Devices and methods of communication
By determining models based on network conditions and transmitting relevant information, the solution addresses the inconsistency between model training and inference, optimizing network assignments and ensuring effective AI/ML model management.
Patent Information
- Application Number
- PCT/CN2024/103307
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-07-03
- Publication Date
- 2026-01-08
AI Technical Summary
Ensuring consistency between model training and model inference in AI/ML models is unclear, particularly in network conditions.
A terminal device determines a set of models based on network conditions and transmits information to a network device, which generates a configuration for model inference to ensure consistency.
Optimizes network assignments and ensures consistency between model training and inference, allowing for effective AI/ML model management.
Smart Images

Figure CN2024103307_08012026_PF_FP_ABST
Abstract
Description
DEVICES AND METHODS OF COMMUNICATIONTECHNICAL FIELD
[0001] Embodiments of the present disclosure generally relate to the field of telecommunication, and in particular, to methods, devices and computer storage media of communication for management of an artificial intelligence (AI) / machine learning (ML) model.BACKGROUND
[0002] It is agreed that consistency between model training and model inference is beneficial from a performance perspective. However, how to ensure the consistency between model training and model inference is still unclear.SUMMARY
[0003] In general, embodiments of the present disclosure provide methods, devices and computer storage media of communication for model management.
[0004] In a first aspect, there is provided a terminal device. The terminal device comprises a processor configured to cause the terminal device to: determine a set of models based at least on a list of configurations for data collection, a configuration in the list of configurations corresponding to an indication associated with a network condition; determine, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions; and transmit the first information to a network device.
[0005] In a second aspect, there is provided a network device. The network device comprises a processor configured to cause the network device to: receive, from a terminal device, first information indicating an assumption of the terminal device for consistency among network conditions; and transmit, to the terminal device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.
[0006] In a third aspect, there is provided a method of communication. The method comprises: determining, at a terminal device, a set of models based at least on a list of configurations for data collection, a configuration in the list of configurations corresponding to an indication associated with a network condition; determining, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions; and transmitting the first information to a network device.
[0007] In a fourth aspect, there is provided a method of communication. The method comprises: receiving, at a network device and from a terminal device, first information indicating an assumption of the terminal device for consistency among network conditions; and transmitting, to the terminal device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.
[0008] In a fifth aspect, there is provided a computer readable medium having instructions stored thereon. The instructions, when executed on at least one processor, cause the at least one processor to perform the method according to third or fourth aspect of the present disclosure.
[0009] Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Through the more detailed description of some 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:
[0011] FIG. 1 illustrates an example communication network in which some embodiments of the present disclosure can be implemented;
[0012] FIG. 2 illustrates a signaling chart illustrating a process of communication according to some embodiments of the present disclosure;
[0013] FIG. 3A illustrates a signaling chart illustrating an example process of model training and feedback according to some embodiments of the present disclosure;
[0014] FIG. 3B illustrates a schematic diagram illustrating an example model development according to some embodiments of the present disclosure;
[0015] FIG. 4A illustrates a signaling chart illustrating another example process of model training and feedback according to some embodiments of the present disclosure;
[0016] FIG. 4B illustrates a schematic diagram illustrating another example model development according to some embodiments of the present disclosure;
[0017] FIG. 5A illustrates a signaling chart illustrating still another example process of model training and feedback according to some embodiments of the present disclosure;
[0018] FIG. 5B illustrates a schematic diagram illustrating still another example model development according to some embodiments of the present disclosure;
[0019] FIG. 6 illustrates a flowchart of a method of communication implemented at a terminal device in accordance with some embodiments of the present disclosure;
[0020] FIG. 7 illustrates a flowchart of a method of communication implemented at a network device in accordance with some embodiments of the present disclosure; and
[0021] FIG. 8 is a simplified block diagram of a device that is suitable for implementing embodiments of the present disclosure.
[0022] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION
[0023] Principle of the present disclosure will now be described with reference to some 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 limitations as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones described below.
[0024] 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.
[0025] 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, device on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure / network, devices for integrated access and backhaul (IAB) , small data transmission (SDT) , mobility, multicast and broadcast services (MBS) , positioning, dynamic / flexible duplex in commercial networks, reduced capability (RedCap) , 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 has ‘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.
[0026] 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) , Network-controlled Repeaters, and the like.
[0027] The terminal device or the network device may have AI / ML 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.
[0028] The terminal or the network device may work on several frequency ranges, e.g. FR1 (410 MHz to 7125 MHz) , FR2 (24.25GHz to 71GHz) , frequency band larger than 100GHz as well as Tera Hertz (THz) . It can further work on licensed / unlicensed / shared spectrum. The terminal device may have more than one connections with the network devices under MR-DC application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.
[0029] The network device may have the function of network energy saving, self-organizing networks (SON) / minimization of drive tests (MDT) . The terminal may have the function of power saving.
[0030] 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.
[0031] In one embodiment, 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 one embodiment, the first network device may be a first RAT device and the second network device may be a second RAT device. In one embodiment, 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 one embodiment, information A may be transmitted to the terminal device from the first network device and information B may be transmitted to the terminal device from the second network device directly or via the first network device. In one embodiment, 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.
[0032] 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’ a re 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. The term ‘and / or’ indicates that there may be three relationships. For example, A and / or B may indicate cases includes ‘only A’ , ‘both A and B’ , and ‘only B’ . The term ‘at least one of the following items’ or a similar expression thereof refers to any combination of these items, including any combination of a single item or a plurality of items. For example, ‘at least one of A, B, or C’ may represent A, B, C, ‘A and B’ , ‘A and C’ , ‘B and C’ , or ‘A, B and C’ . Other definitions, explicit and implicit, may be included below.
[0033] 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.
[0034] Embodiments of the present disclosure provide a solution of communication for model management. In the solution, a terminal device may determine a set of models based at least on a list of configurations for data collection. A configuration in the list of configurations may correspond to an indication associated with a network condition. Based on the set of models, the terminal device may determine first information indicating an assumption of the terminal device for consistency among network conditions, and transmit the first information to a network device. Based on the first information, the network device may generate a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition, and transmit the third configuration to the terminal device. In this way, a network (NW) may be aware of whether a change of a network condition has impact on a model. Accordingly, NW assignment on the network condition may be optimized, and consistency between model training and model inference may be ensured.
[0035] For convenience, definitions of some terms in the present disclosure may be listed as below.
[0036] ·AI / ML-enabled feature: a feature where AI / ML may be used.
[0037] ·AI / ML Model: a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs.
[0038] ·AI / ML model delivery: a generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. An entity may mean a network node / function (e.g., gNB, location management function (LMF) , etc. ) , UE, proprietary server, etc.
[0039] ·AI / ML model inference: a process of using a trained AI / ML model to produce a set of outputs based on a set of inputs.
[0040] ·AI / ML model testing: a sub-process of training, to evaluate 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.
[0041] ·AI / ML model training: a process to train an AI / ML model (e.g., by learning an input / output relationship) in a data driven manner and obtain the trained AI / ML model for inference.
[0042] ·AI / ML model transfer: a delivery of an AI / ML model over an air interface in a manner that is not transparent to the third generation partnership project (3GPP) signalling, either parameters of a model structure known at a receiving end or a new model with parameters. The delivery may contain a full model or a partial model.
[0043] ·AI / ML model validation: a sub-process of training, to evaluate quality of an AI / ML model using a dataset different from one used for model training, that helps selecting model parameters that generalize beyond a dataset used for model training.
[0044] ·data collection: a process of collecting data by a network node, management entity, or UE for the purpose of AI / ML model training, data analytics and inference.
[0045] ·federated learning / federated training: 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.
[0046] ·functionality identification: a process / method of identifying an AI / ML functionality for the common understanding between NW and UE. It is to be noted that information regarding AI / ML functionality may be shared during functionality identification. AI / ML functionality resides depends on the specific use cases and sub use cases.
[0047] ·management instruction: information needed to ensure proper inference operation. This information may include selection / (de) activation / switching of AI / ML models or AI / ML functionalities, fallback to non-AI / ML operation, etc.
[0048] ·model activation: enable an AI / ML model for a specific AI / ML-enabled feature.
[0049] ·model deactivation: disable an AI / ML model for a specific AI / ML-enabled feature.
[0050] ·model download: model transfer from the network to UE.
[0051] ·model identification: a process / method of identifying an AI / ML model for the common understanding between the NW and the UE. The process / method of model identification may or may not be applicable. Information regarding the AI / ML model may be shared during model identification.
[0052] ·model monitoring: a procedure that monitors inference performance of an AI / ML model.
[0053] ·model parameter update: a process of updating model parameters of a model.
[0054] ·model selection: a process of selecting an AI / ML model for activation among multiple models for the same AI / ML enabled feature. Model selection may or may not be carried out simultaneously with model activation.
[0055] ·model switching: a process of deactivating a currently active AI / ML model and activating a different AI / ML model for a specific AI / ML-enabled feature.
[0056] ·model update: a process of updating model parameters and / or model structure of a model.
[0057] ·model upload: model transfer from UE to the network.
[0058] ·network-side (AI / ML) model: an AI / ML model whose inference is performed entirely at the network.
[0059] ·offline field data: data collected from field and used for offline training of an AI / ML model.
[0060] ·offline training: 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.
[0061] ·online field data: data collected from field and used for online training of the AI / ML model.
[0062] ·online training: 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. The notion of (near) real-time vs. non real-time is context-dependent and is relative to the inference time-scale. 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. Fine-tuning / re-training may be done via online or offline training.
[0063] ·reinforcement learning (RL) : a process of training an AI / ML model from an input (also known as a state) and a feedback signal (also known as a reward) resulting from the model’s output (also known as an action) in an environment the model is interacting with.
[0064] ·semi-supervised learning: a process of training a model with a mix of labelled data and unlabelled data.
[0065] ·supervised learning: a process of training a model from input and its corresponding labels.
