Method and apparatus for handling configuration to manage logged measurements for ai / ML during handover

Abstract

The disclosure relates to 5G or 6G communication system for supporting higher data transmission rates and satisfying various service requirements. In an example, a method performed by a UE in a wireless communication system is provided. The method includes: receiving, from a source base station, RRC message for a handover from the source base station to a target base station; performing logging of CSI measurements for an AI / ML; identifying whether to retain a logged CSI measurement for the AI / ML based on the RRC message; and transmitting, to the target base station, a complete message, wherein, based on identification associated with retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained, and wherein, based on identification associated with not retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is discarded.

Description

METHOD AND APPARATUS FOR HANDLING CONFIGURATION TO MANAGE LOGGED MEASUREMENTS FOR AI / ML DURING HANDOVER

[0001] The disclosure relates to the operation of the user equipment (UE) and / or base station in wireless communication networks. More specifically, the disclosure pertains to a method and a apparatus for managing configuration to handle logged measurements for Artificial Intelligence (AI) and Machine Learning (ML) during handover.

[0002] 5G mobile communication technologies define broad frequency bands such that high transmission rates and new services are possible, and can be implemented not only in "Sub 6GHz" bands such as 3.5GHz, but also in "Above 6GHz" bands referred to as mmWave including 28GHz and 39GHz. In addition, it has been considered to implement 6G mobile communication technologies (referred to as Beyond 5G systems) in terahertz bands (for example, 95GHz to 3THz bands) in order to accomplish transmission rates fifty times faster than 5G mobile communication technologies and ultra-low latencies one-tenth of 5G mobile communication technologies.

[0003] At the beginning of the development of 5G mobile communication technologies, in order to support services and to satisfy performance requirements in connection with enhanced Mobile BroadBand (eMBB), Ultra Reliable Low Latency Communications (URLLC), and massive Machine-Type Communications (mMTC), there has been ongoing standardization regarding beamforming and massive MIMO for mitigating radio-wave path loss and increasing radio-wave transmission distances in mmWave, supporting numerologies (for example, operating multiple subcarrier spacings) for efficiently utilizing mmWave resources and dynamic operation of slot formats, initial access technologies for supporting multi-beam transmission and broadbands, definition and operation of BWP (BandWidth Part), new channel coding methods such as a LDPC (Low Density Parity Check) code for large amount of data transmission and a polar code for highly reliable transmission of control information, L2 pre-processing, and network slicing for providing a dedicated network specialized to a specific service.

[0004] Currently, there are ongoing discussions regarding improvement and performance enhancement of initial 5G mobile communication technologies in view of services to be supported by 5G mobile communication technologies, and there has been physical layer standardization regarding technologies such as V2X (Vehicle-to-everything) for aiding driving determination by autonomous vehicles based on information regarding positions and states of vehicles transmitted by the vehicles and for enhancing user convenience, NR-U (New Radio Unlicensed) aimed at system operations conforming to various regulation-related requirements in unlicensed bands, NR UE Power Saving, Non-Terrestrial Network (NTN) which is UE-satellite direct communication for providing coverage in an area in which communication with terrestrial networks is unavailable, and positioning.

[0005] Moreover, there has been ongoing standardization in air interface architecture / protocol regarding technologies such as Industrial Internet of Things (IIoT) for supporting new services through interworking and convergence with other industries, IAB (Integrated Access and Backhaul) for providing a node for network service area expansion by supporting a wireless backhaul link and an access link in an integrated manner, mobility enhancement including conditional handover and DAPS (Dual Active Protocol Stack) handover, and two-step random access for simplifying random access procedures (2-step RACH for NR). There also has been ongoing standardization in system architecture / service regarding a 5G baseline architecture (for example, service based architecture or service based interface) for combining Network Functions Virtualization (NFV) and Software-Defined Networking (SDN) technologies, and Mobile Edge Computing (MEC) for receiving services based on UE positions.

[0006] As 5G mobile communication systems are commercialized, connected devices that have been exponentially increasing will be connected to communication networks, and it is accordingly expected that enhanced functions and performances of 5G mobile communication systems and integrated operations of connected devices will be necessary. To this end, new research is scheduled in connection with eXtended Reality (XR) for efficiently supporting AR (Augmented Reality), VR (Virtual Reality), MR (Mixed Reality) and the like, 5G performance improvement and complexity reduction by utilizing Artificial Intelligence (AI) and Machine Learning (ML), AI service support, metaverse service support, and drone communication.

[0007] Furthermore, such development of 5G mobile communication systems will serve as a basis for developing not only new waveforms for providing coverage in terahertz bands of 6G mobile communication technologies, multi-antenna transmission technologies such as Full Dimensional MIMO (FD-MIMO), array antennas and large-scale antennas, metamaterial-based lenses and antennas for improving coverage of terahertz band signals, high-dimensional space multiplexing technology using OAM (Orbital Angular Momentum), and RIS (Reconfigurable Intelligent Surface), but also full-duplex technology for increasing frequency efficiency of 6G mobile communication technologies and improving system networks, AI-based communication technology for implementing system optimization by utilizing satellites and AI (Artificial Intelligence) from the design stage and internalizing end-to-end AI support functions, and next-generation distributed computing technology for implementing services at levels of complexity exceeding the limit of UE operation capability by utilizing ultra-high-performance communication and computing resources.

[0008] Usage of Artificial intelligence and Machine Learning is used in a variety of fields. 3rd Generation Partnership Project (3GPP) is studying the AI / ML on air interface. The AI / ML based techniques can be used for enhanced performance and / or reduced complexity / overhead. Enhanced performance here depends on the use cases under consideration and could be, e.g., improved throughput, robustness, accuracy or reliability, etc. AI / ML models may reside on a UE or on network entities like base station or Operations, Administration, and Maintenance (OAM). It is also possible that AI / Models are 2 sided models which reside on both UE and network entities.

[0009] Data collection may be performed for different purposes in Life Cycle Management (LCM), e.g., model training, model inference, model monitoring, model selection, model update, etc. For all types of offline model training (i.e., UE- / NW- / two-sided model training), there is no latency requirement for data collection. For model inference, when required data comes from other entities, there is a latency requirement for data collection. For (real-time) performance monitoring, when required monitoring data (e.g., performance metric) comes from other entities, there is a latency requirement for data collection.

[0010] A typical usage of the AI / ML on the air interface is AI / ML for beam management, which includes spatial-domain downlink beam prediction for Set A of beams based on measurement results of Set B of beams, and temporal downlink beam prediction for Set A of beams based on the historic measurement results of Set B of beams.

[0011] The AI / ML on air interface also may be used for positioning, such as direct AI / ML positioning and AI / ML assisted positioning. The direct AI / ML positioning includes UE-based positioning with UE-side model, direct AI / ML positioning, UE-assisted / LMF-based positioning with LMF-side model, direct AI / ML positioning, and NG-RAN node assisted positioning with LMF-side model. The AI / ML assisted positioning includes UE-assisted / LMF-based positioning with UE-side model, AI / ML assisted positioning, and NG-RAN node assisted positioning with gNB-side model, AI / ML assisted positioning. Another use case of the AI / ML on the air interface is for CSI feedback enhancements. This may be used for CSI prediction, CSI compression etc. The AI / ML may be used for other cases such as AI / ML for mobility.

[0012] The training can be online training or offline training. CN (Core Network) / OAM (Operation and Maintenance) / OTT (Over the Top Server) collection of UE-sided model training (and even Network sided model training) data may be supported. The UE may be configured by the network for reporting layer 1 (L1) measurements for AI / ML. These measurements may be logged and reported to the network. The measurements are typically layer 1 measurements, for beam prediction or CSI feedback enhancements. These measurements would be typically used for training the AI / ML models (for instance training the network side models for the AI / ML for physical layer such as for beam management or CSI feedback enhancements), but can be used for other use cases also. Additionally, the UE may be configured to log and report other measurements such as Layer 3 RSRP, Layer 3 RSRQ, Layer 3 SINR etc. for AI / ML purposes. They may be used for other use cases such as Layer 3 mobility. All of these can be referred to as logged measurements for AI / ML. Version 18.5.1 of TS 38.331, 38.321, 38.306, 37.340, 38.300 etc. are considered as background for this invention.

[0013] Further, wireless communication technologies have significantly evolved over the years, with the advent of 5th Generation New Radio (5G NR) representing a major leap forward. 5G NR devices are designed to move seamlessly across different cells within the network. This mobility is crucial for maintaining continuous service as the UE moves geographically. Mobility management in 5G NR is categorized into different modes, with cell reselection being a key procedure in RRC_IDLE mode, and network-controlled mobility applicable to the UEs in RRC_CONNECTED mode. It is expected that the key principles of 5G NR will be applicable to newer technologies also. In the RRC_CONNECTED mode, mobility management involves explicit RRC signaling, which is initiated by the gNB. The handover process in 5G NR generally comprises three main steps: handover preparation, handover execution, and handover completion. During this process, the gNB may configure the UE to report measurements. Based on these measurements or its own understanding of the network topology, the gNB sends an RRC Reconfiguration message to handover the UE from the source cell to a target cell. The UE then accesses the target cell and sends an RRC Reconfiguration complete message to finalize the handover.

