User equipment (UE), and method of ue

Abstract

In one implementation, a user equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model is pro-vided. The UE comprises a transceiver circuit to receive from a network node a set of mon-itoring resources and a set of parameters, the set of parameters comprises at least one of: re-sources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The UE further comprises a memory to store the AI / ML model, and a controller to determine performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML mod-el, wherein the performance metric is reported to the network node.

Description

USER EQUIPMENT (UE), AND METHOD OF UE

[0001] The present disclosure generally relates to a user equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-sided artificial intelligence / machine learning (AI / ML) model and a method of the UE.

[0002] In 3GPP RAN#102, Rel-19, there is a work item (WI) on "New WID on Artificial Intelligence (AI) / Machine Learning (ML) for NR Air Interface". The WI objective includes Beam management - DL Tx beam prediction for both UE-sided model and NW-sided model, encompassing [RAN1 / RAN2]. The beam management includes spatial-domain DL Tx beam prediction for Set A of beams based on measurement results of Set B of beams ("BM-Case1"), Temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams ("BM-Case2"). The beam management further includes specifying necessary signalling / mechanism(s) to facilitate LCM operations specific to the Beam Management use cases, if any and enabling method(s) to ensure consistency between training and inference regarding NW-side additional conditions (if identified) for inference at user equipment (UE).

[0003] The following presents a simplified summary of the subject matter / disclosure in order to provide a basic understanding of some of the aspects of subject matter embodiments. This summary is not an extensive overview of the subject matter. It is not intended to identify key / critical elements of the embodiments or to delineate the scope of the subject matter. Its sole purpose is to present some concepts of the subject matter in a simplified form as a prelude to the more detailed description that is presented later.

[0004] In one implementation, a user equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model is disclosed. The UE comprises a transceiver circuit configured to receive from a network node: a set of monitoring resources, and a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The UE further comprises a memory configured to store the AI / ML model, and a controller coupled to the memory and configured to determine performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model, wherein the transceiver circuit is configured to report the performance metric to the network node.

[0005] In another implementation, a method for a User Equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model is disclosed. The method comprises receiving from a network node: a set of monitoring resources and a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The method further comprises storing, in a memory, the AI / ML model, determining, by a controller, a performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model, and reporting the performance metric to the network node.

[0006] The details of one or more examples are set forth in the accompanying drawings and the description below. Other features, objects, and advantages will be apparent from the description, drawings, and claims.

[0007] The foregoing and further objects, features, and advantages of the present subject matter will become apparent from the following description of exemplary embodiments with reference to the accompanying drawings, wherein like numerals are used to represent like elements.

[0008] It is to be noted, however, that the appended drawings illustrate only typical embodiments of the present subject matter, and are, therefore, not to be considered for limiting of its scope, for the subject matter may admit to other equally effective embodiments.

[0009] For a better understanding of the present disclosure, reference is made to the following description of an exemplary embodiment thereof, considered in conjunction with the accompanying drawings, in which:Fig. 1 schematically illustrates a telecommunication system , in accordance with an embodiment of the present disclosure.Fig. 2 illustrates a block diagram illustrating main components of a user equipment (UE), in accordance with an embodiment of the present disclosure.Fig. 3 illustrates a block diagram illustrating main components of an exemplary (R)AN node, in accordance with an embodiment of the present disclosure.Fig. 4 illustrates a (R)AN node based on O-RAN architecture in accordance with an embodiment of the present disclosure.Fig. 5 illustrates a block diagram of a radio unit, in accordance with an embodiment of the present disclosure.Fig. 6 illustrates a block diagram of a distributed unit, in accordance with one embodiment of the present disclosure.Fig. 7 illustrates a block diagram of a centralized unit, in accordance with an embodiment of the present disclosure.Fig. 8 illustrates a block diagram illustrating the main components of Access Management Function (AMF), in accordance with an embodiment of the present disclosure.Fig. 9 illustrates a representation of N distinct future prediction time instances, in accordance with an embodiment of the present disclosure.Fig. 10 illustrates a representation of N equal future prediction time instances, in accordance with an embodiment of the present disclosure.Fig. 11 illustrates a timing diagram representing timing for performance metric determination, in accordance with an embodiment of the present disclosure.Fig. 12 illustrates a timing diagram illustrating duration of the predicted N future time instances and the differential RSRP report among multiple beams over multiple time instances, in accordance with an embodiment of the present disclosure.Fig. 13 illustrates a flowchart of a method performed by a user equipment, in accordance with an embodiment of the present disclosure.Fig. 14 illustrates a flowchart of a method performed by a network node, in accordance with an embodiment of the present disclosure.Fig. 15 illustrates a block diagram of the user equipment (UE) in accordance with another embodiment of the present disclosure.

[0010] Although specific features of various embodiments may be shown in some drawings and not in others, this is for convenience only. Any feature of any drawing may be referenced and / or claimed in combination with any feature of any other drawing.

[0011] The present disclosure generally relates to wireless communication systems and more particularly relates to techniques for effective AI / ML based beam management and performance monitoring utilizing UE-sided models.

[0012] This disclosure is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The disclosure is capable of other embodiments and of being practiced or of being carried out in various ways. Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having," "containing," "involving," and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

[0013] Various aspects of the proposed apparatus and method are described fully hereinafter with reference to the accompanying drawings. The same reference numbers in different drawings may identify the same or similar elements. The teachings disclosed may, however, be embodied in many different models with variations and should not be construed as limited to any specific structure or function presented throughout this disclosure. Rather, these aspects are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be understood that any aspect disclosed herein may be embodied by one or more elements of a claim, and also that the following detailed description does not limit the claims.

[0014] Also, all logical units described and depicted in the figures include the software and / or hardware components required for the unit to function. Further, each unit may comprise within itself one or more components which are implicitly understood. These components may be operatively coupled to each other and be configured to communicate with each other to perform the function of the said unit.

[0015] In modern wireless communication systems, beam management is a technique used to establish a communication between a user equipment (UE) and base station (for example, gNB). The User Equipment (UE) selects the best beam to establish and maintain a robust connection with the base station (gNB) by evaluating signal quality metrics. To establish or maintain connection (e.g., to maintain optimal connection), the gNB transmits synchronization signals (e.g., synchronization signal blocks (SSBs) , SS / PBCH block etc.) and response signals (or reference signals). The UE measures the signal quality of these beams, typically using metrics such as Reference Signal Received Power (RSRP) and Signal to Interference and noise ratio (SINR) and identifies the beam with the highest quality. After measurement, the UE sends a beam quality report to the gNB, which uses this information to confirm or refine the selected beam. In scenarios of dynamic mobility or fluctuating signal conditions, the process of beam selection and switching becomes critical to ensure uninterrupted connectivity and optimal performance between the UE and the gNB. This particularly is helpful in scenarios where the UE is constantly moving, and an optimal connection is required to be established between the UE and the gNB. To establish initial connection, the UE may select an initial beam to camp on.

[0016] To further enhance the efficiency of beam management, artificial intelligence / Machine learning AI / ML models are being integrated into the UE. These models enable the UE to predict and select the best beams out of the plurality of beams by leveraging historical data, mobility patterns, and environmental context. Such AI / ML-driven predictions reduce the need for exhaustive beam measurements and speed up the decision-making process, particularly in dense urban environments or high-mobility scenarios. The integration of AI / ML models significantly improves beam selection accuracy, reduces beam switching delays, and ensures a more stable connection, enhancing the overall user experience.

[0017] The AI / ML models generates inference report specifying configuration of UE with N future time instance(s) for inference by network (NW). As explained above, the downlink transmit beam prediction for both UE-sided model and the NW-sided model can be performed in two ways- spatial-domain (BM-case 1) and temporal domain (BM-case2). The BM-case 2 involves transmit beam prediction based on historic measurement results of a subset of beams (set B beams). In other words, BM-case2 deals with predicting the beams for N future time instances based on the historical measurement results of the subset of beams (set B beams). The present disclosure deals with the issue of determining duration of the predicted N future time instances, i.e., the time instances at which each prediction happens in between two actual Set B measurements.

[0018] Further, the present disclosure deals with the issue of providing differential RSRP report among multiple beams over multiple time instances. For BM-case 2, the present disclosure provides solution for reporting of the RSRP values of over all the N predicted time instances by providing quantized reporting mechanism.

[0019] If the network triggers performance metric report request, the UE needs to measure the performance metric for the AI / ML Model. For BM-case 2, the AI / ML model predicts beams in N future time instances and hence performance metric is required to be defined for all the predicted future time instances. Further, the UE needs to measure the monitoring resources to determine the performance metric. For determining the performance metric, the network may or may not configure the full set of beams on monitoring resources.

[0020] This disclosure may provide the UE, network node(s), a method of the UE and / or a method of the network node(s) etc. to solve at least one of problems mentioned in this disclosure. For example, this disclosure may provide the UE, network node(s), a method of the UE and / or a method of the network node(s) etc. to solve at least problem that considering the WI in RAN#102, Rel-19, TR 38.843, and / or 3GPP RAN1 Agreement (e.g., RAN1#118b Agreement), it may need to consider at.

[0021] In TR 38.843, for BM-Case1 and BM-Case2 with a UE-side AI / ML model, two types (Type-1 and Type-2) of performance monitoring are defined. Type 1 performance monitoring includes configuration / signalling from gNB to UE for measurement and / or reporting. The UE may have different operations: Option 1 (NW-side performance monitoring) and Option 2 (UE-assisted performance monitoring). Under option 1, UE sends reporting to network (NW) (e.g., for the calculation of performance metric at NW) and under Option 2 (UE-assisted performance monitoring), UE calculates performance metric(s), either reports it to NW or reports an event to NW based on the performance metric(s).

