WO2024036587A1 - Sélection de modèle d'apprentissage automatique pour prédiction de faisceau - Google Patents

Sélection de modèle d'apprentissage automatique pour prédiction de faisceau Download PDF

Info

Publication number
WO2024036587A1
WO2024036587A1 PCT/CN2022/113501 CN2022113501W WO2024036587A1 WO 2024036587 A1 WO2024036587 A1 WO 2024036587A1 CN 2022113501 W CN2022113501 W CN 2022113501W WO 2024036587 A1 WO2024036587 A1 WO 2024036587A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
models
processor
network entity
capability information
Prior art date
Application number
PCT/CN2022/113501
Other languages
English (en)
Inventor
Tianyang BAI
Hua Wang
Yuwei REN
Junyi Li
Taesang Yoo
Original Assignee
Qualcomm Incorporated
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qualcomm Incorporated filed Critical Qualcomm Incorporated
Priority to PCT/CN2022/113501 priority Critical patent/WO2024036587A1/fr
Publication of WO2024036587A1 publication Critical patent/WO2024036587A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H04W8/24—Transfer of terminal data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0495—Quantised networks; Sparse networks; Compressed networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/088—Non-supervised learning, e.g. competitive learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning

Definitions

  • aspects of the present disclosure generally relate to wireless communication and specifically, to techniques and apparatuses for selecting machine learning models for beam prediction.
  • Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, and broadcasts.
  • Typical wireless communication systems may employ multiple-access technologies capable of supporting communication with multiple users by sharing available system resources (for example, bandwidth or transmit power) .
  • multiple-access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, orthogonal frequency division multiple access (OFDMA) systems, single-carrier frequency division multiple access (SC-FDMA) systems, time division synchronous code division multiple access (TD-SCDMA) systems, and Long Term Evolution (LTE) .
  • LTE/LTE-Advanced is a set of enhancements to the Universal Mobile Telecommunications System (UMTS) mobile standard promulgated by the Third Generation Partnership Project (3GPP) .
  • UMTS Universal Mobile Telecommunications System
  • New Radio which may be referred to as 5G, is a set of enhancements to the LTE mobile standard promulgated by the 3GPP.
  • NR is designed to better support mobile broadband internet access by improving spectral efficiency, lowering costs, improving services, making use of new spectrum, and better integrating with other open standards using orthogonal frequency division multiplexing (OFDM) with a cyclic prefix (CP) (CP-OFDM) on the downlink, using CP-OFDM or single-carrier frequency division multiplexing (SC-FDM) (also known as discrete Fourier transform spread OFDM (DFT-s-OFDM) ) on the uplink, as well as supporting beamforming, multiple-input multiple-output (MIMO) antenna technology, and carrier aggregation.
  • OFDM orthogonal frequency division multiplexing
  • SC-FDM single-carrier frequency division multiplexing
  • MIMO multiple-input multiple-output
  • Machine learning may be used for beam management.
  • a UE may use an ML model to predict a beam to use for communication.
  • Inputs to the ML model may include past signal strength measurements of candidate beams or interference information.
  • An output of the ML model may be an identification of a preferred beam.
  • the UE may be configured with an ML model that has too much or not enough complexity for the UE.
  • the UE may use additional overhead when providing information about the use of ML models.
  • the method may include transmitting capability information associated with one or more capabilities of the UE associated with using machine learning (ML) models for beam prediction.
  • the method may include receiving an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the method may include selecting a beam using the selected ML model.
  • the method may include communicating using the selected beam.
  • ML machine learning
  • the method may include receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the method may include selecting an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the method may include transmitting an indication of the ML model.
  • the method may include transmitting capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the method may include receiving an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the method may include selecting an ML model from among the multiple ML models based at least in part on a selection rule.
  • the method may include selecting a beam using the ML model.
  • the method may include communicating using the selected beam.
  • the method may include receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the method may include selecting multiple ML models for beam prediction based at least in part on the capability information.
  • the method may include transmitting an indication of the multiple ML models.
  • the UE may include at least one processor and at least one memory, communicatively coupled with the at least one processor, that stores processor-readable code.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to receive an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to select a beam using the selected ML model.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to communicate using the selected beam.
  • the network entity may include at least one processor and at least one memory, communicatively coupled with the at least one processor, that stores processor-readable code.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to select an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to transmit an indication of the ML model.
  • the UE may include at least one processor and at least one memory, communicatively coupled with the at least one processor, that stores processor-readable code.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to receive an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to select an ML model from among the multiple ML models based at least in part on a selection rule.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to select a beam using the ML model.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the UE to communicate using the selected beam.
  • the network entity may include at least one processor and at least one memory, communicatively coupled with the at least one processor, that stores processor-readable code.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to select multiple ML models for beam prediction based at least in part on the capability information.
  • the processor-readable code when executed by the at least one processor, may be configured to cause the network entity to transmit an indication of the multiple ML models.
  • Some aspects described herein relate to a non-transitory computer-readable medium that stores a set of instructions for wireless communication by a UE.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to receive an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to select a beam using the selected ML model.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to communicate using the selected beam.
  • Some aspects described herein relate to a non-transitory computer-readable medium that stores a set of instructions for wireless communication by a network entity.
  • the set of instructions when executed by one or more processors of the network entity, may cause the network entity to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the set of instructions when executed by one or more processors of the network entity, may cause the network entity to select an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the set of instructions, when executed by one or more processors of the network entity may cause the network entity to transmit an indication of the ML model.
  • Some aspects described herein relate to a non-transitory computer-readable medium that stores a set of instructions for wireless communication by a UE.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to receive an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to select an ML model from among the multiple ML models based at least in part on a selection rule.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to select a beam using the ML model.
  • the set of instructions when executed by one or more processors of the UE, may cause the UE to communicate using the selected beam.
  • Some aspects described herein relate to a non-transitory computer-readable medium that stores a set of instructions for wireless communication by a network entity.
  • the set of instructions when executed by one or more processors of the network entity, may cause the network entity to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the set of instructions when executed by one or more processors of the network entity, may cause the network entity to select multiple ML models for beam prediction based at least in part on the capability information.
  • the set of instructions, when executed by one or more processors of the network entity may cause the network entity to transmit an indication of the multiple ML models.
  • the apparatus may include means for transmitting capability information associated with one or more capabilities of the apparatus associated with using ML models for beam prediction.
  • the apparatus may include means for receiving an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the apparatus may include means for selecting a beam using the selected ML model.
  • the apparatus may include means for communicating using the selected beam.
  • the apparatus may include means for receiving capability information associated with one or more capabilities of another apparatus associated with using ML models for beam prediction.
  • the apparatus may include means for selecting an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the apparatus may include means for transmitting an indication of the ML model.
  • the apparatus may include means for transmitting capability information associated with one or more capabilities of the apparatus associated with using ML models for beam prediction.
  • the apparatus may include means for receiving an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the apparatus may include means for selecting an ML model from among the multiple ML models based at least in part on a selection rule.
  • the apparatus may include means for selecting a beam using the ML model.
  • the apparatus may include means for communicating using the selected beam.
  • the apparatus may include means for receiving capability information associated with one or more capabilities of another apparatus associated with using ML models for beam prediction.
  • the apparatus may include means for selecting multiple ML models for beam prediction based at least in part on the capability information.
  • the apparatus may include means for transmitting an indication of the multiple ML models.
  • aspects generally include a method, apparatus, system, computer program product, non-transitory computer-readable medium, UE, base station, network node, network entity, wireless communication device, or processing system as substantially described with reference to and as illustrated by the drawings and specification.
  • Figure 1 is a diagram illustrating an example of a wireless network in accordance with the present disclosure.
  • Figure 2 is a diagram illustrating an example network node in communication with a user equipment (UE) in a wireless network in accordance with the present disclosure.
  • UE user equipment
  • Figure 3 is a diagram illustrating an example disaggregated base station architecture in accordance with the present disclosure.
  • Figure 4 is a diagram illustrating examples of channel state information reference signal beam management procedures, in accordance with the present disclosure.
  • Figure 5 is a diagram illustrating an example of training and using a machine learning (ML) model in connection with selecting ML models for beam prediction.
  • ML machine learning
  • Figure 6 is a diagram illustrating an example of beam prediction at a UE, in accordance with the present disclosure.
  • Figure 7 is a diagram illustrating an example associated with ML model selection, in accordance with the present disclosure.
  • Figure 8 is a diagram illustrating another example associated with ML model selection, in accordance with the present disclosure.
  • Figure 9 is a flowchart illustrating an example process performed, for example, by a UE that supports ML model selection in accordance with the present disclosure.
  • Figure 10 is a flowchart illustrating an example process performed, for example, by a network entity that supports ML model selection in accordance with the present disclosure.
  • Figure 11 is a flowchart illustrating an example process performed, for example, by a UE that supports ML model selection in accordance with the present disclosure.
  • Figure 12 is a flowchart illustrating an example process performed, for example, by a network entity that supports ML model selection in accordance with the present disclosure.
  • Figure 13 is a diagram of an example apparatus for wireless communication that supports ML model selection in accordance with the present disclosure.
  • Figure 14 is a diagram of an example apparatus for wireless communication that supports ML model selection in accordance with the present disclosure.
  • Figure 15 is a diagram of an example apparatus for wireless communication that supports ML model selection in accordance with the present disclosure.
  • Figure 16 is a diagram of an example apparatus for wireless communication that supports ML model selection in accordance with the present disclosure.
  • Various aspects relate generally to selection of a machine learning (ML) model for beam prediction. Some aspects more specifically relate to ML model selection based on one or more user equipment (UE) capabilities.
  • a UE may transmit, to a network entity, an indication of one or more UE capabilities associated with using ML models for beam prediction.
  • the network entity may select an ML model for the UE based at least in part on the capabilities indicated by the UE.
  • multiple ML models may be previously stored at the UE and the network entity may transmit an indication of the ML model to the UE, which may then use the selected ML model.
  • the network entity may transmit multiple ML models, or indications of the multiple ML models, to the UE, and the UE may select an ML model from among the multiple ML models based at least in part on a selection rule.
  • the selection rule may involve, for example, a battery level of the UE, an indication from a network entity, or a selection algorithm.
  • the UE may then use the ML model to predict a beam.
  • the UE may subsequently select and use the predicted beam for communication.
  • the described techniques can be used to improve beam selection and reduce latency. This may improve communications and conserve battery power, processing resources, and signaling resources. For example, using an ML model associated with one or more capabilities of a UE for beam prediction may enable quicker beam selection. Beam selection by beam prediction may be quicker than beam selection involving multiple beam measurements. Furthermore, using the ML model associated with the one or more capabilities of the UE may enable more efficient use of UE resources for beam selection, because the ML model may be selected to utilize UE resources without straining the UE resources.
  • FIG. 1 is a diagram illustrating an example of a wireless network in accordance with the present disclosure.
  • the wireless network 100 may be or may include elements of a 5G (for example, NR) network or a 4G (for example, Long Term Evolution (LTE) ) network, among other examples.
  • the wireless network 100 may include one or more network entities such as network nodes 110 (shown as a network node (NN) 110a, a network node 110b, a network node 110c, and a network node 110d) , a UE 120 or multiple UEs 120 (shown as a UE 120a, a UE 120b, a UE 120c, a UE 120d, and a UE 120e) , or other network entities.
  • network nodes 110 shown as a network node (NN) 110a, a network node 110b, a network node 110c, and a network node 110d
  • a UE 120 or multiple UEs 120 shown as a
  • a network node 110 is an entity that communicates with UEs 120. As shown, a network node 110 may include one or more network nodes. For example, a network node 110 may be an aggregated network node, meaning that the aggregated network node is configured to utilize a radio protocol stack that is physically or logically integrated within a single RAN node (for example, within a single device or unit) .
  • a network node 110 may be a disaggregated network node (sometimes referred to as a disaggregated base station) , meaning that the network node 110 is configured to utilize a protocol stack that is physically or logically distributed among two or more nodes (such as one or more central units (CUs) , one or more distributed units (DUs) , or one or more radio units (RUs) ) .
  • CUs central units
  • DUs distributed units
  • RUs radio units
  • a network node 110 is or includes a network node that communicates with UEs 120 via a radio access link, such as an RU. In some examples, a network node 110 is or includes a network node that communicates with other network nodes 110 via a fronthaul link or a midhaul link, such as a DU. In some examples, a network node 110 is or includes a network node that communicates with other network nodes 110 via a midhaul link or a core network via a backhaul link, such as a CU.
  • a network node 110 may include multiple network nodes, such as one or more RUs, one or more CUs, or one or more DUs.
  • a network node 110 may include, for example, an NR network node, an LTE network node, a Node B, an eNB (for example, in 4G) , a gNB (for example, in 5G) , an access point, or a transmission reception point (TRP) , a DU, an RU, a CU, a mobility element of a network, a core network node, a network element, a network equipment, a RAN node, or a combination thereof.
