Method and apparatus for handling artificial intelligence / machine learning-based beam management in wireless networks

Abstract

A method performed by a User Equipment (UE) for Artificial Intelligence (AI) / Machine Learning (ML)-based beam management is provided. The method receives, from a base station (BS), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources. The method receives, from the BS, a second RRC configuration indicating time information for a Channel State Information (CSI) report. The method determines at least one future time instance based on the time information for the CSI report. The method generates at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources. The method transmits, to the BS, the CSI report including the at least one predicted result.

Description

METHOD AND APPARATUS FOR HANDLING ARTIFICIAL INTELLIGENCE / MACHINE LEARNING-BASED BEAM MANAGEMENT IN WIRELESS NETWORKSThe present disclosure is related to wireless communication and, more specifically, to a User Equipment (UE), Base Station (BS), and method for handling Artificial Intelligence (AI) / Machine Learning (ML)-based Beam Management (BM) in the wireless communication networks.Various efforts have been made to improve different aspects of wireless communication for the cellular wireless communication systems, such as the 5thGeneration (5G) New Radio (NR), by improving data rate, latency, reliability, and mobility. The 5G NR system is designed to provide flexibility and configurability to optimize network services and types, accommodating various use cases, such as enhanced Mobile Broadband (eMBB), massive Machine-Type Communication (mMTC), and Ultra-Reliable and Low-Latency Communication (URLLC). As the demand for radio access continues to grow, however, there exists a need for further improvements in the next-generation wireless communication systems, such as improvements in a beam management procedure.The present disclosure is related to a UE, a BS, and a method for handling AI / ML-based BM in the wireless communication networks.In a first aspect of the present disclosure, a UE for performing AI / ML-based beam management is provided. The UE includes at least one processor and at least one non-transitory computer-readable medium that is coupled to the at least one processor and that stores one or more computer-executable instructions. The computer-executable instructions, when executed by the at least one processor, cause the UE to: receive, from a base station (BS), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; receive, from the BS, a second RRC configuration indicating time information for a Channel State Information (CSI) report; determine at least one future time instance based on the time information for the CSI report; generate at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources; and transmit, to the BS, the CSI report including the at least one predicted result.In some implementations of the first aspect, the UE receives, from the BS, a parameter indicating the number of the at least one predicted result to be included in the CSI report.In some implementations of the first aspect, the time information for the CSI report includes a time offset, and determining the at least one future time instance includes separating any two consecutive future time instances by the time offset.In some implementations of the first aspect, the second RRC configuration includes a CSI report configuration.In some implementations of the first aspect, the time information for the CSI report indicates the earliest future time instance among the at least one future time instance.In some implementations of the first aspect, the UE activates an AI / ML function before receiving the second RRC configuration.In some implementations of the first aspect, each of the at least one predicted result corresponds to a Reference Signal Received Power (RSRP) value.In some implementations of the first aspect, the second RRC configuration includes an index for association.In a second aspect of the present application, a BS for facilitating AIML-based beam management is provided. The BS includes at least one processor and at least one non-transitory computer-readable medium that is coupled to the at least one processor and that stores one or more computer-executable instructions. The computer-executable instructions, when executed by the at least one processor, cause the BS to: transmit, to a User Equipment (UE), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; transmit, to the UE, a second RRC configuration indicating time information for a Channel State Information (CSI) report; and receive, from the UE, the CSI report including the at least one predicted result. The UE determines at least one future time instance based on the time information for the CSI report. The UE generates at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources.In some implementations of the second aspect, the BS transmits, to the UE, a parameter indicating the number of the at least one predicted result to be included in the CSI report.In some implementations of the second aspect, the time information for the CSI report includes a time offset, and determining the at least one future time instance includes separating any two consecutive future time instances by the time offset.In some implementations of the second aspect, the second RRC configuration includes a CSI report configuration.In some implementations of the second aspect, the time information for the CSI report indicates the earliest future time instance among the at least one future time instance.In some implementations of the second aspect, each of the at least one predicted result corresponds to a Reference Signal Received Power (RSRP) value, and the second RRC configuration includes an index for association.In a third aspect of the present disclosure, a method performed by a UE for AI / ML-based beam management is provided. The method includes receiving, from a base station (BS), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; receiving, from the BS, a second RRC configuration indicating time information for a Channel State Information (CSI) report; determining at least one future time instance based on the time information for the CSI report; generating at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources; and transmitting, to the BS, the CSI report including the at least one predicted result.Aspects of the present disclosure are best understood from the following detailed disclosure when read with the accompanying drawings. Various features are not drawn to scale. Dimensions of various features may be arbitrarily increased or reduced for clarity of discussion.FIG. 1 is a block diagram illustrating functional framework for AI / ML for NR air interface, according to an example implementation of the present disclosure.FIG. 2 is a diagram illustrating a timing pattern for predicted results in a report, according to an example implementation of the present disclosure.FIG. 3 is a diagram illustrating association between predicted results and a report in time domain, according to an example implementation of the present disclosure.FIG. 4 is a flowchart illustrating a method / process performed by a UE for AI / ML-based BM, according to an example implementation of the present disclosure.FIG. 5 is a flowchart illustrating a method / process performed by a BS for facilitating AI / ML-based BM, according to an example implementation of the present disclosure.FIG. 6 is a block diagram illustrating a node for wireless communication, according to an example implementation of the present disclosure.Some of the abbreviations used in the present disclosure include:Abbreviation　　　　　　　　Full name3GPP　　　　　　　　3rd Generation Partnership Project5G　　　　　　　　5th generationACK　　　　　　　　AcknowledgmentAI　　　　　　　　Artificial IntelligenceAL　　　　　　　　Aggregation levelARFCN　　　　　　　　Absolute Radio Frequency Channel NumberBFD　　　　　　　　Beam Failure DetectionBM　　　　　　　　Beam ManagementBWP　　　　　　　　Band Width PartCA　　　　　　　　Carrier AggregationCORESET 　　　　　　　　Control resource setCC　　　　　　　　Component CarrierCCE　　　　　　　　Control Chanel ElementCE　　　　　　　　Control ElementCG　　　　　　　　Configured GrantCP　　　　　　　　Cyclic PrefixCRC　　　　　　　　Cyclic Redundancy CheckCRI　　　　　　　　CSI-RS Resource IndicatorC-RNTI　　　　　　　　Cell Radio Network Temporary IdentifierCS-RNTI　　　　　　　　Configured Scheduling Radio Network Temporary IdentifierCSS　　　　　　　　Common Search SpaceCSI　　　　　　　　Channel State InformationDC　　　　　　　　Dual ConnectivityDCI　　　　　　　　Downlink Control InformationDL　　　　　　　　DownlinkDMRS　　　　　　　　Demodulation Reference SignalFR　　　　　　　　Frequency RangeGC-PDCCH　　　　　　　　Group Common Physical Downlink Control ChannelHARQ　　　　　　　　Hybrid Automatic Repeat RequestID　　　　　　　　IdentifierIE　　　　　　　　Information ElementIIoT　　　　　　　　Industrial Internet of ThingsLSB　　　　　　　　Least Significant BitLTE　　　　　　　　Long Term EvolutionL1　　　　　　　　Layer 1L1-RSRP　　　　　　　　Layer 1 reference signal received powerLCM　　　　　　　　Life Cycle ManagementLMF　　　　　　　　Location Management FunctionMAC　　　　　　　　Medium Access ControlMCG　　　　　　　　Master Cell GroupMCS-C-RNTI　　　　　　　　Modulation Coding Scheme Cell Radio Network Temporary IdentifiermTRP　　　　　　　　Multiple Transmission Reception PointMIMO　　　　　　　　Multiple-input Multiple-outputMSB　　　　　　　　Most Significant BitML　　　　　　　　Machine LearningNACK　　　　　　　　Negative AcknowledgmentNDI　　　　　　　　New Data IndicatorNR　　　　　　　　New RAT / RadioNW　　　　　　　　NetworkNUL　　　　　　　　Normal ULPCI　　　　　　　　Physical Cell IDPCell　　　　　　　　Primary CellPSCell　　　　　　　　Primary Secondary CellPBCH　　　　　　　　Physical Broadcast ChannelPDCCH　　　　　　　　Physical Downlink Control ChannelPDSCH　　　　　　　　Physical Downlink Shared ChannelPDU　　　　　　　　Protocol Data UnitPHY　　　　　　　　PhysicalPRACH　　　　　　　　Physical Random Access ChannelPTAG　　　　　　　　Primary Timing Advance GroupPTRS　　　　　　　　Phase-Tracking Reference SignalPUCCH　　　　　　　　Physical Uplink Control ChannelPUSCH　　　　　　　　Physical Uplink Shared ChannelRA　　　　　　　　Random AccessRAN　　　　　　　　Radio Access NetworkRAR　　　　　　　　Random Access ResponseRel　　　　　　　　ReleaseRMSI　　　　　　　　Remaining Minimum System InformationRNTI　　　　　　　　Radio Network Temporary IdentifierRRC　　　　　　　　Radio Resource ControlRRM　　　　　　　　Radio Resource MeasurementRS　　　　　　　　Reference SignalRSRP　　　　　　　　Reference Signal Received PowerRSRQ　　　　　　　　Reference Signal Received QualityRV　　　　　　　　Redundancy VersionSCell　　　　　　　　Secondary CellSCG　　　　　　　　Secondary Cell GroupSCS　　　　　　　　Subcarrier SpacingSDM　　　　　　　　Spatial Division MultiplexingSFN　　　　　　　　System Frame NumberSINR　　　　　　　　Signal to Interference plus Noise RatioSpCell　　　　　　　　Special CellSR　　　　　　　　Scheduling RequestSRS　　　　　　　　Sounding Reference SignalSRI　　　　　　　　SRS Resource IndicatorSS　　　　　　　　Synchronization SignalSSB　　　　　　　　Synchronization Signal BlockSSBRI　　　　　　　　SS / PBCH Block Resource indicatorSTAG　　　　　　　　Secondary Timing Advance GroupSTxMP　　　　　　　　Simultaneous Transmission on Multiple PanelsSUL　　　　　　　　Supplementary ULTA　　　　　　　　Timing AdvanceTAG　　　　　　　　Timing Advance GroupTB　　　　　　　　Transport BlockTBS　　　　　　　　Transport Block SizeTCI　　　　　　　　Transmission Configuration IndicationTPMI　　　　　　　　Transmission Precoding Matrix IndicatorTR　　　　　　　　Technical ReportTRP　　　　　　　　Transmission Reception PointTS　　　　　　　　Technical SpecificationTx　　　　　　　　TransmissionQCL　　　　　　　　Quasi-CoLocationUE　　　　　　　　User EquipmentUL　　　　　　　　UplinkURLLC　　　　　　　　Ultra Reliable Low Latency CommunicationUSS　　　　　　　　UE-Specific Search SpaceWG　　　　　　　　Working GroupWI　　　　　　　　