[0066] ·test encoder / decoder for tester (TE) : AI / ML model for UE encoder / gNB decoder implemented by TE.
[0067] ·two-sided (AI / ML) model: 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.
[0068] ·UE-side (AI / ML) model: an AI / ML Model whose inference is performed entirely at the UE.
[0069] ·unsupervised learning: a process of training a model without labelled data.
[0070] ·proprietary-format models: 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.
[0071] ·open-format models: 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.
[0072] In the context of the present disclosure, for time domain beam prediction, time domain DL beam prediction for a set of beams (also called as Set A herein) may be performed based on historic measurement results of another set of beams (also called as Set B herein) . The model inputs may be measurement results of K (K≥1) latest measurement instances, which may be in following form (s) : only layer 1-reference signal received power (L1-RSRP) measurement based on Set B; L1-RSRP measurement based on Set B and assistance information; L1-RSRP measurement based on Set B and the corresponding DL Tx and / or Rx beam identity (ID) . The model outputs may be predictions of F (F≥1) future time instances, which may be in following form (s) : Tx and / or Rx Beam ID (s) and / or the predicted L1-RSRP of the N predicted DL Tx and / or Rx beams, e.g., N predicted beams can be the top-N predicted beams; Tx and / or Rx Beam ID (s) of the N predicted DL Tx and / or Rx beams and other information, e.g., N predicted beams can be the top-N predicted beams; Tx and / or Rx Beam angle (s) and / or the predicted L1-RSRP of the N predicted DL Tx and / or Rx beams, e.g., N predicted beams can be the top-N predicted beams.
[0073] In the context of the present disclosure, the terms ‘model’ , ‘functionality’ and ‘model / functionality’ may be used interchangeably. The terms ‘model’ and ‘model group’ may be used interchangeably. The terms ‘functionality’ , ‘functionality group’ , ‘functionality set’ may be used interchangeably. The terms ‘ID’ , ‘index’ , ‘indicator’ and ‘identifier’ may be used interchangeably. The term ‘NW’ herein may refer to ‘operations, administration and maintenance (OAM) ’ , ‘server’ , or ‘advanced mobile location (AML) / LMF’ .
[0074] In the context of the present disclosure, the term ‘conditions’ may refer to configurations supported indicated via UE capability reporting, related to model training, model inference, performance monitoring, validation procedure, fallback, of an AI / ML model / functionality or a group of models / functionalities. The term ‘additional conditions’ may refer to, e.g., application conditions, scenarios, sites, datasets, cell identity (ID) , timestamp, beam shape, etc. For an AI / ML-enabled feature / feature group (FG) , additional conditions refer to any aspects that are assumed for a training of a 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-side additional conditions. Here, the term ‘NW-side additional conditions’ may be interchangeably used with ‘network conditions’ .
[0075] The term ‘application conditions’ may refer to a signal-to-noise ratio (SNR) , line of sight (LOS) / non-line of sight (NLOS) , a channel condition, etc. The term ‘UE internal conditions’ may refer to, e.g., memory, battery, computation resource, overheating and other hardware limitations on functionality or model operations. The term ‘complexity or processing capability’ may refer to Tera operation per second (TOPs) , floating-point operation per second (FLOPs) , multiply-accumulates (MACs) , or number of parameters and / or a size.
[0076] In the context of the present disclosure, the terms ‘beam’ , ‘precoder’ , ‘precoding’ , ‘precoding matrix’ , ‘beam’ , ‘spatial relation information’ , ‘spatial relation information’ , ‘precoding information’ , ‘precoding information and number of layers’ , ‘precoding matrix indicator (PMI) ’ , ‘precoding matrix indicator’ , ‘transmission precoding matrix indication’ , ‘precoding matrix indication’ , ‘transmission configuration indication (TCI) state’ , ‘UL TCI state’ , ‘joint TCI state’ , ‘transmission configuration indicator’ , ‘quasi co-location (QCL) ’ , ‘quasi-co-location’ , ‘QCL parameter’ , ‘QCL assumption’ , ‘QCL relationship’ and ‘spatial relation’ may be used interchangeably.
[0077] A beam may refer to a downlink beam, uplink beam, transmit beam, receive beam, beam pair, reference signal (RS) resource, RS resource set, antenna port, antenna port group, antenna element (s) , antenna array (s) , beam group.
[0078] The terms ‘reference signal received power (RSRP) ’ , ‘L1-RSRP’ ‘layer 3 (L3) -RSRP’ , ‘filtered RSRP’ may be used interchangeably. If ‘RSRP’ is used as a beam quality metric, the methods are readily extended to other metrics like ‘signal to interference and noise ratio (SINR) ’ , ‘reference signal received quality (RSRQ) ’ , ‘received signal strength indicator (RSSI) ’ , etc.
[0079] As used herein, a model may be equivalent to at least one of the following: an AI / ML model, a ML model, an AI model, a data-driven model, a data processing model, an algorithm, a functionality, a procedure, a process, an entity, a function, a feature, a feature group, a model ID, an ID, a functionality ID, a configuration ID, a scenario ID, a site ID, or a dataset ID. As a result, the above terms may be used interchangeably.
[0080] In some embodiments, the model may be represented by or associated with a channel, a resource, a resource set, a RS resource, a RS resource set, a RS port, a set of RS ports, a RS port ID, or a set of RS port IDs.
[0081] In some embodiments, the model may comprise a set of weights values that may be learned during training, for example for a specific architecture or configuration, where a set of weights values may also be called a parameter set.
[0082] In some embodiments, the model may be used to predict a target cell, or measurements of a set of beams of a set of candidate cells in future based on at least historical measurements (e.g., L1-RSRP, L1-SINR) of a set of beams of a set of candidate cells.
[0083] In some embodiments, an input of the ML model (i.e., AI input) may refer to the input of a model and indicate data inputted into the model, which may be equivalent to data.
[0084] In some embodiments, an output of ML model (i.e., AI output) may refers to the output of a model and indicate result (s) outputted by the model, which is equivalent to label / data.
[0085] In some embodiments, a ground truth label of data (or ground-truth label) for monitoring or training the ML model (i.e., AI output) may refers to the authoritative, accepted data, or true answer or outcome for AI / ML model. In some embodiments, ‘ground truth’ , ‘ground truth label’ , ‘ground truth label of data’ , ‘input label’ , ‘input data’ and ‘data’ can be used interchangeably. In some embodiments, the ground truth can be interpreted as actual / factual (i.e. actual / factual measured) data / values / results / collections / parameters, which can be used as reference, compared to prediction or inference.
[0086] Principles and implementations of the present disclosure will be described in detail below with reference to the figures.
[0087] EXAMPLE OF COMMUNICATION NETWORK
[0088] FIG. 1 illustrates a schematic diagram of an example communication network 100 in which some embodiments of the present disclosure can be implemented. As shown in FIG. 1, the communication network 100 may include a terminal device 110 and network devices 120 and 130. In some embodiments, each of the network devices 120 and 130 may provide one or more cells to serve the terminal device 110. In the example of FIG. 1, the network device 120 provides a cell 121, and the network device 130 provides a cell 131. The terminal device 110 is located in the cell 121 and served by the network device 120.
[0089] The terminal device 110 may have a plurality of beams (not shown) , and each of the network devices 120 and 130 may have a plurality of beams (not shown) . A channel (or called as a sub-channel in this case) may be formed between one of the plurality of beams of the terminal device 110 and one of the plurality of beams of the network device 120 or 130. The terminal device 110 may transmit information to the network device 120 or 130, or receive information from the network device 120 or 130 via one or more sub-channels.
[0090] It is to be understood that the number of devices in FIG. 1 is given for the purpose of illustration without suggesting any limitations to the present disclosure. The communication network 100 may include any suitable number of network devices and / or terminal devices and / or other network elements adapted for implementing implementations of the present disclosure.
[0091] As shown in FIG. 1, the terminal device 110 and each of the network devices 120 and 130 may communicate with each other via Uu interface. The network device 120 and the network device 130 may communication with each other via Xn interface. Communication in a direction from the terminal device 110 towards the network device 120 or 130 is referred to as uplink (UL) communication, while communication in a reverse direction from the network device 120 or 130 towards the terminal device 110 is referred to as downlink (DL) communication. A wireless communication channel may comprise 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) .
[0092] The communications in the communication network 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.
[0093] Currently, a concept of an associated ID is proposed to ensure the consistency between model training and model inference. In general, the associated ID is used to reflect NW-side additional conditions or NW implementations which NW is not intended to disclose or standardize. However, how to apply the concept of the associated ID in model management is still unclear.
[0094] Thus, embodiments of the present disclosure provide a solution of communication for model management so as to ensure the consistency between model training and model inference. The solution will be described in detail with reference to FIGs. 2 to 5B below.
[0095] EXAMPLE IMPLEMENTATION OF MODEL MANAGEMENT
[0096] FIG. 2 illustrates a signaling chart illustrating a process 200 of communication according to some embodiments of the present disclosure. For the purpose of discussion, the process 200 will be described with reference to FIG. 1. The process 200 may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 2 are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that the terminal device 110 is served by the network device 120.
[0097] As shown in FIG. 2, the terminal device 110 may transmit 210, to the network device 120, information (for convenience, also referred to as second information herein) indicating capability of the terminal device 110. In some embodiments, the second information may be carried in UE assistance information (UAI) .
[0098] In some embodiments, the second information may comprise number of indications associated with network conditions supported by the terminal device 110. In some embodiments, the second information may comprise the maximum number of indications associated with network conditions supported by the terminal device 110. In some embodiments, the indications associated with network conditions may be IDs or indexes associated with network conditions. Of course, any other suitable forms may also be feasible.
[0099] In some embodiments, the second information may comprise number of indications associated with network conditions supported by the terminal device 110 in a cell. In some embodiments, the second information may comprise the maximum number of indications associated with network conditions supported by the terminal device 110 in the cell.
