Control plane initiated delivery of machine learning model for wireless network via user plane

Abstract

A method includes receiving, by a user equipment from a network node, a control message for controlling transmission of a machine learning (ML) model, the control message including a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol, the protocol is to be used by the user equipment for transmission of the ML model via the user plane; and performing, by the user equipment via a protocol, a transmission of the ML model between the user equipment and the storage location indicated by the storage information.

Description

Technical Field

[0001] This specification relates to wireless communications. Background Art

[0002] A communication system may be a facility that enables communication between two or more nodes or devices (such as fixed or mobile communication devices). Signals may be carried on wired or wireless carriers.

[0003] An example of a cellular communication system is the architecture standardized by the Third Generation Partnership Project (3GPP). The latest developments in this area are often referred to as 4G, or the Long Term Evolution (LTE) of 3G, or the Universal Mobile Telecommunications System (UMTS) radio access technology. E-UTRA (Evolved UMTS Terrestrial Radio Access) is the air interface of the 3GPP Long Term Evolution (LTE) upgrade path for mobile networks. In LTE, base stations or access points (APs) called enhanced Node Bs (eNBs) provide wireless access within a coverage area or cell. In LTE, mobile devices or mobile stations are referred to as user equipment (UE). LTE includes many improvements or developments. Various aspects of LTE are constantly being improved.

[0004] 5G New Radio (NR) development is part of the ongoing mobile broadband evolution process to meet the requirements of 5G, similar to the earlier evolution of 3G and 4G wireless networks. In addition, 5G targets emerging use cases beyond mobile broadband. The goal of 5G is to significantly improve wireless performance, which can include new levels of data rates, latency, reliability, and security. 5G NR can also be extended to efficiently connect the massive Internet of Things (IoT) and can provide new types of mission-critical services. For example, ultra-reliable low-latency communication (URLLC) equipment may require high reliability and very low latency. 6G and other wireless networks are also under development or will be developed in the near future. Summary of the Invention

[0005] A method may include: receiving, by a user equipment, a control message for controlling transmission of a machine learning (ML) model from a network node, the control message including: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and performing, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0006] An apparatus may include: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, cause the apparatus to at least: receive, by a user equipment, a control message for controlling transmission of a machine learning (ML) model from a network node, the control message including: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and perform, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0007] A non-transitory computer-readable storage medium may include instructions stored thereon, which, when executed by at least one processor, cause a computing system to: receive, by a user equipment (UE), a control message for controlling transmission of a machine learning (ML) model from a network node, the control message including: a command for the UE to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the UE for transmission of the ML model via a user plane; and perform, by the UE, transmission of the ML model between the UE and the storage location indicated by the storage information, via the protocol.

[0008] An apparatus may include: a component for receiving, by a user equipment, a control message for controlling transmission of a machine learning (ML) model from a network node, the control message including: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and a component for performing, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0009] The details of one or more examples of embodiments are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a block diagram of a wireless network according to an example embodiment.

[0011] Figure 2 is a flow chart illustrating the operation of a user equipment (or UE) according to an example embodiment.

[0012] Figure 3 is a diagram illustrating a machine learning (ML) model downloaded by a user equipment (UE) or user device according to an example embodiment.

[0013] Figure 4is a diagram illustrating a machine learning (ML) model uploaded by a user equipment (UE) or user device according to an example embodiment.

[0014] Figure 5 is a diagram illustrating an ML model downloading process controlled by an LMF according to an example embodiment.

[0015] Figure 6 is a diagram illustrating an ML model downloading process controlled by an LMF according to an example embodiment.

[0016] Figure 7 is a block diagram of a wireless station or node (e.g., a user node, user equipment or UE, a network node, a relay node, a gNB, or other node). DETAILED DESCRIPTION

[0017] Figure 1 is a block diagram of a wireless network 130 according to an example embodiment. Figure 1 In a wireless network 130, user devices 131, 132, 133, and 135 (also referred to as mobile stations (MSs) or user equipment (UEs)) can connect to (and communicate with) a base station (BS) 134, which can also be referred to as an access point (AP), an enhanced Node B (eNB), a gNB, or a network node. The terms user device and user equipment (UE) can be used interchangeably. The BS can also include or be referred to as a RAN (Radio Access Network) node, and can include a portion of a BS or a portion of a RAN node, such as (for example, a centralized unit (CU) and / or a distributed unit (DU) in the case of a split BS or split gNB). At least a portion of the functionality of a BS (e.g., an access point (AP), a base station (BS), or an (e)Node B (eNB), a gNB, a RAN node) can also be performed by any node, server, or host that can be operably coupled to a transceiver (such as a remote radio head). BS (or AP) 134 provides wireless coverage within cell 136, including to user equipment (or UE) 131, 132, 133, and 135. Although only four user equipment (or UE) are shown connected or attached to BS 134, any number of user equipment may be provided. BS 134 is also connected to core network 150 via S1 interface 151. A location management function (LMF) is also connected to BS / gNB 134 and core network 150. This is just one simple example of a wireless network, and other wireless networks may be used.

[0018] A base station (e.g., such as BS 134) is an example of a radio access network (RAN) node within a wireless network. A BS (or RAN node) may be or may include (or may alternatively be referred to as) an access point (AP), a gNB, an eNB, or a portion thereof (such as a centralized unit (CU) and / or distributed unit (DU) in the case of a split BS or split gNB), or other network nodes. As an example, a network node may refer to or include a BS, an AP, a gNB, a CU and / or a DU, or a RAN node. In addition, at least in some cases, a network node may also refer to or include a core network (e.g., a core network node or core network entity, such as an AMF (Access and Mobility Function) or other core network entity), an LMF (Location Management Function), or other network nodes.