[0014] An alternative handover mechanism introduced in the 3GPP NR release 16 allows the gNB to configure the UE with execution conditions for triggering handover. Once these conditions are met, the UE autonomously moves to the target cell and sends the RRC Reconfiguration complete message. Additionally, UEs may be configured to perform Lower Layer Triggered Mobility (LTM). UE may be configured with ltm-Config for MCG LTM and Intra-SN LTM. UE may be also configured with LTM-ConfigNRDC to provide LTM configurations in NR-DC for inter-SN LTM or Intra-SN LTM which includes MN involvement. UE may be configured to apply full configuration during a Layer 3 (L3) handover, as described in section 5.3.5.11 of TS 38.331. In scenarios involving dual connectivity, the UE may perform PSCellChange or Conditional PSCellChange, which are also considered L3 mobility procedures. In this context, PSCellChange or Conditional PSCellChange may be referred to as Secondary Cell Group (SCG) layer 3 mobility, while handover and Conditional Handover (CHO) are referred to as Master Cell Group (MCG) layer 3 mobility. SCG mobility can include Subsequent PSCell Addition or Change (SPAC). UEs may be configured with full configuration or delta configuration for L3 mobility, with LTM and SCPAC supporting reference configuration-based mobility.

[0015] Despite these advancements, existing solutions present several challenges. For instance, during full configuration, the UE releases the current configuration and applies the target configuration. Similarly, during mobility procedures like LTM cell switch or subsequent Conditional PSCell Addition or Change (CPAC) execution, which support reference configuration, the UE clears the current configuration and applies the target configuration. This target configuration may be generated using a reference configuration and a candidate configuration. Consequently, the UE faces difficulty in determining whether to retain or release Artificial Intelligence / Machine Learning (AI / ML) measurements during a handover. This uncertainty can impact the efficiency and performance of mobility procedures, leading to potential service disruptions and degraded user experience.

[0016] Furthermore, the integration of Artificial Intelligence (AI) and Machine Learning (ML) in wireless communication systems is rapidly advancing, with significant interest from the 3GPP in exploring their application on the air interface. The AI and ML-based solutions hold the potential to enhance network performance and reduce operational complexity and overhead. These enhancements can manifest in various forms, such as improved throughput, robustness, accuracy, and reliability, depending on the specific use cases. AI / ML models can be deployed on the UE, network entities such as base stations, or OAM systems. Additionally, these models can operate in a two-sided manner, residing both on the UE and network entities. The LCM of AI / ML models involve various stages, including model training, inference, monitoring, selection, and updates. Data collection is a critical component of these stages. While offline model training does not impose latency requirements on data collection, real-time model inference and performance monitoring demand low-latency data collection to ensure timely and accurate decision-making. One prominent application of AI / ML on the air interface is in beam management, which includes spatial-domain downlink beam prediction based on measurement results and temporal downlink beam prediction based on historical data. AI / ML techniques are also employed for positioning, encompassing direct AI / ML positioning and AI / ML-assisted positioning, with models residing on the UE, Location Management Function (LMF), or Next Generation Radio Access Network (NG-RAN) nodes.

[0017] The primary goal of integrating AI / ML into the New Radio (NR) air interface is to improve network performance and user experience through the analysis of collected data and the enhancement of AI / ML-enabled features, such as beam management and positioning accuracy. Despite the potential benefits, existing solutions do not adequately address several critical use cases related to the handling of logged measurements for AI / ML including the handling of logged measurements for AI / ML by the UE during Inter-RAT (Radio Access Technology) handovers, and the handling of logged measurements for AI / ML by the UE during full configuration scenarios, the handling of logged measurements for AI / ML by the UE when the Signaling Radio Bearer (SRB) for transmitting these measurements becomes unavailable.

[0018] Thus, there is a need to address the above-mentioned disadvantages and shortcomings or at least provide a useful alternative.

[0019] The disclosure relates to a method and a apparatus for managing configuration to handle logged measurements for AI / ML during handover.

[0020] Accordingly, an aspect of the disclosure is to provide a method, a User Equipment (UE), a source apparatus, and / or a target apparatus for handling Artificial Intelligence (AI) / Machine Learning (ML) logged measurements during a handover procedure.

[0021] Another aspect of the disclosure is to handle configuration to manage logged measurements for AI / ML during handover.

[0022] Yet another aspect of the disclosure is to provide a UE for handling logged measurements during Inter-RAT handover.

[0023] The technical problems to be achieved in the various examples of the disclosure are not limited to those mentioned above, and other technical problems not mentioned can be considered by a person having ordinary skill in the art from the various examples of the disclosure described below.

[0024] According to an aspect of the disclosure, a method performed by a user equipment (UE) in a wireless communication system is provided. The method includes: receiving, from a source base station, a radio resource control (RRC) message for a handover from the source base station to a target base station; performing logging of channel state information (CSI) measurements for an artificial intelligence / machine learning (AI / ML); identifying whether to retain a logged CSI measurement for the AI / ML based on the RRC message; and transmitting, to the target base station, a complete message, wherein, based on identification associated with retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained, and wherein, based on identification associated with not retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is discarded.

[0025] According to an aspect of the disclosure, a method performed by a target base station in a wireless communication system is provided. The method includes: receiving, from a source base station, a handover preparation information message including information on retaining a logged channel state information (CSI) measurement for an artificial intelligence / machine learning (AI / ML); transmitting, to a user equipment (UE) via the source base station, a radio resource control (RRC) message for a handover from the source base station to the target base station, the RRC message including the information on retaining the logged CSI measurement for the AI / ML; and receiving, from the UE, a complete message, wherein, based on the information on retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained.

[0026] According to an aspect of the disclosure, a user equipment (UE) in a wireless communication system is provided. The UE includes: a transceiver; at least one processor coupled to the transceiver; and memory coupled to the at least one processor and storing instructions executable by the at least one processor, wherein the instructions cause the UE to: receive, from a source base station, a radio resource control (RRC) message for a handover from the source base station to a target base station, perform logging of channel state information (CSI) measurements for an artificial intelligence / machine learning (AI / ML), identify whether to retain a logged CSI measurement for the AI / ML based on the RRC message, and transmit, to the target base station, a complete message, wherein, based on identification associated with retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained, and wherein, based on identification associated with not retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is discarded.

[0027] According to an aspect of the disclosure, a target base station in a wireless communication system is provided. The target base station includes: a transceiver; at least one processor coupled to the transceiver; and memory coupled to the at least one processor and storing instructions executable by the at least one processor, wherein the instructions cause the target base station to: receive, from a source base station, a handover preparation information message including information on retaining a logged channel state information (CSI) measurement for an artificial intelligence / machine learning (AI / ML), transmit, to a user equipment (UE) via the source base station, a radio resource control (RRC) message for a handover from the source base station to the target base station, the RRC message including the information on retaining the logged CSI measurement for the AI / ML, and receive, from the UE, a complete message, wherein, based on the information on retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained.

[0028] In an aspect, the objects are achieved by providing a method for handling AI / ML logged measurements during a handover procedure. The method includes receiving by a UE an RRC message including one or more handover configurations from a target apparatus or a source apparatus where the one or more handover configurations comprise whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and information to perform handover and determining by the UE whether the handover configuration is one of an Inter-RAT handover configuration or an intra-RAT handover configuration. Further, the method includes performing by the UE one of discarding the AI / ML logged measurements in the case of the Inter-RAT handover configuration or discarding or retaining the AI / ML logged measurements in the case of the Intra-RAT handover configuration based on the configuration received from the target apparatus or the source apparatus.

[0029] In another aspect, the objects are achieved by providing a method for handling AI / ML logged measurements during a handover procedure. The method includes generating by a source apparatus one or more handover configurations comprising ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover, and RetainMeasDuringHandover, whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and transmitting by the source apparatus one or more handover configurations selected from the group to a target apparatus.

[0030] In yet another aspect, the objects are achieved by providing a method for handling AI / ML logged measurements during a handover procedure. The method includes including by the target apparatus the received one or more handover configurations in its RRC message and sending by the target apparatus the RRC message to the UE where the UE discards or retains logged measurements for use in AI / ML applications based on the received one or more handover configurations.

[0031] In yet another aspect, the objects are achieved by providing a UE for handling AI / ML logged measurements during a handover procedure. The UE includes a processor, a memory, and a measurement log controller connected to the memory and the processor. The measurement log controller receives an RRC message including one or more handover configurations from a target apparatus or a source apparatus where the one or more handover configurations comprise whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and information to perform handover and determines whether the handover configuration is one of an Inter-Radio Access Technology (RAT) handover configuration or an intra-RAT handover configuration. Further, the measurement log controller performs one of discarding the AI / ML logged measurements in the case of the Inter-RAT handover configuration or discarding or retaining the AI / ML logged measurements in the case of the Intra-RAT handover configuration based on the configuration received from the target apparatus or the source apparatus.

[0032] In yet another aspect, the objects are achieved by providing a source apparatus for handling AI / ML logged measurements during a handover procedure. The source apparatus includes a processor, a memory, and a logged measurements configuration controller connected to the memory and the processor. The logged measurements configuration controller generates one or more handover configurations including ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover, and RetainMeasDuringHandover, whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and transmits one or more handover configurations selected from the group to the target apparatus.

[0033] In yet another aspect, the objects are achieved by providing a target apparatus for handling AI / ML logged measurements during a handover procedure. The target apparatus includes a processor, a memory, and a logged measurements configuration controller connected to the memory and the processor. The logged measurements configuration controller includes the received one or more handover configurations in its RRC message and sends the RRC message to a User Equipment (UE) wherein the UE discards or retains logged measurements for use in AI / ML applications based on the received one or more handover configurations.

[0034] These and other aspects of the embodiments will be better understood with the following description and accompanying drawings. The descriptions, while indicating preferred embodiments and specific details, are for illustration and not limitation. Many changes and modifications can be made within the scope of the embodiments, which include all such modifications.

