WO2025138053A1

WO2025138053A1 - Devices and methods for communication

Info

Publication number: WO2025138053A1
Application number: PCT/CN2023/142991
Authority: WO
Inventors: Peng Guan; Yukai GAO; Zhen He; Rao SHI; Zhaobang MIAO; Gang Wang
Original assignee: NEC Corp
Current assignee: NEC Corp
Priority date: 2023-12-28
Filing date: 2023-12-28
Publication date: 2025-07-03
Anticipated expiration: 2026-06-28
Also published as: WO2025138053A9

Abstract

Embodiments of the present disclosure provide a solution for a solution for machine-learning (ML) -based time domain prediction. In the solution, a terminal device receives, from a network device, control information of a report which is based on at least one first resource; performs at least one measurement on the at least one first resource; and transmits the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

Description

DEVICES AND METHODS FOR COMMUNICATION

FIELDS

Example embodiments of the present disclosure generally relate to the field of communication techniques and in particular, to devices and methods for machine-learning (ML) -based time domain prediction.

BACKGROUND

As communication networks and services increase in size, complexity, and number of users, operations in the communication networks may become increasingly more complicated. In order to improve the communication performance, machine learning (ML) /artificial intelligence (AI) technology is proposed to be used in the wireless communication network. For example, the terminal device and the network device may use different ML models to assist communication-related functionalities, such as, time-domain prediction for CSI and time-domain prediction for beam.

SUMMARY

In general, embodiments of the present disclosure provide a solution of ML-based time domain prediction.

In a first aspect, there is provided a terminal device comprising: a processor configured to cause the terminal device to: receive, from a network device, control information of a report which is based on at least one first resource; perform at least one measurement on the at least one first resource; and transmit the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In a second aspect, there is provided a terminal device comprising: a processor configured to cause the terminal device to: receive, from a network device, control information of a report which is based on at least one first resource; transmit, to the network device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In a third aspect, there is provided a network device comprising: a processor configured to cause the network device to: transmit, to a terminal device, control information of a report which is based on at least one first resource; and receive the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In a fourth aspect, there is provided a network device comprising: a processor configured to cause the network device to: transmit, to a terminal device, control information indicating based on one first resource; receive, from the terminal device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In a fifth aspect, there is provided a communication method performed by a terminal device. The method comprises: receiving, from a network device, control information of a report which is based on at least one first resource; performing at least one measurement on the at least one first resource; and transmitting the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In a sixth aspect, there is provided a communication method performed by a terminal device. The method comprises: receiving, from a network device, control information of a report which is based on at least one first resource; transmitting, to the network device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In a seventh aspect, there is provided a communication method performed by a network device. The method comprises: transmitting, to a terminal device, control information of a report which is based on at least one first resource; and receiving the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In an eighth aspect, there is provided a communication method performed by a network device. The method comprises: transmitting, to a terminal device, control information indicating based on one first resource; receiving, from the terminal device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In a ninth aspect, there is provided a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to carry out the method according to the fifth, sixth, seventh, or eighth aspect.

Other features of the present disclosure will become easily comprehensible through the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

Through the more detailed description of some example embodiments of the present disclosure in the accompanying drawings, the above and other objects, features and advantages of the present disclosure will become more apparent, wherein:

FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;

FIG. 2 illustrates an example of the inference procedure for beam management;

FIG. 3A to FIG. 3C illustrate example use cases of beam prediction;

FIG. 4 illustrates an example of the inference procedure for CSI prediction;

FIG. 5 illustrates an example timing for prediction;

FIG. 6 illustrates a signaling flow of communication in accordance with some embodiments of the present disclosure;

FIG. 7 to FIG. 16 illustrate example timings for prediction;

FIG. 17 illustrates a flowchart of a method implemented at a terminal device according to some example embodiments of the present disclosure;

FIG. 18 illustrates a flowchart of a method implemented at a terminal device according to some example embodiments of the present disclosure;

FIG. 19 illustrates a flowchart of a method implemented at a network device according to some example embodiments of the present disclosure;

FIG. 20 illustrates a flowchart of a method implemented at a network device according to some example embodiments of the present disclosure;

FIG. 21 illustrates a simplified block diagram of an apparatus that is suitable for implementing example embodiments of the present disclosure.

Throughout the drawings, the same or similar reference numerals represent the same or similar element.

DETAILED DESCRIPTION

Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. Embodiments described herein can be implemented in various manners other than the ones described below.

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.

As used herein, the term ‘terminal device’ refers to any device having wireless or wired communication capabilities. Examples of the terminal device include, but not limited to, user equipment (UE) , personal computers, desktops, mobile phones, cellular phones, smart phones, personal digital assistants (PDAs) , portable computers, tablets, wearable devices, internet of things (IoT) devices, Ultra-reliable and Low Latency Communications (URLLC) devices, Internet of Everything (IoE) devices, machine type communication (MTC) devices, devices on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure/network, devices for Integrated Access and Backhaul (IAB) , Space borne vehicles or Air borne vehicles in Non-terrestrial networks (NTN) including Satellites and High Altitude Platforms (HAPs) encompassing Unmanned Aircraft Systems (UAS) , eXtended Reality (XR) devices including different types of realities such as Augmented Reality (AR) , Mixed Reality (MR) and Virtual Reality (VR) , the unmanned aerial vehicle (UAV) commonly known as a drone which is an aircraft without any human pilot, devices on high speed train (HST) , or image capture devices such as digital cameras, sensors, gaming devices, music storage and playback appliances, or Internet appliances enabling wireless or wired Internet access and browsing and the like. The ‘terminal device’ can further have ‘multicast/broadcast’ feature, to support public safety and mission critical, V2X applications, transparent IPv4/IPv6 multicast delivery, IPTV, smart TV, radio services, software delivery over wireless, group communications and IoT applications. It may also incorporate one or multiple Subscriber Identity Module (SIM) as known as Multi-SIM. The term “terminal device” can be used interchangeably with a UE, a mobile station, a subscriber station, a mobile terminal, a user terminal or a wireless device.

The term “network device” refers to a device which is capable of providing or hosting a cell or coverage where terminal devices can communicate. Examples of a network device include, but not limited to, a Node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a next generation NodeB (gNB) , a transmission reception point (TRP) , a remote radio unit (RRU) , a radio head (RH) , a remote radio head (RRH) , an IAB node, a low power node such as a femto node, a pico node, a reconfigurable intelligent surface (RIS) , and the like.

The terminal device or the network device may have Artificial intelligence (AI) or Machine learning (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.

The terminal or the network device may work on several frequency ranges, e.g., FR1 (e.g., 450 MHz to 6000 MHz) , FR2 (e.g., 24.25GHz to 52.6GHz) , frequency band larger than 100 GHz as well as Tera Hertz (THz) . It can further work on licensed/unlicensed/shared spectrum. The terminal device may have more than one connection with the network devices under Multi-Radio Dual Connectivity (MR-DC) application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.

The embodiments of the present disclosure may be performed in test equipment, e.g., signal generator, signal analyzer, spectrum analyzer, network analyzer, test terminal device, test network device, channel emulator. In some embodiments, the terminal device may be connected with a first network device and a second network device. One of the first network device and the second network device may be a master node and the other one may be a secondary node. The first network device and the second network device may use different radio access technologies (RATs) . In some embodiments, the first network device may be a first RAT device and the second network device may be a second RAT device. In some embodiments, the first RAT device is eNB and the second RAT device is gNB. Information related with different RATs may be transmitted to the terminal device from at least one of the first network device or the second network device. In some embodiments, first information may be transmitted to the terminal device from the first network device and second information may be transmitted to the terminal device from the second network device directly or via the first network device. In some embodiments, information related with configuration for the terminal device configured by the second network device may be transmitted from the second network device via the first network device. Information related with reconfiguration for the terminal device configured by the second network device may be transmitted to the terminal device from the second network device directly or via the first network device.

As used herein, the singular forms ‘a’ , ‘an’ and ‘the’ are intended to include the plural forms as well, unless the context clearly indicates otherwise. The term ‘includes’ and its variants are to be read as open terms that mean ‘includes, but is not limited to. ’ The term ‘based on’ is to be read as ‘at least in part based on. ’ The term ‘one embodiment’ and ‘an embodiment’ are to be read as ‘at least one embodiment. ’ The term ‘another embodiment’ is to be read as ‘at least one other embodiment. ’ The terms ‘first, ’ ‘second, ’ and the like may refer to different or same objects. Other definitions, explicit and implicit, may be included below.

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.

As used herein, the term “resource, ” “transmission resource, ” “uplink resource, ” or “downlink resource” may refer to any resource for performing a communication, such as a resource in time domain, a resource in frequency domain, a resource in space domain, a resource in code domain, or any other resource enabling a communication, and the like. In the following, unless explicitly stated, a resource in both frequency domain and time domain will be used as an example of a transmission resource for describing some example embodiments of the present disclosure. It is noted that example embodiments of the present disclosure are equally applicable to other resources in other domains.

As used herein, the terms “UE expects” , “UE does not expect, “terminal device expects” , “terminal device does not expect” may imply restrictions on a configuration of a network device (also referred to as NW configuration) . The terms “UE is not expected to” and “terminal device is not expected to” may imply a terminal implementation, also referred to as UE implementation. In some embodiments, the terms “UE does not expect” and “UE is not expected to” may be used equally.

Several use cases about time-domain prediction have been identified, including channeled state information (CSI) prediction and downlink transmitting (TX) beam prediction.

In legacy CSI report or beam report, different time domain behaviors may be configured for measurement reference signal (RS) and for reporting, respectively, including periodic, semi-persistent, aperiodic. In particular, aperiodic report based on aperiodic resource is possible and is the usable manner to acquire a timely CSI/beam report.

As for a measurement report (such as, a CSI report) , the resources for measurement are configured in hierarchical structure. Specifically, a ResourceConfig may be associated with one or more resource sets, and one resource set comprises at least one resource.

In operation, the ReportConfig is linked to one or multiple ResouceConfig, the ResourceConfig is linked to one or multiple resource sets via ResourceSetList, and the ResourceSet contains information of one or multiple resources via ResourceList, where resource is the minimum unit for physical layer configuration.

As for aperiodic report, the aperiodic report may be configured for periodic, semi-persistent aperiodic measurement resources. Further, a list of CSI trigger states may be configured for AP report. When aperiodic CSI-RS is used with aperiodic reporting, the CSI-RS offset may be configured per resource set. Generally speaking, the UE does not expect that aperiodic CSI-RS is transmitted before the OFDM symbol (s) carrying its triggering DCI.

By far, for AI/ML-based beam management, it has been agreed to support below BM-Case1 and BM-Case2:

BM-Case1: Spatial-domain downlink (or uplink) beam prediction for Set A of beams based on measurement results of Set B of beams.

BM-Case2: Temporal downlink (or uplink) beam prediction for Set A of beams based on the historic measurement results of Set B of beams.

Further, for BM-Case1 and BM-Case2, beams in above Set A and Set B may be in the same frequency range (FR) .

In case of BM-Case1 and BM-Case2, the following alternatives for the predicted beams may be supported: downlink transmitting (TX) beam prediction, downlink receiving (RX) beam prediction, beam pair prediction (a beam pair consists of a downlink TX beam and a corresponding downlink RX beam) .

Regarding the sub-use cases of BM-Case1 and BM-Case2, the following alternatives for AI/ML output may be supported: TX and/or RX Beam identity (ies) ID (s) and/or the predicted layer 1 (L1) -reference signal receiving power (RSRP) of the N predicted downlink 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 downlink TX and/or TX beams and other information (e.g., probability for the beam to be the best beam, the associated confidence, beam application time/dwelling time, predicted beam failure) , where N predicted beams may 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, where N predicted beams can be the top-N predicted beams.

In the following, some example implementations about the date collection procedure will be discussed.