[0022] Type 1 performance monitoring also includes indication from NW for UE to perform life cycle management (LCM) operations. Here at least the performance and reporting overhead of model monitoring mechanism is considered. Type 2 performance monitoring includes Indication / request / report from UE to gNB for performance monitoring, configuration / Signalling from gNB to UE for performance monitoring measurement and / or reporting. If it is for UE side model monitoring, the UE makes decision(s) of model selection / activation / deactivation / switching / fallback operation. In TR 38.843, mechanisms that facilitates the UE to detect whether the functionality / model is suitable or no longer suitable is also provided.

[0023] Fig. 1 schematically illustrates a telecommunication system 1 for a mobile (cellular or wireless) to which example embodiments disclosed herein and / or the above aspects are applicable. The telecommunication system 1 represents a system overview in which an end-to-end communication is possible. For example, UE (or user equipment, 'mobile device') communicates with other UEs or service servers in the data network 20 via respective (R)AN (Radio Access Network ) nodes 5 and a core network 7. The (R)AN node 5 supports any radio accesses including a 5G radio access technology (RAT), an E-UTRA (Evolved Universal Terrestrial Radio Access ) Radio Access Technology (RAT), a beyond 5G RAT, a 6G RAT and non-3GPP RAT including wireless local area network (WLAN) technology as defined by the Institute of Electrical and Electronics Engineers (IEEE).

[0024] he (R)AN node 5 may split into a Radio Unit (RU), Distributed Unit (DU) and Centralized Unit (CU). In some aspects, each of the units may be connected to each other and structure the (R)AN node 5 by adopting an architecture as defined by the Open RAN (O-RAN) Alliance, where the units above are referred to as O-RU, O-DU and O-CU respectively. The (R)AN node 5 may be split into control plane function and user plane function. Further, multiple user plane functions can be allocated to support a communication. In some aspects, user traffic may be distributed to multiple user plane functions and user traffic over each user plane functions are aggregated in both the UE 3 and the (R)AN node 5. This split architecture may be called as 'dual connectivity' or 'Multi connectivity'. The (R)AN node 5 can also support a communication using the satellite access. In some aspects, the (R)AN node 5 may support a satellite access and a terrestrial access. In addition, the (R)AN node 5 can also be referred as an access node for a non-wireless access. The non-wireless access includes a fixed line access as defined by the Broadband Forum (BBF) and an optical access as defined by the Innovative Optical and Wireless Network (IOWN).

[0025] The core network 7 may include logical nodes (or 'functions') for supporting a communication in the telecommunication system 1. For example, the core network 7 may be 5G Core Network (5GC) that includes, amongst other functions, control plane functions and user plane functions. Each function in a logical nodes can be considered as a network function. The network function may be provided to another node by adapting the Service Based Architecture (SBA). A Network Function can be deployed as distributed, redundant, stateless, and scalable that provides the services from several locations and several execution instances in each location by adapting the network virtualization technology as defined by the European Telecommunications Standards Institute, Network Functions Virtualization (ETSI NFV). The core network 7 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN).

[0026] As is well known, a UE 3 may enter and leave the areas (i.e. radio cells) served by the (R)AN node 5 as the UE 3 is moving around in the geographical area covered by the telecommunication system 1. In order to keep track of the UE 3 and to facilitate movement between the different (R)AN nodes 5, the core network 7 comprises at least one access and mobility management function (AMF) 70. The AMF 70 is in communication with the (R)AN node 5 coupled to the core network 7. In some core networks, a mobility management entity (MME) or a mobility management node for beyond 5G or a mobility management node for 6G may be used instead of the AMF 70.

[0027] The core network 7 also includes, amongst others, a Session Management Function (SMF) 71, a User Plane Function (UPF) 72, a Policy Control Function (PCF) 73, a Network Exposure Function (NEF) 74, a Unified Data Management (UDM) 75, a Network Data Analytics Function (NWDAF) 76, a Network Slice Selection Function (NSSF) 77 and a Network Repository Function (NRF) 78. When the UE 3 is roaming to a visited Public Land Mobile Network (VPLMN), a home Public Land Mobile Network (HPLMN) of the UE 3 provides the UDM 75 and at least some of the functionalities of the SMF 71, UPF 72, and PCF 73 for the roaming-out UE 3.

[0028] The UE 3 and a respective serving (R)AN node 5 are connected via an appropriate air interface (for example the so-called "Uu" interface and / or the like). Neighboring (R)AN node 5 are connected to each other via an appropriate (R)AN node 5 to (R)AN node interface (such as the so-called "Xn" interface and / or the like). Each (R)AN node 5 is also connected to nodes in the core network 7 (such as the so-called core network nodes) via an appropriate interface (such as the so-called "N2" / "N3" interface(s) and / or the like). From the core network 7, connection to a data network 20 is also provided. The data network 20 can be an internet, a public network, an external network, a private network or an internal network of the PLMN. In case that the data network 20 is provided by a PLMN operator or Mobile Virtual Network Operator (MVNO), the IP Multimedia Subsystem (IMS) service may be provided by that data network 20. The UE 3 can be connected to the data network 20 using IPv4, IPv6, IPv4v6, Ethernet or unstructured data type. The "Uu" interface may include a Control plane of Uu interface and User plane of Uu interface.

[0029] The User plane of Uu interface is responsible to convey user traffic between the UE 3 and a serving (R)AN node 5. The User plane of Uu interface may have a layered structure with SDAP, PDCP, RLC and MAC sublayer over the physical connection. The Control plane of Uu interface is responsible to establish, modify and release a connection between the UE 3 and a serving (R)AN node 5. The Control plane of Uu interface may have a layered structure with RRC, PDCP, RLC and MAC sublayers over the physical connection.

[0030] For example, the following messages are communicated over the RRC layer to support AS signaling. - 　　RRC Setup Request message: This message is sent from the UE 3 to the (R)AN node 5. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be included together in the RRC Setup Request message. 　　>> establishmentCause and ue-Identity. The ue-Identity may have a value of ng-5G-S-TMSI-Part1 or randomValue. -　　RRC Setup message: This message is sent from the (R)AN node 5 to the UE 3. In addi-tion to the parameters that are disclosed by Aspects in this disclosure, following parame-ters may be included together in the RRC Setup message. 　　>> masterCellGroup and radioBearerConfig. -　　RRC Setup Complete message: This message is sent from the UE 3 to the (R)AN node 5. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be included together in the RRC Setup Complete message. 　　>> guami-Type, iab-NodeIndication, idleMeasAvailable, mobilityState, ng-5G-S-TMSI-Part2, registeredAMF, selectedPLMN-Identity.

[0031] The UE 3 and the AMF 70 are connected via an appropriate interface (for example the so-called N1 interface and / or the like). The N1 interface is responsible to provide a communication between the UE 3 and the AMF 70 to support NAS signaling. The N1 interface may be established over a 3GPP access and over a non-3GPP access. For example, the following messages are communicated over the N1 interface. -　　Registration Request message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be included together in the Registration Request message. 　　>> 5GS registration type, ngKSI, 5GS mobile identity, Non-current native NAS key set identifier, 5GMM capability, UE security capability, Requested NSSAI, Last visited registered TAI, S1 UE network capability, Uplink data status, PDU session status, MICO indication, UE status, Additional GUTI, Allowed PDU session status, UE's usage setting, Requested DRX parameters, EPS NAS message container, LADN in-dication, Payload container type, Payload container, Network slicing indication, 5GS update type, Mobile station classmark 2, Supported codecs, NAS message container, EPS bearer context status, Requested extended DRX parameters, T3324 value, UE radio capability ID, Requested mapped NSSAI, Additional information requested, Requested WUS assistance information, N5GC indication and Requested NB-N1 mode DRX parameters. -　　Registration Accept message: This message is sent from the AMF 70 to the UE 3. In ad-dition to the parameters that are disclosed by Aspects in this disclosure, following param-eters may be included together in the Registration Accept message. 　　>> 5GS registration result, 5G-GUTI, Equivalent PLMNs, TAI list, Allowed NSSAI, Rejected NSSAI, configure NSSAI, 5GS network feature support, PDU session sta-tus, PDU session reactivation result, PDU session reactivation result error cause, LADN information, MICO indication, Network slicing indication, Service area list, T3512 value, Non-3GPP de-registration timer value, T3502 value, Emergency num-ber list, Extended emergency number list, SOR transparent container, EAP message, NSSAI inclusion mode, Operator-defined access category definitions, Negotiated DRX parameters, Non-3GPP NW policies, EPS bearer context status, Negotiated ex-tended DRX parameters, T3447 value, T3448 value, T3324 value, UE radio capabil-ity ID, UE radio capability ID deletion indication, Pending NSSAI, Ciphering key data, CAG information list, Truncated 5G-S-TMSI configuration, Negotiated WUS assistance information, Negotiated NB-N1 mode DRX parameters and Extended re-jected NSSAI. -　　Registration Complete message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be included together in the Registration Complete message. 　　>> SOR transparent container. -　　Authentication Request message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be included together in the Authentication Request message. 　　>> ngKSI,ABBA, Authentication parameter RAND (5G authentication challenge), Au-thentication parameter AUTN (5G authentication challenge) and EAP message. -　　Authentication Response message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be populated together in the Authentication Response message. 　　>> Authentication response message identity, Authentication response parameter and EAP message. -　　Authentication Result message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be populated together in the Authentication Result message. 　　>> ngKSI, EAP message and ABBA. -　　Authentication Failure message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be populated together in the Authentication Failure message. 　　>> Authentication failure message identity, 5GMM cause and Authentication failure pa-rameter. -　　Authentication Reject message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, following pa-rameters may be populated together in the Authentication Reject message. 　　>> EAP message. -　　Service Request message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be populated together in the Service Request message. 　　>> ngKSI,Service type, 5G-S-TMSI, Uplink data status, PDU session status, Allowed PDU session status, NAS message container. -　　Service Accept message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be populated together in the Service Accept message. 　　>> PDU session status, PDU session reactivation result, PDU session reactivation result error cause, EAP message and T3448 value. -　　Service Reject message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be populated together in the Service Reject message. 　　>> 5GMM cause, PDU session status, T3346 value, EAP message, T3448 value and CAG information list. -　　Configuration Update Command message: This message is sent from the AMF 70 to the UE 3. In addition to the parameters that are disclosed by Aspects in this disclosure, fol-lowing parameters may be populated together in the Configuration Update Command message. 　　>> Configuration update indication,5G-GUTI, TAI list, Allowed NSSAI, Service area list, Full name for network, Short name for network, Local time zone, Universal time and local time zone, Network daylight saving time, LADN information, MICO indi-cation, Network slicing indication, configured NSSAI, Rejected NSSAI, Operator-defined access category definitions, SMS indication, T3447 value, CAG information list, UE radio capability ID, UE radio capability ID deletion indication, 5GS registra-tion result, Truncated 5G-S-TMSI configuration, Additional configuration indication and Extended rejected NSSAI. -　　Configuration Update Complete message: This message is sent from the UE 3 to the AMF 70. In addition to the parameters that are disclosed by Aspects in this disclosure, following parameters may be populated together in the Configuration Update Complete message. 　　>> Configuration update complete message identity.