[0035] According to the examples of the disclosure, the UE can retain the configuration for releasing or retaining logged CSI measurements for AI / ML during mobility procedures, and the target base station can be informed by the source base station of the retention status via handover preparation information, thereby enabling a joint decision on whether to keep or discard logged CSI measurements for AI / ML and improving the reliability of AI / ML data collection across mobility events.

[0036] In addition, according to the examples of the disclosure, the UE may discard the logged CSI measurements for AI / ML upon performing inter-RAT handover or when the dedicated SRB for transmitting the logged measurements becomes unavailable, thereby reducing memory overhead and implementation complexity of wireless network system.

[0037] The effects that can be obtained from the disclosure are not limited to the effects mentioned in the various examples, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the disclosure belongs from the description below.

[0038] The disclosure is illustrated in the accompanying drawings, where like reference letters indicate corresponding parts. The embodiments will be better understood from the following description with reference to the drawings.

[0039] FIG. 1 illustrates an example system for the usage of AI / ML in beam prediction according to examples disclosed herein.

[0040] FIG. 2 is a block diagram that illustrates the hardware components associated with the UE according to the examples disclosed herein.

[0041] FIG. 3 is a block diagram that illustrates the hardware components associated with the source apparatus according to the examples disclosed herein.

[0042] FIG. 4 is a block diagram that illustrates the hardware components associated with the target apparatus according to the examples disclosed herein.

[0043] FIG. 5 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the UE according to the examples disclosed herein.

[0044] FIG. 6 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the source apparatus according to the examples disclosed herein.

[0045] FIG. 7 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the target apparatus according to the examples disclosed herein.

[0046] FIG. 8 is a flow diagram that illustrates a method for handling configuration to manage logged measurements for AI / ML during handover according to examples disclosed herein.

[0047] FIG. 9 is a flow diagram that illustrates handling configuration to manage logged measurements for AI / ML according to examples disclosed herein.

[0048] FIG. 10 is a flow diagram that illustrates an example of handling logged measurements for AI / ML during mobility according to examples disclosed herein.

[0049] FIG. 11 is a flow diagram that illustrates handling logged measurements for AI / ML during full configuration according to examples disclosed herein.

[0050] FIG. 12 is a flow diagram that illustrates handling logged measurements for AI / ML when SRB becomes unavailable according to examples disclosed herein.

[0051] The embodiments and their features are detailed with reference to the non-limiting examples shown in the drawings and described below. Well-known components and techniques are omitted to avoid unnecessary detail. The described embodiments are not mutually exclusive and can be combined to form new embodiments. The term "or" is used in a non-exclusive sense unless stated otherwise. The examples provided are for illustrative purposes to aid understanding and should not be seen as limiting the scope of the embodiments.

[0052] As is existing in the field, embodiments can be described and illustrated in terms of blocks which carry out a described function or functions.　These blocks, which can be referred to herein as managers, units, modules, hardware components or the like, are physically implemented by analog and / or digital circuits such as logic gates, integrated circuits, microprocessors, microcontrollers, memory circuits, passive electronic components, active electronic components, optical components, hardwired circuits and the like, and can optionally be driven by firmware and software. The circuits may, for example, be embodied in one or more semiconductor chips, or on substrate supports such as printed circuit boards and the like.　The circuits constituting a block can be implemented by dedicated hardware, or by a processor (e.g., one or more programmed microprocessors and associated circuitry), or by a combination of dedicated hardware to perform some functions of the block and a processor to perform other functions of the block.　Each block of the embodiments can be physically separated into two or more interacting and discrete blocks without departing from the scope of the disclosure.　Likewise, the blocks of the embodiments can be physically combined into more complex blocks without departing from the scope of the disclosure.

[0053] The drawings help illustrate the technical features but do not limit the embodiments. The disclosure includes any modifications, equivalents, and substitutes beyond those shown. Terms like first, second, etc., are used for distinction and do not limit the elements.

[0054] FIG. 1 illustrates an example system for the usage of AI / ML in beam prediction. The system depicted in FIG. 1 illustrates an example of the usage of AI / ML in beam prediction. The system includes a network (NW), a UE, and various modules for data collection, model training, model inference, beam prediction, and performance monitoring. The NW is responsible for managing requests and transmitting reference signals (RS transmission) to the UE. The UE collects data, which is then used for model training and model inference at both UE and network. The collected data can include layer 1 (L1) measurements for AI / ML, which are typically used for beam prediction or CSI feedback enhancements. The model training module processes the collected data to train AI / ML models. These trained models are then utilized by the model inference module to make predictions regarding beamforming. The beam prediction module uses the output from the model inference module to predict optimal beam directions. The system also includes performance monitoring modules, namely Type 1 Performance monitoring and Type 2 Performance monitoring. These modules monitor the performance of the beam predictions and overall system efficiency.

[0055] The beam indication module is responsible for indicating the predicted beam directions to the network NW. This information helps the network optimize its beamforming strategies. The system supports both online training and offline training methodologies. Online training involves real-time data collection and model updates, while offline training involves using pre-collected data for model training. The AI / ML operations typically include multiple steps: training, inference, and performance monitoring. The UE may indicate its capabilities for AI / ML-related functionalities to the network. This may be sent in Radio Resource Control (RRC) messages such as UECapabilityInformation. The UE may not be able to perform the AI / ML-related functionality always (for example, inference may not always be possible) even though the functionality is reported as supported. For example, the UE may not be able to perform the inference for the AI / ML-related functionality even though it is supported. This may be due to the absence of a relevant model for the functionality or other reasons such as hardware conditions that cannot support the available model or the available model is not applicable for current UE-side additional conditions or network-side additional conditions. The network may configure the UE for reporting the applicable functionality. The support of AI / ML for the NR air interface is to facilitate AI and ML techniques in the NR air interface. The objective of AI / ML for the NR air interface is to improve network performance and user experience through analyzing the data collected and enhancing AI / ML-enabled features, for example, beam management enhancement and positioning accuracy enhancement.

[0056] The AI / ML-enhanced beam management utilizes intra-cell beam prediction of the serving cell to reduce overhead and latency of beam measurement and to improve the accuracy of beam selection. Two types of beam prediction are supported as follows: Spatial-domain Downlink beam prediction for Set A of beams based on measurement results of Set B of beams, where Set B can be different from Set A or a subset of Set A, and Temporal-domain Downlink beam prediction for Set A of beams based on historic measurement results of Set B of beams. Set B can be the same as Set A or different from Set A or a subset of Set A.

[0057] For the deployment of AI / ML-enhanced beam management, both gNB-side model and UE-side model are supported. Data collection for model training is performed under network control, i.e., the network determines when to start or stop the data collection by provision of data collection configuration. For the gNB-side model, model training can be performed at the OAM or the gNB. The enhanced immediate MDT as specified in TS 37320 is used to support data collection for model training at the OAM. For the UE-side model, the network can configure whether the UE is allowed to initiate a request for training data collection. Data collection-related configuration (e.g., measurement resources configuration) and associated ID can be included in the training data collection configuration.

[0058] The UE can indicate its supported functionalities via UE capability information. For the UE-side model, the UE can decide the applicable functionalities based on NW-side additional conditions (if provided), UE-side additional conditions (internally known by UE), and model availability in the UE. To assist the network in managing the UE-side model, the UE can report its applicable functionalities to the network proactively and / or reactively. During handover, the target gNB can provide the UE with NW-side additional conditions and / or the inference configuration related to the target gNB UE via handover command. Then the UE can report its applicable functionalities to the network after the handover completion via either RRCReconfigurationComplete or UAI. UAI can be sent from the source gNB to the target gNB using existing signaling. For the gNB-side model, the CSI report framework is used to acquire input data for inference. For the gNB-side model, the gNB is responsible for performance monitoring. The UE will not be involved in management decisions (e.g., selection, (de)activation, switching, fallback, etc.) of the gNB-side model except for being configured to provide the required measurement / data. Additionally, the UE will not be informed about any gNB-side management decision. AI / ML-based positioning accuracy enhancement is described in TS 38305. The network may send an indication on whether to release or retain un-retrieved data in RRCReconfiguration during / before handover. The source gNB decides whether the data should be kept, and the indication is provided in RRCReconfiguration, i.e., not in RRC Reconfiguration from the target cell.

[0059] Throughout the specification, the terms NF producer status, NF producer degradation status, NF producer reliability status, and NF producer reliability check are used interchangeably. These terms collectively refer to an assessment (e.g., a value / flag / score) indicative of whether the NF producer is considered reliable for a service operation.

[0060] In an embodiment, the proposed solution describes the invention based on the following variables:

[0061] ReleaseMeasDuringHandover: In this solution, ReleaseMeasDuringHandover is used to indicate the configuration to release un-retrieved logged measurements for AI / ML during handover. ReleaseMeasDuringHandover is an example name and the embodiments referring to ReleaseMeasDuringHandover are applicable for any configuration with the similar functionality, i.e. to indicate the configuration to release un-retrieved logged measurements for AI / ML during handover.

[0062] RetainMeasDuringHandover: In this solution, RetainMeasDuringHandover is used to indicate the configuration to retain un-retrieved logged measurements for AI / ML during handover. RetainMeasDuringHandover is an example name and the embodiments referring to RetainMeasDuringHandover are applicable for any configuration with the similar functionality, i.e. to indicate the configuration to retain un-retrieved logged measurements for AI / ML during handover.

[0063] ReleaseorRetainMeasDuringHandover:In this solution, ReleaseorRetainMeasDuringHandover is used to indicate the configuration to release or retain un-retrieved logged measurements for AI / ML during handover. ReleaseorRetainMeasDuringHandover is an example name and the embodiments referring to ReleaseorRetainMeasDuringHandover are applicable for any configuration with the similar functionality i.e. to indicate the configuration to release or retain un-retrieved logged measurements for AI / ML during handover.