In a case of UE-side AI/ML model, the UE may report to the NW about the supported/preferred configurations of DL RS transmission. Further, data collection initiated/triggered by configuration from the network device, or the UE may request for data collection. Signalling/configuration/measurement/report for data collection may indicate assistance information, reference signals, content/type of the collected data, configuration related to Set A and/or Set B, information on association/mapping of Set A and Set B. Further, assistance information from network to UE for UE data collection may be used for categorizing the data for the purpose of differentiating characteristics of the data. The assistance information should preserve privacy/proprietary information.

In a case of NW-side AI/ML model, signalling/configuration/measurement/report for data collection may indicate assistance information, Reference signals. Further, regarding data collection for NW-side AI/ML model regarding the contents of collected data, below alternatives may be considered: M1 layer 1 (L1) reference signal receiving powers (RSRPs) (corresponding to M1 beams) with the indication of beams (beam pairs) based on the measurement corresponding to a beam set, where M1 can be larger than 4, if applicable; M2 L1-RSRPs (corresponding to M2 beams) based on the measurement corresponding to a beam set, where M2 can be larger than 4, if applicable; M3 beam (beam pair) indices based on the measurement corresponding to a beam set, where M3 can be larger than 4, if applicable.

Regarding data collection for NW-side AI/ML model of BM-Case1 and BM-Case2, the following approaches have been identified for overhead reduction: the omission/selection of collected data; the compression of collected data.

Regarding data collection for NW-side AI/ML model of BM-Case1 and BM-Case2, the following reporting signalling for beam-specific aspects maybe applicable: L1 signalling to report the collected data, higher-layer signalling to report the collected data.

In CSI prediction using UE-side model use case, at least the following aspects have been proposed on data collection: signalling and procedures for the data collection (data collection may be indicated by the network or requested by the UE) ; CSI-RS configuration (including assistance information for categorizing the data, if needed) . Further, the provision of assistance information needs to consider feasibility of disclosing proprietary information to the other side.

In the following, some example implementations about the model inference procedure will be discussed. In order to facilitate the AI/ML model inference, below options may be applied: enhanced or new configurations/UE reporting/UE measurement, e.g., enhanced or new beam measurement and/or beam reporting; enhanced or new signalling for measurement configuration/triggering; and signalling of assistance information.

For BM-Case1 and BM-Case2 with a UE-side AI/ML model, an indication of the associated Set A may be transmitted from network to UE, e.g., association/mapping of beams within Set A and beams within Set B if applicable. Further, predicted L1-RSRP (s) corresponding to the DL Tx beam (s) or beam pair (s) .

For BM-Case1 and BM-Case2 with a UE/NW-side AI/ML model, a beam indication may be transmitted from network for UE reception. Further, at least for BM-Case1 with a UE-side AI/ML model, the legacy TCI state mechanism can be used to perform beam indication of beams.

For BM-Case1 and BM-Case2 with a NW-side AI/ML model, the UE may report the measurement results of more than 4 beams in one reporting instance.

For BM-Case1 with a UE-side AI/ML model, an L1 signalling may be used for reporting the following information of AI/ML model inference to NW: the beam (s) that is based on the output of AI/ML model inference. For BM-Case2 with a UE-side AI/ML model, an L1 signalling to report the following information of AI/ML model inference to NW: the beam (s) of N future time instance (s) that is based on the output of AI/ML model inference; information about the timestamp corresponding the reported beam (s) .

For BM-Case2 with a NW-side AI/ML model, the UE may report information about measurements of multiple past time instances in one reporting instance.

Explicit assistance information from UE to network for NW-side AI/ML model may include UE location, UE moving direction, UE Rx beam shape/direction and so on.

Explicit assistance information from network to UE for UE-side AI/ML model may include NW-side beam shape information, e.g., 3dB beamwidth, beam boresight directions, beam shape, Tx beam angle, and so on.

Currently, CSI compressing (such as, spatial-frequency domain CSI compression using two-sided AI model) and time domain CSI prediction are representative sub use cases for AI/ML CSI report enhancement.

For better descriptions, some terms used herein are listed as below:

● AI/ML model: refers to a data driven algorithm that applies AI/ML techniques to generate a set of outputs based on a set of inputs;

● AI/ML model delivery: refers to a generic term referring to delivery of an AI/ML model from one entity to another entity in any manner. Note: An entity could mean a network node/function (e.g., gNB, LMF, etc. ) , UE, proprietary server, and so on;

● Functionality: refers to an AI/ML-enabled feature/feature group (FG) enabled by configuration (s) , where configuration (s) is (are) supported based on conditions indicated by UE capability;

● Functionality-based LCM: operates based on, at least, one configuration of AI/ML-enabled feature/FG or specific configurations of an AI/ML-enabled feature/feature group;

● Model-ID-based LCM: operates based on identified models, where a model may be associated with specific configurations/conditions associated with UE capability of an AI/ML-enabled Feature/FG and additional conditions (e.g., scenarios, sites, and datasets) as determined/identified between UE-side and NW-sider;

● AI/ML model inference: refers to a process of using a trained AI/ML model to produce a set of outputs based on a set of inputs;

● AI/ML model testing: refers to a subprocess of training, to evaluate the performance of a final AI/ML model using a dataset different from one used for model training and validation. Differently from AI/ML model validation, testing does not assume subsequent tuning of the model;

● AI/ML model training: refers to a process to train an AI/ML Model [by learning the input/output relationship] in a data driven manner and obtain the trained AI/ML Model for inference;

● AI/ML model transfer: refers to delivery of an AI/ML model over the air interface in a manner that is not transparent to 3GPP signalling, either parameters of a model structure known at the receiving end or a new model with parameters. Delivery may contain a full model or a partial model;

● AI/ML model validation: refers to a subprocess of training, to evaluate the quality of an AI/ML model using a dataset different from one used for model training, that helps selecting model parameters that generalize beyond the dataset used for model training;

● Data collection: refers to a process of collecting data by the network nodes, management entity, or UE for the purpose of AI/ML model training, data analytics and inference;

● Federated learning/federated training: refers to a machine learning technique that trains an AI/ML model across multiple decentralized edge nodes (e.g., UEs, gNBs) each performing local model training using local data samples. The technique requires multiple interactions of the model, but no exchange of local data samples;

● Functionality identification: refers to a process/method of identifying an AI/ML functionality for the common understanding between the network and the UE. Note: Information regarding the AI/ML functionality may be shared during functionality identification. Where AI/ML functionality resides depends on the specific use cases and sub use cases;

● Model activation: refers to enable an AI/ML model for a specific AI/ML-enabled feature;

● Model deactivation: refers to disable an AI/ML model for a specific AI/ML-enabled feature;

● Model download: refers to transfer a Model from the network to UE;

● Model identification: refers to a process/method of identifying an AI/ML model for the common understanding between the network and the UE. Note: The process/method of model identification may or may not be applicable;

● Information regarding the AI/ML model may be shared during model identification;

● Model monitoring: refers to a procedure that monitors the inference performance of the AI/ML model;

● Model parameter update: refers to a process of updating the model parameters of a model;

● Model selection: refers to a process of selecting an AI/ML model for activation among multiple models for the same AI/ML enabled feature. Note: Model selection may or may not be carried out simultaneously with model activation;

● Model switching: refers to deactivating a currently active AI/ML model and activating a different AI/ML model for a specific AI/ML-enabled feature;

● Model update: refers to a process of updating the model parameters and/or model structure of a model;

● Model upload: refers to transfer a Model from UE to the network;

● Network-side (AI/ML) model: refers to an AI/ML Model whose inference is performed entirely at the network;

● Offline field data: refers to the data collected from field and used for offline training of the AI/ML model;

● Offline training: refers to an AI/ML training process where the model is trained based on collected dataset, and where the trained model is later used or delivered for inference. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as offline training by commonly accepted conventions;

● Online field data: refers to the data collected from field and used for online training of the AI/ML model;

● Online training: refers to an AI/ML training process where the model being used for inference) is (typically continuously) trained in (near) real-time with the arrival of new training samples. Note: the notion of (near) real-time and non real-time are context-dependent and is relative to the inference time-scale. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as online training by commonly accepted conventions. Note: Fine-tuning/re-training may be done via online or offline training. (This note could be removed when we define the term fine-tuning) ;

● Reinforcement Learning (RL) : refers to a process of training an AI/ML model from input (a.k.a. state) and a feedback signal (a.k.a. reward) resulting from the model’s output (a.k.a. action) in an environment the model is interacting with;

● Semi-supervised learning A process of training a model with a mix of labelled data and unlabeled data;

● Supervised learning: refers to a process of training a model from input and its corresponding labels;

● Two-sided (AI/ML) model: refers to a paired AI/ML Model (s) over which joint inference is performed, where joint inference comprises AI/ML Inference whose inference is performed jointly across the UE and the network, i.e., the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa;

● UE-side (AI/ML) model: refers to an AI/ML Model whose inference is performed entirely at the UE;

● Unsupervised learning: refers to a process of training a model without labelled data;

● 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;

● Open-format models: refers to ML models of specified format that are mutually recognizable across vendors and allow interoperability, from 3GPP perspective. They are mutually recognizable between vendors and do not hide model design information from other vendors when shared;

● Measurement result (s) may include but be not limited to, (L1/L3) -reference signal received power (RSRP) , L1/L3) -SINR, (L1/L3) -received signal strengthen indicator (RSSI) , or (L1/L3) -reference signal received quality (RSRQ) ;

● NW/network device: may be an access network device, or a core network device, such as, “Operation Administration and Maintenance (OAM) ” , “server” , “Access and Mobility Management Function (AMF) /Location Management Function (LMF) ” ;

● “Conditions” : 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;

● A beam may refer to 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.

In the present disclosure,

terms of “model” , “feature” “functionality” and “model/functionality/feature” may be used interchangeably;

terms of “ID” , “index” , “indicator” and “identifier” may be used interchangeably;

terms “model” , “model group” , “group of models” , “model set” , “a set of groups” may be used interchangeably;

terms “feature” , “feature group” , “feature of model” , “feature set” , “a set of features” may be used interchangeably;

terms “functionality” , “functionality group” , “group of functionalities” , “functionality set” and “set of functionalities” may be used interchangeably;

the terms “beam” , “precoder” , “precoding” , “precoding matrix” , “beam” , “spatial relation information” , “spatial relation info” , “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 state (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” can be used interchangeably;

terms “time instance, “timestamp” can be used interchangeably.

In the present disclosure, a beam may correspond to a channel state information-reference signal (CSI-RS) , a synchronization signal and physical broadcast channel (PBCH) block (SSB) , a CSI-RS resource, or an SSB resource. Accordingly, a beam identity (ID) may be a CSI-RS resource indicator (CRI) , an SSB resource indicator (SSBRI) , or a RS ID. It also should be understood that in fact, a beam refers to a resource that enables a spatially directional communication, and thus may be identified by other suitable parameter in other embodiments. In present disclosure is not limited in this regard.

As used herein, the term “ML stage” may be replaced by “ML phase” , “ML procedure” , “LCM stage” , “LCM phase” , “LCM procedure” , including but not limited to, model delivery procedure/phase/stage, model inference procedure/phase/stage, model testing procedure/phase/stage, model training procedure/phase/stage, model monitoring procedure/phase/stage, model transfer procedure/phase/stage, model validation procedure/phase/stage, data collection procedure/phase/stage, model learning procedure/phase/stage and so on.

As used herein, a physical model ID may refer to a real AI/ML model, a real implementation; a logical model ID may be associated with one or a group of physical models for the same purpose. Further, the global model ID and the local model ID also may be used to identity a model.

It is noted that when the term “a set/list of” is used, it may mean one or more elements/items, which may be replaced by terms of “at least one” , “a group of” . For example, “a set/list of X” means “at least one X” or “one or more Xs” .

