related state.
[0229] In some example embodiments, in response to transiting to the first AI-related state, the first device may deactivate at least one AI-related operation; or in response to transiting to a second AI-related state, activate or resume at least one AI-related operation.
[0230] In some example embodiments, the first device may transmit a first register request to the second device; receive, from the second device, a first register response comprising a first identity allocated to the first device; transmit a second register request to the third device; receive, from the third device, a second register response comprising a second identity allocated to the first device, wherein the first identity and second identity are different from each other; and transmit, to the third device, an indication indicative of the first identity.
[0231] In some example embodiments, the first device may transmit, to the second device and / or the third device, an indication indicative of a current AI-related state of the first device.
[0232] In some example embodiments, the first device may transmit the indication indicative of the current AI-related state to the second device in response to at least one of the following: a completion of the transiting, or receiving an indication from the second device for requesting the current AI-related state of the first device; or transmit the indication indicative of the current AI-related state to the third device in response to at least one of the following: a completion of the transiting, receiving an indication from the third device for requesting the current AI-related state of the first device, or receiving an indication from the second device indicating the first device to transmit the current AI-related state of the first device to the third device.
[0233] In some example embodiments, the first device may receive, from the second device or the third device, an indication indicative of a current AI-related state of the first device.
[0234] In some example embodiments, the first device is in the first AI-related state, and in response to determining to transit to a second AI-related state, the first device may select the second device from more than one available second device based on at least one of the following: information about remaining AI-related resources of the more than one available second device, whether both an available second device and the first device are connected to the third device.
[0235] In some example embodiments, the first message comprises at least one of the following: a command indicating a transit to the first AI-related state; or a confirmation confirming a request to transit to the first AI-related state transmitted by the first device.
[0236] In some example embodiments, the first device is in the first AI-related state and the first device is caused to: transmit, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related states if: receive, from the second device, a message for paging the first device; there is need to perform an AI-related operation; receiving, from the third device, a message for activating an AI-related operation at the first device; or the first device is indicated to perform an AI-related operation.
[0237] In some example embodiments, the connection establishment request comprises at least one of the following: amount of AI-related resources requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.
[0238] In some example embodiments, the first device is in a second AI-related state and the first device is caused to: transmit, a connection release request for transiting to the first AI-related state of the plurality of AI-related states if: there is no need to perform an AI-related operation; receive, from the third device, a message for deactivating an AI-related operation at the first device; or the first device is unable to perform an AI-related operation.
[0239] In some example embodiments, the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.
[0240] In some example embodiments, rather than receiving the first message from the second device, the first device may receive a second message from the third device, wherein the second message is used for transiting the first device to the first AI-related state of a plurality of AI-related states; and transit to the first AI-related state in response to receiving the second message.
[0241] In some example embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0242] In some example embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0243] FIG. 11 illustrates a flowchart of a communication method 1100 implemented at a second device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1100 will be described from the perspective of the second device 120 in FIG. 1.
[0244] At block 1110, the second device determines whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device.
[0245] At block 1120, in accordance with a determination that the first transit condition is met, the second device transmits, to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0246] In some example embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0247] In some example embodiments, the second device may transmit, to the third device or the first device, an indication indicative of a current AI-related state of the first device.
[0248] In some example embodiments, the second device may transmit indication indicative of a current AI-related state of the first device to the third device in response to at least one of the following: receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; or receiving, from the third device, an indication for requesting the current AI-related state of the first device.
[0249] In some example embodiments, the second device may transmit, to the first device, an indication indicating the first device to transmit a current AI-related state of the first device to the third device.
[0250] In some example embodiments, the second device may transmit the indication indicating the first device to transmit the current AI-related state of the first device to the third device in response to at least one of the following: receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; or receiving, from the third device, an indication for requesting the current AI-related state of the first device.
[0251] In some example embodiments, the second device may receive, from the first device or the third device, an indication indicative of a current AI-related state of the first device.
[0252] In some example embodiments, the second device may broadcast or multicast at least one of the following: an identity of the second device, or information about remaining AI-related resources of the second device.
[0253] In some example embodiments, the first message comprises at least one of the following: a command indicating a transit to the first AI-related state; or a confirmation confirming a request to transit to the first AI-related state transmitted by the first device.