[0032] Fig. 2 is a block diagram illustrating the main components of the UE 3 (mobile device 3). As shown, the UE 3 includes a transceiver circuit 31 which is operable to transmit signals to and to receive signals from the connected node(s) via one or more antennas 32. Further, the UE 3 may include a user interface 34 for inputting information from outside or outputting information to outside. Although not necessarily shown in the figure, the UE 3 may have all the usual functionality of a conventional mobile device and this may be provided by any one or any combination of hardware, software and firmware, as appropriate. Software may be pre-installed in the memory and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. A controller 33 controls the operation of the UE 3 in accordance with software stored in a memory 36. The software includes, among other things, an operating system 361 and a communications control module 362 having at least a transceiver control module 3621. The communications control module 362 (using its transceiver control module 3621) is responsible for handling (generating / sending / receiving) signalling and uplink / downlink data packets between the UE 3 and other nodes, such as the (R)AN node 5 and the AMF 10. Such signalling may include, for example, appropriately formatted signalling messages (e.g. a registration request message and associated response messages) relating to access and mobility management procedures (for the UE 3). The controller 33 interworks with one or more Universal Subscriber Identity Module (USIM) 35. If there are multiple USIMs 35 equipped, the controller 33 may activate only one USIM 35 or may activate multiple USIMs 35 at the same time.

[0033] The UE 3 may, for example, support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN).

[0034] The UE 3 may, for example, be an item of equipment for production or manufacture and / or an item of energy related machinery (for example equipment or machinery such as: boilers; engines; turbines; solar panels; wind turbines; hydroelectric generators; thermal power generators; nuclear electricity generators; batteries; nuclear systems and / or associated equipment; heavy electrical machinery; pumps including vacuum pumps; compressors; fans; blowers; oil hydraulic equipment; pneumatic equipment; metal working machinery; manipulators; robots and / or their application systems; tools; molds or dies; rolls; conveying equipment; elevating equipment; materials handling equipment; textile machinery; sewing machines; printing and / or related machinery; paper converting machinery; chemical machinery; mining and / or construction machinery and / or related equipment; machinery and / or implements for agriculture, forestry and / or fisheries; safety and / or environment preservation equipment; tractors; precision bearings; chains; gears; power transmission equipment; lubricating equipment; valves; pipe fittings; and / or application systems for any of the previously mentioned equipment or machinery etc.).

[0035] The UE 3 may, for example, be an item of transport equipment (for example transport equipment such as: rolling stocks; motor vehicles; motorcycles; bicycles; trains; buses; carts; rickshaws; ships and other watercraft; aircraft; rockets; satellites; drones; balloons etc.). The UE 3 may, for example, be an item of information and communication equipment (for example information and communication equipment such as: electronic computer and related equipment; communication and related equipment; electronic components etc.). The UE 3 may, for example, be a refrigerating machine, a refrigerating machine applied product, an item of trade and / or service industry equipment, a vending machine, an automatic service machine, an office machine or equipment, a consumer electronic and electronic appliance (for example a consumer electronic appliance such as: audio equipment; video equipment; a loud speaker; a radio; a television; a microwave oven; a rice cooker; a coffee machine; a dishwasher; a washing machine; a dryer; an electronic fan or related appliance; a cleaner etc.).

[0036] The UE 3 may, for example, be an electrical application system or equipment (for example an electrical application system or equipment such as: an x-ray system; a particle accelerator; radio isotope equipment; sonic equipment; electromagnetic application equipment; electronic power application equipment etc.). The UE 3 may, for example, be an electronic lamp, a luminaire, a measuring instrument, an analyzer, a tester, or a surveying or sensing instrument (for example a surveying or sensing instrument such as: a smoke alarm; a human alarm sensor; a motion sensor; a wireless tag etc.), a watch or clock, a laboratory instrument, optical apparatus, medical equipment and / or system, a weapon, an item of cutlery, a hand tool, or the like.

[0037] The UE 3 may, for example, be a wireless-equipped personal digital assistant or related equipment (such as a wireless card or module designed for attachment to or for insertion into another electronic device (for example a personal computer, electrical measuring machine)). The UE 3 may be a device or a part of a system that provides applications, services, and solutions described below, as to "internet of things (IoT)", using a variety of wired and / or wireless communication technologies. Internet of Things devices (or "things") may be equipped with appropriate electronics, software, sensors, network connectivity, and / or the like, which enable these devices to collect and exchange data with each other and with other communication devices. IoT devices may comprise automated equipment that follow software instructions stored in an internal memory. IoT devices may operate without requiring human supervision or interaction. IoT devices might also remain stationary and / or inactive for a long period of time. IoT devices may be implemented as a part of a (generally) stationary apparatus. IoT devices may also be embedded in non-stationary apparatus (e.g. vehicles) or attached to animals or persons to be monitored / tracked.

[0038] It will be appreciated that IoT technology can be implemented on any communication devices that can connect to a communications network for sending / receiving data, regardless of whether such communication devices are controlled by human input or software instructions stored in memory.

[0039] It will be appreciated that IoT devices are sometimes also referred to as Machine-Type Communication (MTC) devices or Machine-to-Machine (M2M) communication devices or Narrow Band-IoT UE (NB-IoT UE). It will be appreciated that a UE 3 may support one or more IoT or MTC applications.

[0040] The UE 3 may be a smart phone or a wearable device (e.g. smart glasses, a smart watch, a smart ring, or a hearable device). The UE 3 may be a car, or a connected car, or an autonomous car, or a vehicle device, or a motorcycle or V2X (Vehicle to Everything) communication module (e.g. Vehicle to Vehicle communication module, Vehicle to Infrastructure communication module, Vehicle to People communication module and Vehicle to Network communication module).

[0041] The memory 36 present in the UE 3 may store AI / ML models for producing inference results. The AI / ML models are configured to perform performance monitoring based on a set of parameters and inference results produced by the AI / ML models. With UE- side AI / ML model, there are two different ways of performing performance monitoring. One is network side-performance monitoring where the UE 3 sends a report to the network (NW) (for the calculation of performance metric at the NW). Another way is of performance monitoring is UE-assisted performance metric where the UE calculates performance metrics and reports it to NW either by self or upon a trigger by the NW.

[0042] For performance monitoring, performance metric is calculated based on at least one parameter received from the network node. The parameters may include, but are not limited to, resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. A set of parameters are received from the network node by the UE 3. The UE 3 also receives a set of monitoring resources from the network node. The performance metric is determined for the set of monitoring resources based on the set of received parameters and the inference results from the AI / ML models. Further, the performance metric is determined for the number of prediction time instances (N). Once the performance metric is determined, the performance metric is reported to the network node and the reporting the performance metric can be the average of T performance metrics for T prediction instances. The T prediction instances will be explained further in following disclosure.

[0043] Fig. 3 is a block diagram illustrating the main components of an exemplary (R)AN node 5, also called, network node, for example a base station ('eNB' in LTE, 'gNB' in 5G, a base station for 5G beyond, a base station for 6G). As shown, the (R)AN node 5 includes a transceiver circuit 51 which is operable to transmit signals to and to receive signals from connected UE(s) 3 via one or more antennas 52 and to transmit signals to and to receive signals from other network nodes (either directly or indirectly) via a network interface 53. A controller 54 controls the operation of the (R)AN node 5 in accordance with software stored in a memory 55. Software may be pre-installed in the memory and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. The software includes, among other things, an operating system 551 and a communications control module 552 having at least a transceiver control module 5521. In one embodiment, the performance metric is calculated by the UE 3 upon receipt of a trigger report sent by the network node 6.

[0044] The communications control module 552 (using its transceiver control sub-module) is responsible for handling (generating / sending / receiving) signalling between the (R)AN node 5 and other nodes, such as the UE 3, another (R)AN node 5, the AMF 70 and the UPF 72 (e.g. directly or indirectly). The signalling may include, for example, appropriately formatted signalling messages relating to a radio connection and a connection with the core network 7 (for a particular UE 3), and in particular, relating to connection establishment and maintenance (e.g. RRC connection establishment and other RRC messages), NG Application Protocol (NGAP) messages (i.e. messages by N2 reference point) and Xn application protocol (XnAP) messages (i.e. messages by Xn reference point), etc. Such signalling may also include, for example, broadcast information (e.g. Master Information and System information) in a sending case.