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, a data processing model, an algorithm, a functionality, a procedure, a process, an entity, a function, a feature, a feature group, a model identifier (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.

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.

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.

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.

Principles and implementations of the present disclosure will be described in detail below with reference to the figures.

Example environment

FIG. 1 illustrates a schematic diagram of an example communication environment 100 in which example embodiments of the present disclosure can be implemented. In the communication environment 100, a plurality of communication devices, including a terminal device 110 and a network device 120, can communicate with each other.

Further, multiple input multiple output (MIMO) may be supported in the communication environment 100, such that the network device 120 and the terminal device 110 may communicate with each other via different beams to enable a directional communication.

In the following, for the purpose of illustration, some example embodiments are described with the terminal device 110 operating as a UE. Further, the network device 120 may operate as a base station, or the network device 120 may have an RAN node function and partial core network function (s) . However, in some example embodiments, operations described in connection with a terminal device may be implemented at a network device or other device, and operations described in connection with a network device may be implemented at a terminal device or other device.

In some example embodiments, if the terminal device 110 is a terminal device and the network device 120 is a network device, a link from the network device 120 to the terminal device 110 is referred to as a downlink (DL) , while a link from the terminal device 110 to the network device 120 is referred to as an uplink (UL) . In DL, the network device 120 is a transmitting (TX) device (or a transmitter) and the terminal device 110 is a receiving (RX) device (or a receiver) . In UL, the terminal device 110 is a TX device (or a transmitter) and the network device 120 is a RX device (or a receiver) .

In FIG. 1, the terminal device 110 and the network device 120 may communicate with each other via one or more beams. As illustrated in FIG. 1, the terminal device 110 may communicate with the network device 120 via the beams 130-1 to 130-3. For purpose of discussion, the beams 130-1 to 130-3 are collectively or individually referred to as beam 130. As illustrated in FIG. 1, the network device 120 may communicate with the terminal device 110 via the one or more of beams 140-1, 140-2 and 140-3. For purpose of discussion, the beams 140-1 to 140-3 are collectively or individually referred to as beam 140.

In some embodiments, one or more models may be deployed at the network device 120 and/or the terminal device 110. As illustrated in FIG. 1, the model 115 is deployed at the terminal device 110 and/or the model 125 is deployed at the network device 120. Further, in the example of FIG. 1, the model 115 and/or the model 125 may assist such as BM, i.e., obtain input data and derive related output.

It is to be understood that the number of devices and their connections shown in FIG. 1 are only for the purpose of illustration without suggesting any limitation. The communication environment 100 may include any suitable number of devices configured to implementing example embodiments of the present disclosure.

In some embodiments, the terminal device 110 and the network device 120 may communicate with each other via a channel such as a wireless communication channel on an air interface (e.g., Uu interface) . The 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) . Of course, any other suitable channels are also feasible.

The communications in the communication environment 100 may conform to any suitable standards including, but not limited to, Global System for Mobile Communications (GSM) , Long Term Evolution (LTE) , LTE-Evolution, LTE-Advanced (LTE-A) , New Radio (NR) , Wideband Code Division Multiple Access (WCDMA) , Code Division Multiple Access (CDMA) , GSM EDGE Radio Access Network (GERAN) , Machine Type Communication (MTC) and the like. The embodiments of the present disclosure may be performed according to any generation communication protocols either currently known or to be developed in the future. Examples of the communication protocols include, but not limited to, the first generation (1G) , the second generation (2G) , 2.5G, 2.75G, the third generation (3G) , the fourth generation (4G) , 4.5G, the fifth generation (5G) communication protocols, 5.5G, 5G-Advanced networks, or the sixth generation (6G) networks.

In FIG. 1, the ML model 115/125 may be used for a channel state information (CSI) prediction or a beam prediction in time domain.

FIG. 2 illustrates an example of the inference procedure 200 for beam management for BM-Case1 and BM-Case2.

In FIG. 2, measurements based on Set B of beams are used as model input. In addition, beam ID information may be also provided as input to the AI/ML model. Based on model output (e.g., probability of each beam in Set A to be the Top-1 beam, predicted L1-RSRPs) , Top-1/N beam (s) among Set A of beams can be predicted and/or potentially with predicted L1-RSRPs (depending on the labeling) . For BM-Case 1, the measurements of Set B (otherwise stated) are used as model input to predict Top-1/N beams from Set A, and for BM-Case2, the measurements from historic time instance (s) are used as model input for temporal DL beam prediction of beams from Set A. The cases that Set A and Set B are different (Set B is NOT a subset of Set A) , and Set B is a subset of Set A for both BM-Case1 and BM-Case2, and case that Set A and Set B are the same for BM-Case2 are considered.

Further, for both BM-Case1 and BM-Case2, UE can report the prediction result to NW based on the output of a UE-side model, or NW can predict the Top-1/N beam (s) based on the reported measurements of Set B for a NW-side model.

Further refer to FIG. 3A to FIG. 3C, which illustrate example use cases of beam prediction 300A to 300C.

In FIG. 3A, beam prediction is based on number of measurements/RSs and prediction time, where T2 is the time duration for beam prediction, Mt is the number of time instances for measurement as AI/ML inputs with a periodicity of Tper and Pt is the number of time instance (s) for prediction with a periodicity of Tper in T2. In IFG. 3A, the historical measurement results may be based on number of measurements/RSs and prediction time.

In FIG. 3B, beam prediction is based on a periodicity T of the required reference signals for measurements to achieve a certain beam prediction accuracy. For non-AI baseline, every T=X ms reference signals for measurements are needed. For AI baseline, every T=Y ms, reference signals for measurements are needed. In FIG. 3B, the historical measurement results may be based on a periodicity T of the required reference signals for measurements.

In FIG. 3C, beam prediction is based on Y times of a given minimal periodicity Tper of the reference signals for measurements. For non-AI baseline, UE measures all the reference signals of Set A every Tper. For AI, UE measures the reference signals of Set B every Y times of Tper. In this case, prediction time is defined as the time from each measurement instance to the latest prediction instance before the next measurement instance. In FIG. 3C, the historical measurement results may be based on Y times of a given minimal periodicity Tper of the reference signals for measurements.

In some embodiments, the historical measurement results may be based on an observation window (number/distance) : e.g., 5/5ms, 10/5ms.

Further, “periodicity of Tper” , “T” , “Y” , “distance” may be called as “interval” , “time interval” between two historical measurements, or between two measurements for historical results, and The time duration related to Mt*Tper, Y*Tper, “distance” “distance*number” mentioned above may be call as “measurement window” for historical measurement results.

FIG. 4 illustrates an example of the inference procedure for CSI prediction 400. In FIG. 4, for generating the input of CSI prediction model, it may need some further pre-processing on the measured channel, and for the output of the CSI prediction model, some further post-processing may also be applied.

For AI/ML based CSI prediction sub use case, the following details of models may be considered:

● the structure of the AI/ML model, e.g., type (such as, feedforward neural network (FCN) , recurrent neural network (RNN) , convolutional neural network (CNN) ) , the number of layers, branches, format of parameters and so on;

● the input CSI type, e.g., raw channel matrix, eigenvector (s) of the raw channel matrix, feedback CSI information, assumptions on the observation window (such as, i.e., number/time distance of historic CSI/channel measurements) ;

● the output CSI type, e.g., channel matrix, eigenvector (s) , feedback CSI information assumptions on the prediction window (such as, number/time distance of predicted CSI/channel) and so on;

● data pre-processing/post-processing;

● loss function.

For the input CSI type, both of the following types may be considered for evaluations: raw channel matrices and eigenvector (s) .

Further, the following aspects may be considered as impacting factors for CSI prediction:

● UE speed,

● input/Output type: Raw channel matrix, eigenvectors,

● observation window (number/distance) : e.g., 5/5ms, 10/5ms,

● prediction window (number/distance between prediction instances/distance from the last observation instance to the 1st prediction instance) : e.g., 1/5ms/5ms,

● performance metric for intermediate Key Performance Indicator (KPI) : squared generalized cosine similarity (SGCS) , Normalized mean squared error (NMSE) can be additionally submitted,

● spatial consistency configuration (optional) : procedure A with 50m decorrelation distance and channel updating periodicity of 1 ms.

Example processes

In legacy CSI framework, aperiodic report may be based on aperiodic RS, which is one-shot measurement and report. However, for time-domain prediction of CSI/beam, usually historical measurements are needed (i.e., as model inputs) for AI/ML model to generate outputs (i.e., prediction for the future time instance) . In this event, one-shot measurement cannot provide sufficient inputs for model inference.

Reference is now made to FIG. 5, which illustrates illustrate an example timing 500 for time domain prediction. In FIG. 5, an aperiodic report is configured, and the RS is based on aperiodic resource. It can be seen, as the time domain prediction requires multiple historical results as inputs, the aperiodic report based on aperiodic resource may be problematic since there may be no opportunity to obtain the historical results.

According to some embodiment of the present discourse, the ML model may obtain enough inputs for model inference.

Example embodiments will be discussed with reference to FIG. 6, which illustrates a signaling flow 600 in accordance with some embodiments of the present disclosure. For the purposes of discussion, the signaling flow 600 will be discussed with reference to FIG. 1, for example, by using the terminal device 110 and the network device 120.

It is to be understood that the operations at the terminal device 110 and the network device 120 should be coordinated. In other words, the network device 120 and the terminal device 110 should have common understanding about configurations, parameters and so on. Such common understanding may be implemented by any suitable interactions between the network device 120 and the terminal device 110 or both the network device 120 and the terminal device 110 applying the same rule/policy. In the following, although some operations are described from a perspective of the terminal device 110, it is to be understood that the corresponding operations should be performed by the network device 120. Similarly, although some operations are described from a perspective of the network device 120, it is to be understood that the corresponding operations should be performed by the terminal device 110. Merely for brevity, some of the same or similar contents are omitted here.

ML model deployed at the terminal device

Generally speaking, by using an aperiodic report, the network device 120 may obtain the prediction result in a more flexibility, timely manner. Further, for AI/ML at UE side, the aperiodic report may suggest/trigger aperiodic model inference, i.e., UE does not always run AI/ML model till it’s been triggered to do so, which reduces complexity and power caused by the AI/ML inference.

When aperiodic report is triggered, the terminal device 110 needs to provide predicted CSI/beam of future time instances based on historical measurement results. However, according to the legacy method, there may only be one resource for measurement between the trigger and the report. The problem is more severe if the aperiodic report is based on aperiodic resource.

One possible solution may be trying to have a long time duration between trigger and report, such that the UE may collect sufficient number of measurement results.

Another possible solution may be providing second resource for the report and asking UE to store the previous measurement results (as shown in FIG. 7) . FIG. 7 illustrates an example procedure 700 for prediction. In FIG. 7, UE may store the measurement results on the second resources and such measurement results may be used for model inference. It should be understood that, in some cases, part of the measurement results may be obtained after the trigger, especially when the second resources is semi-persistent or periodic.

A further possible solution may be that the UE does not expect to be configured with aperiodic report, or be configured with aperiodic report based aperiodic resource, for UE-side time domain prediction.

In the following embodiments, the ML model is deployed at the terminal device 110, where data collection of model inputs and model inference is performed by the terminal device 110.

In some embodiments, the ML model is used for a channel state information (CSI) prediction or a beam prediction in time domain.

In operation, the terminal device 110 receives 640 control information (such as, DCI or MAC CE) of a report from the network device 120, where the report is based on at least one first resource.

For example, the network device 120 needs to terminal device 110 to report the ML-based prediction results, and the network device 120 trasnmits a DCI to the terminal device 110, where the DCI may indicate one or more trigger states.