[0254] In some example embodiments, the second device may determine to transit the first device to a second AI-related state in response to at least of the following: receiving, from the first device, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related state; a remaining AI-related resource at the second device being above a threshold; a determination that there is a need to trigger the first device to perform an AI-related operation; receiving, from the third device, a message for activating an AI-related operation at the first device.
[0255] In some example embodiments, the connection establishment request comprises at least one of the following: AI-related resource requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.
[0256] In some example embodiments, the second device may determine the first transit condition is met in response to at least of the following: receiving, from the first device, a connection release request for transiting to a first AI-related state of the plurality of AI-related state; a remaining AI-related resource at the second device being below a threshold; a determination that there is no need to trigger the first device to perform an AI-related operation; or receiving, from the third device, a message for deactivating an AI-related operation at the first device.
[0257] In some example embodiments, the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.
[0258] In some example embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0259] In some example embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0260] FIG. 12 illustrates a flowchart of a communication method 1200 implemented at a third device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1200 will be described from the perspective of the third device 130 in FIG. 1.
[0261] At block 1210, the third device establishes a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device.
[0262] At block 1220, the third device receives, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.
[0263] In some example embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0264] In some example embodiments, the third device may deactivate an AI-related operation at the first device in response to the first device transiting to the first AI-related state, or active an AI-related operation at the first device if the first device is in a second AI-related state.
[0265] In some example embodiments, the third device may receive a second register request from the first device; transmit, to the first device, a second register response comprising a second identity allocated to the first device; and receive, from the first device, an indication indicative of a first identity of the first device allocated by the second device, wherein the first identity and second identity are different from each other.
[0266] In some example embodiments, the third device may receive, from the first device or the second device, an indication indicative of the current AI-related state of the first device.
[0267] In some example embodiments, the third device may transmit, to the first device or the second device, an indication for requesting the current AI-related state of the first device.
[0268] In some example embodiments, the third device may transmit, to the first device or the second device, an indication indicative of a current AI-related state of the first device.
[0269] In some example embodiments, the third device may determine whether a first transit condition associated with a first AI-related state of a plurality of AI-related states is met for a first device; and in accordance with a determination that the first transit condition is met, transmit, to the first device, a second message for transiting the first device to the first AI-related state.
[0270] In some example embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0271] In some example embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0272] FIG. 13 is a simplified block diagram of a device 1300 that is suitable for implementing embodiments of the present disclosure. The device 1300 can be considered as a further example implementation of any of the devices as shown in FIG. 1. Accordingly, the device 1300 can be implemented at or as at least a part of the first device 110, the second device 120 and the third device 130.
[0273] As shown, the device 1300 includes a processor 1310, a memory 1320 coupled to the processor 1310, a suitable transceiver 1340 coupled to the processor 1310, and a communication interface coupled to the transceiver 1340. The memory 1320 stores at least a part of a program 1330. The transceiver 1340 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 1340 may include at least one of a transmitter 1342 and a receiver 1344. The transmitter 1342 and the receiver 1344 may be functional modules or physical entities. The transceiver 1340 has at least one antenna to facilitate communication, though in practice an Access Node mentioned in this application may have several ones. The communication interface may represent any interface that is necessary for communication with other network elements, such as X2 / Xn interface for bidirectional communications between eNBs / gNBs, S1 / NG interface for communication between a Mobility Management Entity (MME) / Access and Mobility Management Function (AMF) / SGW / UPF and the eNB / gNB, Un interface for communication between the eNB / gNB and a relay node (RN) , or Uu interface for communication between the eNB / gNB and a terminal device.
[0274] The program 1330 is assumed to include program instructions that, when executed by the associated processor 1310, enable the device 1300 to operate in accordance with the embodiments of the present disclosure, as discussed herein with reference to FIGS. 1 to 12. The embodiments herein may be implemented by computer software executable by the processor 1310 of the device 1300, or by hardware, or by a combination of software and hardware. The processor 1310 may be configured to implement various embodiments of the present disclosure. Furthermore, a combination of the processor 1310 and memory 1320 may form processing means 1350 adapted to implement various embodiments of the present disclosure.
[0275] The memory 1320 may be of any type suitable to the local technical network and may be implemented using any suitable data storage technology, such as a non-transitory computer readable storage medium, semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory, as non-limiting examples. While only one memory 1320 is shown in the device 1300, there may be several physically distinct memory modules in the device 1300. The processor 1310 may be of any type suitable to the local technical network, and may include one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multicore processor architecture, as non-limiting examples. The device 1300 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.
