Method and apparatus for measurement reporting for artificial intelligence / machine learning enabled mobility

Abstract

A method performed by a UE for artificial intelligence (AI) / machine learning (ML) enabled mobility is provided. The method receives, from a base station (BS), a configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction. The method in response to receiving the configuration for the RRM prediction, reports, via a measurement report message, one or more first predicted measurement results to the BS. The method in response to receiving the configuration for the measurement event prediction, reports, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, where the timing information is associated with the predicted measurement event.

Description

METHOD AND APPARATUS FOR MEASUREMENT REPORTING FOR ARTIFICIAL INTELLIGENCE / MACHINE LEARNING ENABLED MOBILITY

[0001] The present disclosure is related to wireless communication and, more specifically, to a User Equipment (UE), Base Station (BS), and method for measurement reporting for artificial intelligence (AI) / machine learning (ML) enabled mobility in the wireless communication networks.

[0002] 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.

[0003] The present disclosure is related to a UE, a BS, and a method for measurement reporting for artificial intelligence (AI) / machine learning (ML) enabled mobility in the wireless communication networks.

[0004] In a first aspect of the present disclosure, a UE for measurement reporting for artificial intelligence (AI) / machine learning (ML) enabled mobility 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 configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction, in response to receiving the configuration for the RRM prediction, report, via a measurement report message, one or more first predicted measurement results to the BS, and in response to receiving the configuration for the measurement event prediction, report, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, where the timing information is associated with the predicted measurement event.

[0005] In some implementations of the first aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to include the predicted measurement event in the measurement report message.

[0006] In some implementations of the first aspect, the configuration indicates a periodic reporting or an event-triggered reporting for the one or more first predicted measurement results.

[0007] In some implementations of the first aspect, the configuration includes an event related parameter for the measurement event prediction.

[0008] In some implementations of the first aspect, the configuration indicates a prediction window (PW) for the RRM prediction.

[0009] In some implementations of the first aspect, the configuration includes a synchronization signal block (SSB) configuration for the RRM prediction.

[0010] In some implementations of the first aspect, the one or more first predicted measurement results in the measurement report message are reported on a per-cell basis.

[0011] In a second aspect of the present disclosure, a method performed by a user equipment (UE) for artificial intelligence (AI) / machine learning (ML) enabled mobility is provided. The method includes receiving, from a base station (BS), a configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction, in response to receiving the configuration for the RRM prediction, reporting, via a measurement report message, one or more first predicted measurement results to the BS, and in response to receiving the configuration for the measurement event prediction, reporting, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, where the timing information is associated with the predicted measurement event.

[0012] In a third aspect of the present application, a BS for artificial intelligence (AI) / machine learning (ML) enabled mobility 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 configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction, where the configuration causes the UE to in response to receiving the configuration for the RRM prediction, report, via a measurement report message, one or more first predicted measurement results to the BS, and in response to receiving the configuration for the measurement event prediction, report, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, where the timing information is associated with the predicted measurement event.

[0013] 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.

[0014] FIG. 1 is a diagram illustrating predicted RRM measurement results corresponding to nearest and farthest future time instances relative to a current time, according to an example implementation of the present disclosure.

[0015] FIG. 2 is a diagram illustrating characterizations of a window based on offset and duration, according to an example implementation of the present disclosure.

[0016] FIG. 3 is a flowchart illustrating a method / process performed by a UE for measurement reporting for Artificial Intelligence / Machine Learning (AI / ML) enabled mobility, according to an example implementation of the present disclosure.

[0017] FIG. 4 is a block diagram illustrating a node for wireless communication, according to an example implementation of the present disclosure.

[0018] The 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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.

[0025] A software implementation may include computer-executable instructions and / or Artificial Intelligence (AI) / Machine Learning (ML) module(s) 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 computer-executable instructions and perform the disclosed network function(s), AI / ML module(s), or algorithm(s). The AI / ML module(s) may be implemented with a supervised learning approach, a semi-supervised learning approach, an unsupervised learning approach (e.g., Transductive approach and Inductive approach), a federated learning approach, or a reinforcement learning (RL) approach, but the present disclosure is not limited thereto. The computer-executable instructions associated with the AI module(s) and / or the ML module(s) may include, but are not limited to, data management instructions (e.g., collection instructions, validation instructions…etc.), model monitoring and management instructions (e.g., NW KPIs monitoring, model input / output monitoring, model selection / switching / update / upload / download, model (de)activation, model identification, functionality selection…etc.), and / or pre-process input instructions.

[0026] The microprocessors or general-purpose computers may include Application-Specific Integrated Circuits (ASICs), programmable logic arrays, Central Processing Units (CPUs), Tensor Processing Units (TPUs), Graphics Processing Units (GPUs), General-purpose computing on GPUs (GPGPU, or less often GPGP), and / or one or more Digital Signal Processors (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 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 (e.g., computer-readable instructions related to AI module(s) and / or the ML module(s)), data structures, program modules or data. The computer-readable medium may include, but is not limited to, Random Access Memory (RAM), Dynamic Random Access Memory (DRAM), High Bandwidth Memory (HBM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Resistive Random Access Memory (RRAM), Read-Only Memory (ROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory (or other memory technology), Compact Disc Read-Only Memory (CD-ROM) , Digital Versatile Disks (DVD) (or other optical disk storage), magnetic cassettes, magnetic tape, magnetic disk storage (or other magnetic storage devices), or any other equivalent medium capable of storing computer-readable instructions. Computer-storage media may not include a propagated data signal. Communication media may typically embody computer-readable instructions (e.g., computer-readable instructions related to AI module(s) and / or the ML module(s)), 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.

[0027] 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 previously listed components should also be included within the scope of computer-readable media.

[0028] A radio communication network architecture such as a Long-Term Evolution (LTE) system, an LTE-Advanced (LTE-A) system, an LTE-Advanced Pro system, a 5G NR Radio Access Network (RAN), 5G-Advanced (5G-A) system, or an open radio access network (O-RAN) may typically include at least one base station (BS), at least one UE, and one or more optional network elements that provide connection within a network. The BS and one or more optional network elements enable the UE to access a radio network. Thus, the UE may communicate with the network, such as a Core Network (CN), an Evolved Packet Core (EPC) network, an Evolved Universal Terrestrial RAN (E-UTRAN), a Next-Generation Core (NGC), a 5G Core (5GC), or an internet via a RAN established by one or more BSs and the network elements / functions.

[0029] 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, a virtual reality (VR) device, an augmented (AR) device, an Internet of Things (IoT) device, an unmanned aerial vehicle (UAV), or a Personal Digital Assistant (PDA) with wireless communication capability. The UE may be configured to receive and transmit signals over an air interface to one or more cells in a RAN. In some implementations, the UE may be an AI / ML-enabled device and / or an AI / ML capable device that is equipped with AI module(s) and / or ML module(s).

[0030] 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.

[0031] 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. In some implementations, the BS may be an AI / ML-enabled device and / or an AI / ML capable device that is equipped with AI module(s) and / or ML module(s).

[0032] 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.

[0033] 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.

[0034] A cell may allocate Sidelink (SL) resources for supporting Proximity Service (ProSe), LTE SL services, LTE / NR sidelink communication services, LTE / NR sidelink discovery services, and / or LTE / NR Vehicle-to-Everything (V2X) services. In addition, a cell may allocate DL and / or UL resources for supporting Multicast / Broadcast Service (MBS) services, Non-Terrestrial Networks (NTN) services, positioning services, power serving services and / or Network Energy Saving (NES) services.

[0035] 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.

[0036] The terms, definitions, and abbreviations as given in the present disclosure may be either imported from existing documentation (e.g., European Telecommunications Standards Institute (ETSI), International Telecommunication Union (ITU), or elsewhere) or newly created by 3GPP experts whenever the need for precise vocabulary is identified.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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.

[0042] 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.

[0043] 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.

[0044] The term “A and / or B” within the present disclosure means “A”, “B”, or “A and B”. The term “A and / or B and / or C” within the present disclosure means “A”, “B”, “C”, “A and B”, “A and C”, “B and C”, or “A and B and C”. The term “A / B” within the present disclosure means “A” or “B”.

[0045] In some implementations, the network (NW), cell, camped cell, serving cell, base station, gNB, eNB and ng-eNB may be used interchangeably in the present disclosure. In some implementations, some of these items may refer to the same network entity.

[0046] The RAT may be (but not limited to) NR, LTE, E-UTRA connected to 5GC, LTE connected to 5GC, E-UTRA connected to EPC, and LTE connected to EPC. The proposed mechanism may be applied for UEs in public networks, or in private network (e.g., non-public network (NPN), standalone NPN (SNPN), public network integrated NPN (PNI-NPN)).

[0047] The proposed mechanism may be used for licensed frequency and / or unlicensed frequency. In addition, the proposed mechanism of conditional configuration selection may be applied for the cases that a UE experiences a radio link failure when configured with conditional configurations.

[0048] System information (SI) may refer to MIB, SIB1, and other SI. Minimum SI may include MIB and SIB1. Other SI may refer to SIB3, SIB4, SIB5, and other SIB(s).