As illustrated in FIG. 6, the network device 120 may transmit 650 reference signals (such as, SSB, CSI-RS) to the terminal device 110 on the at least one first resource. Accordingly, the terminal device 110 performs at least one measurement on the at least one first resource.

After that, the terminal device 110 transmits 660 the report to the network device 120, where the report comprising at least one prediction result which is obtained by and based on the following:

● at least one first measurement result measured on the at least one first resource, and

● at least one second measurement result measured on at least one second resource associated with the first resource.

In some embodiments, at least part of the at least one second measurement result may be historical measurements. If so, as illustrated in FIG. 6, the network device 120 may transmit 630 reference signals (such as, SSB, CSI-RS) to the terminal device 110 on the second resource. The terminal device 110 may perform measurement in the second resource and obtain the second measurement result.

In some embodiments, if the at least one first resource is periodic resource or semi-persistent resource, the at least one second resource may be at least one historical transmission occasion of the periodic resource or semi-persistent resource.

It can be seen, by using at least one second measurement result measured on at least one second resource associated with the first resource, the ML model deployed at the terminal device 110 may perform an inference procedure even if the at least one first measurement result measured on the at least one first resource is not sufficient for perform model inference.

In some embodiments, the control information may be used for activating the ML model or activating a model inference of the ML model.

In some embodiments, the control information may further indicate model-related information of the ML model. Further, in some embodiments, the model-related information may be information about the model/functionality/feature, e.g., model ID, functionality identification, feature ID and so on.

In some embodiments, the report may be aperiodic, the at least one first resource may be an aperiodic resource and the at least second resource may be a semi-persistent resource or periodic resource.

In some embodiments, the terminal device 110 may receive configuration information 620 from the network device 120.

In some embodiments, the configuration information may indicate the association between the at least one first resource and the at least one second resource.

In some embodiments, the at least one first resource and the at least one second resource associate with a same trigger state.

Alternatively, in some embodiments, an identity of the at least one second resource may be linked to an identity of the at least one first resource.

Alternatively, in some embodiments, a first time offset between the least one first resource and the at least one second resource may be smaller than or equal to a threshold offset.

Alternatively, in some embodiments, the configuration information may indicate at least one trigger state associated with the at least one first resource and the at least one second resource.

Alternatively, in some embodiments, the configuration information may indicate at least one report configuration associated with the at least one first resource and the at least one second resource.

Alternatively, in some embodiments, the configuration information may indicate model-related information of the ML model. Further, in some embodiments, the model-related information may be information about the model/functionality/feature, e.g., model ID, functionality identification, feature ID and so on.

In some embodiments, the terminal device 110 may receive a trigger state used for triggering a measurement on the at least one second resource.

Alternatively, in some embodiments, the terminal device 110 may receive at least one trigger state each of which used for triggering a measurement on one of the at least one second resource.

In some embodiments, the terminal device 110 may store the at least one second measurement result measured on the at least one second resource.

In some embodiments, the at least one second measurement result may be stored in a variable or log of the terminal device 110.

In some embodiments, the variable or log may be maintained by the terminal device 110 in a first in first out manner.

In some embodiments, if the stored measurement results reach the limit of UE buffer, UE may drop the store measurement results, or UE may report the stored measurement results.

Additionally, in some embodiments, the terminal device 110 may receive an indication from the network device 120, where the indication indicating the terminal device 110 to store one or more of the second measurement results.

In some embodiments, prior to receiving the control information, the terminal device 110 may transmit an indication to the network device 120, the indication indicating a number of measured or stored second measurement results is equal to or larger than a threshold number. Alternatively, in some embodiments, the prior to receiving the control information, the terminal device 110 may transmit an indication to the network device 120, the indication indicate a number of the measured or stored second measurement results.

In some embodiments, the number of second resources may be determined based on at least one of the following:

● a number of measurement results required for determining the at least one prediction result,

● a number of the at least one first resource,

● a measurement window before a time point of a reception of the control information, or

● a capability of the terminal device 110.

In some embodiments, the at least one first resource and the at least one second resource may associate with at least one of the following:

● at least one same measurement parameter,

● at least one same transmitting parameter, or

● at least one same receiving parameter.

Alternatively, in some embodiments, if the at least one first resource and the at least one second resource are transmitted with different setup, the difference may be provided to the terminal device 110 by the network device 120.

In some embodiments, first quasi co-location information associated with the at least one first resource is the same as second quasi co-location information associated with the at least one second resource.

In some embodiments, if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the terminal device 110 may determine at least one third resource, where the number of at least one third resource is determine based on at least one of the following:

● a number of the at least one first resource,

● a number of the at least one second resource, or

● a capability of the terminal device 110.

In some embodiments, if the at least one first resource is aperiodic resource, the at least one third resource may be at least one repetition of the at least one first resource.

Alternatively, in some embodiments, if the at least one first resource is a periodic resource or semi-persistent resource, the at least one third resource and the at least one first resource may belong to a same resource set and the at least one third resource may be at least one transmission occasion of the periodic resource or semi-persistent resource.

In some embodiments, the control information further may indicate a second time offset used for determining a resource for transmitting the report, where the second time offset starts timing from one of the following:

● a time point of a reception of the control information,

● a time point of the first transmission occasion of the at least one first resource,

● a time point of the completion of prediction, or

● a time point of the last transmission occasion of the at least one first resource.

Optionally, in some embodiments, the terminal device 110 may indicate its capability to the network device 120, such that the network device 120 may make proper configuration for the terminal device 110.

As illustrated in FIG. 6, the terminal device 110 may transmit 610 capability-related information to the network device 120, where the capability-related information comprises at least one of the following:

● whether the terminal device 110 supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model,

● whether the terminal device 110 supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model,

● whether the terminal device 110 supports to store measurement results for transmitting a report comprising at least one prediction result,

● a maximum number of measurement results stored by the terminal device 110 for an inference procedure of the ML model,

● a number of repetitions of aperiodic resources needed by the terminal device 110 for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model,

● a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model,

● a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource,

● a time length of a measurement time window supported by the terminal device 110, wherein measurement results obtained within the measurement time window are stored and used for an inference procedure of the ML model.

According to some embodiments of the present discourse, the network device make provide a proper configuration to avoid the scenario where the terminal device 110 cannot obtain enough measurement results to perform inference.

Specifically, in some embodiments, the network device 120 may configure at least one first resource to be semi-persistent or periodic.

Alternatively, or in addition, in some embodiments, the network device 120 may not configure the at least one first resource to be aperiodic.

Accordingly, in some embodiments, the terminal device 110 may expect the at least one first resource indicated by the control information is semi-persistent or periodic.

Alternatively, or in addition, in some embodiments, the terminal device 110 may not expect the at least one first resource indicated by the control information is aperiodic.

For a better understanding, more detail about the above procedure will be discussed in the following, which also will be discussed with reference to FIG. 6.

Optionally, the network device 120 may provide 620 configuration information to the terminal device, such as, via RRC signalling.

In some embodiments, the configuration information may include report-related configuration. The report-related configuration may include information on the time domain behavior of the report, such as, periodic, semi-persistent resource, or aperiodic.

The report-related configuration also may include information on report quantity, such as, the report at least contains predicted CSI/beam for future time instance, where the exact quantity is model/functionality-specific, e.g., based on specific AI/ML model output of the applied/supported AI/ML model. Further, for time domain CSI prediction, predicted CSI or target CSI may be included in the report. For time domain beam prediction, predicted beam or predicted beam and predicted RSRP may be included in the report.

The report-related configuration also may include information on other parameters on the report content, for example, the number of future time instances to be reported, the number of RS per future time instance to be reported and so on.

In some embodiments, the configuration information may include resource-related configuration.

The resource-related configuration may include information on time domain behavior of the resources, such as, periodic, semi-persistent resource, or aperiodic.

The resource-related configuration may include information on resource (s) corresponding to the first resource. In some embodiments, the first resource may be a CSI-RS resource, or a SS/PBCH block, or a set of CSI-RS resources, or SSBs. In some embodiments, the first resource may be aperiodic. In some embodiments, for time domain CSI prediction, the set of CSI-RS resources may contain only one CSI-RS resource. In some embodiments, for time domain beam prediction, the set of CSI-RS resources corresponds to the beams in set B.

The resource-related configuration also may include information on resource (s) corresponding to the second resource. In some embodiments, the second resource may be a set of CSI-RS resources or SS/PBCH blocks.

In some embodiments, if the first resource is aperiodic, the second resource may be periodic or semi-persistent.

In some embodiments, quasi co-location (QCL) information may be provided for periodic resource or semi-persistent resources in the resource configuration for the terminal device 110 to measure.

In some embodiments, the configuration information may include information on the association between first resource and second resource, which may be configured explicitly or implicitly.

In one example, the association may be established by linking both the first resource and second resource to a same CSI trigger state.

In another example, the association may be established by a time-offset between first resource and second resource, where the time-offset is smaller than a threshold offset.

In another example, the association may be established based on the RS ID, resource ID, or resource set ID, such as, the related ID of the second/first resource is linked to ID of the first/second resource.

In some embodiments, if the first resource is periodic or semi-persistent, the second resource may be the previous transmitted occasions/instances of the same resource.

In some embodiments, the configuration information may include trigger state configuration. As one example, the trigger state configuration may include a list of trigger states, where each trigger state may be configured with one or multiple report configurations.

In some embodiments, at least one of report configuration may be configured with first resource and the second resource if any. Alternatively, in some embodiments, more than one trigger states may be activated simultaneously.

In some embodiments, if the first resource is aperiodic CSI-RS resource, QCL information may be provided in the trigger state for UE to measure.

In some embodiments, QCL information for aperiodic CSI-RS resource configured in the trigger state may be aligned with QCL information for periodic/semi-persistent CSI- RS resource provided in the resource configuration.

In some embodiments, the configuration information may include AI/ML related configuration, such as, AI/ML functionality/model related information (e.g., functionality/model ID) , required model inputs and/or model outputs.

In some embodiments, model/functionality related information may be configured in the trigger state.

In the following, how to trigger the first/second resource will be discussed.

Reference is now made to FIG. 8, which illustrates an example procedure 800 for prediction. In FIG. 8, a first trigger state is used to trigger the second resources for measurement, and a second trigger state is used to trigger report of the prediction results. In some embodiments, the first trigger state and the second trigger state may be associated. Alternatively, in some embodiments, the second trigger state may be transmitted based on a UE indication indicating that sufficient historical measurement results have been collected by the terminal device 110.

Reference is now made to FIG. 9, which illustrates an example procedure 900 for prediction. In FIG. 9, a first trigger state is transmitted multiple times (e.g., configured with a repetition value K, or configured with K associated different trigger states as the first trigger state) . Further, for the first K-1 times, the trigger state is only to trigger resource for measurements, and for the last time, the trigger state is to trigger report of the prediction results.

In order to ensure the terminal device 110 may perform valid model inference, the terminal device 110 needs to obtain some historical measurement results, which will be discussed as below.

As illustrated in FIG. 6, the network device 120 may transmit 630 RS on the second resource and the terminal device 110 may performs measurements on the second resource.

In some embodiments, the terminal device 110 may obtain/maintain/keep/store the second measurement results measured in the second resources, which may be used later as part of model inputs (i.e., historical results) if related report is triggered.

The number of second measurement results that the terminal device 110 needs to keep is discussed as below. In some embodiments, the terminal device 110 may obtain K-1 second measurement results (where K is the requirement of the corresponding AI/ML model) of the latest measurement instances, and another measurement result may be obtained based on the first resource.

In some embodiments, the terminal device 110 may obtain K-k1 second measurement results (where K is the requirement of the corresponding AI/ML model) of the latest measurement instances, and the other k1 measurement results may be obtained based on the first resources.

In some embodiments, the terminal device 110 may obtain the second measurement results in a measurement window.

In some embodiments, the second measurement results may be kept in the variable or log of the terminal device 110.