[0019] According to an illustrative example, a BS node (e.g., BS, eNB, gNB, CU / DU, etc.) or a radio access network (RAN) can be part of a mobile telecommunications system. The RAN (radio access network) can include one or more BSs or RAN nodes that implement a radio access technology, for example, to allow one or more UEs to access a network or core network. Thus, for example, a RAN (RAN node, such as a BS or gNB) can reside between one or more user equipment or UEs and a core network. According to an example embodiment, each RAN node (e.g., BS, eNB, gNB, CU / DU, etc.) or BS can provide one or more wireless communication services to one or more UEs or user equipment, for example, to allow the UE to access the network wirelessly via the RAN node. Each RAN node or BS can perform or provide wireless communication services, for example, such as allowing a UE or user equipment to establish a wireless connection to a RAN node and to send data to and / or receive data from one or more UEs. For example, after establishing a connection to a UE, a RAN node or network node (e.g., a BS, eNB, gNB, CU / DU, etc.) may forward data received from the network or core network to the UE, and / or forward data received from the UE to the network or core network. A RAN node or network node (e.g., a BS, eNB, gNB, CU / DU, etc.) may perform a variety of other wireless functions or services, such as broadcasting control information (e.g., such as system information or on-demand system information) to the UE, paging the UE when there is data to be delivered to the UE, assisting the UE in handover between cells, scheduling resources for uplink data transmission from (multiple) UEs and downlink data transmission to (multiple) UEs, issuing control information for configuring one or more UEs, etc. These are just a few examples of one or more functions that a RAN node or BS may perform.

[0020] A user device or user node (user terminal, user equipment (UE), mobile terminal, handheld wireless device, etc.) may refer to a portable computing device, including a wireless mobile communication device that operates with or without a subscriber identity module (SIM), such as, but not limited to, a mobile station (MS), a mobile phone, a cell phone, a smartphone, a personal digital assistant (PDA), a handset, a device using a wireless modem (such as an alarm or measurement device), a laptop and / or touch screen computer, a tablet, a tablet phone, a game console, a notebook computer, a vehicle, a sensor, a multimedia device, or any other wireless device. It should be understood that a user device may also be (or may include) an almost exclusively uplink-only device, an example of which is a camera or video camera that uploads images or video clips to a network. In addition, a user node may include a user equipment (UE), a user device, a user terminal, a mobile terminal, a mobile station, a mobile node, a subscriber device, a subscriber node, a subscriber terminal, or other user nodes. For example, a user node may be configured to communicate wirelessly with one or more network nodes (e.g., gNB, eNB, BS, AP, CU, DU, CU / DU) and / or with one or more other user nodes, regardless of the technology or radio access technology (RAT). In LTE (as an illustrative example), the core network 150 may be referred to as an evolved packet core (EPC), which may include a mobility management entity (MME), one or more gateways, and other control functions or blocks. The MME may handle or assist in the movement / handover of user equipment between BSs, and the gateway may forward data and control signals between the BS and a packet data network or the Internet. Other types of wireless networks, such as 5G (which may be referred to as new radio (NR)), may also include a core network.

[0021] Furthermore, the techniques described herein can be applied to various types of user devices or data service types, or can be applied to user devices that can have multiple applications running on them, each of which can be of different data service types. New Radio (5G) developments can support many different applications or many different data service types, such as, for example, machine type communications (MTC), enhanced machine type communications (eMTC), Internet of Things (IoT) and / or narrowband IoT user devices, enhanced mobile broadband (eMBB), and ultra-reliable low latency communications (URLLC). Many of these new 5G (NR)-related applications may require higher performance than previous wireless networks.

[0022] The IoT can refer to the growing group of objects that can have internet or network connectivity, allowing them to send and receive information to other network devices. For example, many sensor-type applications or devices can monitor physical conditions or states and report to servers or other network devices, such as when events occur. For example, machine-type communication (MTC or machine-to-machine communication) can be characterized as fully automated data generation, exchange, processing, and actuation between intelligent machines, with or without human intervention. Enhanced mobile broadband (eMBB) can support higher data rates than currently available in LTE.

[0023] Ultra-Reliable Low Latency Communication (URLLC) is a new data service type or new use case that the New Radio (5G) system can support. This enables emerging new applications and services such as industrial automation, autonomous driving, vehicle safety, e-health services, etc. As an illustrative example, 3GPP aims to provide a 5G-compatible 5G network with the same performance as 10G. -5 Connectivity with reliability corresponding to a block error rate (BLER) of 100% and a U-plane (user / data plane) latency of up to 1 ms. Thus, for example, a URLLC user equipment / UE may require a significantly lower block error rate and low latency (with or without simultaneous high reliability requirements) than other types of user equipment / UE. Thus, for example, a URLLC UE (or a URLLC application on a UE) may require shorter latency than an eMBB UE (or an eMBB application running on a UE).

[0024] The techniques described herein can be applied to a variety of wireless technologies or wireless networks, such as 5G (New Radio (NR)), cmWave and / or mmWave band networks, IoT, MTC, eMTC, eMBB, URLLC, 6G, etc., or any other wireless network or wireless technology. These example networks, technologies, or data service types are provided as illustrative examples only.

[0025] According to example embodiments, a machine learning (ML) model may be used within a wireless network to perform (or assist in performing) one or more tasks or functions. Generally, one or more nodes within a wireless network (e.g., a BS, gNB, eNB, RAN node, user node, UE, user equipment, relay node, or other wireless node) may use or employ an ML model, such as, for example, a neural network model (e.g., which may be referred to as a neural network, an artificial intelligence (AI) neural network, an AI neural network model, an AI model, an AI machine learning (AIML) model, or an algorithm, model, or other terminology) to perform or assist in performing one or more ML-enabled tasks or functions. ML-enabled tasks may include tasks that can be performed (or assisted in performing) by an ML model, or tasks that an ML model has been trained to perform (or assist in performing).

[0026] ML-based algorithms or ML models can be used to perform and / or assist in performing various wireless-related functions, such as radio resource management (RRM) functions, or functions or tasks for improving network performance, such as, for example, beam prediction (e.g., predicting the best beam or best beam pair based on measured reference signals) in a UE, antenna panel or beam steering, RRM (radio resource measurement) measurement and feedback (channel state information (CSI) feedback), CSI report compression, link monitoring, transmit power control (TPC), etc. In some cases, the use of ML models can be used to improve the performance of a wireless network in one or more aspects, or as measured by one or more performance indicators or performance criteria.

[0027] For example, an ML model can be or include a computational model used in machine learning that is composed of hierarchically organized nodes. Nodes are also called artificial neurons, or simply neurons, and perform a function on a given input to produce a certain output value. A neural network or ML model may typically require a training period to learn the parameters, i.e., weights, used to map inputs to desired outputs. The mapping is performed via a function. Therefore, weights are the weights of the neural network's mapping function. Each neural network model or ML model can be trained for a specific task.