[0049] Dedicated signaling may refer to (but not limited to) RRC message(s). For example, RRC (Connection) Setup Request message, RRC (Connection) Setup message, RRC (Connection) Setup Complete message, RRC (Connection) Reconfiguration message, RRC Connection Reconfiguration message including the mobility control information, RRC Connection Reconfiguration message without the mobility control information inside, RRC Reconfiguration message including the configuration with sync, RRC Reconfiguration message without the configuration with sync inside, RRC (Connection) Reconfiguration Complete message, RRC (Connection) Resume Request message, RRC (Connection) Resume message, RRC (Connection) Resume Complete message, RRC (Connection) Reestablishment Request message, RRC (Connection) Reestablishment message, RRC (Connection) Reestablishment Complete message, RRC (Connection) Reject message, RRC (Connection) Release message, RRC System Information Request message, UE Assistance Information message, UE Capability Enquiry message, and UE Capability Information message.

[0050] The UE may include the RRC_CONNECTED UE, RRC_INACTIVE UE, and RRC_IDLE UE.

[0051] The current time in the present disclosure may be referred to as the time when the UE performs an inference from a model, the time when the UE performs a measurement on a reference signal, or the time when the UE generates a measurement report.

[0052] A serving cell may include a PCell, a PSCell, or an SCell. An spCell may include a PCell or a PSCell. A neighbor cell may include a cell other than the PCell and the PSCell. An SCell may also include a neighbor cell.

[0053] Artificial Intelligence (AI) / Machine Learning (ML) techniques have been increasingly considered for integration into new radio (NR) systems by the third generation partnership project (3GPP) in recent years. By collecting a sufficient amount of data, an AI / ML model and / or functionality can be trained to perform reliable prediction tasks that support efficient air interface configuration. AI / ML-based enhancements are currently being investigated for a variety of NR procedures. These include, but are not limited to, beam management, positioning accuracy enhancement, channel state information (CSI) prediction, CSI reporting compression, and mobility management. By leveraging AI / ML, these procedures may achieve improved performance, reduced signaling overhead, and enhanced adaptability to dynamic radio environments.

[0054] In the present disclosure, the focus is placed on mobility-related use cases in which AI / ML models are utilized to enhance mobility performance. AI / ML models may enable improvements in several aspects of mobility. For instance, by predicting future radio resource management (RRM) measurement results, the AI / ML models may reduce the likelihood of radio link failure (RLF). In another instance, the AI / ML models may reduce latency in the handover procedure by allowing the user equipment (UE) or network to initiate handover preparation in advance based on predicted measurement events (MEs). Additionally, prediction-based mechanisms may reduce the measurement burden on the UE by minimizing the need for continuous reference-signal monitoring.

[0055] To achieve mobility enhancement (e.g., reducing measurement overhead and / or lowering RLF occurrence probability), the AI / ML models may be used to predict future variations in RRM measurements, the potential occurrence of RLF, and / or the MEs. However, when AI / ML inference is performed at the UE, existing NR specifications do not provide a clear mechanism for conveying the inference results to the network. Specifically, the current RRM framework for Layer 3 (L3) measurements does not support the reporting of predictive results associated with future time instances. Furthermore, no existing signaling message or information element (IE) is defined for reporting predicted RLF or predicted MEs. Accordingly, the present disclosure provides a reporting framework enabling the UE to indicate to the network the predicted RRM measurement results, the predicted occurrence of RLF, and the predicted occurrence of MEs.

[0056] The present disclosure focuses on prediction conducted in the temporal domain; that is, predicting future measurement results based on current and / or historical measurement results. A model may be referred to as an AI / ML model in the present disclosure. The model may be applied at either the UE side or the network side (e.g., at a gNB or at the core network), and may be operated (e.g., be used for inference and be managed via monitoring and training / retraining) at the UE side or at the network side. A functionality may be referred to as an AI / ML functionality in the present disclosure.

[0057] In some implementations, from the perspective of the communication system, a functionality may be used to represent one or more configurations. For example, the one or more configurations may include a radio resource control (RRC) configuration. For AI / ML in mobility use cases, the RRC configuration may be related to the radio resource management (RRM) such as the measurement configuration (e.g., the measObject IE), the report configuration (e.g., the reportConfig IE), or both (e.g., the measId IE). In addition, the RRC configuration related to RRM may be enhanced to accommodate the AI / ML-related configuration such that a UE supporting the AI / ML-enabled features may be allowed to perform the AI / ML-related behaviors according to the AI / ML-related configuration. For another example, the one or more configurations may be an inference configuration corresponding to mobility / handover. In some implementations, a functionality may be used to represent a special ID (e.g., an associated ID, a global cell ID, and / or an area ID), a NW-side additional condition, and / or a UE-side additional condition. In some implementations, the associated ID may be used for consistency association for the RRM whereas same associated ID may imply the training results of RRM on a specific timing, or a specific area (e.g., either cell or tracking area), or a specific environment may be applied for current inference configuration.

[0058] From the perspective of inference, a functionality may be regarded as a set of models. When operating AI / ML with a functionality, the UE and / or the network may select different models associated with the functionality (e.g., based on UE implementation and / or NW implementation).

[0059] An additional condition (AC) may be referred to a situation. The situation may be related to the parameters / configurations applied at the network side or at the UE side. The situation may also be related to the communication status (e.g., the congestion occurrence, the mobility of the UE). Additional conditions may be indicated from the UE to the network or from the network to the UE, in order to assist the network and / or the UE to select a proper model for the corresponding situation.

[0060] An associated ID may be referred to an ID guaranteeing the consistency between model training and model inference. In some implementations, during data collection for the model training, the collected data may be categorized by the associated ID associated with the configuration which is used for the data collection. In some implementations, the network / UE may use the associated ID to determine which models / functionality to be applied for inference. In some implementations, an associated ID may represent a combination of one or more ACs.

[0061] For RRM prediction use cases, a UE may measure the reference signal and derive the channel quality (e.g., the reference signal received power (RSRP), the reference signal received quality (RSRQ), and / or the signal to interference plus noise ratio (SINR)). The UE may obtain a beam-level quality by measuring a single reference signal (e.g., an SSB or a CSI-RS). The UE may then obtain a cell-level quality by consolidating one or more beam-level qualities. The UE may use the measured qualities (e.g., beam-level qualities and / or cell-level qualities) to predict the channel qualities in future time instances. To achieve this, there are at least three methods (i)-(iii) as follows.

[0062] (i) Consolidation of predicted beam-level qualities: The UE may obtain a set of beam-level qualities in a series of time instances by measurement or a set of beam-level qualities in a time instance by measurement. The UE may use a model by taking the beam-level qualities as inputs to generate a set of beam-level qualities in a series of future time instances. The UE may then obtain a set of cell-level qualities in a series of future time instances by consolidating the beam-level qualities in the corresponding future time instances. The beam-level qualities and the cell-level qualities may be associated with the same cell (e.g., the serving cell or the neighboring cell).

[0063] (ii) Prediction by cell-level qualities: In some implementations, the UE may obtain a set of beam-level qualities in a series of time instances by measurement, and the UE may then obtain a set of cell-level qualities in the series of time instances by consolidating the beam-level qualities in the corresponding time instances. Then, the UE may use a model by taking the cell-level qualities as inputs to generate a set of cell-level qualities in a series of future time instances. For example, the beam-level qualities, the cell qualities and the generated cell qualities may be associated with the same cell (e.g., the serving cell or the neighboring cell).

[0064] In some implementations, the UE may obtain a set of beam-level qualities in a time instance by measurement, and the UE may then obtain a cell-level quality in that time instance by consolidating the beam-level qualities in that time instance. Then, the UE may use a model by taking the cell-level quality and / or some parameters (e.g., the probability of cell quality becomes better / worse, the probability of switching to this cell, and / or the confidence of measurement results) as inputs to generate a set of cell-level qualities in a series of future time instances.

[0065] (iii) Direct prediction by beam-level qualities: The UE may obtain a set of beam-level qualities in a series of time instances by measurement, or the UE may obtain a set of beam-level qualities in a time instance by measurement. The UE may use a model by taking the beam-level qualities as inputs to generate a set of cell-level qualities in a series of future time instances or a set of cell-level qualities of the current time instance. The beam-level qualities and the cell-level qualities may be associated with the same cell (e.g., the serving cell or the neighboring cell).

[0066] For ME prediction use cases, the UE may use the measured qualities to predict the occurrence of MEs in future time instances. To achieve this, there are at least two methods (i) and (ii) as follows.

[0067] (i) Direct prediction: The UE may obtain a set of beam-level or cell-level qualities in a series of time instances by measurement, or the UE may obtain a set of beam-level qualities or a cell-level quality in a time instance by measurement. The UE may use a model by taking the beam-level or cell-level qualities as inputs to generate a probability that an ME will occur in a future time period. The duration of future time period may be configured by the NW for the UE-side model or the NW-side model. The duration of future time period may be up to implementation and the UE may indicate the probability and the duration upon its inference reporting under the UE-side model. The performance monitoring may be performed based on the same duration of time period.