[0045] The controller 54 is also configured (by software or hardware) to handle related tasks such as, when implemented, UE mobility estimates and / or moving trajectory estimation. The (R)AN node 5 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN).

[0046] Fig. 4 schematically illustrates a (R)AN node 5 based on O-RAN architecture to which the (R)AN node 5 aspects are applicable. The (R)AN node 5 based on O-RAN architecture represents a system overview in which the (R)AN node is split into a Radio Unit (RU) 60, Distributed Unit (DU) 61 and Centralized Unit (CU) 62. In some aspects, each unit may be combined. For example, the RU 60 can be integrated / combined with the DU 61 as an integrated / combined unit, the DU 61 can be integrated / combined with the CU 62 as another integrated / combined unit. Any functionality in the description for a unit (e.g. one of RU 60, DU 61 and CU 62) can be implemented in the integrated / combined unit above. Further, CU 62 can separate into two functional units such as CU Control plane (CP) and CU User plane (UP). The CU CP has a control plane functionality in the (R)AN node 5. The CU UP has a user plane functionality in the (R)AN node 5. Each CU CP is connected to the CU UP via an appropriate interface (such as the so-called "E1" interface and / or the like).

[0047] The UE 3 and a respective serving RU 60 are connected via an appropriate air interface (for example the so-called "Uu" interface and / or the like). Each RU 60 is connected to the DU 61 via an appropriate interface (such as the so-called "Front haul", "Open Front haul", "F1" interface and / or the like). Each DU 61 is connected to the CU 62 via an appropriate interface (such as the so-called "Mid haul", "Open Mid haul", "E2" interface and / or the like). Each CU 62 is also connected to nodes in the core network 7 (such as the so-called core network nodes) via an appropriate interface (such as the so-called "Back haul", "Open Back haul", "N2" / "N3" interface(s) and / or the like). In addition, a user plane part of the DU 61 can also be connected to the core network nodes 7 via an appropriate interface (such as the so-called "N3" interface(s) and / or the like). Depending on functionality split among the RU 60, DU 61 and CU 62, each unit provides some of the functionality that is provided by the (R)AN node 5. For example, the RU 60 may provide a functionalities to communicate with a UE 3 over air interface, the DU 61 may provide functionalities to support MAC layer and RLC layer, the CU 62 may provide functionalities to support PDCP layer, SDAP layer and RRC layer.

[0048] Fig. 5 is a block diagram illustrating the main components of an exemplary RU 60, for example a RU part of base station ('eNB' in LTE, 'gNB' in 5G, a base station for 5G beyond, a base station for 6G). As shown, the RU 60 includes a transceiver circuit 601 which is operable to transmit signals to and to receive signals from connected UE(s) 3 via one or more antennas 602 and to transmit signals to and to receive signals from other network nodes or network unit (either directly or indirectly) via a network interface 603. A controller 604 controls the operation of the RU 60 in accordance with software stored in a memory 605. Software may be pre-installed in the memory and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. The software includes, among other things, an operating system 6051 and a communications control module 6052 having at least a transceiver control module 60521.

[0049] The communications control module 6052 (using its transceiver control sub-module) is responsible for handling (generating / sending / receiving) signalling between the RU 60 and other nodes or units, such as the UE 3, another RU 60 and DU 61 (e.g. directly or indirectly). The signalling may include, for example, appropriately formatted signalling messages relating to a radio connection and a connection with the RU 60 (for a particular UE 3), and in particular, relating to MAC layer and RLC layer.

[0050] The controller 604 is also configured (by software or hardware) to handle related tasks such as, when implemented, UE mobility estimation and / or moving trajectory estimation. The RU 60 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN). As described above, the RU 60 can be integrated / combined with the DU 61 as an integrated / combined unit. Any functionality in the description for the RU 60 can be implemented in the integrated / combined unit above.

[0051] Fig. 6 is a block diagram illustrating the main components of an exemplary DU 61, for example a DU part of a base station ('eNB' in LTE, gNB' in 5G, a base station for 5G beyond, a base station for 6G). As shown, the apparatus includes a transceiver circuit 611 which is operable to transmit signals to and to receive signals from other nodes or units (including the RU 60) via a network interface 612. A controller 613 controls the operation of the DU 61 in accordance with software stored in a memory 614. Software may be pre-installed in the memory 614 and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. The software includes, among other things, an operating system 6141 and a communications control module 6142 having at least a transceiver control module 61421. The communications control module 6142 (using its transceiver control module 61421 is responsible for handling (generating / sending / receiving) signalling between the DU 61 and other nodes or units, such as the RU 60 and other nodes and units.

[0052] The DU 61 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN). As described above, the RU 60 can be integrated / combined with the DU 61 or CU 62 as an integrated / combined unit. Any functionality in the description for DU 61 can be implemented in one of the integrated / combined unit above.

[0053] Fig. 7 is a block diagram illustrating the main components of an exemplary CU 62, for example a CU part of base station ('eNB' in LTE, gNB' in 5G, a base station for 5G beyond, a base station for 6G). As shown, the apparatus includes a transceiver circuit 621 which is operable to transmit signals to and to receive signals from other nodes or units (including the DU 61) via a network interface 622. A controller 623 controls the operation of the CU 62 in accordance with software stored in a memory 624. Software may be pre-installed in the memory 624 and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. The software includes, among other things, an operating system 6241 and a communications control module 6242 having at least a transceiver control module 62421. The communications control module 6242 (using its transceiver control module 62421 is responsible for handling (generating / sending / receiving) signalling between the CU 62 and other nodes or units, such as the DU 61 and other nodes and units. The CU 62 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN).

[0054] As described above, the CU 62 can be integrated / combined with the DU 61 as an integrated / combined unit. Any functionality in the description for the CU 62 can be implemented in the integrated / combined unit above.

[0055] Fig. 8 is a block diagram illustrating the main components of the AMF 70. As shown, the apparatus includes a transceiver circuit 701 which is operable to transmit signals to and to receive signals from other nodes (including the UE 3) via a network interface 702. A controller 703 controls the operation of the AMF 70 in accordance with software stored in a memory 704. Software may be pre-installed in the memory 704 and / or may be downloaded via the telecommunication network or from a removable data storage device (RMD), for example. The software includes, among other things, an operating system 7041 and a communications control module 7042 having at least a transceiver control module 70421. The communications control module 7042 (using its transceiver control module 70421 is responsible for handling (generating / sending / receiving) signalling between the AMF 70 and other nodes, such as the UE 3 (e.g. via the (R)AN node 5) and other core network nodes (including core network nodes in the HPLMN of the UE 3 when the UE 3 is roaming-in. Such signalling may include, for example, appropriately formatted signalling messages (e.g. a registration request message and associated response messages) relating to access and mobility management procedures (for the UE 3). The AMF 70 may support the Non-Public Network (NPN). The NPN may be a Stand-alone Non-Public Network (SNPN) or a Public Network Integrated NPN (PNI-NPN).

[0056] Detailed aspects have been described above. As those skilled in the art will appreciate, a number of modifications and alternatives can be made to the above aspects whilst still benefiting from the disclosures embodied therein. By way of illustration only a number of these alternatives and modifications will now be described.

[0057] In the above description, the UE 3 and the network apparatus are described for ease of understanding as having a number of discrete modules (such as the communication control modules). Whilst these modules may be provided in this way for certain applications, for example where an existing system has been modified to implement the disclosure, in other applications, for example in systems designed with the inventive features in mind from the outset, these modules may be built into the overall operating system or code and so these modules may not be discernible as discrete entities. These modules may also be implemented in software, hardware, firmware or a mix of these.

[0058] Each controller may comprise any suitable form of processing circuitry including (but not limited to), for example: one or more hardware implemented computer processors; microprocessors; central processing units (CPUs); arithmetic logic units (ALUs); input / output (IO) circuits; internal memories / caches (program and / or data); processing registers; communication buses (e.g. control, data and / or address buses); direct memory access (DMA) functions; hardware or software implemented counters, pointers and / or timers; and / or the like.

[0059] In the above aspects, a number of software modules were described. As those skilled in the art will appreciate, the software modules may be provided in compiled or un-compiled form and may be supplied to the UE 3 and the network apparatus as a signal over a computer network, or on a recording medium. Further, the functionality performed by part or all of this software may be performed using one or more dedicated hardware circuits.

[0060] In the above aspects, a 3GPP radio communications (radio access) technology is used. However, any other radio communications technology (e.g. WLAN, Wi-Fi, WiMAX, Bluetooth, etc.) and other fix line communications technology (e.g. BBF Access, Cable Access, optical access, etc.) may also be used in accordance with the above aspects. Items of user equipment might include, for example, communication devices such as mobile telephones, smartphones, user equipment, personal digital assistants, laptop / tablet computers, web browsers, e-book readers and / or the like. Such mobile (or even generally stationary) devices are typically operated by a user, although it is also possible to connect so-called 'Internet of Things' (IoT) devices and similar machine-type communication (MTC) devices to the network. For simplicity, the present application refers to mobile devices (or UEs) in the description but it will be appreciated that the technology described can be implemented on any communication devices (mobile and / or generally stationary) that can connect to a communications network for sending / receiving data, regardless of whether such communication devices are controlled by human input or software instructions stored in memory. Various other modifications will be apparent to those skilled in the art and will not be described in further detail here.

[0061] While the disclosure has been particularly shown and described with reference to exemplary aspects thereof, the disclosure is not limited to these aspects. It will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present disclosure as defined by this document. For example, the aspects above are not limited to 5GS, and the Aspects are also applicable to communication systems other than 5GS.

