Configuring User Devices for Machine Learning

Abstract

Methods, systems, and devices for wireless communications are described. In some examples, a wireless communications system may support machine learning and may configure a user equipment (UE) for machine learning. The UE may send a request message to a base station including an indication of a machine learning model or a neural network function based at least in part on a trigger event. In response to the request message, the base station may send the machine learning model, a set of parameters corresponding to the machine learning model, or a configuration corresponding to the neural network function, and may send an activation message to the UE to implement the machine learning model and the neural network function.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001]

[0001] This patent application claims the benefit of U.S. patent application Ser. No. 17 / 467,156 by ZHU et al., entitled "CONFIGURING A USER EQUIPMENT FOR MACHINE LEARNING," filed Sep. 3, 2021, which is assigned to the present assignee and expressly incorporated by reference herein. [Technical field]

[0002] The following relates to wireless communications, including configuring user equipment (UE) for machine learning. [Background technology]

[0003]

[0003] Wireless communication systems are widely deployed to provide various types of communication content such as voice, video, packet data, messaging, broadcasts, and the like. These systems may be capable of supporting communication with multiple users by sharing available system resources (e.g., time, frequency, and power). Examples of such multiple access systems include fourth generation (4G) systems, such as Long Term Evolution (LTE), LTE Advanced (LTE-A), or LTE-A Pro systems, and fifth generation (5G) systems, which may be referred to as New Radio (NR) systems. These systems may use technologies such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal FDMA (OFDMA), or Discrete Fourier Transform Spread Orthogonal Frequency Division Multiplexing (DFT-S-OFDM). A wireless multiple-access communication system may include one or more base stations, or alternatively, one or more network access nodes, each simultaneously supporting communication for multiple communication devices, which may also be known as user equipment (UE).

[0004] In some examples, a wireless communication system may support machine learning. Machine learning may be described as a branch of artificial intelligence that provides a system with the ability to learn and improve from experience. In some examples, a network may configure a UE for machine learning, and the UE may utilize machine learning to perform tasks such as cell reselection, beam obstruction, beam management, and the like. Summary of the Invention

[0005]

[0005] The described techniques relate to improved methods, systems, devices, and apparatuses that support configuring a user equipment (UE) for machine learning. For example, the described techniques provide for a UE to obtain a set of neural network functions, machine learning models, and corresponding parameters from a network.

[0006]

[0006] In some examples, the UE may transmit capability information to the network. The capability information may include one or more of a list of potential neural network functions, a list of potential machine learning models, or an indication of whether the UE may request machine learning. Based on the capability information, the network may select a set of neural network functions, a set of machine learning models, and a set of corresponding parameters, and in some examples, may indicate them to the UE.

[0007]

[0007] In some examples, the UE may send a message to the network requesting to implement machine learning (e.g., based on some trigger). The message may include an indication of a neural network function, a neural network model, and a corresponding parameter set. In response to the request message, the network may configure the machine learning model and corresponding parameters in the UE. Once the UE obtains the machine learning model and the corresponding parameter set, the network may activate machine learning in the UE, and the UE utilizes the machine learning to perform one or more tasks. The techniques described herein may support machine learning in the UE. Machine learning may enable the UE to perform tasks with little or no instructions from the network. Thus, machine learning in the UE, as supported by the techniques described herein, may result in less signaling overhead and reduce power consumption in the UE.

[0008] A method for wireless communication in a UE is described. The method may include receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning model repository (MR) included in or coupled to a base station, and receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0009]

[0009] An apparatus for wireless communication in a UE is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to receive a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to a base station, and receive an activation message for the machine learning model, the neural network function, or both from the base station.

[0010] Another apparatus for wireless communication in a UE is described. The apparatus may include means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to a base station, and means for receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0011]

[0011] A non-transitory computer-readable medium having stored thereon code for wireless communication in a UE is described. The code may include instructions executable by a processor to receive a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to a base station, and receive an activation message for the machine learning model, the neural network function, or both from the base station.

[0012]

[0012] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for sending a request message to the base station including an indication of the machine learning model, the neural network function, or both, where receiving the machine learning model, the neural network function, or both may be based on the request message.

[0013]

[0013] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include an operation, feature, means, or instruction for receiving signaling from the base station indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, where sending the request message may be based on the machine learning models being included in the whitelist, being excluded from the blacklist, or both.

[0014]

[0014] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, each machine learning model of the one or more machine learning models may be associated with a respective scope corresponding to a location, a network slice, a deep neural network (DNN), a public land mobile network (PLMN), a UE type, a radio resource control (RRC) state, a communication service, a communication configuration, or any combination thereof, and sending the request message may be based on a trigger event including the UE having a condition that may be within the respective scope of the machine learning model.

[0015]

[0015] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the request message includes an indication of the trigger event.

[0016]

[0016] In some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification, sending the request message may include an operation, feature, means, or instruction for sending a UE assistance information message that includes the request message.

[0017]

[0017] In some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification, sending a request message may include an operation, feature, means, or instruction for sending RRC signaling that includes the request message.

[0018]

[0018] Some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification may further include operations, features, means, or instructions whereby sending the request message includes sending the request message to a central unit-control plane (CU-CP) entity included in the base station, and receiving the machine learning model, set of parameters, or configuration includes receiving the machine learning model, set of parameters, or configuration from the CU-CP entity.

[0019]

[0019] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for determining an address for the machine learning model, set of parameters, or configuration based on an associated ID and an associated rule, where receiving the machine learning model, set of parameters, or configuration may be based on a download of the machine learning model, set of parameters, or configuration from the machine learning MR based on the address.

[0020]

[0020] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for determining an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions based on the associated ID and the associated rule, and initiating an upload of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0021]

[0021] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include an operation, feature, means, or instruction for receiving an address for a machine learning model, set of parameters, or configuration from a central unit-machine learning plane (CU-XP) entity included in the base station, where receiving the machine learning model, set of parameters, or configuration may be based on downloading the machine learning model, set of parameters, or configuration from the machine learning MR based on the address.

[0022]

[0022] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for receiving, from a CU-XP entity included in the base station, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions, and initiating uploading of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0023]

[0023] A method for wireless communication in a base station is described. The method may include transmitting to a UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station, and transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0024]

[0024] An apparatus for wireless communication in a base station is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to send to a UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station, and to send to the UE an activation message for the machine learning model, the neural network function, or both.

[0025]

[0025] Another apparatus for wireless communication in a base station is described. The apparatus may include means for transmitting, to a UE, a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station, and means for transmitting, to the UE, an activation message for the machine learning model, the neural network function, or both.

[0026]

[0026] A non-transitory computer-readable medium having stored thereon code for wireless communication in a base station is described. The code may include instructions executable by a processor to send to a UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station, and to send to the UE an activation message for the machine learning model, the neural network function, or both.

[0027]

[0027] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for receiving a request message from the UE including an indication of the machine learning model, the neural network function, or both, where sending the machine learning model, the neural network function, or both may be based on the request message.

[0028]

[0028] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for sending signaling to the UE indicating a first set of machine learning models that are included in a blacklist, a second set of machine learning models that are included in a whitelist, or both, where the machine learning models may be included in the whitelist, may be excluded from the blacklist, or both.

[0029]

[0029] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, each machine learning model of the one or more machine learning models may be associated with a respective scope corresponding to a location, a network slice, a DNN, a PLMN, a UE type, an RRC state, a communication service, a communication configuration, or any combination thereof, and receiving the request message may be based on a trigger event including the UE having a condition that may be within the respective scope of the machine learning model.

[0030]

[0030] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the request message includes an indication of the trigger event.

[0031]

[0031] In some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification, receiving a request message may include an operation, feature, means, or instruction for receiving a UE assistance information message that includes the request message.

[0032]

[0032] In some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification, receiving a request message may include an operation, feature, means, or instruction for receiving RRC signaling that includes the request message.

[0033]

[0033] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for receiving a request message at a CU-CP entity included in the base station, forwarding the request message from the CU-CP entity to a CU-XP entity included in the base station, and downloading a machine learning model, set of parameters, or configuration from the machine learning MR to the CU-CP entity based on the request message, where transmitting the machine learning model, set of parameters, or configuration to the UE may be based on the downloading.

[0034]

[0034] Some examples of the methods, apparatus, and non-transitory computer-readable media described in this specification may further include operations, features, means, or instructions for receiving an address for a machine learning model, set of parameters, or configuration from the UE, and downloading the machine learning model, set of parameters, or configuration from the machine learning MR for the UE based on the address.

[0035]

[0035] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions for receiving from the UE an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions, and uploading the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR.

[0036]

[0036] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions, at a CU-XP entity included in the base station, for receiving an ID associated with the machine learning model, set of parameters, or configuration from the UE, determining an address for the machine learning model, set of parameters, or configuration based at least in part on the ID, and downloading the machine learning model, set of parameters, or configuration from the machine learning MR for the UE based on the address, where transmitting the machine learning model, set of parameters, or configuration to the UE may be based on the downloading.

[0037]

[0037] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may further include operations, features, means, or instructions, at a CU-XP entity included in the base station, for receiving from the UE an ID associated with a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions, determining an address for the second machine learning model, second set of parameters, or second configuration based at least in part on the ID, and uploading the second machine learning model, second set of parameters, or second configuration to the machine learning MR based on the address. [Brief description of the drawings]

[0038] [Figure 1]

[0038] FIG. 1 illustrates an example of a wireless communications system that supports configuring user equipment (UE) for machine learning in accordance with an aspect of the present disclosure. [Figure 2A]

[0039] FIG. 2A illustrates an example of a wireless communications system that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 2B]

[0040] FIG. 2B illustrates an example of a protocol stack that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Diagram 3]

[0041] FIG. 3 illustrates an example of a process flow that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 4] FIG. 4 illustrates an example of a process flow that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Diagram 5] FIG. 5 illustrates an example of a process flow that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 6]FIG. 6 illustrates an example of a process flow that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 7]

[0042] FIG. 7 illustrates a block diagram of a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 8] FIG. 8 illustrates a block diagram of a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 9]

[0043] FIG. 9 illustrates a block diagram of a communications manager that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 10]

[0044] FIG. 10 illustrates a diagram of a system including a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 11]

[0045] FIG. 11 illustrates a block diagram of a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 12] FIG. 12 illustrates a block diagram of a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 13]

[0046] FIG. 13 illustrates a block diagram of a communications manager that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 14]

[0047] FIG. 14 illustrates a diagram of a system including a device that supports configuring a UE for machine learning in accordance with an aspect of the disclosure. [Figure 15]

[0048] FIG. 15 illustrates a flowchart illustrating a method for supporting configuring a UE for machine learning according to an aspect of the disclosure. [Figure 16] FIG. 16 illustrates a flowchart illustrating a method for supporting configuring a UE for machine learning according to an aspect of the disclosure. [Figure 17]FIG. 17 illustrates a flowchart illustrating a method for supporting configuring a UE for machine learning according to an aspect of the disclosure. [Figure 18] FIG. 18 illustrates a flowchart illustrating a method for supporting configuring a UE for machine learning according to an aspect of the disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039]

[0049] A user equipment (UE) may utilize machine learning to perform different communication procedures. For example, the UE may utilize machine learning to perform cell reselection, channel state information (CSI) reporting, etc. To utilize machine learning, the UE may obtain knowledge of neural network functions, machine learning models, and corresponding parameters. Improved solutions to support configuration of machine learning in the UE (e.g., provision of neural network functions, machine learning models, and corresponding parameters to the UE) may be desired.

[0040]

[0050] Described herein are improved architectures and techniques by which a network may configure a UE to utilize machine learning. In some examples, a base station may include multiple network entities, such as a central unit user plane (CU-UP), a central unit control plane (CU-CP), and a distributed unit (DU). In some cases, the base station may additionally or alternatively include another central unit (e.g., a central unit machine learning plane (CU-XP)) configured to facilitate the exchange of messages related to machine learning. Furthermore, the base station may include or be in communication with a model repository (MR) configured to store multiple machine learning models and corresponding parameters. In some cases, the central unit may alternatively be referred to as a centralized unit.

[0041]

[0051] In some cases, the network may provide the UE with the neural network model, the machine learning model, the corresponding parameters, or any combination thereof, in response to a request from the UE to implement machine learning. In some examples, the UE may download the neural network function, the machine learning model, or the corresponding parameters over the user plane (e.g., directly from the MR). In other examples, the UE may download the neural network function, the machine learning model, or the corresponding parameters over the control plane (e.g., obtain the model from the CU-CP). Messages (e.g., request messages) exchanged between the UE and the network related to machine learning may be signaled over radio resource control (RRC) (e.g., on an existing or new radio bearer, using a new container in an RRC message, or any combination thereof). The techniques described herein may enable machine learning in the UE. Machine learning may enable the UE to perform tasks with little or no instructions from the network. Thus, enabling machine learning in the UE may result in less signaling overhead and reduced power consumption in the UE.