[0068] (ii) Indirect prediction: The UE may obtain a set of beam-level or cell-level qualities in a series of time instances by measurement, or the UE may obtain a set of beam-level qualities or a cell-level quality in a time instance by measurement. The UE may use a model by taking the beam-level or cell-level qualities as inputs to generate a series of cell-level qualities in a series of future time instances. The UE may then decide whether an ME will occur at a future time instance or in a future time period according to the generated cell-level qualities in a series of time instances. The performance monitoring may be performed based on the future time instance or in the duration of the future time period.

[0069] For RLF prediction use cases, the UE may use the measured qualities to predict the occurrence of RLF in future time instances. To achieve this, there are at least two methods (i) and (ii) as follows.

[0070] (i) Direct prediction: The UE may obtain a set of beam-level or cell-level qualities in a series of time instances by measurement, or the UE may obtain a set of beam-level qualities or a cell-level quality in a time instance by measurement, and the UE may use a model by taking the beam-level or cell-level qualities as inputs to generate a probability that an RLF will occur in a future time period. The duration of future time period may be configured by the NW for the UE-side model or the NW-side model. The duration of future time period may be up to implementation and the UE may indicate the probability and the duration upon its inference reporting under the UE-side model. The performance monitoring may be performed based on the same duration of time period.

[0071] (ii) Indirect prediction: The UE may obtain a set of beam-level or cell-level qualities in a series of time instances by measurement, or the UE may obtain a set of beam-level qualities or a cell-level quality in a time instance by measurement, and the UE may use a model by taking the beam-level or cell-level qualities as inputs to generate a series of cell-level qualities in a series of future time instances. The UE may then decide whether an RLF will occur at a future time instance or in a future time period according to the generated cell-level qualities in a series of time instances. The performance monitoring may be performed based on the future time instance or in the duration of the future time period.

[0072] The RRM measurement results (e.g., the beam-level and the cell-level measurement results) which are taken as model inputs may be obtained from the measurement on reference signals or from the output of a model in a previous inference.

[0073] In addition to temporal domain prediction, spatial domain and / or frequency domain prediction enabled by AI / ML techniques may also be leveraged to facilitate operations of a cellular system (e.g., NR or technologies beyond NR). In the present disclosure, the focus is placed on measurement reporting in the temporal domain; however, the embodiments described in the present disclosure may likewise be applied to the frequency domain and / or the spatial domain.

[0074] An observation window (OW) may be referred to as a duration in which the obtained measurement results may be taken as inputs of a model.

[0075] In some implementations, the OW may be defined by a time length and an ending point, and the starting point of the OW may be the ending point minus the time length. In some implementations, the OW may be defined by a time length and a starting point, and the ending point of the OW may be the starting point plus the time length. In some implementations, the OW may be defined by a starting point and an ending point, and the length of the OW may be equal to the ending point minus the starting point. In some implementations, the ending point of the OW may be the current time (e.g., the time when the UE performs (or starts to perform) the inference with a model). In some implementations, the ending point of the OW may be the time that the latest measurement result is obtained.

[0076] In some implementations, the starting point of the OW may be the time that the N-th measurement result prior to the latest measurement result is obtained. In some implementations, the value N may be dependent on the UE’s capability, and the UE may transmit the value N to the network (e.g., via RRC signaling such as the UECapabilityInformation message). In some implementations, the network may configure the value N to the UE via RRC signaling (e.g., the RRCReconfiguration message).

[0077] In some implementations, the time length of the OW may be dependent on the UE’s capability, and the UE may transmit the time length to the network (e.g., via RRC signaling such as the UECapabilityInformation message). In some implementations, the network may configure the time length to the UE via RRC signaling (e.g., the RRCReconfiguration message). In some implementations, the time length may be a fixed or preconfigured value (e.g., by default).

[0078] In some implementations, the time length of the OW may be defined by a value and a unit. In some implementations, the time length of the OW may be defined by the duration of a number of consecutive measurement results. In some implementations, the time length of the OW may be defined by an integer multiplicity of a given interval (e.g., the interval of measurement).

[0079] In some implementations, the time length of the OW may vary with the UE mobility speed, frequency band (e.g., intra-frequency / inter-frequency). For instance, based on the cellEdgeEvalaution IE and / or the LowMobilityEvalaution IE, the time length of the OW may be multiplied by different factors. In some implementations, the factors may be a predefined value known to the UE. In some implementations, the factors may be provided to the UE via common signaling (e.g., a SIB). In some implementations, the time length of the OW may reuse current Tmeasure,x, Tdetect,x, and / or Tevaluate,xwith corresponding DRX cycle. The time unit may include system frame number (SFN), UTC time, subframe, slot, symbol, second, and millisecond, but not limited to.

[0080] A prediction window (PW) may be referred to as a duration which may include the corresponding time instances of the predicted measurement results as outputs of a model.

[0081] In some implementations, the PW may be defined by a time length and an ending point, and the starting point of the PW may be the ending point minus the time length. In some implementations, the PW may be defined by a time length and a starting point, and the ending point of the PW may be the starting point plus the time length. In some implementations, the PW may be defined by a starting point and an ending point, and the length of the PW may be equal to the ending point minus the starting point. In some implementations, the starting point of the PW may be the current time (e.g., the time when the UE performs (starts to perform) the inference with a model). In some implementations, the starting point of the PW may be the time that the latest measurement result is obtained. In some implementations, the starting point of the PW may be the time that the next measurement result is expected to be obtained. In some implementations, the starting point of the PW may be the time that the UE activates a functionality (e.g., activates a configuration associated with the AI / ML for mobility).

[0082] In some implementations, the ending point of the PW may be the time that the N-th future measurement result from the latest measurement result is obtained. In some implementations, the value N may be dependent on the UE’s capability, and the UE may transmit the value N to the network (e.g., via RRC signaling such as the UECapabilityInformation message). In some implementations, the network may configure the value N to the UE via RRC signaling (e.g., the RRCReconfiguration message).

[0083] In some implementations, the time length of the PW may be dependent on the UE’s capability, and the UE may transmit the time length to the network (e.g., via RRC signaling such as the UECapabilityInformation message). In some implementations, the network may configure the time length to the UE via RRC signaling (e.g., the RRCReconfiguration message).

[0084] In some implementations, the time length of the PW may be defined by a value and a unit. In some implementations, the time length of the PW may be defined by the duration of a number of consecutive measurement results. In some implementations, the time length of the PW may be defined by an integer multiplicity of a given interval (e.g., the interval of measurement).

[0085] In some implementations, the time length of the PW may vary with the UE mobility speed, frequency band (e.g., intra-frequency / inter-frequency). For instance, based on the cellEdgeEvalaution IE and / or the LowMobilityEvalaution IE, the time length of the OW may be multiplied by different factors. In some implementations, the time length of the PW may reuse current Tmeasure,x, Tdetect,x, and / or Tevaluate,xwith corresponding DRX cycle.

[0086] In some implementations, the time length of the PW may be different from the time length of the OW. In some implementations, the time length of the PW may be a fixed value and the time length of the OW may be a variable value based on the associated scenarios / factors. The time unit may be system frame number (SFN), UTC time, subframe, slot, and / or symbol, but not limited to.

[0087] Report of Predicted RRM Measurement

[0088] In some implementations, the UE may obtain a number (e.g., M) of predicted RRM measurement results, each of which may correspond to a future time instance, from the outputs of a model. The UE may report one or more or all of the predicted RRM measurement results to the network via RRC signaling (e.g., the MeasurementReport message). In some implementations, the value M may be dependent on the UE’s capability (e.g., based on the UE’s functionality and / or the UE’s model for inference). In some implementations, the value M may be dependent on the length of the PW. In some implementations, the value M may be configured by the NW via RRC signaling, MAC CE, and / or DCI. In some implementations, the value M may be a fixed value or a predefined / preconfigured value.

[0089] In some implementations, the UE may report a number of predicted RRM measurement results corresponding to the time instances indicated by the network. In some implementations, the network may indicate the information of the time instances via RRC signaling (e.g., the RRCReconfiguration message) with an IE related to the reporting configuration (e.g., the reportConfig IE).

[0090] In some implementations, the network may indicate a number of offset values, each of which may correspond to a future time instance with the current time as the reference point. In response to the indication, the UE may report the predicted RRM measurement results that correspond to the time instances nearest / closest to the future time instances determined by the offset values. The nearest future time instances may be valid for reporting if it is within the prediction window. In some implementations, the nearest future time instances may be valid for reporting even it is outside the prediction window.

[0091] In some implementations, the offset value may be an integer indicating the predicted RRM measurement results in time order, starting from the current time. For example, if offset values 3, 5, and / or 7 are indicated, the UE may report the third, the fifth, and / or the seventh predicted RRM measurement results. In some implementations, the offset value may be an integer indicating the future time instance starting from the current time. The unit of the offset may be predetermined (e.g., 1 ms, 10 ms, 20 ms, 50 ms, etc.). For example, if offset values 3, 5, and 7 are indicated with the unit of 50ms, the UE may report the predicted RRM measurement results closest to the time instances of 150 ms, 250 ms, and 350 ms from the current time.

[0092] In some implementations, the offset value may be an ENUMERATED value directly indicating the value and the unit of the time offset. For example, if offset values ‘ms500’, ‘ms1500’, and ‘ms2500’ are indicated, the UE may report the predicted RRM measurement results closest to the time instances of 500 ms, 1500 ms, and 2500 ms from the current time. In some implementations, the network may indicate a number of offset values in the format of a sequence of integers, where each integer may correspond to an offset value. In some implementations, the network may indicate a number of offset values in the format of a sequence of ENUMERATED values, where each ENUMERATED value may correspond to an offset value.