Working ItemThe following contains specific information related to implementations of the present disclosure. The drawings and their accompanying detailed disclosure are merely directed to implementations. However, the present disclosure is not limited to these implementations. Other variations and implementations of the present disclosure will be obvious to those skilled in the art.Unless noted otherwise, like or corresponding elements among the drawings may be indicated by like or corresponding reference numerals. Moreover, the drawings and illustrations in the present disclosure are generally not to scale and are not intended to correspond to actual relative dimensions.For the purposes of consistency and ease of understanding, like features may be identified (although, in some examples, not illustrated) by the same numerals in the drawings. However, the features in different implementations may be different in other respects and may not be narrowly confined to what is illustrated in the drawings.References to “one implementation,” “an implementation,” “example implementation,” “various implementations,” “some implementations,” “implementations of the present application,” etc., may indicate that the implementation(s) of the present application so described may include a particular feature, structure, or characteristic, but not every possible implementation of the present application necessarily includes the particular feature, structure, or characteristic. Further, repeated use of the phrase “In some implementations,” or “in an example implementation,” “an implementation,” do not necessarily refer to the same implementation, although they may. Moreover, any use of phrases like “implementations” in connection with “the present application” are never meant to characterize that all implementations of the present application must include the particular feature, structure, or characteristic, and should instead be understood to mean “at least some implementations of the present application” includes the stated particular feature, structure, or characteristic. The term “coupled” is defined as connected, whether directly or indirectly through intervening components, and is not necessarily limited to physical connections. The term “comprising,” when utilized, means “including, but not necessarily limited to”; it specifically indicates open-ended inclusion or membership in the so-described combination, group, series, and the equivalent.The expression “at least one of A, B and C” or “at least one of the following: A, B and C” means “only A, or only B, or only C, or any combination of A, B and C.” The terms “system” and “network” may be used interchangeably. The term “and / or” is only an association relationship for describing associated objects and represents that three relationships may exist such that A and / or B may indicate that A exists alone, A and B exist at the same time, or B exists alone. The character “ / ” generally represents that the associated objects are in an “or” relationship.For the purposes of explanation and non-limitation, specific details, such as functional entities, techniques, protocols, and standards, are set forth for providing an understanding of the disclosed technology. In other examples, detailed disclosure of well-known methods, technologies, systems, and architectures are omitted so as not to obscure the present disclosure with unnecessary details.Persons skilled in the art will immediately recognize that any network function(s) or algorithm(s) disclosed may be implemented by hardware, software, or a combination of software and hardware. Disclosed functions may correspond to modules which may be software, hardware, firmware, or any combination thereof.A software implementation may include computer executable instructions stored on a computer-readable medium, such as memory or other type of storage devices. One or more microprocessors or general-purpose computers with communication processing capability may be programmed with corresponding executable instructions and perform the disclosed network function(s) or algorithm(s).The microprocessors or general-purpose computers may include Application-Specific Integrated Circuits (ASICs), programmable logic arrays, and / or one or more Digital Signal Processor (DSPs). Although some of the disclosed implementations are oriented to software installed and executing on computer hardware, alternative implementations implemented as firmware, as hardware, or as a combination of hardware and software are well within the scope of the present disclosure. The computer-readable medium includes but is not limited to Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, Compact Disc Read-Only Memory (CD-ROM), magnetic cassettes, magnetic tape, magnetic disk storage, or any other equivalent medium capable of storing computer-readable instructions.A radio communication network architecture such as a Long-Term Evolution (LTE) system, an LTE-Advanced (LTE-A) system, an LTE-Advanced Pro system, or a 5G NR Radio Access Network (RAN) typically includes at least one base station (BS), at least one UE, and one or more optional network elements that provide connection within a network. The UE communicates with the network such as a Core Network (CN), an Evolved Packet Core (EPC) network, an Evolved Universal Terrestrial RAN (E-UTRAN), a 5G Core (5GC), or an internet via a RAN established by one or more BSs.A UE may include, but is not limited to, a mobile station, a mobile terminal or device, or a user communication radio terminal. The UE may be a portable radio equipment that includes, but is not limited to, a mobile phone, a tablet, a wearable device, a sensor, a vehicle, or a Personal Digital Assistant (PDA) with wireless communication capability. The UE is configured to receive and transmit signals over an air interface to one or more cells in a RAN.The BS may be configured to provide communication services according to at least a Radio Access Technology (RAT) such as Worldwide Interoperability for Microwave Access (WiMAX), Global System for Mobile communications (GSM) that is often referred to as 2G, GSM Enhanced data rates for GSM Evolution (EDGE) RAN (GERAN), General Packet Radio Service (GPRS), Universal Mobile Telecommunication System (UMTS) that is often referred to as 3G based on basic wideband-code division multiple access (W-CDMA), high-speed packet access (HSPA), LTE, LTE-A, evolved LTE (eLTE) that is LTE connected to 5GC, NR (often referred to as 5G), and / or LTE-A Pro. However, the scope of the present disclosure is not limited to these protocols.The BS may include, but is not limited to, a node B (NB) in the UMTS, an evolved node B (eNB) in LTE or LTE-A, a radio network controller (RNC) in UMTS, a BS controller (BSC) in the GSM / GERAN, an ng-eNB in an Evolved Universal Terrestrial Radio Access (E-UTRA) BS in connection with 5GC, a next generation Node B (gNB) in the 5G-RAN, or any other apparatus capable of controlling radio communication and managing radio resources within a cell. The BS may serve one or more UEs via a radio interface. Although the gNB is used as an example in some implementations within the present disclosure, it should be noted that the disclosed implementations may also be applied to other types of base stations.The BS may be operable to provide radio coverage to a specific geographical area using multiple cells forming the RAN. The BS may support the operations of the cells. Each cell may be operable to provide services to at least one UE within its radio coverage.Each cell (may often referred to as a serving cell) may provide services to one or more UEs within the cell’s radio coverage, such that each cell schedules the DL (and optionally UL resources) to at least one UE within its radio coverage for DL (and optionally UL packet transmissions from the UE). The BS may communicate with one or more UEs in the radio communication system via the cells.A cell may allocate sidelink (SL) resources for supporting the Proximity Services (ProSe) or Vehicle to Everything (V2X) services. Each cell may have overlapped coverage areas with other cells.In Multi-RAT Dual Connectivity (MR-DC) cases, the primary cell of a Master Cell Group (MCG) or a Secondary Cell Group (SCG) may be referred to as a Special Cell (SpCell). A Primary Cell (PCell) may include the SpCell of an MCG. A Primary SCG Cell (PSCell) may include the SpCell of an SCG. MCG may include a group of serving cells associated with the Master Node (MN), including the SpCell and optionally one or more Secondary Cells (SCells). An SCG may include a group of serving cells associated with the Secondary Node (SN), including the SpCell and optionally one or more SCells.As discussed above, the frame structure for NR may support flexible configurations for accommodating various next generation (e.g., 5G) communication requirements, such as Enhanced Mobile Broadband (eMBB), Massive Machine Type Communication (mMTC), and Ultra-Reliable and Low-Latency Communication (URLLC), while fulfilling high reliability, high data rate, and low latency requirements. The Orthogonal Frequency-Division Multiplexing (OFDM) technology in the 3GPP may serve as a baseline for an NR waveform. The scalable OFDM numerology, such as adaptive sub-carrier spacing, channel bandwidth, and Cyclic Prefix (CP), may also be used.Two coding schemes may be considered for NR, specifically, Low-Density Parity-Check (LDPC) code and Polar Code. The coding scheme adaption may be configured based on channel conditions and / or service applications.At least the DL transmission data, a guard period, and UL transmission data should be included in a transmission time interval (TTI) of a single NR frame. The respective portions of the DL transmission data, the guard period, and the UL transmission data should also be configurable based on, for example, the network dynamics of NR. SL resources may also be provided in an NR frame to support ProSe services or V2X services.Any two or more than two of the following paragraphs, (sub)-bullets, points, actions, behaviors, terms, or claims described in the present disclosure may be combined logically, reasonably, and properly to form a specific method.Any sentence, paragraph, (sub)-bullet, point, action, behaviors, terms, or claims described in the present disclosure may be implemented independently and separately to form a specific method.Dependency, e.g., “based on”, “more specifically”, “preferably”, “in one embodiment”, “in some implementations”, etc., in the present disclosure is just one possible example which would not restrict the specific method.A and / or B in following paragraph may refer to either A or B, both A and B, at least one of A and B.The term ‘TRP’ in the present disclosure may be replaced by ‘beam’ or ‘panel’. The term ‘overlap’ in the present disclosure may refer to partial overlap or full overlap in time domain.In some implementations, all the designs / embodiment / implementations introduced within this disclosure are not limited to be applied for dealing with the problems discussed within this disclosure. For example, the described embodiments may be applied to solve other problems that exist in the RAN of wireless communication systems. In some implementations, all of the numbers listed within the designs / embodiment / implementations introduced within this disclosure are just examples and for illustration, for