[0100] In some embodiments, the second information may comprise number of indications associated with network conditions supported by the terminal device 110 for a model (e.g., per model or functionality) . In some embodiments, the second information may comprise the maximum number of indications associated with network conditions supported by the terminal device 110 for the model.
[0101] In some embodiments, the second information may comprise number of indications associated with network conditions supported by the terminal device for a model (e.g., per model or functionality) in a cell. In some embodiments, the second information may comprise the maximum number of indications associated with network conditions supported by the terminal device 110 for the model in the cell.
[0102] In some embodiments, the second information may comprise number of models for an indication associated with a network condition supported by the terminal device 110. In some embodiments, the second information may comprise the maximum number of models for an indication associated with a network condition supported by the terminal device 110.
[0103] In some embodiments, the second information may comprise number of models for an indication associated with a network condition supported by the terminal device 110 in a cell. In some embodiments, the second information may comprise the maximum number of models for an indication associated with a network condition supported by the terminal device 110 in the cell.
[0104] In some embodiments, the second information may comprise number of models maintained at the terminal device 110.
[0105] In some embodiments, the second information may comprise an indication of whether the terminal device 110 supports a simultaneous configuration or a sequential configuration for multiple indications associated with network conditions.
[0106] In some embodiments, the second information may comprise an indication of whether the terminal device 110 supports an assumption (for convenience, also referred to as an inter-cell assumption herein) for consistency among network conditions from different cells.
[0107] It is to be understood that the second information may comprise any other suitable capability information or any combination of the above capability information.
[0108] Continuing to refer to FIG. 2, the terminal device 110 may transmit 220, to the network device 120, information (for convenience, also referred to as first information herein) indicating an assumption of the terminal device 110 for consistency among network conditions.
[0109] With reference to FIG. 2, the terminal device 110 may determine 221 a set of models (i.e., one or more models) based at least on a list of configurations for data collection. The list of configurations may comprise one or more configurations for data collection. In some embodiments, the network device 120 may transmit the list of configurations to the terminal device 110, e.g., via a RRC signaling or any other suitable ways. In some embodiments, the network device 120 may simultaneously configure two or more configurations for data collection to the terminal device 110. In some embodiments, the network device 120 may sequentially configure the two or more configurations for data collection.
[0110] In some embodiments, the terminal device 110 may perform model training based on the one or more configurations for data collection to determine the set of models. In some embodiments, the terminal device 110 may perform model training based on the one or more configurations for data collection and a reference model to determine the set of models. The reference model may be determined in any suitable ways, and the present disclosure does not limit this aspect.
[0111] Continuing to refer to FIG. 2, the terminal device 110 may determine 222, based on the set of models, the first information indicating the assumption of the terminal device 110 for the consistency among network conditions. In some embodiments, the first information may indicate the assumption of the terminal device 110 for the consistency among network conditions during model training. In some embodiments, the first information may indicate the assumption of the terminal device 110 for the consistency among network conditions across model training and model inference. In some embodiments, the first information may indicate the assumption of the terminal device 110 for the consistency among network conditions across two or more of the following: model training, model inference or performance monitoring.
[0112] As shown in FIG. 2, the terminal device 110 may transmit 223 the first information to the network device 120. That is, the terminal device 110 may transmit, to the network device 120, feedback information regarding whether the terminal device 110 assumes the consistency among network conditions.
[0113] For illustration, some example embodiments will be described in connection with Embodiments 1 to 3 below.
[0114] Embodiment 1
[0115] In this embodiment, configurations in the list of configurations are configured simultaneously, and the set of models is determined based on the configurations in the list of configurations. For illustration, it is assumed that the list of configurations comprises a first configuration for data collection and a second configuration for data collection. It is to be noted that more than two configurations for data collection may also be feasible. More details will be described in connection with FIGs. 3A and 3B below.
[0116] FIG. 3A illustrates a signaling chart illustrating an example process 300A of model training and feedback according to some embodiments of the present disclosure. For the purpose of discussion, the process 300A will be described with reference to FIG. 1. The process 300A may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 3A are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that the terminal device 110 is served by the network device 120.
[0117] As shown in FIG. 3A, the network device 120 may transmit 310, to the terminal device 110, a first configuration for data collection and a second configuration for data collection. The first configuration may correspond to a first indication associated with a first network condition, and the second configuration may correspond to a second indication associated with a second network condition.
[0118] In some embodiments, the first and second configurations may be data collection related configurations. In some embodiments, the first and second configurations may be data collection related configurations for one or more of the following: model training, model inference or performance monitoring. In some embodiments, the first and second configurations may be a report configuration, resource configuration or measurement configuration for data collection. In some embodiments, for a beam management use case, an indication associated with a network condition may be provided as one or more IEs within a channel status information (CSI) report configuration, CSI resource configuration or CSI measurement configuration.
[0119] In some embodiments, the first and second configurations may correspond to different network conditions. In some embodiments, for a beam management use case, network conditions may include an antenna layout, beam pattern, transmitting power, Set A / Set B relationship, etc. In some embodiments, the Set A / Set B relationship may comprise Set A / Set B size, an ordering of resources (e.g., resource index consistency) for Set B beams and Set A beams, a QCL relationship of Set A beams with respect to Set B beams, a relative pointing direction and beam-width difference between physical beams with respect to Set A and Set B resources. NW may transmit RSs based on configurations utilizing those different network conditions.
[0120] In some embodiments, the first and second configurations may be configured simultaneously. In this case, time required for model training and model inference may be reduced although NW may adjust its network conditions frequently.
[0121] With reference to FIG. 3A, the terminal device 110 may collect 320 data corresponding to the first and second configurations and the first and second indications. For example, the data may be a measurement result of an RS based on at least one of an RS report configuration or RS resource configuration or RS measurement configuration. Alternatively, the data may be obtained from a dataset transferred to the terminal device 110 corresponding to an indication associated with a network condition. The dataset may be associated with a dataset ID.
[0122] In some embodiments, the terminal device 110 may determine a first set of data based on the first configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the first configuration, and determine the first set of data from the measurement results. In some embodiments, the terminal device 110 may determine the first set of data based on the first indication. For example, the terminal device 110 may determine a dataset corresponding to the first indication, and determine the first set of data from the dataset.
[0123] In some embodiments, the terminal device 110 may determine a second set of data based on the second configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the second configuration, and determine the second set of data from the measurement results. In some embodiments, the terminal device 110 may determine a second set of data based on the second indication. For example, the terminal device 110 may determine a dataset corresponding to the second indication, and determine the second set of data from the dataset.
[0124] It is to be noted that a data collection entity and a model development (training or update) entity may be UE, or UE side, or UE side server (e.g., over the top (OTT) server, or third party server) . The present disclosure does not limit this physical entity.
[0125] With reference to FIG. 3A, the terminal device 110 may determine 330 the set of models based on at least one of the first set of data or the second set of data. In other words, the terminal device 110 may develop (e.g., train or update) an AI / ML model based on the data collected respectively or jointly. For illustration, an example embodiment will be described in connection with FIG. 3B.
[0126] FIG. 3B illustrates a schematic diagram 300B illustrating an example model development according to some embodiments of the present disclosure. As shown in FIG. 3B, a first model 301 may be trained based on a first set of data 302. A second model 303 may be trained based on a second set of data 304. A third model 305 may be trained based on more than one set of data (e.g., the first set of data 302 and the second set of data 304) . Other one or more models (or versions or parameters) 306 may also be possible, e.g., with part of the first set of data 302 and the second set of data 304. It may be up to the terminal device 110 to develop the set of models based on the collected data.
[0127] As shown in FIG. 3B, the terminal device 110 may perform 307 a check and feedback, e.g., by checking performance of the developed models 301, 303, 305 and 306 (or versions or parameters) . The terminal device 110 may check and feedback whether different network conditions have impact on the set of models. e.g., by checking the performance and / or cost of the models (or versions, parameters) developed and decide whether to use separate models or one general model to handle the different network conditions. How to check and / or make the decision may be up to the terminal device 110.
[0128] Refer back to FIG. 3A, the terminal device 110 may determine and transmit 340 the first information indicating the assumption of the terminal device 110 for the consistency among network conditions.
[0129] In some embodiments, the first information may comprise an indication of whether the terminal device 110 assumes that the first network condition and the second network condition are consistent. In some embodiments, the terminal device 110 may signal a list of indications associated with network conditions, e.g., {UE-assumed consistency: indication#1, indication#2, …} . It is to be noted that UE-assumed consistency does not mean real consistency in NW implementation, and basically means that UE apply the same model for those indications.
[0130] In some embodiments, the first information may comprise an indication indicating that a group of indications of network conditions comprising at least the first network condition and the second network condition are handled jointly. For example, the terminal device 110 may signal a group ID and a list of indications associated with network conditions, e.g., {group ID#1: indication#1, indication#2, …} . For indications within a group, the terminal device 110 assumes that network conditions corresponding to the indications are consistent.
[0131] In some embodiments, the first information may comprise a set of model IDs for the group of indications associated with network conditions. In some embodiments, a model ID in the set of model IDs and an indication in the group of indications have a one-to-one mapping relationship. In some embodiments, a model ID in the set of model IDs may be mapped to two or more indications in the group of indications. For example, {model ID#1: indication#1, indication#2, …} . In some embodiments, an indication in the group of indications may be mapped to two or more model IDs in the set of model IDs. In some embodiments, a model ID in the set of model IDs may be mapped to one or more indications in the group of indications, and an indication in the group of indications may be mapped to one or more model IDs in the set of model IDs.
[0132] In some embodiments, the first information may comprise an indication of whether the first indication and the second indication are needed. In some embodiments, the first information may comprise an indication of whether the first indication and the second indication are needed for one or more of the following: model training, model inference or performance monitoring. In some embodiments, the first information may comprise an indication of whether the first indication and the second indication are needed for data collection configuration for one or more of the following: model training, model inference or performance monitoring.