[0093] In some implementations, the network may indicate an offset value that corresponds to a future time instance with the current time as the reference point, and a periodicity value. The periodicity value may be configured by the NW or be applied with concurrent DRX cycle. In response to the indication, the UE may obtain an arithmetic sequence by taking the offset value as the first term and the periodicity as the common difference. The UE may report the predicted RRM measurement results that correspond to the time instances closest to the future time instances corresponding to the arithmetic sequence.

[0094] In some implementations, the offset value may be an integer indicating the future time instance starting from the current time. The unit of the offset may be predetermined (e.g., 1 ms, 10 ms, 20 ms, 50 ms, etc.). For example, if offset value 5 is indicated with the unit of 50ms, the UE may consider the first future time instance to be 250 ms from the current time. In some implementations, the offset value may be an ENUMERATED value directly indicating the value and the unit of the time offset. For example, if offset value ‘ms500’ is indicated, the UE may consider the first future time instance to be 500 ms from the current time. In some implementations, the offset value may be absent, and the UE may consider that the first future time instance to be the current time or to be the current time plus the periodicity value.

[0095] In some implementations, the periodicity value may be an integer with predetermined unit (e.g., 1 ms, 10 ms, 20 ms, 50 ms, etc.). For example, if the periodicity value 3 is indicated with the unit of 50 ms, the UE may consider the time difference between consecutive indicated future time instances to be 150ms. In some implementations, the periodicity value may be an ENUMERATED value directly indicating the value and the unit of the periodicity. For example, if the periodicity value ‘ms150’ is indicated, the UE may consider the time difference between consecutive indicated future time instances to be 150 ms. In some implementations, the network may further indicate an end point of the time instance to be reported. In response to the indication of the end point, the UE may report the predicted RRM measurement results corresponding to the time instances that are before the end point.

[0096] In some implementations, the UE may not include in the measurement report the predicted RRM measurement results corresponding to the time instances which are after the end point (e.g., the UE may only include in the measurement report the predicted RRM measurement results corresponding to the time instances which are earlier than and / or equal to the end point). In some implementations, the end point may be an offset with current time as the reference point. For example, the UE may consider the end point to be the current time plus the offset value. In some implementations, the end point may be an offset with the first indicated time instance as the reference point. For example, the UE may consider the end point to be the first indicated time instance plus the offset value. In some implementations, the end point may be an integer (e.g., L) indicating the number of future time instances. For example, the UE may consider the end point to be the time instance corresponding to the L-th predicted measurement result starting from the current time.

[0097] In some implementations, if the confidence level of the predicted RRM measurement result corresponding to an indicated future time instance is low, for the reporting entry corresponding to the time instance, the UE may report a value that indicates that the RRM measurement result is not available. More specifically, if the confidence level of the predicted RRM measurement result corresponding to an indicated future time instance is lower than a threshold, for the reporting entry corresponding to the time instance, the UE may report a value that indicates that the RRM measurement result is not available. In some implementations, if the confidence levels of the predicted RRM measurement results corresponding to all indicated future time instances are lower than a threshold, the UE may skip the measurement report, and the network not receiving the measurement report may consider that the RRM measurements corresponding to the report may be unavailable. In some implementations, the threshold may be configured, along with the information of indicated future time instances, by the network (e.g., via RRC signaling).

[0098] In some implementations, the UE may select a number (e.g., N) of predicted RRM measurement results from the M predicted RRM measurement results and report the selected predicted RRM measurement results to the network. In some implementations, the UE may perform the selection according to the indication from the network. In some implementations, the network may configure the indication via RRC signaling (e.g., the RRCReconfiguration message) with an IE related to the reporting configuration (e.g., the reportConfig IE).

[0099] In some implementations, the UE may report the predicted RRM measurement results with the first to the N-th highest confidence levels. In some implementations, the value N may be configured by the network. In some implementations, the value M and / or N may be configured by the network or depend on the observation window configured by network or implemented by UE. In some implementations, the value M and / or N may be configured by the NW via RRC signaling, MAC CE, and / or DCI. In some implementations, the value M and / or N may be a fixed value or a predefined / preconfigured value.

[0100] In some implementations, the UE may report the predicted RRM measurement results whose confidence levels are higher than a threshold. In some implementations, the threshold may be configured by the network, via RRC signaling. In some implementations, the value N and the threshold may be preconfigured or a predefined value.

[0101] In some implementations, the UE may check the number of predicted RRM measurement results whose confidence levels are higher than a threshold. The UE may report the predicted RRM measurement results with the first to the N-th highest confidence levels if the number is larger than or equal to N. The UE may report all the predicted RRM measurement results whose confidence levels are higher than the threshold if the number is less than N. In some implementations, the value N and the threshold may be configured by the network (e.g., via RRC signaling). In some implementations, the value N and the threshold may be preconfigured or a predefined value.

[0102] In some implementations, the UE may report the predicted RRM measurement results with the first to the N-th lowest RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value N may be configured by the network. More specifically, the UE may report the predicted RRM measurement results with the first to the N-th lowest RRM measurement quantities if the measured cell is a SpCell (e.g., a PCell or a PSCell). In some implementations, the UE may report the predicted RRM measurement results with the first to the N-th highest RRM measurement quantities. In some implementations, it is configurable whether highest or lowest measurement results be reported.

[0103] In some implementations, the UE may report the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are lower than a threshold. In some implementations, the threshold may be configured by the network. More specifically, the UE may report the predicted RRM measurement results whose RRM measurement quantities are lower than a threshold if the measured cell is a SpCell (e.g., a PCell or a PSCell).

[0104] In some implementations, the UE may check the number of predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are lower than a threshold. The UE may report the predicted RRM measurement results with the first to the N-th lowest RRM measurement quantities (e.g., RSRP, RSRQ, SINR) if the number is larger than or equal to N. The UE may report all the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are lower than the threshold if the number is less than N. In some implementations, the value N and the threshold may be configured by the network. More specifically, if the measured cell is a SpCell (e.g., a PCell or a PSCell), the UE may report (i) the predicted RRM measurement results with the first to the N-th lowest RRM measurement quantities if the number is larger than or equal to N and (ii) all the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are lower than the threshold if the number is less than N.

[0105] In some implementations, the UE may report the predicted RRM measurement results with the first to the N-th highest RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value N may be configured by the network. More specifically, the UE may report the predicted RRM measurement results with the first to the N-th highest RRM measurement quantities if the measured cell is a cell other than the SpCell (e.g., a PCell or a PSCell).

[0106] In some implementations, the UE may report the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are higher than a threshold. In some implementations, the threshold may be configured by the network. More specifically, the UE may report the predicted RRM measurement results whose RRM measurement quantities are higher than a threshold if the measured cell is a cell other than the SpCell (e.g., a PCell or a PSCell).

[0107] In some implementations, the UE may check the number of predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are higher than a threshold. The UE may report the predicted RRM measurement results with the first to the N-th highest RRM measurement quantities (e.g., RSRP, RSRQ, SINR) if the number is larger than or equal to N. The UE may report all the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are higher than the threshold if the number is less than N. In some implementations, the value N and the threshold may be configured by the network. More specifically, if the measured cell is a cell other than the SpCell (e.g., a PCell or a PSCell), the UE may report (i) the predicted RRM measurement results with the first to the N-th highest RRM measurement quantities if the number is larger than or equal to N and (ii) all the predicted RRM measurement results whose RRM measurement quantities (e.g., RSRP, RSRQ, SINR) are higher than the threshold if the number is less than N.

[0108] FIG. 1 is a diagram illustrating predicted RRM measurement results corresponding to nearest and farthest future time instances relative to a current time, according to an example implementation of the present disclosure. In some implementations, the UE may report the predicted RRM measurement results corresponding to the N nearest future time instances relative to the current time (e.g., the group 102 as illustrated in FIG. 1), where the value N may be configured by the network (e.g., N=4 as illustrated in FIG. 1). The group 102 may include the first to the N-th predicted RRM measurement results that correspond to the N nearest future time instances relative to the current time. In some implementations, the UE may report the predicted RRM measurement results corresponding to the N farthest future time instances relative to the current time (e.g., the group 104 as illustrated in FIG. 1), where the value N may be configured by the network (e.g., N=4 as illustrated in FIG. 1). The group 104 may include the first to the N-th predicted RRM measurement results that correspond to the N farthest future time instances relative to the current time.

[0109] In some implementations, from the predicted RRM measurement result corresponding to the farthest future time instance to the predicted RRM measurement result corresponding to the nearest future time instance relative to the current time, the UE may include a predicted RRM measurement result in the measurement report if (i) the confidence level of the predicted RRM measurement result is higher than a first threshold and (ii) the number of included predicted RRM measurement results is less than a second threshold. In some implementations, the first threshold and the second threshold may be configured by the network.