[0042]

[0052] Aspects of the present disclosure are first described in the context of a wireless communication system. Additional aspects are described in the context of protocol stacks and process flows. Aspects of the present disclosure are further illustrated by and described with reference to apparatus diagrams, system diagrams, and flowcharts relating to configuring a UE for machine learning.

[0043]

[0053] 1 illustrates an example of a wireless communication system 100 that supports configuring UEs for machine learning according to aspects of the disclosure. The wireless communication system 100 may include one or more base stations 105, one or more UEs 115, and a core network 130. In some examples, the wireless communication system 100 may be a Long Term Evolution (LTE) network, an LTE Advanced (LTE-A) network, an LTE-A Pro network, or a New Radio (NR) network. In some examples, the wireless communication system 100 may support enhanced broadband communications, ultra-reliable communications, low latency communications, communications with low-cost and low-complexity devices, or any combination thereof.

[0044]

[0054] The base stations 105 may be distributed throughout a geographic area and may be devices in different forms or with different capabilities to form a wireless communication system 100. The base stations 105 and the UEs 115 may communicate wirelessly via one or more communication links 125. Each base station 105 may provide a coverage area 110 over which the UEs 115 and the base stations 105 may establish one or more communication links 125. The coverage area 110 may be an example of a geographic area over which the base stations 105 and the UEs 115 may support communication of signals according to one or more radio access technologies.

[0045]

[0055] The UEs 115 may be distributed throughout the coverage area 110 of the wireless communication system 100, and each UE 115 may be fixed or mobile, or both at different times. The UEs 115 may be devices in different forms or with different capabilities. Some example UEs 115 are illustrated in FIG. 1. The UEs 115 described herein may be capable of communicating with various types of devices, such as other UEs 115, base stations 105, or network equipment (e.g., core network nodes, relay devices, integrated access and backhaul (IAB) nodes, or other network equipment), as shown in FIG. 1.

[0046]

[0056] The base stations 105 may communicate with the core network 130, or with each other, or both. For example, the base stations 105 may interface with the core network 130 through one or more backhaul links 120 (e.g., via an S1, N2, N3, or other interface). The base stations 105 may communicate with each other over the backhaul links 120 (e.g., via an X2, Xn, or other interface), either directly (e.g., directly between the base stations 105) or indirectly (e.g., via the core network 130), or both. In some examples, the backhaul links 120 may be or include one or more wireless links.

[0047]

[0057] The one or more base stations 105 described herein may include or be referred to by one skilled in the art as a base transceiver station, radio base station, access point, radio transceiver, Node B, eNode B (eNB), Next Generation Node B or Giga Node B (any of which may be referred to as a gNB), Home Node B, Home eNode B, or other suitable terminology.

[0048]

[0058] The UE 115 may include or be referred to as a mobile device, a wireless device, a remote device, a handheld device, or a subscriber device, or some other suitable terminology, where a "device" may also be referred to as a unit, a station, a terminal, or a client, among other examples. The UE 115 may also include or be referred to as a personal electronic device, such as a cellular phone, a personal digital assistant (PDA), a tablet computer, a laptop computer, or a personal computer. In some examples, the UE 115 may also include or be referred to as a wireless local loop (WLL) station, an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a machine type communication (MTC) device, among other examples, which may be implemented in various objects, such as appliances, or vehicles, meters, among other examples.

[0049]

[0059] The UEs 115 described herein may be capable of communicating with various types of devices, such as other UEs 115, which may at times act as relays, as shown in FIG. 1, as well as base stations 105 and network equipment, including macro eNBs or gNBs, small cell eNBs or gNBs, or relay base stations, among other examples.

[0050]

[0060] The UE 115 and the base station 105 may communicate wirelessly with each other over one or more communication links 125 on one or more carriers. The term “carrier” may refer to a set of radio frequency spectrum resources having a defined physical layer structure for supporting the communication link 125. For example, a carrier used for the communication link 125 may include a portion (e.g., a bandwidth part (BWP)) of a radio frequency spectrum band operated according to one or more physical layer channels for a given radio access technology (e.g., LTE, LTE-A, LTE-A Pro, NR). Each physical layer channel may carry acquisition signaling (e.g., synchronization signals, system information), control signaling to coordinate operation for the carrier, user data, or other signaling. The wireless communication system 100 may support communication with the UE 115 using carrier aggregation or multi-carrier operation. The UE 115 may be configured with multiple downlink component carriers and one or more uplink component carriers according to a carrier aggregation configuration. Carrier aggregation can be used with both frequency division duplex (FDD) and time division duplex (TDD) component carriers.

[0051]

[0061] A signal waveform transmitted on a carrier may be composed of multiple subcarriers (e.g., using a multicarrier modulation (MCM) technique such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread OFDM (DFT-S-OFDM)). In a system using MCM techniques, a resource element may include one symbol period (e.g., the duration of one modulation symbol) and one subcarrier, where the symbol period and the subcarrier spacing are inversely related. The number of bits carried by each resource element may depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both). Thus, the more resource elements the UE 115 receives and the higher the order of the modulation scheme, the higher the data rate for the UE 115 may be. Wireless communication resources may refer to a combination of radio frequency spectrum resources, time resources, and spatial resources (e.g., spatial layers or beams), and the use of multiple spatial layers may further increase the data rate or data integrity for communications with the UE 115.

[0052]

[0062] The time interval for the base station 105 or the UE 115 is, for example, T s =1 / (Δf max N f ) seconds, where Δf max may represent the maximum supported subcarrier spacing, and N f may represent the maximum supported Discrete Fourier Transform (DFT) size. The communication resource time intervals may be organized according to radio frames, each having a specified duration (e.g., 10 milliseconds (ms)). Each radio frame may be identified by a System Frame Number (SFN) (e.g., ranging from 0 to 1023).

[0053]

[0063] Each frame may include multiple consecutively numbered subframes or slots, and each subframe or slot may have the same duration. In some examples, a frame may be divided (e.g., in the time domain) into subframes, and each subframe may be further divided into several slots. Alternatively, each frame may include a variable number of slots, and the number of slots may depend on the subcarrier spacing. Each slot may include several symbol periods (e.g., depending on the length of a cyclic prefix added to the beginning of each symbol period). In some wireless communications systems 100, a slot may be further divided into multiple minislots containing one or more symbols. Excluding the cyclic prefix, each symbol period may include one or more (e.g., N f The duration of a symbol period may depend on the frequency band of operation or the subcarrier spacing.

[0054]

[0064] A subframe, slot, minislot, or symbol may be the smallest scheduling unit (e.g., in the time domain) of the wireless communications system 100 and may be referred to as a transmission time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol periods in a TTI) may be variable. Additionally or alternatively, the smallest scheduling unit of the wireless communications system 100 may be dynamically selected (e.g., in a burst of shortened TTIs (sTTIs)).

[0055]

[0065] The physical channels may be multiplexed on the carriers according to various techniques. The physical control channels and the physical data channels may be multiplexed on the downlink carriers using, for example, one or more of a time division multiplexing (TDM), a frequency division multiplexing (FDM), or a hybrid TDM-FDM technique. A control region (e.g., a control resource set (CORESET)) for the physical control channel may be defined by a number of symbol periods and may extend across the system bandwidth of the carrier or a subset of the system bandwidth. One or more control regions (e.g., CORESET) may be configured for a set of UEs 115. For example, one or more of the UEs 115 may monitor or search the control region for control information according to one or more search space sets, and each search space set may include one or more control channel candidates at one or more aggregation levels arranged in a cascaded manner. The aggregation level for the control channel candidates may refer to the number of control channel resources (e.g., control channel elements (CCEs)) associated with the coded information for a control information format having a given payload size. The search space sets may include a common search space set configured for sending control information to multiple UEs 115 and a UE-specific search space set for sending control information to a particular UE 115.

[0056]

[0066] In some examples, the base stations 105 may be mobile and thus provide communication coverage for moving geographic coverage areas 110. In some examples, different geographic coverage areas 110 associated with different technologies may overlap, but the different geographic coverage areas 110 may be supported by the same base station 105. In other examples, overlapping geographic coverage areas 110 associated with different technologies may be supported by different base stations 105. The wireless communication system 100 may include a heterogeneous network, for example, where different types of base stations 105 provide coverage for various geographic coverage areas 110 using the same or different radio access technologies.

[0057]

[0067] Some UEs 115 may be configured to use operating modes that reduce power consumption, such as half-duplex communication (e.g., a mode that supports one-way communication via transmission or reception, but not simultaneous transmission and reception). In some examples, half-duplex communication may be performed at a reduced peak rate. Other power saving techniques for the UE 115 include entering a power-saving deep sleep mode when not engaged in active communication, operating on a limited bandwidth (e.g., pursuant to narrowband communication), or a combination of these techniques. For example, some UEs 115 may be configured for operation using a narrowband protocol type associated with a defined portion or range (e.g., a set of subcarriers or resource blocks (RBs)) within a carrier, within a guard band of the carrier, or outside the carrier.

[0058]

[0068] The wireless communication system 100 may be configured to support ultra-reliable or low latency communications, or various combinations thereof. For example, the wireless communication system 100 may be configured to support ultra-reliable low latency communications (URLLC). The UEs 115 may be designed to support ultra-reliable, low latency, or critical functionality. Ultra-reliable communications may include private or group communications and may be supported by one or more services, such as push-to-talk, video, or data. Support for ultra-reliable, low latency functionality may include service prioritization, and such services may be used for public safety or general commercial applications. The terms ultra-reliable, low latency, and ultra-reliable low latency may be used interchangeably herein.

[0059]

[0069] In some examples, the UE 115 may also be able to communicate directly with other UEs 115 over a device-to-device (D2D) communication link 135 (e.g., using a peer-to-peer (P2P) or D2D protocol). One or more UEs 115 utilizing D2D communication may be within the geographic coverage area 110 of the base station 105. Other UEs 115 in such a group may be outside the geographic coverage area 110 of the base station 105 or may not be able to receive transmissions from the base station 105 for other reasons. In some examples, a group of UEs 115 communicating via D2D communication may utilize a one-to-many (1:M) system in which each UE 115 transmits to every other UE 115 in the group. In some examples, the base station 105 facilitates scheduling of resources for D2D communication. In other cases, D2D communication occurs between UEs 115 without the involvement of the base station 105.

[0060]

[0070] The core network 130 may provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network 130 may be an evolved packet core (EPC) or 5G core (5GC), which may include at least one control plane entity (e.g., a mobility management entity (MME), an access and mobility management function (AMF)) that manages access and mobility, and at least one user plane entity (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function (UPF)) that routes packets or interconnects to external networks. The control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management for the UEs 115 served by the base stations 105 associated with the core network 130. User IP packets may be forwarded through the user plane entity, which may provide IP address allocation as well as other functions. The user plane entities may be connected to IP services 150 for one or more network operators. The IP services 150 may include access to the Internet, an intranet(s), an IP Multimedia Subsystem (IMS), or packet-switched streaming services.

[0061]

[0071] Some of the network devices, such as the base stations 105, may include subcomponents, such as an access network entity 140, which may be an example of an access node controller (ANC). Each access network entity 140 may communicate with the UE 115 through one or more other access network transmitting entities 145, which may be referred to as a radio head, a smart radio head, or a transmit / receive point (TRP). Each access network transmitting entity 145 may include one or more antenna panels. In some configurations, various functions of each access network entity 140 or base station 105 may be distributed across various network devices (e.g., radio heads and ANCs) or integrated into a single network device (e.g., the base station 105).

[0062]

[0072] The wireless communication system 100 may operate using one or more frequency bands, for example, in the range of 300 megahertz (MHz) to 300 gigahertz (GHz). The 300 MHz to 3 GHz region is commonly known as the ultra-high frequency (UHF) region or decimeter band because the wavelengths span lengths of approximately 1 decimeter to 1 meter. Although UHF waves may be blocked or redirected by buildings and environmental features, the waves may penetrate structures sufficiently for a macrocell to serve UEs 115 located indoors. Transmission of UHF waves may be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers) compared to transmissions using smaller frequencies and longer waves in the short wave (HF) or very high frequency (VHF) portions of the spectrum below 300 MHz.

