Over-the-air aggregation federated learning with non-connected devices

Abstract

Methods, systems, and devices for wireless communication are described. A user equipment (UE) can receive a first message requesting the UE to perform a federated learning procedure. The first message can indicate a machine learning model and a configuration for the federated learning procedure. The UE can perform, in response to the first message, a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure. The UE can transmit, based on the configuration for the federated learning procedure, a second message indicating one or more gradient values for the one or more model parameters via one or more resources configured for over-the-air (OTA) aggregation.

Description

[0001] Cross-references

[0002] This patent application claims priority to Greek patent application No. 20220100644, filed by STEFANATOS et al. on August 4, 2023, entitled “OVER-THE-AIR AGGREGATION FEDERATED LEARNING WITH NONCONNECTED DEVICES,” which is assigned to the assignee of this application and is expressly incorporated herein by reference. Technical Field

[0003] The following relates to wireless communications, including over-the-air (OTA) aggregated federated learning with non-connected devices. Background Art

[0004] Wireless communication systems are widely deployed to provide various types of communication content, such as voice, video, packet data, messaging, broadcast, and the like. These systems may be able to support 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) systems, advanced LTE (LTE-A) systems, or LTE-A Pro systems) and fifth generation (5G) systems (which may be referred to as new radio (NR) systems). These systems may employ techniques 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, each of which supports wireless communications for communication devices, which may be referred to as user equipment (UE). Summary of the invention

[0005] The described technology relates to improved methods, systems, devices, and apparatuses for supporting over-the-air (OTA) aggregated federated learning with non-connected devices. For example, the described technology allows a user equipment (UE) operating in a non-connected mode to participate (e.g., optionally participate) in training rounds for federated learning. A server for a federated learning procedure may send an invitation message requesting a client (e.g., a UE) to participate in training rounds for a federated learning procedure. The invitation message may indicate a machine learning model and a configuration for the federated learning procedure. The UE may locally train the machine learning model based on a data set collected by the UE to obtain model parameters for the machine learning model. The UE may send gradient values ​​of model parameters on one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. In some examples, the invitation message may indicate a power control configuration for sending gradient values ​​on one or more resources configured for OTA aggregation.

[0006] A method for wireless communication at a UE is described. The method may include: receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; in response to the first message requesting the UE to perform the federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure; and sending a second message indicating one or more gradient values ​​of the one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0007] A device for wireless communication at a UE is described. The device may include a processor, a memory coupled to the processor, and one or more instructions stored in the memory. The one or more instructions may be executed by the processor to cause the device to: receive a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; in response to the first message requesting the UE to perform the federated learning procedure, perform a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure; and based on the configuration for the federated learning procedure, send a second message indicating one or more gradient values ​​of the one or more model parameters via one or more resources configured for OTA aggregation.

[0008] Another apparatus for wireless communication at a UE is described. The apparatus may include: a component for receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; a component for performing a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure in response to the first message requesting the UE to perform the federated learning procedure; and a component for sending a second message indicating one or more gradient values ​​of the one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0009] A non-transitory computer-readable medium storing code for wireless communication at a UE is described. The code may include instructions executable by a processor to perform the following operations: receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; in response to the first message requesting the UE to perform the federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure; and based on the configuration for the federated learning procedure, sending a second message indicating one or more gradient values ​​of the one or more model parameters via one or more resources configured for OTA aggregation.

[0010] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving a first message indicating a range associated with participating in a federated learning procedure, wherein sending a second message may be based on determining that the UE may be within the range.

[0011] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for: receiving a first message indicating a reference signal and a threshold; and measuring the reference signal to obtain a measurement of the reference signal, wherein sending a second message may be based on the measurement of the reference signal exceeding the threshold.

[0012] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving the first message indicating a nominal transmit power for sending a second message on one or more resources configured for OTA aggregation, wherein sending the second message may be based on the nominal transmit power.

[0013] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving an indication of a path loss reference signal and an indication of a reference signal power value; and measuring the path loss reference signal, wherein sending a second message may be based on the measurement of the path loss reference signal and the reference signal power value.

[0014] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving an indication of a location of a server; and determining a path loss estimate for a second message based on a path loss model and a distance between the UE and the location of the server, wherein sending the second message may be based on the path loss estimate.

[0015] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving the first message may include an operation, feature, component, or instruction for receiving an indication of a machine learning model from a set of multiple machine learning models.

[0016] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving an indication of a number of layers in a set of multiple layers, a size of each layer in a set of multiple layers, an order of a set of multiple layers, connectivity of a set of multiple layers, or any combination thereof for a machine learning model, wherein execution of a training procedure may be based on the number of layers, the size of each layer in a set of multiple layers, the order of a set of multiple layers, connectivity of a set of multiple layers, or any combination thereof.

[0017] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include an operation, feature, component, or instruction for receiving an indication of a version of a machine learning model from a set of multiple versions of the machine learning model, wherein executing a training procedure may be based on the version of the machine learning model.

[0018] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving the first message may include operations, features, components, or instructions for receiving an indication of an OTA aggregation scheme from a set of a plurality of OTA aggregation schemes.

[0019] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving, in one or more paging frames, an indication of a set of multiple occasions for a first message requesting the UE to perform a federated learning procedure; and monitoring the first message based on the indication of the set of multiple occasions for the first message.

[0020] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving a wake-up signal associated with a first message; and monitoring the first message requesting the UE to perform a federated learning procedure based on receiving the wake-up signal.

[0021] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: decoding a first message based on a paging radio network temporary identifier; and detecting one or more fields in the first message corresponding to a federated learning procedure, wherein execution of a training procedure may be based on detecting the one or more fields.

[0022] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for decoding a first message based on a radio network temporary identifier associated with a federated learning procedure, wherein executing a training procedure may be based on decoding the first message based on a radio network temporary identifier associated with the federated learning procedure.

[0023] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving the first message during a paging frame or opportunity associated with a federated learning procedure, wherein execution of the training procedure may be based on receiving the first message during a paging frame or opportunity associated with the federated learning procedure.

[0024] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving the first message may include operations, features, components, or instructions for receiving a broadcast transmission or a multicast transmission of sidelink control information on a sidelink channel that includes a request for the UE to perform a federated learning procedure.

[0025] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving a first message indicating one or more criteria for participating in a federated learning procedure, wherein the one or more criteria may be based on a version of the machine learning model, a minimum local dataset size, an acquisition time of the local dataset at the UE, or any combination thereof.

[0026] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, sending a second message may include operations, features, components, or instructions for sending the second message via one or more resources configured for transmission by a set of multiple UEs including the UE, wherein the set of multiple UEs may be scheduled to send a corresponding set of gradient values ​​via one or more resources according to a common transmission power.

[0027] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving a third message indicating an updated version of the machine learning model based on one or more gradient values ​​of one or more model parameters; and updating the machine learning model based on the updated version of the machine learning model.

[0028] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: receiving a third message indicating an identical version model of a machine learning model and a scalar probability threshold for participating in a federated learning procedure; determining to execute a federated learning procedure based on a randomly generated value satisfying the scalar probability threshold; executing a training procedure using the machine learning model to obtain a second one or more model parameters based on a configuration for the federated learning procedure and that the randomly generated value satisfies the scalar probability threshold; and sending a fourth message indicating a second one or more gradient values ​​for the second one or more model parameters based on the configuration for the federated learning procedure.

[0029] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, the UE may operate in an inactive state or an idle state.

[0030] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, receiving a first message may include operations, features, components, or instructions for receiving the first message to configure the UE to send one or more model parameters in a second message, wherein the second message indicates the one or more model parameters.

[0031] A method for wireless communication at a wireless device is described. The method may include: sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; based on the configuration for the federated learning procedure, receiving a set of multiple second messages indicating a set of multiple sets of gradient values ​​for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation; and training the machine learning model using the set of multiple sets of the one or more model parameters.

[0032] An apparatus for wireless communication at a wireless device is described. The apparatus may include a processor, a memory coupled to the processor, and one or more instructions stored in the memory. The one or more instructions may be executed by the processor to cause the apparatus to: send a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; based on the configuration for the federated learning procedure, receive a set of multiple second messages indicating a set of multiple sets of gradient values ​​for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation; and train the machine learning model using the set of multiple sets of the one or more model parameters.

[0033] Another apparatus for wireless communication at a wireless device is described. The apparatus may include: a component for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; a component for receiving a set of multiple second messages indicating a set of multiple sets of gradient values ​​for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure; and a component for training the machine learning model using the set of multiple sets of the one or more model parameters.

[0034] A non-transitory computer-readable medium storing code for wireless communication at a wireless device is described. The code may include instructions executable by a processor to perform the following operations: sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; based on the configuration for the federated learning procedure, receiving a set of multiple second messages indicating a set of multiple sets of gradient values ​​for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation; and training the machine learning model using the set of multiple sets of the one or more model parameters.

[0035] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, sending a first message may include operations, features, components, or instructions for sending a first message indicating a range associated with participating in a federated learning procedure, where a set of multiple second messages may be received from UEs within the range.

[0036] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, sending a first message may include operations, features, components, or instructions for sending an indication of a machine learning model from a set of multiple machine learning models, a version of a machine learning model from a set of multiple versions of a machine learning model, an OTA aggregation scheme from a set of multiple OTA aggregation schemes, a federated learning procedure from a set of multiple federated learning procedures, or any combination thereof.

[0037] In some examples of the methods, apparatus, and non-transitory computer-readable media described herein, sending the first message may include operations, features, components, or instructions for sending an indication of the number of layers in a set of multiple layers of a machine learning model, the size of each layer in the set of multiple layers, the order of the set of multiple layers, the connectivity of the set of multiple layers, or any combination thereof.

[0038] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: sending a third message indicating an updated version of the machine learning model based on training the machine learning model; and receiving a set of multiple fourth messages indicating a second set of multiple sets of gradient values ​​for a second set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on a configuration for a federated learning procedure.

[0039] Some examples of the methods, apparatus, and non-transitory computer-readable media described herein may also include operations, features, components, or instructions for: determining that training a machine learning model using a second set of multiple sets of one or more model parameters may be unsuccessful based on excess received power of a set of multiple fourth messages; and sending a fifth message requesting a set of multiple UEs to perform a federated learning procedure, the fifth message indicating an updated version of the machine learning model and a reduced nominal UE transmit power. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 An example of a wireless communication system supporting over-the-air (OTA) aggregated federated learning with non-connected devices in accordance with one or more aspects of the present disclosure is illustrated.

[0041] Figure 2 An example of a wireless communication system supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is illustrated.

[0042] Figure 3 An example of a process flow supporting OTA aggregated federated learning with non-connected devices in accordance with one or more aspects of the present disclosure is illustrated.

[0043] Figure 4An example of a process flow supporting OTA aggregated federated learning with non-connected devices in accordance with one or more aspects of the present disclosure is illustrated.

[0044] Figure 5 and Figure 6 A block diagram of a device supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown.

[0045] Figure 7 A block diagram of a communication manager supporting OTA aggregated federated learning with non-connected devices is shown in accordance with one or more aspects of the present disclosure.

[0046] Figure 8 A diagram of a system including a device that supports OTA aggregated federated learning with non-connected devices is shown in accordance with one or more aspects of the present disclosure.

[0047] Fig. 9 and Fig.10 A block diagram of a device supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown.

[0048] Fig.11 A block diagram of a communication manager supporting OTA aggregated federated learning with non-connected devices is shown in accordance with one or more aspects of the present disclosure.

[0049] Fig.12 A diagram of a system including a device that supports OTA aggregated federated learning with non-connected devices is shown in accordance with one or more aspects of the present disclosure.

[0050] Figures 13 to 17 A flow chart illustrating a method for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. DETAILED DESCRIPTION

[0051] A wireless communication system may support a federated learning procedure to train a machine learning model. For a federated learning procedure, multiple clients, such as user equipment (UE), may adopt the same machine learning model structure and locally train model parameters for the machine learning model based on local observations or data. The client may transmit information for updated model parameters to a server, which compiles information from all clients to determine global or aggregated model parameters. The server may then transmit updated model parameters or an updated machine learning model to the client. The federated learning procedure may include multiple training rounds, each of which may be dynamically scheduled by the server. The server may receive information for model parameters, update the machine learning model, and send information for an updated version of the machine learning model to the client for another round of the federated learning procedure.

[0052] Some wireless communication systems may use over-the-air (OTA) aggregation techniques to reduce the overhead of client reporting information for locally updated or locally trained model parameters. A machine learning model may have a large number of parameters (e.g., tens of thousands or hundreds of thousands of parameters), and each client reporting the machine learning model parameters individually may have significant overhead. In order to reduce overhead, the client may send the gradients of the locally trained model parameters on the same set of resources. For example, each gradient (e.g., corresponding to a model parameter) may be mapped to a corresponding resource element. Each client may send a locally calculated simulated gradient value corresponding to the model parameter through the same resource element (e.g., corresponding to a model parameter or gradient). A receiver (such as a server) may observe the superposition of the sent gradient values, which may correspond to the average of the gradient values ​​of the model parameters. The receiver may determine the average of the model parameters from all clients participating in the federated learning procedure, and use the average of the model parameters to update the machine learning model value.

[0053] In some systems that implement OTA aggregation of federated learning procedures, each client may perform power control techniques to ensure that gradient values ​​from all clients are received at the same power. These power control techniques may prevent the near-far effect, which may over-represent data reported from clients that are closer to the receiver or that experience less path loss when OTA aggregation of gradients is performed. For example, UEs that are closer to the receiver may reduce transmit power compared to UEs that are farther away so that the information received from each UE at the receiver has the same average power. Therefore, some systems may only support UEs operating in radio resource control (RRC) connected mode to participate in federated learning procedures using OTA aggregation, because RRC connected UEs may apply a default power control configuration. However, this may reduce the potential information collected for the federated learning procedure and slow down the training procedure of the machine learning model.

[0054] The techniques described herein support federated learning procedures for non-connected UEs using OTA aggregation. These techniques may enable non-connected UEs (such as UEs operating in RRC idle or RRC inactive states) to participate in one or more training rounds, sessions, or procedures of federated learning using OTA aggregation without switching to RRC connected mode, which may reduce overhead and latency from RRC establishment procedures. A server (such as a network entity or a UE) may send an invitation message prior to a federated learning procedure. The invitation message may invite or request a UE (e.g., operating in an RRC connected state or a non-connected state) to participate in an upcoming round of a federated learning procedure. The invitation message may indicate a configuration or a set of one or more parameters for a federated learning procedure. For example, the invitation message may indicate a machine learning model, an OTA aggregation scheme, resources for OTA aggregation, and the like. In some examples, the invitation message may indicate a power control instruction for reporting gradient values ​​of model parameters. Indicating a power control instruction may enable a non-RRC connected UE to report a gradient value so that the gradient value is received at the same power as other UEs participating in the federated learning procedure. In some examples, the invitation message can be a system information block (SIB) message and can be received or read by any UE 115.

[0055] A UE receiving the invitation message may determine whether to participate in a federated learning training round. If the UE determines to participate in a federated learning training round, the UE may implement power control according to the indicated power control instructions of the invitation message. In some cases, the UE may participate in a federated learning training round without sending a notification message that the UE is participating. A UE participating in a federated learning training round may perform a training procedure using a machine learning model to determine parameters for the machine learning model, and the UE may implement power control instructions while sending gradients of the machine learning model on resources configured for OTA aggregation.

[0056] Power control instructions for a federated learning procedure may be based on the distance from the server or the path loss, or both. For example, the invitation message may indicate the distance from the server, and UEs within the distance from the server may participate in the training round. Additionally or alternatively, the invitation message may indicate a reference signal and a power threshold (e.g., a reference signal received power (RSRP) threshold). The UE may measure the RSRP of the reference signal, and UEs with RSRP measurements of the reference signal that meet the power threshold may participate in the federated learning procedure or an upcoming training round of the federated learning procedure. In some examples, the invitation message may indicate a nominal UE transmit power to be applied by UEs participating in the training round when sending gradient values ​​of model parameters. Some additional or alternative techniques for power control are described herein.

[0057] Various aspects of the present disclosure are first described in the context of a wireless communication system. Various aspects of the present disclosure are further illustrated and described by and with reference to apparatus diagrams, system diagrams, and flow diagrams related to OTA aggregated federated learning with non-connected devices.

[0058] Figure 1 An example of a wireless communication system 100 that supports OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is illustrated. The wireless communication system 100 may include one or more network entities 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 advanced LTE (LTE-A) network, an LTE-A Pro network, a new radio (NR) network, or a network operating according to other systems and radio technologies, including future systems and radio technologies not explicitly mentioned herein.

[0059] The network entities 105 may be dispersed throughout a geographic area to form a wireless communication system 100, and may include devices in different forms or with different capabilities. In various examples, the network entities 105 may be referred to as network elements, mobility elements, radio access network (RAN) nodes, or network equipment, among other names. In some examples, the network entities 105 and the UE 115 may communicate wirelessly via one or more communication links 125 (e.g., radio frequency (RF) access links). For example, the network entity 105 may support a coverage area 110 (e.g., a geographic coverage area), within which the UE 115 and the network entity 105 may establish one or more communication links 125. The coverage area 110 may be an example of a geographic area within which the network entity 105 and the UE 115 may support signal communications according to one or more radio access technologies (RATs).

[0060] The UEs 115 may be dispersed throughout the coverage area 110 of the wireless communication system 100, and each UE 115 may be stationary or mobile, or stationary and mobile at different times. The UEs 115 may be devices in different forms or with different capabilities. Figure 1 Some example UEs 115 are illustrated in FIG. 1. The UEs 115 described herein may be capable of supporting communications with various types of devices, such as Figure 1 Other UEs 115 or network entities 105 are shown.

[0061] As described herein, a node of the wireless communication system 100 (which may be referred to as a network node or a wireless node) may be a network entity 105 (e.g., any network entity described herein), a UE 115 (e.g., any UE described herein), a network controller, an apparatus, a device, a computing system, one or more components, or another suitable processing entity configured to perform any of the techniques described herein. For example, the node may be a UE 115. As another example, the node may be a network entity 105. As another example, a first node may be configured to communicate with a second node or a third node. In one aspect of this example, the first node may be a UE 115, the second node may be a network entity 105, and the third node may be a UE 115. In another aspect of this example, the first node may be a UE 115, the second node may be a network entity 105, and the third node may be a network entity 105. In other aspects of this example, the first node, the second node, and the third node may be different relative to these examples. Similarly, references to UE 115, network entity 105, apparatus, device, computing system, etc. may include disclosure of UE 115, network entity 105, apparatus, device, computing system, etc. as nodes. For example, a disclosure that UE 115 is configured to receive information from network entity 105 also discloses that the first node is configured to receive information from a second node.