In some embodiments, as the storage size of variable or log is limited, the terminal device 110 may release/delete/remove first K’ stored measurement results after related report or after related prediction, where K’ is a number smaller than or equal to the storage size. In some embodiments, the variable or log may be maintained in a first in first out manner.

In the following, network device 120 may activate/trigger 640 an aperiodic report.

In some embodiments, if aperiodic report is triggered by DCI, a CSI request field may be included in the DCI. Additionally, the number of bits of the CSI request field depends on the number of trigger states configured.

In some embodiments, the trigger (i.e., DCI) also may be used to activate the UE-side AI/ML model, or activate the UE-side AI/ML model inference. In this event, the model/functionality-related information may be included in the DCI.

As shown in FIG. 6, upon the control information, the terminal device 110 may perform measurement 650 on the first resource (s) .

In some embodiments, if the first resource is aperiodic, the terminal device 110 may expect that the first resource is configured with the associated second resources.

In some embodiments, if the first resource is periodic or semi-persistent, the second resource may be the previous transmitted occasions of the same resource.

In some embodiments, the terminal device 110 may assume that the first resource and the second resource are configured with the same parameters/properties, or transmitted/received with the same setup for one or many of the following:

● RS density, antenna port (s) , the number of antenna ports, Tx power,

● TCI state, QCL information, Rx beam (s) ,

● the number of resources,

● Tx beams, Tx beam pattern, the mapping/relationship between set B and set A.

Alternatively, if they are transmitted with different setups, the difference between the setups may be provided to terminal device 110.

Alternatively, in some embodiments, if the terminal device 110 cannot support obtaining sufficient measurements before the trigger, the network device 120 may trigger multiple first resources. Reference is now made to FIG. 10 and FIG. 11, which illustrate example procedures 1000 and 1100 for prediction. In FIG. 10, no historical measurement results have been obtained. In FIG. 11, the obtained historical measurement results (together with the measurement result on the second resource) are not sufficient for model inference. In these event, more measurement results need to be obtained after the trigger, such as, triggering multiple first resources (also referred to as third resources) .

In some embodiments, if the first resource is aperiodic, then multiple first resource may correspond to a set of repeated resource, for example, in FIG. 10, resource #1 to resource #K corresponds to the same RS, or, resource #1 repeated K times.

In addition, as shown in FIG. 11, the number of repetitions of the first resource is k1, where K_store is the terminal device 110 capability (or the terminal device 110 request) on the number of first resource it requires for model inference. Alternatively, in some embodiments the number of repetitions of the first resource is (K -K_store) , where K_store is the terminal device 110 capability (or the terminal device 110 feedback) on the number of stored historical measurement results.

In some embodiments, if the first resource is periodic or semi-persistent, the multiple first resource may correspond to multiple transmission occasions/instances of the same resource (s) .

In the following, the terminal device 110 performs model inference. Specifically, the model inputs may include the measurement results of the first resource and the second resource, and the model outputs may include the predictions of future time instances.

In some embodiments, for time domain CSI prediction, the model inputs may be K (K≥1) historical CSI, which may be in following form (s) : precoding matrix, raw channel matrix, in spatial-frequency domain, using angular-delay domain projection, rank indicator (RI) , PMI, channel quality indicator (CQI) .

In some embodiments, for time domain CSI prediction, the model outputs may be predicted CSI of F (F≥1) future time instances, which may be in following form (s) , precoding matrix, Raw channel matrix, in spatial-frequency domain, using angular-delay domain projection, RI, PMI, CQI.

In some embodiments, for time domain beam prediction, the model inputs may be measurement results of K (K≥1) latest measurement instances, which may be in following form (s) :

● 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 ID.

In some embodiments, for time domain beam prediction, 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.

In some embodiments, the terminal device 110 reports 650 the prediction results.

In some embodiments, a time duration is needed for the terminal device 110 to perform measurement, e.g., T1, to perform model inference, e.g., T2, to prepare reports, e.g., T3, respectively. In view of this, for aperiodic report, the report configuration may also provide a time offset value for report.

In some embodiments, the offset may be from trigger to report, or from first/last transmission occasion/instance of first resource to report.

In some embodiments, if multiple trigger states are used, the offset may be from the first/last trigger to the report.

In some embodiments, the trigger to report offset value shall be larger or equal than the sum of T1, T2 and T3.

● Whether the terminal device 110 can support aperiodic report based on aperiodic resource for model inference;

● Whether the terminal device 110 can support aperiodic report based on aperiodic resource and associated periodic/semi-persistent resource for model inference;

● Whether the terminal device 110 can keep the measurement results for associated periodic/semi-persistent resources for model inference;

● The number of measurement results the terminal device 110 can keep for associated periodic/semi-persistent resources for model inference;

● The number of repetitions of aperiodic resources the terminal device 110 needs for aperiodic report based aperiodic resources for model inference;

● The number of trigger states can be associated with aperiodic resource for model inference;

● The number of trigger states can be associated with aperiodic resource and associated periodic/semi-persistent resource;

● The supported length of measurement time window for model inference.

In some embodiments, the capability may be reported for time-domain CSI prediction and time-domain beam prediction respectively.

In some embodiments, the capability may be reported for each model/functionality respectively.

In some embodiments, the capability may be reported for the terminal device 110 side prediction and the network device 120 side prediction respectively.

ML model deployed at the network device 120

In the following embodiments, the ML model is deployed at the network device 120. Further, for the NW-side model, for data collection of model inputs to be used for the NW-side model inference, an aperiodic report is configured.

When aperiodic report is triggered, the terminal device 110 needs to provide the network device 120 sufficient historical measurement results as model inputs. However, when aperiodic report is triggered, there may only be one report based on the newest/latest measurement, according to legacy method.

One possible solution may be that the aperiodic report is configured with associated report, e.g., periodic or semi-persistent report (as shown in FIG. 12, which illustrates an example procedure 1200 for prediction) . Additionally, resource for associated second report is also associated with resource for aperiodic report.

Another possible solution may be providing second resource and asking the terminal device 110 to store the previous measurement results.

A further possible solution may be allowing a long time duration between trigger and report, such that the terminal device 110 may collect sufficient number of measurement results.

A further possible solution may be that the UE does not expect to be configured with aperiodic report to provide K1 historical measurement results for model inputs for the network device 120 side time domain prediction.

Some example embodiments will be discussed with reference to FIG. 6. In operation, the terminal device 110 receives 640 control information of a report from a network device 120, where the report is based on at least one first resource.

For example, the network device 120 decides to perform inference and thus trasnmits a DCI to the terminal device 110, such that the terminal device 110 may report related measurement results.

After that, based on the control information, the terminal device 110 transmit 660 the following to the network device 120:

● at least one second measurement result measured on at least one second resource.

In some embodiments, the at least one first measurement result may be comprised in a first report, the at least one second measurement result may be comprised in one or more second report associated with the at least one first report.

Alternatively, in some embodiments, the at least one first measurement result and the at least one second measurement result may be comprised in one third report.

In some embodiments, the first report may be an aperiodic reporting, and the at least second report may be a semi-persistent reporting or periodic reporting.

In some embodiments, terminal device 110 may receive 620 configuration information from the network device 120.

In some embodiments, the configuration information may indicate association between the first report and the at least one second report.

Alternatively, in some embodiments, the configuration information may indicate association between the first report and the at least one second report.

Alternatively, in some embodiments, the configuration information may indicate at least one trigger state associated with the first report and the at least one second report.

Alternatively, in some embodiments, the configuration information may indicate a number of the at least one first measurement result.

Alternatively, in some embodiments, the configuration information may indicate a number of the at least one second measurement result.

In some embodiments, the number of first measurement results configured to be comprised in the first report is a first number, and a number of the at least one second measurement result is associated with at least one of the following:

● the first number, or

● a capability of the terminal device 110.

In some embodiments, prior to receiving the control information, the terminal device 110 may perform a measurement on the at least one second resource and store the at least one second measurement result measured on the at least one second resource. Then, after receiving the control information, the terminal device 110 may perform a measurement on the at least one first resource to obtain the at least one first measurement result, and then may transmit the third report to the network device 120, the third report comprising the at least one first measurement result and the at least one second measurement result.

In some embodiments, the terminal device 110 may receive a second trigger state used for triggering a measurement on the at least one second resource, and a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.

In some embodiments, measurements on the at least one first resource and the at least one second resource may be triggered by a plurality of different linked trigger states.

Alternatively, in some embodiments, measurements on the at least one first resource and the at least one second resource may be triggered by a first trigger state and at least one associated trigger state of the first trigger state.

In some embodiments, the control information may indicate the first trigger state, and each of the at least one associated trigger state may be a repetition of the first trigger state.

In some embodiments, the at least one first measurement result and the at least one second measurement result associate with at least one of the following:

● at least one same measurement parameter,

● at least one same transmitting parameter, or

● at least one same receiving parameter.

● whether the terminal device 110 supports to report an aperiodic report for an inference procedure of the ML model,

● whether the terminal device 110 supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report,

● whether the terminal device 110 supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model,

● whether the terminal device 110 supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model,

● a number of repetitions of aperiodic resources needed by the terminal device 110 for an inference procedure of the ML model,

● a number of trigger states associated with an aperiodic report supported by the terminal device 110,

● a maximum number of trigger states associated with an aperiodic report supported by the terminal device 110,

● a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.

According to some embodiments of the present discourse, the network device make provide a proper configuration to avoid the terminal device 110 cannot have enough measurement results to perform inference.

Specifically, in some embodiments, the network device 120 may configure the first report to be a semi-persistent or periodic reporting.

Alternatively, or in addition, in some embodiments, the network device 120 may not configure the first report to be an aperiodic reporting if the at least one first resource is aperiodic.

Accordingly, in some embodiments, the terminal device 110 may expect the first report is configured to be a semi-persistent or periodic reporting.

Alternatively, or in addition, in some embodiments, the terminal device 110 may not expect the first report is configured to be an aperiodic reporting if the at least one first resource is aperiodic.

Alternatively, or in addition, in some embodiments, the terminal device 110 may not expect the at least one first resource is configured to be aperiodic.

The report-related configuration also may include information on report configuration for the first report and the associated second report.

In some embodiments, if the first report is aperiodic, the associated second report may be configured and the associated second report can be periodic/semi-persistent report.

In some embodiments, the configuration information may include the association between aperiodic report and periodic/semi-persistent report, which may be configured explicitly or implicitly.

In one example, the association may be established by linking both first resource and associated second report to the same CSI trigger state. In another example, the association may be established based on the report configuration ID, RS ID, resource ID, or resource set ID.

In some embodiments, if the first report is also periodic or semi-persistent, the associated second report may be the previous transmitted report.

The report-related configuration also may include information on report quantity for first report/associated second report. For example, the CSI/beam report as model inputs used for the network device 120 side model inference.

In some embodiments, the configuration information may include resource configuration, such as, the time domain behavior of the resources: periodic, semi-persistent, or aperiodic.

In some embodiments, the resource configuration may include information on the resource (s) for the first report.

In some embodiments, the first resource is aperiodic, then the first resource is also associated with the resource (s) configured for the associated second report (which may be periodic, semi-persistent) .

In some embodiments, the first resource is periodic, semi-persistent, then resource (s) configured for the associated second report may be the previous transmitted occasions/instances of the same resource.

In some embodiments, the resource configuration may include information on resource (s) for the associated second report, i.e., the second resource. The details of the second resource have been fully discussed in example embodiments of UE-side ML model. For brevity, the same or similar contents are omitted here.

In some embodiments, the configuration information may include trigger state configuration. As one example, the trigger state configuration may include a list of trigger states, where at least one of the trigger states is two report configurations: first report and associated second report.

In some embodiments, the first report may be configured with first resource and the second resource if any. Alternatively, the associated second report may be configured with the second resource.