[0276] According to embodiments of the present disclosure, a first device comprising a circuitry is provided. The circuitry is configured to: receive, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; and transit to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the first device as discussed above.
[0277] According to embodiments of the present disclosure, a second device comprising a circuitry is provided. The circuitry is configured to: determine whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; and in accordance with a determination that the first transit condition is met, transmit, to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the second device as discussed above.
[0278] According to embodiments of the present disclosure, a third device comprising a circuitry is provided. The circuitry is configured to: establish a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; and receive, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the third device as discussed above.
[0279] The term “circuitry” used herein may refer to hardware circuits and / or combinations of hardware circuits and software. For example, the circuitry may be a combination of analog and / or digital hardware circuits with software / firmware. As a further example, the circuitry may be any portions of hardware processors with software including digital signal processor (s) , software, and memory (ies) that work together to cause an apparatus, such as a terminal device or a network device, to perform various functions. In a still further example, the circuitry may be hardware circuits and or processors, such as a microprocessor or a portion of a microprocessor, that requires software / firmware for operation, but the software may not be present when it is not needed for operation. As used herein, the term circuitry also covers an implementation of merely a hardware circuit or processor (s) or a portion of a hardware circuit or processor (s) and its (or their) accompanying software and / or firmware.
[0280] According to embodiments of the present disclosure, a first apparatus is provided. The first apparatus comprises means for receiving, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; and means for transiting to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting. In some embodiments, the first apparatus may comprise means for performing the respective operations of the method 1000. In some example embodiments, the first apparatus may further comprise means for performing other operations in some example embodiments of the method 1000. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.
[0281] According to embodiments of the present disclosure, a second apparatus is provided. The second apparatus comprises means for determining whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; and means for in accordance with a determination that the first transit condition is met, transmitting to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting. In some embodiments, the second apparatus may comprise means for performing the respective operations of the method 1100. In some example embodiments, the second apparatus may further comprise means for performing other operations in some example embodiments of the method 1100. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.
[0282] According to embodiments of the present disclosure, a third apparatus is provided. The third apparatus comprises means for establishing a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; and means for receiving, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states. In some embodiments, the third apparatus may comprise means for performing the respective operations of the method 1200. In some example embodiments, the third apparatus may further comprise means for performing other operations in some example embodiments of the method 1200. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.
[0283] In summary, embodiments of the present disclosure provide the following aspects.
[0284] In an aspect, it is proposed a first device comprising: a processor configured to cause the first device to: receive, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; and transit to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0285] In some embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0286] In some embodiments, the first device is caused to: receive, from the third device, a message for deactivating an AI-related operation at the first device in response to the first device transiting to the first AI-related state; or receive, from the third device, a message for activating an AI-related operation at the first device if the first device is in a second AI-related state.
[0287] In some embodiments, the first device is caused to: in response to transiting to the first AI-related state, deactivate at least one AI-related operation; or in response to transiting to a second AI-related state, activate or resume at least one AI-related operation.
[0288] In some embodiments, the first device is caused to: transmit a first register request to the second device; receive, from the second device, a first register response comprising a first identity allocated to the first device; transmit a second register request to the third device; receive, from the third device, a second register response comprising a second identity allocated to the first device, wherein the first identity and second identity are different from each other; and transmit, to the third device, an indication indicative of the first identity.
[0289] In some embodiments, the first device is caused to: transmit, to the second device and / or the third device, an indication indicative of a current AI-related state of the first device.
[0290] In some embodiments, the first device is caused to: transmit the indication indicative of the current AI-related state to the second device in response to at least one of the following: a completion of the transiting, or receiving an indication from the second device for requesting the current AI-related state of the first device; or transmit the indication indicative of the current AI-related state to the third device in response to at least one of the following: a completion of the transiting, receiving an indication from the third device for requesting the current AI-related state of the first device, or receiving an indication from the second device indicating the first device to transmit the current AI-related state of the first device to the third device.
[0291] In some embodiments, the first device is caused to: receive, from the second device or the third device, an indication indicative of a current AI-related state of the first device.
[0292] In some embodiments, the first device is in the first AI-related state, and the first device is caused to: in response to determining to transit to a second AI-related state, select the second device from more than one available second device based on at least one of the following: information about remaining AI-related resources of the more than one available second device, whether both an available second device and the first device are connected to the third device.