[0133] In some embodiments, the first information may comprise information of the set of models. In some embodiments, the information of the set of models may comprise information of an input of a model in the set of models. In some embodiments, the information of the set of models may comprise information of an output of the model. In some embodiments, the information of the set of models may comprise time required for model inference of the model. It is to be noted that any combinations of the above information may also be feasible.
[0134] In some embodiments, the information of the set of models may be associated with the first indication or the second indication. In other words, the information of the set of models may be reported per indication of a network condition.
[0135] In some embodiments, the information of the set of models may be associated with a model ID in the set of model ID. In other words, the information of the set of models may be reported per model ID.
[0136] In some embodiments, the information of the set of models may be associated with a group ID associated with the group of indications associated with network conditions. In other words, the information of the set of models may be reported per group ID.
[0137] In some embodiments, for a beam management use case, the information of the set of models may comprise at least one of the following: number of reference signals in a first set of reference signals for measurement (i.e., Set A size) ; number of reference signals in a second set of reference signals for prediction (i.e., Set B size) ; a mapping between the first set of reference signals and the second set of reference signals; a beam pattern for at least one of the first set of reference signals or the second set of reference signals (if the network device 120 provides multiple patterns for selection, the terminal device 110 may feedback a pattern ID); number of predicted results (e.g., number of RSRPs) ; a size of an observation window for the measurement (i.e., number of historical measurements or historical instances needed for model inference) ; or a size of a prediction window for the prediction (i.e., number of future instances obtained based on the output of model inference) .
[0138] In some embodiments, the terminal device 110 may transmit the first information together with information indicating that the model training is completed. In some embodiments, the terminal device 110 may transmit the first information together with information for model registration and / or model identification.
[0139] In some embodiments, the terminal device 110 may transmit the first information via a RRC signaling or a medium access control control element (MAC CE) or uplink control information (UCI) . In some embodiments, the terminal device 110 may transmit the first information via a UE capability reporting, UE capability reporting update, dynamic UE capability reporting, or UE assistance information.
[0140] In some embodiments, the terminal device 110 may transmit the first information in an event trigger way (where the event may be defined as that the model training is completed) . In some embodiments, the terminal device 110 may transmit the first information periodically or semi-persistently. In some embodiments, the terminal device 110 may transmit the first information aperiodic triggered by the network device 120.
[0141] Embodiment 2
[0142] In this embodiment, configurations in the list of configurations are configured sequentially, and the set of models is determined based on the configurations in the list of configurations or based on a reference model and the configurations. For illustration, it is assumed that the list of configurations comprises a first configuration for data collection and a second configuration for data collection. It is to be noted that more than two configurations for data collection may also be feasible. More details will be described in connection with FIGs. 4A and 4B below.
[0143] FIG. 4A illustrates a signaling chart illustrating another example process 400A of model training and feedback according to some embodiments of the present disclosure. For the purpose of discussion, the process 400A will be described with reference to FIG. 1. The process 400A may involve the terminal device 110 and the network device 120 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 4A are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added. It is assumed that the terminal device 110 is served by the network device 120.
[0144] As shown in FIG. 4A, the terminal device 110 may determine 410 a reference model. In some embodiments, the network device 120 may transmit 411, to the terminal device 110, at least a first configuration for data collection. In some embodiments, the first configuration may correspond to a first indication associated with a first network condition. In some embodiments, the first configuration may not correspond to the first indication associated with the first network condition.
[0145] With reference to FIG. 4A, the terminal device 110 may collect 412 data corresponding to at least the first configuration and the first indication if any. For example, the data may be a measurement result of an RS based on at least one of an RS report configuration or RS resource configuration or RS measurement configuration. Alternatively, the data may be obtained from a dataset transferred to the terminal device 110 corresponding to the first indication. The dataset may be associated with a dataset ID.
[0146] In some embodiments, the terminal device 110 may determine at least a first set of data based on the first configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the first configuration, and determine the first set of data from the measurement results. In some embodiments, the terminal device 110 may determine at least the first set of data based on the first indication. For example, the terminal device 110 may determine a dataset corresponding to the first indication, and determine the first set of data from the dataset.
[0147] With reference to FIG. 4A, the terminal device 110 may determine 413 the reference model by model training based on at least the first set of data. In other words, the terminal device 110 may develop (e.g., train or update) the reference model based on at least the first set of data. It is to be noted that the reference model may be determined in any other suitable ways and the present disclosure does not limit this aspect.
[0148] As shown in FIG. 4A, in some embodiments, the terminal device 110 may transmit 420, to the network device 120, information indicating that the model training is completed for the first configuration. Based on this information, the network device 120 may provide reconfiguration (e.g., a second configuration as described below) .
[0149] Continuing to refer to FIG. 4A, the network device 120 may transmit 430, to the terminal device 110, the second configuration for data collection. The second configuration may correspond to a second indication associated with a second network condition. In some embodiments, if the first indication associated with the first network condition is not provided for the first configuration, the second indication may be reflected as information that whether the second network condition is the same as the first network condition.
[0150] With reference to FIG. 4A, the terminal device 110 may collect 440 data corresponding to the second configuration and the second indication. For example, the data may be a measurement result of an RS based on at least one of an RS report configuration or RS resource configuration or RS measurement configuration. Alternatively, the data may be obtained from a dataset transferred to the terminal device 110 corresponding to the second indication. The dataset may be associated with a dataset ID.
[0151] In some embodiments, the terminal device 110 may determine a second set of data based on the second configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the second configuration, and determine the second set of data from the measurement results. In some embodiments, the terminal device 110 may determine the second set of data based on the second indication. For example, the terminal device 110 may determine a dataset corresponding to the second indication, and determine the second set of data from the dataset.
[0152] With reference to FIG. 4A, the terminal device 110 may determine 450 the set of models based on at least one of the reference model or the second set of data. In other words, the terminal device 110 may develop (e.g., train or update) an AI / ML model based on the data collected respectively or jointly. For illustration, an example embodiment will be described in connection with FIG. 4B.
[0153] FIG. 4B illustrates a schematic diagram 400B illustrating another example model development according to some embodiments of the present disclosure. As shown in FIG. 4B, a first model 401 may be a reference model 402. In other words, the reference model may be still applicable for the second network conditions of the second configuration. A second model 403 may be trained based on a second set of data 404. A third model 405 may be trained based on the reference model 402 and the second set of data 404. The third model 405 may be suitable for both the first network condition associated with the reference model 402 and the second network condition associated with the second configuration. Other one or more models (or versions or parameters) 406 may also be possible, e.g., with part of the reference model 402 and the second set of data 404. It may be up to the terminal device 110 to develop the set of models based on the collected data.
[0154] As shown in FIG. 4B, the terminal device 110 may perform 407 a check and feedback, e.g., by checking performance of the developed models 401, 403, 405 and 406 (or versions or parameters) . The terminal device 110 may check and feedback whether different network conditions have impact on the set of models, e.g., by checking the performance and / or cost of the models (or versions, parameters) developed and decide whether to use separate models or one general model to handle the different network conditions. How to check and / or make the decision may be up to the terminal device 110.
[0155] Refer back to FIG. 4A, the terminal device 110 may determine and transmit 460 the first information indicating the assumption of the terminal device 110 for the consistency among network conditions.
[0156] In some embodiments, the first information may comprise an indication of whether the terminal device 110 assumes that the second network condition associated with the second indication and a network condition associated with the reference model are consistent. In some embodiments, the terminal device 110 may signal a list of indications associated with network conditions, e.g., {UE-assumed consistency: indication#1, indication#2, …} . It is to be noted that UE-assumed consistency does not mean real consistency in NW implementation, and basically means that UE apply the same model for those indications.
[0157] In some embodiments, the first information may comprise an indication of whether the terminal device assumes that the second network condition associated with the second configuration and a network condition associated with a configuration for data collection of the reference model are consistent, e.g., if the first indication is not provided for the first configuration. For example, the terminal device 110 may signal a list of configuration IDs, e.g., {UE-assumed consistency: configuration ID#1, configuration ID#2, …} . It is to be noted that UE-assumed consistency does not mean real consistency in NW implementation, and basically means that UE apply the same model.
[0158] In some embodiments, the first information may comprise an indication of whether the second indication is needed. In some embodiments, the first information may comprise an indication of whether the second indication is needed for one or more of the following: model training, model inference or performance monitoring. In some embodiments, the first information may comprise an indication of whether the second indication is needed for data collection configuration for one or more of the following: model training, model inference or performance monitoring.
[0159] In some embodiments, the first information may comprise a set of model IDs for the group of indications associated with network conditions. In some embodiments, a model ID in the set of model IDs and an indication in the group of indications have a one-to-one mapping relationship. In some embodiments, a model ID in the set of model IDs may be mapped to two or more indications in the group of indications. For example, {model ID#1: indication#1, indication#2, …} . In some embodiments, an indication in the group of indications may be mapped to two or more model IDs in the set of model IDs. In some embodiments, a model ID in the set of model IDs may be mapped to one or more indications in the group of indications, and an indication in the group of indications may be mapped to one or more model IDs in the set of model IDs.
[0160] In some embodiments, the first information may comprise an indication of whether the reference model is valid. In some embodiments, considering that the terminal device 110 may have limited capabilities to maintain multiple AI / ML models and the terminal device 110 may release the previous developed models, the first information may include information of whether the first indication or the reference model is still valid or maintained at the terminal device 110.
[0161] In some embodiments, the first information may comprise information of the set of models. In some embodiments, the information of the set of models may comprise an input of a model in the set of models. In some embodiments, the information of the set of models may comprise an output of the model. In some embodiments, the information of the set of models may comprise time required for model inference of the model. It is to be noted that any combinations of the above information may also be feasible.
[0162] In some embodiments, the information of the set of models may be associated with the first indication or the second indication. In other words, the information of the set of models may be reported per indication of a network condition.