[0063]

[0073] The wireless communication system 100 may utilize both licensed and unlicensed radio frequency spectrum bands. For example, the wireless communication system 100 may use Licensed Assisted Access (LAA), LTE Unlicensed (LTE-U) radio access technology, or NR technology in an unlicensed band, such as the 5 GHz Industrial, Scientific, and Medical (ISM) band. When operating in an unlicensed radio frequency spectrum band, devices such as the base station 105 and the UE 115 may use carrier sensing for collision detection and avoidance. In some examples, operation in an unlicensed band may be based on a carrier aggregation configuration in conjunction with a component carrier operating in a licensed band (e.g., LAA). Operation in an unlicensed spectrum may include downlink transmissions, uplink transmissions, P2P transmissions, or D2D transmissions, among other examples.

[0064]

[0074] The base station 105 or UE 115 may be equipped with multiple antennas, which may be used to employ techniques such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communications, or beamforming. The antennas of the base station 105 or UE 115 may be located in one or more antenna arrays or antenna panels that may support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays may be co-located in an antenna assembly, such as an antenna tower. In some examples, antennas or antenna arrays associated with the base station 105 may be located in diverse geographic locations. The base station 105 may have an antenna array with several rows and columns of antenna ports that the base station 105 may use to support beamforming of communications with the UE 115. Similarly, the UE 115 may have one or more antenna arrays that may support various MIMO or beamforming operations. Additionally or alternatively, the antenna panels may support radio frequency beamforming for signals transmitted through the antenna ports.

[0065]

[0075] A base station 105 or a UE 115 may use MIMO communication to exploit multipath signal propagation and increase spectral efficiency by transmitting or receiving multiple signals via different spatial layers. Such techniques may be referred to as spatial multiplexing. The multiple signals may be transmitted by a transmitting device, for example, via different antennas or different combinations of antennas. Similarly, the multiple signals may be received by a receiving device via different antennas or different combinations of antennas. Each of the multiple signals may be referred to as a separate spatial stream and may carry bits associated with the same data stream (e.g., the same codeword) or different data streams (e.g., different codewords). The different spatial layers may be associated with different antenna ports used for channel measurement and reporting. MIMO techniques include single-user MIMO (SU-MIMO), in which multiple spatial layers are transmitted to the same receiving device, and multi-user MIMO (MU-MIMO), in which multiple spatial layers are transmitted to multiple devices.

[0066]

[0076] Beamforming, which may also be referred to as spatial filtering, directional transmission, or directional reception, is a signal processing technique that may be used in a transmitting or receiving device (e.g., base station 105, UE 115) to form or steer an antenna beam (e.g., transmit beam, receive beam) along a spatial path between the transmitting device and the receiving device. Beamforming may be achieved by combining signals communicated through antenna elements of an antenna array such that some signals propagating in a particular orientation relative to the antenna array experience constructive interference, while other signals experience destructive interference. Adjustment of signals communicated through antenna elements may include the transmitting or receiving device applying an amplitude offset, a phase offset, or both to signals carried through the antenna elements associated with the device. The adjustments associated with each of the antenna elements may be defined by a beamforming weight set associated with a particular orientation (e.g., relative to the antenna array of the transmitting or receiving device, or to some other orientation).

[0067]

[0077] The wireless communication system 100 may be a packet-based network that operates according to a hierarchical protocol stack. In the user plane, communication at the bearer or Packet Data Convergence Protocol (PDCP) layer may be IP-based. The Radio Link Control (RLC) layer may perform packet segmentation and reassembly to communicate on logical channels. The Medium Access Control (MAC) layer may perform priority handling and multiplexing of logical channels into transport channels. The MAC layer may also use error detection techniques, error correction techniques, or both to support retransmissions at the MAC layer to improve link efficiency. In the control plane, the RRC protocol layer may provide establishment, configuration, and maintenance of an RRC connection between the UE 115 and the base station 105 or core network 130, which supports radio bearers for user plane data. In the physical layer, the transport channels may be mapped to physical channels.

[0068]

[0078] In some examples, the network may configure the UE 115 for machine learning. The UE 115 may send capability information to the network (e.g., base station 105). The capability information may include one or more of a list of potential neural network functions, a list of potential machine learning models, or an indication of whether the UE 115 may request machine learning. Based on the capability information, the network may select a set of neural network functions, a set of machine learning models, and a set of corresponding parameters and indicate them to the UE 115. In some examples, the UE 115 may send a message to the network requesting to implement machine learning (e.g., based on some trigger). The message may include an indication of the neural network functions, the neural network models, and the corresponding parameter sets. In response to the request message, the network may configure the machine learning models and corresponding parameters in the UE 115. Once the UE 115 obtains the machine learning models and corresponding parameter sets, the network may activate machine learning in the UE 115, and the UE 115 utilizes machine learning to perform one or more tasks. Techniques described herein may enable machine learning in the UE 115. Machine learning may enable the UE 115 to perform tasks with little or no instructions from the network. Thus, enabling machine learning in the UE may result in less signaling overhead and reduced power consumption in the UE 115.

[0069]

[0079] 2A illustrates an example of a wireless communication system 201 that supports configuring a UE for machine learning in accordance with aspects of the disclosure. In some examples, the wireless communication system 201 may implement or be implemented by the wireless communication system 100. For example, the wireless communication system 201 may include a base station 105-a and a UE 115-a, which may be examples of the base station 105 and the UE 115 as described with reference to FIG.

[0070]

[0080] 2B illustrates an example of a protocol stack 202 that supports configuring a UE for machine learning as described herein. In some examples, the protocol stack 202 may implement or be implemented by the wireless communication system 100. For example, the protocol stack 202 may be implemented by the base station 105 and the UE 115 as described with reference to FIG.

[0071]

[0081] In some examples, the wireless communication system 201 may support machine learning or artificial intelligence. Using machine learning, a device (e.g., the base station 105-a or the UE 115-a) may perform a task without being explicitly programmed to do so. To perform machine learning, the device may obtain neural network functions and neural network models. A neural network function may be defined as a function supported by one or more neural network models and may be specific to the task being performed. The inputs and outputs of each neural network function may be set (e.g., standardized), and each neural network function may be identified by a neural network function identifier (ID). A neural network model may be defined as a model structure and a parameter set. The model structure may be identified by a model ID, and each model ID may be associated with a neural network function. The model ID may also specify a set of parameters corresponding to the neural network model. The set of parameters may include weights and other configuration parameters of the neural network model. In some examples, the UE 115-a may utilize machine learning for cell reselection, beam management, etc. However, methods for configuring the UE 115-a for machine learning may be lacking or inefficient.

[0072]

[0082] In some examples, the base station 105 may include different network entities. For example, the base station 105-a may include at least the CU-UP 205, the CU-CP 210, the DU 220, and the radio unit (RU) 225. The CU-CP 210 may host the control plane part of the Packet Data Convergence Protocol (PDCP), and the CU-UP 205 may host the user plane part of the PDCP. Meanwhile, the DU 220 may support lower layer signaling (e.g., Medium Access Control (MAC) protocol or Radio Link Control (RLC) protocol), and the RU 225 may support physical layer signaling as well as digital beamforming functions. The CU-UP 205 may be connected to the CU-CP 210 via an E1 interface. Furthermore, the CU-UP 205 may be connected to the DU 220 via an F1-U interface, and the CU-CP 210 may be connected to the DU 220 via an F1-C interface. To support machine learning in the UE 115-a as described herein, the base station 105-a may also include a CU-XP 215. The CU-XP 215 may host a Machine Learning Control (MLC) protocol, as shown in FIG. 2B. The MLC protocol may define control plane messaging for managing machine learning or artificial intelligence in the network. In some examples, the CU-XP 215 may be connected to the CU-UP via an E3 interface. Additionally, the base station 105-a may be in communication with a UE model repository (UE-MR) 230. The UE-MR 230 may be defined as a central location where neural network models are stored (e.g., cloud storage, online storage, etc.).

[0073]

[0083] To implement machine learning in the UE 115-a, MLC messages may be exchanged between the CU-XP 215 and the UE 115-a. The MLC messages may be defined as control messages that facilitate machine learning or artificial intelligence. In some examples, the MLC messages may be exchanged between the UE 115-a and the CU-XP 215 via RRC signaling. For example, the UE 115-a and the CU-CP 210 may exchange RRC signaling including a container carrying an MLC message directed to the CU-XP 215. The CU-CP 210 may then forward the MLC message to the CU-XP 215. The RRC container may be decoded in the CU-XP 215 and sent to the MLC layer. In some examples, the MLC messages may be carried on a signaling radio bearer (SRB) 2 in the RRC. In another example, a new SRB may be defined for machine learning (e.g., SRB X) and the MLC messages may be carried on the newly defined SRB in the RRC. In some examples, the MLC message may be piggybacked by an existing RRC message, for example, the MLC message may be included as a container in an RRC reconfiguration message, an RRC reconfiguration complete message, or an RRC setup / resumption complete message.

[0074]

[0084] FIG. 3 illustrates an example of a process flow 300 supporting configuring a UE for machine learning according to aspects of the disclosure. In some examples, the process flow 300 may be implemented by aspects of the wireless communication system 100 and the wireless communication system 201. For example, the process flow 300 may be performed by the UE 115-b, the CU-CP 305, the CU-XP 310, and the UE-MR 320, which may be examples of the UE 115, the CU-CP 210, the CU-XP 215, and the UE-MR 230 as described with reference to FIG. 2. The process flow 300 may support a network configuring the UE 115-b for machine learning. The following alternative examples may be implemented, where some steps are performed in a different order than described or are not performed at all. In some cases, the steps may include additional features not described below or further steps may be added.

[0075]

[0085] At 325, the UE 115-b may exchange capability information related to machine learning with the network. In some examples, the CU-CP 305 may send a message to the UE 115-b inquiring about the capability information, and the UE 115-b may send the capability information to the CU-CP 305 based on receiving the message inquiring about the capability information. Upon receiving the capability information from the UE 115-b, the CU-CP 305 may forward the capability information to the CU-XP 310. In some examples, the capability information may include a list of neural network functions supported by the UE 115-b, a list of neural network models supported by the UE 115-b (e.g., a list of model IDs), an indication of whether the UE 115-b may request to be configured for machine learning, etc. Based on the capability information, the CU-CP 305 or the CU-XP 310 may determine one or more neural network functions from the list of neural network functions, one or more model IDs from the list of model IDs, and the ID of a corresponding set of parameters. An indication of the neural network function(s), model ID(s), and corresponding parameter set ID(s) may be sent in an MLC container from CU-CP 305 to UE 115-b as part of an RRC message (e.g., an RRC reconfiguration message). UE 115-b may send an RRC message (e.g., an RRC reconfiguration complete message) to CU-CP 305 to acknowledge receipt of the RRC message and the MLC container.

[0076]

[0086] In some examples, the UE 115-b may request to implement machine learning (perform machine learning for a particular task). In such examples, the RRC message from the CU-CP 305 to the UE 115-b may also include a prohibit timer for the machine learning request. The prohibit timer may limit the number of times the UE 115-b may send the machine learning request to the network. Additionally or alternatively, the RRC message may include an indication of a blacklist of neural network functions, model IDs, and corresponding parameter set IDs, and a whitelist of neural network functions, model IDs, and corresponding parameter set IDs. The UE 115-b may be able to request to implement models in the whitelist, but may not be able to request to implement models in the blacklist. Each model ID of the one or more model IDs indicated to the UE 115-b at 325 may be associated with a condition (or applicable scope). The condition may be a location (e.g., cell, cell list, Tracking Area Identity (TAI), Radio Access Network Notification Area (RNA), Multi-broadcast Single Frequency Network (MBSFN) area, or geographic area), network slice, deep neural network (DNN) type, public land mobile network (PLMN) list (e.g., a list of public network integrated non-public network (PNI-NPN) IDs or standalone non-public network (SNPN) IDs), UE type, RRC state, type of service (e.g., multi-broadcast service (MBS) or sidelink), or configuration (e.g., MIMO, dual connectivity / carrier aggregation (DC / CA), or mmW).

[0077]

[0087] At 330, the UE 115-b may potentially determine whether a condition associated with a model ID of one or more model IDs is satisfied. For example, the UE 115-b may determine that the UE 115-b has left a cell associated with a first model ID and entered a cell associated with a second model ID. When the UE 115-b determines that a condition associated with a model ID is satisfied, the UE 115-b may send a machine learning request message to the CU-CP 305 at 335. The machine learning request message may include one or more of a model ID (e.g., a second model ID) whose condition is satisfied, a neural network function of the one or more neural network functions, or an indication of the condition that is satisfied. In some examples, the UE 115-b may include a machine learning request in a UE assistance information message. More specifically, a machine learning assistance information element including information of the machine learning request message may be added to the UE assistance information. In some examples, if the model whose condition is satisfied is included in the blacklist, the UE 115-b may not send a machine learning request.

