Indication of training datasets for lifecycle management

Abstract

Methods, systems, and devices for wireless communication are described. A user equipment (UE) may receive control information from a network entity. The control information may indicate characteristics of a data set for training a machine learning (ML) model at the UE. The UE may determine to activate or deactivate a first ML model for maintaining the wireless communication link based on receiving the control information. Alternatively, the UE may determine, based on receiving the control information, that functionality of a second ML model to be used to maintain the wireless communication link is active or inactive. The UE may send feedback information based on a determination that the first ML model is to be activated or deactivated or functionality of the second ML model is to be validated or invalidated. The feedback information may indicate one or more parameters of the first ML model or one or more parameters of a functionality of the second ML model.

Description

TECHNICAL FIELD

[0001] The following relates to wireless communications, including indications of training datasets for lifecycle management. BACKGROUND

[0002] Wireless communications systems are widely deployed to provide various types of communication content such as voice, video, packet data, messaging, broadcast, etc. These systems can be capable of supporting communication with multiple users by sharing the available system resources (e.g., time, frequency, and power). Examples of such multiple- access systems include fourth generation (4G) systems such as Long Term Evolution (LTE) systems, LTE-Advanced (LTE-A) systems, or LTE-A Pro systems, and fifth generation (5G) systems which can be referred to as New Radio (NR) systems. These systems can employ technologies such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), or discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-S-OFDM).

[0003] A wireless multiple-access communication system can include one or more network entities each supporting wireless communication for communication devices, which can be referred to as user equipments (UEs). In some wireless communication systems, a communication device can support artificial intelligence (AI) or machine learning (ML). In some cases, existing techniques for managing a lifecycle of an AI or ML (AI / ML) model can be deficient. SUMMARY

[0004] The described techniques relate to improved methods, systems, devices, and apparatuses that support indications of training datasets for lifecycle management (LCM). For example, the described techniques provide a framework for indicating characteristics associated with a training dataset to enable machine learning (ML) model activation or deactivation or functionality enablement or disablement. In some examples, a user equipment (UE) can receive control information from a network entity. The control information can indicate characteristics of a dataset used to train a ML model at the UE. In some examples, the ML model can be associated with maintaining a wireless communication link. The UE can determine, based on receiving the control information, to activate or deactivate a first ML model used to maintain the wireless communication link. Alternatively, the UE can determine, based on receiving the control information, to use functionality enablement or disablement of a second ML model used to maintain the wireless communication link. The UE can transmit feedback information to the network entity based on determining to activate or deactivate the first ML model or to use the functionality enablement or disablement of the second ML model. The feedback information can indicate one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model.

[0005] A method for wireless communication at a UE is described. The method can include receiving, from a network entity, control information indicating a characteristic of a dataset for training a ML model at the UE, the ML model being associated with maintaining a wireless communication link; determining, based on receiving the control information, to activate or deactivate a first ML model for maintaining the wireless communication link or to effectuate or nullify functionality of a second ML model for maintaining the wireless communication link; and transmitting, to the network entity, feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model based on the determination.

[0006] An apparatus for wireless communication at a UE is described. The apparatus can include a processor, a memory coupled with the processor, and instructions stored in the memory. The instructions can be executable by the processor to cause the apparatus to receive, from a network entity, control information indicating a characteristic of a dataset for training a ML model at the UE, the ML model being associated with maintaining a wireless communication link; determine, based on receiving the control information, to activate or deactivate a first ML model for maintaining the wireless communication link or to effectuate or nullify functionality of a second ML model for maintaining the wireless communication link; and transmit, to the network entity, feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model based on the determination.

[0007] Another apparatus for wireless communication at a UE is described. The apparatus can include means for receiving, from a network entity, control information indicating a characteristic of a dataset for training a ML model at the UE, the ML model being associated with maintaining a wireless communication link; means for determining, based on receiving the control information, to activate or deactivate a first ML model for maintaining the wireless communication link or to effectuate or nullify functionality of a second ML model for maintaining the wireless communication link; and means for transmitting, to the network entity, feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model based on the determination.

[0008] A non-transitory computer-readable medium storing code for wireless communications at a UE is described. The code can include instructions executable by a processor to receive, from a network entity, control information indicating a characteristic of a dataset for training a ML model at the UE, the ML model being associated with maintaining a wireless communication link, determine, based on receiving the control information, to activate or deactivate a first ML model for maintaining the wireless communication link or to effectuate or nullify functionality of a second ML model for maintaining the wireless communication link, and transmit, to the network entity, feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model based on the determination.

[0009] In some examples of the method, apparatuses, and non-transitory computer- readable media described herein, receiving the control information can include operations, features, means, or instructions for receiving a dataset identifier (ID) corresponding to the dataset or a characteristic ID corresponding to the characteristic or both.

[0010] Some examples of the method, apparatuses, and non-transitory computer-readable media described herein can further include operations, features, means, or instructions for transmitting, to the network entity, an uplink message including one or more dataset IDs corresponding to one or more recommended datasets, one or more characteristic IDs corresponding to one or more recommended characteristics, or both, where the control information can be based on the uplink message.

[0011] In some examples of the method, apparatuses, and non-transitory computer- readable media described herein, the one or more dataset IDs include at least the dataset ID, and the one or more characteristic IDs include at least the characteristic ID.

[0012] In some examples of the method, apparatuses, and non-transitory computer- readable media described herein, the characteristic includes an operating scenario of the wireless communication link, a profile characteristic of the wireless communication link, a parameter of a cell serving the wireless communication link, a transmission parameter for downlink communications via the wireless communication link, a transmission parameter for uplink communications via the wireless communication link, or a distance between the network entity and the UE.

[0013] In some examples of the method, apparatuses, and non-transitory computer- readable media described herein, the first ML model and the second ML model each include a respective beam prediction ML model based on the dataset for training a beam prediction ML model.

[0014] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, the characteristic includes a statistical quantity associated with a received power measurement used as an input to the ML model, a statistical quantity associated with a received power measurement used as a prediction target of the ML model, a performance metric associated with the received power measurement used as the input to the ML model, a performance metric associated with the received power measurement used as the prediction target of the ML model, a UE mobility characteristic, a characteristic of a transmit beam at the network entity, or a characteristic of a receive beam at the UE.

[0015] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, receiving the control information can include operations, features, means, or instructions for receiving radio resource control (RRC) layer signaling, physical (PHY) layer signaling, medium access control (MAC) layer signaling, or application layer signaling that includes the control information.

[0016] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, transmitting the feedback information can include operations, features, means, or instructions for transmitting an indication of a correspondence between the one or more parameters and the characteristic.

[0017] Some examples of the method, apparatuses, and non-transitory computer- readable medium described herein can further include operations, features, means, or instructions for identifying the one or more parameters in response to the determination and based on the correspondence between the one or more parameters and the characteristic, where the feedback information indicates activation or deactivation of the first ML model at the UE or validation or invalidation of the functionality of the second ML model at the UE.

[0018] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, transmitting the feedback information can include operations, features, means, or instructions for transmitting an indication of the one or more parameters.

[0019] Some examples of the method, apparatuses, and non-transitory computer- readable medium described herein can further include operations, features, means, or instructions for activating or deactivating the first ML model in association with the characteristic of the dataset.

[0020] Some examples of the method, apparatuses, and non-transitory computer- readable medium described herein can further include operations, features, means, or instructions for predicting, based on activating the first ML model, a transmit beam at the network entity or a receive beam at the UE for maintaining the wireless communication link using the first ML model, where the dataset can be used to train a beam prediction ML model.

[0021] Some examples of the method, apparatuses, and non-transitory computer-readable medium described herein can further include operations, features, means, or instructions for causing the functionality of the second ML model to be validated or invalidated in association with the characteristic of the data set.

[0022] Some examples of the method, apparatuses, and non-transitory computer-readable medium described herein can further include operations, features, means, or instructions for predicting, based on causing the functionality of the second ML model to be validated, a transmit beam at the network entity or a receive beam at the UE for maintaining the wireless communication link using the second ML model, where the data set can be used to train a beam prediction ML model.

[0023] A method for wireless communications at a network entity is described. The method can include outputting control information indicating a characteristic of a data set for training a ML model at a UE, the ML model being associated with maintaining a wireless communication link, and obtaining, in response to the control information, feedback information indicating one or more parameters of a first ML model or one or more parameters of functionality of a second ML model based on determining to activate or deactivate the first ML model for maintaining the wireless communication link or to validate or invalidate functionality of the second ML model for maintaining the wireless communication link.

[0024] An apparatus for wireless communications at a network entity is described. The apparatus can include a processor, memory coupled with the processor, and instructions stored in the memory. The instructions can be executable by the processor to cause the apparatus to output control information indicating a characteristic of a data set for training a ML model at a UE, the ML model being associated with maintaining a wireless communication link, and obtain, in response to the control information, feedback information indicating one or more parameters of a first ML model or one or more parameters of functionality of a second ML model based on determining to activate or deactivate the first ML model for maintaining the wireless communication link or to validate or invalidate functionality of the second ML model for maintaining the wireless communication link.

[0025] Another apparatus for wireless communications at a network entity is described. The apparatus can include means for outputting control information indicating a characteristic of a data set for training a ML model at a UE, the ML model being associated with maintaining a wireless communication link, and means for obtaining, in response to the control information, feedback information indicating one or more parameters of a first ML model or one or more parameters of functionality of a second ML model based on determining to activate or deactivate the first ML model for maintaining the wireless communication link or to validate or invalidate functionality of the second ML model for maintaining the wireless communication link.

[0026] A non-transitory computer-readable medium storing code for wireless communications at a network entity is described. The code can include instructions executable by a processor to output control information indicating a characteristic of a dataset for training a ML model at a UE, the ML model being associated with maintaining a wireless communication link, and obtain feedback information indicating one or more parameters of the first ML model or one or more parameters of functionality of the second ML model based on determining to activate or deactivate a first ML model for maintaining the wireless communication link or to effectuate or nullify functionality of a second ML model for maintaining the wireless communication link in response to the control information.

[0027] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, outputting the control information can include operations, features, means, or instructions for outputting a dataset ID corresponding to the dataset or a characteristic ID corresponding to the characteristic or both.

[0028] Some examples of the method, apparatuses, and non-transitory computer- readable medium described herein can further include operations, features, means, or instructions for obtaining an uplink message including one or more dataset IDs corresponding to one or more recommended datasets, one or more characteristic IDs corresponding to one or more recommended characteristics, or both, where the control information can be based on the uplink message.

[0029] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, the one or more dataset IDs include at least the dataset ID, and the one or more characteristic IDs include at least the characteristic ID.

[0030] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, the characteristic includes an operating scenario of the wireless communication link, a profile characteristic of the wireless communication link, a parameter of a cell serving the wireless communication link, a transmission parameter for downlink communications via the wireless communication link, a transmission parameter for uplink communications via the wireless communication link, or a distance between the network entity and the UE.

[0031] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, the first ML model and the second ML model each include a respective beam prediction ML model based on the dataset for training a beam prediction ML model.

[0032] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, the characteristic includes a statistical quantity associated with a received power measurement used as an input to the ML model, a statistical quantity associated with a received power measurement used as a prediction target of the ML model, a performance metric associated with the received power measurement used as the input to the ML model, a performance metric associated with the received power measurement used as the prediction target of the ML model, a UE mobility characteristic, a characteristic of a transmit beam at the network entity, or a characteristic of a receive beam at the UE.

[0033] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, outputting the control information can include operations, features, means, or instructions for outputting RRC layer signaling, PHY layer signaling, MAC layer signaling, or application layer signaling that includes the control information.

[0034] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, obtaining the feedback information can include operations, features, means, or instructions for obtaining an indication of a correspondence between the one or more parameters and the characteristic.

[0035] In some examples of the method, apparatuses, and non-transitory computer- readable medium described herein, obtaining the feedback information can include operations, features, means, or instructions for obtaining an indication of the one or more parameters. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 AND Figure 2 Each of the figures illustrates exemplary wireless communications systems that support indication of training datasets for lifecycle management (LCM) in accordance with one or more aspects of the present disclosure.

[0037] Figure 3 An example of a process flow that supports indication of training datasets for LCM in accordance with one or more aspects of the present disclosure is shown.

[0038] Figure 4 AND Figure 5 A block diagram of a device that supports indication of training datasets for LCM in accordance with one or more aspects of the present disclosure is shown.

[0039] Figure 6 A block diagram of a communications manager that supports indication of training datasets for LCM in accordance with one or more aspects of the present disclosure is shown.

[0040] Figure 7 A diagram of a system including a device that supports indication of training datasets for LCM in accordance with one or more aspects of the present disclosure is shown.

[0041] Figure 8 and Figure 9 A block diagram illustrating a device that supports indication of a training data set for LCM in accordance with one or more aspects of the present disclosure is shown.

[0042] Figure 10 A block diagram illustrating a communications manager that supports indication of a training data set for LCM in accordance with one or more aspects of the present disclosure is shown.

[0043] Figure 11 An illustration of a system that includes a device that supports indication of a training data set for LCM in accordance with one or more aspects of the present disclosure is shown.

[0044] Figure 12 and Figure 13 A flow diagram illustrating a method that supports indication of a training data set for LCM in accordance with one or more aspects of the present disclosure is shown. DETAILED DESCRIPTION

[0045] Some wireless communications systems can support artificial intelligence or machine learning (AI / ML) at one or more communication devices, such as user equipment (UE). For example, a UE can support one or more AI / ML models for various functionalities, such as beam prediction for beam management. In some examples, a network entity associated with the UE can enable (or otherwise support) AI / ML model lifecycle management (LCM) at the UE. For example, the network entity can monitor the performance of the UE or one or more AI / ML models deployed at the UE. In some examples, such as based on the performance monitoring, the network entity can make a determination regarding selection, activation, or deactivation of AI / ML models deployed at the UE. Additionally or alternatively, based on the performance monitoring, the network can make a determination regarding selection, validation, or invalidation of functionalities for which the UE can use AI / ML models.

[0046] In some examples, the network entity and the UE can support model-based LCM for AI / ML, in which the network entity can obtain information associated with an AI / ML model at the UE, such as parameters and structure of the AI / ML model. The network entity can use the information obtained for the AI / ML model to make a determination and instruct the UE to activate or deactivate the AI / ML model. For example, the network entity can instruct the UE to activate or deactivate the AI / ML model based on (or using) the information obtained by the network entity for the AI / ML model. However, in some examples, model-based LCM can result in sensitive information being disclosed to the network entity. In some other examples, the UE and the network entity can support functionality-based LCM for AI / ML, in which the network entity can instruct the UE to activate or deactivate the AI / ML model by instructing the UE to validate or invalidate functionality. In other words, the network entity can achieve AI / ML model activation or deactivation by instructing the validation or invalidation of functionality. In such examples, the UE can reduce (e.g., avoid) the likelihood of sensitive information being disclosed to the network entity. However, in some examples, aspects of the functionality (e.g., how the functionality is defined) can be unclear to the UE or the network entity or both. Thus, instructing the validation or invalidation of functionality can be ambiguous to the UE, which can degrade the performance of the LCM at the UE.

[0047] Various aspects of the present disclosure relate to techniques for indication of training datasets for LCM, and more specifically, to a framework for indicating characteristics associated with a training dataset to enable AI / ML model activation or deactivation or functionality enablement or disablement. For example, an AI / ML model deployed at a UE for functionality can be associated with a dataset (e.g., a training dataset) used to train the AI / ML model. That is, the AI / ML model and functionality of the AI / ML model can be associated with the dataset used to train the AI / ML model. Accordingly, a network entity can use characteristics of the dataset used to train the AI / ML model to indicate to the UE to activate or deactivate the AI / ML model or to enable or disable functionality of the AI / ML model. For example, the UE can receive control information indicating characteristics of a dataset used to train an AI / ML model at the UE. In some examples, such as in response to receiving the control information, the UE can determine to activate or deactivate a first AI / ML model in accordance with the indicated characteristics. Additionally or alternatively, the UE can determine to enable or disable functionality of a second AI / ML model (e.g., an activated AI / ML model) in accordance with the indicated characteristics. In some examples, the AI / ML model can be associated with one or more functionalities, such as a functionality associated with maintaining a wireless communication link. In some examples, the UE can transmit feedback information to the network entity based on determining to activate or deactivate the first AI / ML model or determining to enable or disable functionality of the second AI / ML model. The feedback information can indicate one or more parameters of the first AI / ML model or one or more parameters of the functionality of the second AI / ML model.

[0048] Aspects of the subject matter described herein can be implemented to realize one or more of the following potential advantages. For example, techniques employed by the described communication devices can provide benefits and enhancements for operation of the communication devices, including improving LCM for AI / ML operations at a UE. Operations performed by the described communication devices to improve LCM for AI / ML operations at a UE can include indicating, to a UE, characteristics of a dataset used to train an AI / ML model at the UE. In some examples, among other benefits, operations performed by the described communication devices can support improvements in communication reliability within a wireless communication system and other benefits. Aspects of the present disclosure are initially described in the context of wireless communication systems and process flows. Aspects of the present disclosure are further illustrated by and described with reference to apparatus diagrams, system diagrams, and flowcharts that relate to indication of training datasets for LCM.