[0293] In some embodiments, the first message comprises at least one of the following: a command indicating a transit to the first AI-related state; or a confirmation confirming a request to transit to the first AI-related state transmitted by the first device.
[0294] In some embodiments, the first device is in the first AI-related state and the first device is caused to: transmit, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related states if: receive, from the second device, a message for paging the first device; there is need to perform an AI-related operation; receiving, from the third device, a message for activating an AI-related operation at the first device; or the first device is indicated to perform an AI-related operation.
[0295] In some embodiments, the connection establishment request comprises at least one of the following: amount of AI-related resources requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.
[0296] In some embodiments, the first device is in a second AI-related state and the first device is caused to: transmit, a connection release request for transiting to the first AI-related state of the plurality of AI-related states if: there is no need to perform an AI-related operation; receive, from the third device, a message for deactivating an AI-related operation at the first device; or the first device is unable to perform an AI-related operation.
[0297] In some embodiments, the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.
[0298] In some embodiments, the first device is caused to: rather than receiving the first message from the second device, receive a second message from the third device, wherein the second message is used for transiting the first device to the first AI-related state of a plurality of AI-related states; and transit to the first AI-related state in response to receiving the second message.
[0299] In some embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0300] In some embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0301] In an aspect, it is proposed a second device comprising: a processor configured to cause the second device to: determine whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; and in accordance with a determination that the first transit condition is met, transmit, to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.
[0302] In some embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0303] In some embodiments, the second device is caused to: transmit, to the third device or the first device, an indication indicative of a current AI-related state of the first device.
[0304] In some embodiments, the second device is caused to: transmit indication indicative of a current AI-related state of the first device to the third device in response to at least one of the following: receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; or receiving, from the third device, an indication for requesting the current AI-related state of the first device.
[0305] In some embodiments, the second device is caused to: transmit, to the first device, an indication indicating the first device to transmit a current AI-related state of the first device to the third device.
[0306] In some embodiments, the second device is caused to: transmit the indication indicating the first device to transmit the current AI-related state of the first device to the third device in response to at least one of the following: receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; or receiving, from the third device, an indication for requesting the current AI-related state of the first device.
[0307] In some embodiments, the second device is caused to: receive, from the first device or the third device, an indication indicative of a current AI-related state of the first device.
[0308] In some embodiments, the second device is caused to: broadcast or multicast at least one of the following: an identity of the second device, or information about remaining AI-related resources of the second device.
[0309] In some embodiments, the first message comprises at least one of the following: a command indicating a transit to the first AI-related state; or a confirmation confirming a request to transit to the first AI-related state transmitted by the first device.
[0310] In some embodiments, the second device is caused to: determine to transit the first device to a second AI-related state in response to at least of the following: receiving, from the first device, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related state; a remaining AI-related resource at the second device being above a threshold; a determination that there is a need to trigger the first device to perform an AI-related operation; receiving, from the third device, a message for activating an AI-related operation at the first device.
[0311] In some embodiments, the connection establishment request comprises at least one of the following: AI-related resource requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.
[0312] In some embodiments, the second device is caused to: determine the first transit condition is met in response to at least of the following: receiving, from the first device, a connection release request for transiting to a first AI-related state of the plurality of AI-related state; a remaining AI-related resource at the second device being below a threshold; a determination that there is no need to trigger the first device to perform an AI-related operation; or receiving, from the third device, a message for deactivating an AI-related operation at the first device.
[0313] In some embodiments, the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.
[0314] In some embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0315] In some embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0316] In an aspect, it is proposed a third device comprising: a processor configured to cause the third device to: establish a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; and receive, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.
[0317] In some embodiments, the plurality of AI-related states comprise at least one of the following: a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state, a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state, a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state, a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, or a fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.
[0318] In some embodiments, the third device is caused to: deactivate an AI-related operation at the first device in response to the first device transiting to the first AI-related state, or active an AI-related operation at the first device if the first device is in a second AI-related state.
[0319] In some embodiments, the third device is caused to: receive a second register request from the first device; transmit, to the first device, a second register response comprising a second identity allocated to the first device; and receive, from the first device, an indication indicative of a first identity of the first device allocated by the second device, wherein the first identity and second identity are different from each other.
[0320] In some embodiments, the third device is caused to: receive, from the first device or the second device, an indication indicative of the current AI-related state of the first device.
[0321] In some embodiments, the third device is caused to: transmit, to the first device or the second device, an indication for requesting the current AI-related state of the first device.