[0062] In some examples, the network entities 105 may communicate with the core network 130, or communicate with each other, or both. For example, the network entities 105 may communicate with the core network 130 via one or more backhaul communication links 120 (e.g., according to S1, N2, N3 or other interface protocols). In some examples, the network entities 105 may communicate with each other via the backhaul communication link 120 (e.g., according to X2, Xn or other interface protocols) directly (e.g., directly between the network entities 105) or indirectly (e.g., via the core network 130). In some examples, the network entities 105 may communicate with each other via the midhaul communication link 162 (e.g., according to the midhaul interface protocol) or the fronthaul communication link 168 (e.g., according to the fronthaul interface protocol) or any combination thereof. The backhaul communication link 120, the midhaul communication link 162, or the fronthaul communication link 168 may be or include one or more wired links (e.g., electrical links, optical fiber links), one or more wireless links (e.g., radio links, wireless optical links), etc. or various combinations thereof. UE 115 may communicate with core network 130 via communication link 155 .

[0063] One or more of the network entities 105 described herein may include or may be referred to as a base station 140 (e.g., a transceiver base station, a radio base station, an NR base station, an access point, a radio transceiver, a Node B, an evolved Node B (eNB), a next-generation Node B, or a Gigabit Node B (any of which may be referred to as a gNB), a 5G NB, a next-generation eNB (ng-eNB), a Home Node B, a Home Evolved Node B, or other suitable terms). In some examples, the network entity 105 (e.g., a base station 140) may be implemented in an aggregated (e.g., monolithic, stand-alone) base station architecture, which may be configured to utilize a protocol stack physically or logically integrated within a single network entity 105 (e.g., a single RAN node, such as a base station 140).

[0064] In some examples, the network entity 105 may be implemented in a disaggregated architecture (e.g., a disaggregated base station architecture, a disaggregated RAN architecture) that may be configured to utilize a protocol stack that is physically or logically distributed between two or more network entities 105, such as an integrated access backhaul (IAB) network, an open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance), or a virtualized RAN (vRAN) (e.g., a cloud RAN (C-RAN)). For example, the network entity 105 may include one or more of the following: a central unit (CU) 160, a distributed unit (DU) 165, a radio unit (RU) 170, a RAN intelligent controller (RIC) 175 (e.g., a near real-time RIC (near RT RIC), a non-real-time RIC (non-RT RIC)), a service management and orchestration (SMO) 180 system, or any combination thereof. The RU 170 may also be referred to as a radio head, a smart radio head, a remote radio head (RRH), a remote radio unit (RRU), or a transmit receive point (TRP). One or more components of the network entity 105 in the decomposed RAN architecture may be co-located, or one or more components of the network entity 105 may be located in distributed locations (e.g., separate physical locations). In some examples, one or more network entities 105 of the decomposed RAN architecture may be implemented as virtual units (e.g., virtual CU (VCU), virtual DU (VDU), virtual RU (VRU)).

[0065] The functional split between CU 160, DU 165, and RU 170 is flexible and may support different functionalities, depending on which functions (e.g., network layer functions, protocol layer functions, baseband functions, RF functions, and any combination thereof) are performed at CU 160, DU 165, or RU 170. For example, a functional split of a protocol stack may be employed between CU 160 and DU 165 such that CU 160 may support one or more layers of a protocol stack and DU 165 may support one or more different layers of a protocol stack. In some examples, CU 160 may host higher protocol layer (e.g., layer 3 (L3), layer 2 (L2)) functionality and signaling (e.g., RRC, service data adaptation protocol (SDAP), packet data convergence protocol (PDCP)). The CU 160 may be connected to one or more DUs 165 or RUs 170, and the one or more DUs 165 or RUs 170 may host lower protocol layers, such as layer 1 (L1) (e.g., physical (PHY) layer) or L2 (e.g., radio link control (RLC) layer, medium access control (MAC) layer) functionality and signaling, and may each be at least partially controlled by the CU 160. Additionally or alternatively, a functional split of the protocol stack may be employed between the DU 165 and the RU 170, such that the DU 165 may support one or more layers of the protocol stack, and the RU 170 may support one or more different layers of the protocol stack. The DU 165 may support one or more different cells (e.g., via one or more RUs 170). In some cases, the functional split between CU 160 and DU 165 or between DU 165 and RU 170 can be within the protocol layer (e.g., some functions of the protocol layer can be performed by one of CU 160, DU 165, or RU 170, while other functions of the protocol layer are performed by different ones of CU 160, DU 165, or RU 170). CU 160 can be further functionally split into CU control plane (CU-CP) and CU user plane (CU-UP) functions. CU 160 can be connected to one or more DUs 165 via midhaul communication links 162 (e.g., F1, F1-c, F1-u), and DU 165 can be connected to one or more RUs 170 via fronthaul communication links 168 (e.g., open fronthaul (FH) interface). In some examples, midhaul communication link 162 or fronthaul communication link 168 may be implemented based on an interface (e.g., a channel) between layers of a protocol stack supported by respective network entities 105 that communicate via such communication links.

[0066] In some wireless communication systems (e.g., wireless communication system 100), infrastructure and spectrum resources for radio access may support wireless backhaul link capabilities to supplement wired backhaul connections, thereby providing an IAB network architecture (e.g., to core network 130). In some cases, in an IAB network, one or more network entities 105 (e.g., IAB nodes 104) may be partially controlled by each other. One or more IAB nodes 104 may be referred to as donor entities or IAB donors. One or more DUs 165 or one or more RUs 170 may be controlled in part by one or more CUs 160 associated with a donor network entity 105 (e.g., donor base station 140). One or more donor network entities 105 (e.g., IAB donors) may communicate with one or more additional network entities 105 (e.g., IAB nodes 104) via supported access and backhaul links (e.g., backhaul communication links 120). The IAB node 104 may include an IAB mobile terminal (IAB-MT) controlled (e.g., scheduled) by a coupled IAB donor's DU 165. The IAB-MT may include an independent set of antennas for relaying communications with the UE 115, or may share the same antennas of the IAB node 104 (e.g., of the RU 170) for access via the DU 165 of the IAB node 104 (e.g., referred to as a virtual IAB-MT (vIAB-MT)). In some examples, the IAB node 104 may include a DU 165 that supports communication links with additional entities (e.g., IAB node 104, UE 115) within a relay chain or configuration (e.g., downstream) of the access network. In such cases, one or more components of the decomposed RAN architecture (e.g., one or more IAB nodes 104 or components of the IAB node 104) may be configured to operate according to the techniques described herein.

[0067] For example, an access network (AN) or RAN may include an access node (e.g., an IAB host), communications between an IAB node 104, and one or more UEs 115. The IAB host may facilitate a connection between the core network 130 and the AN (e.g., via a wired or wireless connection to the core network 130). That is, the IAB host may refer to a RAN node having a wired or wireless connection to the core network 130. The IAB host may include a CU 160 and at least one DU 165 (e.g., and RU 170), wherein the CU 160 may communicate with the core network 130 via an interface (e.g., a backhaul link). The IAB host and the IAB node 104 may communicate via an F1 interface according to a protocol defining a signaling message (e.g., an F1 AP protocol). Additionally or alternatively, CU 160 may communicate with the core network via an interface (which may be an example of part of a backhaul link) and may communicate with other CUs 160 (e.g., CU 160 associated with an alternative IAB host) via an Xn-C interface (which may be an example of part of a backhaul link).

[0068] An IAB node 104 may refer to a RAN node that provides IAB functionality (e.g., access for UE 115, wireless self-backhaul capabilities, etc.). DU 165 may act as a distributed scheduling node toward child nodes associated with IAB node 104, and IAB-MT may act as a scheduled node toward a parent node associated with IAB node 104. That is, an IAB host may be referred to as a parent node that communicates with one or more child nodes (e.g., an IAB host may relay for UE transmissions through one or more other IAB nodes 104). Additionally or alternatively, depending on the relay chain or configuration of the AN, an IAB node 104 may also be referred to as a parent node or child node of other IAB nodes 104. Thus, the IAB-MT entity of the IAB node 104 may provide a Uu interface for the child IAB node 104 to receive signaling from the parent IAB node 104 , and a DU interface (eg, DU 165 ) may provide a Uu interface for the parent IAB node 104 to signal to the child IAB node 104 or the UE 115 .

[0069] For example, the IAB node 104 may be referred to as a parent node supporting communications for a child IAB node or as a child node associated with the IAB host, or both. The IAB host may include a CU 160 having a wired or wireless connection (e.g., backhaul communication link 120) to the core network 130, and may act as a parent node of the IAB node 104. For example, the DU 165 of the IAB host may relay transmission to the UE 115 via the IAB node 104, or may directly signal the transmission to the UE 115, or both. The CU 160 of the IAB host may signal the establishment of a communication link to the IAB node 104 via the F1 interface, and the IAB node 104 may schedule transmission (e.g., transmission relayed from the IAB host to the UE 115) via the DU 165. That is, data may be relayed to and from the IAB node 104 via signaling via the NR Uu interface to the MT of the IAB node 104. Communications with the IAB node 104 may be scheduled by the DU 165 of the IAB host, and communications with the IAB node 104 may be scheduled by the DU 165 of the IAB node 104 .

[0070] In the case where the techniques described herein are applied to the context of a disaggregated RAN architecture, one or more components of the disaggregated RAN architecture may be configured to support OTA aggregated federated learning with non-connected devices as described herein. For example, some operations described as being performed by a UE 115 or a network entity 105 (e.g., a base station 140) may additionally or alternatively be performed by one or more components of the disaggregated RAN architecture (e.g., an IAB node 104, a DU 165, a CU 160, a RU 170, a RIC 175, a SMO 180).

[0071] UE 115 may include or may be referred to as a mobile device, a wireless device, a remote device, a handheld device, or a subscriber device, or some other suitable term, where a "device" may also be referred to as a unit, a station, a terminal, or a client, etc. UE 115 may also include or may 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, UE 115 may include or may 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, etc., which may be implemented in various objects such as appliances or vehicles, meters, etc.

[0072] The UE 115 described herein may be capable of communicating with various types of devices such as other UEs 115 which may sometimes act as relays, as well as network entities 105 and network equipment including macro eNBs or gNBs, small cell eNBs or gNBs, or relay base stations, etc. Figure 1 shown.

[0073] The UE 115 and the network entity 105 may wirelessly communicate with each other via one or more communication links 125 (e.g., access links) using resources associated with one or more carriers. The term "carrier" may refer to a collection of RF spectrum resources having a physical layer structure defined for supporting the communication link 125. For example, a carrier for the communication link 125 may include a portion of an RF spectrum band (e.g., a bandwidth portion (BWP)) that operates 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 for coordinating carrier operations, user data, or other signaling. The wireless communication system 100 may support communications with the UE 115 using carrier aggregation or multi-carrier operation. Depending on the carrier aggregation configuration, the UE 115 may be configured to have multiple downlink component carriers and one or more uplink component carriers. Carrier aggregation may be used for both frequency division duplex (FDD) and time division duplex (TDD) component carriers. Communication between the network entity 105 and other devices may refer to communication between these devices and any part (e.g., entity, sub-entity) of the network entity 105. For example, the terms "send", "receive", or "communicate" when referring to the network entity 105 may refer to any part of the network entity 105 (e.g., base station 140, CU 160, DU 165, RU 170) of the RAN communicating with another device (e.g., directly or via one or more other network entities 105).

[0074] In some examples, such as in a carrier aggregation configuration, a carrier may also have acquisition signaling or control signaling that coordinates the operation of other carriers. A carrier may be associated with a frequency channel (e.g., an Evolved Universal Mobile Telecommunications System Terrestrial Radio Access (E-UTRA) Absolute RF Channel Number (EARFCN)) and may be identified according to a channel raster for discovery by UE 115. A carrier may operate in a standalone mode, in which case initial acquisition and connection may be performed by UE 115 via the carrier, or a carrier may operate in a non-standalone mode, in which case the connection is anchored using a different carrier (e.g., a different carrier of the same or different radio access technology).

[0075] The communication link 125 shown in the wireless communication system 100 may include downlink transmissions (e.g., forward link transmissions) from the network entity 105 to the UE 115, uplink transmissions (e.g., return link transmissions) from the UE 115 to the network entity 105, or both, as well as other transmission configurations. A carrier may carry either downlink communications or uplink communications (e.g., in FDD mode), or may be configured to carry both downlink communications and uplink communications (e.g., in TDD mode).

[0076] A carrier may be associated with a particular bandwidth of the RF spectrum, and in some examples, the carrier bandwidth may be referred to as a "system bandwidth" of the carrier or wireless communication system 100. For example, the carrier bandwidth may be one of a set of bandwidths (e.g., 1.4, 3, 5, 10, 15, 20, 40, or 80 megahertz (MHz)) of a carrier of a particular radio access technology. A device (e.g., a network entity 105, a UE 115, or both) of the wireless communication system 100 may have a hardware configuration that supports communications using a particular carrier bandwidth, or may be capable of being configured to support communications using one of a set of carrier bandwidths. In some examples, the wireless communication system 100 may include a network entity 105 or a UE 115 that supports concurrent communications using carriers associated with multiple carrier bandwidths. In some examples, each served UE 115 may be configured to operate using a portion (e.g., a subband, a BWP) or all of a carrier bandwidth.

[0077] The signal waveform transmitted via the carrier may include multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques, such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread OFDM (DFT-S-OFDM)). In a system using MCM technology, a resource element may refer to a symbol period (e.g., the duration of a modulation symbol) and a resource of a subcarrier, in which case the symbol period and the subcarrier spacing may be inversely related. The number of bits carried by each resource element may depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both), so that a relatively high number of resource elements (e.g., in the transmission duration) and a relatively high order modulation scheme may correspond to a relatively high communication rate. Wireless communication resources may refer to a combination of RF spectrum resources, time resources, and spatial resources (e.g., spatial layers or beams), and the use of multiple spatial resources may increase the data rate or data integrity used for communication with UE 115.

[0078] One or more parameter sets for a carrier may be supported, and the parameter sets may include subcarrier spacing (Δf) and cyclic prefix. A carrier may be divided into one or more BWPs with the same or different parameter sets. In some examples, a UE 115 may be configured with multiple BWPs. In some examples, a single BWP for a carrier may be active at a given time, and communications of a UE 115 may be constrained to one or more active BWPs.

[0079] The time interval for the network entity 105 or the UE 115 may be expressed as a multiple of a basic time unit, which may be, for example, a sampling period T s =1 / (△f max ·N f ) seconds, for which △f max It can represent the supported subcarrier spacing, and N f The supported discrete Fourier transform (DFT) size may be indicated. The time intervals of the communication resources 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).

[0080] Each frame may include multiple consecutively numbered subframes or time slots, and each subframe or time slot may have the same duration. In some examples, the frame may be divided into subframes (e.g., in the time domain), and each subframe may be further divided into a certain number of time slots. Alternatively, each frame may include a variable number of time slots, and the number of time slots may depend on the subcarrier spacing. Each time slot may include a certain number of symbol periods (e.g., depending on the length of the cyclic prefix appended in front of each symbol period). In some wireless communication systems 100, the time slot may be further divided into a plurality of micro time slots associated with one or more symbols. Excluding the cyclic prefix, each symbol period may be associated with one or more (e.g., N f The duration of a symbol period may depend on the subcarrier spacing or the operating frequency band.

[0081] A subframe, slot, mini-slot, or symbol may be the smallest scheduling unit (e.g., in the time domain) of the wireless communication system 100 and may be referred to as a transmit 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 communication system 100 may be dynamically selected (e.g., in a burst of a shortened TTI (sTTI)).

[0082] According to various techniques, physical channels may be multiplexed using carriers for communication. Physical control channels and physical data channels may be multiplexed for signaling via downlink carriers, for example, using one or more of a time division multiplexing (TDM) technique, a frequency division multiplexing (FDM) technique, or a hybrid TDM-FDM technique. A control region (e.g., a control resource set (CORESET)) of a physical control channel may be defined by a set of symbol periods and may extend across a system bandwidth of a carrier or a subset of that system bandwidth. One or more control regions (e.g., CORESETs) may be configured for a set of UEs 115. For example, one or more UEs in UE 115 may monitor or search a control region to obtain control information according to one or more search space sets, and each search space set may include one or more control channel candidates in one or more aggregation levels arranged in a cascaded manner. The aggregation level of a control channel candidate may refer to the amount of control channel resources (e.g., control channel elements (CCEs)) associated with 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 transmitting control information to multiple UEs 115 , and a UE-specific search space set for transmitting control information to a specific UE 115 .

[0083] The network entity 105 may provide communication coverage via one or more cells (e.g., macro cells, small cells, hot spots, or other types of cells, or any combination thereof). The term "cell" may refer to a logical communication entity used to communicate with the network entity 105 (e.g., using a carrier) and may be associated with an identifier used to distinguish adjacent cells (e.g., a physical cell identifier (PCID), a virtual cell identifier (VCID), or other cell identifier). In some examples, a cell may also refer to a coverage area 110 or a portion of a coverage area 110 (e.g., a sector) on which the logical communication entity operates. Depending on various factors such as the capabilities of the network entity 105, such cells may range from smaller areas (e.g., structures, subsets of structures) to larger areas. For example, a cell may be or may include a building, a subset of a building, or an external space between or overlapping coverage areas 110, and the like.

[0084] A macro cell generally covers a relatively large geographic area (e.g., several kilometers in radius) and may allow unrestricted access to a UE 115 that has a service subscription with a network provider that supports the macro cell. A small cell may be associated with a lower power network entity 105 (e.g., a lower power base station 140) than a macro cell, and the small cell may operate using the same or different (e.g., licensed, unlicensed) frequency band as the macro cell. A small cell may provide unrestricted access to a UE 115 that has a service subscription with a network provider, or may provide restricted access to a UE 115 associated with a small cell (e.g., a UE 115 in a closed subscriber group (CSG), a UE 115 associated with a user in a home or office). A network entity 105 may support one or more cells, and may also use one or more component carriers to support communications via one or more cells.

[0085] In some examples, a carrier may support multiple cells, and different cells may be configured according to different protocol types (e.g., MTC, narrowband IoT (NB-IoT), enhanced mobile broadband (eMBB)) that may provide access to different types of devices.

[0086] In some examples, the network entities 105 (e.g., base stations 140, RUs 170) may be mobile and thus provide communication coverage for mobile coverage areas 110. In some examples, different coverage areas 110 associated with different technologies may overlap, but the different coverage areas 110 may be supported by the same network entity 105. In some other examples, overlapping coverage areas 110 associated with different technologies may be supported by different network entities 105. The wireless communication system 100 may include, for example, a heterogeneous network in which different types of network entities 105 provide coverage for various coverage areas 110 using the same or different radio access technologies.