[0078]

[0088] Upon receiving the machine learning request from the UE 115-b, the network (e.g., the CU-XP 310 or the CU-CP 305) may select a neural network function (e.g., from one or more neural network functions indicated in the machine learning request message received at 335) and a machine learning model (e.g., selecting a machine learning model corresponding to a model ID indicated in the machine learning request message received at 335) and configure the UE 115-b with the machine learning model and a corresponding set of parameters at 340. In some examples, configuring the UE 115-b with the neural network model may include downloading the neural network model from the UE-MR 320. Different aspects of downloading and uploading the neural network model are described in more detail in Figures 4-6.

[0079]

[0089] In addition to the UE 115-b, the other network node 315 (e.g., a distributed unit, CU-UP, or RIC) may be configured with the selected machine learning model and the corresponding set of parameters. To configure the other network node 315, the CU-XP 310 may send a model setup request message to the other network node 315, where the model setup request message may include the model ID and the corresponding parameter set ID of the selected neural network model. The other network node 315 may send the model ID and the parameter set ID to the MDAC via a model query request message, and the MDAC may send a model query response to the other network node including an address (e.g., a web address or URL) corresponding to the model ID and an address corresponding to the parameter ID. Upon receiving the model query response message, the other network node 315 may download the neural network model associated with the model ID and the parameter set associated with the parameter set ID from the UE-MR 320 by their respective web addresses. The other network nodes 315 may then confirm the configuration of the neural network model and corresponding parameter set via a Model Setup Response message to the CU-XP 310, and the CU-XP 310 may forward confirmation of the neural network model configuration to the CU-CP 305 via a Neural Network Response message.

[0080]

[0090] At 345, the network may activate the neural network model. To activate the neural network model in the UE 115-b, the UE 115-b may send a model activation request message to the other network node 315 via the CU-CP 305 to request activation of machine learning, and the other network node may send a model activation response message to the UE 115-b via MAC-CE or RRC signaling to activate the machine learning in the UE 115-b. To activate the neural network model in the network, the CU-CP 305 may send a model activation message to the CU-XP 310, and the CU-XP 310 may send a model activation message to the other network node 315 to activate the machine learning in the other network node 315.

[0081]

[0091] FIG. 4 illustrates an example of a process flow 400 that supports configuring a UE for machine learning according to aspects of the disclosure. In some examples, the process flow 400 may be implemented by aspects of the wireless communication system 100, the wireless communication system 201, and the process flow 300. For example, the process flow 400 may be performed by the UE 115-c and the UE-MR 405, which may be examples of the UE 115 and the UE-MR 230 as described with reference to FIG. 2. The process flow 400 may support uploading and downloading of neural network models and parameter sets in the UE 115-c. The following alternative examples may be implemented, where some steps are performed in a different order than described or not performed at all. In some cases, the steps may include additional features not described below or further steps may be added.

[0082]

[0092] As described with reference to FIG. 3, the network (e.g., CU-CP or CU-XP) may select a neural network function, a neural network model, and a corresponding set of parameters (e.g., based on the capability of UE 115-c or based on a request message from UE 115-c) and indicate the neural network function, the neural network model, and the corresponding set of parameters to UE 115-c so that UE 115-c may perform machine learning. For example, the network may send a message (e.g., an RRC reconfiguration message) including a neural network function ID, a model ID, and a corresponding parameter set ID. UE 115-c may then perform the following procedure to download the indicated neural network model and the corresponding set of parameters.

[0083]

[0093] At 410, the UE 115-c may construct an address (e.g., a web address or URL) of the neural network model and an address (e.g., a web address or URL) of the set of parameters. In some examples, the UE 115-c may construct the address based on a model ID and a parameter set ID. The model ID and parameter set ID may serve as inputs for the predefined rules.

[0084]

[0094] At 415, the UE 115-c may download the model and the corresponding set of parameters from the UE-MR 405 by the address. The UE 115-c may send the address of the neural network model and, in some examples, the address of the set of parameters to the UE-MR 405. For example, the UE 115-c may send an HTTP GET message including the address of the neural network model and the address of the set of parameters. Upon receiving the address, the UE-MR 405 may send the neural network model and the corresponding set of parameters (e.g., in a 200 GET message) to the UE 115-c. That is, the UE 115-c may download the neural network model and the set of parameters from the UE-MR 405 by their respective addresses. In some examples, the UE 115-c may cache frequently used neural network models and parameter sets. In some examples, version tags and timers may be used to evaluate and protect the freshness of the cached neural network models and parameter sets. If the neural network model and parameter set are cached locally, the UE 115-c may not download the neural network model and parameter set from the UE-MR 405, but may instead obtain the neural network model and parameter set from the cached data.

[0085]

[0095] In some examples, the UE 115-c may not obtain the neural network model and set of parameters from the UE-MR 405, but may obtain the neural network model and set of parameters elsewhere (e.g., using a model training configuration). In such examples, the UE 115-c may upload the neural network model and set of parameters to the UE-MR 405 so that the UE-MR 405 may store the neural network model and set of parameters for future use or so that other devices may access the neural network model and set of parameters.

[0086]

[0096] In some examples, the UE 115-c may construct an address (e.g., a web address or URL) for the set of parameters at 420. In another example, the UE 115-c may construct an address (e.g., a web address or URL) for both the neural network model and the set of parameters. The UE 115-c may know the address from a training configuration, or the UE 115-c may determine the address using a predefined rule in which a neural network model ID associated with the neural network model and a parameter set ID associated with the set of parameters are used as inputs.

[0087]

[0097] At 425, the UE 115-c may upload the parameter set, and in some examples, the neural network model, to the UE-MR 405 via the address. In some examples, the UE 115-c may upload the set of parameters or the neural network model to the UE-MR 405 by sending an HTTP PUT message including the neural network model and the set of parameters to the UE-MR 405. To confirm receipt of the set of parameters or the neural network model, the UE-MR may send a confirmation message (e.g., a 200 OK message) to the UE 115-c.

[0088]

[0098] FIG. 5 illustrates an example of a process flow 500 that supports configuring a UE for machine learning according to aspects of the disclosure. In some examples, the process flow 500 may be implemented by aspects of the wireless communication system 100, the wireless communication system 201, the process flow 300, or the process flow 400. For example, the process flow 500 may be performed by the UE 115-d, the CU-CP 505, the CU-XP 510, and the UE-MR 515, which may be examples of the UE 115, the CU-CP 210, the CU-XP 215, and the UE-MR 230 as described with reference to FIG. 2. The process flow 500 may support uploading and downloading of neural network models and parameter sets in the UE 115-d. The following alternative examples may be implemented, where some steps are performed in a different order than described or not performed at all. In some cases, the steps may include additional features not described below or further steps may be added.

[0089]

[0099] As described with reference to FIG. 3, the network (e.g., CU-CP505 or CU-XP510) may select a neural network function, a neural network model, and a corresponding set of parameters (e.g., based on the capability of UE115-d or based on a request message from UE115-d) and indicate the neural network function, the neural network model, and the corresponding set of parameters to UE115-d so that UE115-d may perform machine learning. For example, the network may send a message (e.g., an RRC reconfiguration message) including a neural network function ID, a model ID, and a corresponding parameter set ID. UE115-d may then perform the following procedure to obtain the indicated neural network model and the corresponding set of parameters.

[0090]

[0100] At 520, the UE 115-d may obtain a neural network model and a corresponding set of parameters from the network. First, the UE 115-d may send a model download request message to the CU-CP 505 at 525. The model download request message may include a model ID corresponding to the neural network model and a parameter set ID corresponding to the set of parameters. In some examples, the UE 115-d may send the download request message to the CU-CP 505 via RRC signaling over SRB2.

[0091]

[0101] At 530, the CU-CP 505 may forward the model download request to the CU-XP 510. The CU-XP 510 may then retrieve the neural network model and set of parameters from the UE-MR 515 at 535. Similar to how the UE 115 downloads the neural network model and set of parameters in FIG. 4, the CU-XP 510 may use the model ID and parameter set ID to construct an address for the neural network model and set of parameters and use this address to download the neural network model and set of parameters from the UE-MR 515. Alternatively, the CU-XP 510 may obtain the address from another network entity (e.g., MDAC).

[0092]

[0102] At 540, the CU-XP 510 may send a model download response message to the CU-CP 505. The model download model response message may include the neural network model and a set of parameters.

[0093]

[0103] At 545, the CU-CP 505 may forward the model download response message to the UE 115-d. In some examples, the CU-CP 505 may send the model download response message to the UE 115-d via RRC signaling on a new SRB (e.g., SRB X). If the size of the neural network model and the set of parameters is above a threshold, RRC segmentation may be used to send the neural network model and the set of parameters to the UE 115-d.

[0094]

[0104] In some examples, the UE 115-d may not obtain the neural network model and set of parameters from the UE-MR 515, but may obtain the neural network model and set of parameters elsewhere (e.g., using a model training configuration). In such examples, the UE 115-d may upload the neural network model and set of parameters to the UE-MR 515 so that the UE-MR 515 may store the neural network model and set of parameters for future use or so that other devices may access the neural network model and set of parameters.

[0095]

[0105] At 550, the UE 115-d may upload the neural network model and the set of parameters to the UE-MR 515 via one or more network nodes. For example, the UE 115-d may send a model upload request message to the CU-CP 505 at 555. The model upload request message may include the neural network model and the set of parameters. In some examples, the UE 115-d may send the model upload request message via RRC signaling over SRB2.

[0096]

[0106] At 560, the CU-CP 505 may forward the model upload request to the CU-XP 510. The CU-XP 510 may then upload the neural network model and set of parameters to the UE-MR 515 at 565. Similar to how the UE 115 uploads the neural network model and set of parameter sets in FIG. 4, the CU-XP 510 may use the model ID and parameter set ID to construct an address for the neural network model and set of parameters and use this address to upload the neural network model and set of parameters to the UE-MR 515. Alternatively, the CU-XP 510 may obtain the address from another network entity (e.g., MDAC).

[0097]

[0107] At 570, the CU-XP 510 may send a model upload response message to the CU-CP 505. The model download response message may serve as confirmation that the neural network model and set of parameters have been uploaded to the UE-MR 515.

[0098]

[0108] At 575, the CU-CP 505 may forward the model download response message to the UE 115-d. In some examples, the CU-CP 505 may send the model download response message to the UE 115-d via RRC signaling on a new SRB (e.g., SRB X).

[0099]

[0109] FIG. 6 illustrates an example of a process flow 600 supporting configuring a UE for machine learning according to aspects of the disclosure. In some examples, the process flow 600 may be implemented by aspects of the wireless communication system 100, the wireless communication system 201, the process flow 300, the process flow 400, or the process flow 500. For example, the process flow 500 may be performed by the UE 115-e, the CU-XP 605, and the UE-MR 615, which may be examples of the UE 115, the CU-XP 215, and the UE-MR 230 as described with reference to FIG. 2. The process flow 600 may support uploading and downloading of neural network models and parameter sets in the UE 115-e. The following alternative examples may be implemented, where some steps are performed in a different order than described or not performed at all. In some cases, the steps may include additional features not described below or further steps may be added.

[0100]

[0110] As described with reference to FIG. 3, the network (e.g., CU-CP or CU-XP 605) may select a neural network function, a neural network model, and a corresponding set of parameters (e.g., based on the capabilities of UE 115-e or based on a request message from UE 115-e) and indicate the neural network function, the neural network model, and the corresponding set of parameters to UE 115-e so that UE 115-e may perform machine learning. For example, the network may send a message (e.g., an RRC reconfiguration message) including a neural network function ID, a model ID, and a corresponding parameter set ID. UE 115-e may then perform the following procedure to obtain the indicated neural network model and the corresponding set of parameters.

[0101]

[0111] The UE 115-e may obtain a neural network model and a corresponding set of parameters from the network, at 620. First, the UE 115-e may send a model query request message to the CU-XP 605, at 625. The model query request message may include a model ID corresponding to the neural network model and a parameter set ID corresponding to the set of parameters.

[0102]

[0112] At 630, the CU-XP 605 may forward the model query request message to the MDAC 610. In response to the model query request message, the MDAC 610 may send a model query response message to the CU-XP 605 at 635. The model query request message may include an address for the neural network model and an address for the set of parameters.

[0103]

[0113] At 640, the CU-XP 605 may forward a model query response message to the UE 115-e. The UE 115-e may then send a message (e.g., an HTTP GET message) including an address for the neural network model and an address for the set of parameters to the UE-MR at 645. The UE-MR 615 may receive the message and send the neural network model and the set of parameters to the UE 115-e at 650. That is, the UE 115-e may download the neural network model and the set of parameters from the UE-MR 615 by the address.