  • the network nodes 110 may be interconnected to one another or to one or more other network nodes 110 in the wireless network 100 through various types of fronthaul, midhaul, or backhaul interfaces, such as a direct physical connection, an air interface, or a virtual network, using any suitable transport network.
  • Each network node 110 may provide communication coverage for a particular geographic area.
  • the term “cell” can refer to a coverage area of a network node 110 or a network node subsystem serving this coverage area, depending on the context in which the term is used.
  • a network node 110 may provide communication coverage for a macro cell, a pico cell, a femto cell, or another type of cell.
  • a macro cell may cover a relatively large geographic area (for example, several kilometers in radius) and may allow unrestricted access by UEs 120 with service subscriptions.
  • a pico cell may cover a relatively small geographic area and may allow unrestricted access by UEs 120 with service subscription.
  • a femto cell may cover a relatively small geographic area (for example, a home) and may allow restricted access by UEs 120 having association with the femto cell (for example, UEs 120 in a closed subscriber group (CSG) ) .
  • CSG closed subscriber group
  • a network node 110 for a macro cell may be referred to as a macro network node.
  • a network node 110 for a pico cell may be referred to as a pico network node.
  • a network node 110 for a femto cell may be referred to as a femto network node or an in-home network node.
  • the wireless network 100 may be a heterogeneous network that includes network nodes 110 of different types, such as macro network nodes, pico network nodes, femto network nodes, or relay network nodes. These different types of network nodes 110 may have different transmit power levels, different coverage areas, or different impacts on interference in the wireless network 100.
  • macro network nodes may have a high transmit power level (for example, 5 to 40 watts) whereas pico network nodes, femto network nodes, and relay network nodes may have lower transmit power levels (for example, 0.1 to 2 watts) .
  • the network node 110a may be a macro network node for a macro cell 102a
  • the network node 110b may be a pico network node for a pico cell 102b
  • the network node 110c may be a femto network node for a femto cell 102c.
  • a network node may support one or multiple (for example, three) cells.
  • a cell may not necessarily be stationary, and the geographic area of the cell may move in accordance with the location of a network node 110 that is mobile (for example, a mobile network node) .
  • base station, ” “network entity” , or “network node” may refer to an aggregated base station, a disaggregated base station, an integrated access and backhaul (IAB) node, a relay node, or one or more components thereof.
  • base station, ” “network entity, ” or “network node” may refer to a CU, a DU, an RU, a Near-Real Time (Near-RT) RAN Intelligent Controller (RIC) , or a Non-Real Time (Non-RT) RIC, or a combination thereof.
  • the term “base station, ” “network entity, ” or “network node” may refer to one device configured to perform one or more functions, such as those described herein in connection with the network node 110. In some aspects, the term “base station, ” “network entity, ” or “network node” may refer to a plurality of devices configured to perform the one or more functions.
  • each of a quantity of different devices may be configured to perform at least a portion of a function, or to duplicate performance of at least a portion of the function
  • the term “base station, ” “network entity, ” or “network node” may refer to any one or more of those different devices.
  • the term “base station, ” “network entity, ” or “network node” may refer to one or more virtual base stations or one or more virtual base station functions.
  • two or more base station functions may be instantiated on a single device.
  • the term “base station, ” “network entity, ” or “network node” may refer to one of the base station functions and not another. In this way, a single device may include more than one base station.
  • a network controller 130 may couple to or communicate with a set of network nodes 110 and may provide coordination and control for these network nodes 110.
  • the network controller 130 may communicate with the network nodes 110 via a backhaul communication link.
  • the network nodes 110 may communicate with one another directly or indirectly via a wireless or wireline backhaul communication link.
  • the network controller 130 may be a CU or a core network device, or the network controller 130 may include a CU or a core network device.
  • a cell may not necessarily be stationary, and the geographic area of the cell may move in accordance with the location of a network node 110 that is mobile (for example, a mobile network node) .
  • the network nodes 110 may be interconnected to one another or to one or more other network nodes 110 or network nodes (not shown) in the wireless network 100 through various types of backhaul interfaces, such as a direct physical connection or a virtual network, using any suitable transport network.
  • the wireless network 100 may include one or more relay stations.
  • a relay station is an entity that can receive a transmission of data from an upstream station (for example, a network node 110 or a UE 120) and send a transmission of the data to a downstream station (for example, a UE 120 or a network node 110) .
  • a relay station may be a UE 120 that can relay transmissions for other UEs 120.
  • the network node 110d (for example, a relay network node) may communicate with the network node 110a (for example, a macro network node) and the UE 120d in order to facilitate communication between the network node 110a and the UE 120d.
  • a network node 110 that relays communications may be referred to as a relay station, a relay network node, or a relay.
  • the UEs 120 may be dispersed throughout the wireless network 100, and each UE 120 may be stationary or mobile.
  • a UE 120 may include, for example, an access terminal, a terminal, a mobile station, or a subscriber unit.
  • a UE 120 may be a cellular phone (for example, a smart phone) , a personal digital assistant (PDA) , a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device, a biometric device, a wearable device (for example, a smart watch, smart clothing, smart glasses, a smart wristband, smart jewelry (for example, a smart ring or a smart bracelet) ) , an entertainment device (for example, a music device, a video device, or a satellite radio) , a vehicular component or sensor, a smart
  • Some UEs 120 may be considered machine-type communication (MTC) or evolved or enhanced machine-type communication (eMTC) UEs.
  • An MTC UE or an eMTC UE may include, for example, a robot, a drone, a remote device, a sensor, a meter, a monitor, or a location tag, that may communicate with a network node, another device (for example, a remote device) , or some other entity.
  • Some UEs 120 may be considered Internet-of-Things (IoT) devices or may be implemented as NB-IoT (narrowband IoT) devices.
  • Some UEs 120 may be considered a Customer Premises Equipment.
  • a UE 120 may be included inside a housing that houses components of the UE 120, such as processor components or memory components.
  • the processor components and the memory components may be coupled together.
  • the processor components for example, one or more processors
  • the memory components for example, a memory
  • the processor components and the memory components may be operatively coupled, communicatively coupled, electronically coupled, or electrically coupled.
  • any quantity of wireless networks 100 may be deployed in a given geographic area.
  • Each wireless network 100 may support a particular RAT and may operate on one or more frequencies.
  • a RAT may be referred to as a radio technology or an air interface.
  • a frequency may be referred to as a carrier or a frequency channel.
  • Each frequency may support a single RAT in a given geographic area in order to avoid interference between wireless networks of different RATs.
  • NR or 5G RAT networks may be deployed.
  • two or more UEs 120 may communicate directly using one or more sidelink channels (for example, without using a network node 110 as an intermediary to communicate with one another) .
  • the UEs 120 may communicate using peer-to-peer (P2P) communications, device-to-device (D2D) communications, a vehicle-to-everything (V2X) protocol (for example, which may include a vehicle-to-vehicle (V2V) protocol, a vehicle-to-infrastructure (V2I) protocol, or a vehicle-to-pedestrian (V2P) protocol) , or a mesh network.
  • V2X vehicle-to-everything
  • a UE 120 may perform scheduling operations, resource selection operations, or other operations described elsewhere herein as being performed by the network node 110.
  • Devices of the wireless network 100 may communicate using the electromagnetic spectrum, which may be subdivided by frequency or wavelength into various classes, bands, or channels.
  • devices of the wireless network 100 may communicate using one or more operating bands.
  • two initial operating bands have been identified as frequency range designations FR1 (410 MHz –7.125 GHz) and FR2 (24.25 GHz –52.6 GHz) .
  • FR1 frequency range designations FR1 (410 MHz –7.125 GHz)
  • FR2 24.25 GHz –52.6 GHz)
  • FR1 is often referred to (interchangeably) as a “Sub-6 GHz” band in various documents and articles.
  • FR2 which is often referred to (interchangeably) as a “millimeter wave” band in documents and articles, despite being different from the extremely high frequency (EHF) band (30 GHz –300 GHz) which is identified by the International Telecommunications Union (ITU) as a “millimeter wave” band.
  • EHF extremely high frequency
  • ITU International Telecommunications Union
  • FR3 7.125 GHz –24.25 GHz
  • FR3 7.125 GHz –24.25 GHz
  • Frequency bands falling within FR3 may inherit FR1 characteristics or FR2 characteristics, and thus may effectively extend features of FR1 or FR2 into mid-band frequencies.
  • higher frequency bands are currently being explored to extend 5G NR operation beyond 52.6 GHz.
  • FR4a or FR4-1 52.6 GHz –71 GHz
  • FR4 52.6 GHz –114.25 GHz
  • FR5 114.25 GHz –300 GHz
  • sub-6 GHz may broadly represent frequencies that may be less than 6 GHz, may be within FR1, or may include mid-band frequencies.
  • millimeter wave if used herein, may broadly represent frequencies that may include mid-band frequencies, may be within FR2, FR4, FR4-a or FR4-1, or FR5, or may be within the EHF band. It is contemplated that the frequencies included in these operating bands (for example, FR1, FR2, FR3, FR4, FR4-a, FR4-1, or FR5) may be modified, and techniques described herein are applicable to those modified frequency ranges.
  • the UE 120 may include a communication manager 140.
  • the communication manager 140 may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the communication manager 140 may receive an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the communication manager 140 may select a beam using the selected ML model and communicate using the selected beam.
  • the communication manager 140 may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the communication manager 140 may receive an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the communication manager 140 may select an ML model from among the multiple ML models based at least in part on a selection rule.
  • the communication manager 140 may select a beam using the ML model and communicate using the selected beam. Additionally or alternatively, the communication manager 140 may perform one or more other operations described herein.
  • a network entity may include a communication manager 150.
  • the communication manager 150 may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the communication manager 150 may select an ML model for beam prediction based at least in part on the one or more capabilities of the UE and transmit an indication of the ML model.
  • the communication manager 150 may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the communication manager 150 may select multiple ML models for beam prediction based at least in part on the capability information and transmit an indication of the multiple ML models. Additionally or alternatively, the communication manager 150 may perform one or more other operations described herein.
  • FIG 2 is a diagram illustrating an example network node in communication with a UE in a wireless network in accordance with the present disclosure.
  • the network node may correspond to the network node 110 of Figure 1.
  • the UE may correspond to the UE 120 of Figure 1.
  • the network node 110 may be equipped with a set of antennas 234a through 234t, such as T antennas (T ⁇ 1) .
  • the UE 120 may be equipped with a set of antennas 252a through 252r, such as R antennas (R ⁇ 1) .
  • the network node 110 of depicted in Figure 2 includes one or more radio frequency components, such as antennas 234 and a modem 254.
  • a network node 110 may include an interface, a communication component, or another component that facilitates communication with the UE 120 or another network node. Some network nodes 110 may not include radio frequency components that facilitate direct communication with the UE 120, such as one or more CUs, or one or more DUs.
  • a transmit processor 220 may receive data, from a data source 212, intended for the UE 120 (or a set of UEs 120) .
  • the transmit processor 220 may select one or more modulation and coding schemes (MCSs) for the UE 120 based at least in part on one or more channel quality indicators (CQIs) received from that UE 120.
  • MCSs modulation and coding schemes
  • CQIs channel quality indicators
  • the network node 110 may process (for example, encode and modulate) the data for the UE 120 based at least in part on the MCS (s) selected for the UE 120 and may provide data symbols for the UE 120.
  • the transmit processor 220 may process system information (for example, for semi-static resource partitioning information (SRPI) ) and control information (for example, CQI requests, grants, or upper layer signaling) and provide overhead symbols and control symbols.
  • the transmit processor 220 may generate reference symbols for reference signals (for example, a cell-specific reference signal (CRS) or a demodulation reference signal (DMRS) ) and synchronization signals (for example, a primary synchronization signal (PSS) or a secondary synchronization signal (SSS) ) .
  • reference signals for example, a cell-specific reference signal (CRS) or a demodulation reference signal (DMRS)
  • synchronization signals for example, a primary synchronization signal (PSS) or a secondary synchronization signal (SSS)
  • a transmit (TX) multiple-input multiple-output (MIMO) processor 230 may perform spatial processing (for example, precoding) on the data symbols, the control symbols, the overhead symbols, or the reference symbols, if applicable, and may provide a set of output symbol streams (for example, T output symbol streams) to a corresponding set of modems 232 (for example, T modems) , shown as modems 232a through 232t.
  • each output symbol stream may be provided to a modulator component (shown as MOD) of a modem 232.
  • Each modem 232 may use a respective modulator component to process a respective output symbol stream (for example, for OFDM) to obtain an output sample stream.
  • Each modem 232 may further use a respective modulator component to process (for example, convert to analog, amplify, filter, or upconvert) the output sample stream to obtain a downlink signal.
  • the modems 232a through 232t may transmit a set of downlink signals (for example, T downlink signals) via a corresponding set of antennas 234 (for example, T antennas) , shown as antennas 234a through 234t.
  • a set of antennas 252 may receive the downlink signals from the network node 110 or other network nodes 110 and may provide a set of received signals (for example, R received signals) to a set of modems 254 (for example, R modems) , shown as modems 254a through 254r.
  • each received signal may be provided to a demodulator component (shown as DEMOD) of a modem 254.
  • DEMOD demodulator component
  • Each modem 254 may use a respective demodulator component to condition (for example, filter, amplify, downconvert, or digitize) a received signal to obtain input samples.
  • Each modem 254 may use a demodulator component to further process the input samples (for example, for OFDM) to obtain received symbols.
  • a MIMO detector 256 may obtain received symbols from the modems 254, may perform MIMO detection on the received symbols if applicable, and may provide detected symbols.
  • a receive processor 258 may process (for example, demodulate and decode) the detected symbols, may provide decoded data for the UE 120 to a data sink 260, and may provide decoded control information and system information to a controller/processor 280.
  • controller/processor may refer to one or more controllers, one or more processors, or a combination thereof.
  • a channel processor may determine a reference signal received power (RSRP) parameter, a received signal strength indicator (RSSI) parameter, a reference signal received quality (RSRQ) parameter, or a CQI parameter, among other examples.
  • RSRP reference signal received power
  • RSSI received signal strength indicator
  • RSSRQ reference signal received quality
  • CQI CQI parameter
  • the network controller 130 may include a communication unit 294, a controller/processor 290, and a memory 292.
  • the network controller 130 may include, for example, one or more devices in a core network.
  • the network controller 130 may communicate with the network node 110 via the communication unit 294.
  • One or more antennas may include, or may be included within, one or more antenna panels, one or more antenna groups, one or more sets of antenna elements, or one or more antenna arrays, among other examples.
  • An antenna panel, an antenna group, a set of antenna elements, or an antenna array may include one or more antenna elements (within a single housing or multiple housings) , a set of coplanar antenna elements, a set of non-coplanar antenna elements, or one or more antenna elements coupled to one or more transmission or reception components, such as one or more components of Figure 2.