[0062] Considering the WI in RAN#102, Rel-19, TR 38.843, and / or 3GPP RAN1 Agreement (e.g., 3GPP RAN1#118b Agreement), this disclosure solves issues related to determination of performance metrics, duration of predicted N future time instances and differential RSRP report among multiple beams over multiple time instances. For example, the disclosure can solve a problem that considering the WI in RAN#102, Rel-19, TR 38.843, and / or 3GPP RAN1 Agreement (e.g., 3GPP RAN1#118b Agreement), it is not clear in current 3GPP specification about at least one of the determination of performance metrics, duration of predicted N future time instances and differential RSRP report among multiple beams over multiple time instances.

[0063] As explained above, to ensure beam management and overall network performance, performance monitoring is used. Performance monitoring involves the continuous assessment of key metrics related to signal quality and network such as RSRP, SINR, RSRQ, latency, and throughput. For performance monitoring, performance metric and beam prediction indicator are defined. Definition of the performance metric and the beam prediction indicator depends on various factors, such as RSRP, SINR, RSRQ, Top-K beams, etc. Since the UE evaluates the quality of each beam by using parameters such as RSRP, SINR, RSRQ, latency and throughput, each of the beam can be ranked based on the quality. The Top-K beams are defined as a subset of beams out of the total available beams based on the ranks (top k beams having highest quality). In one embodiment, the processing of the Top-K beams is done such that the output received from the processing of the Top-K beams are applicable to all the available beams.

[0064] The performance monitoring can be defined for Network (NW)-side performance monitoring and / or user equipment (UE)- assisted performance monitoring, with UE having AI / ML model stored in. For each of these, certain options and alternatives have been defined in RAN1 Agreements for defining performance metric. The following options and alternatives are considered for providing various definitions of the performance for BM-Case 2 considering temporal downlink transmit beam prediction. The determination of the performance metric is based on different parameters and will be clearer from the subsequent paragraphs.

[0065] For the UE-sided model, the following options for inference results are defined: -　　Option 1: Beam information on predicted Top-K beam(s) among a set of beams, -　　Option 2: Beam information on predicted Top-K beam(s) among a set of beams and RSRP of predicted Top-K beam(s) among a set of beams, -　　Option 3: Beam information on predicted Top-K beam(s) among a set of beams and probability information of predicted Top-K beam(s) among a set of beams, -　　Option 4: Beam information on predicted Top-K beam(s) among a set of beams, RSRP of predicted Top-K beam(s) among a set of beams, and confidence information of the RSRP.

[0066] With UE-sided model and UE-assisted performance monitoring, the following alternatives are defined: -　　Alt 1: Top 1 or Top-K beam prediction accuracy (with or without margin) by comparing the prediction results and the Top 1 or Top-K beam based on the measurements from a resource set / resources for monitoring. -　　Alt 2: The L1-RSRP difference information based on actual measurement of the L1-RSRP of one or more of Top-K predicted beam, and L1-RSRP measurements from a resource set / resources for monitoring. -　　Alt 3: The RSRP difference information between the predicted RSRP and measured L1-RSRP of corresponding beam(s) of a resource set / resources for monitoring. -　　Alt 4: The probability information of the predicted beam(s) to be the Top 1 or Top-K beam.

[0067] Further, for the UE-sided model for BM-case 2, for inference results report, 3GPP RAN1#118b Agreement provides support to configure UE with N future time instance(s) for inference by NW when applicable. Further, for the UE-sided model, for the quantization of a RSRP value at least for the report of inference results, the 3GPP RAN1#118 Agreement provides support for differential RSRP reporting with legacy quantization step and range for L1-RSRP reporting. Furthermore, for BM-case 1 and BM-case 2 with a UE-sided AI / ML model, the 3GPP RAN1#118 Agreement at least support Alt 1 having Top 1 or Top K beam prediction accuracy (with or without margin) by comparing the prediction results and the Top 1 or Top K beam based on the measurements from a resource set / resources for monitoring.

[0068] Referring to Fig. 9 now, a representation of N distinct future prediction time instances is shown. The network can send, to the UE, RRC signaling including the configuration for prediction. After the UE performs the set B measurements, the UE can calculate monitoring metric based on the configuration received by the network. In BM Case-2, the model can predict the beams in N future time instances. The UE can predict the beams in N future time instances with the AI / ML model. As all the timing instances are distinct, the network configures all the timing instances and the value of N for the prediction. As shown in the figure, a set B beam measurement results are shown at a time instance t. In one embodiment, the set B measurements results may be considered as historical data and is inputted to the AI / ML model. Using the measurement results (e.g., the set B measurements results), the AI / ML model predicts a set of N future time instances. The UE can predict a set of N future time instances with the AI / ML model using the measurement results. The UE can send, to the network, the performance metric report including the results of prediction. The results of prediction can include the set of N future time instances.

[0069] The total time duration comprises N prediction time instances. These prediction time instances can be in reference to either the time instances of the two consecutive predictions of the AI / ML model or time instances of the two consecutive set B measurements. The total time duration can be in reference to either the recent Set B measurement time instance of which are used for the inference or the time instance of the recent CSI-RS / SSB resources corresponding to the inference report. The first prediction time instance out of N time instances can be in reference to either the latest Set B measurement time instance of which are used for the inference or the time instance of the recent CSI-RS / SSB resources corresponding to the inference report.

[0070] Based on the set B measurement results, the AI / ML model predicts the future N time instances (t1, t2…tN), represented as Prediction 1 in the figure. The set B measurement results are in this case the most recent set B measurement time instances as explained above. The time duration between the future N time instances in this case is distinct (alternative 1) or equal (alternative 2). While Fig. 9 shows the duration between the future N time instances as distinct, the Fig. 10 represents the duration between the future N time instances as equal.

[0071] In Alternative 1, the network can select the time instances with reference to the time instances of the two consecutive actual set B measurements (consecutive actual predictions). Further, the network in the case can configure each time instance of the N time instances in the report configuration. For example, the network (e.g., a base station, gNB etc.) may send, to the UE 3, information indicating each time instance of the N time instances. For example, the UE 3 may predict the beams in N future time instances based on the information indicating each time instance of the N time instances. For example, the network may configure, to the UE 3, the times t1, t2…tN that the UE 3 predicts. For example, the network may send, to the UE, information indicating the times t1, t2…tN that the UE 3 predicts. For example, the UE 3 may predict the beams in N future time instances (e.g., the beams at t1, t2, t3 ... tN) based on the information indicating the times t1, t2…tN.

[0072]

[0073] As shown in the figure, a set B beam measurement results is shown at a time instance t. The total time duration comprises N prediction time instances, where the duration value between each of the N prediction time instances are equal. In this case, the k^th prediction time instance out of N prediction time instances can be determined as t+kT, for k∈{1, 2, …N}. The timing offset T can be a time difference to a reference time to the earliest prediction time instance.

[0074]

[0075] In one embodiment, the reference time can be any one of the alternatives in the following: Alt 1: Inference report time instance, Alt 2: Recent Set B measurement time instance used for inference report, Alt 3: Recent CSI-RS / SSB resources corresponding to the inference report. For example, the reference time (or the reference timing) may be determined based on at least one of the above Alt 1, Alt 2 and Alt 3. For example, the network may determine the reference time based on at least one of the above Alt 1, Alt 2 and Alt 3 (then, the network may send the reference time to the UE 3, and the UE 3 may predict the beams in N future time instances based on the reference time). For example, the UE 3 may determine the reference time based on the at least one of the above Alt 1, Alt 2 and Alt 3 (then, the UE 3 may predict the beams in N future time instances based on the reference time).

[0076] In one embodiment, differential Reference Signal Received Power (RSRP) report of the AI / ML model among multiple beams over multiple time instances. For example, the UE 3 may transmit the differential RSRP report to the network. In accordance with TS 38.133 9.5 L1-RSRP measurements for reporting, when configured by the network, the UE shall be able to perform L1-RSRP measurements of configured CSI-RS, SSB or CSI-RS and SSB resources for L1-RSRP. The measurements shall be performed for a serving cell, including PCell, PSCell, or SCell, on the resources configured for L1-RSRP measurements within the active BWP. The UE 3 can use the differential L1-RSRP based reporting, where the largest predicted value of L1-RSRP is quantized to a 7-bit value in the range [-140, -44] dBm with x-dB step size, and the differential L1-RSRP is quantized to a 4-bit value. (in accordance with TS 38.133, 9.5.3 Measurement Reporting Requirements). For example, the UE 3 may use the differential L1-RSRP based reporting for the differential RSRP reporting. Table 1 (Figure A) showing how the quantization is performed for the RSRP report is provided below: In the above table, a report for BM case-2 has been considered as an example with Top-5 beams (along the rows) over 7 future prediction time instances ((along the column). The RSRP values have been considered from legacy RSRP values range as an example. The reporting range of differential CSI-RSRP for L1 reporting is defined from 0 dBm to -30 dB with 2 dB resolution. (in accordance with TS 38.133).

[0077] For UE-side AI / ML model inference and BM-Case2, the quantization of a RSRP value of inference results in a report over multiple future time instances are based on the following options: Option 1: L1-RSRP of earliest / first predicted time instance, Option 2: 　　 A: Differential RSRP values over the time instances (step 1), B: Differential RSRP values over the beams (step 2), Option 3: Largest L1-RSRP value among the beams at that corresponding time instant, Option 4: A: Different RSRP values over the beams (step 1), B: Differential RSRP values over the time instances (step 2), Option 5: Largest L1-RSRP of among the beams and over the multiple time instances, Each of these above options will be explained in further detail in the subsequent paragraphs.