example, of how the described methods are executed.Examples of some selected terms in the present disclosure are provided as follows.DCI: DCI may include downlink control information, and there may be various DCI formats used in a PDCCH. The DCI format may be a predefined format in which the downlink control information may be packed / formed and transmitted in a PDCCH.BWP: A subset of the total cell bandwidth of a cell is referred to as a Bandwidth Part (BWP) and a Bandwidth Adaptation (BA) may be achieved by configuring the UE with BWP(s) and instructing the UE which of the configured BWPs is currently the active one. To enable a BA on the PCell, the BS (e.g., a gNB) configures the UE with UL and DL BWP(s). To enable the BA on SCells, when CA is deployed, the BS configures the UE with one or more DL BWPs. It should be noted that there may be no BWP in the UL. For the PCell, the initial BWP is the BWP used for an initial access. For the SCell(s), the initial BWP is the BWP configured for the UE to operate after an SCell activation. The UE may be configured with a first active uplink BWP by a firstActiveUplinkBWP IE. If the first active uplink BWP is configured for an SpCell, the firstActiveUplinkBWP IE field may contain the ID of the UL BWP to be activated upon performing the RRC (re-)configuration. If such a field is absent, the RRC (re-)configuration may not impose a BWP switching. If the first active uplink BWP is configured for an SCell, the firstActiveUplinkBWP IE field may contain the ID of the uplink bandwidth part to be used upon the MAC-activation of an SCell.TCI state: A TCI state may include parameters for configuring a QCL relationship between one or more DL reference signals and a target reference signal set. For example, a target reference signal set may include the DMRS ports of a PDSCH, a PDCCH, a PUCCH, or a PUSCH. The reference signals may include UL or DL reference signals. In NR Rel-15 / 16, the TCI state may be used for a DL QCL indication, whereas the spatial relation information may be used for providing the UL spatial transmission filter information for the UL signal(s) or channel(s). A TCI state may include the information similar to the spatial relation information, which may be used for UL transmission. In other words, from the UL perspective, a TCI state may provide the UL beam information that may indicate the relationship between a UL transmission and the DL or UL reference signals (e.g., the CSI-RS, the SSB, the SRS, and the PTRS).Beam: The term “beam” here may be replaced by a spatial filter. For example, when a UE reports a preferred gNB Tx beam, the UE is essentially selecting a spatial filter used by the gNB. The term “beam information” may be used to provide information about which beam / spatial filter is being used / selected.Cell: A cell in the present disclosure may refer to a PCell, a PSCell, an SpCell, an SCell, a candidate cell, a target cell, a neighbor cell, a serving cell, or a source cell.In the realm of wireless communication, beam management plays a crucial role in optimizing the performance of communication networks, particularly in the context of emerging technologies like 5G and beyond. Beam management involves the selection, tracking, and optimization of beams in a beamforming system to ensure efficient and reliable data transmission between base stations and user devices. Furthermore, AI / ML-driven beam management enhances spectral efficiency by dynamically adjusting beamforming parameters to optimize resource allocation and / or by maximizing throughput per unit of spectrum. It improves coverage by intelligently steering beams towards active users and high-demand areas, ensuring better signal propagation and reduced dead zones. Additionally, AI / ML algorithms mitigate interference, enhance signal quality and reliability, thereby improving overall quality of service (QoS). Through these mechanisms (e.g., AI / ML algorithms), AI / ML-driven beam management optimizes spectrum usage, extends coverage, and enhances QoS, leading to more efficient and reliable wireless communication systems.For temporal downlink beam prediction based on the historical measurement results, multiple beam prediction results may be included in a single beam report, so how to associate different measurement results and / or predict results to the same report should be addressed. Moreover, not only the time interval for collecting the measurement results to be the Set B beams should be indicated but also the time interval for collecting predicted beam results to be the Set A beams in one report should be indicated to avoid the ambiguity. For example, since the predicted results may be inferred based on the past measurement results during a time interval, how long the interval is and the starting point should be clarified to avoid the report including too less or too many predicted results. On the other hand, whether to introduce the measurement configuration for set A beam (for beam prediction) should be considered.Functional framework for AI / ML for NR air interfaceFIG. 1 is a block diagram 100 illustrating a functional framework for AI / ML for NR air interface, according to an example implementation of the present disclosure. As illustrated in FIG. 1, data collection 102 is a function that provides input data to the model training, management, and inference functions. Training data may include data needed as input for the AI / ML model training function. Monitoring data may include data needed as input for the management of AI / ML models or AI / ML functionalities. Inference data may include data needed as input for the AI / ML inference function.Model training 104 is a function that performs AI / ML model training, validation, and testing, which may generate model performance metrics that may be used as part of the model testing procedure. The model training function is also responsible for data preparation (e.g., data pre-processing and cleaning, formatting, and transformation) based on training data delivered by a data collection function, if required. In some implementations, there may be a model storage 106 in the framework. The model training 104 may deliver a trained / updated model to the model storage 106. In some implementations, the trained / updated model may include trained, validated, and tested AI / ML models. In some implementations, the trained / updated model may include an updated version of the model.Management 108 is a function that oversees the operation (e.g., selection / (de)activation / switching / fallback) and monitoring (e.g., performance) of AI / ML models or AI / ML functionalities. The management 108 is also responsible for making decisions to ensure the proper inference operation based on data received from the data collection function and the inference function. In some implementations, a management instruction represents essential input information to manage the inference function. Concerning information may include selection / (de)activation / switching of AI / ML models or AI / ML-based functionalities, fallback to non-AI / ML operation (e.g., not relying on inference process), and so on. In some implementations, a model transfer / delivery request is used to request model(s) to the model storage 106. In some implementations, performance feedback and retraining request represent essential input information for the model training 104 (e.g., for model (re)training or updating purposes).Inference 110 is a function that provides outputs from the process of applying AI / ML models or AI / ML functionalities, using the data that is provided by the data collection 102 (e.g., inference data) as an input. The inference 110 is also responsible for data preparation (e.g., data pre-processing and cleaning, formatting, and transformation) based on the inference data delivered by the data collection 102, if required. In some implementations, inference output serves as data for the management 108 to monitor the performance of AI / ML models or AI / ML functionalities.Model storage 106 is a function responsible for storing trained / updated models that may be used to perform the inference function. The model storage 106 may transfer / deliver an AI / ML model to the inference 110.L1 measurement and reportA UE may perform, and report measurement (e.g., L1 measurement or L3 measurement) based on the received configuration and / or indication. L1 measurements may be further classified into L1 intra-frequency measurement, or L1 inter-frequency measurement. In some implementations, L1 intra-frequency measurement and L1 inter-frequency measurement may be based on L1-RSRP through measuring SSB (e.g., SS-RSRP) or CSI-RS (e.g., CSI-RSRP). In some implementations, L1 intra-frequency measurement and L1 inter-frequency measurement may be based on L1-SINR through measuring SSB (e.g., SS-SINR) or CSI-RS (CSI-SINR). In some implementations, L1 intra-frequency measurement and L1 inter-frequency measurement may be based on L1-RSRQ through measuring SSB (e.g., SS-RSRQ) or CSI-RS (e.g., CSI-RSRQ).In some implementations, the L1 measurement report may include one or some PCIs (e.g., PCIs of the candidate cells, PCI of the source cell, PCI of the serving cell, or PCI of the target cell). In some implementations, the L1 measurement report may include one or some RS ID.In some implementations, an L1 measurement report, transmitted as UCI on PUCCH or PUSCH, may be considered as the result of measurement from the UE’s perspective. In some implementations, the types of L1 measurement reports may include a periodic report on PUCCH, a semi-persistent report on PUCCH or PUSCH, and an aperiodic report on PUSCH. In some implementations, the L1 measurement report may be transmitted on a MAC-CE.CSI measurement configurationCSI measurement configuration involves setting parameters that configure how CSI measurements are performed and reported by a UE. The CSI measurement configuration configures a list of non-zero power CSI-RS resources for channel measurement, a list of CSI-RS resources for interference management, and a list of SSB resources for CSI measurement and reporting. In some implementation, the CSI measurement configuration may be associated with a CSI report configuration. In some implementations, when the CSI resources in the CSI resource set refer to beam measurement, the corresponding CSI report may only contain CRI, SSBRI, L1-RSRP, L1-SINR for each measurement result based on the configured CSI-RS resources, and one CSI-RS resource or SSB may correspond to one L1-RSRP / L1-SINR / L1-RSRQ value.CSI report contentIn some implementations, a CSI report may include Channel Quality Indicator (CQI), Precoding Matrix Indicator (PMI), CSI-RS resource indicator (CRI), SS / PBCH block resource indicator (SSBRI), Layer Indicator (LI), Rank Indicator (RI), Capability Index, L1-RSRP, L1-SINR, and / or L1-RSRQ. In some implementations, the report may contain top K values among a set of measurement results by measuring the CSI-RS resource or SSB, where K is a positive integer.AI / ML-based Beam Management (BM)The beam management may include DL Tx beam prediction for both the UE-sided model and the NW-sided model. The beam prediction may include predicting a Set A of beams based on a Set B of beams and an AI / ML model. For example, the AI / ML model may receive the Set B of beams as input and generate the Set A of beams as output. In