[0049] Figure 1An example of a wireless communications system 100 that supports indication of a training dataset for LCM is shown, in accordance with one or more aspects of the present disclosure. The wireless communications system 100 can include one or more network entities 105, one or more UEs 115, and a core network 130. In some examples, the wireless communications system 100 can be a Long Term Evolution (LTE) network, an LTE- Advanced (LTE-A) network, an LTE-A Pro network, a New Radio (NR) network, or a network operating according to some other wireless standard, including future iterations of the wireless standards explicitly mentioned herein and future wireless standards not yet developed.

[0050] The network entities 105 can be dispersed throughout the geographic region of the wireless communications system 100, and can each include devices in different forms or having different capabilities. In various examples, the network entities 105 can be referred to as network elements, mobility elements, radio access network (RAN) nodes, or network equipment, among other nomenclature. In some examples, the network entities 105 and the UEs 115 can wirelessly communicate via one or more communication links 125, such as radio frequency (RF) access links. For example, a network entity 105 can support a coverage area 110 (e.g., a geographic coverage area) within which UEs 115 and the network entity 105 can establish one or more communication links 125. The coverage area 110 can be an example of a geographic area over which a network entity 105 and a UE 115 can support communication in accordance with one or more radio access technologies (RATs).

[0051] The UEs 115 can be dispersed throughout the coverage areas 110 of the wireless communications system 100, and each UE 115 can be stationary or mobile, or both at different times. The UEs 115 can be devices in different forms or having different capabilities. Figure 1 Some example UEs 115 are illustrated in FIG. 1. The UEs 115 described herein can be able to communicate with various types of devices, such as other UEs 115 or network entities 105, as shown in FIG. 1. Figure 1 The UEs 115 described herein can be able to communicate with various types of devices, such as other UEs 115 or network entities 105, as shown in FIG. 1.

[0052] As described herein, a node of the wireless communications system 100 (which can be referred to as a network node or a wireless node) can be a network entity 105 (e.g., any of the network entities described herein), a UE 115 (e.g., any of the UEs described herein), a network controller, a device, an apparatus, a computing system, one or more components, or another suitable processing entity configured to perform any of the techniques described herein. For example, a node can be a UE 115. As another example, a node can be a network entity 105. As yet another example, a first node can be configured to communicate with a second node or a third node. In one aspect of this example, the first node can be a UE 115, the second node can be a network entity 105, and the third node can be a UE 115. In another aspect of this example, the first node can be a UE 115, the second node can be a network entity 105, and the third node can be a network entity 105. In other aspects of this example, the first node, the second node, and the third node can be different relative to these examples. Similarly, references to a UE 115, a network entity 105, a device, an apparatus, a computing system, etc. can include the disclosure of the UE 115, the network entity 105, the device, the apparatus, the computing system, etc. as a node. For example, a disclosure that a UE 115 is configured to receive information from a network entity 105 also discloses that a first node is configured to receive information from a second node.

[0053] In some examples, the network entities 105 can be in communication with the core network 130 or with each other or both. For example, the network entities 105 can communicate with the core network 130 via one or more backhaul communication links 120 (e.g., according to an SI, N2, N3, or other interface protocol). In some examples, the network entities 105 can communicate with each other via backhaul communication links 120 (e.g., according to an X2, Xn, or other interface protocol) either directly (e.g., direct point-to- point between network entities 105) or indirectly (e.g., via core network 130). In some examples, the network entities 105 can communicate with each other via mid-cell communication links 162 (e.g., according to a mid-cell interface protocol) or front-haul communication links 168 (e.g., according to a front-haul interface protocol), or any combination thereof. The backhaul communication links 120, the mid-cell communication links 162, or the front-haul communication links 168 can be or include one or more wired links (e.g., electrical, fiber optic), one or more wireless links (e.g., radio, wireless optical), etc., or various combinations thereof. A UE 115 can communicate with the core network 130 via communication links 155.

[0054] One or more of the network entities 105 described herein can include or can be referred to as a base station 140 (e.g., a transceiver base station, a radio base station, an NR base station, an access point, a radio transceiver, a NodeB, an eNodeB (eNB), a next generation NodeB, or a giga-NodeB (either of which can be referred to as a gNB), a 5G NB, a next generation eNB (ng-eNB), a Home NodeB, a Home eNodeB, or other suitable terminology). In some examples, the network entity 105 (e.g., base station 140) can be implemented in an aggregated (e.g., monolithic, self-standing) base station architecture that can be configured to utilize protocol stacks that are physically or logically integrated within a single network entity 105 (e.g., a single RAN node such as a base station 140).

[0055] In some examples, the network entity 105 can be implemented in a disaggregated architecture (e.g., a disaggregated base station architecture, a disaggregated RAN architecture) that can be configured to utilize protocol stacks that are physically or logically distributed between two or more network entities 105 such as an integrated access backhaul (IAB) network, an open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance), or a virtualized RAN (vRAN) (e.g., a cloud RAN (C-RAN)). For example, the network entity 105 can include one or more of a central unit (CU) 160, a distributed unit (DU) 165, a radio unit (RU) 170, a RAN intelligent controller (RIC) 175 (e.g., a near real-time RIC (near-RT RIC), a non-real-time RIC (non-RT RIC)), a service management and orchestration (SMO) 180 system, or any combination thereof. The RU 170 can also be referred to as a radio head, an intelligent radio head, a remote radio head (RRH), a remote radio unit (RRU), or a transmission reception point (TRP). One or more components of the network entity 105 in the disaggregated RAN architecture can be co-located, or one or more components of the network entity 105 can be in distributed locations (e.g., separate physical locations). In some examples, one or more network entities 105 of the disaggregated RAN architecture can be implemented as virtual units (e.g., a virtual CU (VCU), a virtual DU (VDU), a virtual RU (VRU)).

[0056] The functional split between the CU 160, the DU 165, and the RU 170 is flexible and can support different functionality depending on which functions (e.g., network layer functions, protocol layer functions, baseband functions, RF functions, and any combination thereof) are performed at the CU 160, the DU 165, or the RU 170. For example, a functional split of a protocol stack can be employed between the CU 160 and the DU 165, such that the CU 160 can support one or more layers of the protocol stack, and the DU 165 can support one or more different layers of the protocol stack. In some examples, the CU 160 can host higher protocol layer (e.g., Layer 3 (L3), Layer 2 (L2)) functionality and signaling (e.g., Radio Resource Control (RRC), Service Data Adaptation Protocol (SDAP), Packet Data Convergence Protocol (PDCP)). The CU 160 can connect to one or more DUs 165 or RUs 170, and the one or more DUs 165 or RUs 170 can host lower protocol layers, such as Layer 1 (LI) (e.g., Physical (PHY) layer) or L2 (e.g., Radio Link Control (RLC) layer, Medium Access Control (MAC) layer) functionality and signaling, and can each be at least partially controlled by the CU 160. Additionally or alternatively, a functional split of a protocol stack can be employed between the DU 165 and the RU 170, such that the DU 165 can support one or more layers of the protocol stack, and the RU 170 can support one or more different layers of the protocol stack. The DU 165 can support one or more different cells (e.g., via one or more RUs 170). In some cases, the functional split between the CU 160 and the DU 165 or between the DU 165 and the RU 170 can be within a protocol layer (e.g., some functions of a protocol layer can be performed by one of the CU 160, the DU 165, or the RU 170, while other functions of that protocol layer are performed by a different one of the CU 160, the DU 165, or the RU 170). The CU 160 can be further split in functionality into CU control plane (CU-CP) and CU user plane (CU-UP) functions. The CU 160 can connect to one or more DUs 165 via a backhaul communication link 162 (e.g., Fl, Fl-c, Fl-u), and the DU 165 can connect to one or more RUs 170 via a front-haul communication link 168 (e.g., open front-haul (FH) interface). In some examples, the backhaul communication link 162 or the front-haul communication link 168 can be implemented according to an interface (e.g., channel) between layers of a protocol stack that are supported by the respective network entities 105 that communicate via these communication links.

[0057] In some wireless communications systems (e.g., wireless communications system 100), infrastructure and spectrum resources for radio access can support wireless backhaul link capabilities to supplement wired backhaul connections to provide an IAB network architecture (e.g., to core network 130). In some cases, in an IAB network, one or more network entities 105 (e.g., IAB nodes 104) can be partially controlled by one another. One or more IAB nodes 104 can be referred to as a donor entity or IAB donor. One or more DUs 165 or one or more RUs 170 can be partially controlled by one or more CUs 160 associated with a donor network entity 105 (e.g., a donor base station 140). One or more donor network entities 105 (e.g., IAB donors) can communicate with one or more additional network entities 105 (e.g., IAB nodes 104) via supported access and backhaul links (e.g., backhaul communication links 120). An IAB node 104 can include an IAB mobile terminal (IAB-MT) controlled (e.g., scheduled) by a coupled DU 165 of an IAB donor. The IAB-MT can include a separate set of antennas for relaying communications with UEs 115 or can share the same antennas (e.g., of an RU 170) of the IAB node 104 for accessing via the DU 165 of the IAB node 104 (e.g., referred to as a virtual IAB-MT (vIAB-MT)). In some examples, an IAB node 104 can include a DU 165 that supports a communication link with an additional entity (e.g., IAB node 104, UE 115) within a relay chain or configuration (e.g., downstream) of an access network. In such cases, one or more components of the disaggregated RAN architecture (e.g., one or more IAB nodes 104 or components of an IAB node 104) can be configured to operate according to the techniques described herein.

[0058] In cases where the techniques described herein apply in the context of a disaggregated RAN architecture, one or more components of the disaggregated RAN architecture can be configured to support the indication of a training dataset for LCM as described herein. For example, some operations described as being performed by a UE 115 or a network entity 105 (e.g., base station 140) can additionally or alternatively be performed by one or more components of the disaggregated RAN architecture (e.g., IAB node 104, DU 165, CU 160, RU 170, RIC 175, SMO 180).

[0059] A UE 115 can include or can be referred to as a mobile device, a wireless device, a remote device, a handheld device, or a subscriber device, or some other suitable terminology, where the “device” can also be referred to as a unit, a station, a terminal, or a client, among other examples. A UE 115 can also include or can be referred to as a personal electronic device such as a cellular phone, a personal digital assistant (PDA), a tablet computer, a laptop computer, or a personal computer. In some examples, a UE 115 can include or can be referred to as a wireless local loop (WLL) station, an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a machine type communications (MTC) device, among other examples, which can be implemented in various objects such as appliances or vehicles, among other examples.

[0060] The UEs 115 described herein can be able to communicate with various types of devices, such as other UEs 115 that can sometimes act as relays or Figure 1 network equipment including a macro eNB or gNB, a small cell eNB or gNB, or a relay base station, among other examples, as shown in

[0061] The UEs 115 and the network entities 105 can wirelessly communicate with each other using resources associated with one or more carriers via one or more communication links 125 (e.g., access links). The term “carrier” can refer to a set of RF spectrum resources having a defined physical layer structure for supporting communication links 125. For example, a carrier used for a communication link 125 can include a portion of an RF spectrum band (e.g., a bandwidth part (BWP)) operating according to one or more physical layer channels for a given radio access technology (e.g., LTE, LTE-A, LTE-A Pro, NR). Each physical layer channel can carry acquisition signaling (e.g., synchronization signals, system information), control signaling (e.g., control channels), user data (e.g., data channels), or other signaling. The wireless communications system 100 can support communication with a UE 115 using carrier aggregation or multi-carrier operation. According to a carrier aggregation configuration, a UE 115 can be configured to have multiple downlink component carriers and one or more uplink component carriers. Carrier aggregation can be used for both frequency division duplexing (FDD) and time division duplexing (TDD) component carriers. Communication between a network entity 105 and other devices can refer to communication between these devices and any portion of the network entity 105 (e.g., an entity, a sub-entity). For example, the terms “transmit,” “receive,” or “communicate” can refer to any portion of the network entity 105 (e.g., a base station 140, a CU 160, a DU 165, a RU 170) communicating with another device (e.g., directly or via one or more other network entities 105).

[0062] The signal waveform transmitted via a carrier may include multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques, such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform extended OFDM (DFT-S-OFDM)). In a system employing MCM, a resource element may refer to a resource of one symbol period (e.g., the duration of one modulation symbol) and one subcarrier, in which case the symbol period and subcarrier spacing may be inversely related. The number of bits carried by each resource element may depend on the modulation scheme (e.g., the order of the modulation scheme, the decoding rate of the modulation scheme, or both), such that a relatively high number of resource elements (e.g., in the transmission duration) and a relatively high-order modulation scheme can correspond to a relatively high communication rate. Wireless communication resources may refer to a combination of RF spectrum resources, temporal resources, and spatial resources (e.g., spatial layers or beams), and the use of multiple spatial resources can increase the data rate or data integrity used for communication with UE 115.

[0063] The time interval for network entity 105 or UE 115 can be expressed as a multiple of a basic time unit, such as the sampling period T. s =1 / (Δf) max ·N f ) seconds, where Δf max This can represent the supported subcarrier spacing, while N f The supported Discrete Fourier Transform (DFT) size can be represented. The time interval of the communication resources can be organized according to radio frames, each with a specified duration (e.g., 10 milliseconds (ms)). Each radio frame can be identified by a System Frame Number (SFN) (e.g., ranging from 0 to 1023).

[0064] Each frame may include multiple consecutively numbered subframes or time slots, and each subframe or time slot may have the same duration. In some examples, a frame may (e.g., in the time domain) be divided into subframes, and each subframe may be further divided into a number of time slots. Alternatively, each frame may include a variable number of time slots, and the number of time slots may depend on the subcarrier spacing. Each time slot may include a number of symbol periods (e.g., depending on the length of the cyclic prefix appended to each symbol period). In some wireless communication systems 100, time slots may be further divided into multiple micro-time slots associated with one or more symbols. Excluding the cyclic prefix, each symbol period may be associated with one or more (e.g., N) symbols. f The duration of a symbol period is associated with a ( ) sampling period. The duration of a symbol period can depend on the subcarrier spacing or the operating frequency band.

[0065] A subframe, a slot, a mini-slot, or a symbol can be the smallest scheduling unit (e.g., in the time domain) of the wireless communications system 100 and can be referred to as a transmission time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol periods in a TTI) can be variable. Additionally or alternatively, the smallest scheduling unit of the wireless communications system 100 (e.g., in the time domain) can be dynamically selected (e.g., in bursts of shortened TTIs (sTTIs)).

[0066] Physical channels can be multiplexed according to various techniques to communicate using a carrier. For example, physical control channels and physical data channels can be multiplexed using one or more of time division multiplexing (TDM) techniques, frequency division multiplexing (FDM) techniques, or hybrid TDM-FDM techniques, among other techniques. A control region (e.g., a control resource set (CORESET)) of a physical control channel can be defined by a collection of symbol periods and can extend across the system bandwidth or a subset thereof of a carrier. One or more control regions (e.g., CORESETs) can be configured for a set of UEs 115. For example, one or more of the UEs 115 can monitor or search control regions for control information according to one or more search space sets, and each search space set can include one or more control channel candidates arranged in an order of increasing aggregation level. An aggregation level of a control channel candidate can refer to a quantity of control channel resources (e.g., control channel elements (CCEs)) associated with encoded information for a control information format having a given payload size. Search space sets can include common search space sets configured for transmission of control information to multiple UEs 115 and UE-specific search space sets configured for transmission of control information to a specific UE 115.

[0067] The network entity 105 can provide communication coverage for a geographic area 110 via one or more cells (e.g., macro cells, small cells, hot spots, or other types of cells, or any combination thereof). The term “cell” can refer to a logical communication entity used for communication with a network entity 105 (e.g., using a carrier) and can be associated with a identifier (e.g., a physical cell identifier (PCID), a virtual cell identifier (VCID), or other cell identifier) used by a UE 115 to identify a particular cell from which it is receiving the signal. In some examples, the cell can also refer to the coverage area 110, or a portion thereof (e.g., a sector), over which a logical communication entity operates. The size of such a cell can vary from very small (e.g., a structure or a subset of a structure) to very large in scope (e.g., according to coverage area 110). For example, a cell can be or include a building, a subset of a building, or an outdoor space between or overlapping coverage areas 110, among other examples.

[0068] Macro cells can generally cover relatively large geographic areas (e.g., 10s of meters to 100s of meters in radius) and can allow unrestricted access by UEs 115 with service subscriptions with the network provider. Small cell base stations 140, which can also be referred to as femto, pico, or micro base stations, can be associated with a relatively small geographic area (e.g., a private home or a part of a building) and can allow unrestricted access by UEs 115 with service subscriptions with the network provider or restricted access by UEs 115 having an association with the small cell (e.g., by receiving an invitation or some other form of permission).

[0069] In some examples, a carrier can support a number of cells, and different cells can be configured according to different protocol types (e.g., MTC, narrowband IoT (NB-IoT), enhanced mobile broadband (eMBB)) that can provide access for different types of devices.

[0070] In some examples, network entities 105 (e.g., base stations 140, RUs 170) can be mobile and thus provide communication coverage for mobile coverage areas 110. In some examples, different coverage areas 110 associated with different technologies can overlap, but different coverage areas 110 can be supported by the same network entity 105. In some other examples, overlapping coverage areas 110 associated with different technologies can be supported by different network entities 105. Wireless communication system 100 can include, for example, a heterogeneous network in which different types of network entities 105 provide coverage for various coverage areas 110 using the same or different radio access technologies.