[0104]

[0114] In some examples, the UE 115-e may not obtain the neural network model and set of parameters from the UE-MR 615, but may obtain the neural network model and set of parameters elsewhere (e.g., using a model training configuration). In such examples, the UE 115-e may upload the neural network model and set of parameters to the UE-MR 615 so that the UE-MR 615 may store the neural network model and set of parameters for future use or so that other devices may access the neural network model and set of parameters.

[0105]

[0115] At 655, the UE 115-e may upload the neural network model and the set of parameters to the UE-MR 615 via one or more network nodes. For example, the UE 115-e may send a model query request message to the CU-XP 605 at 660. The model query request message may include a model ID associated with the neural network model and a parameter set ID associated with the set of parameters.

[0106]

[0116] At 665, the CU-XP 510 may forward the model query request to the MDAC 610. In response to the model query request message, the MDAC 610 may send a model query response message to the CU-XP 605 at 670. The model query response message may include an address for the neural network model and an address for the set of parameters.

[0107]

[0117] At 675, the CU-XP 605 may forward the model query response message to the UE 115-e. The UE 115-e may then send a message (e.g., an HTTP PUT message) including the neural network model and the set of parameters to the UE-MR 615 at 680. The UE-MR 615 may receive this message and, at 685, a message confirming the upload of the neural network model and the set of parameters to the UE-MR 615. That is, the UE 115-e may upload the neural network model and the set of parameters to the UE-MR 615 by address.

[0108]

[0118] 7 illustrates a block diagram 700 of a device 705 that supports configuring a UE for machine learning according to an aspect of the disclosure. The device 705 may be an example of an aspect of a UE 115 as described herein. The device 705 may include a receiver 710, a transmitter 715, and a communications manager 720. The device 705 may also include a processor. Each of these components may be in communication with one another (e.g., via one or more buses).

[0109]

[0119] The receiver 710 may provide a means for receiving information, such as packets, user data, control information, or any combination thereof, associated with various information channels (e.g., information channels related to configuring the UE for machine learning, data channels, control channels). The information may be passed to other components of the device 705. The receiver 710 may utilize a single antenna or a set of multiple antennas.

[0110]

[0120] The transmitter 715 may provide a means for transmitting signals generated by other components of the device 705. For example, the transmitter 715 may transmit information such as packets, user data, control information, or any combination thereof associated with various information channels (e.g., information channels, data channels, control channels related to configuring the UE for machine learning). In some examples, the transmitter 715 may be co-located with the receiver 710 in a transceiver module. The transmitter 715 may utilize a single antenna or a set of multiple antennas.

[0111]

[0121] The communications manager 720, the receiver 710, the transmitter 715, or various combinations or components thereof may be examples of means for performing various aspects of configuring a UE for machine learning, as described herein. For example, the communications manager 720, the receiver 710, the transmitter 715, or various combinations or components thereof may support a method for performing one or more of the functions described herein.

[0112]

[0122] In some examples, the communications manager 720, the receiver 710, the transmitter 715, or various combinations or components thereof may be implemented in hardware (e.g., in a communications management circuit). The hardware may include a processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof configured as or otherwise supporting a means for performing the functions described in this disclosure. In some examples, the processor and a memory coupled to the processor may be configured to perform one or more of the functions described herein (e.g., by executing, by the processor, instructions stored in the memory).

[0113]

[0123] Additionally or alternatively, in some examples, the communications manager 720, the receiver 710, the transmitter 715, or various combinations or components thereof may be implemented in code executed by a processor (e.g., as communications management software or firmware). If implemented in code executed by a processor, the functions of the communications manager 720, the receiver 710, the transmitter 715, or various combinations or components thereof may be performed by a general purpose processor, a DSP, a central processing unit (CPU), an ASIC, an FPGA, or any combination of these or other programmable logic devices (e.g., configured as or otherwise supporting a means for performing the functions described in this disclosure).

[0114]

[0124] In some examples, the communications manager 720 may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the receiver 710, the transmitter 715, or both. For example, the communications manager 720 may receive information from the receiver 710, send information to the transmitter 715, or may be integrated in combination with the receiver 710, the transmitter 715, or both to receive information, transmit information, or perform various other operations as described herein.

[0115]

[0125] The communications manager 720 may support wireless communications in a UE according to examples as disclosed herein. For example, the communications manager 720 may be configured as or otherwise support a means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The communications manager 720 may be configured as or otherwise support a means for receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0116]

[0126] By including or configuring the communications manager 720 according to examples as described herein, the device 705 (e.g., a processor controlling or otherwise coupled to the receiver 710, the transmitter 715, the communications manager 720, or a combination thereof) may support techniques for reduced power consumption. Methods as described herein may enable the device 705 to utilize machine learning for some communication procedures. Machine learning may enable the device 705 to perform communications without explicit programming, which may reduce power consumption in the UE.

[0117]

[0127] 8 illustrates a block diagram 800 of a device 805 that supports configuring a UE for machine learning according to an aspect of the disclosure. The device 805 may be an example of an aspect of the device 705 or the UE 115 as described herein. The device 805 may include a receiver 810, a transmitter 815, and a communications manager 820. The device 805 may also include a processor. Each of these components may be in communication with one another (e.g., via one or more buses).

[0118]

[0128] The receiver 810 may provide a means for receiving information, such as packets, user data, control information, or any combination thereof, associated with various information channels (e.g., information channels related to configuring the UE for machine learning, data channels, control channels). The information may be passed to other components of the device 805. The receiver 810 may utilize a single antenna or a set of multiple antennas.

[0119]

[0129] The transmitter 815 may provide a means for transmitting signals generated by other components of the device 805. For example, the transmitter 815 may transmit information such as packets, user data, control information, or any combination thereof associated with various information channels (e.g., information channels, data channels, control channels related to configuring the UE for machine learning). In some examples, the transmitter 815 may be co-located with the receiver 810 in a transceiver module. The transmitter 815 may utilize a single antenna or a set of multiple antennas.

[0120]

[0130] The device 805, or various components thereof, may be examples of means for performing various aspects of configuring a UE for machine learning, as described herein. For example, the communications manager 820 may include a UE machine learning manager 830, a UE activation component 835, or any combination thereof. The communications manager 820 may be an example of an aspect of the communications manager 720, as described herein. In some examples, the communications manager 820, or various components thereof, may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the receiver 810, the transmitter 815, or both. For example, the communications manager 820 may receive information from the receiver 810, send information to the transmitter 815, or may be integrated in combination with the receiver 810, the transmitter 815, or both to receive information, transmit information, or perform various other operations as described herein.

[0121]

[0131] The communications manager 820 may support wireless communications in the UE according to examples as disclosed herein. The UE machine learning manager 830 may be configured as or otherwise support a means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The UE activation component 835 may be configured as or otherwise support a means for receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0122]

[0132] 9 illustrates a block diagram 900 of a communications manager 920 that supports configuring a UE for machine learning, according to aspects of the disclosure. Communications manager 920 may be an example of aspects of communications manager 720, communications manager 820, or both, as described herein. Communications manager 920, or various components thereof, may be examples of means for performing various aspects of configuring a UE for machine learning, as described herein. For example, communications manager 920 may include a UE machine learning manager 930, a UE activation component 935, a UE address manager 940, a UE upload component 945, a UE request component 950, or any combination thereof. Each of these components may communicate with each other indirectly or directly (e.g., via one or more buses).

[0123]

[0133] The communications manager 920 may support wireless communications in the UE according to examples as disclosed herein. The UE machine learning manager 930 may be configured as or otherwise support a means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The UE activation component 935 may be configured as or otherwise support a means for receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0124]

[0134] In some examples, the UE request component 950 may be configured or otherwise support a means for sending a request message including an indication of the machine learning models, the neural network functions, or both to the base station, where receiving the machine learning models, the neural network functions, or both is based on the request message. In some examples, the UE request component 950 may be configured or otherwise support a means for receiving signaling from the base station indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, where sending the request message is based on the machine learning models being included in the whitelist, excluded from the blacklist, or both.

[0125]

[0135] In some examples, each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a DNN, a PLMN, a UE type, an RRC state, a communication service, a communication configuration, or any combination thereof, and sending the request message is based on a trigger event including the UE having a condition that is within the respective scope of the machine learning model.

[0126]

[0136] In some examples, the request message includes an indication of the triggering event. In some examples, to support sending the request message, the UE request component 950 may be configured as or otherwise support a means for sending a UE assistance information message that includes the request message.

[0127]

[0137] In some examples, to support transmitting the request message, the UE request component 950 may be configured as or otherwise support a means for transmitting RRC signaling that includes the request message.

[0128]

[0138] In some examples, the UE address manager 940 may be configured as or otherwise support a means for determining an address for a machine learning model, a set of parameters, or a configuration based on an associated identifier and an associated rule, where receiving the machine learning model, set of parameters, or configuration is based on downloading the machine learning model, set of parameters, or configuration from a machine learning MR based on the address.

[0129]

[0139] In some examples, the UE address manager 940 may be configured as or otherwise support a means for determining, based on the associated identifier and the associated rule, an address for the second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions. In some examples, the UE upload component 945 may be configured as or otherwise support a means for initiating an upload of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0130]

[0140] In some examples, sending the request message includes sending the request message to a CU-CP entity included in the base station. In some examples, receiving the machine learning model, set of parameters, or configuration includes receiving the machine learning model, set of parameters, or configuration from the CU-CP entity.

[0131]

[0141] In some examples, the UE address manager 940 may be configured as or otherwise support a means for receiving an address for a machine learning model, a set of parameters, or a configuration from a CU-XP entity included in the base station, where receiving the machine learning model, set of parameters, or configuration is based on a download of the machine learning model, set of parameters, or configuration from the machine learning MR based on the address.

[0132]

[0142] In some examples, the UE address manager 940 may be configured as or otherwise support a means for receiving an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions from a CU-XP entity included in the base station. In some examples, the UE upload component 945 may be configured as or otherwise support a means for initiating an upload of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0133]

[0143] FIG. 10 illustrates a diagram of a system 1000 including a device 1005 that supports configuring a UE for machine learning, according to aspects of the disclosure. The device 1005 may be or include examples of components of device 705, device 805, or UE 115, as described herein. The device 1005 may communicate wirelessly with one or more base stations 105, UE 115, or any combination thereof. The device 1005 may include components for two-way voice and data communication, including components for transmitting and receiving communications, such as a communications manager 1020, an input / output (I / O) controller 1010, a transceiver 1015, an antenna 1025, a memory 1030, code 1035, and a processor 1040. These components may be in electronic communication or may be otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., bus 1045).

[0134]

[0144] The I / O controller 1010 may manage input and output signals for the device 1005. The I / O controller 1010 may also manage peripheral devices that are not built into the device 1005. In some cases, the I / O controller 1010 may represent a physical connection or port to an external peripheral device. In some cases, the I / O controller 1010 may utilize an operating system such as iOS®, ANDROID®, MS-DOS®, MS-WINDOWS®, OS / 2®, UNIX®, LINUX®, or another known operating system. Additionally or alternatively, the I / O controller 1010 may represent or interact with a modem, keyboard, mouse, touch screen, or similar device. In some cases, the I / O controller 1010 may be implemented as part of a processor, such as the processor 1040. In some cases, a user may interact with the device 1005 through the I / O controller 1010 or through hardware components controlled by the I / O controller 1010.

[0135]

[0145] In some cases, the device 1005 may include a single antenna 1025. However, in some other cases, the device 1005 may have more than one antenna 1025, which may be capable of transmitting or receiving multiple wireless transmissions concurrently. The transceiver 1015 may communicate bidirectionally via one or more antennas 1025, wired links, or wireless links as described herein. For example, the transceiver 1015 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver 1015 may also include a modem for modulating packets, providing the modulated packets to one or more antennas 1025 for transmission, and demodulating packets received from the one or more antennas 1025. The transceiver 1015, or the transceiver 1015 and one or more antennas 1025, may be an example of the transmitter 715, the transmitter 815, the receiver 710, the receiver 810, or any combination or components thereof, as described herein.

[0136]

[0146] The memory 1030 may include random access memory (RAM) and read only memory (ROM). The memory 1030 may store computer readable, computer executable code 1035 including instructions that, when executed by the processor 1040, cause the device 1005 to perform various functions described herein. The code 1035 may be stored in a non-transitory computer readable medium, such as a system memory or another type of memory. In some cases, the code 1035 may not be directly executable by the processor 1040, but may (e.g., when compiled and executed) cause a computer to perform functions described herein. In some cases, the memory 1030 may include a basic I / O system (BIOS) that may control basic hardware or software operations, such as interaction with peripheral components or devices, among others.