[0064] A measurement is considered as unretrieved measurement when either the UE has not sent the measurements to the network or the UE has not received acknowledgement for the measurements which are sent. Typically, the measurements may be sent in UEInformationResponse or other RRC messages. They may be also sent over radio bearers such as data radio bearers or over a service plane, in some scenarios.

[0065] The logged measurements for AI / ML may be stored in a UE variable. Unretrieved measurements in the context of this solution are the logged measurements for AI / ML available in the UE, for e.g. in the above mentioned UE variable.

[0066] The proposed solution addresses how the network configures and the UE applies or releases the configurations, RetainMeasDuringHandover, ReleaseorRetainMeasDuringHandover or ReleaseMeasDuringHandover or the equivalent configurations.

[0067] The configurations are sent from network node such as gNB to UE in RRC messages such as RRCReconfiguration. Handover in this solution can include one or more of different mobility procedures such as L3 mobility, conditional handover, LTM cell switch etc. on either MCG or SCG.

[0068] According to the existing mechanisms, the network may send an indication on whether to release or retain un-retrieved data in RRCReconfiguration during / before handover. Source apparatus decides whether the data should be kept and the indication is provided in RRCReconfiguration, i.e. not in RRC Reconfiguration from target cell. When source apparatus provides the indication to the UE on whether to keep the logged measurements in its own RRC Reconfiguration, it may not work properly during various kinds of mobility such as full configuration, LTM cell switch, subsequent CPAC etc. This also leads to complex signaling as the information needs to depend on whether the target apparatus is able to handle the data. In the existing mechanisms, the source apparatus informs the UE whether to retain the logged measurements. Further, in the existing mechanisms, the UE logs the measurements for AI / ML operations and reports to the network when the UE is configured for handover, either Intra-RAT or Inter-RAT handover. But how does the UE handle the logged measurements is not disclosed in these scenarios. In the existing mechanisms, the UE does not release the logged data for AI / ML during Inter-RAT handover.

[0069] The disclosure provides a method, a source apparatus, and a target apparatus for handling AI / ML logged measurements during a handover procedure. The solution proposes that the source apparatus informs the target apparatus whether the UE needs to keep the logged measurements for AI / ML and the target apparatus configures the UE to retain the logged measurements. The solution allows efficient handling of the logged data for AI / ML by the UEs. AI / ML for physical layer and mobility are one of the most important features for R19 / R20 5G-Advanced and 6G. This is mandatory feature in standards for UEs and networks supporting Rel19 AI / ML for physical layer and Rel20 AI / ML for mobility. The solution proposes that upon performing successful Inter-RAT handover, the UE releases the logged measurements. If the handover is Intra-RAT, the UE keeps or discard the logged measurements based on the information from the network.

[0070] Referring now to the drawings, and more particularly to FIGS. 2 through 12, where similar reference characters denote corresponding features consistently throughout the figures, there are shown embodiments.

[0071] FIG. 2 is a block diagram that illustrates the hardware components associated with the UE (200) according to the disclosure. In an embodiment, examples of the UE (200) can include, but are not limited to, Consumer Electronics (such as Mobile Phones and Smartphones), Tablets, Wearable Devices, Computing Devices (such as Laptops, Notebooks, Desktops, Workstations, etc.), IoT Devices, Automotive Systems (such as connected cars, Autonomous Vehicles, Vehicle-to-Everything (V2X) communication devices, etc.), Enterprise Devices such as robotics, Specialized Equipment (such as Medical Devices, Public Safety Devices, etc.), and Media Devices (such as Gaming Consoles, Streaming Devices, etc.).

[0072] As illustrated in FIG. 2, the UE (200) includes a processor (201), a memory (202), a communicator (203), and a Measurement Log Controller (204). The processor (201) is configured to execute instructions stored in the memory (202) and perform operations described herein, including handling AI / ML logged measurements during a handover procedure. The processor (201) is communicatively coupled to the memory (202), the communicator (203), and the Measurement Log Controller (204). In an embodiment, the processor (201) includes one or more processing cores such as a CPU and / or an accelerator (e.g., NPU) suitable for protocol processing, policy evaluation, and control-plane workloads.

[0073] The memory (202) stores an operating system, platform software (including virtualization / container runtime in an embodiment), application software, and data used by the processor (201). In an embodiment, the memory (202) comprises one or more non-transitory computer-readable storage media, including volatile and / or non-volatile memory. In an embodiment, the memory (202) stores the RRC message, including one or more of the handover configurations received from a target apparatus (400).

[0074] The communicator (203) is configured to facilitate communication between the UE (200) and the target apparatus (400), supporting various communication protocols such as Transmission Control Protocol / Internet Protocol (TCP / IP), User Datagram Protocol (UDP), and second-generation Digital Video Broadcasting by Satellite (DVB-S2). It ensures that the UE (200) can communicate effectively with the target apparatus (400), sending and receiving the necessary data for handling AI / ML logged measurements during a handover procedure. Internal communication between hardware components via one or more networks is also facilitated by the communicator (203). An electronic circuit specific to a standard that enables wired or wireless communication is included in the communicator (203). The communicator (203) receives the RRC message, including one or more of the handover configurations from the target apparatus (400).

[0075] The Measurement Log Controller (204) is implemented using dedicated processing circuitry and is operatively coupled to the processor (201), the memory (202), and the communicator (203). The hardware implementation of the Measurement Log Controller (204) ensures efficient and reliable execution of processes integral to handling AI / ML logged measurements during a handover procedure. The Measurement Log Controller (204) includes a multi-core architecture tailored to optimize the handling of AI / ML logged measurements during a handover procedure. Each core within the architecture is specifically designed to execute distinct functions.

[0076] The Measurement Log Controller (204) receives an RRC message, including one or more of the handover configurations from a target apparatus (400) or a source apparatus (300), where the one or more of the handover configurations comprise whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and information to perform handover and determines whether the handover configuration is one of an Inter-RAT handover configuration or an intra-RAT handover configuration. Then the Measurement Log Controller (204) performs one of discarding the AI / ML logged measurements in the case of the Inter-RAT handover configuration or discards or retains the AI / ML logged measurements in the case of the Intra-RAT handover configuration based on the configuration received from the target apparatus (400) or the source apparatus (300).

[0077] Further, the Measurement Log Controller (204) receives the handover configuration in a RRCReconfiguration message comprising reconfigurationWithSync from the target apparatus (400) for intra-RAT handover. The Measurement Log Controller (204) receives the handover configuration in a MobilityFromNRCommand for inter-RAT handover. The Measurement Log Controller (204) discards the AI / ML logged measurements upon successfully completing as Mobility from NR to E-UTRA and discards the AI / ML logged measurements upon performing a full configuration for the handover procedure.

[0078] FIG. 3 is a block diagram that illustrates the hardware components associated with the source apparatus (300) according to the disclosure. Examples of the source apparatus (300) can include, but are not limited to, Base Stations (such as macro cells, small cells, femtocells, picocells), or eNb or gNb for wireless communication, Antennas and RF Units (e.g., MIMO beamforming) to enhance signal coverage and data throughput, Core Network Equipment (e.g., MMEs, HSS, S-GWs, P-GWs in 4G, AMFs, UPFs in 5G) for data routing, mobility, and session control, Network Function Virtualization (NFV) and Software-Defined Networking (SDN) for dynamic resource allocation and scalability, Edge Computing Nodes (e.g., MEC servers) for low-latency processing, Backhaul and Transport Equipment (e.g., fiber-optic links, microwave relays, Ethernet switches) to connect base stations to the core network, Network Management Systems (NMS) and Operation Support Systems (OSS) for network configuration, fault management, and optimization, Radio Network Controllers (RNCs) in 3G, Distributed Units (DUs) and Centralized Units (CUs) in 5G, Network Slicing Components for virtualized resource allocation, Security elements (e.g., Firewalls, IDS, AAA Servers) for secure communication.

[0079] As illustrated in FIG. 3, the source apparatus (300) includes a processor (301), a memory (302), a communicator (303), and a Logged Measurements Configuration Controller (304). The processor (301) is configured to execute instructions stored in the memory (302) and perform operations described herein, including handling AI / ML logged measurements during a handover procedure. The processor (301) is communicatively coupled to the memory (302), the communicator (303), and the Logged Measurements Configuration Controller (304). In an embodiment, the processor (301) includes one or more processing cores such as a CPU and / or an accelerator (e.g., NPU) suitable for protocol processing, policy evaluation, and control-plane workloads.

[0080] The memory (302) stores an operating system, platform software (including virtualization / container runtime in an embodiment), application software, and data used by the processor (301). In an embodiment, the memory (302) comprises one or more non-transitory computer-readable storage media, including volatile and / or non-volatile memory. In an embodiment, the memory (302) stores the generated one or more handover configurations comprising ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover, and RetainMeasDuringHandover.

[0081] The communicator (303) is configured to facilitate communication between the source apparatus (300) and the target apparatus (400), supporting various communication protocols such as TCP / IP, UDP, and DVB-S2. It ensures that the source apparatus (300) can communicate effectively with the target apparatus (400), sending and receiving the necessary data for handling AI / ML logged measurements during a handover procedure. Internal communication between hardware components via one or more networks is also facilitated by the communicator (303). An electronic circuit specific to a standard that enables wired or wireless communication is included in the communicator (303). The communicator (303) transmits the one or more handover configurations selected from the group to the target apparatus (400).