[0071] Wireless communication system 100 can be configured to support ultra-reliable communications or low-latency communications, or various combinations thereof. For example, wireless communication system 100 can be configured to support ultra-reliable low-latency communications (URLCC). UEs 115 can be designed to support ultra-reliable, low-latency, or critical functions. Ultra-reliable communications can include private communication or group communication and can be supported by one or more services, such as push-to-talk, video, or data. Support for ultra-reliable, low-latency functions can include prioritization of services, and such services can be used for public safety or general commercial applications. The terms ultra-reliable, low-latency, and ultra-reliable low-latency can be used interchangeably herein.

[0072] In some examples, UEs 115 can be configured to communicate directly with other UEs 115 via device-to-device (D2D) communication link 135 (e.g., according to a peer-to-peer (P2P), D2D, or sidelink protocol). In some examples, one or more UEs 115 in a group that is performing D2D communication can be within the coverage area 110 of a network entity 105 (e.g., base station 140, RU 170) that can support

[0073] The core network 130 can provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network 130 can be an evolved packet core (EPC) or 5G core (5GC), which can include at least one control plane entity that can manage access and mobility (e.g., a mobility management entity (MME), an access and mobility management function (AMF)) and at least one user plane entity that can route packets or connect to external networks (e.g., a serving gateway (S-GW), a Packet Data Network (PDN) gateway (P-GW), or a user plane function (UPF)). The control plane entity can manage non-access stratum (NAS) functions such as mobility, authentication, and bearer management for UEs 115 served by network entities 105 (e.g., base stations 140) associated with the core network 130. User IP packets can be transferred through the user plane entity, which can provide IP address allocation as well as other functions. The user plane entity can be connected to the IP services 150 of the one or more network operators. The IP services 150 can include access to the Internet, Intranet, IP Multimedia Subsystem (IMS), or a Packet-Switched Streaming Service.

[0074] The wireless communications system 100 can operate using one or more frequency bands, which can be within a range of 300 Megahertz (MHz) to 300 Gigahertz (GHz). Generally, the region from 300 MHz to 3 GHz is known as the ultra-high frequency (UHF) region or decimeter band, since the wavelengths range from approximately one decimeter to one meter in length. The UHF wave s can be blocked or redirected by buildings and environmental features, but the waves can penetrate structures sufficiently for a macro cell to provide service to the UHF- enabled UEs 115 located indoors or in other obstructed areas. In contrast, the smaller frequencies and waves of the high frequency (HF) or very high frequency (VHF) portion of the spectrum below 300 MHz can penetrate buildings and diffraction around buildings and obstacles more readily, but the smaller wavelengths can be scattered by trees and other environmental features, causing poor transmission.

[0075] The wireless communications system 100 can utilize both licensed and unlicensed RF spectrum bands. For example, the wireless communications system 100 can employ LTE License Assisted Access (LAA), LTE-Unlicensed (LTE-U) radio access technology, or NR technology in an unlicensed

[0076] The network entities 105 (e.g., base stations 140, RUs 170) or UEs 115 can be equipped with multiple antennas, which can be used to employ techniques such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communication, or beamforming. The antennas of a network entity 105 or a UE 115 can be located in one or more antenna arrays or antenna panels, which can support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays can be co-located at an antenna assembly, such as an antenna tower. In some examples, the antennas or antenna arrays associated with a network entity 105 can be located at different geographic locations. A network entity 105 can include an antenna array with a set of multiple rows and multiple columns of antenna ports that the network entity 105 can use for beamforming in support of communication with UEs 115. Likewise, a UE 115 can include one or more antenna arrays, which can support various MIMO or beamforming operations. Additionally or alternatively, an antenna panel can support RF beamforming for signals transmitted via the antenna ports.

[0077] Beamforming, which can also be referred to as spatial filtering, directional transmission, or directional reception, is a signal processing technique that can be used at a transmitting device or a receiving device (e.g., a network entity 105, a UE 115) to shape or steer a beam of energy in a specific direction along a surface of a transmitting device or a receiving device. Beamforming can be achieved by combining the signals communicated by antenna elements of an antenna array such that some signals propagating at different angles experience constructive interference while others experience destructive interference. The adjustment of signals communicated by each of the antenna elements can include a transmitting device or a receiving device applying amplitude offsets, phase offsets, or both to signals carried by each of the antenna elements. The adjustments associated with each of the antenna elements can be defined by a beamforming weight set associated with a particular orientation (e.g., with respect to the antenna array of the transmitting device or receiving device, or with respect to some other orientation).

[0078] A network entity 105 or UE 115 can use beam sweeping techniques as part of a beamforming operation. For example, a network entity 105 (e.g., a base station 140, a RU 170) can use multiple antennas or antenna arrays (e.g., antenna panels) to conduct a beamforming operation for directional communications with a UE 115. Some signals (e.g., synchronization signals, reference signals, beam- selection signals, or other control signals) can be transmitted by the network entity 105 multiple times in different directions. For example, the network entity 105 can transmit the signals according to different beamforming weight sets associated with different directions of transmission. Transmissions in different beam directions can be used, for example, to identify (e.g., by a transmitting device such as a network entity 105, or by a receiving device such as a UE 115) a beam direction for subsequent transmission or reception by the network entity 105.

[0079] Some signals, such as data signals associated with a particular receiving device, can be transmitted by a transmitting device (e.g., a transmitting network entity 105, a transmitting UE 115) in a single beam direction (e.g., a direction associated with the receiving device, such as a receiving network entity 105 or a receiving UE 115). In some examples, the beam direction associated with transmissions along a single beam direction can be determined based on a signal that was transmitted in one or more beam directions. For example, a UE 115 can receive one or more of the signals transmitted by the network entity 105 in different directions, and can report to the network entity 105 an indication of the signal that the UE 115 received with a highest signal quality, or other acceptable signal quality.

[0080] In some examples, transmissions by a device (e.g., by a network entity 105 or a UE 115) can be performed using multiple beam directions, and the device can use a combination of digital precoding or beamforming to generate a combined beam for transmissions (e.g., from a network entity 105 to a UE 115). A UE 115 can report feedback that indicates precoding weights for one or more beam directions, and the feedback can correspond to a set of beams that are configured across a system bandwidth or one or more sub-bands. The network entity 105 can transmit a reference signal (e.g., a cell-specific reference signal (CRS), a channel state information reference signal (CSI-RS)), which can or can not be precoded. The UE 115 can provide feedback for beam selection, which can be a precoding matrix indicator (PMI) or codebook-based feedback (e.g., a multi-panel type codebook, a linear combination type codebook, a port selection type codebook). Although these techniques are described with reference to signals transmitted by a network entity 105 (e.g., a base station 140, a RU 170) in one or more directions, a UE 115 can use similar techniques for transmitting signals in different directions multiple times (e.g., for identifying a beam direction for subsequent transmission or reception by the UE 115), or for transmitting a signal in a single direction (e.g., for transmitting data to a receiving device).

[0081] A receiving device (e.g., a UE 115) can perform reception operations according to a number of receive configurations (e.g., directional listening) when receiving various signals from a receiving device (e.g., network entity 105), such as synchronization signals, reference signals, beam selection signals, or other control signals. For example, a receiving device can perform reception according to multiple receive configurations by using different antenna subarrays for reception, by processing received signals according to different receive

[0082] Wireless communications system 100 can be a packet-based network that operates according to a layered protocol stack. In the user plane, communications at the bearer or PDCP layer can be IP -based. A RLC layer can perform packet segmentation and reassembly to communicate over logical channels. A MAC layer can perform priority handling and multiplexing of logical channels into transport channels. The MAC layer can also use error detection techniques, error correction techniques, or both, to support retransmissions to improve link efficiency. In the control plane, the RRC layer can provide establishment, configuration, and maintenance of an RRC connection between a UE 115 and a network entity 105 or core network 130 supporting radio bearers for user plane data. The PHY layer can map transmission channels to physical channels.

[0083] In some examples of the wireless communications system 100, the network entity 105 and the UE 115 can use one or more beam management techniques to improve the capacity of wireless communications between the network entity 105 and the UE 115 (e.g., via the communication link 125). In some examples, the UE 115 and the network entity 105 can use one or more beam management techniques to improve an initial access procedure, a tracking procedure, and to identify a beam pair for wireless communications between the UE 115 and the network entity 105 (e.g., a gNB). For example, the UE 115 can operate in one or more RRC states, such as an idle state (e.g., indicated via an RRCJDLE information element (IE)), an inactive state (e.g., indicated via an RRC_inactive IE), or a connected state (e.g., indicated via an RRC_connected IE). In some examples, the network entity 105 and the UE 115 can perform an initial access procedure after the UE 115 operates in the idle state or the inactive state. For example, the network entity 105 can perform a beam sweeping procedure in which the network entity 105 can transmit a reference signal (e.g., a synchronization signal block (SSB)) using one or more of the beams (e.g., relatively wide beams, such as SSB beams) to the UE 115. The UE 115 can perform an initial access procedure, such as a contention-free random access (CFRA) procedure or a contention-based random access (CBRA) procedure, using information communicated via one or more of the SSBs. During the initial access procedure, the UE 115 can transmit a random access preamble to the network entity 105 using one or more random access occasions, for example, to establish a connection with the network entity 105.

[0084] In some examples, the UE 115 can use a tracking reference signal (TRS) for paging reception, such as at the UE 115, when the UE 115 can be operating in an idle state or an inactive state (e.g., to conserve power), where a configuration for the TRS can be provided to the UE 115 in system information. In a cell where the TRS can be used by the UE 115 when the UE 115 can be operating in an idle state or an inactive state, the availability of the configured TRS can be signaled to the UE 115 via signaling, such as LI signaling (e.g., from a network entity 105).

[0085] In some examples, such as examples where the UE 115 can be operating in a connected state, the UE 115 can receive downlink communications from the network entity 105 via directional beams, such as can be used to transmit one or more reference signals. In some cases, the established connection (e.g., which can also be referred to as a communication link 125 of a radio link or link) can be susceptible to blockage and fading, which can cause an interruption of the radio link or a radio link failure. That is, the downlink communications from the network entity 105 can drop. To reduce the likelihood of a radio link failure occurring or to recover after a radio link failure, the UE 115 can perform one or more beam management procedures, such as a beam failure prevention procedure or a beam failure recovery procedure.

[0086] For example, a UE 115 can perform a beam failure recovery procedure to reestablish a connection with a network entity 105 and select another (e.g., different) beam pair for communicating with the network entity 105. A beam pair can include a beam of the network entity 105 (e.g., a beam associated with a cell supported by the network entity 105) and a beam of the UE 115. In some examples, a beam management procedure can include one or more procedures for downlink beam management, such as beam selection (P1), transmission beam refinement for the network entity 105 (P2), and reception beam refinement for the UE 115 (P3). In some examples, P1, P2, and P3 can include transmission of one or more reference signals (such as SSBs or CSI-RSs) from the network entity 105. Additionally, a beam management procedure can include one or more other procedures for uplink beam management (e.g., U1, U2, U3), which can include transmission of uplink reference signals (e.g., sounding reference signals (SRSs)) from the UE 115. In some examples, a beam management procedure at the UE 115 or the network entity 105 (or both) can include L1 (or L2) based measurement reporting (e.g., L1-RSRP reporting, L1-SINR reporting), transmission configuration indicator (TCI) state configuration (e.g., indication from the network entity 105), component carrier group (CC group) beam update, relatively fast uplink beam update, unified TCI state reporting, L1- or L2-centric mobility reporting, dynamic TCI update, uplink multi-panel selection, and maximum permitted exposure (MPE) mitigation, and other possible examples that can result in reduced beam management latency. The UE 115 and the network entity 105 can support one or more beam management techniques for high speed train (HST), single frequency network (SFN), and multi-TRP (mTRP) deployments, among other examples.

[0087] In some examples, a UE 115 can detect an interruption in a radio link or detect a radio link failure based on a measurement, such as a measurement of a beam failure detection reference signal (BFD-RS) or a physical downlink control channel (PDCCH) block error rate (BLER) measurement. In such examples, the UE 115 can perform a recovery procedure (e.g., a beam failure recovery procedure) to reduce a link interruption time or a link failure time. The recovery procedure can be for a primary cell (PCell), a primary cell in a secondary cell group (PSCell), or a secondary cell (SCell). In some examples, the recovery procedure can be based on a random access procedure (e.g., CFRA). Additionally, in some examples, the recovery procedure can include transmission of a link recovery request (e.g., via a scheduling request). In some examples, the recovery procedure can be a MAC control element (MAC-CE) based beam failure recovery procedure (e.g., for a SCell).

[0088] In some examples, the UE 115 or the network entity 105, or both, can support AI / ML-based beam management. For example, the UE 115 and the network entity 105 can support one or more techniques for predictive beam management using AI / ML. In some examples, the UE 115 (or the network entity 105) can evaluate support for one or more AI / ML-based beam management techniques for characterization and performance (e.g., baseline performance). For example, the UE 115 can support AI / ML-based beam management for performance monitoring. An AI / ML-based beam management technique can include spatial domain downlink beam prediction. For example, the UE 115 can use AI / ML to predict measurements of a first set of downlink beams (e.g., a prediction target, which can be referred to as Set A) based on measurements (e.g., actual measurements) of reference signals transmitted to the UE 115 using a second set of downlink beams (e.g., a measurement source, which can be referred to as Set B). For example, the UE 115 can use AI / ML to predict measurements of a first set of beams (e.g., Set A) based on measurements (e.g., actual measurements) of reference signals transmitted to the UE 115 using a second set of beams (e.g., Set B). The predicted measurements and the actual measurements can include reference signal received power (RSRP) measurements or signal-to-interference-plus-noise (SINR) measurements, among other possible examples of received power measurements. In other words, the predicted measurements and the actual measurements can include received power metrics, such as RSRP values or SINR values. In some examples, one or more beams can be common to Set A and Set B. For example, the network entity 105 can transmit a set of reference signals to the UE 115 using one or more beams, and the UE 115 can predict measurements of the same one or more beams or different one or more beams (e.g., based on measurements of the transmitted set of reference signals).

[0089] In some examples, in the spatial domain, set A can correspond to a first set of reference signal resources (e.g., SSB resources or CSI-RS resources), and set B can correspond to a second set of reference signal resources (e.g., CSI-RS resources or SSB resources). That is, for spatial domain downlink beam prediction, a UE 115 can predict measurements of a first set of reference signal resources (e.g., based on actual measurements of a second set of reference signal resources). A reference signal resource (e.g., each reference signal resource) included in the first set of reference signal resources can correspond to a respective beam included in a first set of beams (e.g., set A). Additionally, the predicted measurements can be based on actual measurements of a set of reference signals transmitted using the second set of reference signal resources. A reference signal resource (e.g., each reference signal resource) included in the second set of reference signal resources can correspond to a respective beam (e.g., used to transmit the corresponding reference signal) included in a second set of beams (e.g., set B). In some other examples, set A can include a subset (e.g., a downsampled version) of set B. That is, the first set of reference signal resources (e.g., the first set of beams) can include a subset of the second set of reference signal resources (e.g., the second set of beams).

[0090] Another AI / ML-based beam management technique can include time domain downlink beam prediction. For example, a UE 115 can use AI / ML to predict measurements (e.g., RSRP measurements, SINR measurements) of a first set of beams (e.g., set A) based on previous (e.g., historical) measurements of a second set of beams (e.g., set B). In some examples, set A can correspond to a set of reference signal resources at a first time occasion, and set B can correspond to the same set of reference signal resources at a second time occasion (e.g., a previous time occasion). In some other examples, set A can correspond to a first set of reference signal resources, and set B can correspond to a second set of reference signal resources that can be different from the first set of reference signals. For example, the second set of reference signals can correspond to SSB resources (e.g., the UE 115 can perform measurements of SSBs transmitted using relatively wide beams), and the first set of reference signals can correspond to CSI-RS resources (e.g., the UE 115 can predict measurements of CSI-RSs that can be transmitted using relatively narrow beams). In some examples, the beams in set A and set B can be within the same frequency range. That is, the first set of reference signal resources and the second set of reference signal resources can include frequencies within the same frequency range.

[0091] In some examples, the UE 115 can be configured to determine a respective number of beams (e.g., reference signal resources) to be included in set A and set B. Additionally, the UE 115 can select set A from the beams (e.g., reference signal resources) in set B (e.g., according to a fixed pattern, a random pattern). For example, the UE 115 can select set A from set B based on the determined number of beams to be included in set A. That is, set A can be a subset of set B. In some examples, the UE 115 can be configured to determine whether set A and set B are to be different (e.g., set A can include relatively narrow beams and set B can include relatively wide beams). Accordingly, the UE 115 can determine a quasi-co-location (QCL) relationship between the beams in set A and the beams in set B. In some examples, set A can be used for downlink beam prediction and set B can be used for downlink beam measurement. Additionally, in some examples, the UE 115 can be configured with one or more codebook constructions for set A and set B.

[0092] In some examples, such as for spatial domain beam prediction, the wireless communications system 100 can support beam management using UE-side AI / ML models that can include Ll signaling from the UE 115 for reporting information associated with AI / ML model inferences (e.g., predictions) to the network entity 105. In such examples, one or more beams for downlink communications with the UE 115 can be based on AI / ML model inferences. That is, one or more beams for downlink communications with the UE 115 can be based on an output of an AI / ML model inference at the UE 115. In some examples, the UE 115 can report predicted Ll-RSRP measurements (or Ll-SINR measurements) corresponding to one or more beams (e.g., one or more reference signal resources).