  • a transmit processor 264 may receive and process data from a data source 262 and control information (for example, for reports that include RSRP, RSSI, RSRQ, or CQI) from the controller/processor 280.
  • the transmit processor 264 may generate reference symbols for one or more reference signals.
  • the symbols from the transmit processor 264 may be precoded by a TX MIMO processor 266 if applicable, further processed by the modems 254 (for example, for DFT-s-OFDM or CP- OFDM) and transmitted to the network node 110.
  • the modem 254 of the UE 120 may include a modulator and a demodulator.
  • the UE 120 includes a transceiver.
  • the transceiver may include any combination of the antenna (s) 252, the modem (s) 254, the MIMO detector 256, the receive processor 258, the transmit processor 264, or the TX MIMO processor 266.
  • the transceiver may be used by a processor (for example, the controller/processor 280) and the memory 282 to perform aspects of any of the methods described herein.
  • the uplink signals from UE 120 or other UEs may be received by the antennas 234, processed by the modem 232 (for example, a demodulator component, shown as DEMOD, of the modem 232) , detected by a MIMO detector 236 if applicable, and further processed by a receive processor 238 to obtain decoded data and control information sent by the UE 120.
  • the receive processor 238 may provide the decoded data to a data sink 239 and provide the decoded control information to the controller/processor 240.
  • the network node 110 may include a communication unit 244 and may communicate with the network controller 130 via the communication unit 244.
  • the network node 110 may include a scheduler 246 to schedule one or more UEs 120 for downlink or uplink communications.
  • the modem 232 of the network node 110 may include a modulator and a demodulator.
  • the network node 110 includes a transceiver.
  • the transceiver may include any combination of the antenna (s) 234, the modem (s) 232, the MIMO detector 236, the receive processor 238, the transmit processor 220, or the TX MIMO processor 230.
  • the transceiver may be used by a processor (for example, the controller/processor 240) and the memory 242 to perform aspects of any of the methods described herein.
  • a controller/processor of a network entity may perform one or more techniques associated with selecting ML models for beam prediction, as described in more detail elsewhere herein.
  • the controller/processor 240 of the network node 110, the controller/processor 280 of the UE 120, or any other component (s) of Figure 2 may perform or direct operations of, for example, process 900 of Figure 9, process 1000 of Figure 10, process 1100 of Figure 11, process 1200 of Figure 12, or other processes as described herein.
  • the memory 242 and the memory 282 may store data and program codes for the network node 110 and the UE 120, respectively.
  • the memory 242 or the memory 282 may include a non-transitory computer-readable medium storing one or more instructions (for example, code or program code) for wireless communication.
  • the one or more instructions when executed (for example, directly, or after compiling, converting, or interpreting) by one or more processors of the network node 110 or the UE 120, may cause the one or more processors, the UE 120, or the network node 110 to perform or direct operations of, for example, process 900 of Figure 9, process 1000 of Figure 10, process 1100 of Figure 11, process 1200 of Figure 12, or other processes as described herein.
  • executing instructions may include running the instructions, converting the instructions, compiling the instructions, or interpreting the instructions, among other examples.
  • the UE 120 includes means for transmitting capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction; means for receiving an indication of a selected ML model for beam prediction based on transmitting the capability information; means for selecting a beam using the selected ML model; or means for communicating using the selected beam.
  • the means for the UE to perform operations described herein may include, for example, one or more of communication manager 140, antenna 252, modem 254, MIMO detector 256, receive processor 258, transmit processor 264, TX MIMO processor 266, controller/processor 280, or memory 282.
  • a network entity (for example, network node 110) includes means for receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction; means for selecting an ML model for beam prediction based at least in part on the one or more capabilities of the UE; or means for transmitting an indication of the ML model.
  • the means for the network entity to perform operations described herein may include, for example, one or more of communication manager 150, transmit processor 220, TX MIMO processor 230, modem 232, antenna 234, MIMO detector 236, receive processor 238, controller/processor 240, memory 242, or scheduler 246.
  • the UE 120 includes means for transmitting capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction; means for receiving an indication of multiple ML models for beam prediction based on transmitting the capability information; means for selecting an ML model from among the multiple ML models based at least in part on a selection rule; means for selecting a beam using the ML model; or means for communicating using the selected beam.
  • the network entity includes means for receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction; means for selecting multiple ML models for beam prediction based at least in part on the capability information; or means for transmitting an indication of the multiple ML models.
  • Deployment of communication systems may be arranged in multiple manners with various components or constituent parts.
  • a network node, a network entity, a mobility element of a network, a RAN node, a core network node, a network element, a base station, or a network equipment may be implemented in an aggregated or disaggregated architecture.
  • a base station such as a Node B (NB) , an evolved NB (eNB) , an NR BS, a 5G NB, an access point (AP) , a TRP, or a cell, among other examples
  • NB Node B
  • eNB evolved NB
  • NR BS NR BS
  • 5G NB 5G NB
  • AP access point
  • TRP TRP
  • a cell a cell, among other examples
  • a base station such as a Node B (NB) , an evolved NB (eNB) , an NR BS, a 5G NB, an access point (AP) , a TRP, or a cell, among other examples
  • AP access point
  • TRP Transmission Protocol
  • a cell a cell
  • a base station such as a Node B (NB) , an evolved NB (eNB) , an NR BS, a 5G NB, an access point (AP) , a TRP
  • An aggregated base station may be configured to utilize a radio protocol stack that is physically or logically integrated within a single RAN node (for example, within a single device or unit) .
  • a disaggregated base station (for example, a disaggregated network node) may be configured to utilize a protocol stack that is physically or logically distributed among two or more units (such as one or more CUs, one or more DUs, or one or more RUs) .
  • a CU may be implemented within a network node, and one or more DUs may be co-located with the CU or alternatively, may be geographically or virtually distributed throughout one or multiple other network nodes.
  • the DUs may be implemented to communicate with one or more RUs.
  • Each of the CU, DU, and RU also can be implemented as virtual units, such as a virtual central unit (VCU) , a virtual distributed unit (VDU) , or a virtual radio unit (VRU) , among other examples.
  • VCU virtual central unit
  • VDU virtual distributed unit
  • VRU virtual radio unit
  • Base station-type operation or network design may consider aggregation characteristics of base station functionality.
  • disaggregated base stations may be utilized in an IAB network, an open radio access network (O-RAN (such as the network configuration sponsored by the O-RAN Alliance) ) , or a virtualized radio access network (vRAN, also known as a cloud radio access network (C-RAN) ) to facilitate scaling of communication systems by separating base station functionality into one or more units that can be individually deployed.
  • a disaggregated base station may include functionality implemented across two or more units at various physical locations, as well as functionality implemented for at least one unit virtually, which can enable flexibility in network design.
  • the various units of the disaggregated base station can be configured for wired or wireless communication with at least one other unit of the disaggregated base station.
  • FIG. 3 is a diagram illustrating an example disaggregated base station architecture 300 in accordance with the present disclosure.
  • the disaggregated base station architecture 300 may include a CU 310 that can communicate directly with a core network 320 via a backhaul link, or indirectly with the core network 320 through one or more disaggregated control units (such as a Near-RT RIC 325 via an E2 link, or a Non-RT RIC 315 associated with a Service Management and Orchestration (SMO) Framework 305, or both) .
  • a CU 310 may communicate with one or more DUs 330 via respective midhaul links, such as through F1 interfaces.
  • Each of the DUs 330 may communicate with one or more RUs 340 via respective fronthaul links.
  • Each of the RUs 340 may communicate with one or more UEs 120 via respective radio frequency (RF) access links.
  • RF radio frequency
  • Each of the units may include one or more interfaces or be coupled with one or more interfaces configured to receive or transmit signals, data, or information (collectively, signals) via a wired or wireless transmission medium.
  • Each of the units, or an associated processor or controller providing instructions to one or multiple communication interfaces of the respective unit, can be configured to communicate with one or more of the other units via the transmission medium.
  • each of the units can include a wired interface, configured to receive or transmit signals over a wired transmission medium to one or more of the other units, and a wireless interface, which may include a receiver, a transmitter or transceiver (such as a RF transceiver) , configured to receive or transmit signals, or both, over a wireless transmission medium to one or more of the other units.
  • a wireless interface which may include a receiver, a transmitter or transceiver (such as a RF transceiver) , configured to receive or transmit signals, or both, over a wireless transmission medium to one or more of the other units.
  • the CU 310 may host one or more higher layer control functions.
  • control functions can include radio resource control (RRC) functions, packet data convergence protocol (PDCP) functions, or service data adaptation protocol (SDAP) functions, among other examples.
  • RRC radio resource control
  • PDCP packet data convergence protocol
  • SDAP service data adaptation protocol
  • Each control function can be implemented with an interface configured to communicate signals with other control functions hosted by the CU 310.
  • the CU 310 may be configured to handle user plane functionality (for example, Central Unit –User Plane (CU-UP) functionality) , control plane functionality (for example, Central Unit –Control Plane (CU-CP) functionality) , or a combination thereof.
  • the CU 310 can be logically split into one or more CU-UP units and one or more CU-CP units.
  • a CU-UP unit can communicate bidirectionally with a CU-CP unit via an interface, such as the E1 interface when implemented in an O-RAN configuration.
  • the CU 310 can be implemented to communicate with a DU 330, as necessary, for network control and signaling.
  • Each DU 330 may correspond to a logical unit that includes one or more base station functions to control the operation of one or more RUs 340.
  • the DU 330 may host one or more of a radio link control (RLC) layer, a MAC layer, and one or more high physical (PHY) layers depending, at least in part, on a functional split, such as a functional split defined by the 3GPP.
  • the one or more high PHY layers may be implemented by one or more modules for forward error correction (FEC) encoding and decoding, scrambling, and modulation and demodulation, among other examples.
  • FEC forward error correction
  • the DU 330 may further host one or more low PHY layers, such as implemented by one or more modules for a fast Fourier transform (FFT) , an inverse FFT (iFFT) , digital beamforming, or physical random access channel (PRACH) extraction and filtering, among other examples.
  • FFT fast Fourier transform
  • iFFT inverse FFT
  • PRACH physical random access channel
  • Each layer (which also may be referred to as a module) can be implemented with an interface configured to communicate signals with other layers (and modules) hosted by the DU 330, or with the control functions hosted by the CU 310.
  • Each RU 340 may implement lower-layer functionality.
  • an RU 340, controlled by a DU 330 may correspond to a logical node that hosts RF processing functions or low-PHY layer functions, such as performing an FFT, performing an iFFT, digital beamforming, or PRACH extraction and filtering, among other examples, based on a functional split (for example, a functional split defined by the 3GPP) , such as a lower layer functional split.
  • each RU 340 can be operated to handle over the air (OTA) communication with one or more UEs 120.
  • OTA over the air
  • real-time and non-real-time aspects of control and user plane communication with the RU (s) 340 can be controlled by the corresponding DU 330.
  • this configuration can enable each DU 330 and the CU 310 to be implemented in a cloud-based RAN architecture, such as a vRAN architecture.
  • the SMO Framework 305 may be configured to support RAN deployment and provisioning of non-virtualized and virtualized network elements.
  • the SMO Framework 305 may be configured to support the deployment of dedicated physical resources for RAN coverage requirements, which may be managed via an operations and maintenance interface (such as an O1 interface) .
  • the SMO Framework 305 may be configured to interact with a cloud computing platform (such as an open cloud (O-Cloud) platform 390) to perform network element life cycle management (such as to instantiate virtualized network elements) via a cloud computing platform interface (such as an O2 interface) .
  • a cloud computing platform such as an open cloud (O-Cloud) platform 390
  • network element life cycle management such as to instantiate virtualized network elements
  • a cloud computing platform interface such as an O2 interface
  • Such virtualized network elements can include, but are not limited to, CUs 310, DUs 330, RUs 340, non-RT RICs 315, and Near-RT RICs 325.
  • the SMO Framework 305 can communicate with a hardware aspect of a 4G RAN, such as an open eNB (O-eNB) 311, via an O1 interface. Additionally, in some implementations, the SMO Framework 305 can communicate directly with each of one or more RUs 340 via a respective O1 interface.
  • the SMO Framework 305 also may include a Non-RT RIC 315 configured to support functionality of the SMO Framework 305.
  • the Non-RT RIC 315 may be configured to include a logical function that enables non-real-time control and optimization of RAN elements and resources, Artificial Intelligence/Machine Learning (AI/ML) workflows including model training and updates, or policy-based guidance of applications/features in the Near-RT RIC 325.
  • the Non-RT RIC 315 may be coupled to or communicate with (such as via an A1 interface) the Near-RT RIC 325.
  • the Near-RT RIC 325 may be configured to include a logical function that enables near-real-time control and optimization of RAN elements and resources via data collection and actions over an interface (such as via an E2 interface) connecting one or more CUs 310, one or more DUs 330, or both, as well as an O-eNB, with the Near-RT RIC 325.
  • the Non-RT RIC 315 may receive parameters or external enrichment information from external servers. Such information may be utilized by the Near-RT RIC 325 and may be received at the SMO Framework 305 or the Non-RT RIC 315 from non-network data sources or from network functions. In some examples, the Non-RT RIC 315 or the Near-RT RIC 325 may be configured to tune RAN behavior or performance. For example, the Non-RT RIC 315 may monitor long-term trends and patterns for performance and employ AI/ML models to perform corrective actions through the SMO Framework 305 (such as reconfiguration via an O1 interface) or via creation of RAN management policies (such as A1 interface policies) .
  • FIG 4 is a diagram illustrating examples of channel state information (CSI) reference signal (CSI-RS) beam management procedures, in accordance with the present disclosure.
  • examples 400, 410, and 420 include a UE 120 in communication with a network entity (for example, network node 110) in a wireless network (for example, wireless network 100) .
  • the devices shown in Figure 4 are provided as examples, and the wireless network may support communication and beam management between other devices (for example, between a UE 120 and a network entity or transmit receive point (TRP) , between a mobile termination node and a control node, between an integrated access and backhaul (IAB) child node and an IAB parent node, or between a scheduled node and a scheduling node) .
  • the UE 120 and the network entity may be in a connected state (for example, an RRC connected state) .
  • example 400 may include a network entity (for example, one or more network node devices such as an RU, a DU, or a CU, among other examples) and a UE 120 communicating to perform beam management using CSI-RSs.
  • Example 400 depicts a first beam management procedure (for example, P1 CSI-RS beam management) .
  • the first beam management procedure may be referred to as a beam selection procedure, an initial beam acquisition procedure, a beam sweeping procedure, a cell search procedure, or a beam search procedure.
  • CSI-RSs may be configured to be transmitted from the network entity to the UE 120.
  • the CSI-RSs may be configured to be periodic (for example, using RRC signaling) , semi-persistent (for example, using media access control (MAC) control element (MAC-CE) signaling) , or aperiodic (for example, using downlink control information (DCI) ) .
  • periodic for example, using RRC signaling