[0163] In some embodiments, the information of the set of models may be associated with a model ID in the set of model ID. In other words, the information of the set of models may be reported per model ID.
[0164] In some embodiments, the information of the set of models may be associated with a group ID associated with the group of indications associated with network conditions. In other words, the information of the set of models may be reported per group ID.
[0165] In some embodiments, the information of the set of models may comprise an indication of whether update of a parameter of a model in the set of models is needed. In some embodiments, the information of the set of models may comprise an updated value of the parameter. In some embodiments, the information of the set of models may comprise both the indication of whether update of the parameter of the model in the set of models is needed and the updated value of the parameter. In some embodiments, the update of the parameter may be reported in an absolute value or an accumulated value.
[0166] In some embodiments, for a beam management use case, the information of the set of models may comprise at least one of the following or an update of at least one of the following: number of reference signals in a first set of reference signals for measurement (Set A size) ; number of reference signals in a second set of reference signals for prediction (Set B size) ; a mapping between the first set of reference signals and the second set of reference signals; a beam pattern for at least one of the first set of reference signals or the second set of reference signals (if the network device 120 provides multiple patterns for selection, the terminal device 110 may feedback a pattern ID) ; number of predicted results (e.g., number of RSRPs) ; a size of an observation window for the measurement (i.e., number of historical measurements or historical instances needed for model inference) ; or a size of a prediction window for the prediction (i.e., number of future instances obtained based on the output of model inference) .
[0167] In some embodiments, the terminal device 110 may transmit the first information via a RRC signaling or a MAC CE or UCI. In some embodiments, the terminal device 110 may transmit the first information via a UE capability reporting, UE capability reporting update, dynamic UE capability reporting, or UE assistance information.
[0168] In some embodiments, the terminal device 110 may transmit the first information in an event trigger way (where the event may be defined as that the model training is completed) . In some embodiments, the terminal device 110 may transmit the first information periodically or semi-persistently. In some embodiments, the terminal device 110 may transmit the first information aperiodic triggered by the network device 120.
[0169] Embodiment 3
[0170] In this embodiment, configurations in the list of configurations are configured sequentially from difference cells. For illustration, it is assumed that the list of configurations comprises a first configuration for data collection and a second configuration for data collection. It is to be noted that more than two configurations for data collection may also be feasible. More details will be described in connection with FIGs. 5A and 5B below.
[0171] FIG. 5A illustrates a signaling chart illustrating another example process 500A of model training and feedback according to some embodiments of the present disclosure. For the purpose of discussion, the process 500A will be described with reference to FIG. 1. The process 500A may involve the terminal device 110 and the network devices 120 and 130 as illustrated in FIG. 1. It is to be understood that the steps and the order of the steps in FIG. 5A are merely for illustration, and not for limitation. For example, the order of the steps may be changed. Some of the steps may be omitted or any other suitable additional steps may be added.
[0172] It is assumed that the terminal device 110 is served by the network device 120. The network device 130 provides a first cell, and the network device 120 provides a second cell. In some embodiments, the network device 120 and the network device 130 may be different network devices. In some embodiments, the network device 120 and the network device 130 may be the same network device. In some embodiments, the first cell and the second cell may be the same cell. In some embodiments, the first cell and the second cell may be different cells.
[0173] In the context of the present disclosure, a term ‘cell’ , ‘network device’ , ‘gNB’ , ‘physical cell’ , ‘bandwidth part (BWP) ’ , ‘component carrier (CC) ’ , ‘CC group’ , ‘band’ or ‘band combination’ may be interchangeably used with their IDs or indexes.
[0174] As shown in FIG. 5A, the network device 120 and the network device 130 may perform 510 information exchange for network conditions. In some embodiments, the network device 120 may transmit, to the network device 130, network conditions of the network device 120 and indications associated with the network conditions of the network device 120. The network device 130 may transmit, to the network device 120, network conditions of the network device 130 and indications associated with the network conditions of the network device 130.
[0175] As shown in FIG. 5A, the terminal device 110 may determine 520 a reference model for the first cell provided by the network device 130. In some embodiments, the network device 130 may transmit 521, to the terminal device 110, at least a first configuration for data collection. In some embodiments, the first configuration may correspond to a first indication associated with a first network condition. In some embodiments, the first configuration may not correspond to the first indication associated with the first network condition.
[0176] With reference to FIG. 5A, the terminal device 110 may collect 522 data corresponding to at least the first configuration and the first indication if any. For example, the data may be a measurement result of an RS based on at least one of an RS report configuration or RS resource configuration or RS measurement configuration. Alternatively, the data may be obtained from a dataset transferred to the terminal device 110 corresponding to the first indication. The dataset may be associated with a dataset ID.
[0177] In some embodiments, the terminal device 110 may determine at least a first set of data based on the first configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the first configuration, and determine the first set of data from the measurement results. In some embodiments, the terminal device 110 may determine at least the first set of data based on the first indication. For example, the terminal device 110 may determine a dataset corresponding to the first indication, and determine the first set of data from the dataset.
[0178] With reference to FIG. 5A, the terminal device 110 may determine 523 the reference model by model training based on at least the first set of data. In other words, the terminal device 110 may develop (e.g., train or update) the reference model based on at least the first set of data. It is to be noted that the reference model may be determined in any other suitable ways and the present disclosure does not limit this aspect.
[0179] Continuing to refer to FIG. 5A, the network device 120 may transmit 530, to the terminal device 110, a second configuration for data collection. In some embodiments, the second configuration may correspond to a second indication associated with a second network condition. In some embodiments, the network device 120 may be aware whether the second network condition of the network device 120 is consistent with the first network condition of the network device 130. In some embodiments, the network device 120 may not be aware whether the second network condition of the network device 120 is consistent with the first network condition of the network device 130. In some alternative embodiments, the second configuration may not correspond to the second indication.
[0180] With reference to FIG. 5A, the terminal device 110 may collect 540 data corresponding to the second configuration and the second indication if any. For example, the data may be a measurement result of an RS based on at least one of an RS report configuration or RS resource configuration or RS measurement configuration. Alternatively, the data may be obtained from a dataset transferred to the terminal device 110 corresponding to the second indication. The dataset may be associated with a dataset ID.
[0181] In some embodiments, the terminal device 110 may determine a second set of data based on the second configuration. For example, the terminal device 110 may obtain measurement results by performing measurements based on the second configuration, and determine the second set of data from the measurement results. In some embodiments where the second indication is provided, the terminal device 110 may determine the second set of data based on the second indication. For example, the terminal device 110 may determine a dataset corresponding to the second indication, and determine the second set of data from the dataset. In some embodiments where the second indication is not provided, the terminal device 110 may determine the second set of data corresponding to the second cell.
[0182] With reference to FIG. 5A, the terminal device 110 may determine 550 the set of models based on at least one of the reference model or the second set of data. In other words, the terminal device 110 may develop (e.g., train or update) an AI / ML model based on the data collected respectively or jointly. For illustration, an example embodiment will be described in connection with FIG. 5B.
[0183] FIG. 5B illustrates a schematic diagram 500B illustrating another example model development according to some embodiments of the present disclosure. As shown in FIG. 5B, a first model 501 may be a reference model 502 for the network device 130. In other words, the reference model for the network device 130 may be still applicable for the second network conditions of the network device 120. A second model 503 may be trained based on a second set of data 504. The second model 503 may be only suitable for the second network condition of the network device 120 and not suitable for the first network condition of the network device 130. A third model 505 may be trained or updated based on the reference model 502 and the second set of data 504. The third model 505 may be suitable for both the first network condition of the network device 130 and the second network condition of the network device 120. Other one or more models (or versions or parameters) 506 may also be possible, e.g., with part of the reference model 502 and the second set of data 504. It may be up to the terminal device 110 to develop the set of models based on the collected data.
[0184] As shown in FIG. 5B, the terminal device 110 may perform 507 a check and feedback, e.g., by checking performance of the developed models 501, 503, 505 and 506 (or versions or parameters) . The terminal device 110 may check and feedback whether different network conditions have impact on the set of models, e.g., by checking the performance and / or cost of the models (or versions, parameters) developed and decide whether to use separate models or one general model to handle the different network conditions. How to check and / or make the decision may be up to the terminal device 110.
[0185] Refer back to FIG. 5A, the terminal device 110 may determine and transmit 560 the first information indicating the assumption of the terminal device 110 for the consistency among network conditions.
[0186] In some embodiments, the first information may comprise an indication of whether the terminal device 110 assumes that the first network condition in the first cell and the second network condition in the second cell are consistent. In some embodiments, the terminal device 110 may signal a list of cell IDs and indications associated with network conditions, e.g., {UE-assumed consistency: cell ID#1, indication#1; cell ID#2, indication#2; …} . It is to be noted that UE-assumed consistency does not mean real consistency in NW implementation, and basically means that UE apply the same model for those indications and cells.
[0187] In some embodiments, the first information may indicate whether the terminal device 110 assumes that network conditions in the first cell and network conditions in the second cell are consistent, if both the first indication and the second indication are not provided. For example, the terminal device 110 may signal a list of cell IDs, e.g., {UE- assumed consistency: cell ID#1, cell ID#2, …} .
[0188] In some embodiments, the first information may indicate whether the terminal device 110 assumes that the first network condition associated with the first indication in the first cell and the second network condition in the second cell are consistent, if the second indication is not provided. For example, the terminal device 110 may signal a list of cell IDs and indications (if any) , e.g., {UE-assumed consistency: cell ID#1, indication#1; cell ID#2; …} .
[0189] In some embodiments, the first information may indicate whether the terminal device 110 assumes that the first network condition in the first cell and the second network condition associated with the second indication in the second cell are consistent, if the first indication is not provided. For example, the terminal device 110 may signal a list of cell IDs and indications (if any) , e.g., {UE-assumed consistency: cell ID#1; cell ID#2, indication #2, …} .
[0190] In some embodiments, the first information may comprise an indication of whether the terminal device 110 has an applicable model for the second cell. In some embodiments, there may be multiple applicable models for the second cell.