[0093] In some other examples, such as for time-domain prediction, wireless communications system 100 can support beam management using a UE-side AI / ML model that can include L1 signaling from UE 115 for reporting information associated with AI / ML model inference to network entity 105. In such examples, one or more beams (e.g., reference signal resources) at a number (N) of future time instances (e.g., time occasions) can be based on AI / ML model inference. That is, one or more beams for downlink communications with UE 115 at a number of future time occasions can be based on an output of AI / ML model inference at UE 115. In some examples, UE 115 can be configured with a value N. Additionally, for time-domain prediction, wireless communications system 100 can support beam management using a UE-side AI / ML model that can include L1 signaling from UE 115 for reporting information associated with AI / ML model inference to network entity 105. In some examples, one or more beams (e.g., reference signal resources) at a number (N) of future time instances (e.g., time occasions) can be based on an output of AI / ML model inference (e.g., at UE 115). In some examples, UE 115 can be configured with a value N. In some examples, UE 115 can report predicted L1-RSRP measurements that can correspond to one or more beams (e.g., one or more reference signal resources). In such examples, UE 115 can also report information regarding a timestamp corresponding to the reported one or more beams (e.g., the reported one or more reference signal resources). The timestamp information can be indicated explicitly or implicitly via a report (e.g., a report for reporting information associated with one or more beams).

[0094] In some examples, such as for spatial domain prediction and time domain prediction with UE-side AI / ML models, wireless communications system 100 can support model monitoring with potential down-selection. For example, wireless communications system 100 can support UE-side model monitoring, where a UE 115 can monitor performance metrics associated with an AI / ML model or with wireless communications between the UE 115 and a network entity 105 (or both). In some examples, the UE 115 can make one or more determinations regarding model selection, activation, deactivation, switching, and fallback operations, among other examples. Additionally or alternatively, wireless communications system 100 can support network-side model monitoring, where a network entity 105 can monitor performance metrics associated with an AI / ML model or with wireless communications between the UE 115 and the network entity 105 (or both). Additionally, in some examples, the network entity 105 can make one or more determinations regarding model selection, activation, deactivation, switching, and fallback operations, among other examples. Wireless communications system 100 can support hybrid model monitoring, where a UE 115 can monitor one or more performance metrics and a network entity 105 can make one or more determinations regarding model selection, activation, deactivation, switching, and fallback operations.

[0095] In some examples, such as for spatial domain prediction or time domain prediction with UE-side AI / ML models and network-side model monitoring, a network entity 105 can monitor one or more performance metrics and make one or more determinations regarding model selection, activation, deactivation, switching, and fallback operations. Additionally, in some examples of network-side model monitoring for network-side AI / ML models (e.g., for spatial domain prediction and for time domain prediction), a UE 115 can be configured to perform beam measurements and transmit a report for model monitoring. In some examples, such as for spatial domain prediction or time domain prediction with network-side AI / ML models, a UE 115 can support one or more Ll beam report enhancements for AI / ML model inference. For example, a UE 115 can report measurements for multiple (e.g., more than 4) beams in one reporting instance. That is, a UE 115 can report measurements for multiple (e.g., more than 4) reference signal resources in one reporting instance.

[0096] In some examples, the UE 115 can use AI / ML to improve performance of some functionality at the UE 115, such as beam prediction. For example, the UE 115 can use AI / ML to make predictions associated with transmit beams (e.g., downlink beams) at the network entity 105 and report such predictions to the network entity 105 to improve beam management (e.g., at the network entity 105). The network entity 105 can assist the UE 115 in managing the lifecycle of one or more AI / ML models. That is, the network entity 105 enables (or otherwise supports) AI / ML model LCM at the UE 115 by indicating to the UE 115 to activate or deactivate one or more AI / ML models based on observations at the network entity 105. For example, the network entity 105 can indicate to the UE 115 to activate or deactivate an AI / ML model at the UE 115 for beam prediction. However, in some examples, to enable the network entity 105 to indicate to the UE 115 to activate or deactivate an AI / ML model, the UE 115 can provide information associated with the AI / ML model to the network entity 105, which can result in a decrease in security at the UE 115. In other words, enabling model-based LCM for AI / ML operations at the UE 115 can result in sensitive information being disclosed to the network entity 105.

[0097] In some other examples, the network entity 105 can indicate to the UE 115 to validate or invalidate functionality at the UE 115. For example, the network entity 105 enables (or otherwise supports) AI / ML model LCM at the UE 115 by indicating to the UE 115 to validate or invalidate functionality (e.g., beam prediction), which can result in the UE 115 activating or deactivating an AI / ML model for the functionality (e.g., for beam prediction at the UE 115). In such examples, the network entity 105 can achieve activation or deactivation of the AI / ML model and the UE 115 can reduce (e.g., avoid) the likelihood of sensitive information being disclosed to the network entity 105. However, in some examples, aspects of the functionality (e.g., how the functionality is defined) can be unclear to the UE 115 or the network entity 105 or both. For example, the indication of the functionality can be ambiguous to the UE 115. Thus, using validation or invalidation of the functionality to achieve activation or deactivation of the AI / ML model can be relatively ineffective and degrade LCM of the AI / ML model at the UE.

[0098] In some examples of the wireless communication system 100, the UE 115 and the network entity 105 can support a framework for indicating characteristics associated with a training dataset to enable AI / ML model activation or deactivation or functionality enablement or disablement. For example, the UE 115 can receive control information from the network entity 105. The control information can indicate characteristics of a dataset used to train an AI / ML model at the UE 115. In some examples, the AI / ML model can be associated with maintaining a wireless communication link. The UE 115 can determine, based on receiving the control information, to activate or deactivate a first AI / ML model used to maintain the wireless communication link. Alternatively, the UE 115 can determine, based on receiving the control information, to use functionality enablement or disablement of a second AI / ML model used to maintain the wireless communication link. The UE 115 can transmit feedback information to the network entity 105 based on determining to activate or deactivate the first AI / ML model or to enable or disable functionality of the second AI / ML model. The feedback information can indicate one or more parameters of the first AI / ML model or one or more parameters of functionality of the second AI / ML model.

[0099] In some examples, as described herein, the indication of a training dataset for LCM can provide improvements to beam management at the UE 115 or the network entity 105 (or both). For example, one or more aspects of adaptive CSI reporting for predictive beam management can provide a framework for AI / ML beam prediction for an air interface (e.g., wireless communication) that can result in improvements in performance and reductions in complexity (e.g., of beam management). The framework can include beam prediction in the time domain or the spatial domain (or both), which can provide reductions in overhead and latency and improvements in beam selection accuracy. In some examples, the framework can enable characterization and baseline performance evaluation using AI / ML. Thus, the framework can provide AI / ML approaches that can be relatively diverse and support constraints on a level of cooperation between the UE 115 and the network entity 105. In some examples, as described herein, the indication of a training dataset for LCM can provide characterization of LCM of AI / ML models, including model training, model deployment, model inference, model monitoring, model updating. In other words, adaptive CSI reporting for predictive beam management can be used for AI-based beam prediction performance monitoring.

[0100] Figure 2 An example of a wireless communications system 200 that supports indication of a training dataset for LCM is shown, in accordance with one or more aspects of the present disclosure. In some examples, the wireless communications system 200 can implement, or can be implemented at, one or more aspects of the wireless communication system 100. For example, the wireless communications system 200 can include a UE 215, which can be a UE 115, as described by and with reference to FIG. 1.Figure 1 An example of the UE 115 (or another network node) that is exemplified and described. The wireless communications system 200 can also include a network entity 205, which can be an example of the network entity 105 by way of Figure 1 An example of one or more of the network entities 105 (e.g., a CU, a DU, a RU, a base station, an IAB node, or one or more other network nodes) that is exemplified and described. The UE 215 and the network entity 205 can communicate with a coverage area 210, which can be an example of the coverage area 110 by way of Figure 1 An example of the coverage area 110 that is exemplified and described. For example, the UE 215 and the network entity 205 can communicate within the coverage area 210 via a communication link 220, which can be an example of the communication link 125 by way of Figure 1 An example of the communication link 125 (e.g., a Uu link) that is exemplified and described.

[0101] The wireless communications system 200 can support a UE-side AI / ML model (e.g., an AI / ML model deployed at the UE 215) for one or more functionalities, such as beam management. For example, the UE 215 can support an AI / ML model for time-domain beam prediction and spatial-domain beam prediction, among other examples. In some examples, the UE 215 or the network entity 205, or both, can monitor a performance (e.g., one or more performance metrics) of the UE 215 or the AI / ML model deployed at the UE 215. In such examples (e.g., based on the monitoring), the network entity 205 or the UE 215, or both, can make one or more determinations regarding AI / ML model selection, activation, deactivation, switching, and fallback operations (e.g., fallback operations at the UE 215 regarding one or more AI / ML models). Additionally or alternatively, the UE 215 or the network entity 205, or both, can make one or more determinations regarding functionality selection, validation, invalidation, switching, and fallback operations (e.g., based on monitoring a performance of the UE 215 or the AI / ML model at the UE 215). In other words, the UE 215 and the network entity 205 can support LCM for AI / ML operations at the UE 215 (e.g., for UE-side model LCM).

[0102] For example, the UE 215 and the network entity 205 can support model-based LCM for AI / ML operations at the UE 215. In some examples, such as to support model-based LCM, the network entity 205 can obtain information associated with AI / ML models at the UE 215. For example, the network entity 205 can obtain information associated with parameters and structures of the AI / ML models or datasets used to train the AI / ML models, or both. For example, the network entity 205 can be configured with an association (e.g., correspondence, mapping) between information (e.g., parameter or structure information) associated with one or more AI / ML models deployed at the UE 215, one or more identifiers (IDs) corresponding to the one or more AI / ML models, or one or more functionalities associated with the AI / ML models, or any combination thereof. The one or more functionalities can include one or more scenarios, one or more UE capabilities, or other information or parameters that can be associated with the AI / ML models deployed at the UE 215.

[0103] In some examples, the network entity 205 can use information obtained for the AI / ML model (e.g., parameter or structural information, ID, functionality) to make a determination for activation, deactivation, or switching of the AI / ML model (and instruct the UE 215 to activate, deactivate, or switch the AI / ML model). That is, the network entity 205 can instruct the UE 215 to activate, deactivate, or switch an AI / ML model (e.g., a particular AI / ML model) for functionality. In other words, the network entity 205 can instruct the UE 215 to activate, deactivate, or switch an AI / ML model for a task, such as beam prediction. For example, the UE 215 can be configured to make predictions according to a first time increment (e.g., 20 ms) and a second time increment (e.g., 200 ms). In this example, the network entity 205 can instruct the UE 215 to use a first AI / ML model for predictions according to the first time increment and a second AI / ML model for predictions according to the second time increment. For example, the network entity 205 can instruct the UE 215 of a first AI / ML model ID corresponding to the first AI / ML model for predictions according to the first time increment (e.g., for a first functionality, for a first task) and a second AI / ML model ID corresponding to the second AI / ML model for predictions according to the second time increment (e.g., for a second functionality, for a second task). That is, the UE 215 and the network entity 205 can support model-based LCM, where the network entity 205 can obtain information associated with one or more AI / ML models deployed at the UE 215 (e.g., can be transparent to the one or more AI / ML models, partially aware of the one or more AI / ML models, fully aware of the one or more AI / ML models). Thus, the AI / ML model can be activated, deactivated, or switched by the network entity 205 (e.g., directly, such as via signaling). However, in some examples, providing the network entity 205 with information associated with AI / ML models deployed at the UE 215 can result in reduced security at the UE 215 (e.g., due to possible disclosure of sensitive information). In other words, model-based LCM can result in sensitive information (e.g., UE-specific information) being disclosed to the network entity 205.

[0104] In some other examples, the UE 215 and the network entity 205 can support a functional-based LCM for AI / ML operations at the UE 215. For example, the UE 215 can support (e.g., enable, use, participate in) one or more functionalities, which can be associated with one or more AI / ML models. That is, the UE 215 can use AI / ML for one or more functionalities. In some examples, the functionalities can include (or otherwise be associated with) sub-functionalities and sub-sub-functionalities. That is, the UE 215 can support functionalities with multiple levels (e.g., multi-level functionalities). For example, the UE 215 can support a beam prediction functionality, which can include one or more sub-functionalities, such as spatial domain beam prediction and time domain beam prediction. In some examples, the functionalities can include (or be associated with, e.g., higher than) UE capability features or conditions, such as over-the-air (OTA) conditions that can trigger or cause the functionalities to take effect. In other words, the functionalities can include UE capabilities that are not reported to the network entity 205 (e.g., beyond the UE reported capabilities), such as UE mobility situations, speed information, or channel profile information, among other examples.

[0105] In some examples, such as for a functional-based LCM, the UE 215 can reduce the likelihood that information associated with the AI / ML model deployed at the UE 215 is disclosed to the network entity 205. In such examples, the network entity 205 can lack control over the AI / ML model deployed at the UE 215. In other words, the network entity 205 can support a functional-based LCM in which an AI / ML feature (or AI / ML use case) can be defined as a functionality (e.g., a multi-level functionality). In such examples, the network entity 205 can indicate (e.g., implicitly indicate) the UE 215 to activate, deactivate, or switch the AI / ML model by indicating (e.g., explicitly indicating) to the UE 215 to validate, invalidate, or switch the functionality. In other words, AI / ML model activation, deactivation, or switching can be achieved through functionality validation, functionality invalidation, or functionality switching. For example, the network entity 205 can indicate to the UE 215 to validate or invalidate a beam prediction (e.g., a functionality), which can result in the UE 215 activating or deactivating an AI / ML model for beam prediction. In such examples, the network entity 205 can achieve activation or deactivation of the AI / ML model at the UE 215, and the UE 215 can reduce the likelihood that sensitive information is disclosed to the network entity 205. However, in some examples, the UE 215 and the network entity 205 can lack an agreement (e.g., a consensus) on how one or more functionalities can be defined. That is, aspects of the functionality (e.g., how the functionality is defined) can be unclear to the UE 215 or the network entity 205, or both. Thus, the indication of the functionality from the network entity 205 (e.g., an instruction to validate, invalidate, or switch the functionality) can be ambiguous to the UE 215 and reduce the performance of the LCM at the UE 215.

[0106] In some examples, one or more techniques of indicating a training dataset for LCM, as described herein, can provide a framework for implementing AI / ML model activation, deactivation, or switching, or functionality enablement, disablement, or switching. For example, use of an AI / ML model can be associated with a dataset used to train the AI / ML model (e.g., a training dataset). That is, the AI / ML model and functionality of the AI / ML model can be associated with a dataset used to train the AI / ML model. Accordingly, a network entity 205 can use a characteristic of a dataset used to train an AI / ML model (e.g., a training dataset) to indicate to a UE 215 to activate, deactivate, or switch an AI / ML model, or to enable, disable, or switch functionality of an AI / ML model (e.g., functionality that the AI / ML model can be applicable to). In other words, one or more techniques of indicating a training dataset for LCM, as described herein, can implement a framework for indication of a training dataset for functional or model-based LCM in AI / ML operations.

[0107] In some examples, indicating a characteristic associated with a training dataset can implement (e.g., implicitly implement) AI / ML model activation, deactivation, or switching, or functionality enablement, disablement, or switching. For example, a network entity 205 can indicate a characteristic of a first dataset to a UE 215. In such an example, the UE 215 can use the indicated characteristic to identify an AI / ML model trained using a second dataset. The second dataset can be the same dataset as the first dataset. Alternatively, the second dataset is different from the first dataset. For example, the second dataset can include the indicated characteristic or another characteristic that is similar to the indicated characteristic. In other words, a network entity 205 can signal a characteristic (e.g., a detail) of a training dataset that can identify one or more AI / ML models for LCM. For example, a UE 215 can select an identified AI / ML model (e.g., an AI / ML model trained using a second dataset) for activation, deactivation, or switching. In some examples, a network entity 205 can signal a characteristic (e.g., a detail) of a training dataset that can identify one or more AI / ML models for one or more functionalities. For example, a network entity 205 can signal a characteristic of a training dataset that can identify one or more AI / ML models for beam prediction. In some examples, by indicating a characteristic of a training dataset, a UE 215 can reduce a likelihood of sensitive information (e.g., UE-specific information, such as for model-based LCM) being disclosed to a network entity 205, and reduce ambiguity associated with an indication from a network entity 205 (e.g., provide a more clear definition than a definition available for a functionality).

[0108] The UE 215 and the network entity 205 can support AI / ML model LCM for training dataset identification. As Figure 2 As illustrated in the example of FIG. 2, the UE 215 can support one or more AI / ML models (e.g., model 235-a, model 235-b) that can be associated with one or more functionalities (e.g., functionality 240-a, functionality 240-b). For example, the UE 215 can use the model 235-a for the functionality 240-a, or the model 235-a can be otherwise associated with the functionality 240-a. Additionally, the UE 215 can use the model 235-b for the functionality 240-b, or the model 235-b can be otherwise associated with the functionality 240-b. In some examples, the UE 215 can receive control information 225 (e.g., a network indication) regarding a characteristic of a dataset (e.g., a particular training dataset). That is, the control information 225 can indicate a characteristic 245 of a dataset used to train an AI / ML model at the UE 215. In some examples, the AI / ML model can be associated with one or more functionalities, such as a functionality associated with maintaining the communication link 220 (e.g., a wireless communication link). In some examples, the control information 225 (e.g., the network indication) can be carried via RRC, MAC-CE, or downlink control information (DCI). In some other examples, the control information 225 can be carried via one or more other upper layer protocols, such as via application layer signaling. In other words, the UE 215 can receive RRC layer signaling, PHY layer signaling, MAC layer signaling, or application layer signaling that includes the control information 225.