  • semi-persistent for example, using media access control (MAC) control element (MAC-CE) signaling
  • MAC-CE media access control element
  • DCI downlink control information
  • the first beam management procedure may include the network entity performing beam sweeping over multiple transmit (Tx) beams.
  • the network entity may transmit a CSI-RS using each transmit beam for beam management.
  • the network node may use a transmit beam to transmit (for example, with repetitions) each CSI-RS at multiple times within the same RS resource set so that the UE 120 can sweep through receive beams in multiple transmission instances. For example, if the network entity has a set of N transmit beams and the UE 120 has a set of M receive beams, the CSI-RS may be transmitted on each of the N transmit beams M times so that the UE 120 may receive M instances of the CSI-RS per transmit beam.
  • the UE 120 may perform beam sweeping through the receive beams of the UE 120.
  • the first beam management procedure may enable the UE 120 to measure a CSI-RS on different transmit beams using different receive beams to support selection of network entity transmit beams/UE 120 receive beam (s) beam pair (s) .
  • the UE 120 may report the measurements to the network entity to enable the network entity to select one or more beam pair (s) for communication between the network entity and the UE 120.
  • the first beam management process may also use synchronization signal blocks (SSBs) for beam management in a similar manner as described above.
  • SSBs synchronization signal blocks
  • example 410 may include a network entity and a UE 120 communicating to perform beam management using CSI-RSs.
  • Example 410 depicts a second beam management procedure (for example, P2 CSI-RS beam management) .
  • the second beam management procedure may be referred to as a beam refinement procedure, a network node beam refinement procedure, a TRP beam refinement procedure, or a transmit beam refinement procedure.
  • CSI-RSs may be configured to be transmitted from the network entity to the UE 120.
  • the CSI-RSs may be configured to be aperiodic (for example, using DCI) .
  • the second beam management procedure may include the network entity performing beam sweeping over one or more transmit beams.
  • the one or more transmit beams may be a subset of all transmit beams associated with the network entity (for example, determined based at least in part on measurements reported by the UE 120 in connection with the first beam management procedure) .
  • the network entity may transmit a CSI-RS using each transmit beam of the one or more transmit beams for beam management.
  • the UE 120 may measure each CSI-RS using a single (for example, the same) receive beam (for example, determined based at least in part on measurements performed in connection with the first beam management procedure) .
  • the second beam management procedure may enable the network entity to select a best transmit beam based at least in part on measurements of the CSI-RSs (for example, measured by the UE 120 using the single receive beam) reported by the UE 120.
  • example 420 depicts a third beam management procedure (for example, P3 CSI-RS beam management) .
  • the third beam management procedure may be referred to as a beam refinement procedure, a UE beam refinement procedure, or a receive beam refinement procedure.
  • one or more CSI-RSs may be configured to be transmitted from the network entity to the UE 120.
  • the CSI-RSs may be configured to be aperiodic (for example, using DCI) .
  • the third beam management process may include the network entity transmitting the one or more CSI-RSs using a single transmit beam (for example, determined based at least in part on measurements reported by the UE 120 in connection with the first beam management procedure or the second beam management procedure) .
  • the network node may use a transmit beam to transmit (for example, with repetitions) CSI-RS at multiple times within the same RS resource set so that UE 120 can sweep through one or more receive beams in multiple transmission instances.
  • the one or more receive beams may be a subset of all receive beams associated with the UE 120 (for example, determined based at least in part on measurements performed in connection with the first beam management procedure or the second beam management procedure) .
  • the third beam management procedure may enable the network entity or the UE 120 to select a best receive beam based at least in part on reported measurements received from the UE 120 (for example, of the CSI-RS of the transmit beam using the one or more receive beams) .
  • beam management may utilize ML, including artificial intelligence (AI) .
  • the UE 120 may use an ML model to predict a beam to use for communication.
  • An ML model may be, for example, a neural network model that is defined with a model structure and a parameter set.
  • the model structure may be identified by a unique model identifier (ID) and use a parameter set.
  • the parameter set may include weights for the neural network model and other configuration parameters.
  • the parameter set may be location specific or configuration specific.
  • Each ML model (identified by a model ID) may be associated with a neural network function (identified by a neural network function ID) that receives inputs and provides an output. Inputs to the neural network function may include, for example, past signal strength measurements of candidate beams or interference information.
  • An output of the neural network function may be an identification of a beam.
  • the beam may be considered a “preferred beam” if it has a positive metric that is greater than other candidate beams, such as a greater signal strength or less interference.
  • One neural network function may be supported by multiple model structures.
  • FIG. 5 is a diagram illustrating an example of training and using an ML model in connection with selecting ML models for beam prediction, in accordance with the present disclosure.
  • the ML model training and usage described herein may be performed using an ML system.
  • the ML system may include or may be included in the UE 120, a network entity (for example, network node 110) , a computing device, a server, or a cloud computing environment.
  • an ML model may be trained using a set of observations.
  • the set of observations may be obtained from training data (for example, historical data) , such as data gathered during one or more processes described herein.
  • the ML system may receive the set of observations (for example, as input) from past beam measurements, as described elsewhere herein.
  • the set of observations includes a feature set.
  • the feature set may include a set of variables, and a variable may be referred to as a feature.
  • a specific observation may include a set of variable values (or feature values) corresponding to the set of variables.
  • the ML system may determine variables for a set of observations or variable values for a specific observation based on input received from UEs. For example, the ML system may identify a feature set (for example, one or more features or feature values) by extracting the feature set from structured data, by performing natural language processing to extract the feature set from unstructured data, or by receiving input from an operator.
  • a feature set for a set of observations may include a first feature of an RSRP for a beam 1, an RSRP for a beam 2, and an RSRP for a beam 3.
  • the first feature may have a first decibel milliwatt (dBm) value (for example, -70 dBm)
  • the second feature may have a second dBm value (for example, -40 dBm)
  • the third feature may have a third dBm value (for example, -80 dBm) , and so forth.
  • the feature set may include one or more of the following features: RSSIs for candidate beams, signal-to-noise-plus-interference ratios (SINRs) for candidate beams, or other type of measurements for beams.
  • SINRs signal-to-noise-plus-interference ratios
  • the set of observations may be associated with a target variable.
  • the target variable may represent a variable having a numeric value, may represent a variable having a numeric value that falls within a range of values or has some discrete possible values, may represent a variable that is selectable from one of multiple options (for example, one of multiples classes, classifications, or labels) or may represent a variable having a Boolean value.
  • a target variable may be associated with a target variable value, and a target variable value may be specific to an observation. In example 500, the target variable is a preferred beam, or a beam that has a greatest dBm value among the observations.
  • the target variable may represent a value that an ML model is being trained to predict
  • the feature set may represent the variables that are input to a trained ML model to predict a value for the target variable.
  • the set of observations may include target variable values so that the ML model can be trained to recognize patterns in the feature set that lead to a target variable value.
  • An ML model that is trained to predict a target variable value may be referred to as a supervised learning model.
  • the ML model may be trained on a set of observations that do not include a target variable. This may be referred to as an unsupervised learning model.
  • the ML model may learn patterns from the set of observations without labeling or supervision, and may provide output that indicates such patterns, such as by using clustering or association to identify related groups of items within the set of observations.
  • the ML system may train an ML model using the set of observations and using one or more ML algorithms, such as a regression algorithm, a decision tree algorithm, a neural network algorithm, a k-nearest neighbor algorithm, a support vector machine algorithm, among other examples.
  • the ML system may store the ML model as a trained ML model 525 to be used to analyze new observations.
  • the ML system may apply the trained ML model 525 to a new observation, such as by receiving a new observation and inputting the new observation to the trained ML model 525.
  • the ML system may apply the trained ML model 525 to the new observation to generate an output (for example, a result) .
  • the type of output may depend on the type of ML model or the type of ML task being performed.
  • the output may include a predicted value of a target variable, such as when supervised learning is employed.
  • the output may include information that identifies a cluster to which the new observation belongs or information that indicates a degree of similarity between the new observation and one or more other observations, such as when unsupervised learning is employed.
  • the trained ML model 525 may predict values of, for example, an RSRP, an RSSI, or an SINR of beams for the target variable of a preferred beam for the new observation, as shown in a sixth operation 535. Based on this prediction, the ML system may provide a first recommendation, may provide output for determination of a first recommendation, may perform a first automated action, or may cause a first automated action to be performed (for example, by instructing another device to perform the automated action) , among other examples.
  • the first recommendation may include, for example, beam 3.
  • the first automated action may include, for example, selecting beam 3.
  • the ML: system may provide a second (for example, different) recommendation (for example, beam 4) or may perform or cause performance of a second (for example, different) automated action (for example, select beam 4) .
  • a feature set may also include dBm values that are based on a location or movement of the UE 120.
  • the trained ML model 525 may classify (for example, cluster) the new observation in a cluster, as shown in a seventh operation 540.
  • the observations within a cluster may have a threshold degree of similarity.
  • the ML system may provide a first recommendation, such as the first recommendation described above.
  • the ML system may perform a first automated action or may cause a first automated action to be performed (for example, by instructing another device to perform the automated action) based on classifying the new observation in the first cluster, such as the first automated action described above.
  • the ML system may provide a second (for example, different) recommendation (for example, beam 5) or may perform or cause performance of a second (for example, different) automated action, such as selection of beam 5.
  • a second (for example, different) recommendation for example, beam 5
  • a second (for example, different) automated action such as selection of beam 5.
  • the recommendation or the automated action associated with the new observation may be based on a target variable value having a particular label (for example, classification or categorization) , may be based on whether a target variable value satisfies one or more thresholds (for example, whether the target variable value is greater than a threshold, is less than a threshold, is equal to a threshold, falls within a range of threshold values, among other values) , or may be based on a cluster in which the new observation is classified.
  • a target variable value having a particular label for example, classification or categorization
  • thresholds for example, whether the target variable value is greater than a threshold, is less than a threshold, is equal to a threshold, falls within a range of threshold values, among other values
  • the trained ML model 525 may be re-trained using feedback information.
  • feedback may be provided to the ML model.
  • the feedback may be associated with actions performed based on the recommendations provided by the trained ML model 525 or automated actions performed, or caused, by the trained ML model 525.
  • the recommendations or actions output by the trained ML model 525 may be used as inputs to re-train the ML model (for example, a feedback loop may be used to train or update the ML model) .
  • the feedback information may include observed beam measurements for predicted beams.
  • the ML system may apply a rigorous and automated process to beam management.
  • the ML system enables recognition or identification of tens, hundreds, thousands, or millions of features or feature values for tens, hundreds, thousands, or millions of observations, thereby increasing accuracy and consistency and reducing delay associated with beam selection relative to requiring computing resources to be allocated for tens, hundreds, or thousands of operators to manually measure and select beams using the features or feature values.
  • FIG. 6 is a diagram illustrating an example of beam prediction at a UE, in accordance with the present disclosure.
  • a network entity may configure a UE (for example, the UE 120) with an ML model configuration to execute ML models at the UE 120.
  • the UE 120 may have more measurement results for beams than the network entity and this may improve an accuracy of beam selection.
  • the ML models at the UE 120 may be configured by the network entity.
  • the network entity may transmit reference signals for measurement.
  • the UE 120 may execute an ML model based on local beam measurements (plus network entity signaling) .
  • the UE 120 may transmit a prediction report of the prediction results of the ML model. The report may be transmitted based on the ML model configuration or triggering conditions.
  • the UE 120 may use an ML model to predict a beam to use for communication
  • the UE 120 may be configured with an ML model that has too much or not enough complexity for the UE 120. For example, if an ML model involves too many computations for the processing capability of the UE 120, the UE 120 may not be able to obtain a beam prediction in a reasonable amount of time. As a result, latency is introduced and the beam selection may be suboptimal. This may waste battery power, processing resources and signaling resources.
  • a UE may transmit capability information that indicates one or more UE capabilities associated with using ML models for beam prediction.
  • a network entity may select an ML model for the UE 120 based at least in part on the one or more UE capabilities.
  • the network entity may provide multiple ML models to the UE 120, and the UE 120 may select the ML model.
  • the UE 120 may select the ML model based at least in part on a selection rule.
  • the selection rule may involve, for example, a battery level of the UE 120, an indication from a network entity, or a selection algorithm.
  • the UE 120 may use the ML model to predict a beam.
  • the UE 120 may select and use the beam for communication.
  • the described techniques can be used to reduce latency, conserve power at the UE 120, and improve beam selection. Quicker and more accurate beam selection improves communications and conserves processing resources and signaling resources.
  • the UE 120 may train an algorithm of an ML model to predict future measurements (for example, RSRP) of a beam or beam set (beam set 1) based on past measurements of a previous beam or beam set (beam set 2) at multiple instances of time, such as at time k, time k-1, up to k + 1 -n for n iterations. Beam set 1 and beam set 2 may be the same, overlapped, or totally different.
  • the algorithm may be a recursive neural network or a traditional algorithm.
  • the algorithm may be trained and maintained by the network entity.
  • the network entity may configured the UE 120 with an ML model that uses the algorithm.
  • the algorithm may be executed by the network entity or the UE 120.
  • the UE 120 may measure a subset of SSBs to predict future SSBs.
  • the UE 120 may measure SSBs to predict refined CSI-RS beams for unicast physical downlink shared channel (PDSCH) or physical downlink control channel (PDCC) communications.
  • an output of an ML model may be a beam ID for a preferred beam (for example, a beam with a greatest signal strength) at a future time or other related metrics at a future time.
  • a preferred beam for example, a beam with a greatest signal strength
  • the UE 120 may conserve reference signal resources because reference signals are transmitted less frequently for beam tracking and channel estimation.
  • the UE 120 may conserve uplink feedback because channel estimation is less frequent.