[0322] In some embodiments, the third device is caused to: transmit, to the first device or the second device, an indication indicative of a current AI-related state of the first device.
[0323] In some embodiments, the third device is caused to: determine whether a first transit condition associated with a first AI-related state of a plurality of AI-related states is met for a first device; and in accordance with a determination that the first transit condition is met, transmit, to the first device, a second message for transiting the first device to the first AI-related state.
[0324] In some embodiments, a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following: a second AI-related state being conditioned on an RRC or CM connected state, being in the first AI-related state if the first device is not in an RRC or CM connected state, transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state; transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state; transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; or transiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; or a current AI-related state is separated from an RRC or CM state by at least one of the following: being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state; being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.
[0325] In some embodiments, the first device comprises a terminal device or an access network device, or the second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, or the third device comprises a terminal device, an access network device or a core network entity.
[0326] In an aspect, a first device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the first device discussed above.
[0327] In an aspect, a second device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the second device discussed above.
[0328] In an aspect, a third device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the third device discussed above.
[0329] In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the first device discussed above.
[0330] In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the second device discussed above.
[0331] In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the third device discussed above.
[0332] In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the first device discussed above.
[0333] In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the second device discussed above.
[0334] In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the third device discussed above.
[0335] Generally, various embodiments of the present disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. While various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other pictorial representation, it will be appreciated that the blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.
[0336] The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer readable storage medium. The computer program product includes computer-executable instructions, such as those included in program modules, being executed in a device on a target real or virtual processor, to carry out the process or method as described above with reference to FIGS. 1 to 13. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments. Machine-executable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.
[0337] Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
[0338] The above program code may be embodied on a machine readable medium, which may be any tangible medium that may contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine readable medium may be a machine readable signal medium or a machine readable storage medium. A machine readable medium may include but not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM) , a read-only memory (ROM) , an erasable programmable read-only memory (EPROM or Flash memory) , an optical fiber, a portable compact disc read-only memory (CD-ROM) , an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0339] Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment may also be implemented in multiple embodiments separately or in any suitable sub-combination.
[0340] Although the present disclosure has been described in language specific to structural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims
1.A first device comprising:a processor configured to cause the first device to:receive, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; andtransit to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.2.The first device of claim 1, wherein the plurality of AI-related states comprise at least one of the following:a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state,a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state,a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state,a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, ora fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.3.The first device of claim 1, wherein the first device is caused to:receive, from the third device, a message for deactivating an AI-related operation at the first device in response to the first device transiting to the first AI-related state; orreceive, from the third device, a message for activating an AI-related operation at the first device if the first device is in a second AI-related state.4.The first device of claim 1, wherein the first device is caused to:in response to transiting to the first AI-related state, deactivate at least one AI-related operation; orin response to transiting to a second AI-related state, activate or resume at least one AI-related operation.5.The first device of claim 1, wherein the first device is caused to:transmit a first register request to the second device;receive, from the second device, a first register response comprising a first identity allocated to the first device;transmit a second register request to the third device;receive, from the third device, a second register response comprising a second identity allocated to the first device, wherein the first identity and second identity are different from each other; andtransmit, to the third device, an indication indicative of the first identity.6.The first device of claim 1, wherein the first device is caused to:transmit, to the second device and / or the third device, an indication indicative of a current AI-related state of the first device.7.The first device of claim 6, wherein the first device is caused to:transmit the indication indicative of the current AI-related state to the second device in response to at least one of the following:a completion of the transiting, orreceiving an indication from the second device for requesting the current AI-related state of the first device; ortransmit the indication indicative of the current AI-related state to the third device in response to at least one of the following:a completion of the transiting,receiving an indication from the third device for requesting the current AI-related state of the first device, orreceiving an indication from the second device indicating the first device to transmit the current AI-related state of the first device to the third device.8.The first device of claim 1, wherein the first device is caused to:receive, from the second device or the third device, an indication indicative of a current AI-related state of the first device.9.The first device of claim 1, wherein the first device is in the first AI-related state, and the first device is caused to:in response to