[0109] In some examples, the characteristic 245 (e.g., a general characteristic) can be associated with (e.g., satisfied by, included in) one or more datasets. For example, the characteristic 245 can be associated with multiple types of datasets. In other words, the characteristic 245 can be an example of a characteristic that is associated with two or more datasets, such as a dataset used to train the model 235-a and a dataset used to train the model 235-b. In some other examples, the characteristic 245 can be associated with a single dataset. For example, multiple (e.g., different) datasets can be associated with multiple (e.g., different) characteristics. In some examples, the characteristic 245 can include a scenario in which the UE 315 or the network entity 305, or both, can operate, which can also be referred to as an operational scenario or a deployment scenario. For example, the characteristic 245 can include a dense urban operational scenario, an indoor operational scenario, or a rural operational scenario.

[0110] In some examples, feature 245 may include a feature of the wireless communication link profile (e.g., a wireless communication link profile feature). For example, feature 245 may include the range of delay spread or Doppler spread associated with the wireless communication link (e.g., or the wireless communication channel). In some examples, feature 245 may include one or more parameters associated with network entity 305 or a cell (such as a cell served by network entity 305). For example, feature 245 may include one or more features of a cell providing coverage area 210 that may serve wireless communication between network entity 205 and UE 215 (e.g., may serve communication link 220). In some examples, feature 245 may include transmission parameters for downlink communication, such as the size, type, or orientation of one or more antenna arrays at network entity 205 (e.g., gNB), the number of beams in a codebook (e.g., a codebook configured for downlink communication at network entity 205), the transmission power at network entity 205 (e.g., gNB transmission power), or one or more capabilities of network entity 205. In some examples, feature 245 may include transmission parameters for uplink communication, such as the size, type, or orientation of one or more antenna arrays at UE 215, the number of beams in a codebook (e.g., a codebook configured for downlink communication at UE 215), the transmission power at UE 215 (e.g., UE transmission power), or one or more capabilities of UE 215. In some examples, feature 245 may include the location of UE 215 relative to network entity 205 (e.g., whether UE 215 is relatively close to or relatively far from network entity 205).

[0111] In some examples, feature 245 may include a feature associated with one or more datasets used for beam prediction AI / ML models. That is, the datasets may be used to train beam prediction AI / ML models. In some examples, a feature may be associated with multiple beam prediction AI / ML models. In other words, a feature used for beam prediction AI / ML models (e.g., including the feature) may be included in two or more datasets (e.g., common to the two or more datasets). In some cases, multiple (e.g., different) datasets used for beam prediction models may be associated with multiple (e.g., different) features. In some examples, feature 245 may include the distribution of measured or reported received power measurements (or both) (e.g., maximum, minimum, mean, variance, standard deviation, probability distribution function, cumulative distribution function). For example, feature 245 may include the distribution of L1 RSRP measurements or L1 SINR measurements to be used as input to the AI / ML model or as a prediction target for the AI / ML model. In other words, feature 245 may include statistics associated with a received power measurement used as input to an AI / ML model or with a received power measurement used as a prediction target of the AI / ML model. In some examples, feature 245 may include the reliability or accuracy of the received power measurement to be used as input to an AI / ML model or as a prediction target of the AI / ML model. That is, feature 245 may include a performance metric associated with a received power measurement used as input to an AI / ML model or with a received power measurement used as a prediction target of the AI / ML model.

[0112] In some examples, feature 245 may include UE mobility characteristics, such as the direction in which UE 215 may be moving (e.g., direction of movement), the speed at which UE 215 may be moving (e.g., speed of movement), the direction in which UE 215 may be rotating (e.g., direction of rotation), the speed at which UE 215 may be rotating (e.g., speed of rotation), or the orientation of UE 215. In some examples, feature 245 may include characteristics of the transmit beam at network entity 205. For example, feature 245 may include the transmit beam shape (e.g., a range of beam pointing directions or beamwidths for measuring resources or predicting targets). In some examples, feature 245 may include characteristics of the receive beam at UE 215. For example, feature 245 may include the receive beam shape (e.g., a range of beam pointing directions or beamwidths for measuring received power, such as L1-RSRP or L1-SINR).

[0113] In some examples, such as in response to receiving control information 225, UE 215 may determine to activate or deactivate model 235-a (e.g., for maintaining communication link 220). For example, UE 215 may activate, deactivate, or switch the application of an AI / ML model to model 235-a in association with indicated features of the training dataset (e.g., control information 225). In other words, UE 215 activates or deactivates model 235-a in association with feature 245 (e.g., features of the dataset). In some other examples, UE 215 may determine to enable or disable functionality (e.g., beam prediction) of a model or another model (e.g., the activated or switched model). For example, model 235-b may be activated, and UE 215 may determine to enable or disable functionality 240-b of model 235-b (e.g., for maintaining communication link 220). That is, UE 215 can apply AI / ML functionality to enable, disable, or switch functionality 240-b in association with the indicated characteristics of the training dataset (e.g., control information 225). In other words, UE 215 activates or deactivates functionality 240-b in association with characteristic 245 (e.g., characteristics of the dataset).

[0114] In some examples, UE 215 may send feedback information 230 to network entity 205 based on this determination (e.g., based on determining to activate or deactivate model 235-a or to enable or disable functionality 240-b of model 235-b). Feedback information 230 may indicate one or more parameters of model 235-a or functionality 240-b. For example, UE 215 may send feedback information 230 to network entity 205 based on determining to activate or deactivate model 235-a. In this example, feedback information 230 may indicate one or more parameters (e.g., characteristics) associated with model 235-a (e.g., the model selected by UE 215 for activation, deactivation, or switching). In some examples, feedback information 230 may indicate a correspondence (e.g., similarity level) between characteristic 245 (e.g., data characteristics identified by the UE) and one or more parameters of model 235-a (e.g., the model selected by UE 215). Additionally or alternatively, UE 215 may send feedback information 230 to network entity 205 based on determining whether to enable or disable functionality 240-b. In this example, feedback information 230 may indicate one or more parameters (e.g., characteristics) associated with functionality 240-b (e.g., the functionality selected by UE 215 for an activated or switched model, the functionality UE 215 selects to enable, disable, or switch). In some examples, feedback information 230 may indicate a correspondence (e.g., similarity level) between characteristic 245 (e.g., data characteristics identified by the UE) and one or more parameters of functionality 240-b (e.g., the functionality selected by UE 215). In some examples, by indicating feedback information 230 to network entity 205, UE 215 may improve the LCM of the AI / ML model and other benefits.

[0115] Figure 3 An example of process flow 300 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. In some examples, process flow 300 may implement one or more aspects of wireless communication system 100 and wireless communication system 200. For example, process flow 300 may include example operations associated with network entity 305 and UE 315, which may be implemented by means of and reference to... Figure 1 and Figure 2Examples of corresponding devices are illustrated and described. Operations performed by network entity 305 and UE 315 can support improvements in communication between UE 315 and network entity 305, as well as other benefits. In the following description of process flow 300, operations between UE 315 and network entity 305 may be performed in a different order than the example order shown. Additionally or alternatively, operations performed by UE 315 and network entity 305 may be performed in a different order or at different times. Some operations may also be omitted or combined. UE 315 and network entity 305 can support a framework for indicating the activation or deactivation, or functional activation or deactivation, of features associated with the training dataset to achieve ML model (e.g., AI / ML model) activation or deactivation.

[0116] At 325, UE 315 can receive control information from network entity 305. The control information can be obtained through and referenced... Figure 1 and Figure 2 Examples of control information illustrated and described. For instance, control information might indicate characteristics of the dataset used to train an ML model (e.g., an AI / ML model) at UE 315. These characteristics could be defined by and reference to... Figure 1 and Figure 2 Examples of illustrative and descriptive features. For instance, a feature may be associated with a dataset used to train an AI / ML model (such as multiple types of AI / ML models (e.g., training AI / ML models for multiple types of functionality) or for one or more functional AI / ML models). For example, a feature may be associated with a dataset used to train an AI / ML model for beam prediction. The AI / ML model may be obtained by referencing... Figure 1 and Figure 2 Examples of AI / ML models are illustrated and described. For instance, an AI / ML model could be associated with maintaining a wireless communication link.

[0117] Network entity 305 may support one or more techniques (e.g., methods) for indicating datasets (e.g., reference datasets, training datasets) to UE 315. For example, UE 315 or network entity 305, or both, may be configured with multiple datasets (e.g., reference datasets, training datasets) for training AI / ML models at UE 315. Multiple datasets (e.g., each of multiple datasets) may be associated with one or more features. Additionally, multiple datasets may be associated with multiple dataset IDs (e.g., defined by or utilizing the multiple dataset IDs). That is, each dataset may be associated with a corresponding dataset ID and one or more corresponding features. In this example, network entity 305 may indicate to UE 315 the dataset ID associated with the dataset. In other words, network entity 305 may send an indication including one or more dataset IDs corresponding to (e.g., identifying) one or more datasets. That is, network entity 305 may indicate (e.g., directly, explicitly) the features of a dataset via the dataset ID. In other words, the control information received at UE 315 (e.g., at 325) may include a dataset ID corresponding to a dataset, which may be associated with the feature (e.g., including the feature, satisfying the feature).

[0118] In some other examples, UE 315 or network entity 305, or both, may be configured with multiple features (e.g., dataset features). These multiple features may be associated with multiple feature IDs (e.g., defined by or utilizing these feature IDs). That is, each feature may be associated with a corresponding feature ID. In other words, UE 315 or network entity 305 may be configured with one or more features (e.g., feature options) of a training dataset (e.g., a reference training dataset) that can be associated with a feature ID (e.g., feature option ID). In this example, network entity 305 may indicate the feature ID associated with the feature to UE 315. In other words, network entity 305 may send an indication that includes one or more feature IDs (e.g., one or more feature options) and, in some examples, a value associated with the corresponding feature (e.g., a corresponding feature value). That is, network entity 305 may indicate (e.g., directly, explicitly) the feature via an indication of the corresponding feature ID. In other words, control information received at UE 315 (e.g., at 325) may include feature IDs corresponding to the features of the dataset. In some examples, UE 315 may identify one or more AI / ML models trained using a dataset that includes features corresponding to (or otherwise associated with) the indicated feature ID. In some examples, the identified AI / ML model may be deactivated at UE 315. In such examples, UE 315 may determine whether to activate the identified AI / ML model.

[0119] In some examples, network entity 305 may configure (e.g., RRC configuration) UE 315 using a subset of IDs, which may include one or more dataset IDs or one or more feature IDs or any combination thereof. For example, network entity 305 may configure a subset including one or more datasets (e.g., and one or more corresponding features) or one or more features of one or more datasets or any combination thereof. In some examples, network entity 305 may configure the subset via a subset of IDs, which includes (e.g., corresponding to one or more datasets) one or more dataset IDs or (e.g., corresponding to one or more features) one or more feature IDs or any combination thereof. In some examples, network entity 305 may configure the subset (e.g., indicating a subset of IDs) via RRC signaling or via one or more other upper-layer protocols. In such examples, network entity 305 may indicate to UE 315 the feature ID or dataset ID associated with the configured subset. In other words, network entity 305 may send an indication including one or more feature IDs (e.g., one or more feature options) or one or more dataset IDs included in the configured subset. That is, the control information received at UE 315 (e.g., at 325) may include a feature ID corresponding to a feature or a dataset ID corresponding to a dataset, or both, and the feature ID or dataset ID or both may be included in a configured subset.

[0120] In some examples, at 320, UE 315 may send feature or dataset recommendations to network entity 305. For example, UE 315 may send an uplink message to network entity 305 that includes one or more dataset IDs corresponding to one or more recommended datasets, one or more feature IDs corresponding to one or more recommended features, or both. In this example, control information (e.g., received at UE at 325) may be based on feature or dataset recommendations (e.g., the uplink message). In other words, UE 315 may recommend (or request) one or more dataset IDs or one or more feature IDs or any combination thereof to network entity 305. For example, UE 315 may report to network entity 305 a subset of one or more recommended datasets (e.g., and corresponding features, such as via dataset IDs) or one or more recommended features (e.g., via corresponding feature IDs) or both.

[0121] In some examples, UE 315 may expect to receive control information (e.g., network indication) associated with a recommended subset (e.g., options recommended by UE 315). In such examples, network entity 305 may indicate to UE 315 a feature ID or dataset ID associated with the recommended subset. In other words, network entity 305 may send an indication that includes one or more feature IDs (e.g., one or more feature options) or one or more dataset IDs included in the recommended subset. That is, the control information received at UE 315 (e.g., at 325) may include a feature ID corresponding to a feature or a dataset ID corresponding to a dataset, or both, and the feature ID or dataset ID, or both, may be included in the recommended subset.

[0122] In some examples, at 330, UE 315 can determine whether to activate or deactivate a first AI / ML model for maintaining the wireless communication link based on received control information. For example, UE 315 can activate or deactivate the first AI / ML model in association with characteristics of the dataset. In some examples, the AI / ML model can be used for beam prediction. That is, the dataset can be used to train the beam prediction AI / ML model. In such an example, UE 315 can activate the first AI / ML model and use it to predict the transmit beam at network entity 305 or the receive beam at UE 315 (e.g., for maintaining the wireless communication link).

[0123] In some examples, UE 315 may use a first AI / ML model to make predictions based on a first time increment (e.g., 200 ms) (e.g., predicting the transmit beam at network entity 305 or the receive beam at UE 315). In some examples, control information may indicate a dataset ID for a dataset used to train another AI / ML model to make predictions based on a second time increment (e.g., 20 ms). In such an example, UE 315 may determine to deactivate the first AI / ML model and activate another AI / ML model that can be used to make predictions based on the second time increment (e.g., 20 ms) or a third time increment that is relatively similar to the second time increment (e.g., the third time increment may have a value closer to 20 than 200). In some other examples, UE 315 may determine to modify (e.g., switch, change) the first AI / ML model such that the first AI / ML model can be used to make predictions based on the second time increment (e.g., 20 ms) or a third time increment that is relatively similar to the second time increment.

[0124] Additionally or alternatively, at 335, UE 315 may determine, based on received control information, whether the functionality of the second AI / ML model to be used for maintaining the wireless communication link is enabled or disabled. For example, UE 315 may enable or disable the functionality of the second AI / ML model in association with characteristics of the dataset. In some examples, the second AI / ML model may be used for beam prediction. That is, the dataset may be used to train the beam prediction AI / ML model. In such examples, UE 315 may use the second AI / ML model to predict the transmit beam at network entity 305 or the receive beam at UE 315 (e.g., for maintaining the wireless communication link) based on enabling the functionality of the second AI / ML model.

[0125] In some examples, UE 315 may use a second AI / ML model to make predictions (e.g., predicting the transmit beam at network entity 305 or the receive beam at UE 315). In some examples, control information may indicate a characteristic ID corresponding to UE mobility characteristics, such as the direction UE 315 may be moving (e.g., direction of movement), the speed UE 315 may be moving (e.g., speed of movement), the direction UE 315 may be rotating (e.g., direction of rotation), the speed UE 315 may be rotating (e.g., speed of rotation), or the orientation of UE 315. In such examples, UE 315 may determine whether to use the second AI / ML model to enable (or disable) beam prediction at UE 315. For example, UE 315 may use a second AI / ML model to enable (or disable) beam prediction at UE 315 to determine whether the second AI / ML model produces a prediction or output with appropriate fidelity for (e.g., relatively reliably) beam prediction based on UE mobility characteristics (e.g., the direction or speed at which UE 315 may be moving or rotating). In some examples, based on its activation, UE 315 may determine to activate, deactivate, or modify the second AI / ML model. For example, UE 315 may determine that the prediction or output of the second AI / ML model fails to meet a threshold fidelity (e.g., which may be based on UE mobility characteristics). In such an example, UE 315 may determine to modify the second AI / ML model such that its prediction or output meets the threshold fidelity. Alternatively, UE 315 may determine to deactivate the second AI / ML model and activate another AI / ML model in which the prediction or output meets the threshold fidelity.

[0126] At 340, UE 315 may send feedback information to network entity 305 based on determining whether to activate or deactivate the first AI / ML model (e.g., at 330) or based on determining whether to enable or disable the functionality of the second AI / ML model (e.g., at 335). The feedback information may be transmitted through and referenced from... Figure 1 and Figure 2 Examples of illustrative and described feedback information. For example, feedback information may indicate one or more parameters of a first AI / ML model or one or more parameters of the functionality of a second AI / ML model. In some examples, UE 315 may identify one or more parameters of the first AI / ML model or one or more parameters of the functionality of the second AI / ML model in response to this determination. The one or more parameters may be based on a correspondence between one or more parameters and features. In some examples, feedback information may indicate (e.g., confirm) the activation or deactivation of the first AI / ML model at UE 315 or the activation or deactivation of the functionality of the second AI / ML model at UE 315. In some examples, by indicating features to UE 315 (e.g., via control information), network entity 305 may improve the LCM of the AI / ML model at UE 315 and other benefits.

[0127] Figure 4 A block diagram 400 of a device 405 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. Device 405 may be an example of aspects of UE 115 as described herein. Device 405 may include a receiver 410, a transmitter 415, and a communications manager 420. Device 405 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0128] Receiver 410 may provide components for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to indications of training datasets for LCM). The information may be passed to other components of device 405. Receiver 410 may utilize a single antenna or a collection of multiple antennas.