[0110] In some implementations, from the predicted RRM measurement result corresponding to the nearest future time instance to the predicted RRM measurement result corresponding to the farthest future time instance relative to the current time, the UE may include a predicted RRM measurement result in the measurement report if (i) the confidence level of the predicted RRM measurement result is higher than a first threshold and (ii) the number of included predicted RRM measurement results is less than a second threshold. In some implementations, the first threshold and the second threshold may be configured by the network.

[0111] In some implementations, the UE may select a number (e.g., N) of predicted RRM measurement results from the M predicted RRM measurement results and report the selected predicted RRM measurement results to the network. In addition, how to select the predicted RRM measurement results to report to the network may be based on UE implementation.

[0112] In some implementations, the UE may include multiple entries in the measurement report, where each entry may include an RRM measurement quantity (e.g., RSRP, RSRQ, SINR), the confidence level corresponding to the RRM measurement quantity, and the timestamp corresponding to the RRM measurement quantity.

[0113] In some implementations, the confidence level may be absent (e.g., the UE may not include the confidence level) if the network has configured a confidence level threshold for the selection of RRM measurement results to be reported. In some implementations, the timestamp may be absent (e.g., the UE may not include the timestamp) if the network has configured the indicated future time instances to report. In some implementations, the UE may include a value indicating that the RRM measurement quantity is invalid or unavailable (e.g., due to low confidence level) for an entry corresponding to the timestamp. In some implementations, the reporting may be triggered by the NW or by specific event or be performed based on the configured periodicity.

[0114] In some implementations, the UE may have the predicted RRM measurement results associated with multiple cells. The selection method may apply to the RRM measurement results associated with each cell. The selection method may also apply to the RRM measurement results associated with the multiple cells.

[0115] For example, the UE may report the RRM measurement results corresponding to the first to the N-th highest confidence levels among all the RRM measurement results associated with the multiple cells. In some implementations, the value N may be configured by the network. In some implementations, the value N may be varied based on UE speed and / or cell edge conditions.

[0116] For example, the UE may report the RRM measurement results corresponding to the first to the N-th lowest RRM measurement quantities among all the RRM measurement results associated with the multiple cells. In some implementations, the value N may be configured by the network. In some implementations, the value N may be varied based on UE speed and / or cell edge conditions.

[0117] For example, the UE may report the RRM measurement results corresponding to the first to the N-th highest RRM measurement quantities among all the RRM measurement results associated with the multiple cells. In some implementations, the value N may be configured by the network. In some implementations, the value N may be varied based on UE speed and / or cell edge conditions.

[0118] For example, the UE may report the RRM measurement results corresponding to the future time instances that are the N nearest to the current time among all the RRM measurement results associated with multiple cells. In some implementations, the value N may be configured by the network. In some implementations, the value N may be varied based on UE speed and / or cell edge conditions.

[0119] For example, the UE may report the RRM measurement results corresponding to the future time instances that are the N farthest to the current time among all the RRM measurement results associated with the multiple cells. In some implementations, the value N may be configured by the network. In some implementations, the value N may be varied based on UE speed and / or cell edge conditions.

[0120] For example, upon different time instance, the UE may report the RRM measurement results associated with multiple cells if the result is above / below a configured threshold. One implementation is several events with corresponding threshold / timer may be configured, and the UE may report the RRM measurement results associated with the multiple cells while event is satisfied.

[0121] Report of Predicted Measurement Events

[0122] In some implementations, the network may configure event-triggered measurement report to the UE. The triggering events may include the following (a)-(f).

[0123] (a) A1 event: The measurement quality of the serving cell becomes better than a threshold.

[0124] (b) A2 event: The measurement quality of the serving cell becomes worse than a threshold.

[0125] (c) A3 event: The measurement quality of a neighbor cell becomes offset better than the measurement quality of the PCell / PSCell.

[0126] (d) A4 event: The measurement quality of a neighbor cell becomes better than a threshold.

[0127] (e) A5 event: The measurement quality of the serving cell becomes worse than a first threshold and the measurement quality of a neighbor cell becomes better than a second threshold.

[0128] (f) A6 event: The measurement quality of a neighbor cell becomes offset better than the measurement quality of a SCell.

[0129] In some implementations, the UE may predict one or more ME occurrences by predicting the probability of the ME occurrence in a time window. More specifically, the one or more predicted MEs may be a subset of MEs configured by the network. In some implementations, the UE may initiate an event-triggered measurement report when the probability is higher than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may initiate an event-triggered measurement report when the time window is included in a period. In some implementations, the period may be configured by the network.

[0130] In some implementations, the period may be configured by a timing offset indicating the starting point and a timing offset indicating the end point. In some implementations, the UE may take the current time as the reference point for the offsets. In some implementations, the period may be configured by a timing offset indicating the starting point and a duration indicating the length of the period. In some implementations, the UE may take the current time as the reference point for the offset, and the UE may consider the end point to be the starting point plus the duration. In some implementations, the period may be configured by a timing offset indicating the end point and a duration indicating the length of the period. In some implementations, the UE may take the current time as the reference point for the offset, and the UE may consider the starting point to be the end point minus the duration.

[0131] In some implementations, the UE may initiate an event-triggered measurement report when the probability is higher than a threshold and the time window is included in a period. In some implementations, the threshold and the period may be configured by the network.

[0132] In some implementations, the period may be configured by a timing offset indicating the starting point and a timing offset indicating the end point. In some implementations, the UE may take the current time as the reference point for the offsets. In some implementations, the period may be configured by a timing offset indicating the starting point and a duration indicating the length of the period. In some implementations, the UE may take the current time as the reference point for the offset, and the UE may consider the end point to be the starting point plus the duration. In some implementations, the period may be configured by a timing offset indicating the end point and a duration indicating the length of the period. In some implementations, the UE may take the current time as the reference point for the offset, and the UE may consider the starting point to be the end point minus the duration.

[0133] In some implementations, the UE may initiate an event-triggered measurement report when a conditional handover, an LTM, or a conditional LTM is configured. In some implementations, the UE may initiate an event-triggered measurement report when a BFR occurs.

[0134] In some implementations, the UE may include the event type (e.g., which configured event(s) is triggered) in the event-triggered measurement report. More specifically, the UE may include the event type in the event-triggered measurement report if the network configures more than one event. More specifically, the UE may not include the event type in the event-triggered measurement report if the network only configures a single event.

[0135] In some implementations, the UE may include the timing information associated with the triggered event in the measurement report. More specifically, the UE may indicate the window during which the ME is predicted to occur with a probability.

[0136] FIG. 2 is a diagram illustrating characterizations of a window based on offset and duration, according to an example implementation of the present disclosure. In some implementations, the UE may report the window with an offset (e.g., Offset 1 of Opt1 as illustrated in FIG. 2) indicating the starting point of the window and an offset (e.g., Offset 2 of Opt1 as illustrated in FIG. 2) indicating the end point of the window. In some implementations, the current time may be taken as the reference point for the offsets. In some implementations, the offset indicating the starting point of the window may be an absolute time. The offset indicating the end point of the window may be an absolute time. In some implementations, the offset indicating the starting point of the window may be a time length (e.g., relative time to the current time). The offset indicating the end point of the window may be a time length (e.g., relative time to the current time or relative time to the starting point of the window).

[0137] In some implementations, the UE may report the window with an offset (e.g., Offset 1 of Opt2 as illustrated in FIG. 2) indicating the starting point of the window and a duration (e.g., Duration 202 of Opt2 as illustrated in FIG. 2) indicating the length of the window. In some implementations, the current time may be taken as the reference point for the offset. In some implementations, the offset indicating the starting point of the window may be an absolute time.

[0138] In some implementations, the UE may report the window with an offset (e.g., Offset 1 of Opt3 as illustrated in FIG. 2) indicating the central point of the window and a duration (e.g., Duration 204 of Opt3 as illustrated in FIG. 2) indicating the half-length of the window (e.g., the window starts at the offset minus the duration and ends at the offset plus the duration). In some implementations, the current time may be taken as the reference point for the offset.

[0139] In some implementations, the network may limit the length of the reported window by configuring the UE with a maximum window length. In some implementations, if the UE is configured with a maximum window length, the UE may report a window with length shorter than or equal to the maximum window length. In some implementations, the network may limit the number of the reported instances by configuring the UE with a maximum instance number. When the UE stored more than maximum instance number predicted report, the UE may report the prediction with number smaller than or equal to the maximum instance number.

[0140] In some implementations, the UE may include the probability of the ME occurrence during the window in the measurement report. In some implementations, the UE may round the probability to a quantized value and report the quantized value with an ENUMERATED format. For example, if the probability obtained from the model is 0.82, the UE may round the probability and report with an ENUMERATED format ‘.8’. If the probability obtained from the model is 0.89, the UE may round the probability and report with an ENUMERATED format ‘.9’. In some implementations, the UE may use an indication indicating whether the probability is high or low. In some implementations, the indication may be a Boolean format with a first value (e.g., 1) indicating that an ME will occur in the window with high probability, and the indication may be a Boolean format with a second value (e.g., 0) indicating that an ME will occur in the window with low probability. In some implementations, the indication may be an ENUMERATED format with a first value (e.g., ‘high’) indicating that an ME will occur in the window with high probability, and the indication may be an ENUMERATED format with a second value (e.g., ‘low’) indicating that an ME will occur in the window with low probability. Whether the probability is high or low may be determined based on the UE’s implementation.