[0082] The Logged Measurements Configuration Controller (304) is implemented using dedicated processing circuitry and is operatively coupled to the processor (301), the memory (302), and the communicator (303). The hardware implementation of the Logged Measurements Configuration Controller (304) ensures efficient and reliable execution of processes integral to handling AI / ML logged measurements during a handover procedure. The Logged Measurements Configuration Controller (304) includes a multi-core architecture tailored to optimize the handling of AI / ML logged measurements during a handover procedure. Each core within the architecture is specifically designed to execute distinct functions.

[0083] The Logged Measurements Configuration Controller (304) generates one or more handover configurations comprising ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover, and RetainMeasDuringHandover, whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover, and transmits one or more handover configurations selected from the group to the target apparatus (400).

[0084] Further, the Logged Measurements Configuration Controller (304) sends one or more handover configurations to the target apparatus (400) in an NR Handover Preparation Information message.

[0085] FIG. 4 is a block diagram that illustrates the hardware components associated with the target apparatus (400) according to the disclosure. Examples of the target apparatus (400) can include, but are not limited to, Base Stations (such as macro cells, small cells, femtocells, picocells), or eNb or gNb for wireless communication, Antennas and RF Units (e.g., MIMO beamforming) to enhance signal coverage and data throughput, Core Network Equipment (e.g., MMEs, HSS, S-GWs, P-GWs in 4G, AMFs, UPFs in 5G) for data routing, mobility, and session control, NFV and SDN for dynamic resource allocation and scalability, Edge Computing Nodes (e.g., MEC servers) for low-latency processing, Backhaul and Transport Equipment (e.g., fiber-optic links, microwave relays, Ethernet switches) to connect base stations to the core network, NMS and OSS for network configuration, fault management, and optimization, RNCs in 3G, DUs and CUs in 5G, Network Slicing Components for virtualized resource allocation, Security elements (e.g., Firewalls, IDS, AAA Servers) for secure communication.

[0086] As illustrated in FIG. 4, the target apparatus (400) includes a processor (401), a memory (402), a communicator (403), and a Logged Measurements Configuration Controller (404). The processor (401) is configured to execute instructions stored in the memory (402) and perform operations described herein, including handling AI / ML logged measurements during a handover procedure. The processor (401) is communicatively coupled to the memory (402), the communicator (403), and the Logged Measurements Configuration Controller (404). In an embodiment, the processor (401) includes one or more processing cores such as a CPU and / or an accelerator (e.g., NPU) suitable for protocol processing, policy evaluation, and control-plane workloads.

[0087] The memory (402) stores an operating system, platform software (including virtualization / container runtime in an embodiment), application software, and data used by the processor (401). In an embodiment, the memory (402) comprises one or more non-transitory computer-readable storage media, including volatile and / or non-volatile memory. In an embodiment, the memory (402) stores the received one or more handover configurations from the source apparatus (300) in its RRC message.

[0088] The communicator (403) is configured to facilitate communication among the target apparatus (400), the source apparatus (300), and the UE (200), supporting various communication protocols such as TCP / IP, UDP, and DVB-S2. It ensures that the target apparatus (400) can communicate effectively with the source apparatus (300) and the UE (200), sending and receiving the necessary data for handling AI / ML logged measurements during a handover procedure. Internal communication between hardware components via one or more networks is also facilitated by the communicator (403). An electronic circuit specific to a standard that enables wired or wireless communication is included in the communicator (403). The communicator (403) receives the one or more handover configurations in its RRC message from the source apparatus (300) and sends the RRC message to the UE (200).

[0089] The Logged Measurements Configuration Controller (404) is implemented using dedicated processing circuitry and is operatively coupled to the processor (401), the memory (402), and the communicator (403). The hardware implementation of the Logged Measurements Configuration Controller (404) ensures efficient and reliable execution of processes integral to handling AI / ML logged measurements during a handover procedure. The Logged Measurements Configuration Controller (404) includes a multi-core architecture tailored to optimize the handling of AI / ML logged measurements during a handover procedure. Each core within the architecture is specifically designed to execute distinct functions.

[0090] The Logged Measurements Configuration Controller (404) includes the received one or more handover configurations in its RRC message and sends the RRC message to the UE (200), where the UE (200) discards or retains logged measurements for use in AI / ML applications based on the received one or more handover configurations. Further, the Logged Measurements Configuration Controller (404) sends one or more handover configurations to the UE (200) in a new RRCReconfiguration message including ReconfigurationWithSync which may be sent to the UE (200) through source apparatus (300). The target apparatus (400) may include the received one or more handover configuration in the RRCReconfiguration message based on its own algorithms. For instance, even if the source apparatus (300) provides the handover configuration for retaining logged measurements during handover, the target apparatus (400) may still decide that the UE (200) needs to delete the logged measurements during handover, if it is not able to retrieve the measurements within a certain time frame and may decide not to include the corresponding handover configuration. Alternatively, the target apparatus (400) may decide that it can retrieve the logged measurements and may include the handover configuration. The inter node exchange between the source apparatus (300) and the target apparatus (400) allows the retrieval of measurements based on the conditions of both the nodes.

[0091] FIG. 5 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the UE (200) according to the embodiments disclosed herein. At step 501, the method includes receiving by a UE (200) an RRC message including one or more of the handover configurations in a RRC Reconfiguration message from a target apparatus (400) or a source apparatus (300), where the one or more of the handover configurations comprise whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover and information to perform handover. RRC Reconfiguration message from the target apparatus (400) may be sent to the UE (200) by the source apparatus (300). At step 502, the method includes determining by the UE (200) whether the handover configuration is one of an Inter-RAT handover configuration or an intra-RAT handover configuration. The inter-RAT handover includes mobility from New Radio (NR) to Evolved Universal Terrestrial Radio Access (E-UTRA) or mobility from 6G to NR. At step 503, the method includes performing by the UE (200) one of discarding the AI / ML logged measurements in the case of the Inter-RAT handover configuration or discarding or retaining the AI / ML logged measurements in the case of the Intra-RAT handover configuration based on the configuration received from the target apparatus (400) or the source apparatus (300). Further, the method includes the handover configuration is received by the UE (200) from the target apparatus (400) in a RRCReconfiguration message for Intra-RAT handover. Further the method includes the handover configuration is received by the UE (200) in a MobilityFromNRCommand in the case of Inter-RAT handover. The inter-RAT handover comprises a mobility from NR to Evolved Universal Terrestrial Radio Access (E-UTRA) or mobility from 6G to NR. Furthermore, the method includes the UE (200) discards the AI / ML logged measurements upon successfully completing as Mobility from NR to E-UTRA and discards the AI / ML logged measurements upon performing a full configuration for the handover procedure.

[0092] FIG. 6 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the source apparatus (300) according to the embodiments disclosed herein. At step 601, the method includes generating by a source apparatus (300) one or more handover configurations including ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover, and RetainMeasDuringHandover, configuration indicating whether to discard or retain the logged measurements for AI / ML during the intra-RAT handover. The one or more handover configurations are sent from the source apparatus (300) to the target apparatus (400) in an NR Handover Preparation Information message (or equivalent message in another RAT). At step 602, the method includes transmitting by the source apparatus (300) one or more handover configurations selected from the group to the target apparatus (400).

[0093] FIG. 7 is a flow diagram that illustrates a proposed method for handling AI / ML logged measurements during a handover procedure by the target apparatus (400) according to the embodiments disclosed herein. At step 701, the method includes including by the target apparatus (400) the received one or more handover configurations in its RRC message. At step 702, the method includes sending by the target apparatus (400) the RRC message to the UE (200), where the UE (200) discards or retains logged measurements for use in AI / ML applications based on the received one or more handover configurations. Further, the method includes the one or more handover configurations are sent from the target apparatus (400) to the UE (200) in a new RRC Reconfiguration message. The RRC Reconfiguration message (e.g. NR RRCReconfiguration message) may include reconfigurationWithSync, i.e. the RRC message may be for synchronous reconfiguration.

[0094] FIG. 8 is a flow diagram illustrating a method for handling configuration to manage logged measurements for AI / ML during handover according to embodiments disclosed herein. In an embodiment, at step 801, the method includes receiving configuration by the UE (200) to release / retain / determine release or retain of logged measurements for AI / ML during handover. At step 802, the method includes performing by the UE (200) one of receiving explicit instruction to release the configuration to release / retain / "determine release or retain" of logged measurements for AI / ML during handover, perform handover, initiate RRC Reestablishment and transition to RRC_IDLE or RRC_INACTIVE. At step 803, the method includes releasing configuration to release / retain / determine release or retain of logged measurements for AI / ML.

[0095] According to TS 38.300, logging of layer1 measurements is described as follows:

[0096] - The UE can be configured by gNB to log L1 measurements in the AS layer memory and report them via a RRC message(s).

[0097] - Both periodic and L3 measurement event-triggered data collection are supported. The UE stores the logged data at the AS layer memory. When the AS layer memory for storing logged data becomes full, the UE stops measurement and logging for data collection. When the AS layer memory reserved for storing logged data becomes full or reaches an absolute threshold (if configured), the UE indicates data availability to the gNB, as specified in TS 38.331.

[0098] - When low power state is detected, the UE can indicate the low power state to the gNB. Upon reception of the low power state indication, the gNB should release the UE data collection configuration for Network-side model.

[0099] - The gNB can indicate UE whether the logged data should be kept or not during handover. When indicated to keep the logged data, UE retains it during handover and indicates its availability after handover.

[0100] The UE may store the logged measurements in UE variables such as VarCSI-LogMeasReport and may report them to the network node when the network node requests for reporting those measurements. An example of the UE variable for storing logged CSI measurements for network-side data collection is shown as Table 1. It is merely an example, and is not limited to the contents listed in the table.