[0028] In order to provide an output given an input, a neural network model or ML model should be trained, which can involve learning appropriate values ​​for a large number of parameters (e.g., weights) of a mapping function. These parameters are also commonly referred to as weights because they are used to weight the terms in the mapping function. This training can be an iterative process in which the values ​​of the weights are adjusted in multiple rounds (e.g., thousands of rounds) of training until the best or most accurate values ​​(or weights) are reached. In the context of a neural network (neural network model) or ML model, the parameters can typically be initialized using random values, and the training optimizer iteratively updates the parameters (weights) of the neural network to minimize the error of the mapping function. In other words, in each round or step of iterative training, the network updates the values ​​of the parameters so that the values ​​of the parameters eventually converge to the optimal values.

[0029] For example, neural network models or ML models can be trained in a supervised or unsupervised manner. In supervised learning, training examples are provided to a neural network model or other machine learning algorithm. The training examples include inputs and expected or previously observed outputs. The training examples are also called labeled data because the inputs are labeled with the expected or observed outputs. In the case of a neural network, the network learns the values ​​of the weights used in the mapping function that most often result in the expected output given the training inputs. In unsupervised training, the neural network model learns to identify structure or patterns in the provided inputs. In other words, the model identifies implicit relationships in the data. Unsupervised learning is used in many machine learning problems and typically requires large amounts of unlabeled data.

[0030] According to example embodiments, the learning or training of a neural network model or ML model can be divided into (or can include) multiple categories (including supervised and unsupervised), depending on whether the model has a learning "signal" or "feedback" available. Thus, for example, in the field of machine learning, there can be two main types of learning or training of a model: supervised and unsupervised. The main difference between these two types is that supervised learning is done using known or prior knowledge, i.e., what the output value of certain data samples should be. Thus, the goal of supervised learning can be to learn a function that, given a data sample and a desired output, best approximates the relationship between the input and output that can be observed in the data. On the other hand, unsupervised learning has no labeled outputs, so its goal is to infer the natural structure that exists within the set of data points. ML model training can also include reinforcement learning.

[0031] Challenges exist regarding the techniques that should be used to transmit or transfer ML models to UEs, such as which messages or information should be provided to initiate ML model transmission, and whether to use the control plane and / or user (or data) plane to initiate and / or transfer ML model transmission. Generally, the control plane may include control messages transmitted to provide control of various aspects or functions of the wireless network, such as control messages that may be transmitted to coordinate or control connection establishment, UE handover or cell change, power control, configuring the UE to perform a certain function, and so on. The data or user plane may generally include the transmission of user data to or from the UE. The control plane may generally include control messages, such as Radio Resource Control (RRC) messages sent to or received from a gNB, CU / DU, or other RAN node. Other types of control messages may include, for example, LPP (LTE Positioning Protocol) control messages sent to or received from a Location Management Function (LMF), or Network Access Stratum (NAS) control messages sent to or received from the core network (e.g., such as the Access and Mobility Function (AMF)). The user plane may include the transmission of data to and / or from the UE, such as using one or more user plane data transmission protocols (e.g., such as File Transfer Protocol (FTP), Hypertext Transfer Protocol (HTTP), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), or other user plane protocols). Furthermore, at least in some cases, packets or messages sent via the control plane may typically have a different priority and / or QoS (Quality of Service) than data transmitted via the user plane. It is currently unclear how the user plane and / or control plane may be used to allow the UE to download the ML model (or otherwise facilitate the transmission of the ML model by the UE).

[0032] Compared to control messages (or control plane messages), the user (or data plane) may generally have a greater capacity to transmit larger files or larger data blocks. According to example embodiments, the control plane (including one or more control messages, such as, for example, RRC messages, NAS messages, and / or LPP messages or other control plane messages) may have very limited capacity to convey large files or important information (e.g., such as ML models or ML model parts), but the control messages or control plane may be advantageously used to provide control and / or communication to coordinate and / or initiate and / or confirm the transmission of ML models. Similarly, the user (or data) plane can accommodate larger data transmissions, and therefore, after the ML model transmission is coordinated, controlled, or initiated via the control plane (e.g., via one or more control messages). Therefore, the user equipment or UE may use the user (or data) plane to perform ML model transmission (e.g., downloading an ML model or uploading an ML model).

[0033] Thus, according to example embodiments, control message(s) (e.g., provided via a control plane) may be used to initiate, coordinate, confirm, and / or otherwise control the transmission of an ML model to or from a UE, while a user (or data) plane may be used by the UE to perform ML model transmission (upload or download). According to example embodiments, the transmitted ML model may include, for example, a trained or untrained model, and may include a complete ML model, a portion of a (or partial) ML model, and / or a delta or difference indicating a change (or delta) to a (or another) ML model.

[0034] Capability Exchange

[0035] The capability exchange may be a bidirectional message flow initiated by a network node or network entity, such as a gNB, an access and mobility function (AMF) in the core network, or a location management function (LMF). The purpose may be to synchronize with the UE to understand which features specified in the 3GPP specifications the UE supports. According to example embodiments, the capability exchange process may be enhanced to allow the UE (and / or network node) to indicate the ability to transfer (e.g., upload and / or download) ML models, and may also indicate one or more user (or data) plane protocols supported by the UE for downloading or uploading ML models.

[0036] Command and response messaging

[0037] According to an example embodiment, the UE and the network node may use a command response message structure to initiate ML model transmission, control confirmation of ML model transmission, and / or indicate failure of ML model transmission via a control plane message (e.g., an RRC message or other control message). For example, the basic protocol structure may use or implement messages in the form of Command, CommandResponse, CommandFailure, and / or CommandComplete. A Command may typically be issued by the network (or network node) to the UE. A CommandResponse may be issued by the UE, for example, to request additional data or additional information. For example, when a process (e.g., ML model transmission) is successfully completed, the network may issue a CommandComplete to the UE for upload, or the UE may issue a CommandComplete to the network for download, and when the process (ML model transmission) fails, the network node may send a CommandFailure to the UE for model transmission, or the UE may send a CommandFailure to the network for ML model download.