[0141] In some implementations, the UE may include the measured RRM results in the measurement report. More specifically, the network may indicate to the UE whether to include the measured RRM results in the measurement report. For example, the UE may include the measured RRM results in the measurement report if the indication is a specific value (e.g., ‘true’).

[0142] In some implementations, the UE may predict one or more MEs based on the predicted RRM measurement results in future time instances. In some implementations, the one or more MEs may be a subset of MEs configured by the network.

[0143] In some implementations, the UE may initiate an event-triggered measurement report when the predicted RRM measurement results (e.g., of the serving cell and / or of the neighbor cells) satisfy the event (e.g., the A1 to A6 events) for a period called time-to-trigger (TTT) at a future time instance.

[0144] In some implementations, if an event has been satisfied and reported, the UE may initiate an event-triggered measurement report when the predicted RRM measurement results satisfy the leaving condition for a TTT at a future time instance. More specifically, the UE may initiate an event-triggered measurement report when the predicted RRM measurement results satisfy the leaving condition for a TTT at a future time instance if the network configures an indication (e.g., the ReportOnLeave IE set to ‘true’). In some implementations, the leaving condition for the counterpart of the events may include the following (a)-(f).

[0145] (a) A1 event (leaving): The measurement quality of the serving cell becomes worse than a second threshold.

[0146] (b) A2 event (leaving): The measurement quality of the serving cell becomes better than a second threshold.

[0147] (c) A3 event (leaving): The measurement quality of a neighbor cell becomes offset worse than the measurement quality of the PCell / PSCell.

[0148] (d) A4 event (leaving): The measurement quality of a neighbor cell becomes worse than a second threshold.

[0149] (e) A5 event (leaving): The measurement quality of the serving cell becomes better than a third threshold, and the measurement quality of a neighbor cell becomes worse than a fourth threshold.

[0150] (f) A6 event: The measurement quality of a neighbor cell becomes offset worse than the measurement quality of a SCell.

[0151] In some implementations, the UE may include the event type (e.g., which configured event(s) is triggered) in the event-triggered measurement report. More specifically, the UE may include the event type in the event-triggered measurement report if the network configures more than one event. More specifically, the UE may not include the event type in the event-triggered measurement report if the network only configures a single event. More specifically, the UE may include the event type in the event-triggered measurement report if the network configures an indication (e.g., with a specific value ‘true’) indicating the UE to include the event type.

[0152] In some implementations, similar events may be configured for actual RRM measurements, and the event reporting of actual RRM measurements may be performed in parallel with the event reporting of predicted RRM measurements. The parameters in actual event reporting and predicted event reporting may be separately configured by report configuration.

[0153] In some implementations, the UE may include the timing information of the predicted ME.

[0154] In some implementations, if a first predicted ME is reported (e.g., including the timing and the related RRM measurement results) via a first measurement report, upon a second predicted ME is predicted, the UE may initiate a second measurement report. In some implementations, the UE may include all or a partial of the information in the first measurement report, along with the information of the second predicted ME, in the second measurement report. In some implementations, the UE may include all or a partial of the information in the first measurement report, along with the information of the second predicted ME, in the second measurement report if the first predicted ME does not satisfy the leaving condition. In some implementations, the UE may include all or a partial of the information in the first measurement report, along with the information of the second predicted ME, in the second measurement report if the network configures an indication to the UE. More specifically, the indication may take ENUMERATED format. For example, if the indication is a first value (e.g., ‘timing’), the UE may include the timing information of the first predicted ME. For example, if the indication is a second value (e.g., ‘RRM result’), the UE may include the RRM result of the first predicted ME. For example, if the indication is a third value (e.g., ‘timing and RRM result’), the UE may include the timing information and the RRM result of the first predicted ME.

[0155] In some implementations, the UE may indicate the timing instance that the ME is predicted to occur. In some implementations, the UE may report the timing instance with an offset value (e.g., taking the current time as the reference point). In some implementations, the offset value may be an integer with predefined unit. For example, if the ME is predicted to occur at 2500 ms from the current time, and if the predefined unit is 500 ms, the UE may report the offset value as ‘5’. In some implementations, the UE may report the timing instance with an integer (e.g., N) indicating that the ME is predicted to occur between the N-th and the (N+1)-th reported time instances from the current time. In some implementations, the UE may report the timing instance with an integer (e.g., N) indicating that the ME is predicted to occur between the (N-1)-th and the N-th reported time instances from the current time.

[0156] In some implementations, the UE may indicate a window during which the ME is predicted to occur. In some implementations, the UE may report the window with an offset indicating the starting point of the window and an offset indicating the end point of the window. In some implementations, the current time may be taken as the reference point for the offsets. In some implementations, the UE may report the window with an offset indicating the starting point of the window and a duration indicating the length of the window. In some implementations, the current time may be taken as the reference point for the offset. In some implementations, the UE may report the window with an offset indicating the central point of the window and a duration indicating the half-length of the window (e.g., the window starts at the offset minus the duration and ends at the offset plus the duration). In some implementations, the current time may be taken as the reference point for the offset. In some implementations, the network may limit the length of the reported window by configuring the UE with a maximum window length. In some implementations, if the UE is configured with a maximum window length, the UE may report a window with length shorter than or equal to the maximum window length.

[0157] In some implementations, the UE may include the entries associated with the predicted RRM results in the measurement report. In some implementations, an entry may include a predicted RRM measurement quantity (e.g., RSRP, RSRQ, SINR), a confidence level corresponding to the predicted RRM measurement quantity, and a timestamp corresponding to the predicted RRM measurement quantity. In some implementations, if the UE reports the time instance of the ME occurrence, the UE may select a number of predicted RRM measurement results corresponding to the future time instances near the time instance of the ME occurrence. The UE may include the entries associated with the selected predicted RRM measurement results in the measurement report.

[0158] In some implementations, the UE may select the predicted RRM measurement result corresponding to the time instance of the ME occurrence. In some implementations, the UE may select a number (e.g., N) of predicted RRM measurement result corresponding to consecutive N future time instances which are nearest and before the time instance of the ME occurrence. In some implementations, the value N may be configured by the network. In some implementations, the UE may select a number (e.g., M) of predicted RRM measurement result corresponding to consecutive M future time instances which are nearest and after the time instance of the ME occurrence. In some implementations, the value M may be configured by the network. In some implementations, the UE may select a number (e.g., Z) of predicted RRM measurement results based on UE implementations. The value Z may be configured by the network.

[0159] In some implementations, if the UE reports the window of the ME occurrence, the UE may select a number of predicted RRM measurement results corresponding to the future time instances in the window. The UE may include the entries associated with the selected predicted RRM measurement results in the measurement report.

[0160] In some implementations, the UE may select a number (e.g., N) of predicted RRM measurement results with the first to the N-th highest confidence levels. In some implementations, the value N may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with confidence levels higher than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may select a number (e.g., M) of predicted RRM measurement results with the first to the M-th lowest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value M may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with the predicted RRM measurement quantities lower than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may select a number (e.g., L) of predicted RRM measurement results with the first to the L-th highest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value L may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with the predicted RRM measurement quantities higher than a threshold. In some implementations, the threshold may be configured by the network.

[0161] In some implementations, the UE may select a number (e.g., K) of predicted RRM measurement results with the first to the K-th highest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value K may be configured by the network. In some implementations, the UE may select a number (e.g., J) of predicted RRM measurement results corresponding to the future time instances which are the first to the J-th nearest to the central point / starting point / end point of the window. In some implementations, the value J may be configured by the network. In some implementations, the UE may select a number (e.g., Z) of predicted RRM measurement results based on UE implementations. The value Z may be configured by the network.

[0162] In some implementations, the UE may select the predicted RRM measurement results corresponding to the future time instances falling in the period that the ME’s event is considered to be satisfied. The UE may include the entries associated with the selected predicted RRM measurement results in the measurement report.

[0163] In some implementations, when selecting the predicting RRM measurement results, the UE may also consider the confidence level of the RRM measurement result. The UE may include the entry with a predicting RRM measurement result if the corresponding confidence level is higher than a threshold. The UE may not include the entry with a predicting RRM measurement result if the corresponding confidence level is lower than a threshold. In some implementations, the threshold may be configured by the network.

[0164] Report of Predicted RLF Occurrence

[0165] In some implementations, the UE may predict an RLF occurrence with a model. In some implementations, the UE may predict the probability of an RLF occurrence in a time window. In some implementations, the UE may consider that an RLF occurrence is predicted if the probability is higher than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may predict an RLF occurrence via the predicted RRM measurement results in future time instances. In some implementations, upon predicting the RLF occurrence, the UE may transmit signaling to the network indicating a predicted RLF occurrence. In some implementations, the signaling may include an RRC message, a MAC CE, or UCI.

[0166] In some implementations, the RRC message may be the RRCReestablishmentRequest message. In some implementations, the UE may set the ReestablishmentCause IE to a specific value (e.g., ‘PredictedFailure’ or ‘otherFailure’) to indicate a predicted RLF occurrence. In some implementations, the RRC message may be the MeasuermentReport message. In some implementations, the UE may include the timing information of the predicted RLF occurrence. In some implementations, the UE may indicate the timing instance that the RLF is predicted to occur. In some implementations, the UE may indicate a window during which the RLF is predicted to occur.