[0087] The wireless communication system 100 may support synchronous or asynchronous operation. For synchronous operation, the network entities 105 (e.g., base stations 140) may have similar frame timing, and transmissions from different network entities 105 may be approximately aligned in time. For asynchronous operation, the network entities 105 may have different frame timing, and in some examples, transmissions from different network entities 105 may not be aligned in time. The techniques described herein may be used for either synchronous operation or asynchronous operation.

[0088] Some UEs 115, such as MTC or IoT devices, may be low-cost or low-complexity devices and may provide automated communication between machines (e.g., via machine-to-machine (M2M) communication). M2M communication or MTC may refer to data communication technology that allows devices to communicate with each other or with a network entity 105 (e.g., base station 140) without human intervention. In some examples, M2M communication or MTC may include communication from devices with integrated sensors or meters to measure or capture information and relay such information to a central server or application that uses the information or presents the information to a person interacting with the application. Some UEs 115 may be designed to collect information or implement automated behavior of machines or other devices. Examples of applications for MTC devices include: smart metering, inventory monitoring, water level monitoring, equipment monitoring, health care monitoring, wildlife monitoring, weather and geographic event monitoring, queue management and tracking, remote security sensing, physical access control, and transaction-based commercial charging.

[0089] Some UEs 115 may be configured to employ a mode of operation that reduces power consumption, such as half-duplex communication (e.g., a mode that supports unidirectional communication via transmission or reception but does not transmit and receive concurrently). In some examples, half-duplex communication may be performed with a reduced peak rate. Other power conservation techniques for UEs 115 include entering a power-saving deep sleep mode when not engaged in active communications, operating using limited bandwidth (e.g., according to narrowband communications), or a combination of these techniques. For example, some UEs 115 may be configured to operate using a narrowband protocol type that is 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 a carrier, or outside a carrier.

[0090] The wireless communication system 100 may be configured to support ultra-reliable communication or low-latency communication or various combinations thereof. For example, the wireless communication system 100 may be configured to support ultra-reliable low-latency communication (URLLC). UE 115 may be designed to support ultra-reliable or low-latency or critical functions. Ultra-reliable communication may include private communication or group communication, and may be supported by one or more services (such as push-to-talk, video, or data). Support for ultra-reliable, low-latency functions may include prioritization of services, 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.

[0091] In some examples, a UE 115 may be configured to support communication directly with other UEs 115 via a device-to-device (D2D) communication link 135 (e.g., according to a peer-to-peer (P2P), D2D, or sidelink protocol). In some examples, one or more UEs 115 in a group that are performing D2D communication may be within a coverage area 110 of a network entity 105 (e.g., a base station 140, a RU 170), which may support aspects of such D2D communication configured (e.g., scheduled) by the network entity 105. In some examples, one or more UEs 115 in such a group may be outside of the coverage area 110 of the network entity 105, or may otherwise be unable or not configured to receive transmissions from the network entity 105. In some examples, a group of UEs 115 communicating via D2D communication may support a one-to-many (1:M) system, in which each UE 115 transmits to each of the other UEs 115 in the group. In some examples, network entity 105 may facilitate scheduling of resources for D2D communications. In some other examples, D2D communications may be performed between UEs 115 without involving network entity 105.

[0092] In some systems, the D2D communication link 135 can be an example of a communication channel (such as a sidelink communication channel) between vehicles (e.g., UE 115). In some examples, the vehicles can communicate using vehicle-to-everything (V2X) communication, vehicle-to-vehicle (V2V) communication, or some combination of these items. The vehicles can signal information related to traffic conditions, signal scheduling, weather, safety, emergency situations, or any other information related to the V2X system. In some examples, vehicles in the V2X system can communicate with roadside infrastructure (such as roadside units), or communicate with the network via one or more network nodes (e.g., network entity 105, base station 140, RU 170) using vehicle-to-network (V2N) communication, or both.

[0093] 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 a 5G core (5GC), which may include at least one control plane entity for managing access and mobility (e.g., a mobility management entity (MME), an access and mobility management function (AMF)) and at least one user plane entity for routing packets or interconnecting to an external network (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function (UPF)). The control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management of UE 115 served by a network entity 105 (e.g., a base station 140) associated with the core network 130. User IP packets may be delivered via a user plane entity, which may provide IP address allocation and other functions. The user plane entity may be connected to the IP service 150 of one or more network operators. IP services 150 may include access to the Internet, an intranet, an IP Multimedia Subsystem (IMS), or packet-switched streaming services.

[0094] The wireless communication system 100 may operate using one or more frequency bands that may be in the range of 300 megahertz (MHz) to 300 gigahertz (GHz). Generally speaking, the region from 300 MHz to 3 GHz is referred to as the ultra-high frequency (UHF) region or decimeter band because the wavelength ranges from about one decimeter to one meter in length. UHF waves may be blocked or redirected by buildings and environmental features (which may be referred to as clusters), but these waves may be sufficient to penetrate structures so that macro cells provide services to UEs 115 located indoors. Communications using UHF waves may be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers) compared to communications using lower frequencies and longer waves in the high frequency (HF) or very high frequency (VHF) portions of the spectrum below 300 MHz.

[0095] The wireless communication system 100 may also operate in a super high frequency (SHF) region (also known as a centimeter band) in the range of 3 GHz to 30 GHz or in an extremely high frequency (EHF) region (e.g., 30 GHz to 300 GHz) (also known as a millimeter band) using a spectrum. In some examples, the wireless communication system 100 may support millimeter wave (mmW) communication between UE 115 and network entity 105 (e.g., base station 140, RU 170), and the EHF antenna of the corresponding device may be smaller and closer than the UHF antenna. In some examples, such technology may facilitate the use of antenna arrays within the device. However, the propagation of EHF transmissions may be affected by greater attenuation and shorter range than SHF or UHF transmissions. The technology disclosed herein may be adopted across transmissions using one or more different frequency regions, and the use of frequency bands specified across these frequency regions may vary by country or regulatory agency.

[0096] The wireless communication system 100 may utilize licensed and unlicensed RF spectrum bands. For example, the wireless communication system 100 may use unlicensed bands (such as the 5 GHz Industrial, Scientific, and Medical (ISM) band) to employ license assisted access (LAA), LTE unlicensed (LTE-U) radio access technology, or NR technology. When operating using unlicensed RF spectrum bands, devices such as network entity 105 and UE 115 may employ carrier sensing for conflict detection and avoidance. In some examples, operations using unlicensed bands may be based on carrier aggregation configuration (e.g., LAA) in conjunction with component carriers operating using licensed bands. Operations using unlicensed spectrum may include downlink transmissions, uplink transmissions, P2P transmissions, or D2D transmissions, and the like.

[0097] The network entity 105 (e.g., base station 140, RU 170) or UE 115 may be equipped with multiple antennas that can be used to employ technologies such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communication, or beamforming. The antennas of the network entity 105 or UE 115 may be located in one or more antenna arrays or antenna panels, which may support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays may be co-located at an antenna assembly, such as an antenna tower. In some examples, antennas or antenna arrays associated with the network entity 105 may be located at different geographical locations. The network entity 105 may include an antenna array having a collection of multiple rows and columns of antenna ports that the network entity 105 can use to support beamforming for communications with the UE 115. Similarly, the UE 115 may include one or more antenna arrays that may support various MIMO or beamforming operations. Additionally or alternatively, the antenna panel may support RF beamforming for signals sent via the antenna ports.

[0098] The network entity 105 or UE 115 may use MIMO communication to utilize multipath signal propagation and improve spectral efficiency by sending or receiving multiple signals via different spatial layers. Such technology may be referred to as spatial multiplexing. Multiple signals may be, for example, sent by a transmitting device via different antennas or different combinations of antennas. Similarly, 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 information associated with the same data stream (e.g., the same codeword) or different data streams (e.g., different codewords). Different spatial layers may be associated with different antenna ports for channel measurement and reporting. MIMO technologies include: single-user MIMO (SU-MIMO), in which multiple spatial layers are sent to the same receiving device; and multi-user MIMO (MU-MIMO), in which multiple spatial layers are sent to multiple devices.

[0099] Beamforming (which may also be referred to as spatial filtering, directional transmission, or directional reception) is a signal processing technique that may be used at a transmitting device or a receiving device (e.g., a network entity 105, a UE 115) to shape or direct an antenna beam (e.g., a transmit beam, a receive beam) along a spatial path between the transmitting device and the receiving device. Beamforming may be achieved by combining signals conveyed via antenna elements of an antenna array so that some signals propagating along a particular orientation relative to the antenna array experience constructive interference, while other signals experience destructive interference. Adjustments to signals conveyed via antenna elements may include the transmitting device or the receiving device applying an amplitude offset, a phase offset, or both to signals carried via antenna elements associated with the device. Adjustments associated with each of these antenna elements may be defined by a set of beamforming weights associated with a particular orientation (e.g., relative to the antenna array of the transmitting device or the receiving device or relative to some other orientation).

[0100] The network entity 105 or the UE 115 may use beam scanning techniques as part of a beamforming operation. For example, the network entity 105 (e.g., base station 140, RU 170) may use multiple antennas or antenna arrays (e.g., antenna panels) to perform beamforming operations for directional communication with the UE 115. Some signals (e.g., synchronization signals, reference signals, beam selection signals, or other control signals) may be sent multiple times by the network entity 105 along different directions. For example, the network entity 105 may send signals according to different sets of beamforming weights associated with different transmission directions. Transmissions along different beam directions may be used to identify (e.g., by a transmitting device such as the network entity 105, or by a receiving device such as the UE 115) the beam direction for later transmission or reception by the network entity 105.

[0101] Some signals (such as data signals associated with a particular receiving device) may be sent by a transmitting device (e.g., transmitting network entity 105, transmitting UE 115) along a single beam direction (e.g., a direction associated with a receiving device such as receiving network entity 105 or receiving UE 115). In some examples, a beam direction associated with transmission along a single beam direction may be determined based on signals sent along one or more beam directions. For example, UE 115 may receive one or more of the signals sent by network entity 105 along different directions, and may report to network entity 105 an indication of the signal received by UE 115 with the highest signal quality or other acceptable signal quality.

[0102] In some examples, transmission by a device (e.g., by network entity 105 or UE 115) may be performed using multiple beam directions, and the device may use a combination of digital precoding or beamforming to generate a combined beam for transmission (e.g., from network entity 105 to UE 115). UE 115 may report feedback indicating precoding weights for one or more beam directions, and the feedback may correspond to a set of configured beams across a system bandwidth or one or more subbands. Network entity 105 may send a reference signal (e.g., a cell-specific reference signal (CRS), a channel state information reference signal (CSI-RS)), which may be precoded or not precoded. UE 115 may provide feedback for beam selection, which may be a precoding matrix indicator (PMI) or codebook-based feedback (e.g., a multi-panel codebook, a linear combination codebook, a port selection codebook). Although these techniques are described with reference to signals sent along one or more directions by a network entity 105 (e.g., base station 140, RU 170), UE 115 may use similar techniques to send signals multiple times along different directions (e.g., to identify a beam direction for subsequent transmission or reception by UE 115), or to send signals along a single direction (e.g., to send data to a receiving device).

[0103] A receiving device (e.g., UE 115) may perform receiving operations according to multiple receiving configurations (e.g., directional listening) when receiving various signals (such as synchronization signals, reference signals, beam selection signals, or other control signals) from a receiving device (e.g., network entity 105). For example, the receiving device may perform reception according to multiple receiving directions by: receiving via different antenna subarrays, processing received signals according to different antenna subarrays, receiving according to different receiving beamforming weight sets (e.g., different directional listening weight sets) applied to signals received at multiple antenna elements of an antenna array, or processing received signals according to different receiving beamforming weight sets applied to signals received at multiple antenna elements of an antenna array, any of which may be referred to as "listening" according to different receiving configurations or receiving directions. In some examples, the receiving device may use a single receiving configuration to receive along a single beam direction (e.g., when receiving a data signal). A single receive configuration may be aligned along a beam direction determined based on listening according to different receive configuration directions (e.g., a beam direction determined to have the highest signal strength, highest signal-to-noise ratio (SNR), or other acceptable signal quality based on listening according to multiple beam directions).

[0104] The wireless communication system 100 may be a packet-based network that operates according to a layered protocol stack. In the user plane, communications at the bearer or PDCP layer may be IP-based. The RLC layer may perform packet segmentation and reassembly to communicate via logical channels. The MAC layer may perform priority processing and multiplexing of logical channels into transport channels. The MAC layer may also implement error detection techniques, error correction techniques, or both to support retransmission to improve link efficiency. In the control plane, the RRC layer may provide the establishment, configuration, and maintenance of an RRC connection that supports a radio bearer for user plane data between the UE 115 and the network entity 105 or the core network 130. The PHY layer may map a transport channel to a physical channel.

[0105] UE 115 and network entity 105 may support retransmission of data to increase the likelihood that the data is successfully received. Hybrid automatic repeat request (HARQ) feedback is a technique for increasing the likelihood of correctly receiving data via a communication link (e.g., communication link 125, D2D communication link 135). HARQ may include a combination of error detection (e.g., using a cyclic redundancy check (CRC)), forward error correction (FEC), and retransmission (e.g., automatic repeat request (ARQ)). HARQ may improve throughput at the MAC layer under poor radio conditions (e.g., low signal-to-noise ratio conditions). In some examples, a device may support same-slot HARQ feedback, in which case the device may provide HARQ feedback in a specific time slot for data received via a previous symbol in the time slot. In some other examples, the device may provide HARQ feedback in a subsequent time slot or according to some other time interval.

[0106] Some wireless communication systems may implement federated learning to train machine learning models. Federated learning may train a machine learning model in a distributed model, such as a global machine learning model. Using federated learning, one or more clients (e.g., UE 115 or other devices) may employ the same machine learning model or machine learning model structure. The machine learning model or machine learning model structure may be associated with a common task associated with or performed by UE 115. UE 115 may train model parameters of a machine learning model independently or locally based on corresponding observations or data collection.

[0107] A UE 115 participating in or performing federated learning may periodically transmit information for updated model parameters to a server. The server may be an example of a network entity 105 or a component thereof as described herein. In some examples, UE 115 may be an example of a server, or UE 115 may execute the techniques of a server. The server may compile information from UE 115 to generate global or aggregated model parameters. In some examples, this may be referred to as updating a machine learning model, where information reported from UE 115 is used to train a machine learning model at the server, and an updated version of the machine learning model is generated using the collected data. The server may transmit or forward the updated model parameters, and UE 115 may update the local machine learning model with the updated model parameters. For example, the server may send an indication of an updated machine learning model to UE 115. In some cases, the server may transmit an updated machine learning model to a UE 115 participating in a federated learning procedure, to a UE 115 not participating in a federated learning procedure, or to both.

[0108] In some examples, the server may implement stochastic gradient descent (SGD) model training. For example, at iteration or epoch i, the server may update the global model parameters according to equation (1), where θ i is the vector of model parameters at iteration i, is the gradient vector calculated by the nth UE 115 at iteration i, N UEs is the number of UEs 115 participating in the i-th round, and n i is the learning rate applied by the server. In some examples, if each UE 115 performs local updates using the same amount of training data, then equation (1) may be applied. In some examples, local training may be performed on data that is unevenly distributed (e.g., in number) among UEs 115, in which case the server may compute a weighted sum of the gradient values ​​(e.g., rather than an average).

[0109]

[0110] A machine learning model may have many parameters, such as tens of thousands or hundreds of thousands of parameters. If a UE 115 were to send the gradients of the model parameters on an uplink shared channel resource or an uplink control channel resource, the payload of the gradient transmission may incur very large overhead when a large number of UEs 115 participate in the federated learning procedure.

[0111] To reduce overhead, some systems may perform OTA aggregation (e.g., OTA simulation aggregation). Each gradient of a model parameter is mapped to a single resource (e.g., a single resource element), and each UE 115 sends a calculated simulated gradient value over the resource. For example, each UE 115 may determine the gradient of a model parameter, and each UE 115 may send a calculated simulated gradient of the model parameter on the same resource. A receiver (e.g., a server) may observe a superposition of the sent gradient values. The superposition of the sent gradient values ​​may provide an indication of an average or sum of the gradient values ​​from each of the UEs 115 participating in the federated learning procedure. Thus, the server may obtain the gradient of the model parameter using much fewer resources than if each UE 115 reported a gradient on a corresponding set of resources.

[0112] Federated learning with OTA analog aggregation can improve the scalability of a large number of UEs 115 that contribute to a federated learning session by using the superposition properties of the wireless medium. The UE 115 that transmits according to the OTA aggregation may be configured with power control instructions because if the value sent by the UE 115 is first normalized, it may be necessary to receive the contribution of the UE 115 at the same power or approximately the same power to avoid the near-far effect. The near-far effect can make the received aggregate signal more representative of the UE 115 experiencing a smaller path loss than other UEs, such as based on distance, channel quality, or line of sight conditions. UEs 115 that experience small path losses (e.g., UEs 115 close to the server) can reduce the transmit power compared to UEs 115 with high path losses (e.g., UEs 115 farther away from the server). By implanting this power control, the gradient can be received at the server with the same average power. In some examples, each UE 115 may be configured with the same nominal transmit power to ensure that the gradient is received at the server with the same average power.

[0113] For uplink power control, the uplink shared channel transmit power for transmission type j can be provided by the following equation (2). In equation (2), P CMAX，f，c (i) may be the pre-configured maximum UE transmit power for PUSCH transmission on BWPb over carrier f and cell c for transmission opportunity i. O-PUSCH,b,f，c (j) may be the nominal UE transmit power per resource block for transmission type j with 15kHz SCS, which may be a value calculated based on a cell-specific nominal power value (e.g., provided via system information or RRC signaling) and in some cases based on a UE-specific offset (e.g., provided via RRC signaling). It can be the number of resource blocks allocated for transmission type j. b,f,c It can be a fractional power control parameter, PL b，f，c (q d ) can be used with the identifier q d The link path loss measured by the reference signal, Δ TF,b,vc (i) may be an MCS-related offset, and f b，f，c (i, l) may be a closed-loop power control component corresponding to power control adjustment state l. In some examples, when f b，f，c When (i, l)=0, power control may be open loop (eg, implemented without assistance from the network entity 105).

[0114]

[0115] Sidelink power control may be similar to uplink power control. However, sidelink power control may be open loop, rather than open loop or closed loop. In some examples, sidelink power control may take into account downlink path loss (e.g., if the transmitting UE 115 is under network coverage) and sidelink path loss.