[0078] Considering option 1 first, the differential RSRP can be determined and reported for each predicted time instance based on the L1-RSRP of the earliest / first predicted time instance. The UE can determine / calculate and report the differential RSRP for each predicted time instance. For this, the predicted L1-RSRP values from the Top K beams of the earliest / first predicted time instance are considered. Further, the differential predicted L1-RSRP values for Top K beams of each time instance based on the predicted L1-RSRP value of the Top K beams of earliest / first time instance of the same inference report are also considered. Table 2 (Figure B) below illustrates the same: As shown in the above table, the first row having L1-RSRP values as {-94, -93, -93, -93, -94, -90, -90} have been quantized to {-94, -1, -1, -1, 0, -4, -4}. Here the L1-RSRP value for the earliest / first predicted time instance is -94. Taking this value as a reference, the rest of the L1-RSRP values have been quantized. For example, the value -93 of the Top 1 beam at the second predicted time instance may be quantized to -1 by calculating as "-94 (this -94 is the reference value) - (-93) = -1". Values in other rows or columns may be also quantized in the same way. The UE can obtain the quantized differential RSRP using option1, based on the prediction and monitoring resources. Table 2 is one example for option1.

[0079] Considering options 2a and 2b now, for UE-side AI / ML model inference and BM-Case 2, for the quantization of a RSRP value of inference results in a report over multiple future time instances, quantization is performed in two steps. Firstly, in step 1, the differential RSRP values are determined over the time instances followed by determining differential RSRP values over the beams by considering the Top 1 predicted RSRP values as reference in step 2. In step 1, the UE can determine / calculate the differential RSRP values over the time instances. In step 2, the UE can determine / calculate the differential RSRP values over the beams. The following tables (Table 3, Figure C) illustrates the options 2a and 2b: Firstly, the differential RSRP are determined over the time instance as shown in second (middle) table. As shown in the above table, the first row having L1-RSRP values as {-94, -93, -93, -93, -94, -90, -90} have been quantized to {-94, -1, -1, -1, 0, -4, -4}. Considering L1-RSRP value of {-94} as a reference, the rest of the L1-RSRP values have been quantized. This first step may be same to the process in the option 1. Once the results of differential RSRP values over the time instances are received or calculated, the differential RSRP values are determined over the beams. Here again considering {-94} as reference, rest of the L1-RSRP values are quantized (third table). For example, the value -92 of the Top 2 beam at the first predicted time instance may be quantized to -2 by calculating as "-94 (this -94 is the reference value) - (-92) = -2". Values in other rows or columns may be also quantized in the same way. The UE can obtain the quantized differential RSRP using option 2a and 2b, based on the prediction and monitoring resources. Table 3 is one example for option 2a and 2b.

[0080] Considering Option 3 now, for the UE-side AI / ML model inference and BM-Case2, for the quantization of a RSRP value of inference results in a report over multiple future time instances, the differential RSRP can be determined and reported for predicted time instance based on the largest L1-RSRP of among the beams at that corresponding time instant. The UE can determine / calculate and report the differential RSRP for predicted time instance. Here, the largest predicted L1-RSRP values from Top K beams of each time instance is considered. Also, the differential predicted L1-RSRP values for each instance based on the largest predicted L1-RSRP value of the corresponding time instance and of the same inference report is also considered. As shown in the above table 4 (Figure D), the entries in the table have been quantized based on the L1-RSRP values of the first row, i.e., largest predicted L1-RSRP values of each time instance. Here, taking the L1-RSRP values {-94, -93, -93, -93, -94, -90, -90} as reference, the rest of the L1-RSRP values in the table have been quantized. For example, the value -92 of the Top 2 beam at the first predicted time instance may be quantized to -2 by calculating as "-94 (this -94 is the reference value) - (-92) = -2". Values in other rows or columns may be also quantized in the same way. The UE can obtain the quantized differential RSRP using option 3, based on the prediction and monitoring resources. Table 4 is one example for option 3.

[0081] Moving to options 4a and 4b now, for UE-side AI / ML model inference and BM-Case2, for the quantization of a RSRP value of inference results in a report over multiple future time instances, quantization is performed in two steps. Firstly, in step 1, the differential RSRP values are determined over the beams followed by determining differential RSRP values over the time instances. In step 2, a reference L1-RSRP values are considered as the L1-RSRP values of a particular time instance at which the largest L1-RSRP value of that particular time instant is higher than the largest L1-RSRP values of all other predicted time instances. Also, based on the reference L1-RSRP values at a particular time instance, the differential values of all other time instances are considered. In step 1, the UE can determine / calculate the differential RSRP values over the beams. In step 2, the UE can determine / calculate the differential RSRP values over the time instances. The following tables (Table 5, Figure E) illustrates the options 4a and 4b: Firstly, the differential RSRP are determined over the beams as shown in second (middle) table. As shown in the above table, the first column having L1-RSRP values as {-94, -92, -90, -89, -87} have been quantized to {-94, -2, -4, -5, -7}. Thereafter the first row having L1-RSRP values as {-94, -93, -93, -93, -94, -90, -90} have been quantized to {-94, -1, -1, -1, 0, -4, -4}. For example, in step 1, the value -92 of the Top 2 beam at the first predicted time instance may be quantized to -2 by calculating as "-94 (this -94 is the reference value) - (-92) = -2". Values in other rows or columns may be also quantized in the same way. For example, in step 2, the value -93 of the Top 1 beam at the second predicted time instance may be quantized to -1 by calculating as "-94 (this -94 is the reference value) - (-93) = -1". Values in other rows or columns may be also quantized in the same way. The UE can obtain the quantized differential RSRP using option 4a and 4b, based on the prediction and monitoring resources. Table 5 is one example for option 4.

[0082] Referring to option 5 now, for UE-side AI / ML model inference and BM-Case2, for the quantization of a RSRP value of inference results in a report over multiple future time instances, the differential RSRP value can be determined based on the largest L1-RSRP of among the beams and over the multiple time instances. The UE can determine / calculate the differential RSRP value based on the largest L1-RSRP of among the beams and over the multiple time instances. For this, the largest predicted L1-RSRP values from the predicted beams over all the time instances are considered. Further, the differential predicted L1-RSRP values for the predicted beams over all the time instances with reference to the largest predicted L1-RSRP value of the same inference report are also considered. The following table 6 (Figure F) illustrates the option 5: In the above table, {-80} being the largest L1-RSRP value among the beams, the quantization of the RSRP has been performed based on this largest L1-RSRP value. For example, the value -94 of the Top 1 beam at the first predicted time instance may be quantized to -14 by calculating as "-94 - (-80) (this -80 is the reference value) = -14". Values in other rows or columns may be also quantized in the same way. The UE can obtain the quantized differential RSRP using option 5, based on the prediction and monitoring resources. Table 6 is one example for option 5.

[0083] In one embodiment, if the RSRP values of Top K beams for each predicted time instance are equal, the above option 1 can be modified. In this case, the differential RSRP values of each predicted time instance are compared with reference to the earliest / first prediction time instance. In this case, the complete report could be a RSRP values of the earliest prediction instance RSRP values and also gNB can indicate the UE regarding the remaining predicted time instance RSRP values through signalling. This has been illustrated in the table 7 (Figure G): As shown in the table 7, the RSRP values of the Top K beams for each predicted time instances are equal, i.e., {-94} in this case. In this case, the RSRP report includes the RSRP values as {-94, -92, -90, -89, -87}. The UE can send, to the network, the performance metric report including the remaining predicted time instance RSRP values, as {-94, -92, -90, -89, -87}. For example, the UE 3 may report the RSRP values as {-94, -92, -90, -89, -87}. In a case where the UE 3 receives an indication to report the remaining RSRP value from the network, the UE 3 may transmit the remaining RSRP value (e.g., the RSRP value as {-0, -0, -0, -0, -0}).

[0084]

[0085]

[0086]

[0087] Referring to Fig. 11 now, timing diagram representing timing for the metric calculation is provided. The figure provides details about the determination of the performance metric using the step 1 and step 2 as explained above. The UE performs the set B measurement. Inference report provides for N time instances labelled as 1…N. For example, the UE may send the differential RSRP report to the network. For example, the UE may send the inference report including at least the differential RSRP report to the network. Calculation of monitoring metric is performed for each N time instances instep 1. The UE can calculate monitoring metric for each N time instances in step 1. Once the monitoring metric is obtained for each of the N time instance, the UE calculates the performance metric for a total of N time instances using the formulas mentioned in step 2. The UE then processes the performance metric and reports the same to the network node (or the network). This process is repeated for each of the inference report (1…N).

[0088]

[0089] The subsequent paragraphs explain the operations performed by the network / network node. The network / network node may be a base station, gNB, etc. The network / network node may include a base station, gNB, etc. For UE-assisted performance monitoring, to report either the performance metric or the decisions such as AI / ML model selection, activation, deactivation, switching, and fallback operations, the network can configure the reporting framework within CSI-framework in csi-ReportConfig or within a dedicated resource configuration. The report quantity can be at least one of the performance metric and / or the recommendation for the model management decisions such as AI / ML model selection, activation, deactivation, switching, and fallback operations.

[0090] The network can configure the gNB to take the model management decisions such as AI / ML model selection, activation, deactivation, switching, and fallback operations based on the report. The network node may configure itself to take the model management decisions such as AI / ML model selection, activation, deactivation, switching, and fallback operations based on the report. The network may configure a waiting time window after the reporting instance for UE to receive at least one of the AI / ML model selection, activation, deactivation, switching, and fallback decision from the gNB. The network or network node may configure the time instances N, the duration between each of the N time instances, etc.

[0091] The performance metric reporting can be periodic, semi-persistent, aperiodic, and event triggered report. Under periodic reporting, periodicity can be k^' multiples of the prediction period. Under period reporting, network can configure the number of prediction instances that UE needs to be considered for calculating the performance metric. Under Aperiodic reporting, network can configure the metric report instance to UE. Further, under event triggered reporting, UE can report by observing a drastic change in performance metric comparing with the desired performance metric, UE can report by observing the L1-RSRP difference of the predicted beams and measured beams lower than a threshold and UE has to report on either MAC-CE or RRC.