[0129] Transmitter 415 may provide components for transmitting signals generated by other components of device 405. For example, transmitter 415 may transmit information associated with various information channels (e.g., control channels, data channels, information channels related to indications of training datasets for LCM), such as packets, user data, control information, or any combination thereof. In some examples, transmitter 415 may be co-located with receiver 410 in a transceiver module. Transmitter 415 may utilize a single antenna or a collection of multiple antennas.

[0130] The communication manager 420, receiver 410, transmitter 415, or various combinations thereof, or various components thereof, may be examples of components for performing various aspects of instructing a training dataset for LCM as described herein. For example, the communication manager 420, receiver 410, transmitter 415, or various combinations thereof, or components thereof, may support methods for performing one or more of the functions described herein.

[0131] In some examples, the communication manager 420, receiver 410, transmitter 415, or various combinations or components thereof may be implemented in hardware (e.g., in communication management circuitry). The hardware may include processors, digital signal processors (DSPs), central processing units (CPUs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, microcontrollers, discrete gate or transistor logic components, discrete hardware components, or any combination thereof, configured as or otherwise to support components for performing the functions described herein. In some examples, the processor and memory coupled to the processor may be configured to perform one or more of the functions described herein (e.g., by executing instructions stored in memory by the processor).

[0132] Additionally or alternatively, in some examples, the communication manager 420, receiver 410, transmitter 415, or various combinations or components thereof may be implemented in code executed by a processor (e.g., implemented as communication management software or firmware). If implemented in code executed by a processor, the functionality of the communication manager 420, receiver 410, transmitter 415, or various combinations or components thereof may be performed by (e.g., a general-purpose processor, DSP, CPU, ASIC, FPGA, microcontroller, or any combination of these or other programmable logic devices configured as or otherwise supporting components for performing the functions described in this disclosure).

[0133] In some examples, the communication manager 420 may be configured to use or otherwise cooperate with the receiver 410, transmitter 415, or both to perform various operations (e.g., receiving, acquiring, monitoring, outputting, transmitting). For example, the communication manager 420 may receive information from the receiver 410, transmit information to the transmitter 415, or be integrated with the receiver 410, transmitter 415, or both to acquire information, output information, or perform various other operations as described herein.

[0134] Communication manager 420 may support wireless communication at the UE (e.g., device 405) according to examples disclosed herein. For example, communication manager 420 may be capable of, configured to, or operable to support components for receiving control information from a network entity indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. Communication manager 420 may be capable of, configured to, or operable to support components for determining, based on received control information, whether to activate or deactivate a first ML model for maintaining a wireless communication link or to use a second ML model for maintaining a wireless communication link. Communication manager 420 may be capable of, configured to, or operable to support components for sending feedback information to a network entity, based on the determination, indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model. By including or configuring communication manager 420 according to the examples described herein, device 405 (e.g., a processor that controls or is otherwise coupled to receiver 410, transmitter 415, communication manager 420, or combinations thereof) may support techniques for reducing processing.

[0135] Figure 5 A block diagram 500 of a device 505 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. Device 505 may be an example of aspects of device 405 or UE 115 as described herein. Device 505 may include receiver 510, transmitter 515, and communication manager 520. Device 505 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0136] Receiver 510 may provide components for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to indications of training datasets for LCM). The information may be passed to other components of device 505. Receiver 510 may utilize a single antenna or a collection of multiple antennas.

[0137] Transmitter 515 may provide components for transmitting signals generated by other components of device 505. For example, transmitter 515 may transmit information associated with various information channels (e.g., control channels, data channels, information channels related to indications of training datasets for LCM), such as packets, user data, control information, or any combination thereof. In some examples, transmitter 515 may be co-located with receiver 510 in a transceiver module. Transmitter 515 may utilize a single antenna or a collection of multiple antennas.

[0138] Device 505 or its various components may be examples of parts for performing various aspects of instructing a training dataset for LCM as described herein. For example, communication manager 520 may include control information component 525, ML model component 530, feedback component 535, or any combination thereof. Communication manager 520 may be examples of aspects of communication manager 420 as described herein. In some examples, communication manager 520 or its various components may be configured to use or otherwise cooperate with receiver 510, transmitter 515, or both to perform various operations (e.g., receiving, acquiring, monitoring, outputting, transmitting). For example, communication manager 520 may receive information from receiver 510, transmit information to transmitter 515, or be integrated in combination with receiver 510, transmitter 515, or both to acquire information, output information, or perform various other operations as described herein.

[0139] Communication manager 520 can support wireless communication at the UE (e.g., device 505) according to examples disclosed herein. Control information component 525 is capable of, configured to, or operable to support components for receiving control information from a network entity indicating characteristics of a dataset used to train an ML model associated with maintaining a wireless communication link. ML model component 530 is capable of, configured to, or operable to support components for determining, based on received control information, whether to activate or deactivate the functionality of a first ML model for maintaining the wireless communication link or a second ML model for maintaining the wireless communication link. Feedback component 535 is capable of, configured to, or operable to support components for sending feedback information to a network entity, based on the determination, indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model.

[0140] Figure 6 A block diagram 600 of a communication manager 620 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. The communication manager 620 may be an example of aspects of the communication manager 420, communication manager 520, or both as described herein. The communication manager 620 or its various components may be examples of components for performing various aspects of instructions for a training dataset for LCM as described herein. For example, the communication manager 620 may include a control information component 625, an ML model component 630, a feedback component 635, a recommendation component 640, a parameter component 645, a beam prediction component 650, or any combination thereof. Each of these components may communicate directly or indirectly with each other (e.g., via one or more buses).

[0141] According to the examples disclosed herein, the communication manager 620 can support wireless communication at the UE. The control information component 625 is capable of, configured to, or operable to support components for receiving control information from a network entity indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining the wireless communication link. The ML model component 630 is capable of, configured to, or operable to support components for determining, based on received control information, whether to activate or deactivate the functionality of a first ML model for maintaining the wireless communication link or a second ML model for maintaining the wireless communication link. The feedback component 635 is capable of, configured to, or operable to support components for sending feedback information to a network entity, based on the determination, indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model.

[0142] In some examples, in order to support receiving control information, the control information component 625 can be configured or operated to support a component for receiving a dataset ID corresponding to a dataset or a feature ID corresponding to a feature, or both.

[0143] In some examples, the recommendation component 640 is capable of, configured to, or operable to support components for sending uplink messages to network entities, the uplink messages including one or more dataset IDs corresponding to one or more recommended datasets, one or more feature IDs corresponding to one or more recommended features, or both, wherein control information is based on the uplink messages. In some examples, the one or more dataset IDs include at least a dataset ID. In some examples, the one or more feature IDs include at least a feature ID.

[0144] In some examples, this characteristic includes the operating scenario of the wireless communication link, the profile characteristics of the wireless communication link, the parameters of the cell serving the wireless communication link, the transmission parameters for downlink communication via the wireless communication link, the transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE. In some examples, the first ML model and the second ML model each include a corresponding beam prediction ML model based on a dataset used to train the beam prediction ML model.

[0145] In some examples, this characteristic includes statistics associated with received power measurements used as input to the ML model, statistics associated with received power measurements used as prediction targets of the ML model, performance metrics associated with received power measurements used as input to the ML model, performance metrics associated with received power measurements used as prediction targets of the ML model, UE mobility characteristics, characteristics of the transmit beam at the network entity, or characteristics of the receive beam at the UE.

[0146] In some examples, to support the reception of control information, the control information component 625 is capable of, configured to, or operable to support components for receiving RRC layer signaling, PHY layer signaling, MAC layer signaling, or application layer signaling that includes control information. In some examples, to support the transmission of feedback information, the feedback component 635 is capable of, configured to, or operable to support components for transmitting indications of the correspondence between one or more parameters and characteristics.

[0147] In some examples, parameter component 645 is capable of, configured to, or operable to support components for identifying one or more parameters in response to determining and based on a correspondence between one or more parameters and characteristics, wherein feedback information indicates activation or deactivation of a first ML model at the UE or activation or deactivation of functionality of a second ML model at the UE. In some examples, to support the transmission of feedback information, feedback component 635 is capable of, configured to, or operable to support components for transmitting indications of one or more parameters.

[0148] In some examples, the ML model component 630 is capable of, configured to, or operable to support components for activating or deactivating a first ML model in association with characteristics of the dataset. In some examples, the beam prediction component 650 is capable of, configured to, or operable to support components for using the first ML model, based on the activation of the first ML model, to predict the transmit beam at a network entity or the receive beam at a UE for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction ML model.

[0149] In some examples, the ML model component 630 is capable of, configured to, or operable to support components for enabling or disabling the functionality of a second ML model in association with characteristics of the dataset. In some examples, the beam prediction component 650 is capable of, configured to, or operable to support components for using the second ML model to predict the transmit beam at a network entity or the receive beam at a UE based on enabling the functionality of the second ML model for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction ML model.

[0150] Figure 7A diagram of a system 700 including device 705 supporting instructions for a training dataset for LCM, according to one or more aspects of this disclosure, is shown. Device 705 may be an example of device 405, device 505, or UE 115 as described herein, or a component including such devices. Device 705 may communicate with one or more network entities 105, one or more UEs 115, or any combination thereof (e.g., wirelessly). Device 705 may include components for bidirectional voice and data communication, including components for transmitting and receiving communications, such as a communication manager 720, an input / output (I / O) controller 710, a transceiver 715, an antenna 725, a memory 730, a code 735, and a processor 740. These components may communicate electronically via one or more buses (e.g., bus 745) or be coupled in other ways (e.g., operational ground, communication ground, functional ground, electronic ground, electrical ground).

[0151] The I / O controller 710 manages the input and output signals of the device 705. The I / O controller 710 can also manage peripheral devices not integrated into the device 705. In some cases, the I / O controller 710 may represent a physical connection or port to an external peripheral device. In some cases, the I / O controller 710 may utilize an operating system such as... Alternatively, it may be another known operating system. Additionally or alternatively, the I / O controller 710 may represent or interact with a modem, keyboard, mouse, touchscreen, or similar device. In some cases, the I / O controller 710 may be implemented as part of a processor such as processor 740. In some cases, a user may interact with device 705 via the I / O controller 710 or via hardware components controlled by the I / O controller 710.

[0152] In some cases, device 705 may include a single antenna 725. However, in other cases, device 705 may have more than one antenna 725, which may be capable of concurrently transmitting or receiving multiple wireless transmissions. Transceiver 715 may communicate bidirectionally via one or more antennas 725, a wired link, or a wireless link as described herein. For example, transceiver 715 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. Transceiver 715 may also include a modem for: modulating packets; providing the modulated packets to one or more antennas 725 for transmission; and demodulating packets received from one or more antennas 725. Transceiver 715, or transceiver 715 and one or more antennas 725, may be an example of transmitter 415, transmitter 515, receiver 410, receiver 510, or any combination thereof or components thereof as described herein.

[0153] Memory 730 may include random access memory (RAM) and read-only memory (ROM). Memory 730 may store computer-readable, computer-executable code 735, including instructions that, when executed by processor 740, cause device 705 to perform the various functions described herein. Code 735 may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some cases, code 735 may not be directly executable by processor 740, but may (e.g., when compiled and executed) cause the computer to perform the functions described herein. In some cases, in addition, memory 730 may also include a basic I / O system (BIOS) that controls basic hardware or software operations, such as interaction with peripheral components or devices.

[0154] Processor 740 may include intelligent hardware devices (e.g., general-purpose processors, DSPs, CPUs, microcontrollers, ASICs, FPGAs, programmable logic devices, discrete gate or transistor logic components, discrete hardware components, or any combination thereof). In some cases, processor 740 may be configured to use a memory controller to operate a memory array. In other cases, the memory controller may be integrated into processor 740. Processor 740 may be configured to execute computer-readable instructions stored in memory (e.g., memory 730) to cause device 705 to perform various functions (e.g., functions or tasks supporting instructions for a training dataset used for LCM). For example, device 705 or components of device 705 may include processor 740 and memory 730 coupled to or coupled to processor 740, processor 740 and memory 730 being configured to perform the various functions described herein.

[0155] Communication manager 720 may support wireless communication at a UE (e.g., device 705) according to examples disclosed herein. For example, communication manager 720 may be capable of, configured to, or operable to support components for receiving control information from a network entity indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. Communication manager 720 may be capable of, configured to, or operable to support components for determining, based on received control information, whether to activate or deactivate a first ML model for maintaining a wireless communication link, or to enable or disable the functionality of a second ML model for maintaining a wireless communication link. Communication manager 720 may be capable of, configured to, or operable to support components for sending feedback information to a network entity, based on the determination, indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model.

[0156] By including or configuring a communication manager 720 according to an example as described herein, device 705 can support techniques for improving communication reliability, reducing latency, improving and reducing the user experience associated with processing, and improving the utilization of processing power.

[0157] In some examples, the communication manager 720 may be configured to use or otherwise coordinate with the transceiver 715, one or more antennas 725, or any combination thereof to perform various operations (e.g., receiving, monitoring, transmitting). Although the communication manager 720 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 720 may be supported or performed by the processor 740, memory 730, code 735, or any combination thereof. For example, code 735 may include instructions that can be executed by the processor 740 to cause the device 705 to perform various aspects of the indications for the training dataset for LCM as described herein, or the processor 740 and memory 730 may be otherwise configured to perform or support such operations.

[0158] Figure 8 A block diagram 800 of a device 805 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. Device 805 may be an example of aspects of network entity 105 as described herein. Device 805 may include receiver 810, transmitter 815, and communication manager 820. Device 805 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0159] Receiver 810 may provide components for acquiring (e.g., receiving, determining, identifying) information (such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units)) associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack). The information may be passed to other components of device 805. In some examples, receiver 810 may support acquiring information by receiving signals via one or more antennas. Additionally or alternatively, receiver 810 may support acquiring information by receiving signals via one or more wired (e.g., electrical, fiber optic) interfaces, wireless interfaces, or any combination thereof.

[0160] Transmitter 815 may provide components for outputting (e.g., transmitting, providing, conveying, transmitting) information generated by other components of device 805. For example, transmitter 815 may output information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). In some examples, transmitter 815 may support outputting information by transmitting signals via one or more antennas. Additionally or alternatively, transmitter 815 may support outputting information by transmitting signals via one or more wired (e.g., electrical, fiber optic) interfaces, wireless interfaces, or any combination thereof. In some examples, transmitter 815 and receiver 810 may be co-located in a transceiver, which may include or be coupled to a modem.

[0161] The communication manager 820, receiver 810, transmitter 815, or various combinations thereof, or various components thereof, may be examples of components for performing various aspects of instructing a training dataset for LCM as described herein. For example, the communication manager 820, receiver 810, transmitter 815, or various combinations thereof, or components thereof, may support methods for performing one or more of the functions described herein.

[0162] In some examples, the communication manager 820, receiver 810, transmitter 815, or various combinations or components thereof may be implemented in hardware (e.g., in communication management circuitry). The hardware may include processors, DSPs, CPUs, ASICs, FPGAs, or other programmable logic devices, microcontrollers, discrete gate or transistor logic components, discrete hardware components, or any combination thereof, configured as or otherwise to support components for performing the functions described herein. In some examples, the processor and memory coupled to the processor may be configured to perform one or more of the functions described herein (e.g., by executing instructions stored in memory by the processor).

[0163] Additionally or alternatively, in some examples, the communication manager 820, receiver 810, transmitter 815, or various combinations or components thereof may be implemented in code executed by a processor (e.g., implemented as communication management software or firmware). If implemented in code executed by a processor, the functionality of the communication manager 820, receiver 810, transmitter 815, or various combinations or components thereof may be performed by (e.g., a general-purpose processor, DSP, CPU, ASIC, FPGA, microcontroller, or any combination of these or other programmable logic devices configured as or otherwise supporting components for performing the functions described in this disclosure).

[0164] In some examples, the communication manager 820 may be configured to use or otherwise cooperate with the receiver 810, transmitter 815, or both to perform various operations (e.g., receiving, acquiring, monitoring, outputting, transmitting). For example, the communication manager 820 may receive information from the receiver 810, transmit information to the transmitter 815, or be integrated with the receiver 810, transmitter 815, or both to acquire information, output information, or perform various other operations as described herein.

[0165] Communication manager 820 may support wireless communication at a network entity (e.g., device 805) according to examples disclosed herein. For example, communication manager 820 may be capable of, configured to, or operable to support components for outputting control information indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. Communication manager 820 may be capable of, configured to, or operable to support components for obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based on determining whether to activate or deactivate a first ML model for maintaining a wireless communication link or to use a second ML model for maintaining a wireless communication link.

[0166] By including or configuring a communication manager 820 according to the examples described herein, device 805 (e.g., a processor that controls or is otherwise coupled to receiver 810, transmitter 815, communication manager 820, or a combination thereof) can support techniques for reducing processing.

[0167] Figure 9 A block diagram 900 of a device 905 supporting indications for a training dataset for LCM according to one or more aspects of this disclosure is shown. Device 905 may be an example of aspects of device 805 or network entity 105 as described herein. Device 905 may include receiver 910, transmitter 915, and communication manager 920. Device 905 may also include a processor. Each of these components may communicate with each other (e.g., via one or more buses).