[0101]

[0102] In an embodiment, a network node configures the UE (200) with a configuration on whether to release or retain un-retrieved logged measurements for AI / ML during handover (configuration such as ReleaseorRetainMeasDuringHandover) through RRC messages such as RRCReconfiguration or RRCResume. The configuration ReleaseorRetainMeasDuringHandover may be configured during or before the handover. If the configuration ReleaseorRetainMeasDuringHandover informs the UE (200) to retain the un-retrieved logged measurements for AI / ML during handover, UE keeps the unretrieved logged measurements during LTM cell switch and L3 mobility. If the configuration ReleaseorRetainMeasDuringHandover informs the UE (200) to release the un-retrieved logged measurements for AI / ML during handover, the UE (200) releases the un retrieved logged measurements during LTM cell switch and L3 mobility.

[0103] In an embodiment, the UE (200) which has received the configuration ReleaseorRetainMeasDuringHandover maintains the configuration. If the UE (200) which has received an RRCReconfiguration including ReleaseorRetainMeasDuringHandover receives another RRCReconfiguration without including the ReleaseorRetainMeasDuringHandover, the UE (200) performs releasing / retaining the logged measurements for AI / ML during handover based on the ReleaseorRetainMeasDuringHandover received in the first RRCReconfiguration. In an embodiment, network node may explicitly configure the UE (200) to release the ReleaseorRetainMeasDuringHandover. Network node may send an explicit configuration to release ReleaseorRetainMeasDuringHandover. Network node may send the ReleaseorRetainMeasDuringHandover as SetupRelease. If the configuration ReleaseorRetainMeasDuringHandover is received as Release, the UE (200) releases the configuration. In an embodiment, upon performing a handover or LTM cell switch or any mobility, the UE (200) releases ReleaseorRetainMeasDuringHandover.

[0104] In an embodiment, upon performing ReconfigurationWithSync, the UE (200) releases the ReleaseorRetainMeasDuringHandover. In an embodiment, upon moving to RRC_IDLE, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon receiving the RRC message RRCRelease, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon moving to RRC_INACTIVE, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon receiving RRC Release with suspendConfig, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, during radio link failure, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, during RRC Reestablishment, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon initiating RRC reestablishment procedure, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon initiating RRC Reestablishment, the UE (200) releases ReleaseorRetainMeasDuringHandover before attempting the LTM based or CHO based recovery. LTM based recovery means applying LTM candidate configuration after MCG radio link failure instead of sending RRCReestablishment (for e.g. as specified in TS 38.331). CHO based recovery means applying CHO configuration after MCG radio link failure instead of sending RRCReestablishment (for e.g. as specified in TS 38.331).

[0105] In an embodiment, upon initiation of RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps the ReleaseorRetainMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases ReleaseorRetainMeasDuringHandover. If the UE (200) is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases ReleaseorRetainMeasDuringHandover following cell selection while T311 is running.

[0106] In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases ReleaseorRetainMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps ReleaseorRetainMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, upon initiation of RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps ReleaseorRetainMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases ReleaseorRetainMeasDuringHandover. If the UE (200) is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases ReleaseorRetainMeasDuringHandover following cell selection while T311 is running.

[0107] In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases ReleaseorRetainMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps ReleaseorRetainMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases ReleaseorRetainMeasDuringHandover. In an embodiment, network node configures the UE (200) with a configuration to release un-retrieved logged measurements for AI / ML during handover (ReleaseMeasDuringHandover) through RRC messages such as RRCReconfiguration or RRCResume. This may be received from the source cell of the handover (mobility) before or during the handover. If the configuration ReleaseMeasDuringHandover is received, the UE (200) releases the un-retrieved logged measurements for AI / ML during handover. If the configuration ReleaseMeasDuringHandover is not received before or during the handover, the UE (200) retains the un-retrieved logged measurements for AI / ML during handover.

[0108] In an embodiment, the UE (200) which has received the configuration ReleaseMeasDuringHandover maintains this configuration. If the UE (200) receives an RRCReconfiguration including ReleaseMeasDuringHandover receives another RRCReconfiguration without including the ReleaseMeasDuringHandover, the UE (200) still releases the logged measurements for AI / ML during handover based on the ReleaseMeasDuringHandover received in the first RRCReconfiguration. In an embodiment, network node may explicitly configure the UE (200) to release the configuration, ReleaseMeasDuringHandover. Network node may send an explicit configuration to release ReleaseMeasDuringHandover. Network node may send the ReleaseMeasDuringHandover as SetupRelease. If it is received as Release, the UE (200) releases the configuration ReleaseMeasDuringHandover. In an embodiment, upon performing a handover or LTM cell switch or any other mobility, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon performing ReconfigurationWithSync, UE releases the ReleaseMeasDuringHandover.

[0109] In an embodiment, upon moving to RRC_IDLE, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon receiving the RRC message RRCRelease, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon moving to RRC_INACTIVE, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon receiving RRC Release with suspendConfig, UE releases ReleaseMeasDuringHandover. In an embodiment, during radio link failure, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, during RRC Reestablishment, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon initiating RRC Reestablishment, the UE (200) releases ReleaseMeasDuringHandover before attempting the LTM based or CHO based recovery. In an embodiment, upon initiating RRC reestablishment procedure, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon initiation of RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps the ReleaseMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases ReleaseMeasDuringHandover. If the UE is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases ReleaseMeasDuringHandover following cell selection while T311 is running.

[0110] In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases ReleaseMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps ReleaseMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, upon initiation of RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps ReleaseMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases ReleaseMeasDuringHandover. If the UE is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases ReleaseMeasDuringHandover following cell selection while T311 is running.

[0111] In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases ReleaseMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps ReleaseMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases ReleaseMeasDuringHandover. In an embodiment, network node configures the UE (200) with a configuration to retain un-retrieved logged measurements for AI / ML during handover (RetainMeasDuringHandover) through RRC messages such as RRCReconfiguration or RRCResume. This may be received from the source cell of the handover (mobility) before or during the handover. If the configuration RetainMeasDuringHandover is received, the UE retains the un-retrieved logged measurements for AI / ML during handover. If the configuration RetainMeasDuringHandover is not received, the UE (200) releases the un-retrieved logged measurements for AI / ML during handover. In an embodiment, the UE (200) which has received RetainMeasDuringHandover maintains the configuration. If the UE which has received a RRCReconfiguration including RetainMeasDuringHandover receives another RRCReconfiguration without including the RetainMeasDuringHandover, UE releases the logged measurements for AI / ML during handover based on the RetainMeasDuringHandover received in the first RRCReconfiguration.

[0112] In an embodiment, network node may configure the UE (200) to release the configuration, RetainMeasDuringHandover. Network node may send an explicit configuration to release RetainMeasDuringHandover. Network node may send the RetainMeasDuringHandover as SetupRelease. If RetainMeasDuringHandover is received as Release, UE releases the configuration, RetainMeasDuringHandover. In an embodiment, upon performing a handover or LTM cell switch or any other mobility, UE releases RetainMeasDuringHandover. In an embodiment, upon performing ReconfigurationWithSync, the UE (200) releases the RetainMeasDuringHandover. In an embodiment, upon moving to RRC_IDLE, the UE (200) releases RetainMeasDuringHandover. In an embodiment, upon receiving the RRC message RRCRelease, the UE (200) releases RetainMeasDuringHandover. In an embodiment, upon moving to RRC_INACTIVE, the UE (200) releases RetainMeasDuringHandover. In an embodiment, upon receiving RRC Release with suspendConfig, the UE (200) releases RetainMeasDuringHandover. In an embodiment, during radio link failure, the UE (200) releases RetainMeasDuringHandover. In an embodiment, during RRC Reestablishment, the UE (200) releases RetainMeasDuringHandover. In an embodiment, upon initiating RRC reestablishment procedure, the UE (200) releases RetainMeasDuringHandover. In an embodiment, upon initiating RRC Reestablishment, the UE (200) releases RetainMeasDuringHandover before attempting the LTM based or CHO based recovery.

[0113] In an embodiment, upon initiation of the RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps the RetainMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases RetainMeasDuringHandover. If the UE is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases RetainMeasDuringHandover following cell selection while T311 is running. In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases RetainMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps RetainMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases RetainMeasDuringHandover.

[0114] In an embodiment, upon initiation of RRC re-establishment procedure, the UE (200) checks if it is configured for attempting conditional reconfiguration for RLF recovery (for e.g. NR RRC IE attemptCondReconfig is configured) or if it is configured for attempting LTM for RLF recovery (for e.g. NR RRC IE attemptLTM-Switch is configured) and if any of these configurations are available (i.e. if any of these flags are configured), the UE (200) keeps RetainMeasDuringHandover. If none of these configurations for the recovery are available (i.e. none of these flags such as attemptCondReconfig or attemptLTM-Switch are configured), the UE (200) releases RetainMeasDuringHandover. If the UE is not able to perform CHO based recovery or LTM based recovery while the timer for cell selection in the RRC Reestablishment procedure (such as T311 timer in NR is running), the UE (200) releases RetainMeasDuringHandover following cell selection while T311 is running. In an embodiment, if both attemptCondReconfig and attemptLTM-Switch are not configured, the UE (200) releases RetainMeasDuringHandover while initiating RRC connection reestablishment. If any of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) keeps RetainMeasDuringHandover. In an embodiment, if the UE (200) is not able to apply LTM candidate configuration or conditional reconfiguration upon cell selection while T311 is running and if at least one of attemptCondReconfig and attemptLTM-Switch are configured, the UE (200) releases RetainMeasDuringHandover.