determining to transit to a second AI-related state, select the second device from more than one available second device based on at least one of the following:information about remaining AI-related resources of the more than one available second device,whether both an available second device and the first device are connected to the third device.10.The first device of claim 1, wherein the first message comprises at least one of the following:a command indicating a transit to the first AI-related state; ora confirmation confirming a request to transit to the first AI-related state transmitted by the first device.11.The first device of claim 1, wherein the first device is in the first AI-related state and the first device is caused to:transmit, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related states if:receive, from the second device, a message for paging the first device;there is need to perform an AI-related operation;receive, from the third device, a message for activating an AI-related operation at the first device; orthe first device is indicated to perform an AI-related operation.12.The first device of claim 11, wherein the connection establishment request comprises at least one of the following: amount of AI-related resources requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.13.The first device of claim 1, wherein the first device is in a second AI-related state and the first device is caused to:transmit, a connection release request for transiting to the first AI-related state of the plurality of AI-related states if:there is no need to perform an AI-related operation;receive, from the third device, a message for deactivating an AI-related operation at the first device; orthe first device is unable to perform an AI-related operation.14.The first device of claim 1, wherein the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.15.The first device of claim 1, wherein the first device is caused to:rather than receiving the first message from the second device, receive a second message from the third device, wherein the second message is used for transiting the first device to the first AI-related state of a plurality of AI-related states; andtransit to the first AI-related state in response to receiving the second message.16.The first device of claim 1, wherein,a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following:a second AI-related state being conditioned on an RRC or CM connected state,being in the first AI-related state if the first device is not in an RRC or CM connected state,transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state;transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state;transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; ortransiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; ora current AI-related state is separated from an RRC or CM state by at least one of the following:being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.17.The first device of claim 1, wherein,the first device comprises a terminal device or an access network device, orthe second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, orthe third device comprises a terminal device, an access network device or a core network entity.18.A second device comprising:a processor configured to cause the second device to:determine whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; andin accordance with a determination that the first transit condition is met, transmit, to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.19.The second device of claim 18, wherein the plurality of AI-related states comprise at least one of the following:a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state,a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state,a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state,a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, ora fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.20.The second device of claim 18, wherein the second first device is caused to:transmit, to the third device or the first device, an indication indicative of a current AI-related state of the first device.21.The second device of claim 20, wherein the second device is caused to:transmit indication indicative of a current AI-related state of the first device to the third device in response to at least one of the following:receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; orreceiving, from the third device, an indication for requesting the current AI-related state of the first device.22.The second device of claim 18, wherein the second device is caused to:transmit, to the first device, an indication indicating the first device to transmit a current AI-related state of the first device to the third device.23.The second device of claim 22, wherein the second device is caused to:transmit the indication indicating the first device to transmit the current AI-related state of the first device to the third device in response to at least one of the following:receiving, from the first device, an indication indicative of the current AI-related state of the first device or the completion of the transiting; orreceiving, from the third device, an indication for requesting the current AI-related state of the first device.24.The second device of claim 18, wherein the second device is caused to:receive, from the first device or the third device, an indication indicative of a current AI-related state of the first device.25.The second device of claim 18, wherein the second device is caused to:broadcast or multicast at least one of the following: an identity of the second device, or information about remaining AI-related resources of the second device.26.The second device of claim 18, wherein the first message comprises at least one of the following:a command indicating a transit to the first AI-related state; ora confirmation confirming a request to transit to the first AI-related state transmitted by the first device.27.The second device of claim 18, wherein the second device is caused to:determine to transit the first device to a second AI-related state in response to at least of the following:receiving, from the first device, a connection establishment request for transiting to a second AI-related state of the plurality of AI-related state;a remaining AI-related resource at the second device being above a threshold;a determination that there is a need to trigger the first device to perform an AI-related operation;receiving, from the third device, a message for activating an AI-related operation at the first device.28.The second device of claim 27, wherein,the connection establishment request comprises at least one of the following: AI-related resource requested by the first device, an identity of the first device, at least one use cases for which the first device requests an AI-related resource, or a cause of the connection establishment request.29.The second device of claim 18, wherein the second device is caused to:determine the first transit condition is met in response to at least of the following:receiving, from the first device, a connection release request for transiting to a first AI-related state of the plurality of AI-related state;a remaining AI-related resource at the second device being below a threshold;a determination that there is no need to trigger the first device to perform an AI-related operation; orreceiving, from the third device, a message for deactivating an