[0137]

[0147] The processor 1040 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or any combination thereof). In some cases, the processor 1040 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be incorporated into the processor 1040. The processor 1040 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1030) to cause the device 1005 to perform various functions (e.g., functions or tasks supporting configuring a UE for machine learning). For example, the device 1005 or a component of the device 1005 may include a processor 1040 and a memory 1030 coupled to the processor 1040, where the processor 1040 and the memory 1030 are configured to perform various functions described herein.

[0138]

[0148] The communications manager 1020 may support wireless communications in the UE according to examples as disclosed herein. For example, the communications manager 1020 may be configured as or otherwise support a means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The communications manager 1020 may be configured as or otherwise support a means for receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0139]

[0149] By including or configuring a communications manager 1020 in accordance with examples as described herein, the device 1005 may support techniques for reduced power consumption and improved coordination between devices.

[0140]

[0150] In some examples, the communications manager 1020 may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the transmitter 1015, one or more antennas 1025, or any combination thereof. Although the communications manager 1020 is illustrated as a separate component, in some examples, one or more functions described with reference to the communications manager 1020 may be supported or performed by the processor 1040, the memory 1030, the code 1035, or any combination thereof. For example, the code 1035 may include instructions executable by the processor 1040 to cause the device 1005 to perform various aspects of configuring a UE for machine learning, as described herein, or the processor 1040 and the memory 1030 may be otherwise configured to perform or support such operations.

[0141]

[0151] 11 illustrates a block diagram 1100 of a device 1105 that supports configuring a UE for machine learning according to an aspect of the disclosure. The device 1105 may be an example of an aspect of a base station 105 as described herein. The device 1105 may include a receiver 1110, a transmitter 1115, and a communications manager 1120. The device 1105 may also include a processor. Each of these components may be in communication with one another (e.g., via one or more buses).

[0142]

[0152] The receiver 1110 may provide a means for receiving information, such as packets, user data, control information, or any combination thereof, associated with various information channels (e.g., information channels related to configuring the UE for machine learning, data channels, control channels). The information may be passed to other components of the device 1105. The receiver 1110 may utilize a single antenna or a set of multiple antennas.

[0143]

[0153] The transmitter 1115 may provide a means for transmitting signals generated by other components of the device 1105. For example, the transmitter 1115 may transmit information such as packets, user data, control information, or any combination thereof associated with various information channels (e.g., information channels, data channels, control channels related to configuring the UE for machine learning). In some examples, the transmitter 1115 may be co-located with the receiver 1110 in a transceiver module. The transmitter 1115 may utilize a single antenna or a set of multiple antennas.

[0144]

[0154] The communications manager 1120, the receiver 1110, the transmitter 1115, or various combinations or components thereof may be examples of a means for performing various aspects of configuring a UE for machine learning as described herein. For example, the communications manager 1120, the receiver 1110, the transmitter 1115, or various combinations or components thereof may support a method for performing one or more of the functions described herein.

[0145]

[0155] In some examples, the communications manager 1120, the receiver 1110, the transmitter 1115, or various combinations or components thereof may be implemented in hardware (e.g., in a communications management circuit). The hardware may include a processor, a DSP, an ASIC, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof configured as or otherwise supporting a means for performing the functions described in this disclosure. In some examples, the processor and a memory coupled to the processor may be configured to perform one or more of the functions described herein (e.g., by executing, by the processor, instructions stored in the memory).

[0146]

[0156] Additionally or alternatively, in some examples, the communications manager 1120, the receiver 1110, the transmitter 1115, or various combinations or components thereof may be implemented in code executed by a processor (e.g., as communications management software or firmware). If implemented in code executed by a processor, the functions of the communications manager 1120, the receiver 1110, the transmitter 1115, or various combinations or components thereof may be performed by a general purpose processor, a DSP, a CPU, an ASIC, an FPGA, or any combination of these or other programmable logic devices (e.g., configured as or otherwise supporting a means for performing the functions described in this disclosure).

[0147]

[0157] In some examples, the communications manager 1120 may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the receiver 1110, the transmitter 1115, or both. For example, the communications manager 1120 may receive information from the receiver 1110, send information to the transmitter 1115, or may be integrated in combination with the receiver 1110, the transmitter 1115, or both to receive information, transmit information, or perform various other operations as described herein.

[0148]

[0158] The communications manager 1120 may support wireless communications in the base station according to examples as disclosed herein. For example, the communications manager 1120 may be configured as or otherwise support a means for transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The communications manager 1120 may be configured as or otherwise support a means for transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0149]

[0159] By including or configuring a communications manager 1120 according to examples as described herein, the device 1105 (e.g., a processor controlling or otherwise coupled to the receiver 1110, the transmitter 1115, the communications manager 1120, or a combination thereof) may support techniques for reduced processing and reduced power consumption.

[0150]

[0160] 12 illustrates a block diagram 1200 of a device 1205 that supports configuring a UE for machine learning according to an aspect of the disclosure. The device 1205 may be an example of an aspect of the device 1105 or a base station 105 as described herein. The device 1205 may include a receiver 1210, a transmitter 1215, and a communications manager 1220. The device 1205 may also include a processor. Each of these components may be in communication with one another (e.g., via one or more buses).

[0151]

[0161] The receiver 1210 may provide a means for receiving information, such as packets, user data, control information, or any combination thereof, associated with various information channels (e.g., information channels related to configuring the UE for machine learning, data channels, control channels). The information may be passed to other components of the device 1205. The receiver 1210 may utilize a single antenna or a set of multiple antennas.

[0152]

[0162] The transmitter 1215 may provide a means for transmitting signals generated by other components of the device 1205. For example, the transmitter 1215 may transmit information such as packets, user data, control information, or any combination thereof associated with various information channels (e.g., information channels, data channels, control channels related to configuring the UE for machine learning). In some examples, the transmitter 1215 may be co-located with the receiver 1210 in a transceiver module. The transmitter 1215 may utilize a single antenna or a set of multiple antennas.

[0153]

[0163] The device 1205, or various components thereof, may be examples of means for performing various aspects of configuring a UE for machine learning, as described herein. For example, the communications manager 1220 may include a machine learning manager 1230, an activation component 1235, or any combination thereof. The communications manager 1220 may be an example of an aspect of the communications manager 1120, as described herein. In some examples, the communications manager 1220, or various components thereof, may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the receiver 1210, the transmitter 1215, or both. For example, the communications manager 1220 may receive information from the receiver 1210, send information to the transmitter 1215, or may be integrated in combination with the receiver 1210, the transmitter 1215, or both to receive information, transmit information, or perform various other operations as described herein.

[0154]

[0164] The communications manager 1220 may support wireless communications in the base station according to examples as disclosed herein. The machine learning manager 1230 may be configured as or otherwise support a means for transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The activation component 1235 may be configured as or otherwise support a means for transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0155]

[0165] FIG. 13 illustrates a block diagram 1300 of a communications manager 1320 that supports configuring a UE for machine learning, according to aspects of the disclosure. The communications manager 1320 may be an example of aspects of the communications manager 1120, the communications manager 1220, or both, as described herein. The communications manager 1320, or various components thereof, may be an example of a means for performing various aspects of configuring a UE for machine learning, as described herein. For example, the communications manager 1320 may include a machine learning manager 1330, an activation component 1335, an address component 1340, a download component 1345, an upload component 1350, an identifier component 1355, a request component 1360, or any combination thereof. Each of these components may communicate with each other indirectly or directly (e.g., via one or more buses).

[0156]

[0166] The communication manager 1320 may support wireless communications in the base station according to examples as disclosed herein. The machine learning manager 1330 may be configured as or otherwise support a means for transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The activation component 1335 may be configured as or otherwise support a means for transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0157]

[0167] In some examples, the request component 1360 may be configured as or otherwise support a means for receiving, from the UE, a request message including an indication of the machine learning models, the neural network functions, or both, where transmitting the machine learning models, the neural network functions, or both is based on the request message. In some examples, the request component 1360 may be configured as or otherwise support a means for transmitting, to the UE, signaling indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, where the machine learning models are included in the whitelist, excluded from the blacklist, or both. In some examples, the request message includes an indication of a trigger event.

[0158]

[0168] In some examples, each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a DNN, a PLMN, a UE type, an RRC state, a communication service, a communication configuration, or any combination thereof, and receiving the request message is based on a trigger event including the UE having a condition that is within the respective scope of the machine learning model.

[0159]

[0169] In some examples, to support receiving the request message, the request component 1360 may be configured as or otherwise support a means for receiving a UE assistance information message that includes the request message.

[0160]

[0170] In some examples, to support receiving a request message, the request component 1360 may be configured as or otherwise support a means for receiving RRC signaling that includes the request message.

[0161]

[0171] In some examples, the address component 1340 may be configured or otherwise support a means for receiving an address for a machine learning model, a set of parameters, or a configuration from the UE. In some examples, the download component 1345 may be configured or otherwise support a means for downloading a machine learning model, a set of parameters, or a configuration from the machine learning MR based on the address for the UE.

[0162]

[0172] In some examples, the address component 1340 may be configured as or otherwise support a means for receiving, from the UE, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions. In some examples, the upload component 1350 may be configured as or otherwise support a means for uploading, to the machine learning MR, the second machine learning model, the second set of parameters, or the second configuration.

[0163]

[0173] In some examples, the request component 1360 may be configured or otherwise support a means for receiving a request message at a CU-CP entity included in the base station. In some examples, the request component 1360 may be configured or otherwise support a means for forwarding a request message from the CU-CP entity to a CU-XP entity included in the base station. In some examples, the download component 1345 may be configured or otherwise support a means for downloading a machine learning model, a set of parameters, or a configuration from the machine learning MR to the CU-CP entity based on the request message, where transmitting the machine learning model, the set of parameters, or the configuration to the UE is based on the downloading.

[0164]

[0174] In some examples, the identifier component 1355 may be configured as or otherwise support a means for receiving an identifier associated with a machine learning model, a set of parameters, or a configuration from a UE at a CU-XP entity included in the base station. In some examples, the address component 1340 may be configured as or otherwise support a means for determining an address for a machine learning model, a set of parameters, or a configuration based at least in part on the identifier. In some examples, the download component 1345 may be configured as or otherwise support a means for downloading a machine learning model, a set of parameters, or a configuration from the machine learning MR based on the address for the UE, where transmitting the machine learning model, the set of parameters, or the configuration to the UE is based on the downloading.

[0165]

[0175] In some examples, the identifier component 1355 may be configured as or otherwise support a means for receiving, at a CU-XP entity included in the base station, from the UE, an identifier associated with a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions. In some examples, the address component 1340 may be configured as or otherwise support a means for determining an address for the second machine learning model, the second set of parameters, or the second configuration based at least in part on the identifier. In some examples, the upload component 1350 may be configured as or otherwise support a means for uploading, to the machine learning MR, the second machine learning model, the second set of parameters, or the second configuration based on the address.

[0166]

[0176] FIG. 14 illustrates a diagram of a system 1400 including a device 1405 that supports configuring a UE for machine learning, according to aspects of the disclosure. The device 1405 may be or include an example of a component of the device 1105, device 1205, or base station 105, as described herein. The device 1405 may wirelessly communicate with one or more base stations 105, UEs 115, or any combination thereof. The device 1405 may include components for two-way voice and data communication, including components for transmitting and receiving communications, such as a communications manager 1420, a network communications manager 1410, a transceiver 1415, an antenna 1425, a memory 1430, code 1435, a processor 1440, and an inter-station communications manager 1445. These components may be in electronic communication or may be otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., bus 1450).

[0167]

[0177] The network communications manager 1410 may manage communications with the core network 130 (e.g., via one or more wired backhaul links). For example, the network communications manager 1410 may manage the transfer of data communications for client devices, such as one or more UEs 115.

[0168]

[0178] In some cases, the device 1405 may include a single antenna 1425. However, in some other cases, the device 1405 may have more than one antenna 1425, which may be capable of transmitting or receiving multiple wireless transmissions simultaneously. The transceiver 1415 may communicate bidirectionally via one or more antennas 1425, wired links, or wireless links as described herein. For example, the transceiver 1415 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver 1415 may also include a modem for modulating packets, providing the modulated packets to one or more antennas 1425 for transmission, and demodulating packets received from the one or more antennas 1425. The transceiver 1415, or the transceiver 1415 and one or more antennas 1425, may be an example of the transmitter 1115, the transmitter 1215, the receiver 1110, the receiver 1210, or any combination or components thereof, as described herein.