[0168] Receiver 910 may provide components for acquiring (e.g., receiving, determining, identifying) information (such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units)) associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack). The information may be passed to other components of device 905. In some examples, receiver 910 may support acquiring information by receiving signals via one or more antennas. Additionally or alternatively, receiver 910 may support acquiring information by receiving signals via one or more wired (e.g., electrical, fiber optic) interfaces, wireless interfaces, or any combination thereof.

[0169] Transmitter 915 may provide components for outputting (e.g., transmitting, providing, conveying, transmitting) information generated by other components of device 905. For example, transmitter 915 may output information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). In some examples, transmitter 915 may support outputting information by transmitting signals via one or more antennas. Additionally or alternatively, transmitter 915 may support outputting information by transmitting signals via one or more wired (e.g., electrical, fiber optic) interfaces, wireless interfaces, or any combination thereof. In some examples, transmitter 915 and receiver 910 may be co-located in a transceiver, which may include or be coupled to a modem.

[0170] Device 905 or its various components may be examples of parts for performing various aspects of indicative instruction to a training dataset for LCM as described herein. For example, communication manager 920 may include feature indication component 925, feedback information component 930, or any combination thereof. Communication manager 920 may be examples of aspects of communication manager 820 as described herein. In some examples, communication manager 920 or its various components may be configured to use or otherwise cooperate with receiver 910, transmitter 915, or both to perform various operations (e.g., receive, acquire, monitor, output, transmit). For example, communication manager 920 may receive information from receiver 910, transmit information to transmitter 915, or be integrated in combination with receiver 910, transmitter 915, or both to acquire information, output information, or perform various other operations as described herein.

[0171] Communication manager 920 can support wireless communication at a network entity (e.g., device 905) according to examples disclosed herein. Feature indication component 925 is capable of, configured to, or operable to support components for outputting control information indicating the characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. Feedback information component 930 is capable of, configured to, or operable to support components for obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based on determining whether to activate or deactivate a first ML model for maintaining a wireless communication link or to use a second ML model for maintaining a wireless communication link.

[0172] Figure 10 A block diagram 1000 of a communication manager 1020 supporting instructions for a training dataset for LCM according to one or more aspects of this disclosure is shown. The communication manager 1020 may be an example of aspects of a communication manager 820, a communication manager 920, or both as described herein. The communication manager 1020 or its various components may be examples of components for performing various aspects of instructions for a training dataset for LCM as described herein. For example, the communication manager 1020 may include a feature indication component 1025, a feedback information component 1030, a control information indication component 1035, a correspondence indication component 1040, a parameter indication component 1045, a feature recommendation component 1050, or any combination thereof. Each of these components may communicate directly or indirectly with each other (e.g., via one or more buses), and such communication may include communication within the protocol layers of the protocol stack, communication associated with logical channels of the protocol stack (e.g., between the protocol layers of the protocol stack, within devices, components or virtualization components associated with network entity 105, between devices, components or virtualization components associated with network entity 105), or any combination thereof.

[0173] Based on the examples disclosed herein, the communication manager 1020 can support wireless communication at network entities. The feature indication component 1025 is capable of, configured to, or operable to support components for outputting control information indicating the characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. The feedback information component 1030 is capable of, configured to, or operable to support components for obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based on determining whether to activate or deactivate a first ML model for maintaining the wireless communication link or to use a second ML model for maintaining the wireless communication link.

[0174] In some examples, in order to support output control information, the feature indication component 1025 can be configured or operated to support components for outputting a dataset ID corresponding to a dataset or a feature ID corresponding to a feature, or both.

[0175] In some examples, the feature recommendation component 1050 is capable of, configured to, or operable to support components for obtaining uplink messages, which include one or more dataset IDs corresponding to one or more recommended datasets, one or more feature IDs corresponding to one or more recommended features, or both, wherein control information is based on the uplink message. In some examples, the one or more dataset IDs include at least a dataset ID. In some examples, the one or more feature IDs include at least a feature ID.

[0176] In some examples, this feature includes the operating scenario of the wireless communication link, the configuration file characteristics of the wireless communication link, the parameters of the cell serving the wireless communication link, the transmission parameters for downlink communication via the wireless communication link, the transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE.

[0177] In some examples, the first ML model and the second ML model each include a corresponding beam prediction ML model based on a dataset used to train the beam prediction ML model. In some examples, the feature includes statistics associated with received power measurements used as inputs to the ML model, statistics associated with received power measurements used as prediction targets of the ML model, performance metrics associated with received power measurements used as inputs to the ML model, performance metrics associated with received power measurements used as prediction targets of the ML model, UE mobility characteristics, characteristics of the transmitted beam at the network entity, or characteristics of the received beam at the UE.

[0178] In some examples, in order to support the output of control information, the control information instruction component 1035 is capable of, configured to, or able to operate to support components for outputting RRC layer signaling, PHY layer signaling, MAC layer signaling, or application layer signaling that includes control information.

[0179] In some examples, in order to support obtaining feedback information, the correspondence indication component 1040 is capable of, configured to, or able to operate to support a component for obtaining an indication of the correspondence between one or more parameters and characteristics.

[0180] In some examples, in order to support obtaining feedback information, the parameter indication component 1045 is capable of, configured to, or able to operate to support components for obtaining indications of one or more parameters.

[0181] Figure 11A diagram of a system 1100 including device 1105 supporting instructions for a training dataset for LCM, according to one or more aspects of this disclosure, is shown. Device 1105 may be an example of device 805, device 905, or network entity 105 as described herein, or a component including such devices. Device 1105 may communicate with one or more network entities 105, one or more UEs 115, or any combination thereof, and such communication may include communication via one or more wired interfaces, one or more wireless interfaces, or any combination thereof. Device 1105 may include components supporting output and acquisition of communication, such as a communication manager 1120, transceiver 1110, antenna 1115, memory 1125, code 1130, and processor 1135. These components may communicate electronically via one or more buses (e.g., bus 1140) or otherwise coupled (e.g., operational ground, communication ground, functional ground, electronic ground, electrical ground).

[0182] Transceiver 1110 may support bidirectional communication via a wired link, a wireless link, or both, as described herein. In some examples, transceiver 1110 may include a wired transceiver and may communicate bidirectionally with another wired transceiver. Additionally or alternatively, in some examples, transceiver 1110 may include a wireless transceiver and may communicate bidirectionally with another wireless transceiver. In some examples, device 1105 may include one or more antennas 1115 that may be capable of (e.g., concurrently) transmitting or receiving wireless transmissions. Transceiver 1110 may also include a modem for: modulating a signal; providing the modulated signal for transmission (e.g., via one or more antennas 1115, via a wired transmitter); receiving the modulated signal (e.g., from one or more antennas 1115, from a wired receiver); and demodulating the signal. In some embodiments, transceiver 1110 may include one or more interfaces, such as one or more interfaces coupled to one or more antennas 1115 configured to support various receive or acquire operations, or one or more interfaces coupled to one or more antennas 1115 configured to support various transmit or output operations, or combinations thereof. In some embodiments, transceiver 1110 may include one or more processor or memory components or be configured to couple to such processor or memory components, which are operable to perform or support operations based on received or acquired information or signals, or generate information or other signals for transmission or other output, or any combination thereof. In some embodiments, transceiver 1110, or transceiver 1110 and one or more antennas 1115, or transceiver 1110 and one or more antennas 1115 and one or more processor or memory components (e.g., processor 1135 or memory 1125 or both) may be included in a chip or chip assembly mounted in device 1105. In some examples, the transceiver may be able to operate to support communication via one or more communication links (e.g., communication link 125, backhaul communication link 120, midhaul communication link 162, and fronthaul communication link 168).

[0183] Memory 1125 may include RAM and ROM. Memory 1125 may store computer-readable, computer-executable code 1130, including instructions that, when executed by processor 1135, cause device 1105 to perform the various functions described herein. Code 1130 may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some cases, code 1130 may not be directly executable by processor 1135, but may (e.g., when compiled and executed) cause the computer to perform the functions described herein. In some cases, in addition, memory 1125 may also contain a BIOS that controls basic hardware or software operations, such as interaction with peripheral components or devices.

[0184] Processor 1135 may include intelligent hardware devices (e.g., general-purpose processors, DSPs, ASICs, CPUs, FPGAs, microcontrollers, programmable logic devices, discrete gate or transistor logic units, discrete hardware components, or any combination thereof). In some cases, processor 1135 may be configured to use a memory controller to operate a memory array. In other cases, the memory controller may be integrated into processor 1135. Processor 1135 may be configured to execute computer-readable instructions stored in memory (e.g., memory 1125) to cause device 1105 to perform various functions (e.g., functions or tasks supporting instructions for a training dataset used for LCM). For example, device 1105 or components thereof may include processor 1135 and memory 1125 coupled to processor 1135, wherein processor 1135 and memory 1125 are configured to perform the various functions described herein. Processor 1135 may be an example of a cloud computing platform (e.g., one or more physical nodes and supporting software such as an operating system, virtual machine, or container instance) that can (e.g., by executing code 1130) host functions to perform the functions of device 1105. Processor 1135 may be any one or more suitable processors capable of executing scripts or instructions of one or more software programs stored in device 1105 (such as within memory 1125). In some specific implementations, processor 1135 may be a component of a processing system. A processing system can generally refer to a system or a series of machines or components that receive input and process that input to produce a set of outputs (which may be passed to other systems or components of device 1105, for example). For example, the processing system of device 1105 may refer to a system that includes various other components or subcomponents of device 1105, such as processor 1135, or transceiver 1110, or communication manager 1120, or other components or combinations of components of device 1105. The processing system of device 1105 can interface with other components of device 1105 and can process information (such as inputs or signals) received from other components or output information to other components. For example, the chip or modem of device 1105 may include a processing system and one or more interfaces for outputting information or for receiving information, or both. The one or more interfaces may be implemented as or otherwise include a first interface configured to output information and a second interface configured to receive information, or the same interface configured to both output and receive information, and other specific implementations. In some specific implementations, the one or more interfaces may refer to the interface between the processing system of the chip or modem and the transmitter, enabling device 1105 to transmit information output from the chip or modem.Additionally or alternatively, in some embodiments, the one or more interfaces may refer to an interface between the processing system of a chip or modem and the receiver, enabling device 1105 to receive information or signal input, and such information to be transmitted to the processing system. Those skilled in the art will readily recognize that the first interface may also receive information or signal input, and the second interface may also output information or signal output.

[0185] In some examples, bus 1140 may support communication at the protocol layer of the protocol stack (e.g., within a protocol layer). In some examples, bus 1140 may support communication associated with logical channels of the protocol stack (e.g., between protocol layers of the protocol stack), which may include communication performed within components of device 1105, or communication performed between different components of device 1105 that may be co-located or located in different locations (e.g., where device 1105 may refer to a system in which one or more of communication manager 1120, transceiver 1110, memory 1125, code 1130, and processor 1135 may be located in one of the different components or partitioned between the different components).

[0186] In some examples, the communication manager 1120 may manage (e.g., via one or more wired or wireless backhaul links) various aspects of communication with the core network 130. For example, the communication manager 1120 may manage the transfer of data communication for client devices (such as one or more UEs 115). In some examples, the communication manager 1120 may manage communication with other network entities 105 and may include a controller or scheduler for coordinating with other network entities 105 to control communication with UE 115. In some examples, the communication manager 1120 may support an X2 interface within LTE / LTE-A wireless communication network technology to provide communication between network entities 105.

[0187] Communication manager 1120 may support wireless communication at a network entity (e.g., device 1105) according to examples disclosed herein. For example, communication manager 1120 may be capable of, configured to, or operable to support components for outputting control information indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. Communication manager 1120 may be capable of, configured to, or operable to support components for obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based on determining whether to activate or deactivate a first ML model for maintaining a wireless communication link or to use a second ML model for maintaining a wireless communication link.

[0188] By including or configuring a communication manager 1120 according to an example as described herein, device 1105 can support techniques for improving communication reliability, reducing latency, improving and reducing the user experience associated with processing, and improving the utilization of processing power.

[0189] In some examples, the communication manager 1120 may be configured to use or otherwise coordinate with the transceiver 1110, one or more antennas 1115 (e.g., where applicable), or any combination thereof to perform various operations (e.g., receive, acquire, monitor, output, transmit). Although the communication manager 1120 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 1120 may be supported or performed by the transceiver 1110, processor 1135, memory 1125, code 1130, or any combination thereof. For example, code 1130 may include instructions that can be executed by the processor 1135 to cause the device 1105 to perform various aspects of the indications for the training dataset for LCM as described herein, or the processor 1135 and memory 1125 may be otherwise configured to perform or support such operations.

[0190] Figure 12 A flowchart illustrating a method 1200 for instructing a training dataset for LCM according to various aspects of this disclosure is shown. The operation of method 1200 can be implemented by a UE or its components as described herein. For example, the operation of method 1200 can be implemented by, as referenced... Figures 1 to 7 The UE 115 described herein performs the following: In some examples, the UE may execute a set of instructions to control the functional elements of the wireless UE to perform the described functions. Additionally or alternatively, the wireless UE may use dedicated hardware to perform aspects of the described functions.

[0191] At 1205, the method may include: receiving from a network entity control information indicating characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. The operation of 1205 may be performed according to examples as disclosed herein. In some examples, aspects of the operation of 1205 may be provided by reference to... Figure 6 The control information component 625 described herein shall perform the operation.

[0192] At 1210, the method may include: determining, based on received control information, whether to activate or deactivate a first ML model for maintaining a wireless communication link, or to enable or disable the functionality of a second ML model for maintaining a wireless communication link. The operation of 1210 may be performed according to examples as disclosed herein. In some examples, aspects of the operation of 1210 may be derived from references... Figure 6 The ML model component 630 described is used for execution.

[0193] At 1215, the method may include: sending feedback information to a network entity based on determination, indicating one or more parameters of a first ML model or one or more parameters of the functionality of a second ML model. The operation of 1215 may be performed according to examples as disclosed herein. In some examples, aspects of the operation of 1215 may be derived from references... Figure 6 The described feedback component 635 is executed.

[0194] Figure 13 A flowchart illustrating method 1300, which provides instructions for a training dataset for LCM according to various aspects of this disclosure, is shown. The operation of method 1300 may be implemented by a network entity or its components as described herein. For example, the operation of method 1300 may be implemented by, as referenced... Figure 1 and Figure 2 as well as Figures 8 to 11 The described network entity performs the functions. In some examples, the network entity may execute a set of instructions to control the functional elements of the wireless network entity to perform the described functions. Additionally or alternatively, the wireless network entity may use dedicated hardware to perform aspects of the described functions.

[0195] At 1305, the method may include: outputting control information indicating the characteristics of a dataset used to train an ML model at the UE, the ML model being associated with maintaining a wireless communication link. The operation of 1305 may be performed according to examples as disclosed herein. In some examples, aspects of the operation of 1305 may be derived from references... Figure 10 The described characteristics are instructing component 1025 to perform.

[0196] At 1310, the method may include: in response to control information, obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based on determining whether to activate or deactivate a first ML model for maintaining a wireless communication link or to use a second ML model for maintaining a wireless communication link. The operation of 1310 may be performed according to examples as disclosed herein. In some examples, aspects of the operation of 1310 may be provided by reference to... Figure 10 The described feedback information component 1030 is used for execution.

[0197] The following provides an overview of the various aspects of this disclosure:

[0198] Aspect 1: A method for wireless communication at a UE, the method comprising: receiving from a network entity control information indicating characteristics of a dataset for training an ML model at the UE, the ML model being associated with maintaining a wireless communication link; determining, at least in part based on the received control information, whether to activate or deactivate a first ML model for maintaining the wireless communication link or to enable or disable the functionality of a second ML model for maintaining the wireless communication link; and at least in part based on the determination, sending to the network entity feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model.

[0199] Aspect 2: According to the method of aspect 1, receiving the control information includes: receiving a dataset ID corresponding to the dataset or a feature ID corresponding to the feature, or both.

[0200] Aspect 3: According to the method of aspect 2, the method further includes: sending an uplink message to the network entity, the uplink message including one or more dataset IDs corresponding to one or more recommended datasets, one or more feature IDs corresponding to one or more recommended features, or both, wherein the control information is at least partially based on the uplink message.

[0201] Aspect 4: According to the method of aspect 3, wherein the one or more dataset IDs include at least the dataset ID, and the one or more feature IDs include at least the feature ID.

[0202] Aspect 5: The method according to any one of Aspects 1 to 4, wherein the features include the operating scenario of the wireless communication link, the configuration file features of the wireless communication link, the parameters of the cell serving the wireless communication link, the transmission parameters for downlink communication via the wireless communication link, the transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE.

[0203] Aspect 6: The method according to any one of Aspects 1 to 4, wherein the first ML model and the second ML model each comprise a corresponding beam prediction ML model based at least in part on the dataset used to train the beam prediction ML mode.

[0204] Aspect 7: The method according to aspect 6, wherein the characteristics include statistics associated with received power measurements used as input to the ML model, statistics associated with received power measurements used as prediction targets of the ML model, performance metrics associated with the received power measurements used as input to the ML model, performance metrics associated with the received power measurements used as prediction targets of the ML model, UE mobility characteristics, characteristics of the transmit beam at the network entity, or characteristics of the receive beam at the UE.

[0205] Aspect 8: The method according to any one of Aspects 1 to 7, wherein receiving the control information includes: receiving RRC layer signaling, PHY layer signaling, MAC layer signaling or application layer signaling including the control information.