[0167] In some implementations, the UE may report the timing instance with an offset value (e.g., taking the current time as the reference point). In some implementations, the offset value may be an integer with predefined unit. For example, if the RLF is predicted to occur at 2500 ms from the current time, and if the predefined unit is 500 ms, the UE may report the offset value as ‘5’.

[0168] In some implementations, the UE may report the timing instance with an integer (e.g., N) indicating that the RLF is predicted to occur between the N-th and the (N+1)-th reported time instances from the current time.

[0169] In some implementations, the UE may report the timing instance with an integer (e.g., N) indicating that the RLF is predicted to occur between the (N-1)-th and the N-th reported time instances from the current time.

[0170] In some implementations, the UE may report the window with an offset indicating the starting point of the window and an offset indicating the end point of the window, as illustrated in Opt1 of FIG. 2. In some implementations, the current time may be taken as the reference point for the offsets.

[0171] In some implementations, the UE may report the window with an offset indicating the starting point of the window and a duration indicating the length of the window, as illustrated in Opt2 of FIG. 2. In some implementations, the current time may be taken as the reference point for the offset.

[0172] In some implementations, the UE may report the window with an offset indicating the central point of the window and a duration indicating the half-length of the window (e.g., the window starts at the offset minus the duration and ends at the offset plus the duration, as illustrated in Opt3 of FIG. 2). In some implementations, the current time may be taken as the reference point for the offset.

[0173] In some implementations, the network may limit the length of the reported window by configuring the UE with a maximum window length. In some implementations, if the UE is configured with a maximum window length, the UE may report a window with length shorter than or equal to the maximum window length.

[0174] In some implementations, the UE may include one or more entries in the RRC message, where each of which may be associated with the predicted RRM measurement result in a future time instance. In some implementations, an entry may include the predicted RRM measurement quantity (e.g., RSRP, RSRQ, SINR), the corresponding time instance, and the corresponding confidence level.

[0175] In some implementations, an entry may include a predicted RRM measurement quantity (e.g., RSRP, RSRQ, SINR), the corresponding time instance, and the corresponding confidence level. In some implementations, the UE may select a number (e.g., N) of predicted RRM measurement results with the first to the N-th highest confidence levels. In some implementations, the value N may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with confidence levels higher than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may select a number (e.g., M) of predicted RRM measurement results with the first to the M-th lowest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value M may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with the predicted RRM measurement quantities lower than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may select a number (e.g., L) of predicted RRM measurement results with the first to the L-th highest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value L may be configured by the network. In some implementations, the UE may select the predicted RRM measurement results with the predicted RRM measurement quantities higher than a threshold. In some implementations, the threshold may be configured by the network. In some implementations, the UE may select a number (e.g., K) of predicted RRM measurement results with the first to the K-th highest predicted RRM measurement quantities (e.g., RSRP, RSRQ, SINR). In some implementations, the value K may be configured by the network. In some implementations, the UE may select a number (e.g., J) of predicted RRM measurement results corresponding to the future time instances which are the first to the J-th nearest to the central point / starting point / end point of the window. In some implementations, the value J may be configured by the network. In some implementations, the UE may select a number (e.g., Z) of predicted RRM measurement results based on UE implementations. The value Z may be configured by the network.

[0176] In some implementations, the UE may include in the RRC message the entries of RRM measurement results if the network configures an indication indicating the UE to include the entries of RRM measurement results. For example, if the network configures an indication with a specific value (e.g., ‘true’), the UE may include the entries of RRM measurement results in the RRC message.

[0177] In some implementations, the UE may include in the RRC message the probability that an RLF will occur in the window.

[0178] In some implementations, the UE may round the probability to a quantized value and report the quantized value with an ENUMERATED format. For example, if the probability obtained from the model is 0.82, the UE may round the probability and report with an ENUMERATED format ‘.8’. If the probability obtained from the model is 0.89, the UE may round the probability and report with an ENUMERATED format ‘.9’.

[0179] In some implementations, the UE may use an indication indicating whether the probability is high or low. In some implementations, the indication may be a Boolean format with a first value (e.g., 1) indicating that an RLF will occur in the window with high probability, and the indication may be a Boolean format with a second value (e.g., 0) indicating that an RLF will occur in the window with low probability. In some implementations, the indication may be an ENUMERATED format with a first value (e.g., ‘high’) indicating that an RLF will occur in the window with high probability, and the indication may be an ENUMERATED format with a second value (e.g., ‘low’) indicating that an RLF will occur in the window with low probability. Whether the probability is high or low may be determined based on the UE’s implementation.

[0180] In some implementations, the MAC CE may include zero bits. The UE may use a specific logical channel ID value for the MAC subheader to indicate that an RLF occurrence is predicted. In some implementations, the MAC CE / UCI may include the window during which the RLF is predicted to occur and the probability that the RLF is predicted to occur in the window.

[0181] In some implementations, the window may be determined based on a field indicating the starting time of the window and a field indicating the duration of the window. In some implementations, the window may be determined based on a field indicating the starting time of the window and a field indicating the ending time of the window. In some implementations, the window may be determined based on a field indicating the ending time of the window and a field indicating the duration of the window. In some implementations, the probability may be determined based on a field whose codepoint bits are mapped to a specific value of probability. In some implementations, the probability may be determined based on a field with single bit, where a first value (e.g., 1) may indicate high probability, and a second value (e.g., 0) may indicate low probability.

[0182] In some implementations, upon predicting the RLF occurrence, the UE may perform a cell pre-reselection procedure. During the cell pre-reselection procedure, the UE may select a cell, and / or store the information for initial access to the cell (e.g., MIB and SIB1).

[0183] In some implementations, upon the actual RLF occurrence, if the UE has selected a cell and has stored the information for initial access to the cell, the UE may perform random access towards the selected cell using the stored information for initial access to the selected cell. In some implementations, upon the actual RLF occurrence, if the UE has selected a cell, the UE may acquire the information for initial access to the cell and perform random access towards the selected cell using the stored information for initial access to the selected cell. In some implementations, the UE may recognize that it can’t report the inference before the actual RLF occurrence (e.g., no available resource for reporting), and then the UE may perform early-synchronization with an appropriate cell. Note this may rely on UE implementation and no standardization impacts here. Specifically, UE may report RLF to the network and the new cause may be no available resource for inference reporting and then RLF. In some implementations, the UE may select the cell according to the cell’s priority, the cell’s current RRM measurement quantities (e.g., RSRP, RSRQ, SINR), and / or the cell’s predicted RRM measurement quantities (more specifically, the measurement quantities near the time instance of RLF occurrence). More specifically, the UE may select the cell if the cell’s predicted RRM measurement quantity is offset better than a threshold.

[0184] In some implementations, the offset and the threshold may be configured by the network (e.g., via SIBs). In some implementations, for cells with different priorities, the offsets for the cells may have different values. In some implementations, the UE may consider using the cell’s predicted RRM measurement quantities for the cell pre-reselection if the cell’s predicted RRM measurement quantities are available. Otherwise (e.g., the cell’s predicted RRM measurement quantities are not available), the UE may consider using the cell’s current RRM measurement quantities for the cell pre-reselection. In some implementations, the parameters for cell pre-reselection may be different from those for cell reselection. For example, the network may configure the UE with a first set of parameters for cell pre-selection and a second set of parameters for cell reselection. For another example, the value of a parameter for cell pre-reselection may be the value of the counterpart for cell reselection plus an offset value. In some implementations, the offset value may be configured by the network.

[0185] In some implementations, upon or after the initial access procedure, the UE may transmit an RRCReestablishmentRequest procedure to the selected cell, where the UE may set the ReestablishmentCause to a specific value (e.g., ‘PredictedFailure’).

[0186] In some implementations, upon predicting the RLF occurrence, the UE may perform a cell reselection procedure. During the cell reselection procedure, the UE may select a cell and perform random access towards the selected cell. In some implementations, upon predicting the RLF occurrence, the UE may transmit signaling to the network indicating a predicted RLF occurrence if the time instance of the predicted RLF occurrence falls after a threshold time instance. The UE may perform a cell pre-reselection procedure if the time instance of the predicted RLF occurrence falls before a threshold time instance.

[0187] More specifically, the threshold time instance may be an offset taking the current time as the reference point, and the offset may be configured by the network. More specifically, the threshold time instance may be dependent on the UE’s capability and may be reported in the RRC message related to UE’s capability (e.g., the UECapabilityInformation message). In some implementations, the UE may expect that the threshold time instance configured by the network is after the threshold time instance reported by the UE. In some implementations, the UE may initiate an RRC re-establishment procedure to the network if the threshold time instance configured by the network is before the threshold time instance reported by the UE.

[0188] In some implementations, upon predicting the RLF occurrence, the UE may transmit signaling to the network indicating a predicted RLF occurrence if the time instance of the predicted RLF occurrence falls after a threshold time instance. The UE may initiate an RRC re-establishment procedure if the time instance of the predicted RLF occurrence falls before a threshold time instance.