[0092] In one embodiment, the time relation between the time instance of the performance monitoring and performance metric reporting can be as per the following table:

[0093] The table above shows that with periodic performance metric monitoring, the performance metric reporting can be periodic, semi-persistent and aperiodic. When the performance metric monitoring is semi-persistent, the performance metric reporting can be semi-persistent and aperiodic (period reporting not supported). Further, when the performance metric monitoring is Aperiodic, the performance metric reporting will only be aperiodic (periodic and semi-persistent not supported).

[0094] Continuing with the different types of reporting of performance metric, the network can configure the monitoring resources for UE-sided AI / ML model performance monitoring. The network can configure either all the available beams or subset of the available beams in the monitoring resources based on the UE capability. A bit-map can be signaled to UE, which conveys the information of the subset of the available beams configured in the monitoring resources / resource set. Network can configure the CSI-RS beams using the existing CSI-RS framework.

[0095] Referring to Fig. 12 now, a timing flow illustrating the procedure of the performance monitoring (e.g. duration of the predicted N future time instances and the differential RSRP report among multiple beams over multiple time instances). The network can send, to the UE, RRC signalling. The network can send the RRC signalling includes the parameters to configure the prediction of the UE. Referring to Fig. 11, The network can configure each time instance of the N time instances based on either the Alternative 1 or Alternative 2 (as explained above) in the report configuration (e.g., RRC signalling). In other words, when the duration values between the each of the N time instances are distinct, the network configures each time instance of the N time instances. Further, when the duration values between the each of the N time instances are equal, the network configures the value of T as explained above in alternative 2.

[0096] Referring to Fig. 12, the UE can perform the set B measurements. After the inference report time instance, the UE can calculate monitoring metric for each N instances. The UE can calculate the monitoring metric based on the configuration. The configuration of the network is shown as the initial step in the figure. There is a time gap between time instance 1 and time instance 2. The time gap can be called monitoring resource period.

[0097] Referring to Fig. 12, for example, the network may send the RRC signalling to the UE. The RRC signalling may include at least one of CSI-RS / SSB RS Set B resources, monitoring Resource period, the number of prediction time instances (N), the number of measured beams (K'), inference report time instance, time gap between the predicted time instances. The RRC signalling may include at least one of information indicating CSI-RS / SSB RS Set B resources, information indicating monitoring Resource period, information indicating the number of prediction time instances (N), information indicating the number of measured beams (K'), information indicating inference report time instance, information indicating time gap between the predicted time instances. The predicted time instance may indicate all the times (e.g., t1, t2, t3 ... tN in Fig. 9) in case of Alt 1. The predicted time instance may indicate the time gap T in case of Alt 2. Then the UE may predict the beams in N future time instances based on the set B measurement results. For example, the UE may perform set B measurements based on CSI-RS / SSB RS Set B resources and may predict the beams in N future time instances based on the set B measurement results. Then the UE may perform the monitoring resource set Tx at each time instance (e.g., at t1, t2, t3 ... tN in case of Fig. 9, at t+T, t+2T, t+3T ... t+NT in case of Fig. 10). For example, the UE 3 may monitor the monitoring resource set based on the monitoring Resource period.

[0098] As shown in Fig. 12, the gap between the monitoring resources may be indicated by the monitoring Resource period. For example, the UE 3 may monitor the monitoring resource set at each time instance and may calculate the actual monitoring results (e.g., RSRP value) at each time instance. For example, the UE 3 may monitor the monitoring resource set at t1 and may calculate the actual monitoring results (e.g., RSRP value) at t1. For example, the UE may monitor the monitoring resource set using information received in the RRC signalling. For example, the UE may determine the monitoring resource set to be measured, the measurement timing of that monitoring resource set, the number of monitoring resource sets to be measure, etc., based on the information included in the RRC signalling. After monitoring the monitoring resources and / or calculating the actual monitoring results, the UE may calculate the performance metric in the manner as mentioned above, and then the UE may report the performance metric to the network.

[0099] The UE can determine / calculate the performance metric. Differential RSRP report among multiple beams over multiple time instances is determined. The UE can obtain differential RSRP report using either of the options 1-5 as explained above and based on the prediction and monitoring resources, the UE will determine performance metric and reports the performance metric to the network node for life cycle management (LCM). The UE can send, to the network, performance metric report.

[0100] Referring to Fig. 13 now, a flowchart of a method 1300 of a user equipment is illustrated. At step 1302, the method comprises receiving a set of monitoring resources from a network node. At step 1304, the method comprises receiving a set of parameters from the network node, the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. At step 1306, the method comprises storing the AI / ML model in a memory. At step 1308, the method comprises determining performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model. At step 1310, the method comprises reporting the performance metric to the network node.

[0101] Referring to Fig. 14 now, a flowchart of a method 1400 performed by a network node is provided. At step 1402, the method comprises transmitting a set of monitoring resources and a set of parameters to the user equipment (UE). The set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. At step 1404, the method comprises receiving performance metric for the set of monitoring resources from the UE, where the performance metric is determined by the UE based on the set of received parameters.

[0102] Referring to Fig. 15 now, a block diagram of the user equipment (UE) 3 in accordance with an embodiment of the present subject matter is provided. The various modules in the UE can be embodied as a hardware that includes, without limitation, the at least one memory 1502, the at least one processor 1504, transmitter / receiver circuitry 1506, and programmable logic or software.

[0103] The at least one memory 1502, which may include both read-only memory (ROM) and random access memory (RAM), can provide instructions and data to the at least one processor 1204. The at least one memory 1502 and the at least one processor 1504 may be operatively coupled. The at least one memory 1502 may store computer readable instructions / computer program code. The at least one processor 1504 in the UE 3 may train the first, second and the third classification models.

[0104] In the context of this document, the "memory" (also referred to as "computer-readable media" or "computer-readable medium") may be any non-transitory media or medium or means that can contain, store, communicate, propagate or transport the instructions for use by or in connection with an instruction execution system, apparatus, or device, such as a computer. The term "non-transitory," as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM).

[0105] The transmitter / receiver (TX / RX) circuitry 1506 may comprise a transmitter and a receiver that can enable the UE to transmit data to or receive data (e.g., the input image of the crop) from the network or plurality of databases. The at least one processor 1204 can be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. The processor can include the logic circuitry with hardware, firmware, and software architecture frameworks for facilitating image processing.

[0106] The steps of a method (e.g., method 1300, 1400) described in connection with the embodiments disclosed herein may be embodied directly in hardware (e.g., UE), in a software module executed by the at least one processor 1504, or in a combination of the two. If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a tangible, non-transitory computer-readable medium (e.g., the at least one memory 1502). A software module may reside in Random Access Memory (RAM), flash memory, Read Only Memory (ROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), registers, hard disk, a removable disk, a CD ROM, or any other form of storage medium known in the art. A storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium.

[0107] In one example, a user equipment (UE) is provided. The UE comprises a transceiver circuit (31) configured to receive from a network node a set of monitoring resources and a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The UE further comprises a memory (36) configured to store the AI / ML model and a controller (33) coupled to the memory (36) and configured to determine performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model, wherein the transceiver circuit (31) is configured to report the performance metric to the network node.

[0108] In another example, the performance metric is determined for the number of prediction time instances (N) and the reporting the performance metric can be the average of T performance metrics for T prediction instances.

[0109] In another example, the duration value of each of the consecutive prediction time instance (N) is either equal or distinct.

[0110] In another example, the inference results of the AI / ML model is configured to perform quantization of RSRP values of inference results in a report over multiple future predicted time instances in the following alternatives: a)　　L1-RSRP values of earliest / first predicted time instance is the reference, the differential values computed based on the reference, b)　　Differential RSRP values over time instances by considering the any predicted time instance L1-RSRP values as reference followed by differential RSRP values over the beams by considering the top beams L1-RSRP values as reference, c)　　Largest L1-RSRP value of among the beams over all predicted time instances is the reference, the differential values computed based on the reference, d)　　Differential RSRP values over the beams by considering the top beams L1-RSRP values as reference followed by Differential RSRP values over the time instant by considering the any predicted time instance L1-RSRP values as reference.

[0111] In another example, a method for a User Equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model is provided. The method comprises receiving from a network node a set of monitoring resources and a set of parameters, wherein the set of parameters comprises at least one of: re-sources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The method further comprises storing, in a memory, the AI / ML model, determining, by a controller, a performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML mod-el, and reporting the performance metric to the network node.

[0112] In one example, a method performed by a network node is provided. The method comprises transmitting a set of monitoring resources and a set of parameters to the user equipment (UE). The set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances. The method further comprises receiving performance metric for the set of monitoring resources from the UE, where the performance metric is determined by the UE based on the set of received parameters.

[0113] In the several embodiments provided in this application, the disclosed system, device, and method may be implemented in another manner. For example, some features of the method embodiments described above may be ignored or not performed. The described device embodiments are merely examples.

[0114] The term based on is not exclusive and allows for being based on additional factors not described unless the context clearly dictates otherwise. In addition, throughout the specification, the meaning of "a", "an" and "the" include plural references. The meaning of "in" includes "in" and "on".

[0115] As used herein the terms "and" and "or" may be used interchangeably to refer to a set of items in both the conjunctive and disjunctive in order to encompass the full description of combinations and alternatives of the items. In either case, the set is to be interpreted as meaning each of the items singularly as alternatives, as well as any combination of the listed items.

[0116] The description above merely illustrating the technical spirit of the present disclosure, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present disclosure. Therefore, the embodiments of the present disclosure described above may be implemented separately or in combination with each other.