[0038] Transmission

[0039] According to example embodiments, one or more user (or data) plane protocols may be used for ML model transmission, for example, raw data transmission such as HTTP, FTP, TCP, and UDP, as a non-exhaustive list of example protocols. Each of these protocols is a different well-known base protocol for data transmission, for example for downloading and uploading. They each have trade-offs that need to be considered, such as reliability and overhead of transmission, but these depend on implementation details. The key is that there will be a data transmission protocol, and the UE and relevant network entities will need to communicate the ability to use the data transmission protocol. According to example embodiments, the control message provided by the network node to the UE for initiating or requesting ML model transmission may indicate the (e.g., user or data plane) protocol (e.g., indicating one of these protocols) that the UE should use to perform ML model transmission.

[0040] Control plane and user plane

[0041] According to example embodiments, the control plane may, for example, include or may provide signaling radio bearers (SRBs) and may be used to send and receive generally structured messages to initiate, request, configure, coordinate, confirm and / or control the UE or user equipment to perform network connection procedures, make and report measurements (e.g., issue CSI reports), and establish data connections, and perform other functions. Typically, the control plane is highly reliable and is used for small amounts of control data. The user plane supports large amounts of data through data radio bearers (DRBs) and is typically used for application layer services, such as the above-mentioned data transmission protocols. According to example embodiments, as described in more detail below, the various embodiments described herein may use a control plane (e.g., one or more control messages to initiate / request ML model transmission, and / or indicate complete / successful ML model transmission, and / or indicate failure of ML model transmission) and a user (or data) plane to perform ML model transmission (e.g., upload or download of ML models). In addition, the user (or data) plane may use a UPF (User Plane Function).

[0042] Figure 2 2 is a flow chart illustrating the operation of a user equipment (or UE) according to an example embodiment. Operation 210 includes receiving, by the user equipment, a control message from a network node for controlling the transmission of a machine learning (ML) model, the control message including: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane. Furthermore, operation 220 includes performing, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via a protocol. The ML model transmission may be performed by the UE via the user (or data) plane using the protocol indicated by the protocol information included in the control message.

[0043] about Figure 2 In the method, the storage information indicating the storage location of the ML model may include: a network address of the host node; and at least one of the following: a path for the ML model on the host node; or a file name.

[0044] about Figure 2 In the method, the control message may further include an ML model metadata container, the ML model metadata container including: at least one of the following ML model metadata: an ML model identifier, uniquely identifying the ML model in the user equipment or in the network; a function identifier, identifying a function to be performed by the ML model; and / or an area indication, identifying an area or one or more cells for which the ML model is valid or to be used.

[0045] about Figure 2 In the method, the ML model may include at least one of: a complete ML model; a portion of the ML model; or a delta or change in the ML model indicating a change or difference in the ML model, for example, compared to another ML model that the UE may already have.

[0046] about Figure 2 In the method, the network node may include at least one of the following: a gNB; an eNB; a base station or an access point; a centralized unit (CU) and / or a distributed unit (DU); a radio access network (RAN) node; an access and mobility function (AMF); a core network or a core network node; or a location management function (LMF).

[0047] about Figure 2 In the method, the control message may include at least one of the following: a first radio resource control (RRC) control message received from a gNB or a RAN (radio access network) node; a first LPP (LTE positioning protocol) control message received from a location management function (LMF); or a first NAS (network access layer) control message received from a core network.

[0048] about Figure 2 The method may further include: the UE sending at least one of the following messages regarding the transmission of the ML model: a second radio resource control (RRC) control message sent to a gNB or a RAN (radio access network) node in response to the first RRC control message; a second LPP (LTE Positioning Protocol) control message sent to a location management function (LMF) in response to the first LPP control message; or a second NAS (Network Access Stratum) control message sent to a core network in response to the first NAS control message.

[0049] about Figure 2In the method, a protocol to be used for transmission of the ML model between the user device and the storage location indicated by the storage information via the user plane may include, by way of example, at least one of the following: File Transfer Protocol (FTP); Hypertext Transfer Protocol (HTTP); Transmission Control Protocol (TCP); or User Datagram Protocol (UDP). Other protocols may be used to transmit the ML model.

[0050] about Figure 2 The method may further include: receiving, by the user equipment, a capability request from the network node; and sending, by the user equipment, a capability response to the network node, where the capability response indicates that the user equipment has the capability to perform ML model transmission.

[0051] about Figure 2 The method may include a download command that instructs the user device to download the ML model from the storage location for the ML model indicated by the storage information to the user device; the method further includes: downloading the ML model from the storage location for the ML model by the user device.

[0052] about Figure 2 The method may include a control message including an ML model metadata container, the ML model metadata container including at least a function identifier, the function identifier identifying: a function to be performed by the ML model, wherein the method further includes: performing, by the user device, the function specified by the function identifier using the downloaded ML model.

[0053] about Figure 2 The method includes storing information indicating a storage location from which the ML model can be downloaded, including a network address of a host node storing the ML model, and at least one of a path for the ML model on the host node or a file name associated with the ML model.

[0054] about Figure 2 The method may include a download command, the download command instructing a user equipment (e.g., UE) to download the ML model from a storage location for the ML model; the method may also include providing, by the user equipment, a download completion indication or a download failure indication to the network node, the download completion indication indicating that the download of the ML model is completed by the user equipment, and the download failure indication indicating that the download of the ML model fails.

[0055] about Figure 2The method of claim 1, wherein the command may include a download command instructing the user equipment to download the ML model from a storage location for the ML model; and wherein the control message further indicates a size of the ML model; the method further includes: determining, by the user equipment, whether there are sufficient storage resources at the user equipment to store and / or use the ML model; and if there are insufficient storage resources at the user equipment to store and / or use the ML model, sending, by the user equipment, a control message to the network node, the control message including: a failure indication indicating that downloading of the ML model failed, including providing a failure reason indicating insufficient storage resources.

[0056] about Figure 2 In a method for transmitting an ML model to a user device, the command may include a download command instructing the user device to download the ML model from a storage location for the ML model; wherein the control message may include a first checksum or hash for the ML model, the first checksum or hash being used by the user device to verify the integrity of the ML model that can be downloaded by the user device. Furthermore, for example, performing the transmission of the ML model may include: downloading the ML model from the storage location for the ML model by the user device. The method may also include: calculating a second checksum or hash of the downloaded ML model; comparing the second checksum or hash with the first checksum or hash to determine whether there is a match; determining whether the integrity of the downloaded ML model is verified based on whether the second checksum or hash matches the first checksum or hash; and transmitting, by the user device to the network node, any one of the following: a model download completion indication, confirming completion of the download of the ML model if the integrity of the ML model is verified; or a model download failure indication, indicating a failure to download the ML model if the integrity of the ML model is not verified, including providing a failure reason indicating the verification failure.