some implementations, the measurement results may correspond to the inference results.BM-Case1: BM-Case1 may refer to a spatial-domain DL Tx beam prediction for Set A of beams based on measurement results of Set B of beams. In some implementations, the Set B of beams may include beams received from one direction and the Set A of beams may include beams received from another direction. In some implementations, the Set B of beams may include wide beams (e.g., SSB) and the Set A of beams may include narrow beams (e.g., CSI-RS). In some implementations, the AI / ML model training and inference may be performed at the NW side. In some implementations, the AI / ML model training and inference may be performed at the UE side. In some implementations, Set A and Set B may be different. In some implementations, Set B may be a subset of Set A.BM-Case2: BM-Case2 may refer to a temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams. In some implementations, the Set B of beams may include beams that were received in the past and the Set A of beams may include beams that are expected to be received in the future. In some implementations, the AI / ML model training and inference may be performed at the NW side. In some implementations, the AI / ML model training and inference may be performed at the UE side. In some implementations, Set A and Set B may be different. In some implementations, Set B may be a subset of Set A. In some implementations, Set A and Set B may be the same.Measurement and report configurationIn some implementations, an RRC signaling may configure a first measurement configuration for Set A of beams and a second measurement configuration for Set B of beams. In some implementations, the first measurement configuration may include different content from the second measurement configuration, and the first measurement configuration may be dedicated to AI / ML function. In other words, the first measurement configuration may be configured only when the AI / ML function is activated / configured / applied. In some implementations, the first configuration and the second configuration may be configured to a single cell. In some implementations, the first measurement configuration may be configured together with the second configuration, but the first measurement configuration and its corresponding measurement behavior (e.g., dedicated to AI / ML) may be active while the corresponding AI / ML function is activated. On the contrary, the second measurement configuration and its corresponding measurement behavior may be active immediately after the UE receives the second measurement configuration.In some implementations, the first measurement configuration may be associated with the second measurement configuration. More specifically, the first measurement configuration may include a second measurement configuration index to associate the first measurement configuration with the second measurement configuration. In some implementations, the second measurement configuration may be associated with the first measurement configuration. More specifically, a second measurement configuration may include a first measurement configuration index to associate the second measurement configuration with the first measurement configuration.In some implementations, the first measurement configuration may include a set of CSI-RS resources and / or a set of SSB resources.In some implementations, the set of CSI-RS resources may be referred to as the CSI-RS resources in the second measurement configuration.For example, the UE may receive the first measurement configuration, where the first measurement configuration may include an RRC field / parameter indicating another measurement configuration (e.g., the second measurement configuration) different from the first measurement configuration. In some implementations, the set of CSI-RS resource (e.g., CSI-RS resources with the smallest-N ID) configured in the first measurement configuration may be equal to the CSI-RS resources configured in the second measurement configuration, where the number of N may be determined by the number of CSI-RS resources configured in the second measurement configuration. In some implementations, the set of CSI-RS resources in the first measurement configuration may be totally different from the set of CSI-RS resources in the second measurement configuration.In some implementations, whether the set of CSI-RS resources in the first measurement configuration is same or different from the set of CSI-RS resources in the second measurement configuration may be indicated by an RRC parameter. For example, if the RRC parameter indicates ‘enable’, the set of CSI-RS resources in the first measurement configuration may be the same as or a subset of the set of CSI-RS resources in the second measurement configuration. In some implementations, there may be no CSI-RS resource in the first measurement configuration. The set of CSI-RS resources for the first measurement configuration may be configured from the second measurement configuration. More specifically, the CSI-RS resource indices to be added in the CSI-RS resources list may be the CSI-RS resource indices configured in the second measurement configuration. On the contrary, if the RRC parameter indicates ‘disable’, the set of CSI-RS resources in the first measurement configuration may be different from the set of CSI-RS resources in the second measurement configuration.In some implementations, the set of SSB resources may be referred to as the SSB resource in the second measurement configuration.For example, the UE may receive the first measurement configuration, where the first measurement configuration may include a RRC field / parameter indicating another measurement configuration (e.g., the second measurement configuration) being different from the first measurement configuration. In some implementations, the set of SSB resources (e.g., SSB resources with the smallest-M ID) configured in the first measurement configuration may be equal to the SSB resources configured in the second measurement configuration, where the number of M may be determined by the number of SSB resources configured in the second measurement configuration.In some implementations, whether the set of SSB resources in the first measurement configuration is same or different from the set of SSB resources in the second measurement configuration may be indicated by an RRC parameter. For example, if the RRC parameter indicates ‘enable’, the set of SSB resources in the first measurement configuration may be the same as or a subset of the set of SSB resources in the second measurement configuration. In some implementations, there is no SSB resource in the first measurement configuration. The set of SSB resources for the first measurement configuration may be configured from the second measurement configuration. More specifically, the SSB resource indices to be added in the SSB resources list may be the SSB resource indices configured in the second measurement configuration. On the contrary, if the RRC parameter indicates ‘disable’, the set of SSB resources in the first measurement configuration may be different from the set of SSB resources in the second measurement configuration.In some implementations, the second measurement configuration may correspond to CSI for channel measurement configuration.In some implementations, the second measurement configuration may correspond to CSI for beam management configuration.In some implementations, the second measurement configuration may correspond to CSI for interference management configuration.In some implementations, an RRC signaling may configure a first report configuration for Set A of beams and a second report configuration for Set B of beams.In some implementations, the first report configuration may be associated with the first measurement configuration for Set A of beams when the first measurement configuration and the second measurement configuration are both configured. In some implementations, the first report configuration may include the first measurement configuration index to associate the first report configuration with the first measurement configuration. In some implementations, the first measurement configuration may include the first report configuration index to associate the first report configuration with the first measurement configuration.In some implementations, the first report configuration may be associated with the second measurement configuration for Set B of beams when the first measurement configuration and the second measurement configuration are both configured. In some implementations, the first report configuration may include the second measurement configuration index to associate the first report configuration with the second measurement configuration. In some implementations, the first report configuration may include the second measurement configuration index to associate the first report configuration with the second measurement configuration. In some implementations, the first report configuration may include the first and the second measurement configuration indexes where the corresponding measurement objects may be different. Specifically, if the corresponding measurement object is the same, the gNB may provide one of the configuration indexes.In some implementations, an RRC parameter corresponding to a flag may be used to indicate that the first report configuration is associated with the first measurement configuration or the second measurement configuration. In some implementations, when the flag is set to ‘1’, the first report configuration may be associated with the first measurement configuration. When the flag is set to ‘0’, the first report configuration may be associated with the second measurement configuration. In some implementations, the flag may be set to the opposite value. For example, the first report configuration may be associated with the first measurement configuration when the flag is set to ‘0’.In some implementations, the second report configuration may correspond to CSI report configuration.In some implementations, the first report configuration may be configured when the AI / ML function is activated / configured / applied. Otherwise, the first report configuration may not be configured.In some implementations, a list of reports may correspond to a list of report indexes in the first report configuration.In some implementations, the first report configuration may include one or more than one timing pattern to configure where the corresponding predicated result may be included in the report. For example, the timing pattern may correspond to a set of bits with values of 0 and 1, where 0 indicates that no predicated result is generated and 1 indicates that a predicated result is generated.FIG. 2 is a diagram 200 illustrating timing patterns for predicted results in a report, according to an example implementation of the disclosure. In some implementations, each timing pattern may be associated with a report index. In the example shown in FIG. 2, there are two timing patterns. The first timing pattern may be associated with report #1, and the second timing pattern may be associated with report #2. Each timing pattern may indicate one or more future time instances to be included in the respective report. For example, the first timing pattern may be a bit string having a value of