Alternatively, in some embodiments, the network device 120 may configure the first report to include multiple measurement results.

Reference is now made to FIG. 13, which illustrates an example procedure 1300 for prediction. Example report configurations for this embodiment are discussed below. In some embodiments, the first report includes k2 measurement results, where k2 is the number of measurement results can be obtained based on the first resource, and the associated second report which provides the network device 120 (K -k2) measurement results .

Example resource configurations for this embodiment are discussed below. In some embodiments, the first resource for the first report may be, for example, k2 repetition of the first resource transmission, and the second resources may be configured for the associated second report.

Alternatively, in some embodiments, the network device 120 may configure the first report to include sufficient historical measurement results (as shown in FIG. 14, which illustrates an example procedure 1400 for prediction) . Example report configurations for this embodiment are discussed below. In some embodiments, the report includes K historical measurement corresponds to K latest transmission occasions/instances of the same resource (s) . Additionally, the report quantity may be historical CSI/beam information, the number (e.g., K) to be reported.

Example resource configurations for this embodiment are discussed below. In some embodiments, the resource (s) for the first report may consist the first resource and the second resources.

Additionally, in some embodiments, if the first resource is aperiodic, the second resource may be configured for (K-1) measurement results, which may be periodic or semi-persistent.

In some embodiments, if the first RS is periodic or semi-persistent, the report includes K historical measurement corresponds to K latest transmission occasions/instances of the same resource (s) .

In the following, how to trigger the first/second resource will be discussed.

In some embodiments, more than one trigger states may be activated simultaneously.

Reference is now made to FIG. 15, which illustrates an example procedure 1500 for prediction. In FIG. 15, a first trigger state is used to trigger the second resources for measurement, and a second trigger state is used to trigger report of the prediction results. The first trigger state and the second trigger state can be associated. Or the second trigger state is based on the terminal device 110 indication on the sufficient historical measurement results have been collected.

Reference is now made to FIG. 16, which illustrates an example procedure 1600 for prediction. In FIG. 16, a first trigger state is transmitted multiple times (e.g., configured with a repetition value K, or configured with K associated different trigger states as the first trigger state) , and for the first K-1 times, the trigger state is only to trigger resource for measurements, and for the last time, the trigger state is to trigger report of the prediction results.

In order to ensure the network device 120 may perform valid model inference, the terminal device 110 needs to obtain some historical measurement results, which will be discussed as below.

As for the terminal device 110, if the associated second report is configured, the terminal device 110 performs measurements for the associated second report, and provide the network device 120 the configured report.

Then, the network device 120 may obtain/maintain/keep/store the measurement results, which may be used later as part of model inputs (i.e., historical results) .

Alternatively, if the associated second report is not configured, the terminal device 110 may obtain/maintain/keep/store the measurement results for later report (i.e., first report) .

In the following, the network device 120 may activate/trigger 640 an aperiodic report.

In some embodiments, if the first report is aperiodic, it may be triggered by DCI, where a CSI request field is included in the DCI.

In some embodiments, if the first report is semi-persistent, it may be activated by MAC CE, a report configuration ID may be included in the MAC CE.

As shown in FIG. 6, upon the control information, the terminal device 110 may perform measurement 650 on the first resource (s) , and provide 660 the network device 120 with the configured first report.

In some embodiments, if the first resource is aperiodic, the terminal device 110 may expect that it is configured with the first associated RS.

Alternatively, in some embodiments, the network device 120 may trigger multiple first resource (e.g., k2) .

In some embodiments, the terminal device 110 may assume that the first resource and the resource associated with the associated second report, or the second resource are configured with the same parameters/properties, or transmitted/received with the same setup for one or many of the following:

● RS density, antenna port (s) , the number of antenna ports, Tx power,

● TCI state, QCL information, Rx beam (s) ,

● the number of resources,

Then, the network device 120 performs model inference. In the inference procedure, the model inputs may include the measurement results of the first resource and the second resource, and the model outputs may include the predictions of future time instances.

In some embodiments, the network device 120 take corresponding implementation based on the prediction results.

In some embodiments, the network device 120 may adopt the prediction for better scheduling, faster link adaptation, more suitable beam selection and so on.

Additionally, the network device 120 may provide (part of) prediction information to the terminal device.

● Whether the terminal device 110 can support aperiodic report for model inference;

● Whether the terminal device 110 can support aperiodic report for model inference if aperiodic report is configured with associated periodic/semi-persistent report;

● Whether the terminal device 110 can report the measurement results for associated periodic/semi-persistent resources for model inference;

● The number of measurement results the terminal device 110 can report for associated periodic/semi-persistent resources for model inference;

● The number of trigger states can be associated with aperiodic report for model inference;

● The number of trigger states can be associated with aperiodic report and associated periodic/semi-persistent report for model inference.

Restrictions on configuration

Below table 1 illustrates example restriction configuration, which defines triggering/activation of reporting for the possible resource Configurations for AI/ML based time domain prediction.

Table 1 Triggering/Activation of Reporting for the possible resource Configurations for AI/ML based time domain prediction

According to the above restriction configuration, the aperiodic resource is provided associated periodic/semi-persistent resource, and/or the aperiodic report is provided associated periodic/semi-persistent report.

In some embodiments, when the CSI request field on a DCI triggers a CSI report (s) for predicted CSI/beam, if the resource configuration does not include periodic/semi-persistent resources, the terminal device 110 is not required to perform model inference for CSI/beam prediction.

Example methods

FIG. 17 illustrates a flowchart of a communication method 1700 implemented at a terminal device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1700 will be described from the perspective of the terminal device 110 in FIG. 1.

At block 1710, the terminal device receives, from a network device, control information of a report which is based on at least one first resource.

At block 1720, the terminal device performs at least one measurement on the at least one first resource.

At block 1730, the terminal device transmits the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In some example embodiments, the terminal device may receive, from the network device, configuration information indicating at least one of the following: association between the at least one first resource and the at least one second resource, at least one trigger state associated with the at least one first resource and the at least one second resource, at least one report configuration associated with the at least one first resource and the at least one second resource, or model-related information of the ML model.

In some example embodiments, the at least one first resource and the at least one second resource associate with a same trigger state, an identity of the at least one second resource is linked to an identity of the at least one first resource, or a first time offset between the least one first resource and the at least one second resource is smaller than or equal to a threshold offset.

In some example embodiments, the terminal device may receive, from the network device, one of the following: a trigger state used for triggering a measurement on the at least one second resource, at least one trigger state each of which used for triggering a measurement on one of the at least one second resource.

In some example embodiments, the terminal device may store the at least one second measurement result measured on the at least one second resource.

In some example embodiments, prior to receiving the control information, the terminal device may transmit an indication to the network device, the indication indicating a number of measured or stored second measurement results is equal to or larger than a threshold number.

In some example embodiments, a number of second resources is determined based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a measurement window before a time point of a reception of the control information, or a capability of the terminal device.

In some example embodiments, the at least one first resource and the at least one second resource associate with at least one of the following: at least one same measurement parameter, at least one same transmitting parameter, or at least one same receiving parameter.

In some example embodiments, if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the terminal device may determine at least one third resource, wherein the number of at least one third resource is determine based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a number of the at least one second resource, or a capability of the terminal device.

In some example embodiments, if the at least one first resource is aperiodic resource, the at least one third resource is at least one repetition of the at least one first resource, if the at least one first resource is a periodic resource or semi-persistent resource, the at least one third resource and the at least one first resource belong to a same resource set and the at least one third resource is at least one transmission occasion of the periodic resource or semi-persistent resource.

In some example embodiments, the control information further indicates a second time offset used for determining a resource for transmitting the report, wherein the second time offset starts timing from one of the following: a time point of a reception of the control information, a time point of the first transmission occasion of the at least one first resource, a time point of the completion of prediction, or a time point of the last transmission occasion of the at least one first resource.

In some example embodiments, the terminal device may transmit, to the network device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to store measurement results for transmitting a report comprising at least one prediction result, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource, a time length of a measurement time window supported by the terminal device, wherein measurement results obtained within the measurement time window are stored and used for an inference procedure of the ML model.

In some example embodiments, the report is aperiodic, the at least one first resource is an aperiodic resource and the at least second resource is a semi-persistent resource or periodic resource.

In some example embodiments, first quasi co-location information associated with the at least one first resource is the same as second quasi co-location information associated with the at least one second resource.

In some example embodiments, if the at least one first resource is periodic resource or semi-persistent resource, the at least one second resource is at least one historical transmission occasion of the periodic resource or semi-persistent resource.

In some example embodiments, the at least one second measurement result is stored in a variable or log of the terminal device.

In some example embodiments, the variable or log is maintained by the terminal device in a first in first out manner.

In some example embodiments, the ML model is used for a channel state information (CSI) prediction or a beam prediction in time domain.

In some example embodiments, the control information is used for activating the ML model or activating a model inference of the ML model.

In some example embodiments, the control information further may indicate model- related information of the ML model.

In some example embodiments, the terminal device may expect the at least one first resource indicated by the control information is semi-persistent or periodic, or not expect the at least one first resource indicated by the control information is aperiodic.

FIG. 18 illustrates a flowchart of a communication method 1800 implemented at a terminal device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1800 will be described from the perspective of the terminal device 110 in FIG. 1.

At block 1810, the terminal device receives, from a network device, control information of a report which is based on at least one first resource.

At block 1820, the terminal device may transmit, to the network device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In some example embodiments, the terminal device may receive, from the network device, configuration information indicating at least one of the following: association between the first report and the at least one second report, at least one trigger state associated with the first report and the at least one second report, a number of the at least one first measurement result, a number of the at least one second measurement result, or model-related information of the ML model.

In some example embodiments, a number of first measurement results configured to be comprised in the first report is a first number, and a number of the at least one second measurement result is associated with at least one of the following: a number of measurement results required for determining the at least one prediction result, the first number, or a capability of the terminal device.

In some example embodiments, prior to receiving the control information, perform a measurement on the at least one second resource and store the at least one second measurement result measured on the at least one second resource; and after receiving the control information, the terminal device may perform a measurement on the at least one first resource to obtain the at least one first measurement result; and transmit the third report to the network device, the third report comprising the at least one first measurement result and the at least one second measurement result.

In some example embodiments, the terminal device may receive, from the network device, the following: a second trigger state used for triggering a measurement on the at least one second resource, and a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.

In some example embodiments, measurements on the at least one first resource and the at least one second resource is triggered by a plurality of trigger states comprising one of the following: a plurality of different linked trigger states, or a first trigger state and at least one associated trigger state of the first trigger state.

In some example embodiments, the control information indicates the first trigger state, and each of the at least one associated trigger state is a repetition of the first trigger state.

In some example embodiments, the at least one first measurement result and the at least one second measurement result associate with at least one of the following: at least one same measurement parameter, at least one same transmitting parameter, or at least one same receiving parameter.

In some example embodiments, the terminal device may transmit, to the network device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report, whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an inference procedure of the ML model, a number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.

In some example embodiments, the first report is an aperiodic reporting, and the at least second report is a semi-persistent reporting or periodic reporting.

In some example embodiments, the terminal device may expect the first report is configured to be a semi-persistent or periodic reporting, not expect the first report is configured to be an aperiodic reporting if the at least one first resource is aperiodic, or not expect the at least one first resource is configured to be aperiodic.

FIG. 19 illustrates a flowchart of a communication method 1900 implemented at a network device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 1900 will be described from the perspective of the network device 120 in FIG. 1.

At block 1910, the network device transmits, to a terminal device, control information of a report which is based on at least one first resource.

At block 1920, the network device receives the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In some example embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, configuration information indicating at least one of the following: association between the at least one first resource and the at least one second resource, at least one trigger state associated with the at least one first resource and the at least one second resource, at least one report configuration associated with the at least one first resource and the at least one second resource, or model-related information of the ML model.