[0115] FIG. 9 is a flow diagram that illustrates the handling configuration to manage logged measurements for AI / ML according to embodiments as disclosed herein. At step 901, the method includes a configuration is received to release, retain, or determine the release or retention of logged measurements for AI / ML. At step 902, the method includes the solution includes performing a full configuration, executing an LTM cell switch, or carrying out a subsequent CPAC associated with the MCG. At step 903, the method includes the configuration to release, retain, or determine the release or retention of logged measurements for AI / ML is retained while performing the full configuration, executing the LTM cell switch, or carrying out the subsequent CPAC associated with the MCG.

[0116] In an embodiment, the UE (200) retains (i.e., avoids releasing) ReleaseorRetainMeasDuringHandover upon performing full configuration. The UE (200) releases all the dedicated configuration except specific configurations such as ReleaseorRetainMeasDuringHandover upon performing full configuration. In an embodiment, the above is applied when the full configuration is for handover (such as reconfigurationWithSync in NR). In an embodiment, the UE (200) retains (i.e., avoids releasing) ReleaseorRetainMeasDuringHandover upon performing full configuration. The UE (200) releases all the dedicated configuration except specific configurations such as ReleaseorRetainMeasDuringHandover upon performing full configuration. In an embodiment, the above is applied when the full configuration is for handover (such as reconfigurationWithSync in NR).

[0117] In an embodiment, the UE (200) retains (i.e., avoids releasing) RetainMeasDuringHandover upon performing full configuration. The UE (200) releases all the dedicated configuration except specific configurations such as RetainMeasDuringHandover upon performing full configuration. In an embodiment, the above is applied when the full configuration is for handover (such as reconfigurationWithSync in NR). An example of the dedicated radio configurations that the UE retains during full configuration is shown as Table 2. It is merely an example, and is not limited to the contents listed in the table.

[0118]

[0119] In an embodiment, the UE (200) keeps (i.e., avoids releasing) the ReleaseorRetainMeasDuringHandover upon LTM cell switch. In an embodiment, the UE (200) keeps (i.e., avoids releasing) the ReleaseorRetainMeasDuringHandover upon LTM cell switch for MCG. In yet another embodiment, the UE (200) keeps (i.e., avoids releasing) the ReleaseorRetainMeasDuringHandover upon LTM cell switch for SCG when the configuration applied at the time of LTM cell switch include MCG RRC Reconfiguration also. In an embodiment, the UE (200) keeps (i.e., avoids releasing) the ReleaseMeasDuringHandover upon LTM cell switch.

[0120] In an embodiment, the UE (200) keeps (i.e., avoids releasing) the ReleaseMeasDuringHandover upon LTM cell switch for MCG. In yet another embodiment, UE keeps the ReleaseMeasDuringHandover upon LTM cell switch for SCG when the configuration applied at the time of LTM cell switch include MCG RRC Reconfiguration also.

[0121] In an embodiment, the UE (200) keeps (i.e., avoids releasing) the RetainMeasDuringHandover upon LTM cell switch.

[0122] In an embodiment, the UE (200) keeps the RetainMeasDuringHandover upon LTM cell switch for MCG. In yet another embodiment, the UE (200) keeps the RetainMeasDuringHandover upon LTM cell switch for SCG when the configuration applied at the time of LTM cell switch include MCG RRC Reconfiguration also. An example of the dedicated and common radio configurations that the UE retains during LTM cell switch execution is shown as Table 3. It is merely an example, and is not limited to the contents listed in the table.

[0123]

[0124] In an embodiment, the UE (200) keeps (i.e., avoids releasing) ReleaseorRetainMeasDuringHandover upon MN configured subsequent CPAC execution. In NR, if the selected subsequent CPAC candidate configuration is stored in MCG VarConditionalReconfig, the UE (200) releases / clears all current dedicated radio configuration except specific configuration such as the ReleaseorRetainMeasDuringHandover. An example of the dedicated radio configurations that the UE retains during conditional reconfiguration execution for subsequent CPAC is shown as Table 4. It is merely an example, and is not limited to the contents listed in the table.

[0125]

[0126] In an embodiment, the UE (200) retains the configuration which informs the UE (200) to release or retain un-retrieved logged measurements for AI / ML during handover, during mobility procedures such as full configuration or the mobility procedures such as LTM cell switch or subsequent CPAC associated to MCG. The UE (200) may release other configurations related to AI / ML during these procedures. Releasing of other configurations while retaining one or more of ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover or RetainMeasDuringHandover may be typically performed before applying the target configuration during execution of mobility involving full configuration or mobility procedures which supports reference configuration. The UE (200) performs the applicable operations using this retained configuration itself during the full configuration, LTM cell switch, subsequent CPAC associated to MCG etc. (i.e. the UE (200) determines whether to release or retain the logged measurements for AI / ML based on the retained configuration.).

[0127] In an alternative embodiment, source network node (source apparatus (300)) informs the target network node (target apparatus (400)) the configuration related to releasing or retaining logged measurements for AI / ML during handover. The source apparatus (300) includes this configuration in HandoverPreparationInformation and sends to the target apparatus (400). Target apparatus (400) further sends it to the UE (200). The UE (200) releases or retains the logged measurements for AI / ML based on the received configuration. In an embodiment, the source apparatus (300) sends one or more of ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover or RetainMeasDuringHandover to the target apparatus (400). This may be sent in messages such as NR HandoverPreparationInformation or equivalent message in another RAT used for transferring UE access stratum context from a source network node to target network node during handover. The target apparatus (400) includes the received configuration in its own RRC message (RRCReconfiguration including ReconfigurationWithSync) and will send to the UE (200). It is also possible that target apparatus (400) does not include this field in its RRC Reconfiguration including ReconfigurationWithSync even though the source apparatus (300) includes this field in the HandoverPreparationInformation message depending on its own logic. The UE (200) releases or retains the logged measurements for AI / ML based on the received configuration.

[0128] Technical advantage of source informing the target about the configuration for retaining logged measurements for AI / ML:

[0129] Source node informing the target node about whether to retain logged measurements for AI / ML and target node informing this in its own RRC message helps to make a joint decision by both the nodes. For many use cases like AI / ML for beam management, the data closer to the handover may not be the best training data. So, the source node can decide whether it needs this data. On the other hand, the target node also needs to be involved as it needs to retrieve and send the data using its own resources (resources on air interface and network interface). The embodiments allow both the nodes to be part of the decision. An example of the HandoverPreparationInformation message structure including the indication for retaining logged measurements is shown as Table 5. It is merely an example, and is not limited to the contents listed in the table.

[0130]

[0131] retainLoggedMeasurements: Indication to the target gNB that the UE shall retain logged measurements available in VarCSI-LogMeasReport at execution of the handover. If included, the target gNB is allowed to include the corresponding indication to the UE within its RRCReconfiguration message including ReconfigurationWithSync message. An example of the UE procedure upon successful completion of the mobility from NR is shown as Table 6. It is merely an example, and is not limited to the contents listed in the table.

[0132]

[0133] Technical advantages of discarding logged measurements for AI / ML during Inter-RAT handover.

[0134] Discarding logged measurements for AI / ML after successfully completing Inter-RAT handover helps the UE to save the memory. Typically, UE implementations may have separate memories for different RATs. Logging data from another RAT (e.g.NR) consumes memory in the new RAT (e.g. LTE) and may affect the LTE performance. Moreover, it will increase the UE implementation complexity as the UE needs to handle the radio link failure or transition to RRC_IDLE from LTE. It will increase the complexity of network implementation also, as the LTE network needs to implement steps to retrieve NR data for AI / ML.

[0135] In an embodiment, the UE (200) releases one or more of the configurations ReleaseorRetainMeasDuringHandover, ReleaseMeasDuringHandover or RetainMeasDuringHandover upon receiving explicit instruction to release these configurations or upon initiating RRC Reestablishment or upon transitioning to RRC_IDLE or RRC_INACTIVE etc.

[0136] In an embodiment, the logged measurements for AI / ML in this solution can be logged measurements for AI / ML for physical layer, or the logged measurements for other purposes such as AI / ML for mobility, AI / ML for Radio Access Network operations or any other purpose related to AI / ML. It is also possible that the measurements for AI / ML are reported immediately instead of logged and then reported. The embodiments of the invention are equally applicable in that case also. "Logged measurements for AI / ML", "Logged data for AI / ML" and "data for AI / ML" can be used interchangeably in the context of this invention.

[0137] The UE (200) may report this logged AI / ML data to the network in UEInformationResponse using a low priority SRB. The disclosure will refer to this SRB as SRBx in this disclosure. Logged AI / ML data is sent over SRBx to ensure that this data doesn't block more important signaling (RRC or NAS) messages.

[0138] An example of the configuration for logging the measurements for AI / ML is shown as Table 7. It is merely an example, and is not limited to the contents listed in the table.

[0139]

[0140] An example of the field descriptions for theCSI-LoggedMeasurementConfigis shown as Table 8. It is merely an example, and is not limited to the contents listed in the table.

[0141]

[0142] The above example is one of the example definitions which may be used for AI / ML for physical layer. Similarly, there can be other configuration for logging of measurements for AI / ML for mobility.

[0143] FIG. 10 is a flow diagram that illustrates an example of handling logged measurements for AI / ML during mobility at the UE (200) according to embodiments as disclosed herein. At step 911, the method includes the UE (200) logs the measurements for the AI / ML. At step 912, the UE (200) triggers to perform mobility operation. At step 913, the UE (200) determines whether the type of mobility is an Inter-RAT or an Intra-RAT. If the configuration is Inter-RAT handover, the UE (200) discards the logged measurements at step 914, or if the configuration is Intra-RAT handover, the UE (200) discards or retains the logged measurements based on configuration from the network, at step 915.