[0116] The present power control framework may be implemented when each UE 115 contributing to a federated learning session is in an RRC connected state (e.g., via an uplink to a network entity 105 or via a sidelink to another UE 115 or a roadside unit). A server (e.g., a network entity 105, a UE 115, or a roadside unit) may provide explicit instructions to each UE 115 by configuring a nominal receive power for open-loop power control and, in some cases, additional closed-loop adaptation commands.

[0117] The wireless communication system 100 may support non-connected UEs 115 to participate in or perform federated learning using OTA aggregation. For federated learning over an uplink, since federated learning updates are independent of other communication activities, UEs 115 that are not in RRC connection mode may still participate in federated learning training rounds. For federated learning over a side link, the mobility of UE 115 may result in significant overhead for establishing and releasing RRC connections. Therefore, using non-RRC connected UEs 115 for federated learning may be similarly efficient. For example, if a fixed-position RSU is a server that receives contributions from a passing vehicle, establishing and tracking an RRC connection with a UE 115 that facilitates each training round may have a large overhead. Using a non-coherent OTA federated learning scheme, the strict timing, phase, and frequency requirements of coherent OTA federated learning are not required (e.g., satisfied via RRC connection mode). However, these techniques may be similarly implemented for coherent OTA federated learning.

[0118] In some examples, a UE 115 participating in a federated learning training round may send a gradient value to a server. In some cases, the UE 115 may normalize the gradient value and send the normalized gradient value to the server. In some cases, the UE 115 may still collect data, train a machine learning model to obtain model parameters, and receive updates to the global machine learning model even if it does not participate (e.g., send gradient values) in a training round or a federated learning procedure. In some cases, participating in a federated learning training round may be referred to as executing a federated learning procedure.

[0119] For example, the wireless communication system 100 can implement a procedure to invite or request non-connected UEs 115 to participate in a federated learning training round. The procedure can align the UEs 115 to operate on the same model and apply the same federated learning aggregation scheme.

[0120] The wireless communication system 100 may also support power control techniques to support non-RRC connected UEs 115 to participate in the OTA federated learning procedure. In some examples, the server may not be aware of the path loss conditions experienced by the contributing UE 115 to determine the target receive power (e.g., nominal transmit power) for the participating UE 115. Additionally or alternatively, the server may not be aware of the number of UEs 115 participating in the federated learning training round. For example, facilitating a federated learning training round when not in RRC connected mode may be optional for the UE 115, and a UE 115 that is not in RRC connected mode may participate without explicitly announcing participation. If the server or server receiver has a limited dynamic range, the aggregated power may overwhelm the server in the event of a very large number of UEs 115 participating. The power control techniques described herein may enable a non-RRC connected UE 115 to determine whether to participate or determine the transmit power for participation to avoid complicating power control at the server. For example, the wireless communication system 100 may support a power control mechanism that allows non-RRC connected UEs 115 to participate in federated learning training rounds, adapt to varying path loss conditions, and adapt to varying numbers of UEs 115 between training rounds.

[0121] Figure 2 An example of a wireless communication system 200 that supports OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is illustrated. The wireless communication system 200 may include UE 115-a, UE 115-b, and UE 115-c, each of which may be as described in reference Figure 1 The wireless communication system 200 may include a network entity 105-a, which may be as described in reference Figure 1 Examples of network entities described. In some examples, network entity 105-a may be an example of a server for a federated learning procedure as described herein. Additionally or alternatively, UE 115 may be an example of a server.

[0122] The wireless communication system 200 may support a federated learning procedure to train a machine learning model. For the federated learning procedure, multiple clients (UE 115) may employ the same machine learning model structure to locally train model parameters for the machine learning model based on observations or data. The UE 115 may transmit information for updated model parameters to a server (such as a network entity 105-a), which compiles information from all UEs 115 to determine global or aggregated model parameters. The network entity 105-a may then transmit the updated model parameters or the updated machine learning model to the client. The federated learning procedure may include multiple training rounds, each of which may be dynamically scheduled by the network entity 105-a. The network entity 105-a may receive information for model parameters, update the machine learning model, and send information for an updated version of the machine learning model to the client for another round of the federated learning procedure.

[0123] The wireless communication system 200 may use OTA aggregation techniques to reduce the overhead of UE 115 reporting information for updated or locally trained model parameters. A machine learning model may have a large number of parameters (e.g., tens of thousands or hundreds of thousands of parameters), and each UE 115 individually reporting the machine learning model parameters may have significant overhead. To reduce overhead, the UE 115 may send gradients of the locally trained model parameters on the same set of resources, such as OTA aggregated resources 215. For example, each gradient may be mapped to a corresponding resource element, and each UE 115 may send a gradient value 210 via the OTA aggregated resource 215. In an example, UE 115-a may send a gradient value 210-a on the OTA aggregated resource 215, UE 115-b may send a gradient value 210-b on the OTA aggregated resource 215, and UE 115-c may send a gradient value 210-c on the OTA aggregated resource 215.

[0124] A receiver (such as network entity 105-a or UE 115 acting as a server) may observe a superposition of gradient values ​​210. In some examples, the superposition of gradient values ​​210 may correspond to an average of the gradient values ​​of the model parameters. Network entity 105-a may determine the average of the model parameters from all clients (e.g., UE 115) participating in the federated learning procedure and use the average of the model parameters to train and update the machine learning model.

[0125] Techniques described herein support federated learning procedures for non-connected UEs 115 using OTA aggregation. These techniques may enable a non-connected UE 115 (such as a UE 115 operating in an RRC idle or RRC inactive state) to participate in one or more training rounds of federated learning using OTA aggregation without switching to an RRC connected mode. For example, UE 115-a may be an example of a UE 115 operating in a non-connected state (such as an RRC inactive or RRC idle state). These techniques may include a power control procedure for the non-connected UE 115 to send a gradient value 210 on an OTA aggregated resource 215 so that the gradient value 210 is received at the network entity 105-a at the same power.

[0126] The network entity 105-a may send an invitation message 205 before a federated learning training round to invite (e.g., request) the UE 115 to participate in an upcoming training round of the federated learning procedure. The invitation message 205 may invite or request the UE 115 to participate in an upcoming round of the federated learning procedure. The invitation message 205 may invite both UEs 115 in a connected mode (e.g., RRC connected) and a non-connected mode (e.g., RRC inactive or RRC idle). In some examples, such as for uplink federated learning, the invitation message 205 may be sent via downlink control information. For example, the invitation message may be referred to as an invitation downlink control information or control signal requesting the UE 115 to perform a federated learning procedure. In some cases, the invitation message 205 may be sent during a paging occasion or paging frame. In some examples, the invitation message 205 may be used to broadcast a global update of a machine learning model. Additionally or alternatively, the server may broadcast a global update or version of the machine learning model separately (e.g., to participating UEs 115 or non-participating UEs 115).

[0127] The invitation message 205 may indicate a configuration or set of parameters for the federated learning procedure. For example, the invitation message may indicate a machine learning model, an OTA aggregation scheme, resources for OTA aggregation, etc. For example, a UE 115 participating in the federated learning procedure may update a local model starting from a global model based on results (e.g., updates) from a previous training round. For example, the network entity 105-a may broadcast an initial global machine learning model and updates (e.g., updated parameters) to the machine learning model based on a previous training round.

[0128] To avoid mismatches in the event that UE 115 participates or joins a training round but uses a different or outdated version of a machine learning model, the invite message 205 may indicate the machine learning model to use in the next training round. In some cases, the invite message 205 may indicate a machine learning model from a set of machine learning models. For example, UE 115 may be configured (e.g., pre-configured via RRC) with a set of machine learning models, and an identifier in the invite message 205 may indicate a machine learning model to use from the set of machine learning models. In some cases, the invite message 205 may include a pointer to an entry of a table (e.g., a lookup table) configured at UE 115.

[0129] Additionally or alternatively, the invitation message 205 may explicitly indicate the model structure of the machine learning model. For example, the invitation message 205 may indicate the number of layers of the machine learning model, the size of each layer, the order of the layers, or the connectivity of the layers, or any combination thereof.

[0130] In some examples, the invitation message 205 may indicate a version of the machine learning model. For example, when the network entity 105-a updates the global model and broadcasts a message of the result (e.g., an updated machine learning model), the network entity 105-a may include a version identifier of the broadcast message. In some examples, the invitation message 205 may include a version identifier of the most recently broadcast model update. UEs 115 whose last received version identifier does not match the latest version model identifier may not participate in upcoming training rounds. UEs 115 may receive the latest version in a subsequent broadcast message and participate after receiving the current version of the updated machine learning model.

[0131] In some examples, the invitation message 205 may indicate a federated learning scheme for the federated learning procedure. In some examples, the invitation message 205 may indicate an OTA aggregation scheme. For example, the invitation message 205 may indicate a resource through which the UE update is sent. For example, the invitation message 205 may indicate an OTA aggregate resource 215. In some examples, the federated learning scheme may be indicated from a set of federated learning schemes (e.g., pre-configured via RRC signaling). In some examples, the OTA aggregation scheme may be indicated from a set of OTA aggregation schemes (e.g., pre-configured via RRC signaling).

[0132] In some examples, the invitation message may indicate a power control instruction for reporting a gradient value of a model parameter. Indicating a power control instruction may enable a non-RRC connected UE 115 to report a gradient value so that the gradient value is received at the same power as other UEs 115 participating in the federated learning procedure. The server may not be aware of the path loss of a non-RRC connected mode UE 115 participating in the federated learning procedure to provide a power control recommendation.

[0133] In some examples, UEs 115 within a certain range of the network entity 105-a may participate in the federated learning procedure. For example, the network entity 105-a may identify where the farthest contributing UE 115 is located and what the received power from the farthest contributing UE 115 would be if transmitting at full power. In some cases, the network entity 105-a may determine the received power of a worst-case UE 115 (e.g., a UE 115 at the farthest distance from the network entity 105-a that may still participate in the federated learning procedure). In some cases, the network entity 105-a may then determine a nominal transmit power for the UEs 115 participating in the federated learning procedure.

[0134] For example, the invitation message 205 may indicate a distance (e.g., a maximum distance) from the network entity 105-a, and UEs 115 within the distance from the network entity 105-a may participate in the federated learning procedure or the next round of the federated learning procedure. In some cases, the network entity 105-a may indicate a larger distance to enable a large number of UEs 115 to participate. In some examples, the network entity 105-a may indicate a short distance to avoid limiting power due to the worst case UEs 115 (e.g., UEs 115 that will contribute with very low transmit power).

[0135] In some examples, the invitation message 205 may indicate a reference signal and a power threshold. In some examples, the RSRP threshold may be an example of a power threshold. A UE 115 that measures the RSRP of a reference signal above the RSRP threshold may participate in the federated learning procedure or the next training round of the federated learning procedure. For example, UE 115-a may measure the reference signal indicated by the invitation message 205. UE 115-a may compare the RSRP measurement of the reference signal to the RSRP threshold indicated by the invitation message 205. If the RSRP measurement of the reference signal meets (e.g., exceeds) the RSRP threshold indicated by the invitation message, UE 115-a may participate in the federated learning procedure.

[0136] In some examples, the invitation message 205 may indicate a nominal UE transmit power. For example, the invitation message 205 may indicate a nominal UE transmit power to be applied by the participating UEs 115 when transmitting the gradient values ​​210. In some cases, the nominal UE transmit power indicated by the invitation message 205 may be used for OTA federated learning procedures instead of regular transmissions (e.g., other PUSCH or PUCCH transmissions).

[0137] In some examples, a federated learning procedure with a non-RRC connected UE 115 or a federated learning procedure in a mode 2 side link operation may use open loop power control. In some cases, for open loop power control, the UE 115 may use a path loss estimate with an indicated nominal UE transmit power. In some cases, the invitation message 205 may indicate a path loss reference signal. The participating UE 115 may measure the path loss reference signal (e.g., a higher layer filtered RSRP of the path loss reference signal) and subtract the measurement from the reference signal power value indicated by the higher layer signaling. For example, the UE 115-a may measure the indicated path loss reference signal to obtain the RSRP of the path loss reference signal. The UE 115-a may compare the RSRP of the path loss reference signal with the reference signal power indicated via RRC signaling to obtain a path loss estimate to the network entity 105-a. The UE 115-a may use the nominal UE transmit power and the path loss estimate to determine the transmit power for transmitting the gradient value 210-a on the OTA aggregated resource 215.

[0138] In another example, the network entity 105-a may indicate the location of the network entity 105, and the UE 115 may use the location of the network entity 105-a to estimate the path loss. For example, the UE 115-a may receive an invitation message 205 indicating the location of the network entity 105-a. The UE 115-a may know its own location (e.g., the location of the UE 115-a), and the UE 115-a may calculate the distance between the network entity 105-a and the UE 115-a to estimate the path loss from the UE 115-a to the network entity 105-a. In some examples, the UE 115-a may use a path loss model to estimate the path loss. In some examples, the path loss model may be pre-configured (e.g., via RRC signaling). In some cases, the path loss model may be indicated via the invitation message 205. For example, the invitation message may include a pointer or identifier to a path loss model from a path loss model table or a path loss model parameter table. In some examples, the path loss estimate may be different for different UEs 115 or may depend on the UE specific implementation.

[0139] In some examples, a non-RRC connected UE 115 participating in a federated learning procedure over an uplink may operate in a discontinuous reception (DRX) mode. The invitation message 205 may include aspects of a wake-up signal or a paging signal for enabling the non-connected UE 115 to participate in a federated learning training round. In some cases, the UE 115 may be configured with a paging frame or paging opportunity associated with the invitation message 205. For example, a paging frame and opportunity for monitoring the invitation message 205 may be configured at the UE 115. In some examples, the paging frame and opportunity associated with the invitation message 205 may not overlap with the paging frame and opportunity of conventional DRX or non-invitation message signaling.

[0140] In some cases, the invitation message 205 may be sent to UEs 115 that are interested in participating in the federated learning procedure. Therefore, the invitation message 205 may not be sent to UEs 115 that are not interested in the federated learning procedure or do not want to participate in the federated learning procedure. In some cases, the network entity 105-a may send a wake-up signal for the invitation message 205, which is broadcast before the invitation message 205. The UE 115 interested in receiving the invitation message 205 may wake up (e.g., enter DRX on mode) to monitor and receive the invitation message 205. Additionally or alternatively, the invitation message 205 may be sent using a paging radio network temporary identifier (P-RNTI), and the P-RNTI may include an additional field indicating that the invitation message 205 is associated with the federated learning procedure. For example, the invitation message 205 may include an additional field indicating that the invitation message will initiate a federated learning training round. UEs that do not want to participate in the training round may avoid decoding the associated PDSCH. In some examples, the invitation message 205 may be sent using an RNTI that is specific to or dedicated to the federated learning procedure. The UE 115 that will participate in the federated learning procedure may receive the invitation message 205 by decoding the invitation message 205 using the dedicated RNTI.

[0141] In some cases, some paging frames or occasions may be associated with the invitation message 205. For example, the invitation message 205 may be sent during paging frames or paging occasions associated with a federated learning procedure. These paging frames or paging occasions may not overlap with paging frames and paging occasions associated with other types of signaling (e.g., non-federated learning signaling).

[0142] When a federated learning procedure is performed over a sidelink channel (e.g., UE 115 is a server), invite message 205 may be sent as different types of signaling. For example, invite message 205 may be sent as a broadcast transmission with sidelink control information indicating that the signaling carrying invite message 205 is an invitation to a federated learning procedure. In another example, invite message 205 may be sent via a multicast transmission with sidelink control information indicating that the signaling carrying invite message 205 is an invitation to a federated learning procedure. In some cases, invite message 205 may be sent as a multicast transmission or a broadcast transmission, and the payload of invite message 205 may indicate that the signaling is an invitation to a federated learning procedure.

[0143] In some examples, the invitation message 205 may include a condition for the UE 115 to participate in the next training round. For example, UE 115-a may receive the invitation message 205 and determine whether to join the next training round of the federated learning procedure based on whether UE 115-a meets the condition. If UE 115-a meets the condition, UE 115-a may join the training round. If UE 115-a does not meet the condition, UE 115-a may not join the training round. For example, the invitation message 205 may indicate the conditions that the participating UE 115 has an identifier of the latest version of the machine learning model, has a minimum local data set size, has a recently acquired data set, or any combination thereof. In some examples, the recently acquired data set may be a data set acquired no earlier than a time indication threshold or no earlier than an indicated time.

[0144] In some cases, limiting the distance of UEs 115 participating in the federated learning procedure can provide a server (e.g., network entity 105-a) with information about the path loss experienced by participating UEs 115. The network entity 105-a can then use an estimate of how many UEs 115 (e.g., on average) are participating to set a nominal UE transmit power in the invitation message 205. The estimate can be based on an estimate of the density of UEs 115 in an area (such as an area served by the network entity 105-a). However, in the case of supporting UE 115 participation without formally announcing or indicating participation, the number of participating UEs 115 may be so high that the aggregate receive power is increased beyond the receiver dynamic range. The wireless communication system 200 can support techniques for reducing the aggregate receive power in subsequent training rounds.

[0145] For example, when the server identifies that an aggregation round failed due to excessive received power, the server may not update the global model of the machine learning model, and the server may not broadcast the global model update. For the next training round, the server may send an invite message 205 pointing to the same model version identifier as in the previous invite message. However, the invite message 205 for the next training round may indicate a reduced nominal UE transmit power (e.g., compared to the previous round). When the UE 115 receives the invite message 205 with the same version as the previous invite message, the UE 115 may not perform any local updates to the local machine learning model. The UE 115 may send the same signal as the previous round, but according to the updated power control configuration (e.g., reduced nominal UE transmit power). In some examples, if the UE 115 obtains additional data samples between two rounds, the UE 115 may retrain the local model using the extended data set, or refrain from training the local model until the global model is updated.

[0146] In another example, when the server identifies that an aggregation round failed due to excessive received power, the server may indicate a probability (e.g., a scalar value) between 0 and 1. The UE 115 may participate in the next round with the indicated probability. For example, the UE 115 may determine a random value between 0 and 1, and if the determined random value exceeds the indicated scalar value, the UE 115 may participate in the next training round. These techniques may reduce the number of UEs participating in the next round without reducing the UE transmit power.

[0147] Figure 3 An example of a process flow 300 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is illustrated. The process flow 300 may be performed by one or more UEs 115 (such as UE 115-d) or a server 305, or any combination thereof. The server 305 may be as described in reference Figure 1 and Figure 2 An example of a UE 115 or network entity 105 is described. UE 115-d may be as described in reference Figure 1 and Figure 2 Examples of UE 115 described. In some examples, it may occur Figure 3 Some additional processes or signaling not shown. Additionally or alternatively, Figure 3 Some of the processes or signaling shown may not occur. In some examples, some of the processes or signaling of process flow 300 may occur in an order different from that shown.