[10101] , and thus the report #1 may include predicted results (e.g., corresponding to respective beam IDs) on future time instances including slot#1, slot#3, and slot#5. The second timing pattern may be a bit string having a value of

[0011] , and thus the report #2 may include predicted results (e.g., corresponding to respective beam IDs) on future time instances including slot#9 and slot#10.In some implementations, a set of timing pattern may be included in a timing pattern set. Each timing pattern in the same timing pattern set may correspond to same periodicity and / or the offset. In some implementations, the periodicity of the timing pattern may be the same as the periodicity of the report. In some implementations, the periodicity of the timing pattern may be different from the periodicity of the report.In some implementations, the timing pattern may include a starting point in time domain. In some implementations, an offset with respect to the SFN#0 may be configured. In some implementations, two offsets may be configured, where a first offset is configured in the timing pattern set and a second offset is configured in the timing pattern. The first offset may be with respect to the SFN#0 and the second offset may be with respect to the starting point of the timing pattern set. In some implementations, the starting point of the timing pattern may be on the slot in which the DCI for scheduling an aperiodic report is received.In some implementations, each timing pattern may correspond to a timing pattern index.In some implementations, the unit of timing pattern may correspond to a slot / symbol / sub-slot.In some implementations, each bit in the timing pattern may correspond to a preconfigured slot / sub-slot / symbol.In some implementations, the unit of timing pattern may be based on the periodicity of the report. Thus, there may be multiple timing patterns based on the number of configured reports.In some implementations, the unit of timing pattern may be based on the periodicity of the reference signal configured by the second measurement configuration. Thus, the number of predicted results may be based on the number of measured RSs. Specifically, the maximum bit length of timing pattern may be different for respective frequency range (e.g., FR1, FR2).In some implementations, the first report configuration may include a parameter configuring the periodicity and / or offset of the report.In some implementations, the report may be transmitted on PUCCH, PUSCH, CG PUSCH, UE assistance information, and / or a UL MAC CE.In some implementations, the report may be transmitted periodically or semi-persistently.In some implementations, the report may include one or multiple measurement / predicted results within each periodicity. For example, if the periodicity of the report is 10ms / slot / sub-slot / symbol, the report may include all measurement / predicted results within that time interval (e.g., 10ms / slot / sub-slot / symbol).In some implementations, the first report configuration may include a parameter configuring the periodicity and / or the offset of the measurement / predicted results corresponding to a report.In some implementations, the report may correspond to a larger periodicity and the measurement results may correspond to a smaller periodicity. For example, the measurement results may be performed every 5ms and the report may be transmitted every 10ms.In some implementations, the report may correspond to an aperiodic report. In some implementations, if the report is aperiodic, the DCI scheduling the aperiodic report may include a field to indicate the associated timing pattern index. The aperiodic report may include the predicted results between the time interval starting from the slot receiving the DCI scheduling the aperiodic report to the slot transmitting the aperiodic report. In some implementations, if the report is aperiodic, the number of predicted results in the report (e.g., N) may be fixed / configured. In this case, the report may include the predicted results starting from the first slot in which the DCI scheduling the aperiodic report is received and continuing through slots preceding the first slot, until the number of predicted results in the report reaches N. In some implementations, a time offset K may be indicated or configured to indicate the time interval from the slot n in which the DCI scheduling is received to the slot n-K where the predicted results are included in the report.In some implementations, the number of predicted results in the report (e.g., N) may be configured to the UE via an RRC signaling. Specifically, the report configuration, the measurement configuration and / or the CSI-RS configuration may include a field / parameter to indicate the number of predicted results in the report (e.g., N). In some implementations, the number of predicted results in the report (e.g., N) may be indicated to the UE via a MAC CE. Specifically, the MAC CE may be used to activate the AI / ML based measurement. In some implementations, the number of predicted results in the report (e.g., N) may be indicated to the UE via a DCI, where the DCI may instruct the UE to transmit a CSI report.In some implementations, if the report is aperiodic, the DCI scheduling the aperiodic report may include at least one of a first field used to indicate the number of predicted results and a second field used to indicate the timing offset between each predicted result. For example, if the DCI schedules the N-slot to a UE for transmitting the predicted beam management result (e.g., aperiodic report), the DCI field used to indicate the number of predicted results indicates M and the field used to indicate the timing offset between each predicted result indicates K, the aperiodic report may include the beam management results related to N-slot, N+K slot, N+2K slot, …, and N+(M-1)K slot.In some implementations, if the report is aperiodic and scheduled by the DCI, the RRC report configuration corresponding to that aperiodic report may include at least one of a first field used to indicate the number of predicted results and a second field used to indicate the timing offset between each predicted result. For example, if the DCI schedules the N-slot to a UE for transmitting the predicted beam management result (e.g., aperiodic report), the DCI field used to indicate the number of predicted results indicates M and the field used to indicate the timing offset between each predicted result indicates K, the aperiodic report may include the beam management results related to N-slot, N+K slot, N+2K slot, …, and N+(M-1)K slot.In some implementations, the report may include all the measurement / predicted results obtained in the interval from the previous report to this report.In some implementations, the first report configuration may include a parameter configuring time duration between a reference signal and a report. Based on the given time duration, the UE may assume the starting point of this duration is based on the location of the reference signal.In some implementations, the reference signal may be configured in the first measurement configuration. In some implementations, the reference signal in the first measurement configuration may be pre-configured.In some implementations, if there are multiple reference signals configured in the first measurement configuration, each time duration value may be associated with one of the reference signals. More specifically, the reference signal associated with a resource index may be associated with a time duration.In some implementations, the reference signal may be configured in the second measurement configuration.In some implementations, if there are multiple reference signals configured in the second measurement configuration, each time duration value may be associated with one of the reference signals. More specifically, the reference signal associated with a resource index may be associated with a time duration.In some implementations, the report may include all the measurement results obtained in the interval from the reference signal to this report.In some implementations, one reference signal may be associated with one or more than one measurement configuration.In some implementations, the first report configuration may include a parameter configuring time duration between the received DCI and a report. In some implementations, the received DCI may be used to schedule a report. In some implementations, the received DCI may be used to indicate the start of generating one or more predicted results in the subsequent slot.In some implementations, the first report configuration may include a parameter configuring time duration between the received MAC CE and a report. In some implementations, the received MAC CE may be used to activate the first measurement configuration. In some implementations, the received MAC CE may be used to activate the first report configuration. In some implementations, the received MAC CE may be used to activate the timing pattern for AI / ML.In some implementations, the unit of time duration may correspond to a slot.In some implementations, the unit of time duration may correspond to a sub-slot.In some implementations, the unit of time duration may correspond to a symbol.In some implementations, the parameter may correspond to a single value.In some implementations, the parameter may correspond to multiple values. For example, each of the values may indicate a time instant for the report.In some implementations, the parameter may configure one or more than one time duration between successive reporting periods.In some implementations, the time duration may correspond to a table with row indexes. For example, the first row may correspond to a first time duration (e.g., 1 slot) and a second row may correspond to a second time duration (e.g., 5 slots), etc.In some implementations, the time interval for including one or more predicted results in one report may have a starting point and an ending point to explicitly inform the UE where to start collecting the predicted result and where to stop collecting the predicted result in the report, and the first report configuration may include a reference starting point to configure where the measurement starts from for a report.FIG. 3 is a diagram 300 illustrating association between predicted results and a report in time domain, according to an example implementation of the present disclosure. In some implementations, the report may contain the predicted results starting from the configured starting point. In some implementations, the reference starting point may be configured per report configuration. For example, if the reference starting point is configured as slot#1, the UE may determine that the predicted results starting from the slot#1 within the report periodicity (e.g., 10 slot) are included in the report. For example, if the reference starting point is configured as slot#1, the UE may determine that the predicted results starting from the slot#1 within each periodicity of the configured measurement configuration are included in the report.The reference starting point may correspond to a symbol index, a slot index, or a sub-slot index.In some implementations, the time interval for including one or more predicted results in one report may have a starting point and an ending point to explicitly inform the UE where to start collecting the predicted result and where to stop collecting the predicted result in the report, and the first report configuration may include a reference ending point to configure where the measurement ends for the report. As shown in FIG. 3, the UE may be provided or configured with the starting point and the ending point, from which the UE may determine the time duration between them. The report may then include predicted results that fall within the determined time duration. In some implementations, the reference ending point may refer to the slot where the report is. In some implementations, the reference ending point may be configured per report configuration. For example, if the reference starting point is configured as slot#1 and the ending point is configured as slot#n, the UE may determine that the predicted results starting from the slot#1 to the slot#n within the report periodicity (e.g., 10 slot) are included in the report. In some implementations, the starting point and the ending point may be within each periodicity of the report. In some implementations, when the predicted result is generated on the same slot where the starting point is, the report may also contain the predicted result on this slot. In some implementations, when the predicted result is generated on the same slot where the starting point is, the report may not contain the predicted result on this slot.The reference ending point may correspond to a symbol index or a slot index.In some implementations, different predicted results may be inferred from the same or different reference signal in the second measurement configuration (e.g., configuration for set B beams). More specifically, more than one predicted results may be associated with the same resource provided by the first or the second measurement configuration. On the other hand, one predicted result may be associated with more than one resources provided by the first or the second measurement configuration.In some implementations, different predicted results may be inferred from the same or different measurement configuration. More specifically, more than one predicted results may be associated with the same measurement configuration. On the other hand, one predicted result may be associated with more than one measurement configurations.In some implementations, the maximum number of measurement results in a single report may be configured or based on a UE capability. In some implementations, a UE may report its capability of the maximum number of measurement results in a single report to the network via an RRC signaling.Report contentIn some implementations, the report content may include one or more than one (e.g., top K) beam information (e.g., beam index, TCI state index, CRI, SSBRI), where K is a positive integer.In some implementations, the report content may include one or more than one (e.g., top K) RSRP value.In some implementations, the report content may include one or more than one (e.g., top K) differential RSRP value.In some implementations, the report content may include one absolute RSRP value (e.g., top 1) and one or more than one (e.g., top 2 to top K) differential RSRP values.In some implementations, the report content may include one or more than one (e.g., top K) probability information of each beam. More specifically, the probability information may indicate the probability of a measured beam being among the top K beams. In some implementations, the sum of all probability values may be equal to one.In some implementations, the report content may include one or more than one (e.g., top K) confidence information. More specifically, the confidence information may indicate how accurate the inferred value (e.g., measurement result) is. In some implementations, each confidence information may be a value between a first value (e.g., 0) and a second value (e.g., 100). In some implementations, each confidence information may be represented as a 7-bit string, with values from 101 to 127 remaining unused. In some implementations, the confidence information with the highest value may be reported as a first value and being an absolute value (e.g., from 0 to 100). The rest K-1 confidence information may be reported as differential value(s) compared to the highest confidence value. The differential value may be represented as a bit string including 1, 2, 3, 4, 5 or 6 bits.In some implementations, the report may be a measurement result codebook containing multiple measurement results in different time instances.In some implementations, the value of K may be configurable by the RAN, the network, or the (AI) service provider. In some implementations, the value of K may be pre-defined by a technical specification or pre-installed in the memory module of the UE. In some implementations, the UE may report the value of K to the network or (AI) service provider via an RRC signaling or a NAS signaling.In some implementations, the