[0191] In some embodiments, the first information may comprise an indication of whether the second indication is needed. In some embodiments, the first information may comprise an indication of whether the second indication is needed for one or more of the following: model training, model inference or performance monitoring. In some embodiments, the first information may comprise an indication of whether the second indication is needed for data collection configuration for one or more of the following: model training, model inference or performance monitoring.
[0192] In some embodiments, the first information may comprise a model ID of the applicable model. In some embodiments, the first information may comprise the first indication with the applicable model. Thereby, the network device 120 may be aware that the terminal device 110 has applicable models for the second cell and decide contents to be signaled for the terminal device 110 to do the model inference, e.g., an indication associated with a network condition.
[0193] In some embodiments, the first information may comprise an indication of whether the reference model is valid. In some embodiments, considering that the terminal device 110 may have limited capabilities to maintain multiple AI / ML models and the terminal device 110 may release the previous developed models, the first information may include information of whether the first indication or the reference model is still valid or maintained at the terminal device 110.
[0194] In some embodiments, the first information may comprise information of the set of models. In some embodiments, the information of the set of models may comprise an input of a model in the set of models. In some embodiments, the information of the set of models may comprise an output of the model. In some embodiments, the information of the set of models may comprise time required for model inference of the model. It is to be noted that any combinations of the above information may also be feasible.
[0195] In some embodiments, the information of the set of models may be associated with the first indication or the second indication. In other words, the information of the set of models may be reported per indication of a network condition.
[0196] In some embodiments, the information of the set of models may be associated with a model ID in the set of model ID. In other words, the information of the set of models may be reported per model ID.
[0197] In some embodiments, the information of the set of models may be associated with a group ID associated with the group of indications associated with network conditions. In other words, the information of the set of models may be reported per group ID.
[0198] In some embodiments, the information of the set of models may comprise an indication of whether update of a parameter of a model in the set of models is needed. In some embodiments, the information of the set of models may comprise an updated value of the parameter. In some embodiments, the information of the set of models may comprise both the indication of whether update of the parameter of the model in the set of models is needed and the updated value of the parameter. In some embodiments, the update of the parameter may be reported in an absolute value or an accumulated value.
[0199] In some embodiments, for a beam management use case, the information of the set of models may comprise at least one of the following or an update of at least one of the following: number of reference signals in a first set of reference signals for measurement (Set A size) ; number of reference signals in a second set of reference signals for prediction (Set B size) ; a mapping between the first set of reference signals and the second set of reference signals; a beam pattern for at least one of the first set of reference signals or the second set of reference signals (if the network device 120 provides multiple patterns for selection, the terminal device 110 may feedback a pattern ID) ; number of predicted results (e.g., number of RSRPs) ; a size of an observation window for the measurement (i.e., number of historical measurements or historical instances needed for model inference) ; or a size of a prediction window for the prediction (i.e., number of future instances obtained based on the output of model inference) .
[0200] In some embodiments, the terminal device 110 may transmit the first information together with information that the model training of the second cell is completed. In some embodiments, the terminal device 110 may transmit the first information together with information that the cell switch is completed.
[0201] So far, the model training at the terminal device 110 and the feedback from the terminal device 110 to the network device 120 are described. It is to be understood that operations described in Embodiments 1 to 3 may be carried out separately or in any suitable combinations.
[0202] Refer back to FIG. 2, the network device 120 may transmit 230 a configuration (for convenience, also referred to as a third configuration herein) for model inference to the terminal device 110. In some embodiments, the network device 120 may generate the configuration for model inference based on the first information so as to ensure consistency between the model training and the model inference. In some embodiments, the configuration may be a data collection configuration.
[0203] In some embodiments, the third configuration may indicate at least a presence of an indication (for convenience, also referred to as a third indication herein) associated with a network condition (for convenience, also referred to as a third network condition herein) . In some embodiments, an information element (IE) may be used to indicate the presence of the third indication associated with the third network condition. For example, a value of the IE may be set as ‘not configured’ (i.e., absent) , or the third indication, or a list of indications associated with network conditions comprising the third indication. In some embodiments, two IEs may be used to indicate the presence of the third indication associated with the third network condition. A first IE in the two IEs is used to enable the model inference with the third indication, and the value of the first IE may be set as ‘enable’ or ‘disable’ . A second IE in the two IE is used to signal the third indication or a list of indications associated with network conditions comprising the third indication, if the value of the first IE is set as ‘enable’ .
[0204] In some embodiments, if the value of the first IE is set as ‘not configured’ or the value of the first IE is set as ‘disable’ , the consistency is not assumed by the network device and the consistency is not assumed by the terminal device. In some embodiments, if the value of the first IE is set as ‘not configured’ or the value of the first IE is set as ‘disable’ , the consistency is not assumed by the network device but the consistency may be assumed by the terminal device.
[0205] In some embodiments, if the first information indicates that the third indication is not needed, the network device 120 may generate the third configuration without the third indication. That is, if the terminal device 110 indicates that an indication associated with a network condition is not needed, the network device 120 does not need to signal the indication associated with the network condition for model inference. For example, if only one IE is used, the network device 120 may not configure this IE. If the two IEs are used, the network device 120 may set the first IE as ‘disable’ . The network device 120 may be aware that different network conditions do not have impacts on a choice of an AI / ML model for model inference at the terminal device 110. It is up to the terminal device 110 to apply a suitable model for model inference.
[0206] In some embodiments, if the first information indicates that the third indication is needed, the network device 120 may generate the third configuration with the third indication. That is, if the terminal device 110 indicates that an indication associated with a network condition is needed, the network device 120 needs to signal the indication associated with the network condition for model inference. For example, if only one IE is used, the network device 120 may configure this IE. If the two IEs are used, the network device 120 may set the first IE as ‘enable’ . The network device 120 may be aware that different network conditions have impacts on a choice of an AI / ML model for model inference at the terminal device 110. The terminal device 110 may need to apply a suitable model which is developed based on data collected corresponding to an indication associated with a network condition in the model training.
[0207] In some embodiments, the third configuration may further indicate a group ID associated with a group of indications associated with network conditions. In some embodiments, if the first information indicates the group of indications associated with network conditions, the network device 120 may generate the third configuration with the group ID.
[0208] In some embodiments, the third configuration may further indicate a model ID. In some embodiments, if the first information indicates a mapping between a model ID and the third indication, the network device 120 may generate the third configuration with the third indication. In some embodiments, if the mapping is not one-to-one mapping, the network device 120 may need to signal the model ID together with the third indication.
[0209] In some embodiments, if the first information indicates the third indication is invalid, the network device 120 may generate the third configuration without the third indication. In some embodiments, if the first information indicates a group of indications associated with network conditions is not valid or is not maintained, the network device 120 may not signal the group of indications associated with network conditions, or it is expected to have a longer application timing for model inference.
[0210] In some embodiments, if the first information indicates information of a model, the network device 120 may generate the third configuration based on the information of the model. For example, for a beam management use case, the third configuration may comprise at least one of the following: number of reference signals in a first set of reference signals for measurement (i.e., Set B size) ; number of reference signals in a second set of reference signals for prediction (i.e., Set A size) ; a mapping between the first set of reference signals and the second set of reference signals (i.e., a mapping between Set A and Set B) ; a beam pattern for at least one of the first set of reference signals or the second set of reference signals; number of predicted results (e.g., number of predicted beam information (e.g., RSRPs to be reported) ) ; a size of an observation window for the measurement (i.e., number of historical measurements or historical instances needed for model inference) ; or a size of a prediction window for the prediction (i.e., number of future instances obtained based on the output of model inference) .
[0211] Continuing to refer to FIG. 2, the terminal device 110 may perform 240 the model inference based on the third configuration. In some embodiments, if the third configuration indicates the third indication is present, the terminal device 110 may assume that a network condition associated with the third configuration and the third network condition associated with a configuration corresponding to the third indication in the list of configurations are consistent. In some embodiments, if the third configuration indicates the third indication is not present, the terminal device 110 may assume that a network condition associated with the third configuration and any network condition associated with any configuration corresponding to any indication in the list of configurations are consistent.
[0212] With the process 200, consistency between model training and model inference may be ensured. The consistency between model training and model inference across cells may also be ensured, and processing complexity across cells may be reduced.
[0213] It is to be understood that operations described in the process 200 may be carried out separately or in any suitable combinations with that described in Embodiments 1 to 3.
[0214] EXAMPLE IMPLEMENTATION OF METHODS
[0215] Corresponding to the above processes, embodiments of the present disclosure provide methods of communication implemented at a terminal device and a network device. These methods will be described below with reference to FIGs. 6 and 7.
[0216] FIG. 6 illustrates a flowchart of a method 600 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. For example, the method 600 may be performed at the terminal device 110 as shown in FIG. 1. For the purpose of discussion, in the following, the method 600 will be described with reference to FIG. 1. It is to be understood that the method 600 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard.
[0217] At block 610, the terminal device 110 may determine a set of models based at least on a list of configurations for data collection. A configuration in the list of configurations corresponds to an indication associated with a network condition.
[0218] At block 620, the terminal device 110 may determine, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions.
[0219] At block 630, the terminal device 110 may transmit the first information to the network device 120.
[0220] In some embodiments, the terminal device 110 may determine the set of models by: receiving at least a first configuration for data collection and a second configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition, the second configuration corresponding to a second indication associated with a second network condition; determining at least a first set of data based on the first configuration or the first indication and a second set of data based on the second configuration or the second indication; and determining the set of models based on at least one of at least the first set of data or the second set of data.
[0221] In some embodiments, the terminal device 110 may determine the first set of data by:determining the first set of data from measurement results obtained based on the first configuration, or from a dataset corresponding to the first indication. In some embodiments, the terminal device 110 may determine the second set of data by: determining the second set of data from measurement results obtained based on the second configuration, or from a dataset corresponding to the second indication.