[0148] At 310, the server 305 may send a first message requesting the UE 115-d to perform a federated learning procedure, and the UE 115-d may receive the first message. The first message may indicate a machine learning model and a configuration for the federated learning procedure. For example, the server 305 may send an invitation message to the UE 115-d, which may invite or request the UE 115-d to participate in a training round 320 of federated learning (e.g., send local updates to the machine learning model). In some examples, the server 305 may send an indication of a global machine learning model, which may be indicated via the first message or via a separate transmission.

[0149] At 315, UE 115-d may, in response to the first message requesting the UE to perform a federated learning procedure, perform a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure. For example, UE 115-d may locally train the machine learning model using a data set collected by UE 115-d. UE 115-d may determine updates to parameters for the machine learning model based on training the machine learning model using the data set. In some examples, UE 115-d may locally train the machine learning model regardless of whether UE 115-d participates in a training round 320 of federated learning.

[0150] In some cases, UE 115-d may determine whether to perform a federated learning procedure based on the first message. For example, UE 115-d may determine whether UE 115-d meets the conditions for performing a federated learning procedure or participating in the next round of federated learning. In some cases, the first message may indicate a version of the machine learning model. UE 115-d may determine that the version of the machine learning model at UE 115-d matches the version indicated by the first message, so that UE 115-d has the latest version of the machine learning model. UE 115-d may be able to participate in training round 320 based on having the latest version of the machine learning model. In some examples, the criteria for participating in training round 320 may be based on the version of the machine learning model, the local data set size (e.g., the minimum local data set size), the acquisition time of the local data set at the UE, or any combination thereof.

[0151] In some examples, the first message may indicate a range associated with participating in the federated learning procedure. For example, the invitation message may indicate a maximum distance from the server 305 for the UE 115 to participate in a training round. The UE 115-d may determine whether the UE 115-d is within the range, and the UE 115-d may participate in the training round 320 of the federated learning procedure based on being within the range.

[0152] In some examples, the first message may indicate a reference signal and a threshold. For example, the threshold may be an RSRP threshold. The UE 115-d may measure the indicated reference signal to obtain a measurement of the reference signal and compare the measurement to the threshold. If the measurement meets the threshold (e.g., exceeds the RSRP threshold), the UE 115-d may participate in training round 320.

[0153] In some examples, the first message may indicate a machine learning model from a set of machine learning models (e.g., a set of one or more machine learning models). For example, UE 115-d may be configured with the set of machine learning models (e.g., via RRC signaling). In some examples, the first message may include a pointer to a machine learning model in a machine learning model table. Additionally or alternatively, the first message may indicate a version of the machine learning model from a set of versions of the machine learning model. Additionally or alternatively, the first message may indicate an OTA aggregation scheme from a set of OTA aggregation schemes or a federated learning scheme from a set of federated learning schemes, or both.

[0154] In some examples, UE 115-d may determine to participate in training round 320. At 325, UE 115-d may send a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. For example, multiple UEs 115 may participate in training round 320, and each UE 115 may report gradient values ​​of machine learning model parameters by sending simulated gradient values ​​of the machine learning model parameters over the same set of resources. Server 305 may observe the superposition of simulated gradient values ​​to determine a set of parameters for training the machine learning model.

[0155] UE 115-d may perform a power control technique to send a second message indicating one or more gradient values. In some cases, the power control technique may ensure that signals from participating UEs 115 are received at the server 305 at the same power.

[0156] In some examples, the first message may indicate a nominal transmit power for sending the second message on one or more resources configured for OTA aggregation. In some examples, UE 115-d may send the second message based on the nominal transmit power.

[0157] For example, UE 115-d may receive an indication of a path loss reference signal and an indication of a reference signal power value. The indication of the path loss reference signal or the indication of the reference signal power value, or both, may be indicated via a first message or via other signaling (e.g., RRC signaling), or both. UE 115-d may measure the path loss reference signal and compare the measurement of the path loss reference signal with the reference signal power value (e.g., subtracting the measurement from the reference signal power value). UE 115-d may determine the path loss to server 305 based on the comparison of the path loss reference signal with the reference signal power value. UE 115-d may send a second message using a transmit power based on the path loss to server 305 and the nominal transmit power indicated by the first message.

[0158] In another example, UE 115-d may receive an indication of the location of server 305. UE 115-d may determine a path loss estimate for the second message based on the path loss model and a distance between UE 115-d and the location of server 305. UE 115-d may send the second message based on the path loss estimate and the nominal transmit power indicated by the first message.

[0159] The server 305 may receive a plurality of second messages from the plurality of UEs 115 on one or more resources configured for OTA aggregation. The plurality of second messages may indicate a plurality of sets of gradient values ​​for machine learning model parameters. At 330, the server 305 may train the machine learning model using the plurality of sets of one or more model parameters. For example, the server 305 may update the global machine learning model based on the reported gradient values ​​from the participating UEs 115.

[0160] In some cases, the server 305 may broadcast an updated global model of the machine learning model at 335. For example, after updating the machine learning model based on the gradient values ​​from the participating UEs 115, the server 305 may broadcast or send the updated machine learning model. The updated machine learning model may be used by the participating UEs 115 for the next federated learning training round. In some examples, at 340, the server 305 may send another invitation message for the next federated learning training round.

[0161] Figure 4 An example of a process flow 400 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is illustrated. The process flow 400 may be performed by one or more UEs 115 (such as UE 115-e) or a server 405, or any combination thereof. The server 405 may be as described in reference Figure 1 and Figure 2An example of a UE 115 or network entity 105 is described. The UE 115-e may be as described in reference Figure 1 and Figure 2 Examples of UE 115 described. In some examples, it may occur Figure 4 Some additional processes or signaling not shown. Additionally or alternatively, Figure 4 Some of the processes or signaling shown may not occur. In some examples, some of the processes or signaling of process flow 400 may occur in an order different from that shown.

[0162] At 410, the server 405 may send a first message requesting the UE 115-e to perform a federated learning procedure, and the UE 115-e may receive the first message. The first message may indicate a machine learning model and a configuration for the federated learning procedure. For example, the server 405 may send an invitation message to the UE 115-e, which may invite or request the UE 115-e to participate in a training round 420 of federated learning (e.g., send local updates to the machine learning model). In some examples, the server 405 may send an indication of a global machine learning model, which may be indicated via the first message or via a separate transmission.

[0163] At 415, the UE 115-e may, in response to the first message requesting the UE 115-e to perform a federated learning procedure, perform a training procedure using the machine learning model to obtain one or more model parameters based on a configuration for the federated learning procedure. For example, the UE 115-e may locally train the machine learning model using a data set collected by the UE 115-e. The UE 115-e may determine updates to parameters for the machine learning model based on training the machine learning model using the data set. In some cases, the UE 115-e may determine whether to perform a federated learning procedure based on the first message. In some examples, the criteria for participating in the training round 420 may be based on a version of the machine learning model, a local data set size (e.g., a minimum local data set size), an acquisition time of the local data set at the UE, or any combination thereof.

[0164] In some examples, UE 115-e may determine to participate in training round 420. At 425, UE 115-e may send a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. For example, multiple UEs 115 may participate in training round 420, and each UE 115 may report the gradient values ​​of the machine learning model parameters by sending simulated gradient values ​​of the machine learning model parameters over the same set of resources. Server 405 may observe the superposition of simulated gradient values ​​to determine a set of parameters for training the machine learning model.

[0165] The server 405 may receive a plurality of second messages from the plurality of UEs 115 on one or more resources configured for OTA aggregation. However, in some cases, the received power of the signaling may be too large, causing the aggregated received power to exceed the receiver dynamic range. This may cause the aggregation to fail, and the server 405 may be unable to update the global model. For example, at 430, the server 405 may determine that training the machine learning model using the set of model parameters is unsuccessful based on the excessive received power of the plurality of second messages.

[0166] At 435, the server 305 may send a second invitation message requesting the plurality of UEs 115 to perform a federated learning procedure, wherein the second invitation message indicates the same machine learning model version as the previous training round and a reduced nominal UE transmit power. By indicating the reduced nominal UE transmit power, the gradient values ​​in the next training round may be sent at the reduced transmit power, which may prevent aggregation failure.

[0167] In some examples, the UE 115-e may identify a second invitation message having the same model version as the previous training round. In some cases, the UE 115-e may refrain from performing local updates for the machine learning model. The UE 115-e may send the gradient value on the next OTA aggregate resource set based on the reduced nominal UE transmit power.

[0168] In some other examples, when the server 405 identifies that an aggregation round failed due to excessive received power, the server 405 may indicate a probability (e.g., a scalar value) between 0 and 1 in the invitation message. The UE 115 may participate in the next round with the indicated probability. For example, the UE 115-e may determine a random value between 0 and 1, and if the determined random value exceeds the indicated scalar value, the UE 115-e may participate in the next training round. These techniques may reduce the number of UEs 115 participating in the next round without reducing the UE transmit power.

[0169] Figure 5 A block diagram 500 of a device 505 supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 505 can be an example of aspects of the UE 115 as described herein. The device 505 may include a receiver 510, a transmitter 515, and a communication manager 520. The device 505 may also include a processor. Each of these components can communicate with each other (e.g., via one or more buses).

[0170] The receiver 510 may provide means for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to OTA aggregated federated learning with non-connected devices). The information may be passed to other components of the device 505. The receiver 510 may utilize a single antenna or a collection of multiple antennas.

[0171] The transmitter 515 may provide means for transmitting signals generated by other components of the device 505. For example, the transmitter 515 may transmit information associated with various information channels (e.g., control channels, data channels, information channels related to OTA aggregated federated learning with non-connected devices), such as packets, user data, control information, or any combination thereof. In some examples, the transmitter 515 may be co-located with the receiver 510 in a transceiver module. The transmitter 515 may utilize a single antenna or a collection of multiple antennas.

[0172] The communication manager 520, the receiver 510, the transmitter 515, or various combinations thereof or various components thereof may be examples of means for performing various aspects of OTA aggregated federated learning with non-connected devices as described herein. For example, the communication manager 520, the receiver 510, the transmitter 515, or various combinations thereof or components thereof may support methods for performing one or more of the functions described herein.

[0173] In some examples, the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be implemented in hardware (e.g., in a communication management circuit). The hardware may include a processor, a digital signal processor (DSP), a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a microcontroller, discrete gate or transistor logic, discrete hardware components, or any combination thereof that is configured to or otherwise supports components for performing the functions described in the present disclosure. In some examples, a 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 instructions stored in the memory by the processor).

[0174] Additionally or alternatively, in some examples, the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be implemented in code executed by a processor (e.g., as communication management software or firmware). When implemented in code executed by a processor, the functionality of the communication manager 520, the receiver 510, the transmitter 515, or various combinations or components thereof may be performed by a general purpose processor (e.g., configured as or otherwise supporting components for performing the functions described in the present disclosure), a DSP, a CPU, an ASIC, an FPGA, a microcontroller, or any combination of these or other programmable logic devices.

[0175] In some examples, communication manager 520 can be configured to perform various operations (e.g., receive, obtain, monitor, output, send) using or otherwise cooperating with receiver 510, transmitter 515, or both. For example, communication manager 520 can receive information from receiver 510, transmit information to transmitter 515, or be integrated in conjunction with receiver 510, transmitter 515, or both to obtain information, output information, or perform various other operations as described herein.

[0176] According to the examples disclosed herein, the communication manager 520 may support wireless communications at the UE. For example, the communication manager 520 may be configured to or otherwise support a component for receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 520 may be configured to or otherwise support a component for performing a training procedure using the machine learning model in response to the first message requesting the UE to perform a federated learning procedure to obtain one or more model parameters based on the configuration for the federated learning procedure. The communication manager 520 may be configured to or otherwise support a component for sending a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0177] Additionally or alternatively, according to examples as disclosed herein, the communication manager 520 may support wireless communications at a wireless device. For example, the communication manager 520 may be configured to or otherwise support a component for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 520 may be configured to or otherwise support a component for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The communication manager 520 may be configured to or otherwise support a component for training a machine learning model using a set of multiple sets of one or more model parameters.

[0178] By including or configuring a communication manager 520 according to examples as described herein, a device 505 (e.g., a processor controlling or otherwise coupled with a receiver 510, a transmitter 515, a communication manager 520, or a combination thereof) may support techniques for more efficiently utilizing communication resources by using OTA aggregation for federated learning with non-connected UEs 115. By increasing the number of UEs 115 participating in federated learning, these techniques provide greater reductions in resource overhead and utilization by enabling non-connected UEs 115 to perform OTA aggregation using efficient power control techniques.

[0179] Figure 6 A block diagram 600 of a device 605 supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 605 can be an example of aspects of the device 505 or UE 115 as described herein. The device 605 may include a receiver 610, a transmitter 615, and a communication manager 620. The device 605 may also include a processor. Each of these components can communicate with each other (e.g., via one or more buses).

[0180] The receiver 610 may provide means for receiving information (such as packets, user data, control information, or any combination thereof) associated with various information channels (e.g., control channels, data channels, information channels related to OTA aggregated federated learning with non-connected devices). The information may be passed to other components of the device 605. The receiver 610 may utilize a single antenna or a collection of multiple antennas.

[0181] The transmitter 615 may provide means for transmitting signals generated by other components of the device 605. For example, the transmitter 615 may transmit information associated with various information channels (e.g., control channels, data channels, information channels related to OTA aggregated federated learning with non-connected devices), such as packets, user data, control information, or any combination thereof. In some examples, the transmitter 615 may be co-located with the receiver 610 in a transceiver module. The transmitter 615 may utilize a single antenna or a collection of multiple antennas.

[0182] Device 605 or its various components may be examples of components for performing various aspects of OTA aggregated federated learning with non-connected devices as described herein. For example, communication manager 620 may include an invitation receiving component 625, a training procedure component 630, a gradient sending component 635, an invitation sending component 640, a gradient receiving component 645, a model training component 650, or any combination thereof. Communication manager 620 may be an example of various aspects of communication manager 520 as described herein. In some examples, communication manager 620 or its various components may be configured to use or otherwise cooperate with receiver 610, transmitter 615, or both to perform various operations (e.g., receive, obtain, monitor, output, send). For example, communication manager 620 may receive information from receiver 610, transmit information to transmitter 615, or integrate with receiver 610, transmitter 615, or both to obtain information, output information, or perform various other operations as described herein.

[0183] According to an example as disclosed herein, the communication manager 620 may support wireless communication at the UE. The invitation receiving component 625 may be configured to or otherwise support a component for receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The training procedure component 630 may be configured to or otherwise support a component for performing a training procedure using the machine learning model in response to the first message requesting the UE to perform a federated learning procedure to obtain one or more model parameters based on the configuration for the federated learning procedure. The gradient sending component 635 may be configured to or otherwise support a component for sending a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0184] Additionally or alternatively, the communication manager 620 may support wireless communications at the wireless device according to examples as disclosed herein. The invitation sending component 640 may be configured to or otherwise support components for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The gradient receiving component 645 may be configured to or otherwise support components for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The model training component 650 may be configured to or otherwise support components for training a machine learning model using a set of multiple sets of one or more model parameters.

[0185] Figure 7 A block diagram 700 of a communication manager 720 supporting OTA aggregate federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The communication manager 720 may be an example of aspects of the communication manager 520, the communication manager 620, or both as described herein. The communication manager 720 or its various components may be examples of components for performing various aspects of OTA aggregate federated learning with non-connected devices as described herein. For example, the communication manager 720 may include an invitation receiving component 725, a training procedure component 730, a gradient sending component 735, an invitation sending component 740, a gradient receiving component 745, a model training component 750, a reference signal measurement component 755, a model update component 760, a model updating component 765, a path loss estimation component 770, an excess power determination component 775, or any combination thereof. Each of these components may communicate directly or indirectly with each other (e.g., via one or more buses).

[0186] According to an example as disclosed herein, the communication manager 720 may support wireless communication at the UE. The invitation receiving component 725 may be configured to or otherwise support a component for receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The training procedure component 730 may be configured to or otherwise support a component for performing a training procedure using the machine learning model in response to the first message requesting the UE to perform a federated learning procedure to obtain one or more model parameters based on the configuration for the federated learning procedure. The gradient sending component 735 may be configured to or otherwise support a component for sending a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0187] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured to or otherwise support components for receiving a first message indicating a range associated with participating in a federated learning procedure, wherein sending the second message is based on determining that the UE is within the range.

[0188] In some examples, to support receiving the first message, the invitation receiving component 725 can be configured or otherwise support means for receiving the first message indicating a reference signal and a threshold. In some examples, to support receiving the first message, the reference signal measuring component 755 can be configured or otherwise support means for measuring the reference signal to obtain a measurement of the reference signal, wherein sending the second message is based on the measurement of the reference signal exceeding the threshold.

[0189] In some examples, to support receiving a first message, the invitation receiving component 725 may be configured as or otherwise support components for receiving a first message indicating a nominal transmit power for sending a second message on one or more resources configured for OTA aggregation, wherein sending the second message is based on the nominal transmit power.

[0190] In some examples, reference signal measurement component 755 can be configured or otherwise support components for receiving an indication of a path loss reference signal and an indication of a reference signal power value. In some examples, reference signal measurement component 755 can be configured or otherwise support components for measuring a path loss reference signal, wherein sending the second message is based on the measurement of the path loss reference signal and the reference signal power value.

[0191] In some examples, path loss estimation component 770 can be configured or otherwise support components for receiving an indication of a location of a server. In some examples, path loss estimation component 770 can be configured or otherwise support components for determining a path loss estimate for a second message based on a path loss model and a distance between the UE and the location of the server, wherein sending the second message is based on the path loss estimate.

[0192] In some examples, to support receiving the first message, invitation receiving component 725 may be configured as or otherwise support components for receiving an indication of a machine learning model from a set of multiple machine learning models.

[0193] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured as or otherwise support a component for receiving an indication of the number of layers in a set of multiple layers, the size of each layer in the set of multiple layers, the order of the set of multiple layers, the connectivity of the set of multiple layers, or any combination thereof for a machine learning model, wherein execution of a training procedure is based on the number of layers, the size of each layer in the set of multiple layers, the order of the set of multiple layers, the connectivity of the set of multiple layers, or any combination thereof.

[0194] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured as or otherwise support components for receiving an indication of a version of the machine learning model from a set of multiple versions of the machine learning model, wherein execution of the training procedure is based on the version of the machine learning model.

[0195] In some examples, to support receiving the first message, invitation receiving component 725 can be configured or otherwise support means for receiving an indication of an OTA aggregation scheme from a set of multiple OTA aggregation schemes.