order of measurement results in a codebook may be based on the timing order. More specifically, the measurement result with the largest time duration may be placed first and the measurement result with the smallest time duration may be placed last.MAC procedure or MAC CE for AI / ML BMIn some implementations, a counter for AI / ML BM may be set to count the number of measurement results in a report for Set A of beams and determine when the UE transmits the report. In some implementations, the MAC entity of UE may maintain a counter to trace the number of measurement results within a report for Set A of beams and the UE may indicate particular conditions to the low layer and / or upper layer to make adjustments in LCM.In some implementations, in the beginning, the counter may be set to an initial value (e.g., 0), and the counter may be incremented by 1 when the UE generates predicted results. After the UE performs the prediction or generates the predicted results, an indication from the physical layer may be sent to the MAC entity. The indication from the physical layer may not be sent to the MAC entity if the prediction or generation of the predicted results is not sufficiently good. For example, if the prediction belongs to a certain confidence level or lower than an accurate probability, the prediction may be identified as not sufficiently good. Then the counter may be incremented by 1 if the indication from the lower layer has been received. Afterwards, when the counter value is greater than or equal to a maximum value or a preconfigured value, the MAC entity may instruct the physical layer to transmit the report for Set A of beams. Specifically, the maximum value or the preconfigured value may refer to the maximum number of predicted results included in the report. In some implementations, the predicted results may be generated periodically or aperiodically. In some implementations, the MAC entity may restart the counter when the UE receives the MAC CE or DCI to indicate the report request.In some implementations, in the beginning, the counter may be set to an initial value (e.g., 0), and the counter may be incremented by 1 when the UE receives one or more configurations for Set A of beams. After the UE receives the one or more configurations, an indication from the physical layer may be sent to the MAC entity. Then the counter may be incremented by 1 if the indication from the lower layer has been received. Afterwards, when the counter value is greater than or equal to a maximum value or a preconfigured value, the MAC entity may instruct the physical layer to transmit the report for Set A of beams. Specifically, the maximum value or the preconfigured value may refer to the maximum number of predicted results included in the report. In some implementations, the UE may receive each of the configurations randomly in different time locations.In some implementations, in the beginning, the counter may be set to an initial value (e.g., 0), and the counter may be incremented by 1 when the UE receives one or more configurations for Set B of beams. After the UE receives the one or more configurations, an indication from the physical layer may be sent to the MAC entity. Then the counter may be incremented by 1 if the indication from the lower layer has been received. Afterwards, when the counter value is greater than or equal to a maximum value or a preconfigured value, the MAC entity may instruct the physical layer to transmit the report for Set B of beams. Specifically, the maximum value or the preconfigured value may refer to the maximum number of predicted results included in the report. In some implementations, the UE may receive each of the configurations randomly in different time locations.In some implementations, a timer for AI / ML BM may be set to instruct the UE to transmit the report. The detailed procedure may be described as follows.First step: When the UE receives the first measurement configuration or the first report configuration, the MAC entity may start the timer. In some implementations, the timer may be configured in the first measurement configuration or the first report configuration.Second step: Check if the indication from the lower layer has been received by the MAC entity of the UE. The timer may continue running regardless of whether the indication from the lower layer has been received.Second step: Check if the MAC CE or the DCI for scheduling / triggering the report or the DCI for indicating the ending of collecting the predicted result is received by the UE. The MAC entity may stop the timer in response to receiving the MAC CE or the DCI.Third step: When the timer expires, the MAC entity of the UE may trigger the report for Set A of beams.In some implementations, the timer may correspond to a periodicity. More specifically, the timer may expire periodically (e.g., every 10ms).In some implementations, the value of the timer may be configured or predefined.In some implementations, a first timer and a second counter may be set to instruct the UE to transmit report including one or more predicted results. In some implementations, the first timer may be used to trigger the report for Set A of beams and the second counter may be used to count the number of predicted results. The detailed procedure may be described as follows.First step: When the UE receives the first measurement configuration or the first report configuration, the MAC entity may start the first timer and the second counter.Second step: If the indication from the lower layer used to indicate the stop of generating the predicted results has been received by the UE MAC entity, the MAC entity may restart the second counter. In some implementations, the criteria for sending the indication may be that the counter value is larger than or is equal to a configured / predefined value.Second step: If the UE receives the MAC CE or DCI for indicating the report request, the MAC entity may restart the second counter and the first timer.Second step: Else, the MAC entity may start the second counter and set the value to ‘1’ as the initial value.Third step: When the UE generates the predicated result and sends an indication from physical layer, the counter may be incremented by 1.Fourth step: If the first timer expires, the MAC entity may stop the second counter and trigger the report for Set A of beams.In some implementations, the first timer may correspond to a periodicity.In some implementations, the value of the first timer may be configured or predefined, and the first timer may be configured in the first measurement configuration or the first report configuration.In some implementations, a MAC CE for AI / ML BM may be used to instruct the UE to transmit the report for Set A of beams.In some implementations, a first MAC CE may be used to indicate the timing pattern of the predicted results (e.g., timing pattern index) that is configured by RRC and start generating the predicted results based on the received timing pattern index. A second MAC CE may be used to instruct the UE to transmit the report containing the predicted results.In some implementations, a MAC CE for AI / ML BM may be used to indicate the timing pattern of the predicted results and trigger the report containing the predicted results.In some implementations, one or more than one MAC CEs may include the following information: a cell index, a BWP index, a measurement configuration index, a report configuration index, a bitmap for the timing of the measurement result, and an indication of the report.In some implementations, the measurement configuration index may be associated with the first measurement configuration or the second measurement configuration. In some implementations, the measurement configuration may configure the relationship between the time and predicted result, and thus the reception of the MAC CE may be the trigger / activation of performing measurement or inference. For example, upon receiving the MAC CE, the UE may start performing generating the predicted result according to the measurement configuration that is indicated by the measurement configuration index in the received MAC CE.Report configuration index: In some implementations, the report configuration may be associated with the first report configuration. In some implementations, the MAC CE may activate the indicated report configuration. For example, upon receiving the MAC CE, the UE may start reporting the measurement results according to the report configuration that corresponds to the report configuration index in the received MAC CE. In some implementations, the MAC CE may be used to indicate a periodic report, an aperiodic report, or activate a semi-persistent report.Bitmap for the timing of the measurement result: The bitmap may have a length of P and each bit in the bitmap may be represented as Ti. Ti (0 <= i < P) may correspond to the ith slot / sub-slot / symbol / ms, starting from the time in which the MAC CE is received. In some implementations, Ti = 1 may represent the location of one measurement result on Ti and Ti=0 may represent that no corresponding measurement result is on Ti. In some implementations, Ti=1 (Ti may be the value corresponding to the ith entry of a table (preconfigured by an RRC signaling) with unit in slot / sub-slot / symbol / ms) may represent the location of one predicted result on the value corresponding to Ti and Ti=0 may represent that no corresponding measurement result is on the value corresponding to Ti.Indication of the report: The indication of the report may be a bit field. In some implementations, the bit value =1 may represent that the beam report is transmitted after x ms or x slot / sub-slot / symbol of the reception of the MAC CE. The bit value=0 may represent that the beam report is not transmitted. In some implementations, if the beam report is not needed, this field may be reserved.DCI for AI / ML BMIn some implementations, a DCI format dedicated to scheduling AI / ML BM may be used to indicate the report for AI / ML BM. In other words, a new DCI format may be used to indicate the report containing one or more predicted results. In some implementations, the UE may transmit a report containing only predicted results for set A of beams.In some implementations, a DCI format dedicated to scheduling AI / ML BM may be used to indicate the measurement result occasion. In some implementations, the DCI format may include a field to indicate the timing pattern index, and thus the UE may perform generating one or more predicted results based on the indicated timing pattern.In some implementations, a DCI format may include a field to indicate the report request. In some implementations, the UE may transmit a report based on the field and the report may contain both predicted results and measurement results based on the second measurement configuration. In some implementations, the UE may transmit a report based on the field and the report may contain only predicted results.In some implementations, a DCI format may include a field to indicate the time duration between the DCI format and the report. In some implementations, the time duration indicated by the DCI format may override the configured time duration. In some implementations, the UE may include one or more generated predicted results during the indicated time period in the report.In some implementations, if a DCI format including a field to indicate the time duration for generating the predicated results and a DCI format including a field to indicate the time gap between DCI and an UL transmission are the same DCI, the value of time duration for generating the predicated results may be smaller than the value of time gap between the DCI and the UL transmission.In some implementations, if a DCI format including a field to indicate the time duration for generating the predicated results and a DCI format including a field to indicate the time gap between DCI and an UL transmission are different DCI, the value of time duration for generating the predicated results may be smaller, larger than, or equal to the value of time gap between DCI and the UL transmission.In some implementations, a DCI format may include a field to indicate time duration between the DCI format and the predicted result. In some implementations, the time duration indicated by the DCI format may override the configured time duration. In some implementations, the DCI format may be used to instruct the UE to generate the predicted result after receiving the DCI format, and the time offset between the DCI and the predicted result may be based on the indicated field. More specifically, the time at which the UE receives the DCI may be the starting point of the time duration.In some implementations, the field in the DCI format may indicate one or more than one time duration. When there are more than one time duration values, the time duration may correspond to a list or a table. For example, if the table with index #0 contains the time duration values [0 2 5 7], the UE may generate the predicted result on the same slot in which the DCI is received, 2 slots with respect to the received DCI, 5 slots with respect to the received DCI, 7 slots with respect to the received DCI. Thus, different entries in the table may indicate different timing patterns for one or more predicted results.In some implementations, a DCI format may include a field to indicate the measurement configuration index.In some implementations, a DCI format may include a field to indicate the report configuration index.In some implementations, a DCI format may include a field to indicate the starting point of collecting predicted results and / or the ending point of collecting predicted results. For example, if the indicated starting point is 1 slot, the UE may start collecting the predicted result on 1 slot after the DCI is received. Furthermore, if the indicated ending point is 10 slots, then the UE may stop collecting the predicted result on 10 slots later than the slot in which the DCI is received.In some implementations, a DCI format may be scrambled by a CRC with an AI / ML BM-specific RNTI to schedule an AI / ML BM report, allowing differentiation between the report for AI / ML and the report for CSI measurement.In some implementations, a DCI format may be scrambled by a CRC with an AI / ML BM-specific RNTI to schedule inferred measurement results for AI / ML BM.UE behaviorIn some implementations, a UE may perform beam report based on a received RRC message (e.g., RRC reconfiguration message), MAC-CE, and / or DCI to generate one or more than one measurement results in a single report for AL / ML BM.In some implementations, a UE may report a UE capability to report whether the UE supports including multiple measurement results from different time instances in a single report.In some implementations, a UE may receive an RRC message (e.g., measurement configuration for AI / ML and / or report configuration for AI / ML), a MAC-CE, and / or DCI including measurement / reporting information from a BS.In some implementations, when a UE receives a time duration value (e.g., T) from RRC / MAC CE / DCI in slot n, the UE may transmit a beam report in the slot n+T and the beam report may include the measurement results from slots n to n+T.In some implementations, the UE may receive an RRC message, a MAC-CE, and DCI in a predetermined order.In some implementations, the UE may receive an RRC message first and receive a MAC CE (e.g., activation command) after receiving the RRC message.In some implementations, the UE may receive an RRC message first and receive DCI after receiving the RRC message.In some implementations, the UE may receive a MAC CE first and receive DCI after receiving the MAC CE.In some implementations, the UE may first receive a RRC message first, followed by a MAC CE, and then receive DCI.In some implementations, the UE may expect one or more than one predicted result in a beam report, starting from S0 and ending at Sf, without no further indication or higher layer signaling indicating where to start collecting the predicted results and where to finish collecting the predicted result.In some implementations, S0 may not be expected to start before a symbol with CP, and the symbol with CP may start after a processing time, where the processing time