  • the UE 120 may conserve power and processing resources because the UE 120 does not need to measure and transmit feedback as frequently.
  • the UE 120 may measure 24 wide SSB beams in a cell and predict which narrow refined beam is the best out of 48 beam candidates.
  • the UE 120 may use an ML model with an FC layer neural network (24, 48) followed by a softmax output layer.
  • a softmax layer is a type of neural network layer where the output is a probability distribution.
  • the output may be a probability of each narrow beam becoming the preferred beam.
  • the FC layer neural network (24, 96) may be followed by a rectified linear unit (ReLU) function, an FC layer (96, 48) , and a softmax layer to output a predicted probability distribution of the preferred narrow refined beam ID.
  • the parameter size may be (25 ⁇ 96) + (97 ⁇ 48) , which is about 7000 parameters.
  • the multiplication before the softmax layer may be (24 ⁇ 96) + (96 ⁇ 48) , or a little less than 7000 parameters. There may be additional computations due to the ReLU function.
  • an ML model may be a convolutional neural network (CNN) that uses convolution layers.
  • a convolution layer may perform the function of pattern detection.
  • An FC layer neural network may perform classification by combining different types of patterns detected by convolution layers.
  • a CNN may be a convolution transposition (for example, with 10 channels and 4 kernels) that uses an ReLU function for a matrix transpose and a convolution (for example, with 48 channels and 10 kernels) with a softmax output layer.
  • Some ML models may have fewer parameters but with less effective performance. Other models may have better performance but with a larger parameter size. Another factor may be the computational complexity during the inference determination or the online training. Even for the same ML model, when implemented at the UE 120, the trained parameter set and the model structure may be modified for better inference efficiency at the UE 120.
  • parameters may be quantized with different bit widths or pruned. Some parameters in dense connected NNs may be omitted in inference. Different levels of quantization or pruning may lead to different performance degradations. Therefore, by selecting ML models based at least in part on the UE capabilities, ML model performance and beam prediction improve.
  • Figure 7 is a diagram illustrating an example associated with ML model selection, in accordance with the present disclosure.
  • a network entity 710 for example, a network node 110
  • a UE 720 for example, UE 120
  • a wireless network for example, wireless network 100
  • the network entity 710 may select the ML model that the UE 720 is to use.
  • the UE 720 may transmit capability information 726 associated with ML models for beam prediction.
  • the capability information 726 may indicate one or more UE capabilities associated with using ML models for beam prediction.
  • the capability information 726 may indicate a limitation of computation power, power consumption, or memory. For example, the more limited the UE capabilities, the lower the complexity of the ML model.
  • a computation power capability may be, for example, how many operations may be performed given a timeline from inference determination to reporting.
  • a power consumption capability may be a battery level over time.
  • a memory capability may be a memory size.
  • the capability information may also include a graphics processing unit (GPU) availability.
  • GPU graphics processing unit
  • hardware of the UE 720 may support only quantized versions of neural networks. This may include support of floating point quantization or fixed point quantization. This may include support for a bit width (for example, int32 or int16) . The bit width can be different for a parameter set or activation function (for example, a ReLU function) .
  • the capability information 726 may indicate one or more supported architecture types for ML models.
  • the UE 720 may support a certain type of ML model architecture.
  • the ML model architecture types may include FC neural networks, CNNs, recurrent neural networks (RNNs) , or transformer-based neural networks that use sequential data feeding.
  • the ML model architecture may have a maximum quantity of layers and a maximum width of each layer (for example, in terms of the quantity of neuron units in each layer) . That is, the quantity of neurons in each layer may be limited by the UE capability.
  • the capability information 726 may indicate one or more supported parameters for ML models.
  • the capability information 726 may indicate a supported parameter size.
  • the network entity 710 may select an ML model 732 based at least in part on the UE capabilities.
  • the network entity 710 may transmit an indication 736 of the ML model 732.
  • the network entity 710 may configure (for example, via RRC signaling) the UE 720 with multiple ML models that may be activated, and the network entity 710 may transmit the indication 736 of the ML model 732 in DCI or a MAC CE.
  • Model configuration data exchange may occur at a higher layer (for example, an application layer) .
  • the ML model 732 that is configured may change over time.
  • the UE 720 may select a beam 742 using the ML model 732. This may include executing the ML model 732 with past beam measurements (for example, RSRPs for candidate beams) or other inputs.
  • the ML model 732 may output an identification of the beam 742 to select.
  • the beam 742 may be a preferred beam from among multiple neighboring beams (for example, narrow beams) .
  • the beam 742 may be associated with a beam ID that is a relative ID with respect to other candidate beams in a subset of all candidate (narrow) beams. For example, the subset may be the neighboring beams of the beam with a greatest RSRP among measured beam whose RSRPs are used as input of the ML model.
  • the network entity 710 may define a rule to determine the subset of beams (for example, neighboring beams) and transmit information about the rule to UE 720.
  • a rule may specify a one-to-one mapping between the relative ID and an absolute ID for a beam.
  • the relative ID may be defined within the subset of beams.
  • the absolute ID may be defined among all candidate beams.
  • the UE 720 may transmit a beam identifier 746 of the selected beam 742.
  • the network entity 710 and the UE 720 may communicate using the beam 742.
  • Figure 8 is a diagram illustrating another example associated with ML model selection, in accordance with the present disclosure.
  • the network entity 710 may provide multiple ML models to the UE 720 and the UE 720 will select the ML model.
  • the UE 720 may transmit the capability information 726 associated with beam prediction ML models.
  • the capability information 726 may indicate one or more UE capabilities associated with using ML models for beam prediction.
  • the capability information may be transmitted in RRC signaling for more stationary parameters. However, some parameters such as battery level and GPU occupancy may be reported in a MAC CE for quicker updates.
  • the network entity 710 may select multiple ML models 832 based at least in part on the UE capabilities.
  • the network entity 710 may transmit an indication 836 of the multiple ML models 832.
  • the network entity 710 may transmit an indication of a selected ML model to use, selected by the network entity 710.
  • the network entity 710 may transmit information associated with supporting one or more of the multiple ML models 832. This may include a minimum memory, a computation complexity, or performance statistics for each of the multiple ML models 832.
  • the UE 720 may select an ML model 842 of the multiple ML models 832.
  • the UE 720 may select the ML model 842 based at least in part on a selection rule.
  • the selection rule may specify selecting the ML model 842 based at least in part on a beam report configuration.
  • the beam report configuration may specify a timeline or a periodicity for selecting the ML model 842 and for transmitting an identification of a beam as an output of the ML model 842.
  • the UE 720 may use other configured selection rules.
  • the network entity 710 may configure report resources for reporting prediction results and define a timeline between inference and the report. Therefore, the UE 120 may need to select an ML model that can finish the inference within the given timeline.
  • the UE 720 may transmit a model ID 846 of the ML model 842.
  • the network entity 710 may know what performance to expect and to monitor. For example, when the UE 720 transmits feedback for an inference error, the network entity 710 may determine which ML model to update, activate, or deactivate based on the error from the feedback. In another example, if the UE 720 selects a simplified ML model with relatively poor performance (for example, less reliable prediction) , the network entity 710 may schedule more reference signals for beam measurement rather than rely mostly on beam predictions.
  • the UE 720 may receive one or more performance parameters associated with supporting one or more ML models of the multiple ML models 832, and select the ML model 842 based on the UE 720 satisfying performance parameters associated with supporting the ML model 842.
  • the UE 720 may select a beam 852 using the ML model 842. This may include executing the ML model 842.
  • the UE 720 may transmit an output of the ML model 842.
  • the output may be, for example, an identification of the beam 852 to select.
  • the UE 720 may transmit a beam ID 856 of the selected beam 852 in a seventh operation 855.
  • the network entity 710 and the UE 720 may communicate using the selected beam 852.
  • the UE 720 may transmit feedback based at least in part on a performance of the ML model 842. This may involve a comparison of a predicted beam with actual beam measurements.
  • FIG. 9 is a flowchart illustrating an example process 900 performed, for example, by a UE that supports ML model selection in accordance with the present disclosure.
  • Example process 900 is an example where the UE (for example, UE 120, UE 720) performs operations associated with selecting ML models for beam prediction.
  • process 900 may include transmitting capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction (block 910) .
  • the UE (such as by using communication manager 1308 or transmission component 1304 depicted in Figure 13) may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction, as described above.
  • process 900 may include receiving an indication of a selected ML model for beam prediction based on transmitting the capability information (block 920) .
  • the UE (such as by using communication manager 1308 or reception component 1302 depicted in Figure 13) may receive an indication of a selected ML model for beam prediction based on transmitting the capability information, as described above.
  • process 900 may include selecting a beam using the selected ML model (block 930) .
  • the UE such as by using communication manager 1308 or selection component 1310 depicted in Figure 13
  • process 900 may include communicating using the selected beam (block 940) .
  • the UE such as by using communication manager 1308, transmission component 1304, and reception component 1302 depicted in Figure 13
  • Process 900 may include additional aspects, such as any single aspect or any combination of aspects described below or in connection with one or more other processes described elsewhere herein.
  • process 900 includes transmitting a beam ID of the selected beam.
  • the capability information indicates a limitation of one or more of computation power, power consumption, or memory.
  • the capability information indicates one or more supported architecture types for ML models.
  • the capability information indicates one or more supported parameters for ML models.
  • one or more inputs for the selected ML model include RSRP measurements for multiple beams, and an output of the selected ML model is an identification of the selected beam.
  • process 900 includes receiving a configuration for multiple ML models via RRC signaling, and receiving the indication of the selected ML model includes receiving the indication of the selected ML model via DCI or a MAC CE.
  • process 900 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Figure 9. Additionally or alternatively, two or more of the blocks of process 900 may be performed in parallel.
  • FIG 10 is a flowchart illustrating an example process 1000 performed, for example, by a network entity that supports ML model selection in accordance with the present disclosure.
  • Example process 1000 is an example where the network entity (for example, network node 110, network entity 710) performs operations associated with selecting ML models for beam prediction.
  • the network entity for example, network node 110, network entity 710 performs operations associated with selecting ML models for beam prediction.
  • process 1000 may include receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction (block 1010) .
  • the network entity (such as by using communication manager 1408 or reception component 1402 depicted in Figure 14) may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction, as described above.
  • process 1000 may include selecting an ML model for beam prediction based at least in part on the one or more capabilities of the UE (block 1020) .
  • the network entity (such as by using communication manager 1408 or selection component 1410 depicted in Figure 14) may select an ML model for beam prediction based at least in part on the one or more capabilities of the UE, as described above.
  • process 1000 may include transmitting an indication of the ML model (block 1030) .
  • the network entity (such as by using communication manager 1408 or transmission component 1404 depicted in Figure 14) may transmit an indication of the ML model, as described above.
  • Process 1000 may include additional aspects, such as any single aspect or any combination of aspects described below or in connection with one or more other processes described elsewhere herein.
  • process 1000 includes receiving a beam ID of a beam selected by the UE and communicating using the selected beam.
  • the capability information indicates a limitation of one or more of computation power, power consumption, or memory
  • selecting the ML model includes selecting the ML model based at least in part on the limitation of one or more of computation power, power consumption, or memory.
  • the capability information indicates one or more supported architecture types for ML models
  • selecting the ML model includes selecting the ML model based at least in part on the one or more supported architecture types for ML models.
  • the capability information indicates one or more supported parameters for ML models
  • selecting the ML model includes selecting the ML model based at least in part on the one or more supported parameters for ML models.
  • process 1000 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Figure 10. Additionally or alternatively, two or more of the blocks of process 1000 may be performed in parallel.
  • FIG 11 is a flowchart illustrating an example process 1100 performed, for example, by a UE that supports ML model selection in accordance with the present disclosure.
  • Example process 1100 is an example where the UE (for example, UE 120) performs operations associated with selecting ML models for beam prediction.
  • process 1100 may include transmitting capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction (block 1110) .
  • the UE (such as by using communication manager 1508 or transmission component 1504 depicted in Figure 15) may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction, as described above.
  • process 1100 may include receiving an indication of multiple ML models for beam prediction based on transmitting the capability information (block 1120) .
  • the UE (such as by using communication manager 1508 or reception component 1502 depicted in Figure 15) may receive an indication of multiple ML models for beam prediction based on transmitting the capability information, as described above.
  • process 1100 may include selecting an ML model from among the multiple ML models based at least in part on a selection rule (block 1130) .
  • the UE (such as by using communication manager 1508 or selection component 1510 depicted in Figure 15) may select an ML model from among the multiple ML models based at least in part on a selection rule, as described above.
  • process 1100 may include selecting a beam using the ML model (block 1140) .
  • the UE such as by using communication manager 1508 or selection component 1510 depicted in Figure 15
  • process 1100 may include communicating using the selected beam (block 1150) .
  • the UE such as by using communication manager 1508, transmission component 1504, and reception component 1502 depicted in Figure 15
  • Process 1100 may include additional aspects, such as any single aspect or any combination of aspects described below or in connection with one or more other processes described elsewhere herein.
  • process 1100 includes transmitting a model identifier of the ML model.
  • process 1100 includes transmitting feedback based at least in part on performance of the ML model.
  • process 1100 includes transmitting an output of the ML model.
  • the output includes an identification of the selected beam.
  • the indication indicates a selected ML model to use, and selecting the ML model includes selecting the selected ML model.
  • the capability information indicates a limitation of one or more of computation power, power consumption, or memory.
  • the capability information indicates one or more supported architecture types for ML models.
  • the capability information indicates one or more supported parameters for ML models.
  • process 1100 includes receiving one or more performance parameters associated with supporting one or more ML models of the multiple ML models, and selecting the ML model includes selecting the ML model based on the UE satisfying performance parameters associated with supporting the ML model.
  • the selection rule specifies selecting the ML model based at least in part on a beam report configuration.