[0144] In an embodiment, the UE (200) discards the logged measurements for AI / ML during Inter-RAT handover. The UE (200) may discard or retain the logged measurements for AI / ML during Intra-RAT handover based on configuration from the network. But according to this embodiment, the UE (200) will always discard during Inter-RAT handover. In an embodiment, the Inter-RAT handover may be from 5G NR to 4G LTE or from 6G to 5G NR In an embodiment, upon receiving an RRC message for mobility from one RAT to another, the UE (200) discards the logged measurements for AI / ML. In an embodiment, upon successfully completing inter-RAT mobility, such as Mobility from NR to E-UTRA or Mobility from 6G to NR, UE discards the logged measurements for AI / ML. For instance, during the procedure for successful completion of the mobility from NR, upon successfully completing the handover, UE (200) discards any logged measurements for AI / ML. In an embodiment, the logged measurements discarded can be the logged measurements which were not yet fully submitted to lower layers for transmission. In an embodiment, upon performing inter-RAT mobility, the UE (200) considers itself not to be configured for performing logged measurements for AI / ML.

[0145] In an embodiment, upon Inter-RAT mobility failure, such as failure for Mobility from NR to E-UTRA or Mobility from 6G to NR, the UE (200) discards the logged measurements for AI / ML. In an embodiment, the logged measurements discarded can be the logged measurements which were not yet fully submitted to lower layers for transmission. In an embodiment, network node informs the UE whether to release or retain the logged measurements for AI / ML during Inter-RAT handover. Network node may include information such as an enumerated or a flag in RRC message such as MobilityFromNRCommand or RRC message which commands the UE (200) to perform mobility from 6G to 5G NR, and the information informs the UE (200) whether to release or retain the logged measurements for AI / ML during Inter-RAT handover. The UE (200) releases or retains the logged measurements for AI / ML during / after the Inter-RAT handover based on this information. The UE (200) may further release the logged measurements for AI / ML if the UE moves to RRC_IDLE or RRC_INACTIVE in the new RAT or if the UE experiences radio link failure in the new RAT. Alternatively, the UE (200) may still keep the logged measurements for AI / ML during transitioning to RRC_IDLE or RRC_INACTIVE in the new RAT or if the UE experiences radio link failure in the new RAT and may release the same after moving back to old RAT in RRC_IDLE or RRC_INACTIVE.

[0146] FIG. 11 is a flow diagram that illustrates the handling of logged measurements for AI / ML during full configuration according to embodiments as disclosed herein. At step 921, the method includes the UE (200) log the measurements for AI / ML. At step 922, the method includes the UE (200) performs full configuration. At step 923, the method includes the UE (200) discards the logged measurements.

[0147] In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing full configuration. In an embodiment, this discard is performed when the full configuration is for handover (such as reconfigurationWithSync in NR). This simplifies the network's management of logged measurements for AI / ML during full configuration as the target gNB doesn't need to handle the data from the source gNB. This also helps the UE (200) as the UE (200) can perform AI / ML on a clean slate after a full configuration. In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing full configuration if the configuration for logging measurements for AI / ML is not present in the new configuration.

[0148] FIG. 12 is a flow diagram that illustrates the handling of logged measurements for AI / ML when SRB becomes unavailable according to embodiments as disclosed herein. At step 931, the method includes the UE (200) log the measurements for AI / ML. At step 932, the method includes the UE (200) determines that the SRB for sending the logged measurements becomes unavailable. At step 933, the method includes the UE (200) discards the logged measurements.

[0149] In an embodiment, the UE (200) discards logged measurements for AI / ML upon release of the signaling radio bearer (SRB) configuration of the SRB used for sending the logged measurements for AI / ML to the network (SRBx in the background). In an embodiment, the UE (200) discards logged measurements for AI / ML when the signaling radio bearer (SRB) configuration for the specific SRB used for sending the logged measurements for AI / ML (SRBx in the background) to the network becomes unavailable at the UE (200). This may be due to SRB release or due to the absence of SRB configuration in the new configuration for a full configuration or the absence of SRBx configuration in candidate configuration or reference configuration of subsequent LTM or any other type of mobility such as L3 handover or subsequent CPAC. In an embodiment, when the signaling radio bearer (SRB) configuration of the SRB used for sending the logged measurements for AI / ML to the network (SRBx in the background) is not available at the UE (200) and the logged measurements for AI / ML are available at the UE (200) and the UE receives a request to send the logged measurements for AI / ML, the UE (200) sends the logged measurements for AI / ML using SRB2. In NR, the UE (200) sends the UEInformationResponse including logged measurements for AI / ML over SRB2 if SRBx is not available.

[0150] In an embodiment, when SRB configuration for the SRB for sending the logged measurements for AI / ML to the network (SRBx in the background) is not available at the UE (200) and the logged measurements for AI / ML are available at the UE (200) and the UE (200) receives a request to send the logged measurements for AI / ML, UE doesn't send the logged measurements for AI / ML. The UE (200) may keep the logged measurements for AI / ML and send them when the specific SRB used for sending the logged measurements for AI / ML (SRBx in the background) becomes available and the network requests to send the logged measurements for AI / ML. In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing full configuration if the configuration of the SRB (SRBx in the background) for sending the logged measurements for AI / ML is not present in the new configuration.

[0151] In an embodiment, UE discards logged measurements for AI / ML upon performing mobility if the configuration of the SRB (SRBx in the background) for sending the logged measurements for AI / ML is not present in the UE configuration after the handover. In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing LTM cell switch if the configuration of the SRB (SRBx in the background) for sending the logged measurements for AI / ML is not present in the LTM candidate configuration or the reference configuration (if the candidate configuration is not a complete configuration).

[0152] In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing mobility if the configuration for logging measurements for AI / ML is not present in the UE (200) configuration after the handover. In an embodiment, the UE (200) discards logged measurements for AI / ML upon performing LTM cell switch if the configuration for logging measurements for AI / ML is not present in the LTM candidate configuration or the reference configuration (if the candidate configuration is not a complete configuration).

[0153] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and / or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. While the preferred embodiments have been described, those skilled in the art will recognize that modifications can be made within the scope of the described embodiments.

Claims

1.A method performed by a user equipment (UE) in a wireless communication system, the method comprising:receiving, from a source base station, a radio resource control (RRC) message for a handover from the source base station to a target base station;performing logging of channel state information (CSI) measurements for an artificial intelligence / machine learning (AI / ML);identifying whether to retain a logged CSI measurement for the AI / ML based on the RRC message; andtransmitting, to the target base station, a complete message,wherein, based on identification associated with retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained, andwherein, based on identification associated with not retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is discarded.2.The method of claim 1,wherein the RRC message includes information on retaining the logged CSI measurement for the AI / ML.3.The method of claim 1,wherein the RRC message includes an RRC reconfiguration with synchronization, andwherein the complete message is an RRC reconfiguration complete message.4.The method of claim 1,wherein the handover is an intra-radio access technology (RAT) handover.5.A method performed by a target base station in a wireless communication system, the method comprising:receiving, from a source base station, a handover preparation information message including information on retaining a logged channel state information (CSI) measurement for an artificial intelligence / machine learning (AI / ML);transmitting, to a user equipment (UE) via the source base station, a radio resource control (RRC) message for a handover from the source base station to the target base station, the RRC message including the information on retaining the logged CSI measurement for the AI / ML; andreceiving, from the UE, a complete message,wherein, based on the information on retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained.6.The method of claim 5,wherein the RRC message includes an RRC reconfiguration with synchronization, andwherein the complete message is an RRC reconfiguration complete message.7.The method of claim 5,wherein the handover is an intra-radio access technology (RAT) handover.8.A user equipment (UE) in a wireless communication system, the UE comprising:a transceiver;at least one processor coupled to the transceiver; andmemory coupled to the at least one processor and storing instructions executable by the at least one processor,wherein the instructions cause the UE to:receive, from a source base station, a radio resource control (RRC) message for a handover from the source base station to a target base station,perform logging of channel state information (CSI) measurements for an artificial intelligence / machine learning (AI / ML),identify whether to retain a logged CSI measurement for the AI / ML based on the RRC message, andtransmit, to the target base station, a complete message,wherein, based on identification associated with retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained, andwherein, based on identification associated with not retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is discarded.9.The UE of claim 8,wherein the RRC message includes information on retaining the logged CSI measurement for the AI / ML.10.The UE of claim 8,wherein the RRC message includes an RRC reconfiguration with synchronization, andwherein the complete message is an RRC reconfiguration complete message.11.The UE of claim 8,wherein the handover is an intra-radio access technology (RAT) handover.12.A target base station in a wireless communication system, the target base station comprising:a transceiver;at least one processor coupled to the transceiver; andmemory coupled to the at least one processor and storing instructions executable by the at least one processor,wherein the instructions cause the target base station to:receive, from a source base station, a handover preparation information message including information on retaining a logged channel state information (CSI) measurement for an artificial intelligence / machine learning (AI / ML),transmit, to a user equipment (UE) via the source base station, a radio resource control (RRC) message for a handover from the source base station to the target base station, the RRC message including the information on retaining the logged CSI measurement for the AI / ML, andreceive, from the UE, a complete message,wherein, based on the information on retaining the logged CSI measurement for the AI / ML, the logged CSI measurement for the AI / ML is retained.13.The target base station of claim 12,wherein the RRC message includes an RRC reconfiguration with synchronization, andwherein the complete message is an RRC reconfiguration complete message.14.The target base station of claim 12,wherein the handover is an intra-radio access technology (RAT) handover.

Images