[0057] about Figure 2 In the method, the command may include an upload command instructing the user device to upload the ML model to a storage location of the ML model.

[0058] about Figure 2 In the method, the storage information may indicate a storage location to which the user device may upload the ML model, including a network address of a host node for storing the ML model, and at least one of a path for the ML model on the host node and a file name associated with the ML model.

[0059] about Figure 2 The method may include an upload command that instructs the user device to upload the ML model to a storage location of the ML model; the method further includes: uploading, by the user device, the ML model to the storage location via a protocol indicated by the protocol information included in the control message.

[0060] about Figure 2 The method may further include: calculating, by the user device, a checksum or hash of the uploaded ML model; and sending, by the user device, the checksum or hash of the uploaded ML model to the network node to allow the network node to determine the integrity of the uploaded ML model.

[0061] about Figure 2 The method may further include: receiving, by the user equipment from the network node, any one of the following: an upload completion indication indicating that the upload of the ML model is completed by the user equipment and is successfully received at the storage location, or an upload failure indication indicating that the upload of the ML model fails.

[0062] about Figure 2 The method may further include: if the user equipment receives an indication of a failure to upload the ML model, performing the following: the user equipment re-uploading the ML model to the storage location via the protocol indicated by the protocol information; and the user equipment sending or re-sending a checksum or hash of the re-uploaded ML model to the network node to allow the network node to determine the integrity of the re-uploaded ML model.

[0063] Various techniques are described that can provide or enable ML model delivery or transmission between a UE and a network (such as a 3GPP wireless network), which can be or include, for example, a RAN node or gNodeB / gNB (e.g., which can be or include a split arrangement of a centralized unit (CU) and / or a distributed unit (DU)), a LMF or core network function (CNF), or other network node. For example, it can be desirable or advantageous for a 3GPP network (or network node) to be able to manage or control the ML model lifecycle, one aspect of which can be or include ML model delivery / transmission, which is the download or upload of the ML model (which can be or include a portion of the ML model or a delta or difference for the ML model, the delta or difference indicating a change or difference with respect to the ML model). Various example embodiments and techniques described herein can leverage the strengths of both the control plane (e.g., using control message(s) to request, command, initiate, control, manage, confirm, etc.) and the user (or data) plane to perform ML model transmission (e.g., upload by or download to a UE).

[0064] As used herein, the terms delivery and transmission may mean the same thing and / or may be used interchangeably, e.g., it may refer to or include or represent the communication, delivery, transmission, or transfer of an ML model (e.g., uploading the ML model from a UE to a storage location, or downloading the ML model from a storage location to a UE).

[0065] Thus, for example, the control plane and the user (or data) plane can be used in conjunction to initiate the delivery / transmission of an ML model. For example, a network node can initiate, request, or instruct a UE to perform ML model transmission, for example, by sending a control message to the UE via the control plane. The control message may include a command for the UE to upload or download the ML model (e.g., an upload command or a download command). The control message sent by the network node to the UE to initiate, control, or cause ML model transmission may also include storage information indicating the storage location of the ML model (e.g., where the ML model can be downloaded from or where the ML model should be uploaded), and protocol information indicating the protocol to be used by the UE for transmission of the ML model via the user (or data) plane. The storage information may indicate the storage location of the ML model and may, for example, indicate or include: a network address of a host node (e.g., an IP address or other network address of a server or node within the network); and at least one of: a path to the ML model on the host node (e.g., indicating a location on or within the host node where the ML model is stored for download or should be stored for upload); and / or a file name associated with the ML model.

[0066] In addition, for example, the control message may also include an ML model metadata container, which may include ML model metadata, including one or more of the following ML model metadata: an ML model identifier (ML model ID), which uniquely identifies the ML model in the UE or in the network; a function identifier, which identifies the function to be used to perform the ML model (for example, a function ID indicating the function to be performed by the UE using the ML model, such as power control, CSI report compression, beam selection, etc.); and / or an area indication (for example, such as PCI (physical cell identifier), TAC (tracking area code, which identifies a tracking area that may include one or more PCIs or cells), or a geo-fence indication identifying a geographical area), which identifies the area or one or more cells for which the ML model is valid or to be used.

[0067] Control messages may also be exchanged or transmitted between the UE and the network node to manage and / or verify ML model transmission, such as confirming that the ML model transmission is complete (indicating that the ML model transmission is successful) or fails (indicating that the ML model transmission fails). For example, ML model transmission failure may be due to (or caused by) insufficient storage resources, or due to integrity verification failure (e.g., due to a mismatch between a checksum or hash provided for the transmitted ML model and a checksum / hash calculated for the transmitted / received ML model).

[0068] Furthermore, a control message received by the UE including an upload command or a download command may indicate or include a storage location for the ML model. The storage location of the ML model may indicate any storage location within the network, such as a storage location on a network node, server, cloud, core network, RAN node, etc. The storage location of the ML model data may indicate the storage location to which the ML model will (or should) be uploaded (for an ML model upload command) or the storage location from which the UE may download the ML model (for an ML model download command). For example, the storage location of the ML model may indicate or include the network address of a host node (e.g., a RAN node, a server, a storage device or node in the cloud, a core network entity, an LMF, or any node, server, or storage device that is or can store the ML model), and at least one of: a path (e.g., indicating the file location of the ML model within the host node) and / or a file name of the ML model (or a file name associated with the ML model). In other words, for example, the ML model storage location is not necessarily located on the network node that sends the control message to initiate or request ML model transmission by the UE.

[0069] Typically, for example, the storage location of the ML model may be (or typically is) located at a node, device, or location that is different from the network node that sends the control message to the UE to request or initiate the ML model transmission. The ML model storage location is (at least in some cases) typically different from and / or independent of the network node that requests the ML model transmission, which can provide improved network flexibility, for example, by allowing the ML model to be stored in different locations (e.g., independent of the specific network node that can request the ML model transmission), and can, for example, allow regionally applicable ML models to be centrally stored without being limited to a specific location, a specific network node, or a specific protocol. Although in some cases, the storage location of the ML model can be the network node that initiates or requests (sends the control message to the UE) the UE to perform the ML model transmission, it can be provided on such a network node, or can be the same as such a network node.