[0117] The embodiments disclosed in the present disclosure are intended to illustrate rather than limit the scope of the present disclosure, and the scope of the technical spirit of the present disclosure is not limited by these embodiments. The scope of the present disclosure should be construed by claims below, and all technical spirits within a range equivalent to claims should be construed as being included in the right scope of the present disclosure.

[0118] While only certain features have been illustrated and described herein, many modifications and changes will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the disclosure.

[0119] <Supplementary notes> 　　(Supplementary note 1) 　　A user equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model, the UE compris-ing: a transceiver circuit (31) configured to receive from a network node: 　　a set of monitoring resources; 　　a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances, a memory (36) configured to store the AI / ML model; a controller (33) coupled to the memory (36) and configured to determine performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model, wherein the transceiver circuit (31) is configured to report the performance metric to the network node. 　　(Supplementary note 2) 　　The UE according to supplementary note 1, wherein the performance metric is determined for the number of prediction time instances (N) and the reporting the performance metric can be the average of T performance metrics for T prediction instances. 　　(Supplementary note 3) 　　The UE according to supplementary note 1, wherein the duration value of each of the consecutive prediction time instance (N) is either equal or distinct. 　　(Supplementary note 4) 　　The UE according to supplementary note 3, wherein the inference results of the AI / ML model is con-figured to perform quantization of RSRP values of inference results in a report over multiple future predicted time instances in the following alternatives: a)　　L1-RSRP values of earliest / first predicted time instance is the reference, the differential values computed based on the reference, b)　　Differential RSRP values over time instances by considering the any predicted time instance L1-RSRP values as reference followed by differential RSRP values over the beams by considering the top beams L1-RSRP values as reference, c)　　Largest L1-RSRP value of among the beams over all predicted time instances is the reference, the differential values computed based on the reference, d)　　Differential RSRP values over the beams by considering the top beams L1-RSRP values as reference followed by Differential RSRP values over the time instant by considering the any predicted time instance L1-RSRP values as reference. 　　(Supplementary note 5) 　　The UE according to supplementary note 4, wherein the controller is further configured to determine the actual RSRP values for number of predicted time instances (N) and number of top K beams from the differential RSRP values. 　　(Supplementary note 6) 　　A method for a User Equipment (UE) for performing transmit beam prediction in a tem-poral domain with a UE-side artificial intelligence / machine learning (AI / ML) model, the method comprising: receiving from a network node: 　　　　a set of monitoring resources; 　　　　a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances, storing, in a memory, the AI / ML model; determining, by a controller, a performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model; and reporting the performance metric to the network node. 　　(Supplementary note 7) 　　The method according to supplementary note 6, wherein the performance metric is determined for the number of prediction time instances (N) and the reporting the performance metric can be the average of T performance metrics for T prediction instances. 　　(Supplementary note 8) 　　The method according to supplementary note 6, wherein the duration value of each of the consecu-tive prediction time instance (N) is either equal or distinct. 　　(Supplementary note 9) 　　The method according to supplementary note 6, wherein the inference results of the AI / ML model is configured to perform quantization of RSRP values of inference results in a report over multiple future predicted time instances in the following alternatives: a)　　L1-RSRP values of earliest / first predicted time instance is the reference, the differential values computed based on the reference, b)　　Differential RSRP values over time instances by considering the any predicted time instance L1-RSRP values as reference followed by differential RSRP values over the beams by considering the top beams L1-RSRP values as reference, c)　　Largest L1-RSRP value of among the beams over all predicted time instances is the reference, the differential values computed based on the reference, d)　　Differential RSRP values over the beams by considering the top beams L1-RSRP values as reference followed by Differential RSRP values over the time instant by considering the any predicted time instance L1-RSRP values as reference. 　　(Supplementary note 10) 　　The method according to supplementary note 9, further comprising determining the actual RSRP values for number of predicted time instances (N) and number of top K beams from the differential RSRP values.

[0120] This application is based upon and claims the benefit of priority from Indian Patent Application No. 202511022714, filed on March 13, 2025, the disclosure of which is incorporated herein in its entirety by reference.

[0121] 10　　CORE NETWORK 20　　DATA NETWORK 3　　USER EQUIPMENT(UE) 31　　TRANSCEIVER CIRCUIT 32　　ANTENNA 33　　CONTROLLER 34　　USER INTERFACE 35　　USIM 36　　MEMORY 361　　OPERATING SYSTEM 362　　COMMUNICATIONS CONTROL MODULE 3621　　TRANSCEIVER CONTROL MODULE 5　　RADIO ACCESS NETWORK (RAN) 51　　TRANSCEIVER CIRCUIT 52　　ANTENNA 53　　NETWORK INTERFACE 54　　CONTROLLER 55　　MEMORY 551　　OPERATING SYSTEM 552　　COMMUNICATIONS CONTROL MODULE 5521　　TRANSCEIVER CONTROL MODULE 60　　RU 601　　TRANSCEIVER CIRCUIT 602　　ANNTENA 603　　NETWORK INTERFACE 604　　CONTROLLER 605　　MEMORY 6051　　OPERATING SYSTEM 6052　　COMMUNICATIONS CONTROL MODULE 60521　　TRANSCEIVER CONTROL MODULE 61　　DU 611　　TRANSCEIVER CIRCUIT 612　　NETWORK INTERFACE 613　　CONTROLLER 614　　MEMORY 6141　　OPERATING SYSTEM 6142　　COMMUNICATIONS CONTROL MODULE 61421　　TRANSCEIVER CONTROL MODULE 62　　CU 621　　TRANSCEIVER CIRCUIT 622　　NETWORK INTERFACE 623　　CONTROLLER 624　　MEMORY 6241　　OPERATING SYSTEM 6242　　COMMUNICATIONS CONTROL MODULE 62421　　TRANSCEIVER CONTROL MODULE 7　　CORE NETWORK 70　　AMF 701　　TRANSCEIVER CIRCUIT 702　　NETWORK INTERFACE 703　　CONTROLLER 704　　MEMORY 7041　　OPERATING SYSTEM 7042　　COMMUNICATIONS CONTROL MODULE 70421　　TRANSCEIVER CONTROL MODULE 71　　SMF 72　　UPF 73　　PCF 74　　NEF 75　　UDM 76　　NWDAF 77　　NSSF 78　　NRF 1502　　MEMORY 1504　　PROCESSOR 1506　　TX / RX CIRCUITRY

Claims

1. A user equipment (UE) for performing transmit beam prediction in a temporal domain with a UE-side artificial intelligence / machine learning (AI / ML) model, the UE compris-ing: 　　a transceiver circuit configured to receive from a network node: 　　　　a set of monitoring resources; 　　　　a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances, 　　a memory configured to store the AI / ML model; 　　a controller coupled to the memory and configured to determine performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model, 　　wherein the transceiver circuit is configured to report the performance metric to the network node.

2. The UE according to claim 1, wherein the performance metric is determined for the number of prediction time instances (N) and the reporting the performance metric can be the average of T performance metrics for T prediction instances.

3. The UE according to claim 1, wherein the duration value of each of the consecutive prediction time instance (N) is either equal or distinct.

4. The UE according to claim 3, wherein the inference results of the AI / ML model is con-figured to perform quantization of RSRP values of inference results in a report over multiple future predicted time instances in the following alternatives: a)　　L1-RSRP values of earliest / first predicted time instance is the reference, the differential values computed based on the reference, b)　　Differential RSRP values over time instances by considering the any predicted time instance L1-RSRP values as reference followed by differential RSRP values over the beams by considering the top beams L1-RSRP values as reference, c)　　Largest L1-RSRP value of among the beams over all predicted time instances is the reference, the differential values computed based on the reference, d)　　Differential RSRP values over the beams by considering the top beams L1-RSRP values as reference followed by Differential RSRP values over the time instant by considering the any predicted time instance L1-RSRP values as reference.

5. The UE according to claim 4, wherein the controller is further configured to determine the actual RSRP values for number of predicted time instances (N) and number of top K beams from the differential RSRP values.

6. A method for a User Equipment (UE) for performing transmit beam prediction in a tem-poral domain with a UE-side artificial intelligence / machine learning (AI / ML) model, the method comprising: 　　receiving from a network node: 　　　　a set of monitoring resources; 　　　　a set of parameters, wherein the set of parameters comprises at least one of: resources for channel state information reference signal (CSI-RS) / synchronization signal block reference signal (SSS-RS), monitoring resource period, number of prediction time instances (N), number of predicted beams (K), number of measured beams (K'), inference result time instance, time gap between the predicted time instances, 　　storing, in a memory, the AI / ML model; 　　determining, by a controller, a performance metric for the set of monitoring resources based on the set of parameters and inference results from the AI / ML model; and reporting the performance metric to the network node.

7. The method according to claim 6, wherein the performance metric is determined for the number of prediction time instances (N) and the reporting the performance metric can be the average of T performance metrics for T prediction instances.

8. The method according to claim 6, wherein the duration value of each of the consecu-tive prediction time instance (N) is either equal or distinct.

9. The method according to claim 6, wherein the inference results of the AI / ML model is configured to perform quantization of RSRP values of inference results in a report over multiple future predicted time instances in the following alternatives: a)　　L1-RSRP values of earliest / first predicted time instance is the reference, the differential values computed based on the reference, b)　　Differential RSRP values over time instances by considering the any predicted time instance L1-RSRP values as reference followed by differential RSRP values over the beams by considering the top beams L1-RSRP values as reference, c)　　Largest L1-RSRP value of among the beams over all predicted time instances is the reference, the differential values computed based on the reference, d)　　Differential RSRP values over the beams by considering the top beams L1-RSRP values as reference followed by Differential RSRP values over the time instant by considering the any predicted time instance L1-RSRP values as reference.

10. The method according to claim 9, further comprising determining the actual RSRP values for number of predicted time instances (N) and number of top K beams from the differential RSRP values.

Images