  • the beam report configuration specifies a timeline or a periodicity for selecting the ML model and for transmitting an identification of a beam as an output of the ML model.
  • process 1100 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Figure 11. Additionally or alternatively, two or more of the blocks of process 1100 may be performed in parallel.
  • Figure 12 is a flowchart illustrating an example process 1200 performed, for example, by a network entity that supports ML Model selection in accordance with the present disclosure.
  • Example process 1200 is an example where the network entity (for example, network node 110, network entity 710) performs operations associated with selecting ML models for beam prediction.
  • the network entity for example, network node 110, network entity 710 performs operations associated with selecting ML models for beam prediction.
  • process 1200 may include receiving capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction (block 1210) .
  • the network entity (such as by using communication manager 1608 or reception component 1602 depicted in Figure 16) may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction, as described above.
  • process 1200 may include selecting multiple ML models for beam prediction based at least in part on the capability information (block 1220) .
  • the network entity (such as by using communication manager 1608 or selection component 1610 depicted in Figure 16) may select multiple ML models for beam prediction based at least in part on the capability information, as described above.
  • process 1200 may include transmitting an indication of the multiple ML models (block 1230) .
  • the network entity (such as by using communication manager 1608 or transmission component 1604 depicted in Figure 16) may transmit an indication of the multiple ML models, as described above.
  • Process 1200 may include additional aspects, such as any single aspect or any combination of aspects described below or in connection with one or more other processes described elsewhere herein.
  • process 1200 includes receiving a model ID of an ML model selected by the UE.
  • process 1200 includes receiving feedback that indicates a performance of the ML model.
  • the indication indicates a selected ML model to use.
  • process 1200 includes transmitting one or more performance parameters associated with supporting one or more ML models of the multiple ML models.
  • process 1200 includes transmitting a selection rule for selecting an ML model.
  • process 1200 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Figure 12. Additionally or alternatively, two or more of the blocks of process 1200 may be performed in parallel.
  • FIG 13 is a diagram of an example apparatus 1300 for wireless communication that supports ML model selection in accordance with the present disclosure.
  • the apparatus 1300 may be a UE (for example, UE 120, UE 720) , or a UE may include the apparatus 1300.
  • the apparatus 1300 includes a reception component 1302, a transmission component 1304, and a communication manager 1308, which may be in communication with one another (for example, via one or more buses) .
  • the communication manager 1308 may control and/or otherwise manage one or more operations of the reception component 1302 and/or the transmission component 1304.
  • the communication manager 1308 may include one or more antennas, a modem, a controller/processor, a memory, or a combination thereof, of the UE described in connection with Figure 2.
  • the communication manager 1308 may be, or be similar to, the communication manager 140 depicted in Figures 1 and 2.
  • the communication manager 1308 may be configured to perform one or more of the functions described as being performed by the communication manager 140.
  • the communication manager 1308 may include the reception component 1302 and/or the transmission component 1304.
  • the apparatus 1300 may communicate with another apparatus 1306 (such as a UE, a network node, or another wireless communication device) using the reception component 1302 and the transmission component 1304.
  • the apparatus 1300 may be configured to perform one or more operations described herein in connection with Figures 1-8. Additionally or alternatively, the apparatus 1300 may be configured to perform one or more processes described herein, such as process 900 of Figure 9. In some aspects, the apparatus 1300 may include one or more components of the UE described above in connection with Figure 2.
  • the reception component 1302 may receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 1306.
  • the reception component 1302 may provide received communications to one or more other components of the apparatus 1300, such as the communication manager 140.
  • the reception component 1302 may perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples) , and may provide the processed signals to the one or more other components.
  • the reception component 1302 may include one or more antennas, a modem, a demodulator, a MIMO detector, a receive processor, a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • the transmission component 1304 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 1306.
  • the communication manager 140 may generate communications and may transmit the generated communications to the transmission component 1304 for transmission to the apparatus 1306.
  • the transmission component 1304 may perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) , and may transmit the processed signals to the apparatus 1306.
  • the transmission component 1304 may include one or more antennas, a modem, a modulator, a transmit MIMO processor, a transmit processor, a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2. In some aspects, the transmission component 1304 may be co-located with the reception component 1302 in a transceiver.
  • the communication manager 1308 may transmit or may cause the transmission component 1304 to transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the communication manager 1308 may receive or may cause the reception component 1302 to receive an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the communication manager 1308 may select a beam using the selected ML model.
  • the communication manager 1308 may communicate using the selected beam.
  • the communication manager 1308 may perform one or more operations described elsewhere herein as being performed by one or more components of the communication manager 1308.
  • the communication manager 1308 may include a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • the communication manager 1308 includes a set of components, such as a selection component 1310.
  • the set of components may be separate and distinct from the communication manager 1308.
  • one or more components of the set of components may include or may be implemented within a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • one or more components of the set of components may be implemented at least in part as software stored in a memory.
  • a component (or a portion of a component) may be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or a processor to perform the functions or operations of the component.
  • the transmission component 1304 may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the reception component 1302 may receive an indication of a selected ML model for beam prediction based on transmitting the capability information.
  • the selection component 1310 may select a beam using the selected ML model.
  • the transmission component 1304 and the reception component 1302 may communicate using the selected beam.
  • the transmission component 1304 may transmit a beam ID of the selected beam.
  • the reception component 1302 may receive a configuration for multiple ML models via RRC signaling and receive the indication of the selected ML model via DCI or a MAC CE.
  • FIG. 13 The quantity and arrangement of components shown in Figure 13 are provided as an example. In practice, there may be additional components, fewer components, different components, or differently arranged components than those shown in Figure 13. Furthermore, two or more components shown in Figure 13 may be implemented within a single component, or a single component shown in Figure 13 may be implemented as multiple, distributed components. Additionally or alternatively, a set of (one or more) components shown in Figure 13 may perform one or more functions described as being performed by another set of components shown in Figure 13.
  • FIG 14 is a diagram of an example apparatus 1400 for wireless communication that supports ML model selection in accordance with the present disclosure.
  • the apparatus 1400 may be a network entity (for example, network node 110, network entity 710) , or a network entity may include the apparatus 1400.
  • the apparatus 1400 includes a reception component 1402, a transmission component 1404, and a communication manager 1408, which may be in communication with one another (for example, via one or more buses) .
  • the communication manager 1408 may control and/or otherwise manage one or more operations of the reception component 1402 and/or the transmission component 1404.
  • the communication manager 1408 may include one or more antennas, a modem, a controller/processor, a memory, or a combination thereof, of the network entity described in connection with Figure 2.
  • the communication manager 1408 may be, or be similar to, the communication manager 150 depicted in Figures 1 and 2.
  • the communication manager 1408 may be configured to perform one or more of the functions described as being performed by the communication manager 150.
  • the communication manager 1408 may include the reception component 1402 and/or the transmission component 1404.
  • the apparatus 1400 may communicate with another apparatus 1406 (such as a UE, a network node, a network entity, or another wireless communication device) using the reception component 1402 and the transmission component 1404.
  • the apparatus 1400 may be configured to perform one or more operations described herein in connection with Figures 1-8. Additionally or alternatively, the apparatus 1400 may be configured to perform one or more processes described herein, such as process 1000 of Figure 10. In some aspects, the apparatus 1400 may include one or more components of the network entity described above in connection with Figure 2.
  • the reception component 1402 may receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 1406.
  • the reception component 1402 may provide received communications to one or more other components of the apparatus 1400, such as the communication manager 1408.
  • the reception component 1402 may perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples) , and may provide the processed signals to the one or more other components.
  • the reception component 1402 may include one or more antennas, a modem, a demodulator, a MIMO detector, a receive processor, a controller/processor, a memory, or a combination thereof, of the network entity described above in connection with Figure 2.
  • the transmission component 1404 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 1406.
  • the communication manager 1408 may generate communications and may transmit the generated communications to the transmission component 1404 for transmission to the apparatus 1406.
  • the transmission component 1404 may perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) , and may transmit the processed signals to the apparatus 1406.
  • the transmission component 1404 may include one or more antennas, a modem, a modulator, a transmit MIMO processor, a transmit processor, a controller/processor, a memory, or a combination thereof, of the network entity described above in connection with Figure 2. In some aspects, the transmission component 1404 may be co-located with the reception component 1402 in a transceiver.
  • the communication manager 1408 may receive or may cause the reception component 1402 to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the communication manager 1408 may select an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the communication manager 1408 may transmit or may cause the transmission component 1404 to transmit an indication of the ML model.
  • the communication manager 1408 may perform one or more operations described elsewhere herein as being performed by one or more components of the communication manager 1408.
  • the communication manager 1408 may include a controller/processor, a memory, a scheduler, a communication unit, or a combination thereof, of the network entity described above in connection with Figure 2.
  • the communication manager 1408 includes a set of components, such as a selection component 1410.
  • the set of components may be separate and distinct from the communication manager 1408.
  • one or more components of the set of components may include or may be implemented within a controller/processor, a memory, a scheduler, a communication unit, or a combination thereof, of the network entity described above in connection with Figure 2.
  • one or more components of the set of components may be implemented at least in part as software stored in a memory.
  • a component (or a portion of a component) may be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or a processor to perform the functions or operations of the component.
  • the reception component 1402 may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the selection component 1410 may select an ML model for beam prediction based at least in part on the one or more capabilities of the UE.
  • the transmission component 1404 may transmit an indication of the ML model.
  • the reception component 1402 may receive a beam ID of a beam selected by the UE.
  • the transmission component 1404 and the reception component 1402 may communicate using the selected beam.
  • FIG. 14 The quantity and arrangement of components shown in Figure 14 are provided as an example. In practice, there may be additional components, fewer components, different components, or differently arranged components than those shown in Figure 14. Furthermore, two or more components shown in Figure 14 may be implemented within a single component, or a single component shown in Figure 14 may be implemented as multiple, distributed components. Additionally or alternatively, a set of (one or more) components shown in Figure 14 may perform one or more functions described as being performed by another set of components shown in Figure 14.
  • FIG. 15 is a diagram of an example apparatus 1500 for wireless communication that supports ML model selection in accordance with the present disclosure.
  • the apparatus 1500 may be a UE (for example, UE 120, UE 720) , or a UE may include the apparatus 1500.
  • the apparatus 1500 includes a reception component 1502, a transmission component 1504, and a communication manager 1508, which may be in communication with one another (for example, via one or more buses) .
  • the communication manager 1508 may control and/or otherwise manage one or more operations of the reception component 1502 and/or the transmission component 1504.
  • the communication manager 1508 may include one or more antennas, a modem, a controller/processor, a memory, or a combination thereof, of the UE described in connection with Figure 2.
  • the communication manager 1508 may be, or be similar to, the communication manager 140 depicted in Figures 1 and 2.
  • the communication manager 1508 may be configured to perform one or more of the functions described as being performed by the communication manager 140.
  • the communication manager 1508 may include the reception component 1502 and/or the transmission component 1504.
  • the apparatus 1500 may communicate with another apparatus 1506 (such as a UE, a network node, a network entity, or another wireless communication device) using the reception component 1502 and the transmission component 1504.
  • the apparatus 1500 may be configured to perform one or more operations described herein in connection with Figures 1-8. Additionally or alternatively, the apparatus 1500 may be configured to perform one or more processes described herein, such as process 1100 of Figure 11. In some aspects, the apparatus 1500 may include one or more components of the UE described above in connection with Figure 2.
  • the reception component 1502 may receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 1506.
  • the reception component 1502 may provide received communications to one or more other components of the apparatus 1500, such as the communication manager 140.
  • the reception component 1502 may perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples) , and may provide the processed signals to the one or more other components.
  • the reception component 1502 may include one or more antennas, a modem, a demodulator, a MIMO detector, a receive processor, a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • the transmission component 1504 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 1506.
  • the communication manager 140 may generate communications and may transmit the generated communications to the transmission component 1504 for transmission to the apparatus 1506.
  • the transmission component 1504 may perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) , and may transmit the processed signals to the apparatus 1506.
  • the transmission component 1504 may include one or more antennas, a modem, a modulator, a transmit MIMO processor, a transmit processor, a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2. In some aspects, the transmission component 1504 may be co-located with the reception component 1502 in a transceiver.
  • the communication manager 1508 may receive or may cause the reception component 1502 to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the communication manager 1508 may select multiple ML models for beam prediction based at least in part on the capability information.
  • the communication manager 1508 may transmit or may cause the transmission component 1504 to transmit an indication of the multiple ML models.
  • the communication manager 1508 may perform one or more operations described elsewhere herein as being performed by one or more components of the communication manager 1508.