AI-related operation at the first device.30.The second device of claim 29, wherein the connection release request comprises at least one of the following: a releasing cause, an identity of the first device, amount of AI-related resources to be released, or at least one use case for which the first device requests to release an AI-related resource.31.The second device of claim 18, wherein,a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following:a second AI-related state being conditioned on an RRC or CM connected state,being in the first AI-related state if the first device is not in an RRC or CM connected state,transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state;transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state;transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; ortransiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; ora current AI-related state is separated from an RRC or CM state by at least one of the following:being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.32.The second device of claim 18, wherein,the first device comprises a terminal device or an access network device, orthe second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, orthe third device comprises a terminal device, an access network device or a core network entity.33.A third device comprising:a processor configured to cause the third device to:establish a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; andreceive, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.34.The third device of claim 33, wherein the plurality of AI-related states comprise at least one of the following:a first AI-related state, wherein an AI-related resource is not available to the first device if the first device is in the first AI-related state,a second AI-related state, wherein an AI-related resource is available to the first device if the first device is in the second AI-related state,a third AI-related state, wherein a first set of AI-related operations is supported by the first device if the first device is in the third AI-related state, and the first set of AI-related operations is a subset of a second set of AI-related operations supported by the first device if the first device is in the second AI-related state,a fourth AI-related state, wherein the first device is known to the second device if the first device is in the fourth AI-related state, ora fifth AI-related state, wherein the first device is unknown to the second device if the first device is in the fifth AI-related state.35.The third device of claim 33, wherein the third device is caused to:deactivate an AI-related operation at the first device in response to the first device transiting to the first AI-related state, oractive an AI-related operation at the first device if the first device is in a second AI-related state.36.The third device of claim 33, wherein the third device is caused to:receive a second register request from the first device;transmit, to the first device, a second register response comprising a second identity allocated to the first device; andreceive, from the first device, an indication indicative of a first identity of the first device allocated by the second device, wherein the first identity and second identity are different from each other.37.The third device of claim 33, wherein the third device is caused to:receive, from the first device or the second device, an indication indicative of the current AI-related state of the first device.38.The third device of claim 37, wherein the third device is caused to:transmit, to the first device or the second device, an indication for requesting the current AI-related state of the first device.39.The third device of claim 33, wherein the third device is caused to:transmit, to the first device or the second device, an indication indicative of a current AI-related state of the first device.40.The third device of claim 33, wherein the third device is caused to:determine whether a first transit condition associated with a first AI-related state of a plurality of AI-related states is met for a first device; andin accordance with a determination that the first transit condition is met, transmit, to the first device, a second message for transiting the first device to the first AI-related state.41.The third first device of claim 33, wherein,a current AI-related state is dependent on a radio resource control (RRC) or connection management (CM) state by at least one of the following:a second AI-related state being conditioned on an RRC or CM connected state,being in the first AI-related state if the first device is not in an RRC or CM connected state,transiting to the first AI-related state in response to the first device transiting to an RRC or CM idle state;transiting to the first AI-related state in response to the first device transiting to in RRC or CM inactive state;transiting out of the second AI-related state in response to the first device transiting to an RRC or CM idle state; ortransiting out of the second AI-related state in response to the first device transiting to in RRC or CM inactive state; ora current AI-related state is separated from an RRC or CM state by at least one of the following:being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM connected state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in an RRC or CM idle state;being allowed to be either in the first AI-related state or in the second AI-related state if the first device is in RRC or CM inactive state.42.The third device of claim 33, wherein,the first device comprises a terminal device or an access network device, orthe second device comprises a server responsible for at least one of the following: AI-related model management, AI-related data management, or AI-related resource management, orthe third device comprises a terminal device, an access network device or a core network entity.43.A communication method implemented at a first device, comprising:receiving, from a second device that provide an artificial intelligence (AI) -related resource for the first device, a first message for transiting the first device to a first AI-related state of a plurality of AI-related states; andtransiting to the first AI-related state in response to receiving the first message, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.44.A communication method implemented at a second device, comprising:determining whether a first transit condition associated with a first artificial intelligence (AI) -related state of a plurality of AI-related states is met for a first device, wherein the second device provides an AI-related resource for the first device; andin accordance with a determination that the first transit condition is met, transmitting to the first device, a first message for transiting the first device to the first AI-related state, wherein a third device is informed of the transiting to the first AI-related state based on a completion of the transiting.45.A communication method implemented at a third device, comprising:establishing a connection with a first device, wherein the first device obtains an artificial intelligence (AI) -related resource from a second device; andreceiving, from the first device or the second device, a message that the first device has transited to a first AI-related state of a plurality of AI-related states.46.A computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method according to any of claims 43-45.