[0189] More specifically, the threshold time instance may be an offset taking the current time as the reference point, and the offset may be configured by the network. More specifically, the threshold time instance may be dependent on the UE’s capability and may be reported in the RRC message related to UE’s capability (e.g., the UECapabilityInformation message). In some implementations, the UE may expect that the threshold time instance configured by the network is after the threshold time instance reported by the UE. In some implementations, the UE may initiate an RRC re-establishment procedure to the network if the threshold time instance configured by the network is before the threshold time instance reported by the UE.

[0190] In some implementations, if the applicable RRM measurement prediction becomes -inapplicable, the UE may inform the network of this change but continuing generating the prediction (with a specific value) to the network. In some implementations, whether the RRM measurement prediction is applicable may be determined based on the length of OW and PW.

[0191] FIG. 3 is a flowchart illustrating a method / process 300 performed by a UE for measurement reporting for Artificial Intelligence / Machine Learning (AI / ML) enabled mobility, according to an example implementation of the present disclosure.

[0192] In the action 302, the process 300 may start by receiving, from a base station (BS), a configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction.

[0193] In the action 304, the process 300 may in response to receiving the configuration for the RRM prediction, report, via a measurement report message, one or more first predicted measurement results to the BS.

[0194] In the action 306, the process 300 may in response to receiving of the configuration for the measurement event prediction, report, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, where the timing information may be associated with the predicted measurement event. The process 300 may then end.

[0195] In some implementations, the UE may include the predicted measurement event in the measurement report message.

[0196] In some implementations, the configuration may indicate a periodic reporting or an event-triggered reporting for the one or more first predicted measurement results.

[0197] In some implementations, the configuration may include an event related parameter for the measurement event prediction.

[0198] In some implementations, the configuration may indicate a prediction window (PW) for the RRM prediction.

[0199] In some implementations, the configuration may include a synchronization signal block (SSB) configuration for the RRM prediction.

[0200] In some implementations, the one or more first predicted measurement results in the measurement report message are reported on a per-cell basis.

[0201] The steps / actions shown in FIG. 3 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. 3 may be omitted in some implementations and one or more actions shown in FIG. 3 may be combined.

[0202] The technical problem addressed by the method illustrated in FIG. 3 is how to enable efficient and proactive mobility management in wireless communication systems by leveraging Artificial Intelligence / Machine Learning (AI / ML) capabilities to predict future Radio Resource Management (RRM) measurements and measurement events. Specifically, the problem involves how a user equipment can effectively report predicted measurement information to a base station in a manner that allows the network to anticipate and prepare for mobility-related events before they occur, rather than relying solely on reactive measurements taken at the time of the event. This requires establishing appropriate configurations for different types of predictions (RRM prediction versus measurement event prediction) and determining what information should be reported for each prediction type to enable the base station to make informed decisions about mobility management, handover preparation, and resource allocation. The technical challenge includes how to structure the reporting mechanism to convey not only predicted measurement values but also predicted events and their associated timing information in a coordinated manner that facilitates AI / ML-enabled proactive network optimization.

[0203] The advantageous technical effect achieved by the method illustrated in FIG. 3 is the optimization of mobility management and network resource utilization through AI / ML-enabled predictive reporting that allows proactive network decision-making. By enabling the UE to report predicted RRM measurement results and predicted measurement events with associated timing information based on configured prediction types, the base station can anticipate future mobility scenarios and prepare appropriate resources, handover procedures, and network configurations before the actual mobility event occurs. This predictive approach reduces handover failures, minimizes service interruptions, and improves overall user experience by allowing the network to proactively optimize radio resources and connectivity. The differentiated reporting mechanisms for RRM predictions (reporting predicted measurement results) versus measurement event predictions (reporting predicted events with timing information) provide flexibility in how the network leverages AI / ML capabilities, enabling more efficient signaling overhead management while ensuring the base station receives the most relevant predictive information for decision-making. Ultimately, this AI / ML-enabled predictive mobility management enhances network efficiency, reduces latency in mobility procedures, and improves quality of service by transforming mobility management from a reactive process to a proactive, prediction-based optimization framework.

[0204] FIG. 4 is a block diagram illustrating a node 400 for wireless communication in accordance with various aspects of the present disclosure. As illustrated in FIG. 4, a node 400 may include a transceiver 420, a processor 428, a memory 434, one or more presentation components 438, and at least one antenna 436. The node 400 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. 4).

[0205] Each of the components may directly or indirectly communicate with each other over one or more buses 440. The node 400 may be a UE or a BS that performs various functions disclosed with reference to FIGS. 1 through 3.

[0206] The transceiver 420 has a transmitter 422 (e.g., transmitting / transmission circuitry) and a receiver 424 (e.g., receiving / reception circuitry) and may be configured to transmit and / or receive time and / or frequency resource partitioning information. The transceiver 420 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 420 may be configured to receive data and control channels.

[0207] The node 400 may include a variety of computer-readable media. Computer-readable media may be any available media that may be accessed by the node 400 and include volatile (and / or non-volatile) media and removable (and / or non-removable) media.

[0208] 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, Artificial Intelligence (AI) / Machine Learning (ML) module(s), or data.

[0209] Computer-storage media may include RAM, DRAM, HBM, MRAM, FRAM, PRAM, 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.

[0210] 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.

[0211] The memory 434 may include computer-storage media in the form of volatile and / or non-volatile memory. The memory 434 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. 4, the memory 434 may store a computer-readable and / or computer-executable instructions 432 (e.g., software codes) that are configured to, when executed, cause the processor 428 to perform various functions disclosed herein, for example, with reference to FIGS. 1 through 3. Alternatively, the instructions 432 may not be directly executable by the processor 428 but may be configured to cause the node 400 (e.g., when compiled and executed) to perform various functions disclosed herein. The AI / ML module(s) may be implemented with a supervised learning approach or an unsupervised learning approach (e.g., Transductive approach and Inductive approach).

[0212] The processor 428 (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 428 may include memory. The processor 428 may process the data 430 and the instructions 432 received from the memory 434, and information transmitted and received via the transceiver 420, the baseband communications module, and / or the network communications module. The processor 428 may also process information to send to the transceiver 420 for transmission via the antenna 436 to the network communications module for transmission to a CN.

[0213] One or more presentation components 438 may present data indications to a person or another device. Examples of presentation components 438 may include a display device, a speaker, a printing component, a vibrating component, etc.

[0214] 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 measurement reporting for Artificial Intelligence / Machine Learning (AI / ML) enabled mobility, the UE comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to 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 configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction; 　　　　　　　　in response to receiving of the configuration for the RRM prediction, report, via a measurement report message, one or more first predicted measurement results to the BS; and 　　　　　　　　in response to receiving of the configuration for the measurement event prediction, report, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, wherein 　　　　　　　　　　　　the timing information is associated with the predicted measurement event.

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: 　　　　include the predicted measurement event in the measurement report message.

3. The UE of claim 1, wherein the configuration indicates a periodic reporting or an event-triggered reporting for the one or more first predicted measurement results.

4. The UE of claim 1, wherein the configuration comprises an event related parameter for the measurement event prediction.

5. The UE of claim 1, wherein the configuration indicates a prediction window (PW) for the RRM prediction.

6. The UE of claim 1, wherein the configuration comprises a synchronization signal block (SSB) configuration for the RRM prediction.

7. The UE of claim 1, wherein the one or more first predicted measurement results in the measurement report message are reported on a per-cell basis.

8. A method performed by a user equipment (UE) for measurement reporting for Artificial Intelligence / Machine Learning (AI / ML) enabled mobility, the method comprising: 　　　　receiving, from a base station (BS), a configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction; 　　　　in response to receiving the configuration for the RRM prediction, reporting, via a measurement report message, one or more first predicted measurement results to the BS; and 　　　　in response to receiving the configuration for the measurement event prediction, reporting, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, wherein 　　　　　　　　the timing information is associated with the predicted measurement event.

9. A base station (BS) for measurement reporting for Artificial Intelligence / Machine Learning (AI / ML) enabled mobility, the BS comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to 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 configuration for at least one of a Radio Resource Management (RRM) prediction and a measurement event prediction, wherein the configuration causes the UE to: 　　　　　　　　　　　　in response to receiving the configuration for the RRM prediction, report, via a measurement report message, one or more first predicted measurement results to the BS; and 　　　　　　　　　　　　in response to receiving the configuration for the measurement event prediction, report, via the measurement report message, second predicted measurement results, a predicted measurement event, and timing information to the BS, wherein 　　　　　　　　　　　　the timing information is associated with the predicted measurement event.

10. The BS of claim 9, wherein the configuration further causes the UE to: 　　　　include the predicted measurement event in the measurement report message.

11. The BS of claim 9, wherein the configuration indicates a periodic reporting or an event-triggered reporting for the one or more first predicted measurement results.

12. The BS of claim 9, wherein the configuration comprises an event related parameter for the measurement event prediction.

13. The BS of claim 9, wherein the configuration indicates a prediction window (PW) for the RRM prediction.

14. The BS of claim 9, wherein the configuration comprises a synchronization signal block (SSB) configuration for the RRM prediction.

15. The BS of claim 9, wherein the one or more first predicted measurement results in the measurement report message are reported on a per-cell basis.

Images