[0206] Aspect 9: The method according to any one of Aspects 1 to 8, wherein sending the feedback information includes: sending an indication of the correspondence between the one or more parameters and the characteristic.

[0207] Aspect 10: The method according to aspect 9, the method further comprising: identifying the one or more parameters in response to the determination and at least in part based on the correspondence between the one or more parameters and the feature, wherein the feedback information indicates activation or deactivation of the first ML model at the UE or activation or deactivation of the functionality of the second ML model at the UE.

[0208] Aspect 11: The method according to any one of aspects 1 to 8, wherein sending the feedback information includes: sending an indication of the one or more parameters.

[0209] Aspect 12: The method according to any one of aspects 1 to 11, the method further comprising: activating or deactivating the first ML model in association with the characteristics of the dataset.

[0210] Aspect 13: The method according to aspect 12, the method further comprising: using the first ML model, at least in part, to predict a transmit beam at the network entity or a receive beam at the UE, based on activating the first ML model, for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction ML model.

[0211] Aspect 14: The method according to any one of Aspects 1 to 11, the method further comprising: enabling or disabling the functionality of the second ML model in association with the characteristics of the dataset.

[0212] Aspect 15: The method according to aspect 14, the method further comprising: using the second ML model to predict a transmit beam at the network entity or a receive beam at the UE, at least in part based on enabling the functionality of the second ML model, for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction ML model.

[0213] Aspect 16: A method for wireless communication at a network entity, the method comprising: outputting control information indicating characteristics of a dataset for training an ML model at a UE, the ML model being associated with maintaining a wireless communication link; and, in response to the control information, obtaining feedback information indicating one or more parameters of the first ML model or one or more parameters of the functionality of the second ML model, based at least in part on determining whether to activate or deactivate a first ML model for maintaining the wireless communication link or to use a second ML model for maintaining the wireless communication link.

[0214] Aspect 17: According to the method of aspect 16, outputting the control information includes: outputting a dataset ID corresponding to the dataset or a feature ID corresponding to the feature, or both.

[0215] Aspect 18: The method according to aspect 17, the method further comprising: obtaining an uplink message, the uplink message including one or more dataset IDs corresponding to one or more recommended datasets, one or more feature IDs corresponding to one or more recommended features, or both, wherein the control information is at least partially based on the uplink message.

[0216] Aspect 19: According to the method of aspect 18, wherein the one or more dataset IDs include at least the dataset ID, and the one or more feature IDs include at least the feature ID.

[0217] Aspect 20: The method according to any one of Aspects 16 to 19, wherein the features include the operating scenario of the wireless communication link, the configuration file features of the wireless communication link, the parameters of the cell serving the wireless communication link, the transmission parameters for downlink communication via the wireless communication link, the transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE.

[0218] Aspect 21: The method according to any one of Aspects 16 to 19, wherein the first ML model and the second ML model each comprise a corresponding beam prediction ML model based at least in part on the dataset used to train beam prediction ML patterns.

[0219] Aspect 22: According to the method of aspect 21, wherein the characteristics include statistics associated with received power measurements used as input to the ML model, statistics associated with received power measurements used as prediction targets of the ML model, performance metrics associated with the received power measurements used as input to the ML model, performance metrics associated with the received power measurements used as prediction targets of the ML model, UE mobility characteristics, characteristics of the transmit beam at the network entity, or characteristics of the receive beam at the UE.

[0220] Aspect 23: The method according to any one of Aspects 16 to 22, wherein outputting the control information includes: outputting RRC layer signaling, PHY layer signaling, MAC layer signaling or application layer signaling including the control information.

[0221] Aspect 24: The method according to any one of aspects 16 to 23, wherein obtaining the feedback information includes obtaining an indication of the correspondence between the one or more parameters and the characteristic.

[0222] Aspect 25: The method according to any one of aspects 16 to 23, wherein obtaining the feedback information includes obtaining an indication of the one or more parameters.

[0223] Aspect 26: An apparatus for wireless communication at a UE, the apparatus comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform the method according to any one of aspects 1 to 15.

[0224] Aspect 27: An apparatus for wireless communication at a UE, the apparatus comprising at least one component for performing the method according to any one of aspects 1 to 15.

[0225] Aspect 28: A non-transitory computer-readable medium storing code for wireless communication at a UE, said code including instructions executable by a processor to perform the method according to any one of aspects 1 to 15.

[0226] Aspect 29: An apparatus for wireless communication at a network entity, the apparatus comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform the method according to any one of aspects 16 to 25.

[0227] Aspect 30: An apparatus for wireless communication at a network entity, the apparatus comprising at least one component for performing the method according to any one of aspects 16 to 25.

[0228] Aspect 31: A non-transitory computer-readable medium storing code for wireless communication at a network entity, the code including instructions executable by a processor to perform a method according to any one of aspects 16 to 25.

[0229] It should be noted that the methods described herein describe possible specific implementations, and the operations and steps can be rearranged or otherwise modified, and other specific implementations are also possible. Furthermore, aspects from two or more methods can be combined.

[0230] While aspects of LTE, LTE-A, LTE-A Pro, or NR systems may be described for illustrative purposes, and the terms LTE, LTE-A, LTE-A Pro, or NR may be used in most of the description, the techniques described herein are also applicable to networks outside of LTE, LTE-A, LTE-A Pro, or NR networks. For example, the techniques described are applicable to a variety of other wireless communication systems, such as Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash-OFDM, and other systems and radio technologies not explicitly mentioned herein.

[0231] The information and signals described herein can be represented using any of a variety of different techniques and skills. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the description can be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, light fields or optical particles, or any combination thereof.

[0232] The various exemplary blocks and components described herein can be implemented or performed using a general-purpose processor, DSP, ASIC, CPU, FPGA or other programmable logic device, discrete gate or transistor logic unit, discrete hardware component, or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but in alternative embodiments, the processor may be any 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, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration).

[0233] The functions described herein can be implemented using hardware, software executed by a processor, firmware, or any combination thereof. If implemented using software executed by a processor, these functions can be stored as one or more instructions or code on a computer-readable medium, or transmitted using one or more instructions or code on a computer-readable medium. Other examples and specific implementations are within the scope of this disclosure and the appended claims. For example, due to the nature of software, the functions described herein can be implemented using software executed by a processor, hardware, firmware, hardwiring, or any combination of these. Features implementing the functions can also be physically located in different locations, including various portions distributed such that the functions are implemented in different physical locations.

[0234] Computer-readable media include both non-transitory computer storage media and communication media, with the latter including any medium that facilitates the transfer of a computer program from one location to another. Non-transitory storage media can be any available medium accessible by a general-purpose or special-purpose computer. By way of example, and not limitation, non-transitory computer-readable media may include RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory, compressed optical disc (CD) ROM or other optical disc storage devices, magnetic disk storage devices or other magnetic storage devices, or any other non-transitory medium that can be used to carry or store desired program code components in the form of instructions or data structures, and accessible by a general-purpose or special-purpose computer or a general-purpose or special-purpose processor. Furthermore, any connection is appropriately referred to as computer-readable media. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included within the definition of computer-readable media. As used herein, disks and optical discs include CDs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs. Disks can magnetically reproduce data, while optical discs can optically reproduce data using lasers. Combinations of the above are also included within the scope of computer-readable media.

[0235] As used herein (including in the claims), the word "or" used in an enumeration of items (e.g., an enumeration of items accompanied by phrases such as "at least one of" or "one or more of") indicates an inclusive enumeration, such that an enumeration of at least one of, for example, A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C). Additionally, as used herein, the phrase "based on" should not be construed as a reference to a closed set of conditions. For example, an example step described as "based on condition A" could be based on both condition A and condition B without departing from the scope of this disclosure. In other words, as used herein, the phrase "based on" should be interpreted in the same manner as the phrase "at least partially based on".

[0236] The term "determine" encompasses a variety of actions, and therefore, "determine" can include calculation, computation, processing, derivation, investigation, lookup (such as by searching in a table, database, or other data structure), identification, and similar actions. Furthermore, "determine" can include receiving (e.g., receiving information), accessing (e.g., accessing data stored in memory), etc. Moreover, "determine" can include parsing, acquiring, selecting, choosing, building, and other similar actions.

[0237] In the accompanying drawings, similar components or features may have the same reference numerals. Furthermore, various components of the same type can be distinguished by adding a dash after the reference numeral and a second numeral for differentiation between similar components. If only the first reference numeral is used in the specification, the description can be applied to any of the similar components having the same first reference numeral, regardless of the second or other subsequent reference numerals.

[0238] The description herein, illustrated with reference to the accompanying drawings, describes an example configuration and does not represent all achievable examples or those within the scope of the claims. The term "example" as used herein means "serving as an example, instance, or illustration," not "preferred" or "advantageous over other examples." The detailed description includes specific details used to provide an understanding of the described techniques. However, these techniques can be practiced without these specific details. In some cases, known structures and devices are shown in block diagram form to avoid obscuring the concept of the described examples.

[0239] The description provided herein is intended to enable those skilled in the art to implement or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the scope of this disclosure. Therefore, this disclosure is not limited to the examples and designs described herein, but should be granted the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An apparatus for wireless communication at a user equipment (UE), the apparatus comprising: processor; A memory coupled to the processor; and Instructions, which are stored in the memory and can be executed by the processor, to cause the device to: Receive control information from a network entity indicating the characteristics of a dataset used to train a machine learning model at the UE, the machine learning model being associated with maintaining a wireless communication link; The determination of whether to activate or deactivate the first machine learning model for maintaining the wireless communication link or the functionality of the second machine learning model for maintaining the wireless communication link is based at least in part on the received control information. as well as Feedback information indicating one or more parameters of the first machine learning model or one or more parameters of the functionality of the second machine learning model is sent to the network entity, at least in part based on the determination.

2. The apparatus of claim 1, wherein the instruction for receiving the control information is executable by the processor to cause the apparatus to: Receive a dataset identifier corresponding to the dataset or a feature identifier corresponding to the feature, or both.

3. The apparatus of claim 2, wherein the instructions are further executable by the processor to cause the apparatus to: An uplink message is sent to the network entity, the uplink message including one or more dataset identifiers corresponding to one or more recommended datasets, one or more feature identifiers corresponding to one or more recommended features, or both, wherein the control information is based at least in part on the uplink message.

4. The apparatus of claim 3, wherein the one or more dataset identifiers include at least the dataset identifier, and the one or more feature identifiers include at least the feature identifier.

5. The apparatus of claim 1, wherein the features include the operating scenario of the wireless communication link, the configuration file features of the wireless communication link, the parameters of the cell serving the wireless communication link, the transmission parameters for downlink communication via the wireless communication link, the transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE.

6. The apparatus of claim 1, wherein the first machine learning model and the second machine learning model each comprise a corresponding beam prediction machine learning model that is at least partially based on the dataset used to train the beam prediction machine learning model.

7. The apparatus of claim 6, wherein the characteristics include statistics associated with a received power measurement used as input to the machine learning model, statistics associated with a received power measurement used as a prediction target of the machine learning model, performance metrics associated with the received power measurement used as input to the machine learning model, performance metrics associated with the received power measurement used as a prediction target of the machine learning model, UE mobility characteristics, characteristics of the transmit beam at the network entity, or characteristics of the receive beam at the UE.

8. The apparatus of claim 1, wherein the instruction for receiving the control information is executable by the processor to cause the apparatus to: Receive radio resource control layer signaling, physical layer signaling, media access control layer signaling, or application layer signaling that includes the control information.

9. The apparatus of claim 1, wherein the instruction for sending the feedback information is executable by the processor to cause the apparatus to: Send an indication of the correspondence between the one or more parameters and the characteristic.

10. The apparatus of claim 9, wherein the instructions are further executable by the processor to cause the apparatus to: In response to the determination and at least in part based on the correspondence between the one or more parameters and the feature, the one or more parameters are identified, wherein the feedback information indicates the activation or deactivation of the first machine learning model at the UE or the activation or deactivation of the functionality of the second machine learning model at the UE.

11. The apparatus of claim 1, wherein the instruction for sending the feedback information is executable by the processor to cause the apparatus to: Send instructions for the one or more parameters.

12. The apparatus of claim 1, wherein the instructions are further executable by the processor to cause the apparatus to: The first machine learning model is activated or deactivated in association with the characteristics of the dataset.

13. The apparatus of claim 12, wherein the instructions are further executable by the processor to cause the apparatus to: At least in part, based on activating the first machine learning model, the first machine learning model is used to predict the transmit beam at the network entity or the receive beam at the UE for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction machine learning model.

14. The apparatus of claim 1, wherein the instructions are further executable by the processor to cause the apparatus to: The functionality of the second machine learning model is enabled or disabled in connection with the characteristics of the dataset.

15. The apparatus of claim 14, wherein the instructions are further executable by the processor to cause the apparatus to: At least in part, based on enabling the functionality of the second machine learning model, the second machine learning model is used to predict the transmit beam at the network entity or the receive beam at the UE for maintaining the wireless communication link, wherein the dataset is used to train the beam prediction machine learning model.

16. An apparatus for wireless communication at a network entity, the apparatus comprising: processor; A memory coupled to the processor; and Instructions, which are stored in the memory and can be executed by the processor, to cause the device to: The output indicates control information for the characteristics of a dataset used to train a machine learning model at a user equipment (UE), the machine learning model being associated with maintaining a wireless communication link; as well as In response to the control information, feedback information indicating one or more parameters of the first machine learning model or one or more parameters of the functionality of the second machine learning model is obtained, at least in part, based on determining whether to activate or deactivate the first machine learning model for maintaining the wireless communication link or to enable or disable the functionality of the second machine learning model for maintaining the wireless communication link.

17. The apparatus of claim 16, wherein the instructions for outputting the control information are executable by the processor to cause the apparatus to: Output the dataset identifier corresponding to the dataset or the feature identifier corresponding to the feature, or both.

18. The apparatus of claim 17, wherein the instructions are further executable by the processor to cause the apparatus to: An uplink message is obtained, the uplink message including one or more dataset identifiers corresponding to one or more recommended datasets, one or more feature identifiers corresponding to one or more recommended features, or both, wherein the control information is at least partially based on the uplink message.

19. The apparatus of claim 18, wherein the one or more dataset identifiers include at least the dataset identifier, and wherein the one or more feature identifiers include at least the feature identifier.

20. The apparatus of claim 16, wherein the features include the operating scenario of the wireless communication link, the configuration file features of the wireless communication link, parameters of the cell serving the wireless communication link, transmission parameters for downlink communication via the wireless communication link, transmission parameters for uplink communication via the wireless communication link, or the distance between the network entity and the UE.

21. The apparatus of claim 16, wherein the first machine learning model and the second machine learning model each comprise a corresponding beam prediction machine learning model that is at least partially based on the dataset used to train the beam prediction machine learning model.

22. The apparatus of claim 21, wherein the characteristics include statistics associated with a received power measurement used as input to the machine learning model, statistics associated with a received power measurement used as a prediction target of the machine learning model, performance metrics associated with the received power measurement used as input to the machine learning model, performance metrics associated with the received power measurement used as a prediction target of the machine learning model, UE mobility characteristics, characteristics of the transmit beam at the network entity, or characteristics of the receive beam at the UE.

23. The apparatus of claim 16, wherein the instructions for outputting the control information are executable by the processor to cause the apparatus to: The output includes radio resource control layer signals, physical layer signaling, media access control layer signaling, or application layer signaling that contain the control information.

24. The apparatus of claim 16, wherein the instructions for obtaining the feedback information are executable by the processor to cause the apparatus to: Obtain an indication of the correspondence between the one or more parameters and the characteristic.

25. The apparatus of claim 16, wherein the instructions for obtaining the feedback information are executable by the processor to cause the apparatus to: Obtain an indication of one or more of the parameters.

26. A method for conducting wireless communication at a user equipment (UE), the method comprising: Receive control information from a network entity indicating the characteristics of a dataset used to train a machine learning model at the UE, the machine learning model being associated with maintaining a wireless communication link; The determination of whether to activate or deactivate the first machine learning model for maintaining the wireless communication link or the functionality of the second machine learning model for maintaining the wireless communication link is based at least in part on the received control information. as well as Feedback information indicating one or more parameters of the first machine learning model or one or more parameters of the functionality of the second machine learning model is sent to the network entity, at least in part based on the determination.

27. The method of claim 26, wherein receiving the control information comprises: Receive a dataset identifier corresponding to the dataset or a feature identifier corresponding to the feature, or both.

28. The method of claim 27, further comprising: An uplink message is sent to the network entity, the uplink message including one or more dataset identifiers corresponding to one or more recommended datasets, one or more feature identifiers corresponding to one or more recommended features, or both, wherein the control information is based at least in part on the uplink message.

29. A method for conducting wireless communication at a network entity, the method comprising: The output indicates control information for the characteristics of a dataset used to train a machine learning model at a user equipment (UE), the machine learning model being associated with maintaining a wireless communication link; as well as In response to the control information, feedback information indicating one or more parameters of the first machine learning model or one or more parameters of the functionality of the second machine learning model is obtained, at least in part, based on determining whether to activate or deactivate the first machine learning model for maintaining the wireless communication link or to enable or disable the functionality of the second machine learning model for maintaining the wireless communication link.

30. The method of claim 29, wherein outputting the control information comprises: Output the dataset identifier corresponding to the dataset or the feature identifier corresponding to the feature, or both.

Images