  • the communication manager 1508 may include a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • the communication manager 1508 includes a set of components, such as a selection component 1510, or a combination thereof.
  • the set of components may be separate and distinct from the communication manager 1508.
  • one or more components of the set of components may include or may be implemented within a controller/processor, a memory, or a combination thereof, of the UE described above in connection with Figure 2.
  • one or more components of the set of components may be implemented at least in part as software stored in a memory.
  • a component (or a portion of a component) may be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or a processor to perform the functions or operations of the component.
  • the transmission component 1504 may transmit capability information associated with one or more capabilities of the UE associated with using ML models for beam prediction.
  • the reception component 1502 may receive an indication of multiple ML models for beam prediction based on transmitting the capability information.
  • the selection component 1510 may select an ML model from among the multiple ML models based at least in part on a selection rule.
  • the selection component 1510 may select a beam using the ML model.
  • the transmission component 1504 and the reception component 1502 may communicate using the selected beam.
  • the transmission component 1504 may transmit a model ID of the ML model.
  • the transmission component 1504 may transmit feedback based at least in part on performance of the ML model.
  • the transmission component 1504 may transmit an output of the ML model.
  • the reception component 1502 may receive one or more performance parameters associated with supporting one or more ML models of the multiple ML models and select the ML model based on the UE satisfying performance parameters associated with supporting the ML model.
  • FIG. 15 The quantity and arrangement of components shown in Figure 15 are provided as an example. In practice, there may be additional components, fewer components, different components, or differently arranged components than those shown in Figure 15. Furthermore, two or more components shown in Figure 15 may be implemented within a single component, or a single component shown in Figure 15 may be implemented as multiple, distributed components. Additionally or alternatively, a set of (one or more) components shown in Figure 15 may perform one or more functions described as being performed by another set of components shown in Figure 15.
  • FIG 16 is a diagram of an example apparatus 1600 for wireless communication that supports ML model selection in accordance with the present disclosure.
  • the apparatus 1600 may be a network entity (for example, network node 110, network entity 710) , or a network entity may include the apparatus 1600.
  • the apparatus 1600 includes a reception component 1602, a transmission component 1604, and a communication manager 1608, which may be in communication with one another (for example, via one or more buses) .
  • the communication manager 1608 may control and/or otherwise manage one or more operations of the reception component 1602 and/or the transmission component 1604.
  • the communication manager 1608 may include one or more antennas, a modem, a controller/processor, a memory, or a combination thereof, of the network entity described in connection with Figure 2.
  • the communication manager 1608 may be, or be similar to, the communication manager 150 depicted in Figures 1 and 2.
  • the communication manager 1608 may be configured to perform one or more of the functions described as being performed by the communication manager 150.
  • the communication manager 1608 may include the reception component 1602 and/or the transmission component 1604.
  • the apparatus 1600 may communicate with another apparatus 1606 (such as a UE, a network node, a network entity, or another wireless communication device) using the reception component 1602 and the transmission component 1604.
  • the apparatus 1600 may be configured to perform one or more operations described herein in connection with Figures 1-8. Additionally or alternatively, the apparatus 1600 may be configured to perform one or more processes described herein, such as process 1200 of Figure 12. In some aspects, the apparatus 1600 may include one or more components of the network entity described above in connection with Figure 2.
  • the reception component 1602 may receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 1606.
  • the reception component 1602 may provide received communications to one or more other components of the apparatus 1600, such as the communication manager 1608.
  • the reception component 1602 may perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples) , and may provide the processed signals to the one or more other components.
  • the reception component 1602 may include one or more antennas, a modem, a demodulator, a MIMO detector, a receive processor, a controller/processor, a memory, or a combination thereof, of the network entity described above in connection with Figure 2.
  • the transmission component 1604 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 1606.
  • the communication manager 1608 may generate communications and may transmit the generated communications to the transmission component 1604 for transmission to the apparatus 1606.
  • the transmission component 1604 may perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) , and may transmit the processed signals to the apparatus 1606.
  • the transmission component 1604 may include one or more antennas, a modem, a modulator, a transmit MIMO processor, a transmit processor, a controller/processor, a memory, or a combination thereof, of the network entity described above in connection with Figure 2. In some aspects, the transmission component 1604 may be co-located with the reception component 1602 in a transceiver.
  • the communication manager 1608 may receive or may cause the reception component 1602 to receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the communication manager 1608 may select multiple ML models for beam prediction based at least in part on the capability information.
  • the communication manager 1608 may transmit or may cause the transmission component 1604 to transmit an indication of the multiple ML models.
  • the communication manager 1608 may perform one or more operations described elsewhere herein as being performed by one or more components of the communication manager 1608.
  • the communication manager 1608 may include a controller/processor, a memory, a scheduler, a communication unit, or a combination thereof, of the network entity described above in connection with Figure 2.
  • the communication manager 1608 includes a set of components, such as a selection component 1610.
  • the set of components may be separate and distinct from the communication manager 1608.
  • one or more components of the set of components may include or may be implemented within a controller/processor, a memory, a scheduler, a communication unit, or a combination thereof, of the network entity described above in connection with Figure 2.
  • one or more components of the set of components may be implemented at least in part as software stored in a memory.
  • a component (or a portion of a component) may be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or a processor to perform the functions or operations of the component.
  • the reception component 1602 may receive capability information associated with one or more capabilities of a UE associated with using ML models for beam prediction.
  • the selection component 1610 may select multiple ML models for beam prediction based at least in part on the capability information.
  • the transmission component 1604 may transmit an indication of the multiple ML models.
  • the reception component 1602 may receive a model ID of an ML model selected by the UE.
  • the reception component 1602 may receive feedback that indicates a performance of the ML model.
  • the transmission component 1604 may transmit one or more performance parameters associated with supporting one or more ML models of the multiple ML models.
  • the transmission component 1604 may transmit a selection rule for selecting an ML model.
  • FIG. 16 The quantity and arrangement of components shown in Figure 16 are provided as an example. In practice, there may be additional components, fewer components, different components, or differently arranged components than those shown in Figure 16. Furthermore, two or more components shown in Figure 16 may be implemented within a single component, or a single component shown in Figure 16 may be implemented as multiple, distributed components. Additionally or alternatively, a set of (one or more) components shown in Figure 16 may perform one or more functions described as being performed by another set of components shown in Figure 16.
  • a method of wireless communication performed by a user equipment (UE) comprising: transmitting capability information associated with one or more capabilities of the UE associated with using machine learning (ML) models for beam prediction; receiving an indication of a selected ML model for beam prediction based on transmitting the capability information; selecting a beam using the selected ML model; and communicating using the selected beam.
  • ML machine learning
  • Aspect 2 The method of Aspect 1, further comprising transmitting a beam identifier of the selected beam.
  • Aspect 3 The method of Aspect 1 or 2, wherein the capability information indicates a limitation of one or more of computation power, power consumption, or memory.
  • Aspect 4 The method of any of Aspects 1-3, wherein the capability information indicates one or more supported architecture types for ML models.
  • Aspect 5 The method of any of Aspects 1-4, wherein the capability information indicates one or more supported parameters for ML models.
  • Aspect 6 The method of any of Aspects 1-5, wherein one or more inputs for the selected ML model include reference signal received power (RSRP) measurements for multiple beams, and wherein an output of the selected ML model is an identification of the selected beam.
  • RSRP reference signal received power
  • Aspect 7 The method of any of Aspects 1-6, further comprising receiving a configuration for multiple ML models via radio resource control (RRC) signaling, wherein receiving the indication of the selected ML model includes receiving the indication of the selected ML model via downlink control information (DCI) or a medium access control control element (MAC CE) .
  • RRC radio resource control
  • a method of wireless communication performed by a network entity comprising: receiving capability information associated with one or more capabilities of a user equipment (UE) associated with using machine learning (ML) models for beam prediction; selecting an ML model for beam prediction based at least in part on the one or more capabilities of the UE; and transmitting an indication of the ML model.
  • UE user equipment
  • ML machine learning
  • Aspect 9 The method of Aspect 8, further comprising: receiving a beam identifier of a beam selected by the UE; and communicating using the selected beam.
  • Aspect 10 The method of Aspect 8 or 9, wherein the capability information indicates a limitation of one or more of computation power, power consumption, or memory, and wherein selecting the ML model includes selecting the ML model based at least in part on the limitation of one or more of computation power, power consumption, or memory.
  • Aspect 11 The method of any of Aspects 8-10, wherein the capability information indicates one or more supported architecture types for ML models, and wherein selecting the ML model includes selecting the ML model based at least in part on the one or more supported architecture types for ML models.
  • Aspect 12 The method of any of Aspects 8-11, wherein the capability information indicates one or more supported parameters for ML models, and wherein selecting the ML model includes selecting the ML model based at least in part on the one or more supported parameters for ML models.
  • a method of wireless communication performed by a user equipment (UE) comprising: transmitting capability information associated with one or more capabilities of the UE associated with using machine learning (ML) models for beam prediction; receiving an indication of multiple ML models for beam prediction based on transmitting the capability information; selecting an ML model from among the multiple ML models based at least in part on a selection rule; selecting a beam using the ML model; and communicating using the selected beam.
  • ML machine learning
  • Aspect 14 The method of Aspect 13, further comprising transmitting a model identifier of the ML model.
  • Aspect 15 The method of Aspect 14, further comprising transmitting feedback based at least in part on performance of the ML model.
  • Aspect 16 The method of any of Aspects 13-15, further comprising transmitting an output of the ML model.
  • Aspect 17 The method of Aspect 16, wherein the output includes an identification of the selected beam.
  • Aspect 18 The method of any of Aspects 13-17, wherein the indication indicates a selected ML model to use, and wherein selecting the ML model includes selecting the selected ML model.
  • Aspect 19 The method of any of Aspects 13-18, wherein the capability information indicates a limitation of one or more of computation power, power consumption, or memory.
  • Aspect 20 The method of any of Aspects 13-19, wherein the capability information indicates one or more supported architecture types for ML models.
  • Aspect 21 The method of any of Aspects 13-20, wherein the capability information indicates one or more supported parameters for ML models.
  • Aspect 22 The method of any of Aspects 13-21, further comprising receiving one or more performance parameters associated with supporting one or more ML models of the multiple ML models, wherein selecting the ML model includes selecting the ML model based on the UE satisfying performance parameters associated with supporting the ML model.
  • Aspect 23 The method of any of Aspects 13-22, wherein the selection rule specifies selecting the ML model based at least in part on a beam report configuration.
  • Aspect 24 The method of Aspect 23, wherein the beam report configuration specifies a timeline or a periodicity for selecting the ML model and for transmitting an identification of a beam as an output of the ML model.
  • a method of wireless communication performed by a network entity comprising: receiving capability information associated with one or more capabilities of a user equipment (UE) associated with using machine learning (ML) models for beam prediction; selecting multiple ML models for beam prediction based at least in part on the capability information; and transmitting an indication of the multiple ML models.
  • UE user equipment
  • ML machine learning
  • Aspect 26 The method of Aspect 25, further comprising receiving a model identifier of an ML model selected by the UE.
  • Aspect 27 The method of Aspect 26, further comprising receiving feedback that indicates a performance of the ML model.
  • Aspect 28 The method of any of Aspects 25-27, wherein the indication indicates a selected ML model to use.
  • Aspect 29 The method of any of Aspects 25-28, further comprising transmitting one or more performance parameters associated with supporting one or more ML models of the multiple ML models.
  • Aspect 30 The method of any of Aspects 25-29, further comprising transmitting a selection rule for selecting an ML model.
  • Aspect 31 An apparatus for wireless communication at a device, comprising a processor; memory coupled with the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform the method of one or more of Aspects 1-30.
  • Aspect 32 A device for wireless communication, comprising a memory and one or more processors coupled to the memory, the one or more processors configured to perform the method of one or more of Aspects 1-30.
  • Aspect 33 An apparatus for wireless communication, comprising at least one means for performing the method of one or more of Aspects 1-30.
  • Aspect 34 A non-transitory computer-readable medium storing code for wireless communication, the code comprising instructions executable by a processor to perform the method of one or more of Aspects 1-30.
  • Aspect 35 A non-transitory computer-readable medium storing a set of instructions for wireless communication, the set of instructions comprising one or more instructions that, when executed by one or more processors of a device, cause the device to perform the method of one or more of Aspects 1-30.
  • the term “component” is intended to be broadly construed as hardware or a combination of hardware and software.
  • “Software” shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, or functions, among other examples, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.
  • a “processor” is implemented in hardware or a combination of hardware and software. It will be apparent that systems or methods described herein may be implemented in different forms of hardware or a combination of hardware and software.
  • satisfying a threshold may, depending on the context, refer to a value being greater than the threshold, greater than or equal to the threshold, less than the threshold, less than or equal to the threshold, equal to the threshold, or not equal to the threshold, among other examples.
  • “at least one of: a, b, or c” is intended to cover a, b, c, a + b, a + c, b + c, and a + b + c, as well as any combination with multiples of the same element (for example, a + a, a + a + a, a + a + b, a + a + c, a +b + b, a + c + c, b + b, b + b + b, b + b + c, c + c, and c + c + c, or any other ordering of a, b, and c) .
  • the terms “has, ” “have, ” “having, ” and similar terms are intended to be open-ended terms that do not limit an element that they modify (for example, an element “having” A may also have B) .
  • the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.
  • the term “or” is intended to be inclusive when used in a series and may be used interchangeably with “or, ” unless explicitly stated otherwise (for example, if used in combination with “either” or “only one of” ) .