[0196] In some examples, the invitation receiving component 725 can be configured or otherwise support a component for receiving an indication of a set of multiple occasions for a first message requesting the UE to perform a federated learning procedure in one or more paging frames. In some examples, the invitation receiving component 725 can be configured or otherwise support a component for monitoring the first message based on the indication of the set of multiple occasions for the first message.

[0197] In some examples, the invitation receiving component 725 may be configured or otherwise support a component for receiving a wake-up signal associated with the first message. In some examples, the invitation receiving component 725 may be configured or otherwise support a component for monitoring the first message requesting the UE to perform a federated learning procedure based on receiving the wake-up signal.

[0198] In some examples, the invitation receiving component 725 can be configured or otherwise support means for decoding the first message based on the paging radio network temporary identifier. In some examples, the invitation receiving component 725 can be configured or otherwise support means for detecting one or more fields in the first message corresponding to a federated learning procedure, wherein performing the training procedure is based on detecting the one or more fields.

[0199] In some examples, the invitation receiving component 725 may be configured as or otherwise support components for decoding the first message based on a radio network temporary identifier associated with a federated learning procedure, wherein executing the training procedure is based on decoding the first message based on the radio network temporary identifier associated with the federated learning procedure.

[0200] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured as or otherwise support components for receiving the first message during a paging frame or opportunity associated with a federated learning procedure, wherein execution of the training procedure is based on receiving the first message during a paging frame or opportunity associated with the federated learning procedure.

[0201] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured as or otherwise support components for receiving a broadcast transmission or a multicast transmission of sidelink control information on a sidelink channel that includes a request for the UE to perform a federated learning procedure.

[0202] In some examples, to support receiving the first message, the invitation receiving component 725 may be configured as or otherwise support a component for receiving a first message indicating one or more criteria for participating in a federated learning procedure, wherein the one or more criteria are based on a version of the machine learning model, a minimum local dataset size, an acquisition time of the local dataset at the UE, or any combination thereof.

[0203] In some examples, to support sending a second message, the gradient sending component 735 may be configured as or otherwise support components for sending a second message via one or more resources configured for transmission by a set of multiple UEs including the UE, wherein the set of multiple UEs are scheduled to send a corresponding set of gradient values ​​via one or more resources according to a common transmission power.

[0204] In some examples, model update component 760 may be configured or otherwise support components for receiving a third message indicating an updated version of the machine learning model based on one or more gradient values ​​of one or more model parameters. In some examples, model update component 760 may be configured or otherwise support components for updating the machine learning model based on the updated version of the machine learning model.

[0205] In some examples, the invitation receiving component 725 may be configured as or otherwise support a component for receiving a third message indicating the same version model of the machine learning model and a scalar probability threshold for participating in the federated learning procedure. In some examples, the invitation receiving component 725 may be configured as or otherwise support a component for determining to execute the federated learning procedure based on the randomly generated value satisfying the scalar probability threshold. In some examples, the training procedure component 730 may be configured as or otherwise support a component for executing a training procedure using the machine learning model to obtain a second one or more model parameters based on the configuration for the federated learning procedure and the randomly generated value satisfying the scalar probability threshold. In some examples, the gradient sending component 735 may be configured as or otherwise support a component for sending a fourth message indicating a second one or more gradient values ​​of the second one or more model parameters based on the configuration for the federated learning procedure.

[0206] In some examples, the UE operates in an inactive state or an idle state.

[0207] In some examples, to support receiving a first message, gradient sending component 735 may be configured or otherwise support components for receiving a first message that configures the UE to send one or more model parameters in a second message, where the second message indicates the one or more model parameters.

[0208] Additionally or alternatively, according to examples as disclosed herein, the communication manager 720 may support wireless communications at the wireless device. The invitation sending component 740 may be configured to or otherwise support components for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The gradient receiving component 745 may be configured to or otherwise support components for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The model training component 750 may be configured to or otherwise support components for training a machine learning model using a set of multiple sets of one or more model parameters.

[0209] In some examples, to support sending the first message, the invitation sending component 740 may be configured to or otherwise support components for sending a first message indicating a range associated with participating in a federated learning procedure, wherein a set of multiple second messages are received from UEs within the range.

[0210] In some examples, to support sending the first message, the invitation sending component 740 may be configured as or otherwise support components for sending an indication of a machine learning model from a set of multiple machine learning models, a version of a machine learning model from a set of multiple versions of the machine learning model, an OTA aggregation scheme from a set of multiple OTA aggregation schemes, a federated learning procedure from a set of multiple federated learning procedures, or any combination thereof.

[0211] In some examples, to support sending the first message, the invitation sending component 740 may be configured as or otherwise support components for sending an indication of the number of layers in a set of multiple layers of a machine learning model, the size of each layer in the set of multiple layers, the order of the set of multiple layers, the connectivity of the set of multiple layers, or any combination thereof.

[0212] In some examples, model update component 765 can be configured or otherwise support means for sending a third message indicating an updated version of the machine learning model based on the training machine learning model. In some examples, gradient receiving component 745 can be configured or otherwise support means for receiving a set of multiple fourth messages indicating a second set of multiple sets of gradient values ​​for a second set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0213] In some examples, excess power determining component 775 may be configured to or otherwise support components for determining that training the machine learning model using a second set of multiple sets of one or more model parameters was unsuccessful based on the excess received power of the set of multiple fourth messages. In some examples, invitation sending component 740 may be configured to or otherwise support components for sending a fifth message requesting the set of multiple UEs to perform a federated learning procedure, the fifth message indicating an updated version of the machine learning model and a reduced nominal UE transmit power.

[0214] Figure 8A diagram of a system 800 including a device 805 that supports OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 805 may be an example of a device 505, a device 605, or a UE 115 as described herein, or include components thereof. The device 805 may communicate (e.g., wirelessly) with one or more network entities 105, one or more UEs 115, or any combination thereof. The device 805 may include components for two-way voice and data communications, including components for sending and receiving communications, such as a communication manager 820, an input / output (I / O) controller 810, a transceiver 815, an antenna 825, a memory 830, a code 835, and a processor 840. These components may be electronically communicated or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., a bus 845).

[0215] I / O controller 810 can manage input and output signals for device 805. I / O controller 810 can also manage peripheral devices that are not integrated into device 805. In some cases, I / O controller 810 can represent a physical connection or port to an external peripheral device. In some cases, I / O controller 810 can utilize an operating system, such as MS- or another known operating system. Additionally or alternatively, I / O controller 810 may represent or interact with a modem, keyboard, mouse, touch screen, or similar device. In some cases, I / O controller 810 may be implemented as part of a processor, such as processor 840. In some cases, a user may interact with device 805 via I / O controller 810 or via hardware components controlled by I / O controller 810.

[0216] In some cases, the device 805 may include a single antenna 825. However, in some other cases, the device 805 may have more than one antenna 825, which may be capable of concurrently transmitting or receiving multiple wireless transmissions. The transceiver 815 may communicate bidirectionally via one or more antennas 825, wired or wireless links, as described herein. For example, the transceiver 815 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver 815 may also include a modem for: modulating packets; providing the modulated packets to one or more antennas 825 for transmission; and demodulating packets received from one or more antennas 825. The transceiver 815, or the transceiver 815 and one or more antennas 825 may be examples of transmitters 515, transmitters 615, receivers 510, receivers 610, or any combination thereof or components thereof as described herein.

[0217] The memory 830 may include random access memory (RAM) and read-only memory (ROM). The memory 830 may store computer-readable, computer-executable code 835 including instructions that, when executed by the processor 840, cause the device 805 to perform various functions described herein. The code 835 may be stored in a non-transitory computer-readable medium (such as system memory or another type of memory). In some cases, the code 835 may not be directly executable by the processor 840, but may (e.g., when compiled and executed) cause the computer to perform the functions described herein. In some cases, the memory 830 may include, among other things, a basic I / O system (BIOS) that may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0218] The processor 840 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 840 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be integrated into the processor 840. The processor 840 may be configured to execute computer-readable instructions stored in a memory (e.g., a memory 830) to enable the device 805 to perform various functions (e.g., functions or tasks that support OTA aggregated federated learning using non-connected devices). For example, the device 805 or a component of the device 805 may include a processor 840 and a memory 830 coupled to or coupled to the processor 840, and the processor 840 and the memory 830 are configured to perform the various functions described herein.

[0219] According to the examples disclosed herein, the communication manager 820 may support wireless communications at the UE. For example, the communication manager 820 may be configured to or otherwise support a component for receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 820 may be configured to or otherwise support a component for performing a training procedure using the machine learning model in response to the first message requesting the UE to perform a federated learning procedure to obtain one or more model parameters based on the configuration for the federated learning procedure. The communication manager 820 may be configured to or otherwise support a component for sending a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure.

[0220] Additionally or alternatively, according to examples as disclosed herein, the communication manager 820 may support wireless communications at a wireless device. For example, the communication manager 820 may be configured to or otherwise support a component for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 820 may be configured to or otherwise support a component for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The communication manager 820 may be configured to or otherwise support a component for training a machine learning model using a set of multiple sets of one or more model parameters.

[0221] By including or configuring a communication manager 820 according to examples as described herein, the device 805 can support techniques for faster machine learning model training and federated learning. These techniques can greatly increase the number of UEs 115 participating in the federated learning procedure by enabling non-connected UEs 115 to participate and report gradient values ​​of locally determined machine learning model parameters. Additionally, techniques for power control enable non-connected UEs 115 to efficiently report gradient values ​​using OTA aggregation techniques without interrupting the receiving power of the gradient values ​​on the OTA aggregation resources at the server.

[0222] In some examples, the communication manager 820 may be configured to perform various operations (e.g., receive, monitor, transmit) using or otherwise cooperating with the transceiver 815, one or more antennas 825, or any combination thereof. Although the communication manager 820 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 820 may be supported or performed by the processor 840, the memory 830, the code 835, or any combination thereof. For example, the code 835 may include instructions that may be executed by the processor 840 to cause the device 805 to perform various aspects of OTA aggregated federated learning with non-connected devices as described herein, or the processor 840 and the memory 830 may be otherwise configured to perform or support such operations.

[0223] Fig. 9 A block diagram 900 of a device 905 supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 905 can be an example of aspects of the network entity 105 as described herein. The device 905 may include a receiver 910, a transmitter 915, and a communication manager 920. The device 905 may also include a processor. Each of these components can communicate with each other (e.g., via one or more buses).

[0224] The receiver 910 may provide means for obtaining (e.g., receiving, determining, identifying) information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). The information may be passed to other components of the device 905. In some examples, the receiver 910 may support obtaining information by receiving signals via one or more antennas. Additionally or alternatively, the receiver 910 may support obtaining information by receiving signals via one or more wired (e.g., electrical, optical) interfaces, wireless interfaces, or any combination thereof.

[0225] The transmitter 915 may provide a means for outputting (e.g., sending, providing, conveying, transmitting) information generated by other components of the device 905. For example, the transmitter 915 may output information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). In some examples, the transmitter 915 may support outputting information by sending signals via one or more antennas. Additionally or alternatively, the transmitter 915 may support outputting information by sending signals via one or more wired (e.g., electrical, optical) interfaces, wireless interfaces, or any combination thereof. In some examples, the transmitter 915 and the receiver 910 may be co-located in a transceiver, which may include a modem or be coupled to a modem.

[0226] The communication manager 920, the receiver 910, the transmitter 915, or various combinations thereof or various components thereof may be examples of means for performing various aspects of OTA aggregated federated learning with non-connected devices as described herein. For example, the communication manager 920, the receiver 910, the transmitter 915, or various combinations thereof or components thereof may support methods for performing one or more of the functions described herein.

[0227] In some examples, the communication manager 920, the receiver 910, the transmitter 915, or various combinations or components thereof may be implemented in hardware (e.g., in a communication management circuit). The hardware may include a processor, a DSP, a CPU, an ASIC, an FPGA or other programmable logic device, a microcontroller, discrete gate or transistor logic, discrete hardware components, or any combination thereof that is configured as or otherwise supports components for performing the functions described in the present disclosure. In some examples, a 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 instructions stored in the memory by the processor).

[0228] Additionally or alternatively, in some examples, the communication manager 920, the receiver 910, the transmitter 915, or various combinations or components thereof may be implemented in code executed by a processor (e.g., as communication management software or firmware). If implemented in code executed by a processor, the functionality of the communication manager 920, the receiver 910, the transmitter 915, or various combinations or components thereof may be performed by a general purpose processor (e.g., configured as or otherwise supporting components for performing the functions described in the present disclosure), a DSP, a CPU, an ASIC, an FPGA, a microcontroller, or any combination of these or other programmable logic devices.

[0229] In some examples, the communication manager 920 may be configured to perform various operations (e.g., receive, obtain, monitor, output, send) using or otherwise cooperating with the receiver 910, the transmitter 915, or both. For example, the communication manager 920 may receive information from the receiver 910, transmit information to the transmitter 915, or be integrated with the receiver 910, the transmitter 915, or both to obtain information, output information, or perform various other operations as described herein.

[0230] According to examples as disclosed herein, the communication manager 920 may support wireless communications at a wireless device. For example, the communication manager 920 may be configured to or otherwise support a component for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 920 may be configured to or otherwise support a component for receiving a set of multiple second messages indicating a set of multiple gradient value sets of a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The communication manager 920 may be configured to or otherwise support a component for training a machine learning model using a set of multiple sets of one or more model parameters.

[0231] By including or configuring a communication manager 920 according to examples as described herein, a device 905 (e.g., a processor controlling or otherwise coupled to a receiver 910, a transmitter 915, a communication manager 920, or a combination thereof) may support techniques for more efficiently utilizing communication resources by performing federated learning with non-connected UEs 115 using OTA aggregation. By increasing the number of UEs 115 participating in federated learning, these techniques provide greater reductions in resource overhead and utilization by enabling non-connected UEs 115 to perform OTA aggregation using efficient power control techniques. Additionally, power control techniques may reduce the likelihood that received power at a server will impact or overwhelm the server and render training rounds useless.

[0232] Fig.10 A block diagram 1000 of a device 1005 supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 1005 can be an example of aspects of the device 905 or network entity 105 as described herein. The device 1005 may include a receiver 1010, a transmitter 1015, and a communication manager 1020. The device 1005 may also include a processor. Each of these components can communicate with each other (e.g., via one or more buses).

[0233] Receiver 1010 may provide means for obtaining (e.g., receiving, determining, identifying) information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). The information may be passed to other components of device 1005. In some examples, receiver 1010 may support obtaining information by receiving signals via one or more antennas. Additionally or alternatively, receiver 1010 may support obtaining information by receiving signals via one or more wired (e.g., electrical, optical) interfaces, wireless interfaces, or any combination thereof.

[0234] The transmitter 1015 may provide a means for outputting (e.g., transmitting, providing, conveying, transmitting) information generated by other components of the device 1005. For example, the transmitter 1015 may output information associated with various channels (e.g., control channels, data channels, information channels, channels associated with a protocol stack), such as user data, control information, or any combination thereof (e.g., I / Q samples, symbols, packets, protocol data units, service data units). In some examples, the transmitter 1015 may support outputting information by sending signals via one or more antennas. Additionally or alternatively, the transmitter 1015 may support outputting information by sending signals via one or more wired (e.g., electrical, optical) interfaces, wireless interfaces, or any combination thereof. In some examples, the transmitter 1015 and the receiver 1010 may be co-located in a transceiver, which may include a modem or be coupled to a modem.

[0235] Device 1005 or its various components may be examples of components for performing various aspects of OTA aggregated federated learning with non-connected devices as described herein. For example, communication manager 1020 may include an invitation sending component 1025, a gradient receiving component 1030, a model training component 1035, or any combination thereof. Communication manager 1020 may be an example of various aspects of communication manager 920 as described herein. In some examples, communication manager 1020 or its various components may be configured to use or otherwise cooperate with receiver 1010, transmitter 1015, or both to perform various operations (e.g., receive, obtain, monitor, output, send). For example, communication manager 1020 may receive information from receiver 1010, transmit information to transmitter 1015, or integrate with receiver 1010, transmitter 1015, or both to obtain information, output information, or perform various other operations as described herein.

[0236] According to examples as disclosed herein, the communication manager 1020 may support wireless communications at a wireless device. The invitation sending component 1025 may be configured to or otherwise support components for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The gradient receiving component 1030 may be configured to or otherwise support components for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The model training component 1035 may be configured to or otherwise support components for training a machine learning model using a set of multiple sets of one or more model parameters.

[0237] Fig.11A block diagram 1100 of a communication manager 1120 supporting OTA aggregate federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The communication manager 1120 may be an example of aspects of the communication manager 920, the communication manager 1020, or both as described herein. The communication manager 1120 or its various components may be examples of components for performing various aspects of OTA aggregate federated learning with non-connected devices as described herein. For example, the communication manager 1120 may include an invitation sending component 1125, a gradient receiving component 1130, a model training component 1135, a model updating component 1140, an excess power determination component 1145, or any combination thereof. Each of these components may communicate directly or indirectly with each other (e.g., via one or more buses), which communication may include communication within a protocol layer of a protocol stack, communication associated with a logical channel of a protocol stack (e.g., between protocol layers of a protocol stack, within a device, component, or virtualized component associated with a network entity 105, between devices, components, or virtualized components associated with a network entity 105), or any combination thereof.

[0238] According to examples as disclosed herein, the communication manager 1120 may support wireless communications at a wireless device. The invitation sending component 1125 may be configured to or otherwise support components for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The gradient receiving component 1130 may be configured to or otherwise support components for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The model training component 1135 may be configured to or otherwise support components for training a machine learning model using a set of multiple sets of one or more model parameters.

[0239] In some examples, to support sending the first message, the invitation sending component 1125 may be configured to or otherwise support components for sending a first message indicating a range associated with participating in a federated learning procedure, wherein a set of multiple second messages are received from UEs within the range.

[0240] In some examples, to support sending the first message, the invitation sending component 1125 may be configured as or otherwise support components for sending an indication of a machine learning model from a set of multiple machine learning models, a version of a machine learning model from a set of multiple versions of a machine learning model, an OTA aggregation scheme from a set of multiple OTA aggregation schemes, a federated learning procedure from a set of multiple federated learning procedures, or any combination thereof.

[0241] In some examples, to support sending the first message, the invitation sending component 1125 may be configured as or otherwise support components for sending an indication of the number of layers in a set of multiple layers of a machine learning model, the size of each layer in the set of multiple layers, the order of the set of multiple layers, the connectivity of the set of multiple layers, or any combination thereof.