In some example embodiments, the network device may transmit, to the terminal device, one of the following: a trigger state used for triggering a measurement on the at least one second resource, or at least one trigger state each of which used for triggering a measurement on one of the at least one second resource.

In some example embodiments, the network device may transmit, to the terminal device, an indication indicating the terminal device to store the at least one second measurement result measured on the at least one second resource.

In some example embodiments, prior to transmitting the control information, receive an indication to the network device, the indication indicating a number of measured second measurement results is equal to or larger than a threshold number.

In some example embodiments, if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the network device may determine at least one third resource, wherein the number of at least one third resource is determine based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a number of the at least one second resource, or a capability of the terminal device.

In some example embodiments, if the at least one first resource is aperiodic resource, the at least one third resource is at least one repeated resource of the at least one first resource, if the at least one first resource is a periodic resource or semi-persistent resource, the at least one third resource and the at least one first resource belong to a same resource set and the at least one third resource is at least one transmission occasion of the periodic resource or semi-persistent resource.

In some example embodiments, the network device may receive, from the terminal device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to store measurement results for transmitting a report comprising at least one prediction result, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource, a time length of a measurement time window supported by the terminal device for an inference procedure of the ML model.

In some example embodiments, the control information further may indicate model-related information of the ML model.

In some example embodiments, the network device may configure at least one first resource to be semi-persistent or periodic, or not configure the at least one first resource to be aperiodic.

FIG. 20 illustrates a flowchart of a communication method 2000 implemented at a network device in accordance with some embodiments of the present disclosure. For the purpose of discussion, the method 2000 will be described from the perspective of the network device 120 in FIG. 1.

At block 2010, the network device may transmit, to a terminal device, control information indicating based on one first resource.

At block 2020, the network device may receive, from the terminal device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In some example embodiments, the network device may transmit, to the terminal device, configuration information indicating at least one of the following: association between the first report and the at least one second report, at least one trigger state associated with the first report and the at least one second report, a number of the at least one first measurement result, a number of the at least one second measurement result, or model-related information of the ML model.

In some example embodiments, the network device may transmit, to the terminal device, the following: a second trigger state used for triggering a measurement on the at least one second resource, and a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.

In some example embodiments, measurements on the at least one first resource and the at least one second resource is triggered by a plurality of trigger states comprising one of the following: a plurality of different linked trigger states, or a first trigger state and at least one associated trigger state of the trigger state.

In some example embodiments, the network device may receive, from the terminal device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report, whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an inference procedure of the ML model, a number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.

In some example embodiments, the network device may configure the first report to be a semi-persistent or periodic reporting, not configure the first report to be an aperiodic reporting, or not configure the at least one first resource to be aperiodic.

Example devices and apparatuses

FIG. 21 is a simplified block diagram of a device 2100 that is suitable for implementing embodiments of the present disclosure. The device 2100 can be considered as a further example implementation of any of the devices as shown in FIG. 1. Accordingly, the device 2100 can be implemented at or as at least a part of the terminal device 110 or the network device 120.

As shown, the device 2100 includes a processor 2110, a memory 2120 coupled to the processor 2110, a suitable transceiver 2140 coupled to the processor 2110, and a communication interface coupled to the transceiver 2140. The memory 2120 stores at least a part of a program 2130. The transceiver 2140 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 2140 may include at least one of a transmitter 2142 and a receiver 2144. The transmitter 2142 and the receiver 2144 may be functional modules or physical entities. The transceiver 2140 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.

The program 2130 is assumed to include program instructions that, when executed by the associated processor 2110, enable the device 2100 to operate in accordance with the embodiments of the present disclosure, as discussed herein with reference to FIGS. 1 to 21. The embodiments herein may be implemented by computer software executable by the processor 2110 of the device 2100, or by hardware, or by a combination of software and hardware. The processor 2110 may be configured to implement various embodiments of the present disclosure. Furthermore, a combination of the processor 2110 and memory 2120 may form processing means 2150 adapted to implement various embodiments of the present disclosure.

The memory 2120 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 2120 is shown in the device 2100, there may be several physically distinct memory modules in the device 2100. The processor 2110 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 2100 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.

According to embodiments of the present disclosure, a terminal device comprising a circuitry is provided. The circuitry is configured to: receive, from a network device, control information of a report which is based on at least one first resource; perform at least one measurement on the at least one first resource; and transmit the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the terminal device as discussed above.

According to embodiments of the present disclosure, a terminal device comprising a circuitry is provided. The circuitry is configured to: receive, from a network device, control information of a report which is based on at least one first resource; transmit, to the network device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the terminal device as discussed above.

According to embodiments of the present disclosure, a network device comprising a circuitry is provided. The circuitry is configured to: transmit, to a terminal device, control information of a report which is based on at least one first resource; and receive the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the network device as discussed above.

According to embodiments of the present disclosure, a network device comprising a circuitry is provided. The circuitry is configured to: transmit, to a terminal device, control information indicating based on one first resource; receive, from the terminal device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report. According to embodiments of the present disclosure, the circuitry may be configured to perform any method implemented by the network device as discussed above.

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.

According to embodiments of the present disclosure, a terminal apparatus is provided. The terminal apparatus comprises means for receiving, from a network device, control information of a report which is based on at least one first resource; means for performing at least one measurement on the at least one first resource; and means for transmitting the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: means for at least one first measurement result measured on the at least one first resource, and means for at least one second measurement result measured on at least one second resource associated with the first resource. In some embodiments, the first apparatus may comprise means for performing the respective operations of the method 1700. In some example embodiments, the first apparatus may further comprise means for performing other operations in some example embodiments of the method 1700. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

According to embodiments of the present disclosure, a terminal apparatus is provided. The terminal apparatus comprises means for receiving, from a network device, control information of a report which is based on at least one first resource; means for transmitting, to the network device, the following: means for at least one first measurement result measured on the at least one first resource, and means for at least one second measurement result measured on at least one second resource, means for wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, means for or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report. In some embodiments, the second apparatus may comprise means for performing the respective operations of the method 1800. In some example embodiments, the second apparatus may further comprise means for performing other operations in some example embodiments of the method 1800. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

According to embodiments of the present disclosure, a network apparatus is provided. The network apparatus comprises means for transmitting, to a terminal device, control information of a report which is based on at least one first resource; and means for receiving the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: means for at least one first measurement result measured on the at least one first resource, and means for at least one second measurement result measured on at least one second resource associated with the first resource. In some embodiments, the third apparatus may comprise means for performing the respective operations of the method 1900. In some example embodiments, the third apparatus may further comprise means for performing other operations in some example embodiments of the method 1900. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

According to embodiments of the present disclosure, a network apparatus is provided. The network apparatus comprises means for transmitting, to a terminal device, control information indicating based on one first resource; means for receiving, from the terminal device, the following: means for at least one first measurement result measured on the at least one first resource, and means for at least one second measurement result measured on at least one second resource, means for wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, means for or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report. In some embodiments, the fourth apparatus may comprise means for performing the respective operations of the method 2000. In some example embodiments, the fourth apparatus may further comprise means for performing other operations in some example embodiments of the method 2000. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

In summary, embodiments of the present disclosure provide the following aspects.

In an aspect, it is proposed a terminal device comprising: a processor configured to cause the terminal device to: receive, from a network device, control information of a report which is based on at least one first resource; perform at least one measurement on the at least one first resource; and transmit the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In some embodiments, the processor is further configured to cause the terminal device to: receive, from the network device, configuration information indicating at least one of the following: association between the at least one first resource and the at least one second resource, at least one trigger state associated with the at least one first resource and the at least one second resource, at least one report configuration associated with the at least one first resource and the at least one second resource, or model-related information of the ML model.

In some embodiments, the at least one first resource and the at least one second resource associate with a same trigger state, an identity of the at least one second resource is linked to an identity of the at least one first resource, or a first time offset between the least one first resource and the at least one second resource is smaller than or equal to a threshold offset.

In some embodiments, the processor is further configured to cause the terminal device to: receive, from the network device, one of the following: a trigger state used for triggering a measurement on the at least one second resource, at least one trigger state each of which used for triggering a measurement on one of the at least one second resource.

In some embodiments, the processor is further configured to cause the terminal device to: store the at least one second measurement result measured on the at least one second resource.

In some embodiments, the processor is further configured to cause the terminal device to: prior to receiving the control information, transmit an indication to the network device, the indication indicating a number of measured or stored second measurement results is equal to or larger than a threshold number.

In some embodiments, a number of second resources is determined based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a measurement window before a time point of a reception of the control information, or a capability of the terminal device.

In some embodiments, the at least one first resource and the at least one second resource associate with at least one of the following: at least one same measurement parameter, at least one same transmitting parameter, or at least one same receiving parameter.

In some embodiments, if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the processor is further configured to cause the terminal device to: determine at least one third resource, wherein the number of at least one third resource is determine based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a number of the at least one second resource, or a capability of the terminal device.

In some embodiments, if the at least one first resource is aperiodic resource, the at least one third resource is at least one repetition of the at least one first resource, if the at least one first resource is a periodic resource or semi-persistent resource, the at least one third resource and the at least one first resource belong to a same resource set and the at least one third resource is at least one transmission occasion of the periodic resource or semi-persistent resource.

In some embodiments, the control information further indicates a second time offset used for determining a resource for transmitting the report, wherein the second time offset starts timing from one of the following: a time point of a reception of the control information, a time point of the first transmission occasion of the at least one first resource, a time point of the completion of prediction, or a time point of the last transmission occasion of the at least one first resource.

In some embodiments, the processor is further configured to cause the terminal device to: transmit, to the network device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to store measurement results for transmitting a report comprising at least one prediction result, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource, a time length of a measurement time window supported by the terminal device, wherein measurement results obtained within the measurement time window are stored and used for an inference procedure of the ML model.

In some embodiments, the report is aperiodic, the at least one first resource is an aperiodic resource and the at least second resource is a semi-persistent resource or periodic resource.

In some embodiments, if the at least one first resource is periodic resource or semi-persistent resource, the at least one second resource is at least one historical transmission occasion of the periodic resource or semi-persistent resource.

In some embodiments, the at least one second measurement result is stored in a variable or log of the terminal device.

In some embodiments, the variable or log is maintained by the terminal device in a first in first out manner.

In some embodiments, the control information is used for activating the ML model or activating a model inference of the ML model.

In some embodiments, the control information further may indicate model-related information of the ML model.

In some embodiments, the processor is further configured to cause the terminal device to: expect the at least one first resource indicated by the control information is semi-persistent or periodic, or not expect the at least one first resource indicated by the control information is aperiodic.

In an aspect, it is proposed a terminal device comprising: a processor configured to cause the terminal device to: receive, from a network device, control information of a report which is based on at least one first resource; transmit, to the network device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In some embodiments, the processor is further configured to cause the terminal device to: receive, from the network device, configuration information indicating at least one of the following: association between the first report and the at least one second report, at least one trigger state associated with the first report and the at least one second report, a number of the at least one first measurement result, a number of the at least one second measurement result, or model-related information of the ML model.

In some embodiments, a number of first measurement results configured to be comprised in the first report is a first number, and a number of the at least one second measurement result is associated with at least one of the following: a number of measurement results required for determining the at least one prediction result, the first number, or a capability of the terminal device.

In some embodiments, the processor is further configured to cause the terminal device to: prior to receiving the control information, perform a measurement on the at least one second resource and store the at least one second measurement result measured on the at least one second resource; and after receiving the control information, perform a measurement on the at least one first resource to obtain the at least one first measurement result; and transmit the third report to the network device, the third report comprising the at least one first measurement result and the at least one second measurement result.

In some embodiments, the processor is further configured to cause the terminal device to: receive, from the network device, the following: a second trigger state used for triggering a measurement on the at least one second resource, and a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.