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

Selon divers aspects, la présente divulgation porte sur le domaine de la communication sans fil. Selon certains aspects, un équipement utilisateur (UE) peut transmettre des informations de capacité associées à une ou à plusieurs capacités de l'UE associées à l'utilisation de modèles d'apprentissage automatique (ML) pour une prédiction de faisceau. L'UE peut recevoir une indication d'un modèle ML sélectionné pour une prédiction de faisceau sur la base de la transmission des informations de capacité. L'UE peut sélectionner un faisceau à l'aide du modèle ML sélectionné. L'UE peut communiquer au moyen du faisceau sélectionné. L'invention concerne également de nombreux autres aspects.
PCT/CN2022/113501 2022-08-19 2022-08-19 Sélection de modèle d'apprentissage automatique pour prédiction de faisceau WO2024036587A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/113501 WO2024036587A1 (fr) 2022-08-19 2022-08-19 Sélection de modèle d'apprentissage automatique pour prédiction de faisceau

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/113501 WO2024036587A1 (fr) 2022-08-19 2022-08-19 Sélection de modèle d'apprentissage automatique pour prédiction de faisceau

Publications (1)

Publication Number Publication Date
WO2024036587A1 true WO2024036587A1 (fr) 2024-02-22

Family

ID=89940416

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/113501 WO2024036587A1 (fr) 2022-08-19 2022-08-19 Sélection de modèle d'apprentissage automatique pour prédiction de faisceau

Country Status (1)

Country Link
WO (1) WO2024036587A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024213187A1 (fr) * 2024-07-18 2024-10-17 Lenovo (Beijing) Limited Indication de condition supplémentaire côté réseau

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210328630A1 (en) * 2020-04-16 2021-10-21 Qualcomm Incorporated Machine learning model selection in beamformed communications
WO2021215995A1 (fr) * 2020-04-23 2021-10-28 Telefonaktiebolaget Lm Ericsson (Publ) Amélioration de l'accès aléatoire sur la base de l'intelligence artificielle/apprentissage automatique (ia/ml)
WO2022015221A1 (fr) * 2020-07-14 2022-01-20 Telefonaktiebolaget Lm Ericsson (Publ) Gestion d'un dispositif sans fil utilisable pour se connecter à un réseau de communication
US20220038349A1 (en) * 2020-10-19 2022-02-03 Ziyi LI Federated learning across ue and ran

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210328630A1 (en) * 2020-04-16 2021-10-21 Qualcomm Incorporated Machine learning model selection in beamformed communications
WO2021215995A1 (fr) * 2020-04-23 2021-10-28 Telefonaktiebolaget Lm Ericsson (Publ) Amélioration de l'accès aléatoire sur la base de l'intelligence artificielle/apprentissage automatique (ia/ml)
WO2022015221A1 (fr) * 2020-07-14 2022-01-20 Telefonaktiebolaget Lm Ericsson (Publ) Gestion d'un dispositif sans fil utilisable pour se connecter à un réseau de communication
US20220038349A1 (en) * 2020-10-19 2022-02-03 Ziyi LI Federated learning across ue and ran

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
VIVO: "New use case for prediction of AI/ML model distribution", 3GPP TSG-SA WG1 MEETING #93E S1-210251R3, 15 March 2021 (2021-03-15), XP093140079, Retrieved from the Internet <URL:https://www.3gpp.org/ftp/tsg_sa/WG1_Serv/TSGS1_93e_Electronic_Meeting/Docs/S1-210251r3.zip> [retrieved on 20240312] *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024213187A1 (fr) * 2024-07-18 2024-10-17 Lenovo (Beijing) Limited Indication de condition supplémentaire côté réseau

Similar Documents

Publication Publication Date Title
US11751072B2 (en) User equipment behavior when using machine learning-based prediction for wireless communication system operation
US20240267710A1 (en) Machine learning group switching
WO2024036587A1 (fr) Sélection de modèle d&#39;apprentissage automatique pour prédiction de faisceau
US20240129750A1 (en) Disabling beam prediction outputs
WO2023028974A1 (fr) Groupe de caractéristiques d&#39;apprentissage automatique de capacité d&#39;équipement utilisateur
US20240073721A1 (en) Measurement compression for multiple reference signals
WO2024065251A1 (fr) Prédictions de caractéristiques de canal sur la base au moins en partie d&#39;un sous-ensemble de ressources de signal de référence de liaison descendante
WO2024000385A1 (fr) Rapport de prédiction de blocage
WO2024092494A1 (fr) Rapport de paire de faisceaux pour mesures de faisceau prédites
WO2023226007A1 (fr) Rapport d&#39;informations d&#39;état de canal pour de multiples groupes de ressources de mesure de canaux
WO2024065375A1 (fr) Transmission d&#39;un rapport de capacité indiquant une capacité de prédiction de faisceau d&#39;un équipement utilisateur
WO2023197205A1 (fr) Prédiction de faisceau de domaine temporel à l&#39;aide d&#39;un rapport d&#39;informations d&#39;état de canal
WO2024007281A1 (fr) Entraînement multi-vendeur hors ligne pour apprentissage automatique inter-nœuds
US20240171991A1 (en) User equipment grouping for federated learning
WO2024092762A1 (fr) Indication de précision pour informations d&#39;état de canal de référence
WO2023206392A1 (fr) Enregistrement de mesures de canal descendant associées à une ou plusieurs instances temporelles au niveau d&#39;un équipement utilisateur
WO2024178525A1 (fr) Caractéristiques d&#39;équipement d&#39;utilisateur pour prédiction de faisceau
WO2024007234A1 (fr) Techniques de fourniture d&#39;informations de rapport d&#39;informations d&#39;état de canal
WO2024098174A1 (fr) Surveillance de modèle à l&#39;aide d&#39;échantillons d&#39;entrée
US20230262568A1 (en) Inter-base station coordination for interference management
WO2024007257A1 (fr) Techniques de prédiction d&#39;états d&#39;indicateur de configuration de transmission de nœud de réseau
WO2024077504A1 (fr) Réalisation de mesures associées à des ressources de mesure de canal à l&#39;aide de sous-ensembles de faisceaux de réception restreints
US20230274181A1 (en) Federated learning model splitting and combining
WO2024060173A1 (fr) Demande de caractéristiques de faisceau prises en charge par un équipement utilisateur pour une gestion prédictive de faisceau
WO2024065655A1 (fr) Recommandation de ressources de signal de référence pour une prédiction de faisceau

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22955367

Country of ref document: EP

Kind code of ref document: A1