[0070] According to example embodiments, a control plane (which may be, but is not limited to, RRC (e.g., such as RRC messages), NAS, or LPP) may be used to instruct or command a UE to download or upload an ML model via a user plane by indicating a download or upload protocol (such as, but not limited to, HTTP, FTP, TCP, or raw UDP data transmission), the host node address of the ML model, the file location (e.g., path and / or file name), a verification hash or checksum, and a container for ML model metadata (e.g., an ML model metadata container). The user plane (e.g., which may be via a UPF and using the indicated protocol) may be used to deliver / transmit the ML model bytes. By using the user plane to deliver / transmit the ML model, the disadvantages or concerns of control plane partitioning or compromised control plane functionality are mitigated or overcome. Furthermore, for example, various example embodiments may allow synchronization of ML model availability status for UEs in the network while placing little burden on the control plane and allowing full utilization of the user plane's data capacity for ML model transmission.

[0071] Figure 3 is a diagram illustrating a machine learning (ML) model downloaded by a user equipment (UE) or user device according to an example embodiment. Figure 3 As shown, UE 310 can communicate with gNB 312, a user plane function (UPF) 314 associated with communication of user data, and an access and mobility function (AMF) 316, where the UPF and AMF are part of the core network. In step 1, the UE can perform a capability exchange with gNB 312. For example, in response to receiving a capability request from gNB 312, UE 310 can send a capability response to gNB 312, indicating that UE 310 has the capability to perform ML model transmission.

[0072] The network or network node (e.g., gNB 312) may have an ML model (e.g., a trained ML model, or a portion of an ML model, or a delta or difference of the model, or one or more parameters that the UE may use to configure the ML model) that it wishes to transmit or provide to the UE 310. Figure 3In step 2, gNB 312 sends a control message (e.g., an RRC message) to UE 310 (and UE 310 receives it from gNB 312) for controlling the transmission of the ML model. In this example, the control message including a download command is sent to UE 310 to control, instruct, or cause UE 310 to download the ML model from a storage location. For example, the control message sent to UE 310 may include a command for UE 310 to download the ML model, storage information indicating the storage location of the ML model, and protocol information indicating the protocol to be used by the user equipment for transmission of the ML model via the user plane (e.g., via UPF 314).

[0073] like Figure 3 As shown, information 2A includes information (all or part thereof) that may be included in a control message sent by the gNB 312 to the UE 310 in step 2. As described above, the control message may include a download command, protocol information, and storage information. As shown in information 2A, in this example, the protocol information may indicate an FTP protocol, which the UE 310 should use to download the ML model from the storage location. For example, the storage information in this example (as shown in information 2A) may include a network address of the host node (e.g., 10.10.10.10), and at least one of the following: a path for the ML model on the host node (e.g., path: / models / as shown in information 2A) (e.g., in this case, a path to the location of the ML model on the host node) and a file name (e.g., in this example, csiModel0 as shown in information 2A). As Figure 3 As shown in information 2A, the control message may also include: a checksum or hash of the ML model (for example, in this example, Figure 2 2A), the UE 310 may use the checksum or hash to verify or validate the integrity of the downloaded ML model (e.g., to confirm or validate that the downloaded ML model is complete and accurate without errors). In addition, the information included in the control message may include the size of the ML model (e.g., the number of bytes).

[0074] In addition, if Figure 3As shown in information 2A of FIG2A , the control message received by the UE 310 from the gNB 312 may include an ML model metadata container, which may include, for example, one or more of the following metadata: an ML model identifier (e.g., modelID:xxx shown in information 2A), which uniquely identifies the ML model in the user equipment or in the network; a functional (or function) identifier (e.g., functionID:xx shown in information 2A), which identifies the function that the ML model is to be used to perform or assist in performing; and / or an area indication (e.g., validArea:{(multiple)TACs,(multiple)PCIs, geofences}, such as Figure 3 2A), which identifies the area or one or more cells for which the ML model is valid or will be used.

[0075] like Figure 3 As shown, at step 3A, UE 310 may determine whether UE 310 has sufficient memory resources to store and / or use the ML model. If sufficient memory resources are available, UE 310 does not send an acknowledgement or response to gNB 312 at this time. However, if the memory resources at UE 310 are insufficient, UE 310 may send a control message with a failure indication to gNB 312. Therefore, at step 3, if UE 310 does not have sufficient memory resources to store and / or use the ML model, UE 310 sends a control message including a failure indication (Model Download Failure) to gNB 312, indicating that the UE failed to download the ML model. UE 310 may indicate a failure reason in the control message it sends to gNB 312, indicating insufficient memory resources (e.g., failureReason: Insufficient Memory).

[0076] exist Figure 3 In step 4, if there are sufficient storage resources at the UE 310 to store and / or use the ML model, the process continues and the UE 310 establishes a PDU (Protocol Data Unit) session with the UPF 314 and / or AMF 316.

[0077] exist Figure 3 In step 5, the UE 310 downloads the indicated ML model from the storage location indicated by the storage information (e.g., from the host node address via the indicated path or file name) via the protocol indicated by the protocol information (indicating the FTP protocol in this example) (via the user plane function UPF 314).

[0078] exist Figure 3In step 6A of step 6A, the UE 310 may verify or confirm the integrity of the downloaded ML model. The received hash received in the control message of step 2 (in information 2A) may be a first checksum or hash. The UE 310 may calculate a second checksum or hash of the downloaded ML model as a second checksum or hash. The UE 310 may compare the two checksums or hashes to verify the integrity of the downloaded ML model. For example, the UE 310 may: 1) compare the calculated second checksum or hash with the received first checksum or hash to determine whether there is a match, and 2) determine whether the integrity of the downloaded ML model is verified based on whether the second checksum or hash matches the first checksum or hash. If the integrity of the downloaded ML model is verified or correct, e.g., without errors, the two checksums or hashes should match.