[0169]

[0179] The memory 1430 may include RAM and ROM. The memory 1430 may store computer-readable, computer-executable code 1435 including instructions that, when executed by the processor 1440, cause the device 1405 to perform various functions described herein. The code 1435 may be stored in a non-transitory computer-readable medium, such as a system memory or another type of memory. In some cases, the code 1435 may not be directly executable by the processor 1440, but may (e.g., when compiled and executed) cause the computer to perform functions described herein. In some cases, the memory 1430 may include a BIOS that may control basic hardware or software operations, such as interaction with peripheral components or devices, among other things.

[0170]

[0180] The processor 1440 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or any combination thereof). In some cases, the processor 1440 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be incorporated into the processor 1440. The processor 1440 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1430) to cause the device 1405 to perform various functions (e.g., functions or tasks supporting configuring a UE for machine learning). For example, the device 1405 or a component of the device 1405 may include the processor 1440 and the memory 1430 coupled to the processor 1440, where the processor 1440 and the memory 1430 are configured to perform various functions described herein.

[0171]

[0181] The inter-station communications manager 1445 may manage communications with other base stations 105 and may include a controller or scheduler for controlling communications with the UE 115 in cooperation with the other base stations 105. For example, the inter-station communications manager 1445 may coordinate scheduling for transmissions to the UE 115 for various interference mitigation techniques, such as beamforming or joint transmission. In some examples, the inter-station communications manager 1445 may provide an X2 interface within the LTE / LTE-A wireless communications network technology to provide communications between the base stations 105.

[0172]

[0182] The communications manager 1420 may support wireless communications in the base station according to examples as disclosed herein. For example, the communications manager 1420 may be configured as or otherwise support a means for transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The communications manager 1420 may be configured as or otherwise support a means for transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0173]

[0183] By including or configuring a communications manager 1420 in accordance with examples as described herein, the device 1405 may support techniques for reduced power consumption and improved coordination between devices.

[0174]

[0184] In some examples, the communications manager 1420 may be configured to perform various operations (e.g., receiving, monitoring, transmitting) using or otherwise cooperating with the transmitter 1415, one or more antennas 1425, or any combination thereof. Although the communications manager 1420 is illustrated as a separate component, in some examples, one or more functions described with reference to the communications manager 1420 may be supported or performed by the processor 1440, the memory 1430, the code 1435, or any combination thereof. For example, the code 1435 may include instructions executable by the processor 1440 to cause the device 1405 to perform various aspects of configuring a UE for machine learning, as described herein, or the processor 1440 and the memory 1430 may be otherwise configured to perform or support such operations.

[0175]

[0185] FIG. 15 shows a flowchart illustrating a method 1500 for supporting configuring a UE for machine learning according to aspects of the disclosure. The operations of method 1500 may be performed by a UE or components thereof as described herein. For example, the operations of method 1500 may be performed by a UE 115 as described with reference to FIGS. 1-10. In some examples, the UE may execute a set of instructions to control functional elements of the UE to perform the described functions. Additionally or alternatively, the UE may perform aspects of the described functions using dedicated hardware.

[0176]

[0186] At 1505, the method may include receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The operations of 1505 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1505 may be performed by a UE machine learning manager 930 as described with reference to FIG. 9.

[0177]

[0187] At 1510, the method may include receiving, from the base station, an activation message for the machine learning model, the neural network function, or both. The operations of 1510 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1510 may be performed by a UE activation component 935, as described with reference to FIG.

[0178]

[0188] FIG. 16 shows a flowchart illustrating a method 1600 for supporting configuring a UE for machine learning according to aspects of the disclosure. The operations of the method 1600 may be performed by a UE or components thereof as described herein. For example, the operations of the method 1600 may be performed by the UE 115 as described with reference to FIGS. 1-10. In some examples, the UE may execute a set of instructions to control functional elements of the UE to perform the described functions. Additionally or alternatively, the UE may perform aspects of the described functions using dedicated hardware.

[0179]

[0189] At 1605, the method may include transmitting a request message to the base station comprising an indication of a machine learning model of the one or more machine learning models, a neural network function of the one or more neural network functions, or both, where the one or more machine learning models, the one or more neural network functions, or any combination thereof are associated with a machine learning model repository included in or coupled to the base station. The operations of 1605 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1605 may be performed by a UE request component 950 as described with reference to FIG. 9.

[0180]

[0190] At 1610, the method may include receiving, from the base station, a machine learning model, a set of parameters corresponding to the machine learning model, or a configuration corresponding to the neural network function based at least in part on transmitting the request message. The operations of 1610 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1610 may be performed by a UE machine learning manager 930 as described with reference to FIG.

[0181]

[0191] At 1615, the method may include receiving, from the base station, an activation message for the machine learning model, the neural network function, or both. The operations of 1615 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1615 may be performed by a UE activation component 935, as described with reference to FIG.

[0182]

[0192] FIG. 17 shows a flowchart illustrating a method 1700 for supporting configuring a UE for machine learning according to aspects of the disclosure. The operations of method 1700 may be performed by a base station or components thereof as described herein. For example, the operations of method 1700 may be performed by a base station 105 as described with reference to FIGS. 1-6 and 11-14. In some examples, the base station may execute a set of instructions to control functional elements of the base station to perform the described functions. Additionally or alternatively, the base station may perform aspects of the described functions using dedicated hardware.

[0183]

[0193] At 1705, the method may include transmitting to the UE a machine learning model of the one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of the one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof may be associated with a machine learning MR included in or coupled to the base station. The operations of 1705 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1705 may be performed by the machine learning manager 1330 as described with reference to FIG. 13.

[0184]

[0194] At 1710, the method may include transmitting to the UE an activation message for the machine learning model, the neural network function, or both. The operations of 1710 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1710 may be performed by the activation component 1335 as described with reference to FIG.

[0185]

[0195] FIG. 18 shows a flowchart illustrating a method 1800 for supporting configuring a UE for machine learning according to aspects of the disclosure. The operations of the method 1800 may be performed by a base station or components thereof as described herein. For example, the operations of the method 1800 may be performed by a base station 105 as described with reference to FIGS. 1-6 and 11-14. In some examples, the base station may execute a set of instructions to control functional elements of the base station to perform the described functions. Additionally or alternatively, the base station may perform aspects of the described functions using dedicated hardware.

[0186]

[0196] At 1805, the method may include receiving a request message from the UE comprising an indication of a machine learning model of one or more machine learning models, a neural network function of one or more neural network functions, or both, where the one or more machine learning models, the one or more neural network functions, or any combination thereof are associated with a machine learning model repository included in or coupled to the base station. The operations of 1805 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1805 may be performed by a request component 1360 as described with reference to FIG. 13.

[0187]

[0197] At 1810, the method may include transmitting, to the UE, the machine learning model, a set of parameters corresponding to the machine learning model, or a configuration corresponding to the neural network function based at least in part on receiving the request message. The operations of 1810 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1810 may be performed by the machine learning manager 1330 as described with reference to FIG.

[0188]

[0198] At 1815, the method may include transmitting, to the UE, an activation message for the machine learning model, the neural network function, or both. The operations of 1815 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1815 may be performed by the activation component 1335, as described with reference to FIG.

[0189]

[0199] The following provides a summary of aspects of the present disclosure:

[0190]

[0200] Aspect 1: A method for wireless communication in a UE, comprising receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, wherein the one or more machine learning models, the one or more neural network functions, or any combination thereof are associated with a machine learning MR included in or coupled to a base station, and receiving an activation message for the machine learning model, the neural network function, or both from the base station.

[0191]

[0201] Aspect 2: The method of aspect 1, further comprising: sending a request message to a base station comprising an indication of the machine learning model, the neural network function, or both, wherein receiving the machine learning model, the neural network function, or both is based at least in part on the request message.

[0192]

[0202] Aspect 3: The method of aspect 2, further comprising receiving signaling from a base station indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, wherein sending the request message is based at least in part on the machine learning models being included in the whitelist, being excluded from the blacklist, or both.

[0193]

[0203] Aspect 4: A method according to any of aspects 2 to 3, wherein each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a DNN, a PLMN, a UE type, an RRC state, a communication service, a communication configuration, or any combination thereof, and wherein sending the request message is based at least in part on a trigger event comprising the UE having a condition that is within the respective scope of the machine learning model.

[0194]

[0204] Aspect 5: The method of aspect 4, wherein the request message comprises an indication of the trigger event.

[0195]

[0205] Aspect 6: The method of any of aspects 2-5, wherein transmitting the request message comprises transmitting a UE assistance information message comprising the request message.

[0196]

[0206] Example 7: The method of any of Examples 2-6, wherein transmitting the request message comprises transmitting RRC signaling comprising the request message.

[0197]

[0207] Aspect 8: A method according to any of aspects 2 to 7, wherein sending the request message comprises sending the request message to a CU-CP entity included in the base station, and receiving the machine learning model, set of parameters, or configuration comprises receiving the machine learning model, set of parameters, or configuration from the CU-CP entity.

[0198]

[0208] Aspect 9: A method as described in any of aspects 1 to 7, further comprising determining an address for a machine learning model, set of parameters, or configuration based at least in part on the associated ID and the associated rule, wherein receiving the machine learning model, set of parameters, or configuration is based at least in part on a download of the machine learning model, set of parameters, or configuration from the machine learning MR based at least in part on the address.

[0199]

[0209] Aspect 10: A method as described in any of aspects 1 to 7, further comprising: determining an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions based at least in part on the associated ID and the associated rule; and initiating uploading of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based at least in part on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0200]

[0210] Aspect 11: A method as described in any of aspects 1 to 7, further comprising receiving an address for a machine learning model, a set of parameters, or a configuration from a CU-XP entity included in the base station, wherein receiving the machine learning model, set of parameters, or configuration is based at least in part on the address and at least in part on a download of the machine learning model, set of parameters, or configuration from the machine learning MR.

[0201]

[0211] Aspect 12: A method as described in any of aspects 1 to 7, further comprising: receiving, from a CU-XP entity included in the base station, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; and initiating uploading of the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based at least in part on the address for the second machine learning model, the second set of parameters, or the second configuration.

[0202]

[0212] Aspect 13: A method for wireless communication in a base station, comprising: transmitting to a UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, where the one or more machine learning models, the one or more neural network functions, or any combination thereof are associated with a machine learning MR included in or coupled to the base station; and transmitting to the UE an activation message for the machine learning model, the neural network function, or both.

[0203]

[0213] Aspect 14: The method of aspect 13, further comprising receiving a request message from the UE including an indication of the machine learning model, the neural network function, or both, wherein transmitting the machine learning model, the neural network function, or both is based at least in part on the request message.

[0204]

[0214] Aspect 15: The method of aspect 14, further comprising sending signaling to the UE indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, wherein the machine learning models are included in the whitelist, excluded from the blacklist, or both.

[0205]

[0215] Aspect 16: A method according to any of aspects 14 to 15, wherein each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a DNN, a PLMN, a UE type, an RRC state, a communication service, a communication configuration, or any combination thereof, and receiving the request message is based at least in part on a trigger event comprising the UE having a condition that is within the respective scope of the machine learning model.

[0206]

[0216] Aspect 17: The method of aspect 16, wherein the request message comprises an indication of the trigger event.

[0207]

[0217] Example 18: The method of any of Examples 14-17, wherein receiving the request message comprises receiving a UE assistance information message comprising the request message.

[0208]

[0218] Example 19: The method of any of examples 14-18, wherein receiving the request message comprises receiving RRC signaling comprising the request message.

[0209]

[0219] Aspect 20: A method as described in any of aspects 14 to 19, further comprising: receiving a request message at a CU-CP entity included in the base station; forwarding the request message from the CU-CP entity to a CU-XP entity included in the base station; and downloading a machine learning model, a set of parameters, or a configuration from the machine learning MR to the CU-CP entity based at least in part on the request message, wherein transmitting the machine learning model, the set of parameters, or the configuration to the UE is based at least in part on the downloading.

[0210]

[0220] Aspect 21: A method as described in any of aspects 13 to 19, further comprising receiving an address for a machine learning model, a set of parameters, or a configuration from the UE, and downloading the machine learning model, the set of parameters, or the configuration from the machine learning MR for the UE based at least in part on the address.

[0211]

[0221] Aspect 22: A method as described in any of aspects 13 to 19, further comprising: receiving from the UE an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function among the one or more neural network functions; and uploading the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR.