may start after a last symbol of corresponding reception of the RRC message.In some implementations, S0 may be expected to start before a symbol with CP, and the symbol with CP may start after a processing time, where the processing time may start after a last symbol of corresponding reception of the MAC CE.In some implementations, S0 may not be expected to start before a symbol with CP, and the symbol with CP may start after a processing time, where the processing time may start after a last symbol of corresponding reception of the DCI.In some implementations, Sf may be the slot in which the UE receives the MAC CE for indicating AI / ML report.In some implementations, Sf may be the slot in which the UE receives DCI for scheduling AI / ML report.In some implementations, Sf may be located at a slot that is a fixed time duration after the S0 (e.g., Sf = S0+ X value).In some implementations, Sf may be the slot in which the UE transmits the report.In some implementations, S0 may correspond to a fixed value after the UE receives the RRC message / MAC CE / DCI (e.g., x slots after the UE receives the RRC / MAC CE / DCI).In some implementations, if a UE receives an RRC message, a MAC CE, and / or a DCI to indicate the time duration for collecting measurement results with different values or patterns, the UE may apply the indication according to the priority. In some implementations, the DCI may have the highest priority order when the DCI, the MAC CE and the RRC message carry the information for the same purpose (e.g., time duration indication). In some implementations, the MAC CE may have a higher priority than the RRC message when the MAC CE and the RRC message carry the information for the same purpose (e.g., time duration indication). In other words, the UE may keep the stored time duration value for collecting measurement results before it is overwritten by a message with a higher priority or an equivalent priority.BS behaviorIn some implementations, a BS may transmit a first measurement configuration / indication and / or a second measurement configuration / indication including timing pattern information via an RRC message, a MAC-CE and / or a DCI to inform a UE to perform inferring / measurement for AI / ML BM.In some implementations, a BS may transmit a first report configuration / indication and / or a second report configuration / indication via an RRC message, a MAC-CE and / or a DCI to inform a UE to trigger the report for AI / ML BM.In some implementations, a BS may transmit a MAC CE to indicate the time duration during which the UE collects one or more predicted results in the report.In some implementations, a BS may transmit a MAC CE to activate one timing pattern in the RRC configuration.In some implementations, a BS may transmit a DCI format to indicate the time duration during which the UE collects one or more predicted results in the report.In some implementations, a BS may receive a UE capability that indicates whether the UE supports including more than one predicted results from different time instances in a single report.In some implementations, a BS may transmit an RRC message, a MAC-CE, and / or a DCI including measurement timing / report information to a UE.In some implementations, a BS may transmit an RRC message, a MAC-CE, and a DCI in a predetermined order.In some implementations, a BS may transmit an RRC message first and transmit a MAC CE after transmitting the RRC message.In some implementations, a BS may transmit an RRC message first and transmit a DCI after transmitting the RRC message.In some implementations, a BS may transmit a MAC CE first and transmit a DCI after transmitting the MAC CE.In some implementations, a BS may transmit an RRC message first, followed by a MAC CE, and then transmit DCI.FIG. 4 is a flowchart illustrating a method / process 400 performed by a UE for AI / ML-based BM, according to an example implementation of the present disclosure. In the action 402, the process 400 may start by receiving, from the BS, a first RRC configuration indicating a set of reference signal resources. The first RRC configuration may correspond to a measurement configuration. The set of reference signal resources may include CSI-RS resources, SSB resources, or both CSI-RS resources and SSB resources. The set of reference signal resources to be measured by the UE may be referred to as Set B beams.In the action 404, the process 400 may receive, from the BS, a second RRC configuration indicating time information for a CSI report. The second RRC configuration may correspond to a report configuration. The time information of the CSI report may be used specifically for the AI / ML purpose. For example, the time information may be associated with future time instances that correspond to predicted results in the CSI report. In the action 406, the process 400 may determine at least one future time instance based on the time information for the CSI report.In the action 408, the process 400 may generate at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources. The at least one predicted result generated via the AI / ML model may be referred to as Set A beams. In the action 410, the process 400 may transmit, to the BS, the CSI report including the at least one predicted result.In some implementations, the process 400 may receive, from the BS, a parameter indicating the number of the at least one predicted result to be included in the CSI report. In some implementations, the parameter, which indicates N predicted results in the CSI report, may correspond to the time information in the second RRC configuration, where N may be a positive integer. In some implementations, the parameter may be carried in a DCI from the BS. In some implementations, the parameter (e.g., N) may be carried in a DCI that instructs the UE to transmit the CSI report.In some implementations, the time information for the CSI report may include a time offset. In the action 406, determining the at least one future time instance may include separating any two consecutive future time instances by the time offset. The unit of the time offset may be a slot, sub-slot, symbol, or ms. For example, if the CSI report includes 4 predicted results (e.g., N=4) and the time offset is 10 slots, each pair of successive predicted results in the CSI report may correspond to time instances that are separated by 10 slots.In some implementations, the second RRC configuration may include a CSI report configuration.In some implementations, the time information for the CSI report may indicate the earliest future time instance among the at least one future time instance. For example, if the CSI report includes 4 predicted results (e.g., N=4), the time information may indicate the earliest future time instance corresponding to the first predicted result among the 4 predicted results. In some implementations, the earliest future time instance may be configured on a per-report basis. For example, each report configuration may indicate its own time information for the earliest time instance.In some implementations, the process 400 may activate an AI / ML function before receiving the second RRC configuration. In some implementations, the UE may receive the second RRC configuration only when the AI / ML function is activated / configured / applied. The activation of the AI / ML function may be notified by the UE to the NW, or initiated by the NW.In some implementations, each of the at least one predicted result may correspond to an RSRP value. In some implementations, one of the at least one predicted result may correspond to an absolute RSRP value, while the remaining predicted results may correspond to differential RSRP values relative to the absolute RSRP value.In some implementations, the second RRC configuration may include an index for associating the at least one predicted result with the at least one measurement result. In some implementations, the second RRC configuration may include an index for the first RRC configuration to associate the first RRC configuration with the second RRC configuration. In some implementations, the first RRC configuration may include an index for the second RRC configuration to associate the second RRC configuration with the first RRC configuration. Due to the association between the at least one predicted result and the at least one measurement result (e.g., the association between the first RRC configuration and the second RRC configuration), the UE may generate the CSI report that includes the least one predicted result based on the at least one measurement result obtained from the reference signal resources. The CSI report may be generated according to the report configuration corresponding to the second RRC configuration. The at least one measurement result may be obtained from the reference signal resources corresponding to the first RRC configuration.The steps / actions shown in FIG. 4 should not be construed as necessarily order dependent. The order in which the process is described is not intended to be construed as a limitation. Moreover, some of the actions shown in FIG. 4 may be omitted in some implementations and one or more actions shown in FIG. 4 may be combined.The technical problem addressed by the method illustrated in FIG. 4 is enhanced beam management. The method illustrated in FIG. 4 introduces a reporting mechanism for temporal downlink beam prediction that effectively handles multiple predicted beam results within a single beam report. The technical problem addressed by this method is the need for an efficient and accurate method for AI / ML-based beam management in a UE. The problem lies in how the UE generates a CSI report that incorporates one or more predicted results for future time instances based on received RRC configurations. By utilizing AI / ML models to generate predictions based on measurement results and time information configured via RRC signaling, the UE is able to provide more accurate and timely CSI reports. This improves beam management, which may enhance overall system performance, such as more efficient beam switching, better throughput, and reduced latency in wireless communication systems. Specifically, by providing the time information for the CSI report via RRC signaling, the UE is able to generate the required CSI report with predicted results at proper future time instances.FIG. 5 is a flowchart illustrating a method / process 500 performed by a BS for facilitating AI / ML-based BM, according to an example implementation of the present disclosure. In the action 502, the process 500 may start by transmitting, to a UE, a first RRC configuration indicating a set of reference signal resources. In the action 504, the process 500 may transmit, to the UE, a second RRC configuration indicating time information for a CSI report. In the action 506, the process 500 may receive, from the UE, the CSI report including at least one predicted result. The process 500 may then end.The UE may determine at least one future time instance based on the time information for the CSI report. The UE may generate the at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources. The method illustrated in FIG. 5 is similar to that in FIG. 4, except that it is described from the perspective of the BS (instead of the UE).FIG. 6 is a block diagram illustrating a node 600 for wireless communication in accordance with various aspects of the present disclosure. As illustrated in FIG. 6, a node 600 may include a transceiver 620, a processor 628, a memory 634, one or more presentation components 638, and at least one antenna 636. The node 600 may also include a radio frequency (RF) spectrum band module, a BS communications module, a network communications module, and a system communications management module, Input / Output (I / O) ports, I / O components, and a power supply (not illustrated in FIG. 6).Each of the components may directly or indirectly communicate with each other over one or more buses 640. The node 600 may be a UE or a BS that performs various functions disclosed with reference to FIGS. 1 through 5.The transceiver 620 has a transmitter 622 (e.g., transmitting / transmission circuitry) and a receiver 624 (e.g., receiving / reception circuitry) and may be configured to transmit and / or receive time and / or frequency resource partitioning information. The transceiver 620 may be configured to transmit in different types of subframes and slots including, but not limited to, usable, non-usable, and flexibly usable subframes and slot formats. The transceiver 620 may be configured to receive data and control channels.The node 600 may include a variety of computer-readable media. Computer-readable media may be any available media that may be accessed by the node 700 and include volatile (and / or non-volatile) media and removable (and / or non-removable) media.The computer-readable media may include computer-storage media and communication media. Computer-storage media may include both volatile (and / or non-volatile media), and removable (and / or non-removable) media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or data.Computer-storage media may include RAM, ROM, EPROM, EEPROM, flash memory (or other memory technology), CD-ROM, Digital Versatile Disks (DVD) (or other optical disk storage), magnetic cassettes, magnetic tape, magnetic disk storage (or other magnetic storage devices), etc. Computer-storage media may not include a propagated data signal. Communication media may typically embody computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave, or other transport mechanisms and include any information delivery media.The term “modulated data signal” may mean a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. Communication media may include wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above listed components should also be included within the scope of computer-readable media.The memory 634 may include computer-storage media in the form of volatile and / or non-volatile memory. The memory 634 may be removable, non-removable, or a combination thereof. Example memory may include solid-state memory, hard drives, optical-disc drives, etc. As illustrated in FIG. 6, the memory 634 may store a computer-readable and / or computer-executable instructions 632 (e.g., software codes) that are configured to, when executed, cause the processor 628 to perform various functions disclosed herein, for example, with reference to FIGS. 1 through 5. Alternatively, the instructions 632 may not be directly executable by the processor 628 but may be configured to cause the node 600 (e.g., when compiled and executed) to perform various functions disclosed herein.The processor 628 (e.g., having processing circuitry) may include an intelligent hardware device, e.g., a Central Processing Unit (CPU), a microcontroller, an ASIC, etc. The processor 628 may include memory. The processor 628 may process the data 630 and the instructions 632 received from the memory 634, and information transmitted and received via the transceiver 620, the baseband communications module, and / or the network communications module. The processor 628 may also process information to send to the transceiver 620 for transmission via the antenna 636 to the network communications module for transmission to a CN.One or more presentation components 638 may present data indications to a person or another device. Examples of presentation components 638 may include a display device, a speaker, a printing component, a vibrating component, etc.In view of the present disclosure, it is obvious that various techniques may be used for implementing the disclosed concepts without departing from the scope of those concepts. Moreover, while the concepts have been disclosed with specific reference to certain implementations, a person of ordinary skill in the art may recognize that changes may be made in form and detail without departing from the scope of those concepts. As such, the disclosed implementations are to be considered in all respects as illustrative and not restrictive. It should also be understood that the present disclosure is not limited to the particular implementations disclosed and many rearrangements, modifications, and substitutions are possible without departing from the scope of the present disclosure.