[0079] Depending on whether the two checksums or hashes match, the UE 310 will: if the integrity of the ML model is verified (e.g., if the two checksums or hash values ​​match), then in step 6B, send a model download completion indication (ModelDownloadComplete, as shown in step 6B) to the gNB 312, confirming that the UE 310 has completed downloading the ML model from the storage location; or if the integrity of the ML model is not verified (e.g., if the two checksums or hash values ​​do not match, indicating that the downloaded ML model is corrupted, inaccurate, or has errors), then in step 6C, send a model download failure indication (e.g., ModelDownloadFailure) to the gNB 312, indicating that the download of the ML model failed, including providing a failure reason indicating the verification failure (e.g., MD5VerificationFailure, as shown in step 6C).

[0080] Therefore, if Figure 3As shown, the gNB instructs the UE to download the ML model via the user plane (e.g., via a specified protocol). The capability exchange procedure is used to determine whether the UE is capable of ML model delivery / transfer via download and which protocols it supports. The gNB issues a ModelDownloadCommand specifying one of the download protocols indicated by the UE as available, the host (or host node) address of the ML model, the path to the ML model, the size of the ML model, and an MD5 hash used by the UE to verify the success of the model download (verifying the integrity of the downloaded ML model). In addition, an ML model metadata container may also be sent to the UE 310 (e.g., within the control message of step 2), which may include information used to identify the model for network control purposes. For example, a validity area is shown, which facilitates autonomous decisions about model activation, deactivation, or selection. The included metadata may contain any information specified by 3GPP, and the included parameters are not an exhaustive list.

[0081] If UE 310 does not have sufficient storage for the ML model, it issues or sends a ModelDownloadFailure with the cause "insufficientStorage" to gNB 312 to cancel the model delivery / transfer process. Otherwise, if not already established by attempting the ML model download process, a PDU session establishment sufficient to access the model's host address is triggered. UE 310 initiates the ML model download process using the specified protocol. The download is either completed or declared as a failure by UE 310. UE 310 calculates the MD5 hash (or other checksum) of the downloaded model and compares it with the hash provided in the ModelDownloadCommand to verify the download success. If verification is successful, the UE sends a ModelDownloadComplete to gNB 312. Similarly, if the download fails, for example, due to integrity verification failure of the ML model, UE 310 sends a ModelDownloadFailure to gNB 312 indicating the download failure, along with the failure reason, which may include, but is not limited to, insufficient storage or MD5 verification failure.

[0082] By replacing the capability exchange between gNB 312 and UE 310 in step 1 with NAS capability exchange between UE 310 and AMF 316, Figure 3The example embodiment shown for ML model download can be implemented in the core network via NAS messaging. The content of the ModelDownloadCommand in step 2 will remain unchanged, and the PDU session can be established in step 3 as before. The remaining steps 4-5 will remain unchanged, except that the model download response (ModelDownloadComplete or ModelDownloadFailure) will be directed to (sent by UE 310 to) AMF 316 instead of gNB 312.

[0083] Similarly, by replacing the capability exchange between gNB 312 and UE in step 1 with LPP capability exchange between UE and LMF ( Figure 3 not shown), Figure 3 The example embodiment shown for ML model download can also be implemented in the LMF using LPP protocol messaging. The content of the ModelDownloadCommand in step 2 will remain unchanged, and the PDU session will be established in step 3 as before. The remaining steps 4-5 will remain unchanged, except that the model download response (ModelDownloadComplete or ModelDownloadFailure) will be directed to (or sent by UE 310 to) the LMF instead of gNB 312.

[0084] In example embodiments, collocation of the ML model storage location is not required. That is, the storage location (e.g., on a host node) can be on the same network node that initiates or requests the ML model download (collocated), or the storage location can be provided on a different node, where the host node (or the storage location of the ML model) is different from the network node that requested the ML model download (the storage location is not collocated with the gNB or network node that requested the ML model download). In addition, the UE should have access to the host node via the UPF so that the ML model can be transferred to or from the storage location on the host node. In addition, the control entity (e.g., gNB, LMF, or AMF) that requests the UE to download or upload the ML model can typically access the internal or external storage of the host node or storage location so that the control entity can also access these ML models, for example, to verify the integrity of the uploaded model.

[0085] Figure 4 is a diagram illustrating a machine learning (ML) model uploaded by a user equipment (UE) or user device according to an example embodiment. Figure 4 An upload example similar to Figure 3 The ML model download example and the difference between the two figures will be described. Figure 4As shown, UE 310 can communicate with gNB 312, a user plane function (UPF) 314 associated with communication of user data, and an access and mobility function (AMF) 316, where the UPF and AMF are part of the core network. In step 1, the UE can perform a capability exchange with gNB 312. For example, in response to receiving a capability request from gNB 312, UE 310 can send a capability response to gNB 312, indicating that UE 310 has the capability to perform ML model transmission.

[0086] The network or a network node (e.g., gNB 312) may want the UE to provide or upload the ML model to a storage location. Figure 4 In step 2, gNB 312 sends a control message (e.g., an RRC message) to UE 310 (and UE 310 receives it from gNB 312) for controlling the transmission of the ML model. In this example, the control message including an upload command is sent to UE 310 to control, instruct, or cause UE 310 to upload the ML model to a storage location. For example, the control message sent to UE 310 may include a command for UE 310 to upload the ML model, storage information indicating the storage location to which the ML model should be uploaded, and protocol information indicating the protocol to be used by the user equipment for uploading the ML model via the user plane (e.g., via UPF 314).

[0087] like Figure 4 As shown, the control message may include all or part of information 2A, for example, including an upload command, protocol information indicating the protocol to be used by the UE for uploading, and storage information. Figure 4 As shown in information 2A, in this example, the protocol information may indicate the FTP protocol, which UE 310 should use to upload the ML model to the storage location. For example, the storage information in this example (as shown in information 2A) may include the network address of the host node (e.g., 10.10.10.10), and at least one of the following: a path for the ML model on the host node (e.g., path: / models / shown in information 2A) (e.g., in this case, the path to the location on the host node where the ML model should be uploaded or stored) and a file name (e.g., csiModel0), which may indicate the name of the ML model to be stored or uploaded.

[0088] In addition, Figure 4For this upload process, the control message received from the gNB does not include a checksum or hash of the ML model. This is because the gNB 312 or other network entity (rather than the UE) will verify the integrity of the uploaded ML model. Therefore, the control message with the upload command in step 2 typically does not (and need not) include a checksum or hash value.