In some embodiments, measurements on the at least one first resource and the at least one second resource is triggered by a plurality of trigger states comprising one of the following: a plurality of different linked trigger states, or a first trigger state and at least one associated trigger state of the first trigger state.

In some embodiments, the control information indicates the first trigger state, and each of the at least one associated trigger state is a repetition of the first trigger state.

In some embodiments, the at least one first measurement result and the at least one second measurement result associate with at least one of the following: at least one same measurement parameter, at least one same transmitting parameter, or at least one same receiving parameter.

In some embodiments, the processor is further configured to cause the terminal device to: transmit, to the network device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report, whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an inference procedure of the ML model, a number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.

In some embodiments, the first report is an aperiodic reporting, and the at least second report is a semi-persistent reporting or periodic reporting.

In some embodiments, the processor is further configured to cause the terminal device to: expect the first report is configured to be a semi-persistent or periodic reporting, not expect the first report is configured to be an aperiodic reporting if the at least one first resource is aperiodic, or not expect the at least one first resource is configured to be aperiodic.

In an aspect, it is proposed a network device comprising: a processor configured to cause the network device to: transmit, to a terminal device, control information of a report which is based on at least one first resource; and receive the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource associated with the first resource.

In some embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, configuration information indicating at least one of the following: association between the at least one first resource and the at least one second resource, at least one trigger state associated with the at least one first resource and the at least one second resource, at least one report configuration associated with the at least one first resource and the at least one second resource, or model-related information of the ML model.

In some embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, one of the following: a trigger state used for triggering a measurement on the at least one second resource, or at least one trigger state each of which used for triggering a measurement on one of the at least one second resource.

In some embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, an indication indicating the terminal device to store the at least one second measurement result measured on the at least one second resource.

In some embodiments, the processor is further configured to cause the network device to: prior to transmitting the control information, receive an indication to the network device, the indication indicating a number of measured second measurement results is equal to or larger than a threshold number.

In some embodiments, if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the processor is further configured to cause the network device to: determine at least one third resource, wherein the number of at least one third resource is determine based on at least one of the following: a number of measurement results required for determining the at least one prediction result, a number of the at least one first resource, a number of the at least one second resource, or a capability of the terminal device.

In some embodiments, if the at least one first resource is aperiodic resource, the at least one third resource is at least one repeated resource of the at least one first resource, if the at least one first resource is a periodic resource or semi-persistent resource, the at least one third resource and the at least one first resource belong to a same resource set and the at least one third resource is at least one transmission occasion of the periodic resource or semi-persistent resource.

In some embodiments, the processor is further configured to cause the network device to: receive, from the terminal device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to store measurement results for transmitting a report comprising at least one prediction result, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model, a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource, a time length of a measurement time window supported by the terminal device for an inference procedure of the ML model.

In some embodiments, the processor is further configured to cause the network device to: configure at least one first resource to be semi-persistent or periodic, or not configure the at least one first resource to be aperiodic.

In an aspect, it is proposed a network device comprising: a processor configured to cause the network device to: transmit, to a terminal device, control information indicating based on one first resource; receive, from the terminal device, the following: at least one first measurement result measured on the at least one first resource, and at least one second measurement result measured on at least one second resource, wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report, or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.

In some embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, configuration information indicating at least one of the following: association between the first report and the at least one second report, at least one trigger state associated with the first report and the at least one second report, a number of the at least one first measurement result, a number of the at least one second measurement result, or model-related information of the ML model.

In some embodiments, the processor is further configured to cause the network device to: transmit, to the terminal device, the following: a second trigger state used for triggering a measurement on the at least one second resource, and a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.

In some embodiments, measurements on the at least one first resource and the at least one second resource is triggered by a plurality of trigger states comprising one of the following: a plurality of different linked trigger states, or a first trigger state and at least one associated trigger state of the trigger state.

In some embodiments, the processor is further configured to cause the network device to: receive, from the terminal device, capability-related information comprising at least one of the following: whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report, whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model, whether the terminal device supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, whether the terminal device supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model, a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model, a number of repetitions of aperiodic resources needed by the terminal device for an inference procedure of the ML model, a number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with an aperiodic report supported by the terminal device, a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.

In some embodiments, the processor is further configured to cause the network device to: configure the first report to be a semi-persistent or periodic reporting, not configure the first report to be an aperiodic reporting, or not configure the at least one first resource to be aperiodic.

In an aspect, a terminal device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the terminal device discussed above.

In an aspect, a network device comprises: at least one processor; and at least one memory coupled to the at least one processor and storing instructions thereon, the instructions, when executed by the at least one processor, causing the device to perform the method implemented by the network device discussed above.

In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the terminal device discussed above.

In an aspect, a computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the network device discussed above.

In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the terminal device discussed above.

In an aspect, a computer program comprising instructions, the instructions, when executed on at least one processor, causing the at least one processor to perform the method implemented by the network device discussed above.

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.

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 21. 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.

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.

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.

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.

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

A terminal device comprising:

a processor configured to cause the terminal device to:

receive, from a network device, control information of a report which is based on at least one first resource;

perform at least one measurement on the at least one first resource; and

transmit the report to the network device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following:

at least one first measurement result measured on the at least one first resource, and

at least one second measurement result measured on at least one second resource associated with the first resource.
The terminal device of claim 1, wherein the processor is further configured to cause the terminal device to:

receive, from the network device, configuration information indicating at least one of the following:

association between the at least one first resource and the at least one second resource,

at least one trigger state associated with the at least one first resource and the at least one second resource,

at least one report configuration associated with the at least one first resource and the at least one second resource, or

model-related information of the ML model.
The terminal device of claim 1, wherein,

the at least one first resource and the at least one second resource associate with a same trigger state,

an identity of the at least one second resource is linked to an identity of the at least one first resource, or

a first time offset between the least one first resource and the at least one second resource is smaller than or equal to a threshold offset.
The terminal device of the claim 1, wherein the processor is further configured to cause the terminal device to:

store the at least one second measurement result measured on the at least one second resource.
The terminal device of the claim 1, wherein the processor is further configured to cause the terminal device to:

prior to receiving the control information, transmit an indication to the network device, the indication indicating a number of measured or stored second measurement results is equal to or larger than a threshold number.
The terminal device of claim 1, wherein a number of second resources is determined based on at least one of the following:

a number of measurement results required for determining the at least one prediction result,

a number of the at least one first resource,

a measurement window before a time point of a reception of the control information, or

a capability of the terminal device.
The terminal device of claim 1, wherein the at least one first resource and the at least one second resource associate with at least one of the following:

at least one same measurement parameter,

at least one same transmitting parameter, or

at least one same receiving parameter.
The terminal device of claim 1, wherein if a sum of a number of the at least one first resource and a number of the at least one second resource is smaller than a number of measurement results required for determining the at least one prediction result, the processor is further configured to cause the terminal device to:

determine at least one third resource, wherein the number of at least one third resource is determine based on at least one of the following:

a number of measurement results required for determining the at least one prediction result,

a number of the at least one first resource,

a number of the at least one second resource, or

a capability of the terminal device.
The terminal device of claim 1, wherein the processor is further configured to cause the terminal device to:

transmit, to the network device, capability-related information comprising at least one of the following:

whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model,

whether the terminal device supports to report an aperiodic report based on an aperiodic resource and an associated semi-persistent or periodic resource for an inference procedure of the ML model,

whether the terminal device supports to store measurement results for transmitting a report comprising at least one prediction result,

a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model,

a number of repetitions of aperiodic resources needed by the terminal device for an aperiodic report based on an aperiodic resource for an inference procedure of the ML model,

a maximum number of trigger states associated with an aperiodic resource for an inference procedure of the ML model,

a maximum number of trigger states associated with both an aperiodic resource and at least one associated semi-persistent or periodic resource,

a time length of a measurement time window supported by the terminal device, wherein measurement results obtained within the measurement time window are stored and used

for an inference procedure of the ML model.
The terminal device of claim 1, wherein the report is aperiodic, the at least one first resource is an aperiodic resource and the at least second resource is a semi-persistent resource or periodic resource.
A terminal device comprising:

a processor configured to cause the terminal device to:

receive, from a network device, control information of a report which is based on at least one first resource;

transmit, to the network device, the following:

at least one first measurement result measured on the at least one first resource, and

at least one second measurement result measured on at least one second resource,

wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report,

or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.
The terminal device of claim 11, wherein the processor is further configured to cause the terminal device to:

receive, from the network device, configuration information indicating at least one of the following:

association between the first report and the at least one second report,

at least one trigger state associated with the first report and the at least one second report,

a number of the at least one first measurement result,

a number of the at least one second measurement result, or

model-related information of the ML model.
The terminal device of claim 11, wherein the processor is further configured to cause the terminal device to:

prior to receiving the control information, perform a measurement on the at least one second resource and store the at least one second measurement result measured on the at least one second resource; and

after receiving the control information, perform a measurement on the at least one first resource to obtain the at least one first measurement result; and

transmit the third report to the network device, the third report comprising the at least one first measurement result and the at least one second measurement result.
The terminal device of claim 11, wherein the processor is further configured to cause the terminal device to:

receive, from the network device, the following:

a second trigger state used for triggering a measurement on the at least one second resource, and

a first trigger state used for triggering a measurement on the at least one first resource and reporting the at least one first measurement result and the at least one second measurement result.
The terminal device of claim 11, wherein the processor is further configured to cause the terminal device to:

store the at least one second measurement result measured on the at least one second resource.
The terminal device of claim 11, wherein the at least one first measurement result and the at least one second measurement result associate with at least one of the following:

at least one same measurement parameter,

at least one same transmitting parameter, or

at least one same receiving parameter.
The terminal device of claim 11, wherein the processor is further configured to cause the terminal device to:

transmit, to the network device, capability-related information comprising at least one of the following:

whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model,

whether the terminal device supports to report an aperiodic report for an inference procedure of the ML model if the aperiodic report is configured with at least one associated semi-persistent or periodic report,

whether the terminal device supports to report an aperiodic report based on an aperiodic resource for an inference procedure of the ML model,

whether the terminal device supports to report an aperiodic report based on an aperiodic resource and at least one associated semi-persistent or periodic resource for an inference procedure of the ML model,

whether the terminal device supports to report measurement results on at least one associated semi-persistent or periodic resource for an inference procedure of the ML model,

a maximum number of measurement results stored by the terminal device for an inference procedure of the ML model,

a number of repetitions of aperiodic resources needed by the terminal device for an inference procedure of the ML model,

a number of trigger states associated with an aperiodic report supported by the terminal device,

a maximum number of trigger states associated with an aperiodic report supported by the terminal device,

a maximum number of trigger states associated with both an aperiodic report and at least one associated semi-persistent or periodic report.
The terminal device of claim 11, wherein the first report is an aperiodic reporting, and the at least second report is a semi-persistent reporting or periodic reporting.
A network device comprising:

a processor configured to cause the network device to:

transmit, to a terminal device, control information of a report which is based on at least one first resource; and

receive the report from the terminal device, the report comprising at least one prediction result which is obtained by a machine-learning (ML) model deployed at the terminal device and based on the following:

at least one first measurement result measured on the at least one first resource, and

at least one second measurement result measured on at least one second resource associated with the first resource.
A network device comprising:

a processor configured to cause the network device to:

transmit, to a terminal device, control information of a report which is based on at least one first resource;

receive, from the terminal device, the following:

at least one first measurement result measured on the at least one first resource, and

at least one second measurement result measured on at least one second resource,

wherein the at least one first measurement result is comprised in a first report, the at least one second measurement result is comprised in one or more second report associated with the at least one first report,

or wherein, the at least one first measurement result and the at least one second measurement result are comprised in one third report.