[0242] In some examples, model update component 1140 may be configured or otherwise support means for sending a third message indicating an updated version of the machine learning model based on the training machine learning model. In some examples, gradient receiving component 1130 may be configured or otherwise support means for receiving a set of a plurality of fourth messages indicating a second set of a plurality of sets of gradient values ​​for a second set of a plurality of sets of one or more model parameters via one or more resources configured for OTA aggregation based on a configuration for a federated learning procedure.

[0243] In some examples, excess power determining component 1145 can be configured to or otherwise support components for determining that training the machine learning model using a second set of multiple sets of one or more model parameters was unsuccessful based on the excess received power of the set of multiple fourth messages. In some examples, invitation sending component 1125 can be configured to or otherwise support components for sending a fifth message requesting the set of multiple UEs to perform a federated learning procedure, the fifth message indicating an updated version of the machine learning model and a reduced nominal UE transmit power.

[0244] Fig.12 A diagram of a system 1200 including a device 1205 supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The device 1205 may be an example of a device 905, a device 1005, or a network entity 105 as described herein, or include components thereof. The device 1205 may communicate with one or more network entities 105, one or more UEs 115, or any combination thereof, which may include communication through one or more wired interfaces, through one or more wireless interfaces, or any combination thereof. The device 1205 may include components that support output and acquisition of communications, such as a communication manager 1220, a transceiver 1210, an antenna 1215, a memory 1225, a code 1230, and a processor 1235. These components may be electronically communicated or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more buses (e.g., bus 1240).

[0245] As described herein, the transceiver 1210 may support bidirectional communication via a wired link, a wireless link, or both. In some examples, the transceiver 1210 may include a wired transceiver and may communicate bidirectionally with another wired transceiver. Additionally or alternatively, in some examples, the transceiver 1210 may include a wireless transceiver and may communicate bidirectionally with another wireless transceiver. In some examples, the device 1205 may include one or more antennas 1215, which may be capable of (e.g., concurrently) sending or receiving wireless transmissions. The transceiver 1210 may also include a modem for: modulating a signal; providing the modulated signal for transmission (e.g., via one or more antennas 1215, via a wired transmitter); receiving the modulated signal (e.g., from one or more antennas 1215, from a wired receiver); and demodulating the signal. In some implementations, the transceiver 1210 may include one or more interfaces, such as one or more interfaces coupled to one or more antennas 1215 configured to support various receiving or obtaining operations, or one or more interfaces coupled to one or more antennas 1215 configured to support various transmitting or outputting operations, or a combination thereof. In some implementations, the transceiver 1210 may include or be configured to be coupled to one or more processors or memory components, which may be operable to perform or support operations based on received or obtained information or signals, or to generate information or other signals for transmission or other output, or any combination of the above. In some implementations, the transceiver 1210, or the transceiver 1210 and one or more antennas 1215, or the transceiver 1210 and one or more antennas 1215 and one or more processors or memory components (e.g., processor 1235 or memory 1225 or both) may be included in a chip or chip assembly installed in the device 1205. Where applicable, the transceiver 1210, or the transceiver 1210 and one or more antennas 1215, or a wired interface may be an example of a transmitter 915, a transmitter 1015, a receiver 910, a receiver 1010, or any combination thereof or components thereof as described herein. In some examples, the transceiver may be operable to support communication via one or more communication links (e.g., communication link 125, backhaul communication link 120, midhaul communication link 162, fronthaul communication link 168).

[0246] Memory 1225 may include RAM and ROM. Memory 1225 may store computer-readable, computer-executable code 1230 including instructions that, when executed by processor 1235, cause device 1205 to perform various functions described herein. Code 1230 may be stored in a non-transitory computer-readable medium (such as system memory or another type of memory). In some cases, code 1230 may not be directly executable by processor 1235, but may (e.g., when compiled and executed) cause a computer to perform the functions described herein. In some cases, memory 1225 may include, among other things, a BIOS that may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0247] Processor 1235 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, an ASIC, a CPU, an FPGA, a microcontroller, a programmable logic device, a discrete gate or transistor logic element, a discrete hardware component, or any combination thereof). In some cases, processor 1235 may be configured to operate a memory array using a memory controller. In some other cases, the memory controller may be integrated into processor 1235. Processor 1235 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1225) to enable device 1205 to perform various functions (e.g., functions or tasks that support OTA aggregated federated learning using non-connected devices). For example, device 1205 or a component of device 1205 may include processor 1235 and memory 1225 coupled to processor 1235, and processor 1235 and memory 1225 are configured to perform the various functions described herein. The processor 1235 may be an example of a cloud computing platform (e.g., one or more physical nodes and supporting software (such as an operating system, a virtual machine, or a container instance)) that can host functions (e.g., by executing code 1230) to perform functions of the device 1205. The processor 1235 may be any one or more suitable processors capable of executing scripts or instructions of one or more software programs stored in the device 1205 (such as within the memory 1225). In some specific implementations, the processor 1235 may be a component of a processing system. A processing system may generally refer to a system or series of machines or components that receive inputs and process these inputs to produce a set of outputs (which may be passed to, for example, other systems or components of the device 1205). For example, the processing system of the device 1205 may refer to a system that includes various other components or subcomponents of the device 1205 (such as the processor 1235, or the transceiver 1210, or the communication manager 1220, or other components or combinations of components of the device 1205). The processing system of device 1205 may interface with other components of device 1205, and may process information (such as input or signal) received from other components or output information to other components. For example, a chip or modem of device 1205 may include a processing system and an interface for outputting information or for obtaining information or both. The interface may be implemented as or otherwise include a first interface configured to output information and a second interface configured to obtain information. In some specific implementations, the first interface may refer to an interface between the processing system of the chip or modem and a transmitter, so that device 1205 can send information output from the chip or modem. In some specific implementations, the second interface may refer to an interface between the processing system of the chip or modem and a receiver, so that device 1205 can obtain information or signal input, and the information can be passed to the processing system.One of ordinary skill in the art will readily recognize that the first interface may also obtain information or signal input, and the second interface may also output information or signal output.

[0248] In some examples, bus 1240 may support communications of a protocol layer (e.g., within a protocol layer) of a protocol stack. In some examples, bus 1240 may support communications associated with a logical channel of a protocol stack (e.g., between protocol layers of a protocol stack), which may include communications performed within a component of device 1205, or communications performed between different components of device 1205 that may be co-located or located in different locations (e.g., where device 1205 may refer to a system in which one or more of communication manager 1220, transceiver 1210, memory 1225, code 1230, and processor 1235 may be located in one of these different components or divided between different components).

[0249] In some examples, the communication manager 1220 can manage aspects of communications with the core network 130 (e.g., via one or more wired or wireless backhaul links). For example, the communication manager 1220 can manage the delivery of data communications for client devices, such as one or more UEs 115. In some examples, the communication manager 1220 can manage communications with other network entities 105 and can include a controller or scheduler for controlling communications with the UE 115 in coordination with the other network entities 105. In some examples, the communication manager 1220 can support an X2 interface within an LTE / LTE-A wireless communication network technology to provide communications between network entities 105.

[0250] According to examples as disclosed herein, the communication manager 1220 may support wireless communications at a wireless device. For example, the communication manager 1220 may be configured to or otherwise support a component for sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The communication manager 1220 may be configured to or otherwise support a component for receiving a set of multiple second messages indicating a set of multiple gradient value sets for a set of multiple sets of one or more model parameters via one or more resources configured for OTA aggregation based on the configuration for the federated learning procedure. The communication manager 1220 may be configured to or otherwise support a component for training a machine learning model using a set of multiple sets of one or more model parameters.

[0251] By including or configuring a communication manager 1220 according to examples as described herein, the device 1205 can support techniques for faster machine learning model training and federated learning. These techniques can greatly increase the number of UEs 115 participating in the federated learning procedure by enabling non-connected UEs 115 to participate and report gradient values ​​of locally determined machine learning model parameters. Additionally, techniques for power control enable non-connected UEs 115 to efficiently report gradient values ​​using OTA aggregation techniques without interrupting the receiving power of the gradient values ​​on the OTA aggregation resources at the server.

[0252] In some examples, the communication manager 1220 may be configured to perform various operations (e.g., receive, obtain, monitor, output, transmit) using or otherwise cooperating with the transceiver 1210, one or more antennas 1215 (e.g., where applicable), or any combination thereof. Although the communication manager 1220 is illustrated as a separate component, in some examples, one or more functions described with reference to the communication manager 1220 may be supported or performed by the processor 1235, the memory 1225, the code 1230, the transceiver 1210, or any combination thereof. For example, the code 1230 may include instructions that are executable by the processor 1235 to cause the device 1205 to perform various aspects of OTA aggregated federated learning with non-connected devices as described herein, or the processor 1235 and the memory 1225 may be otherwise configured to perform or support such operations.

[0253] Fig.13 A flowchart illustrating a method 1300 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The operations of the method 1300 may be implemented by a UE or a component thereof as described herein. For example, the operations of the method 1300 may be implemented by a UE or a component thereof as described in reference to Figures 1 to 8 The described UE 115 may be performed. In some examples, the UE may execute an instruction set to control the functional elements of the UE to perform the described functions. Additionally or alternatively, the UE may use dedicated hardware to perform various aspects of the described functions.

[0254] At 1305, the method may include receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The operations of 1305 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1305 may be performed as described in reference to Figure 7 The described invitation receiving component 725 is executed.

[0255] At 1310, the method may include, in response to a first message requesting the UE to perform a federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based on a configuration for the federated learning procedure. The operations of 1310 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1310 may be performed as described in reference to Figure 7 The described training procedure component 730 is performed.

[0256] At 1315, the method may include sending, based on the configuration for the federated learning procedure, a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation. The operations of 1315 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1315 may be performed as described in reference to Figure 7 The described gradient sending component 735 is executed.

[0257] Fig.14 A flowchart illustrating a method 1400 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The operations of the method 1400 may be implemented by a UE or a component thereof as described herein. For example, the operations of the method 1400 may be implemented by a UE or a component thereof as described in reference to Figures 1 to 8 The described UE 115 may be performed. In some examples, the UE may execute an instruction set to control the functional elements of the UE to perform the described functions. Additionally or alternatively, the UE may use dedicated hardware to perform various aspects of the described functions.

[0258] At 1405, the method may include receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model, a configuration for the federated learning procedure, a reference signal, and a threshold. The operations of 1405 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1405 may be performed as described in reference to Figure 7 The described invitation receiving component 725 is executed.

[0259] At 1410, the method may include measuring a reference signal to obtain a measurement of the reference signal, wherein sending the second message is based on the measurement of the reference signal exceeding a threshold. The operations of 1410 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1410 may be performed as described in reference Figure 7 The described reference signal measurement component 755 is performed.

[0260] At 1415, the method may include, in response to the first message requesting the UE to perform a federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure. The operations of 1415 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1415 may be performed as described in reference to Figure 7 The described training procedure component 730 is performed.

[0261] At 1420, the method may include sending, based on the configuration for the federated learning procedure, a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation. The operations of 1420 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1420 may be performed as described in reference to Figure 7 The described gradient sending component 735 is executed.

[0262] At 1425, the method may include receiving a first message indicating a reference signal and a threshold. The operations of 1425 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1425 may be performed by Figure 7 The described invitation receiving component 725 is executed.

[0263] Fig.15 A flowchart illustrating a method 1500 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The operations of the method 1500 may be implemented by a UE or a component thereof as described herein. For example, the operations of the method 1500 may be implemented by a UE or a component thereof as described in reference to Figures 1 to 8 The described UE 115 may be performed. In some examples, the UE may execute an instruction set to control the functional elements of the UE to perform the described functions. Additionally or alternatively, the UE may use dedicated hardware to perform various aspects of the described functions.

[0264] At 1505, the method may include receiving a wake-up signal associated with a first message requesting the UE to perform a federated learning procedure. 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 as described in reference to Figure 7 The described invitation receiving component 725 is executed.

[0265] At 1510, the method may include monitoring the first message based on receiving the wake-up signal. 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 as described in reference to Figure 7 The described invitation receiving component 725 is executed.

[0266] At 1515, the method may include receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. The operations of 1515 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1515 may be performed as described in reference to Figure 7 The described invitation receiving component 725 is executed.

[0267] At 1520, the method may include, in response to the first message requesting the UE to perform a federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based on the configuration for the federated learning procedure. The operations of 1520 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1520 may be performed as described in reference to Figure 7 The described training procedure component 730 is performed.

[0268] At 1525, the method may include sending, based on the configuration for the federated learning procedure, a second message indicating one or more gradient values ​​of one or more model parameters via one or more resources configured for OTA aggregation. The operations of 1525 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1525 may be performed as described in reference to Figure 7 The described gradient sending component 735 is executed.

[0269] Fig.16 A flowchart illustrating a method 1600 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The operations of the method 1600 may be implemented by a UE or a network entity or a component thereof as described herein. For example, the operations of the method 1600 may be implemented by a UE or a network entity or a component thereof as described herein. Figures 1 to 8 UE 115 as described or as referenced Figures 1 to 4 and Figures 9 to 12 In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0270] At 1605, the method may include sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. 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 as described in reference to Figure 7 and Fig.11 The described invitation sending component 740 or the invitation sending component 1125 is executed.

[0271] At 1610, the method may include receiving, based on a configuration for a federated learning procedure, a set of a plurality of second messages indicating a set of a plurality of sets of gradient values ​​for a set of a plurality of sets of one or more model parameters via one or more resources configured for OTA aggregation. 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 as described in reference to Figure 7 and Fig.11 The described gradient receiving component 745 or the gradient receiving component 1130 is performed.

[0272] At 1615, the method may include training the machine learning model using a set of multiple sets of one or more model parameters. 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 as described in reference to Figure 7 and Fig.11 The described model training component 750 or model training component 1135 is performed.

[0273] Fig.17 A flowchart illustrating a method 1700 for supporting OTA aggregated federated learning with non-connected devices according to one or more aspects of the present disclosure is shown. The operations of the method 1700 may be implemented by a UE or a network entity or a component thereof as described herein. For example, the operations of the method 1700 may be implemented by a UE or a network entity or a component thereof as described herein. Figures 1 to 8 UE 115 as described or as referenced Figures 1 to 4 and Figures 9 to 12 In some examples, the UE or network entity may execute an instruction set to control the functional elements of the UE or network entity to perform the described functions. Additionally or alternatively, the UE or network entity may use dedicated hardware to perform various aspects of the described functions.

[0274] At 1705, the method may include sending a first message requesting a set of multiple UEs to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure. 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 as described in reference to Figure 7 and Fig.11 The described invitation sending component 740 or the invitation sending component 1125 is executed.

[0275] At 1710, the method may include receiving, based on a configuration for a federated learning procedure, a set of a plurality of second messages indicating a set of a plurality of sets of gradient values ​​for a set of a plurality of sets of one or more model parameters via one or more resources configured for OTA aggregation. 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 as described in reference to Figure 7 and Fig.11 The described gradient receiving component 745 or the gradient receiving component 1130 is performed.

[0276] At 1715, the method may include training the machine learning model using a set of multiple sets of one or more model parameters. The operations of 1715 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1715 may be performed as described in reference to Figure 7 and Fig.11 The described model training component 750 or model training component 1135 is performed.

[0277] At 1720, the method may include sending a third message indicating an updated version of the machine learning model based on the training machine learning model. The operations of 1720 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1720 may be performed as described in reference to Figure 7 and Fig.11 The described model update component 765 or model update component 1140 is performed.

[0278] At 1725, the method may include receiving, based on the configuration for the federated learning procedure, via one or more resources configured for OTA aggregation, a set of a plurality of fourth messages indicating a second set of a plurality of sets of gradient values ​​for a second set of a plurality of sets of one or more model parameters. The operations of 1725 may be performed according to examples as disclosed herein. In some examples, aspects of the operations of 1725 may be performed as described in reference to Figure 7 and Fig.11 The described gradient receiving component 745 or the gradient receiving component 1130 is performed.

[0279] The following provides an overview of various aspects of the disclosure:

[0280] Aspect 1: A method for performing wireless communications at a UE, comprising: receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; in response to the first message requesting the UE to perform the federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based at least in part on the configuration for the federated learning procedure; and sending a second message indicating one or more gradient values ​​of the one or more model parameters via one or more resources configured for over-the-air aggregation based at least in part on the configuration for the federated learning procedure.

[0281] Aspect 2: The method according to aspect 1, wherein receiving the first message includes: receiving the first message indicating a range associated with participating in the federated learning procedure, wherein sending the second message is based at least in part on determining that the UE is within the range.

[0282] Aspect 3: A method according to any one of Aspects 1 to 2, wherein receiving the first message comprises: receiving the first message indicating a reference signal and a threshold; and measuring the reference signal to obtain a measurement of the reference signal, wherein sending the second message is at least partially based on the measurement of the reference signal exceeding the threshold.

[0283] Aspect 4: A method according to any one of Aspects 1 to 3, wherein receiving the first message includes: receiving the first message indicating a nominal transmit power for sending the second message on the one or more resources configured for the air aggregation, wherein sending the second message is at least partially based on the nominal transmit power.

[0284] Aspect 5: The method according to Aspect 4 further includes: receiving an indication of a path loss reference signal and an indication of a reference signal power value; and measuring the path loss reference signal, wherein sending the second message is at least partially based on the measurement of the path loss reference signal and the reference signal power value.

[0285] Aspect 6: The method according to any one of Aspects 4 to 5 further includes: receiving an indication of the location of the server; and determining a path loss estimate for the second message based at least in part on a path loss model and a distance between the UE and the location of the server, wherein sending the second message is based at least in part on the path loss estimate.

[0286] Aspect 7: A method according to any one of Aspects 1 to 6, wherein receiving the first message comprises: receiving an indication of the machine learning model from a plurality of machine learning models.

[0287] Aspect 8: A method according to any one of Aspects 1 to 7, wherein receiving the first message includes: receiving an indication of the number of layers in a plurality of layers of the machine learning model, the size of each of the plurality of layers, the order of the plurality of layers, the connectivity of the plurality of layers, or any combination thereof, wherein executing the training procedure is at least partially based on the number of layers, the size of each of the plurality of layers, the order of the plurality of layers, the connectivity of the plurality of layers, or any combination thereof.

[0288] Aspect 9: A method according to any one of Aspects 1 to 8, wherein receiving the first message includes: receiving an indication of a version of the machine learning model from multiple versions of the machine learning model, wherein executing the training procedure is at least partially based on the version of the machine learning model.

[0289] Aspect 10: The method according to any one of aspects 1 to 9, wherein receiving the first message comprises: receiving an indication of an air aggregation scheme from a plurality of air aggregation schemes.

[0290] Aspect 11: The method according to any one of Aspects 1 to 10 further includes: receiving, in one or more paging frames, an indication of multiple occasions for the first message requesting the UE to perform the federated learning procedure; and monitoring the first message based at least in part on the indication of the multiple occasions for the first message.