Claims

1. A User Equipment (UE) for performing Artificial Intelligence (AI) / Machine Learning (ML)-based Beam Management (BM), the UE comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to the at least one processor and storing one or more computer-executable instructions that, when executed by the at least one processor, cause the UE to: 　　　　receive, from a base station (BS), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; 　　　　receive, from the BS, a second RRC configuration indicating time information for a Channel State Information (CSI) report; 　　　　determine at least one future time instance based on the time information for the CSI report; 　　　　generate at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources; and 　　　　transmit, to the BS, the CSI report including the at least one predicted result.

2. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　receive, from the BS, a parameter indicating the number of the at least one predicted result to be included in the CSI report.

3. The UE of claim 1, wherein: 　　　　the time information for the CSI report comprises a time offset, and 　　　　determining the at least one future time instance comprises separating any two consecutive future time instances by the time offset.

4. The UE of claim 1, wherein: 　　　　the second RRC configuration comprises a CSI report configuration.

5. The UE of claim 1, wherein: 　　　　the time information for the CSI report indicates the earliest future time instance among the at least one future time instance.

6. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　activate an AI / ML function before receiving the second RRC configuration.

7. The UE of claim 1, wherein: 　　　　each of the at least one predicted result corresponds to a Reference Signal Received Power (RSRP) value.

8. The UE of claim 1, wherein: 　　　　the second RRC configuration comprises an index for associating the at least one predicted result with the at least one measurement result.

9. A Base Station (BS) for facilitating Artificial Intelligence (AI) / Machine Learning (ML)-based beam management, the BS comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to the at least one processor and storing one or more computer-executable instructions that, when executed by the at least one processor, cause the BS to: 　　　　　　　　transmit, to a User Equipment (UE), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; 　　　　　　　　transmit, to the UE, a second RRC configuration indicating time information for a Channel State Information (CSI) report; and 　　　　　　　　receive, from the UE, the CSI report including at least one predicted result, wherein: 　　　　the UE determines at least one future time instance based on the time information for the CSI report, 　　　　the UE generates the at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources.

10. The BS of claim 9, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to: 　　　　transmit, to the UE, a parameter indicating the number of the at least one predicted result to be included in the CSI report.

11. The BS of claim 9, wherein: 　　　　the time information for the CSI report comprises a time offset, and 　　　　the UE determines the at least one future time instance by separating any two consecutive future time instances by the time offset.

12. The BS of claim 9, wherein: 　　　　the second RRC configuration comprises a CSI report configuration.

13. The BS of claim 9, wherein: 　　　　the time information for the CSI report indicates the earliest future time instance among the at least one future time instance.

14. The BS of claim 9, wherein: 　　　　each of the at least one predicted result corresponds to a Reference Signal Received Power (RSRP) value, and 　　　　the second RRC configuration comprises an index for associating the at least one predicted result with the at least one measurement result.

15. A method performed by a User Equipment (UE) for Artificial Intelligence (AI) / Machine Learning (ML)-based beam management, the method comprising: 　　　　receiving, from a base station (BS), a first Radio Resource Control (RRC) configuration indicating a set of reference signal resources; 　　　　receiving, from the BS, a second RRC configuration indicating time information for a Channel State Information (CSI) report; 　　　　determining at least one future time instance based on the time information for the CSI report; 　　　　generating at least one predicted result for the at least one future time instance using an AI / ML model based on at least one measurement result obtained from the set of reference signal resources; and 　　　　transmitting, to the BS, the CSI report including the at least one predicted result.

Images