[0222] In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that at least the first network condition and the second network condition are consistent; an indication indicating that a group of indications of network conditions comprising at least the first network condition and the second network condition are handled jointly; a set of model identities for the group of indications associated with network conditions; an indication of whether at least the first indication and the second indication are needed; or information of the set of models. In some embodiments, the information of the set of models is associated with one of the following: at least the first indication or the second indication; a model identity in the set of model identities; or a group identity associated with the group of indications associated with network conditions.
[0223] In some embodiments, the terminal device 110 may determine the set of models by: receiving a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition; determining a second set of data based on the second configuration or the second indication; and determining the set of models based on at least one of a reference model or the second set of data. In some embodiments, the terminal device 110 may receive at least a first configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition; determine at least a first set of data based on the first configuration or the first indication; and determine the reference model based at least on the first set of data.
[0224] In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that the second network condition associated with the second indication and a network condition associated with the reference model are consistent; an indication of whether the terminal device assumes that the second network condition associated with the second configuration and a network condition associated with a configuration for data collection of the reference model are consistent; an indication of whether the second indication is needed; a set of model identities for a group of indications associated with network conditions; an indication of whether the reference model is valid; or information of the set of models. In some embodiments, the information of the set of models may comprise at least one of the following: an indication of whether update of a parameter of a model in the set of models is needed; or an updated value of the parameter.
[0225] In some embodiments, the terminal device 110 may receive the first configuration by receiving the first configuration from a first cell. In some embodiments, the terminal device 110 may receive the second configuration by receiving the second configuration from a second cell. In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that the first network condition in the first cell and the second network condition in the second cell are consistent; an indication of whether the terminal device has an applicable model for the second cell; an indication of whether the second indication is needed; a model identity of the applicable model; the first indication; or an indication of whether the reference model is valid.
[0226] In some embodiments, the information of the set of models may comprise at least one of the following: an input of a model in the set of models; an output of the model; or time required for model inference of the model.
[0227] In some embodiments, the information of the set of models may comprise at least one of the following: number of reference signals in a first set of reference signals for measurement; number of reference signals in a second set of reference signals for prediction; a mapping between the first set of reference signals and the second set of reference signals; a beam pattern for at least one of the first set of reference signals or the second set of reference signals; number of predicted results; a size of an observation window for the measurement; or a size of a prediction window for the prediction.
[0228] In some embodiments, the terminal device 110 may receive, from the network device 120, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition. In accordance with a determination that the third configuration indicates the third indication is present, the terminal device 110 may assume that a network condition associated with the third configuration and the third network condition associated with a configuration corresponding to the third indication in the list of configurations are consistent.
[0229] In some embodiments, the third configuration may further indicate at least one of a model identity or a group identity associated with a group of indications associated with network conditions.
[0230] In some embodiments, the terminal device 110 may transmit, to the network device 120, second information indicating capability of the terminal device. The second information comprises at least one of the following: number of indications associated with network conditions supported by the terminal device; number of indications associated with network conditions supported by the terminal device in a cell; number of indications associated with network conditions supported by the terminal device for a model; number of indications associated with network conditions supported by the terminal device for a model in a cell; number of models maintained at the terminal device; an indication of whether the terminal device supports a simultaneous configuration or a sequential configuration for multiple indications associated with network conditions; or an indication of whether the terminal device supports an inter-cell assumption for the consistency among network conditions.
[0231] With the method 600, a terminal device may feedback whether a change of a network condition has impact on a model, and consistency between model training and model inference may be facilitated. It is to be understood that operations of the method 600 correspond to that described in connection with FIGs. 2 to 5B, and other details are omitted here for conciseness.
[0232] FIG. 7 illustrates a flowchart of a method 700 of communication implemented at a network device in accordance with some embodiments of the present disclosure. For example, the method 700 may be performed at the network device 120 as shown in FIG. 1. For the purpose of discussion, in the following, the method 700 will be described with reference to FIG. 1. It is to be understood that the method 700 may include additional blocks not shown and / or may omit some blocks as shown, and the scope of the present disclosure is not limited in this regard.
[0233] At block 710, the network device 120 may receive, from the terminal device 110, first information indicating an assumption of the terminal device 110 for consistency among network conditions.
[0234] At block 720, the network device 120 may transmit, to the terminal device 110, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.
[0235] In some embodiments, the third configuration may further indicate at least one of a model identity or a group identity associated with a group of indications associated with network conditions.
[0236] In some embodiments, the network device 120 may transmit, to the terminal device 110, at least a first configuration for data collection and a second configuration for data collection. The first configuration may correspond to a first indication associated with a first network condition, and the second configuration may correspond to a second indication associated with a second network condition.
[0237] In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that at least the first network condition and the second network condition are consistent; an indication indicating that a group of indications associated with network conditions comprising at least the first network condition and the second network condition are handled jointly; a set of model identities for the group of indications associated with network conditions; an indication of whether at least the first indication and the second indication are needed; or information of the set of models.
[0238] In some embodiments, the information of the set of models may be associated with one of the following: at least the first indication or the second indication; a model identity in the set of model identities; or a group identity associated with the group of indications associated with network conditions.
[0239] In some embodiments, the network device 120 may transmit at least a first configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition; receive information indicating that a model training is completed for the first configuration; and transmit a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition.
[0240] In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that the second network condition associated with the second indication and a network condition associated with a reference model are consistent; an indication of whether the terminal device assumes that the second network condition associated with the second configuration and a network condition associated with a configuration for data collection of the reference model are consistent; an indication of whether the second indication is needed; a set of model identities for a group of indications associated with network conditions; an indication of whether the reference model is valid; or information of the set of models. In some embodiments, the information of the set of models may comprise at least one of the following: an indication of whether update of a parameter of a model in the set of models is needed; or an updated value of the parameter.
[0241] In some embodiments, the network device 120 may transmit, via a second cell, a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition.
[0242] In some embodiments, the first information may comprise at least one of the following: an indication of whether the terminal device assumes that a first network condition in a first cell and the second network condition in the second cell are consistent; an indication of whether the terminal device has an applicable model for the second cell; an indication of whether the second indication is needed; a model identity of the applicable model; a second indication of the second network condition; or an indication of whether the reference model is valid.
[0243] In some embodiments, the network device 120 may receive, from a further network device (e.g., the network device 130) providing the first cell, the first network condition and the first indication associated with the first network condition; and transmit, to the further network device, the second network condition and the second indication associated with the second network condition.
[0244] In some embodiments, the information of the set of models may comprise at least one of the following: an input of a model in the set of models; an output of the model; or time required for model inference of the model.
[0245] In some embodiments, the information of the set of models may comprise at least one of the following: number of reference signals in a first set of reference signals for measurement; number of reference signals in a second set of reference signals for prediction; a mapping between the first set of reference signals and the second set of reference signals; a beam pattern for at least one of the first set of reference signals or the second set of reference signals; number of predicted results; a size of an observation window for the measurement; or a size of a prediction window for the prediction.
[0246] In some embodiments, the network device 120 may receive, from the terminal device 110, second information indicating capability of the terminal device. The second information comprises at least one of the following: number of indications associated with network conditions supported by the terminal device; number of indications associated with network conditions supported by the terminal device in a cell; number of indications associated with network conditions supported by the terminal device for a model; number of indications associated with network conditions supported by the terminal device for a model in a cell; number of models maintained at the terminal device; an indication of whether the terminal device supports a simultaneous configuration or a sequential configuration for multiple indications associated with network conditions; or an indication of whether the terminal device supports an inter-cell assumption for the consistency among network conditions.
[0247] In some embodiments, the network device 120 may be further caused to at least one of the following: in accordance with a determination that the first information indicates that the third indication is not needed, generate the third configuration without the third indication; in accordance with a determination that the first information indicates that the third indication is needed, generate the third configuration with the third indication; in accordance with a determination that the first information indicates a mapping between a model identity and the third indication, generate the third configuration with the third indication; in accordance with a determination that the first information indicates the third indication is invalid, generate the third configuration without the third indication; or in accordance with a determination that the first information indicates information of a model, generate the third configuration based on the information of the model.
[0248] With the method 700, NW may be aware of whether a change of a network condition has impact on a model. Thus, NW assignment on the network condition may be optimized, and consistency between model training and model inference may be ensured.
[0249] It is to be understood that operations of the methods 600 and 700 correspond to that described in connection with FIGs. 2 to 5B, and other details are omitted here for conciseness.
[0250] EXAMPLE IMPLEMENTATION OF DEVICES
[0251] FIG. 8 is a simplified block diagram of a device 800 that is suitable for implementing embodiments of the present disclosure. The device 800 can be considered as a further example implementation of the terminal device 110 or the network device 120 as shown in FIG. 1. Accordingly, the device 800 can be implemented at or as at least a part of the terminal device 110 or the network device 120.
[0252] As shown, the device 800 includes a processor 810, a memory 820 coupled to the processor 810, a suitable transceiver 840 coupled to the processor 810, and a communication interface coupled to the transceiver 840. The memory 810 stores at least a part of a program 830. The transceiver 840 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 840 may include at least one of a transmitter 842 or a receiver 844. The transmitter 842 and the receiver 844 may be functional modules or physical entities. The transceiver 840 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.
[0253] The program 830 is assumed to include program instructions that, when executed by the associated processor 810, enable the device 800 to operate in accordance with the embodiments of the present disclosure, as discussed herein with reference to FIGs. 1 to 7. The embodiments herein may be implemented by computer software executable by the processor 810 of the device 800, or by hardware, or by a combination of software and hardware. The processor 810 may be configured to implement various embodiments of the present disclosure. Furthermore, a combination of the processor 810 and memory 820 may form processing means 850 adapted to implement various embodiments of the present disclosure.
[0254] The memory 820 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 820 is shown in the device 800, there may be several physically distinct memory modules in the device 800. The processor 810 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 800 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.