[0291] Aspect 12: The method according to any one of Aspects 1 to 11 further includes: receiving a wake-up signal associated with the first message; and monitoring the first message requesting the UE to perform the federated learning procedure based at least in part on receiving the wake-up signal.

[0292] Aspect 13: The method according to any one of Aspects 1 to 12 further includes: decoding the first message based at least in part on a paging radio network temporary identifier; and detecting one or more fields in the first message corresponding to the federated learning procedure, wherein execution of the training procedure is based at least in part on detecting the one or more fields.

[0293] Aspect 14: The method according to any one of Aspects 1 to 13 further includes: decoding the first message based at least in part on a radio network temporary identifier associated with the federated learning procedure, wherein executing the training procedure is based at least in part on decoding the first message based at least in part on the radio network temporary identifier associated with the federated learning procedure.

[0294] Aspect 15: A method according to any one of Aspects 1 to 14, wherein receiving the first message includes: receiving the first message during a paging frame or opportunity associated with the federated learning procedure, wherein performing the training procedure is at least partially based on receiving the first message during the paging frame or opportunity associated with the federated learning procedure.

[0295] Aspect 16: A method according to any one of Aspects 1 to 15, wherein receiving the first message includes: receiving a broadcast transmission or a multicast transmission of sidelink control information including a request for the UE to perform the federated learning procedure on a sidelink channel.

[0296] Aspect 17: A method according to any one of Aspects 1 to 16, wherein receiving the first message includes: receiving the first message indicating one or more criteria for participating in the federated learning procedure, wherein the one or more criteria are based at least in part on the version of the machine learning model, the minimum local data set size, the acquisition time of the local data set at the UE, or any combination thereof.

[0297] Aspect 18: A method according to any one of Aspects 1 to 17, wherein sending the second message includes: sending the second message via the one or more resources configured for transmission by multiple UEs including the UE, wherein the multiple UEs are scheduled to send corresponding gradient value sets via the one or more resources according to a common transmission power.

[0298] Aspect 19: The method according to any one of Aspects 1 to 18 further includes: receiving a third message indicating an updated version of the machine learning model based at least in part on the one or more gradient values ​​of the one or more model parameters; and updating the machine learning model based at least in part on the updated version of the machine learning model.

[0299] Aspect 20: The method according to any one of Aspects 1 to 19 further includes: receiving a third message indicating the same version model of the machine learning model and a scalar probability threshold for participating in the federated learning procedure; determining to execute the federated learning procedure based at least in part on a randomly generated value satisfying the scalar probability threshold; executing the training procedure using the machine learning model to obtain a second one or more model parameters based at least in part on the configuration for the federated learning procedure and the randomly generated value satisfying the scalar probability threshold; and sending a fourth message indicating a second one or more gradient values ​​of the second one or more model parameters based at least in part on the configuration for the federated learning procedure.

[0300] Aspect 21: The method according to any one of aspects 1 to 20, wherein the UE operates in an inactive state or an idle state.

[0301] Aspect 22: A method according to any one of Aspects 1 to 21, wherein receiving the first message includes: receiving the first message configuring the UE to send the one or more model parameters in the second message, wherein the second message indicates the one or more model parameters.

[0302] Aspect 23: A method for performing wireless communications at a wireless device, comprising: sending a first message requesting multiple user equipment (UE) to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; receiving multiple second messages indicating multiple sets of gradient values ​​for multiple sets of one or more model parameters via one or more resources configured for air aggregation, based at least in part on the configuration for the federated learning procedure; and training the machine learning model using the multiple sets of the one or more model parameters.

[0303] Aspect 24: The method according to aspect 23, wherein sending the first message includes: sending the first message indicating a range associated with participating in the federated learning procedure, wherein the plurality of second messages are received from UEs within the range.

[0304] Aspect 25: A method according to any one of Aspects 23 to 24, wherein sending the first message includes: sending an indication of the machine learning model from multiple machine learning models, a version of the machine learning model from multiple versions of the machine learning model, an air aggregation scheme from multiple air aggregation schemes, a federated learning procedure from multiple federated learning procedures, or any combination thereof.

[0305] Aspect 26: A method according to any one of Aspects 23 to 25, wherein sending the first message includes: sending an indication of the number of layers in a plurality of layers of the machine learning model, the size of each of the plurality of layers, the order of the plurality of layers, the connectivity of the plurality of layers, or any combination thereof.

[0306] Aspect 27: The method according to any one of Aspects 23 to 26 further includes: sending a third message indicating an updated version of the machine learning model based at least in part on training the machine learning model; and receiving multiple fourth messages indicating a second plurality of sets of gradient values ​​for one or more model parameters via the one or more resources configured for air aggregation, based at least in part on the configuration for the federated learning procedure.

[0307] Aspect 28: The method according to Aspect 27 also includes: determining that training the machine learning model using a second plurality of sets of the one or more model parameters is unsuccessful based at least in part on the excess received power of the plurality of fourth messages; and sending a fifth message requesting the plurality of UEs to execute the federated learning procedure, the fifth message indicating the updated version of the machine learning model and the reduced nominal UE transmit power.

[0308] Aspect 29: An apparatus for performing wireless communications at a UE, comprising: a processor; a memory coupled to the processor; and one or more instructions, the one or more instructions being stored in the memory and executable by the processor to cause the apparatus to perform a method according to any one of Aspects 1 to 22.

[0309] Aspect 30: An apparatus for wireless communication at a UE, comprising: at least one component for performing a method according to any one of aspects 1 to 22.

[0310] Aspect 31: A non-transitory computer-readable medium storing a code for wireless communication at a UE, the code comprising one or more instructions executable by a processor to perform the method according to any one of aspects 1 to 22.

[0311] Aspect 32: An apparatus for performing wireless communications at a wireless device, comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform a method according to any one of Aspects 23 to 28.

[0312] Aspect 33: An apparatus for wireless communication at a wireless device, comprising: at least one component for performing a method according to any one of aspects 23 to 28.

[0313] Aspect 34: A non-transitory computer-readable medium storing code for wireless communication at a wireless device, the code comprising instructions executable by a processor to perform the method according to any one of aspects 23 to 28.

[0314] It should be noted that the methods described herein describe possible implementations, and that the various operations and steps may be rearranged or otherwise modified and that other implementations are possible. Furthermore, aspects from two or more methods may be combined.

[0315] Although aspects of LTE, LTE-A, LTE-A Pro, or NR systems may be described for example purposes, and LTE, LTE-A, LTE-A Pro, or NR terminology may be used in much of the description, the techniques described herein may also be applicable to networks other than LTE, LTE-A, LTE-A Pro, or NR networks. For example, the described techniques may be applicable to various other wireless communication 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.

[0316] The information and signals described herein may be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the specification may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0317] The various illustrative blocks and components described in conjunction with the disclosure herein may be implemented or performed using a general purpose processor, DSP, ASIC, CPU, FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in an alternative, the processor may be any processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0318] The functions described herein may be implemented using hardware, software executed by a processor, firmware, or any combination thereof. When implemented using software executed by a processor, the functions may be stored as one or more instructions or codes of a computer-readable medium or sent using one or more instructions or codes of a computer-readable medium. Other examples and specific implementations are within the scope of the present disclosure and the appended claims. For example, due to the nature of software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these items. Features that implement the functions may also be physically located at different locations, including being distributed so that the various parts of the functions are implemented at different physical locations.

[0319] Computer-readable medium includes both non-transient computer storage medium and communication medium, and it includes any medium that facilitates computer program to be transmitted from one position to another position.Non-transient storage medium can be any available medium that can be accessed by general or special-purpose computer.By way of example and not limitation, non-transient computer-readable medium can include RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory, compact disc (CD) ROM or other optical disc storage device, disk storage device or other magnetic storage device, or can be used for carrying or storing desired program code parts and any other non-transient medium that can be accessed by general or special-purpose computer or general or special-purpose processor in the form of instruction or data structure.In addition, any connection is appropriately referred to as computer-readable medium.For example, if software is sent from website, server or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) or wireless technology such as infrared, radio and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL or wireless technology such as infrared, radio and microwave are included in the definition of computer-readable medium. Disks and optical disks used herein include CDs, laser optical disks, optical optical disks, digital versatile disks (DVDs), floppy disks, and Blu-ray disks. Disks can reproduce data magnetically, and optical disks can reproduce data optically using lasers. Combinations of the above are also included within the scope of computer-readable media.

[0320] As used herein (including in the claims), "or" used in a list of items (e.g., a list of items followed by a phrase such as "at least one of" or "one or more of") indicates an inclusive list, so that, 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). In addition, as used herein, the phrase "based on" should not be interpreted as a reference to a closed set of conditions. For example, an example step described as "based on condition A" can 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" should be interpreted in the same manner as the phrase "based at least in part on."

[0321] The terms "determine" or "determining" encompass a variety of actions, and thus, "determining" may include calculating, computing, processing, deriving, investigating, looking up (such as via looking up in a table, database or other data structure), ascertaining, and the like. Furthermore, "determining" may include receiving (e.g., receiving information), accessing (e.g., accessing data stored in a memory), etc. Additionally, "determining" may include parsing, obtaining, selecting, choosing, establishing, and other such similar actions.

[0322] In the drawings, similar components or features may have the same reference label. In addition, various components of the same type may be distinguished by following the reference label with a dash and a second label to distinguish between similar components. If only the first reference label is used in the specification, the description may apply to any of the similar components having the same first reference label, regardless of the second or other subsequent reference labels.

[0323] The descriptions set forth herein in conjunction with the accompanying drawings describe example configurations and do not represent all examples that may be implemented or within the scope of the claims. The term "example" as used herein means "used as an example, instance, or illustration," rather than "preferred" or "advantageous over other examples." The detailed description includes specific details to provide an understanding of the described techniques. However, these techniques may be practiced without these specific details. In some cases, known structures and devices are shown in block diagram form to avoid obscuring the concepts of the described examples.

[0324] The description herein is provided to enable one of ordinary skill in the art to implement or use the present disclosure. Various modifications to the present disclosure will be apparent to one of ordinary skill 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 should be granted the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An apparatus for wireless communication at a user equipment (UE), the apparatus comprising: processor; a memory coupled to the processor; and One or more instructions, the one or more instructions are stored in the memory and can be executed by the processor to cause the apparatus to perform the following operations based at least in part on the one or more instructions: receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; In response to the first message requesting the UE to perform the federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based at least in part on the configuration for the federated learning procedure; as well as Based at least in part on the configuration for the federated learning procedure, a second message indicating one or more gradient values ​​of the one or more model parameters is sent via one or more resources configured for over-the-air aggregation.

2. The apparatus of claim 1 , wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: The first message is received indicating a range associated with participating in the federated learning procedure, wherein sending the second message is based at least in part on determining that the UE is within the range.

3. The apparatus of claim 1 , wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: receiving the first message indicating a reference signal and a threshold; and The reference signal is measured to obtain a measurement of the reference signal, wherein sending the second message is based at least in part on the measurement of the reference signal exceeding the threshold.

4. The apparatus of claim 1 , wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: The first message is received indicating a nominal transmit power for transmitting the second message on the one or more resources configured for the air aggregation, wherein transmitting the second message is based at least in part on the nominal transmit power.

5. The apparatus of claim 4, wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving an indication of a path loss reference signal and an indication of a reference signal power value; and The path loss reference signal is measured, wherein sending the second message is based at least in part on the measurement of the path loss reference signal and the reference signal power value.

6. The apparatus of claim 4, wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving an indication of the location of the server; and A path loss estimate for the second message is determined based at least in part on a path loss model and a distance between the UE and the location of the server, wherein sending the second message is based at least in part on the path loss estimate.

7. The apparatus of claim 1 , wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: Receive an indication of the machine learning model from a plurality of machine learning models.

8. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: Receiving an indication of a number of layers in a plurality of layers of the machine learning model, a size of each of the plurality of layers, an order of the plurality of layers, connectivity of the plurality of layers, or any combination thereof, wherein performing the training procedure is based at least in part on the number of layers, the size of each of the plurality of layers, the order of the plurality of layers, the connectivity of the plurality of layers, or any combination thereof.

9. The apparatus of claim 1 , wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: Receiving an indication of a version of the machine learning model from a plurality of versions of the machine learning model, wherein executing the training procedure is based at least in part on the version of the machine learning model.

10. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: An indication of an air aggregation scheme from a plurality of air aggregation schemes is received.

11. The apparatus of claim 1 , wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving, in one or more paging frames, an indication of a plurality of occasions of the first message requesting the UE to perform the federated learning procedure; and The first message is monitored based at least in part on the indication of the plurality of opportunities for the first message.

12. The apparatus of claim 1, wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving a wake-up signal associated with the first message; and Monitoring for the first message requesting the UE to perform the federated learning procedure based at least in part on receiving the wake-up signal.

13. The apparatus of claim 1, wherein the one or more instructions are further executable by the processor to cause the apparatus to: decoding the first message based at least in part on a paging radio network temporary identifier; and One or more fields corresponding to the federated learning procedure are detected in the first message, wherein performing the training procedure is based at least in part on detecting the one or more fields.

14. The apparatus of claim 1, wherein the one or more instructions are further executable by the processor to cause the apparatus to: The first message is decoded based at least in part on a radio network temporary identifier associated with the federated learning procedure, wherein performing the training procedure is based at least in part on decoding the first message based at least in part on the radio network temporary identifier associated with the federated learning procedure.

15. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: The first message is received during a paging frame or opportunity associated with the federated learning procedure, wherein performing the training procedure is based at least in part on receiving the first message during the paging frame or opportunity associated with the federated learning procedure.

16. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: A broadcast transmission or a multicast transmission including sidelink control information requesting the UE to perform the federated learning procedure is received on a sidelink channel.

17. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: Receive the first message indicating one or more criteria for participating in the federated learning procedure, wherein the one or more criteria are based at least in part on a version of the machine learning model, a minimum local data set size, an acquisition time of the local data set at the UE, or any combination thereof.

18. The apparatus of claim 1, wherein the one or more instructions for sending the second message are executable by the processor to cause the apparatus to: The second message is sent via the one or more resources configured for transmission by a plurality of UEs including the UE, wherein the plurality of UEs are scheduled to send respective sets of gradient values ​​via the one or more resources according to a common transmit power.

19. The apparatus of claim 1, wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving a third message indicating an updated version of the machine learning model based at least in part on the one or more gradient values ​​of the one or more model parameters; and updating the machine learning model based at least in part on the updated version of the machine learning model.

20. The apparatus of claim 1, wherein the one or more instructions are further executable by the processor to cause the apparatus to: receiving a third message indicating an identical version model of the machine learning model and a scalar probability threshold for participating in the federated learning procedure; determining to execute the federated learning procedure based at least in part on the randomly generated value satisfying the scalar probability threshold; performing the training procedure using the machine learning model to obtain second one or more model parameters based at least in part on the configuration for the federated learning procedure and that the randomly generated value satisfies the scalar probability threshold; and A fourth message is sent indicating second one or more gradient values ​​of the second one or more model parameters based at least in part on the configuration for the federated learning procedure.

21. The apparatus of claim 1, wherein the UE operates in an inactive state or an idle state.

22. The apparatus of claim 1, wherein the one or more instructions for receiving the first message are executable by the processor to cause the apparatus to: The first message is received to configure the UE to send the one or more model parameters in the second message, wherein the second message indicates the one or more model parameters.

23. An apparatus for wireless communication at a wireless device, the apparatus comprising: processor; a memory coupled to the processor; and One or more instructions, the one or more instructions are stored in the memory and can be executed by the processor to cause the apparatus to perform the following operations based at least in part on the one or more instructions: sending a first message requesting a plurality of user equipments (UEs) to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; Based at least in part on the configuration for the federated learning procedure, receiving, via one or more resources configured for over-the-air aggregation, a plurality of second messages indicating a plurality of sets of gradient values ​​for a plurality of sets of one or more model parameters; as well as The machine learning model is trained using multiple sets of the one or more model parameters.

24. The apparatus of claim 23, wherein the one or more instructions for sending the first message are executable by the processor to cause the apparatus to: The first message is sent indicating a range associated with participating in the federated learning procedure, wherein the plurality of second messages are received from UEs within the range.

25. The apparatus of claim 23, wherein the one or more instructions for sending the first message are executable by the processor to cause the apparatus to: Sending instructions for the machine learning model from multiple machine learning models, a version of the machine learning model from multiple versions of the machine learning model, an over-the-air aggregation scheme from multiple over-the-air aggregation schemes, a federated learning procedure from multiple federated learning procedures, or any combination thereof.

26. The apparatus of claim 23, wherein the one or more instructions for sending the first message are executable by the processor to cause the apparatus to: Sending an indication of the number of layers in the plurality of layers of the machine learning model, the size of each of the plurality of layers, the order of the plurality of layers, the connectivity of the plurality of layers, or any combination thereof.

27. The apparatus of claim 23, wherein the one or more instructions are further executable by the processor to cause the apparatus to: sending a third message indicating an updated version of the machine learning model based at least in part on training the machine learning model; and Based at least in part on the configuration for the federated learning procedure, a plurality of fourth messages indicative of a second plurality of sets of gradient values ​​for a second plurality of sets of one or more model parameters are received via the one or more resources configured for over-the-air aggregation.

28. The apparatus of claim 27, wherein the one or more instructions are further executable by the processor to cause the apparatus to: determining that training the machine learning model using the second plurality of sets of the one or more model parameters was unsuccessful based at least in part on the excess received power of the plurality of fourth messages; and Sending a fifth message requesting the plurality of UEs to perform the federated learning procedure, the fifth message indicating the updated version of the machine learning model and the reduced nominal UE transmit power.

29. A method for wireless communication at a user equipment (UE), the method comprising: receiving a first message requesting the UE to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; In response to the first message requesting the UE to perform the federated learning procedure, performing a training procedure using the machine learning model to obtain one or more model parameters based at least in part on the configuration for the federated learning procedure; as well as Based at least in part on the configuration for the federated learning procedure, a second message indicating one or more gradient values ​​of the one or more model parameters is sent via one or more resources configured for over-the-air aggregation.

30. A method for wireless communication at a wireless device, the method comprising: sending a first message requesting a plurality of user equipments (UEs) to perform a federated learning procedure, the first message indicating a machine learning model and a configuration for the federated learning procedure; Based at least in part on the configuration for the federated learning procedure, receiving, via one or more resources configured for over-the-air aggregation, a plurality of second messages indicating a plurality of sets of gradient values ​​for a plurality of sets of one or more model parameters; as well as The machine learning model is trained using multiple sets of the one or more model parameters.

Images