[0212]

[0222] Aspect 23: A method as described in any of aspects 13 to 19, further comprising: at a CU-XP entity included in a base station, receiving from a UE an ID associated with a machine learning model, a set of parameters, or a configuration; determining an address for the machine learning model, the set of parameters, or the configuration based at least in part on the ID; and downloading for the UE the machine learning model, the set of parameters, or the configuration from the machine learning MR based at least in part on the address, wherein transmitting the machine learning model, the set of parameters, or the configuration to the UE is based at least in part on the downloading.

[0213]

[0223] Aspect 24: A method as described in any of aspects 13 to 19, further comprising: receiving, at a CU-XP entity included in the base station, from the UE, an ID associated with a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; determining an address for the second machine learning model, the second set of parameters, or the second configuration based at least in part on the ID; and uploading the second machine learning model, the second set of parameters, or the second configuration to the machine learning MR based at least in part on the address.

[0214]

[0224] Aspect 25: An apparatus for wireless communication in a UE, comprising: a processor; a memory coupled to the processor; and instructions executable by the processor and stored in the memory to cause the apparatus to perform a method according to any of aspects 1-12.

[0215]

[0225] Example 26: An apparatus for wireless communication in a UE, comprising at least one means for performing the method according to any one of Examples 1 to 12.

[0216]

[0226] Aspect 27: A non-transitory computer-readable medium having stored thereon code for wireless communication in a UE, the code comprising instructions executable by a processor to perform a method as recited in any of aspects 1-12.

[0217]

[0227] Aspect 28: An apparatus for wireless communication in a base station, comprising: a processor; a memory coupled to the processor; and instructions executable by the processor and stored in the memory to cause the apparatus to perform a method according to any of aspects 13-24.

[0218]

[0228] Example 29: An apparatus for wireless communication in a base station, comprising at least one means for performing the method according to any of Examples 13 to 24.

[0219]

[0229] Aspect 30: A non-transitory computer-readable medium having stored thereon code for wireless communication in a base station, the code comprising instructions executable by a processor to perform a method as described in any of aspects 13-24.

[0220]

[0230] It should be noted that the methods described herein describe possible implementations, and that the operations and steps may be rearranged or otherwise modified, and that other implementations are possible. Additionally, aspects from two or more of these methods may be combined.

[0221]

[0231] Aspects of LTE, LTE-A, LTE-A Pro, or NR systems may be described for purposes of example, and LTE, LTE-A, LTE-A Pro, or NR terminology may be used in much of the description, but the techniques described herein may be applicable beyond LTE, LTE-A, LTE-A Pro, or NR networks. For example, the techniques described may also be applicable to various other wireless communications systems, such as Ultra Mobile Broadband (UMB), Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash OFDM, and other systems and radio technologies not explicitly mentioned herein.

[0222]

[0232] The information and signals described herein may be represented using any of a variety of different technologies and techniques. For example, the data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0223]

[0233] The various example blocks and components described in connection with the disclosure herein may be implemented or performed using a general purpose processor, a DSP, an ASIC, a CPU, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but alternatively, the processor may be any processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0224]

[0234] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. When implemented in software executed by a processor, the functions may be stored or transmitted as one or more instructions or codes on a computer-readable medium. Other examples and implementations are within the scope of the appended claims and this disclosure. For example, depending on the nature of the software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hardwiring, or any combination thereof. Features implementing the functions may also be physically located in various positions, including being distributed such that portions of the functions are implemented in different physical locations.

[0225]

[0235] Computer-readable media includes both communication media and non-transitory computer storage media, including any medium that facilitates transfer of a computer program from one place to another. Non-transitory storage media may be any available medium that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, non-transitory computer-readable media may include RAM, ROM, Electrically Erasable Programmable ROM (EEPROM), Flash memory, Compact Disk (CD) ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a general purpose or special purpose computer, or a general purpose or special purpose processor. Also, any connection is strictly referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included within the definition of computer readable media. As used herein, disk and disc include CDs, laser discs, optical disks, digital versatile disks (DVDs), floppy disks, and Blu-ray discs, where disks usually reproduce data magnetically while discs reproduce data optically using lasers. Combinations of the above are also included within the scope of computer readable media.

[0226]

[0236] As used herein, including in the claims, "or" in a list of items (e.g., a list of items preceded by the phrase "at least one of" or "one or more of") indicates an inclusive list, such as, for example, a list of at least one of A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C). Also, as used herein, the phrase "based on" is not to be construed as a reference to a closed set of conditions. For example, an example step described as "based on condition A" may be based on both condition A and condition B without departing from the scope of the present disclosure. In other words, as used herein, the phrase "based on" is to be construed in the same way as the phrase "based at least in part on". Also, as used herein, the phrase "set" is to be construed in the same way as "one or more".

[0227]

[0237] The terms "determine" or "determining" encompass a wide variety of actions, and thus "determining" can include calculating, computing, processing, deriving, investigating, looking up (such as via looking up in a table, database, or another data structure), ascertaining, and the like. "Determining" can also include receiving (such as receiving information), accessing (such as accessing data in a memory), and the like. "Determining" can also include resolving, selecting, choosing, establishing, and other such similar actions.

[0228]

[0238] In the accompanying drawings, similar components or features may have the same reference label. Furthermore, various components of the same type may be distinguished by following the reference label with a dash and a second label that distinguishes between the similar components. If only a first reference label is used herein, the description is applicable to any one of the similar components having the same first reference label, regardless of the second reference label, or any other subsequent reference label.

[0229]

[0239] The description set forth herein with reference to the accompanying drawings describes example configurations and does not represent every example that may be implemented or is within the scope of the claims. As used herein, the term "example" means "providing an example, instance, or illustration," rather than "preferred" or "advantageous over other examples." The detailed description includes specific details for the purpose of providing an understanding of the described techniques. However, these techniques may be practiced without these specific details. In some instances, well-known structures and devices are shown in block diagram form in order to avoid obscuring the concepts of the described examples.

[0230]

[0240] The description herein is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to the present disclosure will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the examples and designs described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. 1. A method for wireless communication in a user equipment (UE), comprising: transmitting capability information comprising a list of neural network functions supported by the UE, a list of machine learning models supported by the UE, or any combination thereof; receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, wherein the one or more machine learning models, the one or more neural network functions, or any combination thereof, are associated with a machine learning model repository included in or coupled to the base station; receiving, from the base station, an activation message for the machine learning model, the neural network function, or both, wherein the method comprises: determining an address for the machine learning model, the set of parameters, or the configuration based at least in part on the associated identifier and the associated rule; and receiving the machine learning model, the set of parameters, or the configuration based at least in part on a download of the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address; A method further comprising:

2. sending a request message to the base station comprising an indication of the machine learning model, the neural network function, or both, wherein receiving the machine learning model, the neural network function, or both is based at least in part on the request message; The method of claim 1 further comprising:

3. receiving signaling from the base station indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, wherein sending the request message is based at least in part on the machine learning models being included in the whitelist, being excluded from the blacklist, or both; or each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a deep neural network, a public land mobile network, a UE type, a radio resource control state, a communication service, a communication configuration, or any combination thereof; The method further comprises sending the request message based at least in part on a trigger event comprising the UE having a condition that is within the respective scope of the machine learning models, the request message comprising an indication of the trigger event; or transmitting a UE Assistance Information message comprising the request message; or transmitting radio resource control signaling comprising the request message; or sending said request message to a central unit-control plane entity included in said base station; receiving the machine learning model, the set of parameters, or the configuration from the central unit—a control plane entity; The method of claim 2.

4. determining, based at least in part on the associated identifier and the associated rule, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; Initiating an upload of the second machine learning model, the second set of parameters, or the second configuration to the machine learning model repository based at least in part on the address for the second machine learning model, the second set of parameters, or the second configuration; The method of claim 1 further comprising:

5. receiving an address for the machine learning model, the set of parameters, or the configuration from a central unit-machine learning plane entity included in the base station; receiving the machine learning model, the set of parameters, or the configuration based at least in part on a download of the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address; The method of claim 1 further comprising:

6. receiving, from a central unit - machine learning plane entity included in the base station, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; Initiating an upload of the second machine learning model, the second set of parameters, or the second configuration to the machine learning model repository based at least in part on the address for the second machine learning model, the second set of parameters, or the second configuration; The method of claim 1 further comprising:

7. 1. A method for wireless communication in a base station, comprising: receiving capability information from a user equipment (UE) comprising a list of neural network functions supported by the UE, a list of machine learning models supported by the UE, or any combination thereof; transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, wherein the one or more machine learning models, the one or more neural network functions, or any combination thereof, are associated with a machine learning model repository included in or coupled to the base station; sending to the UE an activation message for the machine learning model, the neural network function, or both; receiving an address for the machine learning model, the set of parameters, or the configuration from the UE; downloading, for the UE, the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address; A method further comprising:

8. receiving a request message from the UE including an indication of the machine learning model, the neural network function, or both, wherein transmitting the machine learning model, the neural network function, or both is based at least in part on the request message; The method of claim 7 further comprising:

9. sending signaling to the UE indicating a first set of machine learning models included in a blacklist, a second set of machine learning models included in a whitelist, or both, wherein receiving the request message is based at least in part on the machine learning models being included in the whitelist, being excluded from the blacklist, or both; or each machine learning model of the one or more machine learning models is associated with a respective scope corresponding to a location, a network slice, a deep neural network, a public land mobile network, a UE type, a radio resource control state, a communication service, a communication configuration, or any combination thereof; receiving the request message based at least in part on a trigger event comprising the UE having a condition that is within the respective scope of the machine learning models; the request message further comprises an indication of the trigger event; or receiving a UE Assistance Information message comprising the request message; or receiving radio resource control signaling comprising the request message; or receiving the request message at a central unit - a control plane entity included in the base station; forwarding the request message from the central unit - control plane entity to a central unit - machine learning plane entity included in the base station; downloading the machine learning model, the set of parameters, or the configuration from the machine learning model repository to the central unit—control plane entity based at least in part on the request message, wherein transmitting the machine learning model, the set of parameters, or the configuration to the UE is based at least in part on the downloading. The method of claim 8.

10. receiving, from the UE, an address for a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; uploading the second machine learning model, the second set of parameters, or the second configuration to the machine learning model repository; The method of claim 7 further comprising:

11. receiving, at a central unit—machine learning plane entity included in the base station, from the UE, an identifier associated with the machine learning model, the set of parameters, or the configuration; determining an address for the machine learning model, the set of parameters, or the configuration based at least in part on the identifier; and downloading, for the UE, the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address, and wherein transmitting the machine learning model, the set of parameters, or the configuration to the UE is based at least in part on the downloading. or receiving, in a central unit—machine learning plane entity included in the base station, from the UE, an identifier associated with a second machine learning model, a second set of parameters corresponding to the second machine learning model, or a second configuration corresponding to a second neural network function of the one or more neural network functions; determining an address for the second machine learning model, the second set of parameters, or the second configuration based at least in part on the identifier; and uploading the second machine learning model, the second set of parameters, or the second configuration to the machine learning model repository based at least in part on the address; and The method of claim 7 further comprising:

12. An apparatus for wireless communication in a user equipment (UE), comprising: means for transmitting capability information comprising a list of neural network functions supported by the UE, a list of machine learning models supported by the UE, or any combination thereof; means for receiving a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, wherein the one or more machine learning models, the one or more neural network functions, or any combination thereof, are associated with a machine learning model repository included in or coupled to the base station; means for receiving an activation message for the machine learning model, the neural network function, or both from the base station, wherein the apparatus comprises: means for determining an address for the machine learning model, the set of parameters, or the configuration based at least in part on an associated identifier and an associated rule; means for receiving the machine learning model, the set of parameters, or the configuration based at least in part on downloading the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address; The apparatus further comprises:

13. The apparatus of claim 12, further comprising means for performing the method of any one of claims 2 to 6.

14. An apparatus for wireless communication in a base station, comprising: means for receiving capability information from a user equipment (UE), the capability information comprising a list of neural network functions supported by the UE, a list of machine learning models supported by the UE, or any combination thereof; means for transmitting to the UE a machine learning model of one or more machine learning models, a set of parameters corresponding to the machine learning model, or a configuration corresponding to a neural network function of one or more neural network functions, wherein the one or more machine learning models, the one or more neural network functions, or any combination thereof, are associated with a machine learning model repository included in or coupled to the base station; means for sending to the UE an activation message for the machine learning model, the neural network function, or both; means for receiving an address for the machine learning model, the set of parameters, or the configuration from the UE; means for downloading, for the UE, the machine learning model, the set of parameters, or the configuration from the machine learning model repository based at least in part on the address; The apparatus further comprises:

15. The apparatus of claim 14, further comprising means for carrying out the method of any one of claims 8 to 11.

Images