[0255] In some embodiments, a terminal device comprises a circuitry configured to: determine a set of models based at least on a list of configurations for data collection, a configuration in the list of configurations corresponding to an indication associated with a network condition; determine, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions; and transmit the first information to a network device.
[0256] In some embodiments, a network device comprises a circuitry configured to: receive, from a terminal device, first information indicating an assumption of the terminal device for consistency among network conditions; and transmit, to the terminal device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.
[0257] 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.
[0258] 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.
[0259] 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 7. 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.
[0260] 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.
[0261] 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.
[0262] 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.
[0263] 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 terminal device, comprising:a processor configured to cause the terminal device to:determine a set of models based at least on a list of configurations for data collection, a configuration in the list of configurations corresponding to an indication associated with a network condition;determine, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions; andtransmit the first information to a network device.2.The terminal device of claim 1, wherein the terminal device is caused to determine the set of models by:receiving at least a first configuration for data collection and a second configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition, the second configuration corresponding to a second indication associated with a second network condition;determining at least a first set of data based on the first configuration or the first indication and a second set of data based on the second configuration or the second indication; anddetermining the set of models based on at least one of at least the first set of data or the second set of data.3.The terminal device of claim 2, wherein the terminal device is caused to determine the first set of data by: determining the first set of data from measurement results obtained based on the first configuration, or from a dataset corresponding to the first indication, andwherein the terminal device is caused to determine the second set of data by: determining the second set of data from measurement results obtained based on the second configuration, or from a dataset corresponding to the second indication.4.The terminal device of claim 2, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that at least the first network condition and the second network condition are consistent;an indication indicating that a group of indications of network conditions comprising at least the first network condition and the second network condition are handled jointly;a set of model identities for the group of indications associated with network conditions;an indication of whether at least the first indication and the second indication are needed; orinformation of the set of models.5.The terminal device of claim 4, wherein the information of the set of models is associated with one of the following:at least the first indication or the second indication;a model identity in the set of model identities; ora group identity associated with the group of indications associated with network conditions.6.The terminal device of claim 1, wherein the terminal device is caused to determine the set of models by:receiving a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition;determining a second set of data based on the second configuration or the second indication; anddetermining the set of models based on at least one of a reference model or the second set of data.7.The terminal device of claim 6, wherein the terminal device is further caused to:receive at least a first configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition;determine at least a first set of data based on the first configuration or the first indication; anddetermine the reference model based at least on the first set of data.8.The terminal device of claim 6, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that the second network condition associated with the second indication and a network condition associated with the reference model are consistent;an indication of whether the terminal device assumes that the second network condition associated with the second configuration and a network condition associated with a configuration for data collection of the reference model are consistent;an indication of whether the second indication is needed;a set of model identities for a group of indications associated with network conditions;an indication of whether the reference model is valid; orinformation of the set of models.9.The terminal device of claim 8, wherein the information of the set of models comprises at least one of the following:an indication of whether update of a parameter of a model in the set of models is needed; oran updated value of the parameter.10.The terminal device of claim 7, wherein the terminal device is caused to receive the first configuration by receiving the first configuration from a first cell; andwherein the terminal device is caused to receive the second configuration by receiving the second configuration from a second cell.11.The terminal device of claim 10, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that the first network condition in the first cell and the second network condition in the second cell are consistent;an indication of whether the terminal device has an applicable model for the second cell;an indication of whether the second indication is needed;a model identity of the applicable model;the first indication; oran indication of whether the reference model is valid.12.The terminal device of claim 4 or 8, wherein the information of the set of models comprises at least one of the following:an input of a model in the set of models;an output of the model; ortime required for model inference of the model.13.The terminal device of claim 4 or 8, wherein the information of the set of models comprises at least one of the following:number of reference signals in a first set of reference signals for measurement;number of reference signals in a second set of reference signals for prediction;a mapping between the first set of reference signals and the second set of reference signals;a beam pattern for at least one of the first set of reference signals or the second set of reference signals;number of predicted results;a size of an observation window for the measurement; ora size of a prediction window for the prediction.14.The terminal device of claim 1, wherein the terminal device is further caused to:receive, from the network device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition; andin accordance with a determination that the third configuration indicates the third indication is present, assume that a network condition associated with the third configuration and the third network condition associated with a configuration corresponding to the third indication in the list of configurations are consistent.15.The terminal device of claim 14, wherein the third configuration further indicates at least one of a model identity or a group identity associated with a group of indications associated with network conditions.16.The terminal device of claim 1, wherein the terminal device is further caused to:transmit, to the network device, second information indicating capability of the terminal device, the second information comprising at least one of the following:number of indications associated with network conditions supported by the terminal device;number of indications associated with network conditions supported by the terminal device in a cell;number of indications associated with network conditions supported by the terminal device for a model;number of indications associated with network conditions supported by the terminal device for a model in a cell;number of models maintained at the terminal device;an indication of whether the terminal device supports a simultaneous configuration or a sequential configuration for multiple indications associated with network conditions; oran indication of whether the terminal device supports an inter-cell assumption for the consistency among network conditions.17.A network device, comprising:a processor configured to cause the network device to:receive, from a terminal device, first information indicating an assumption of the terminal device for consistency among network conditions; andtransmit, to the terminal device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.18.The network device of claim 17, wherein the third configuration further indicates at least one of a model identity or a group identity associated with a group of indications associated with network conditions.19.The network device of claim 17, wherein the network device is further caused to:transmit, to the terminal device, at least a first configuration for data collection and a second configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition, the second configuration corresponding to a second indication associated with a second network condition.20.The network device of claim 19, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that at least the first network condition and the second network condition are consistent;an indication indicating that a group of indications associated with network conditions comprising at least the first network condition and the second network condition are handled jointly;a set of model identities for the group of indications associated with network conditions;an indication of whether at least the first indication and the second indication are needed; orinformation of the set of models.21.The network device of claim 20, wherein the information of the set of models is associated with one of the following:at least the first indication or the second indication;a model identity in the set of model identities; ora group identity associated with the group of indications associated with network conditions.22.The network device of claim 17, wherein the network device is further caused to:transmit at least a first configuration for data collection, the first configuration corresponding to a first indication associated with a first network condition;receive information indicating that a model training is completed for the first configuration; andtransmit a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition.23.The network device of claim 22, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that the second network condition associated with the second indication and a network condition associated with a reference model are consistent;an indication of whether the terminal device assumes that the second network condition associated with the second configuration and a network condition associated with a configuration for data collection of the reference model are consistent;an indication of whether the second indication is needed;a set of model identities for a group of indications associated with network conditions;an indication of whether the reference model is valid; orinformation of the set of models.24.The network device of claim 23, wherein the information of the set of models comprises at least one of the following:an indication of whether update of a parameter of a model in the set of models is needed; oran updated value of the parameter.25.The network device of claim 17, wherein the network device is further caused to:transmit, via a second cell, a second configuration for data collection, the second configuration corresponding to a second indication associated with a second network condition.26.The network device of claim 25, wherein the first information comprises at least one of the following:an indication of whether the terminal device assumes that a first network condition in a first cell and the second network condition in the second cell are consistent;an indication of whether the terminal device has an applicable model for the second cell;an indication of whether the second indication is needed;a model identity of the applicable model;a second indication of the second network condition; oran indication of whether the reference model is valid.27.The network device of claim 26, wherein the network device is further caused to:receive, from a further network device providing the first cell, the first network condition and the first indication associated with the first network condition; andtransmit, to the further network device, the second network condition and the second indication associated with the second network condition.28.The network device of claim 20 or 23, wherein the information of the set of models comprises at least one of the following:an input of a model in the set of models;an output of the model; ortime required for model inference of the model.29.The network device of claim 20 or 23, wherein the information of the set of models comprises at least one of the following:number of reference signals in a first set of reference signals for measurement;number of reference signals in a second set of reference signals for prediction;a mapping between the first set of reference signals and the second set of reference signals;a beam pattern for at least one of the first set of reference signals or the second set of reference signals;number of predicted results;a size of an observation window for the measurement; ora size of a prediction window for the prediction.30.The network device of claim 17, wherein the network device is further caused to:receive, from the terminal device, second information indicating capability of the terminal device, the second information comprising at least one of the following:number of indications associated with network conditions supported by the terminal device;number of indications associated with network conditions supported by the terminal device in a cell;number of indications associated with network conditions supported by the terminal device for a model;number of indications associated with network conditions supported by the terminal device for a model in a cell;number of models maintained at the terminal device;an indication of whether the terminal device supports a simultaneous configuration or a sequential configuration for multiple indications associated with network conditions; oran indication of whether the terminal device supports an inter-cell assumption for the consistency among network conditions.31.The network device of claim 17, wherein the network device is further caused to at least one of the following:in accordance with a determination that the first information indicates that the third indication is not needed, generate the third configuration without the third indication;in accordance with a determination that the first information indicates that the third indication is needed, generate the third configuration with the third indication;in accordance with a determination that the first information indicates a mapping between a model identity and the third indication, generate the third configuration with the third indication;in accordance with a determination that the first information indicates the third indication is invalid, generate the third configuration without the third indication; orin accordance with a determination that the first information indicates information of a model, generate the third configuration based on the information of the model.32.A method of communication, comprising:determining, at a terminal device, a set of models based at least on a list of configurations for data collection, a configuration in the list of configurations corresponding to an indication associated with a network condition;determining, based on the set of models, first information indicating an assumption of the terminal device for consistency among network conditions; andtransmitting the first information to a network device.33.A method of communication, comprising:receiving, at a network device and from a terminal device, first information indicating an assumption of the terminal device for consistency among network conditions; andtransmitting, to the terminal device, a third configuration for model inference indicating at least a presence of a third indication associated with a third network condition.
Citation Information
Patent Citations
Network mobility management optimization method, base station, device, system and related equipment
CN117560650A
Conditional artificial intelligence, machine learning model, and parameter set configurations
US20230412470A1