[0089] In addition, if Figure 4 As shown in information 2A of FIG2 , the control message received by the UE 310 from the gNB 312 may include an ML model metadata container, which may include, for example, one or more of the following metadata: an ML model identifier, which uniquely identifies the ML model in the user equipment or in the network; a functional (or function) identifier, which identifies the function that the ML model is to be used to perform or assist in performing; and / or an area indication, which identifies the area or one or more cells for which the ML model is valid or to be used.

[0090] exist Figure 4 In step 3, UE 310 establishes a PDU (Protocol Data Unit) session with UPF 314.

[0091] exist Figure 4 In step 4, the UE 310 uploads the indicated ML model to a storage location (e.g., to an indicated path or file name on a host node) via the protocol indicated by the protocol information (indicating the FTP protocol in this example) (via the user plane function UPF 314).

[0092] exist Figure 4 In step 5, if the UE detects an error in the ML model upload, the UE 310 sends a model upload response indicating an upload failure at 5A, and the gNB may respond with a model upload failure at 5B.

[0093] exist Figure 4 In step 6, if the UE 310 does not detect a failure in the upload process, the UE 310 sends a checksum or hash of the uploaded ML model to the gNB 312, which can be used by a network node, gNB, or other network entity to verify the integrity of the uploaded ML model.

[0094] exist Figure 4 In step 7A of the UE 310, the gNB 312 (or other network entity) may verify the integrity of the uploaded ML model, for example, by calculating a checksum or hash of the received ML model and comparing the calculated checksum or hash (calculated by the gNB or network entity based on the uploaded ML model) with the received checksum / hash (provided by the UE 310 to the gNB 312) to verify the integrity of the uploaded ML model. If the integrity of the ML model is verified, then Figure 4 In step 7B, gNB 312 sends a model upload completion indication to UE 310. If the integrity verification of the uploaded ML model fails, Figure 4 In step 7C, gNB 312 sends a model upload failure indication to UE 310 and may specify a failure reason for the verification failure. UE 310 may then re-upload the ML model (e.g., in response to the upload failure indication from gNB 312), or UE 310 may wait to receive another upload command from gNB 312 before re-uploading the ML model to the storage location.

[0095] Figure 5 and Figure 6 2 are diagrams respectively illustrating ML model download and upload processes that may be controlled or requested by a location management function (LMF) according to example embodiments. Figure 5 is a diagram illustrating an ML model downloading process controlled by an LMF according to an example embodiment. Figure 6 is a diagram illustrating an ML model downloading process controlled by an LMF according to an example embodiment. Figure 5 (Download) and Figure 6 The message flow and operation of (upload) are usually respectively Figure 3 and Figure 4 Same or very similar as shown, for example, where the LMF is the controlling entity rather than the gNB, and control messages are provided via LPP rather than RRC.

[0096] The LPP protocol is used by the LMF entity to send and receive messages. The LPP protocol specifies most of its bidirectional messaging in two message types: RequestAssistanceData and ProvideAssistanceData. The UE requests data from the LMF, and the LMF provides data to the UE. The LMF can request location information from the UE using the RequestLocationInformation and ProvideLocationInformation messages, and the UE provides location information to the LMF. In addition, the LPP protocol supports the following message types to indicate errors or stop a procedure: Abort allows cancellation of a procedure. Error allows propagation of errors.

[0097] An example method of model delivery / transmission of LMF may include embedding ModelDownload and ModelUpload messages or ModelDelivery messages into Request / ProvideLocationInformation messages (to / from LMF). In order to command the UE to download or upload the ML model, the LMF may issue a RequestLocationInformation with a command to download or upload the ML model. Figure 5 A possible LMF adaptation using the ModelDelivery message option embedded in the LocationInformation message and using LPP Error for error delivery is shown. The data contained in the ModelDownload and ModelUpload or ModelDelivery messages embedded in the LPP messages can be the same as previously defined.

[0098] Similarly, the ML model upload and download procedures can be applied for use with NAS messages issued to / from the core network (e.g., AMF).

[0099] According to another example embodiment, a message set may be used, with different contents or commands in such messages to accommodate the upload or download process, rather than providing separate messages for the upload process and the download process.

[0100] The ModelDownload and ModelUpload protocol messages can be merged into the ModelDelivery command with the following changes. ModelDownloadCommand and ModelUploadCommand will be merged into ModelDeliveryCommand, which will include a direction field that supports both "download" and "upload" values, and an optional MD5 hash field for the "download" direction only. ModelDownloadComplete and ModelUploadComplete will be merged into ModelDeliveryComplete.

[0101] ModelDownloadFailure and ModelUploadFailure will be merged into ModelDeliveryFailure. ModelUploadResponse will be changed to ModelDeliveryResponse.

[0102] Some additional examples will be provided.

[0103] Example 1. A method, comprising: receiving, by a user equipment, a control message for controlling transmission of a machine learning (ML) model from a network node, the control message comprising: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and performing, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0104] Example 2. The method of Example 1, wherein the storage information indicating the storage location of the ML model comprises: a network address of a host node; and at least one of: a path for the ML model on the host node; or a file name.

[0105] Example 3. The method according to any one of Examples 1 to 2, wherein the control message further includes an ML model metadata container, the ML model metadata container including: at least one of the following ML model metadata: an ML model identifier, uniquely identifying the ML model in the user equipment or in the network; a function identifier, identifying the function to be performed by the ML model; and / or an area indication, identifying an area or one or more cells for which the ML model is valid or to be used.

[0106] Example 4. The method of Example 1, wherein the ML model comprises at least one of: a complete ML model; a portion of an ML model; or an increment or change to the ML model indicating a change or difference in the ML model.

[0107] Example 5. A method according to any one of Examples 1 to 4, wherein the network node includes at least one of the following: a gNB; an eNB; a base station or an access point; a centralized unit (CU) and / or a distributed unit (DU); a radio access network (RAN) node; an access and mobility function (AMF); a core network or a core network node; or a location management function (LMF).

[0108] Example 6. A method according to any one of Examples 1 to 5, wherein the control message includes at least one of the following: a first radio resource control (RRC) control message received from a gNB or a RAN (radio access network) node; a first LPP (LTE Positioning Protocol) control message received from a location management function (LMF); or a first NAS (network access layer) control message received from the core network.

[0109] Example 7. The method according to Example 6 further includes: the UE sending at least one of the following messages regarding the transmission of the ML model: a second radio resource control (RRC) control message sent to the gNB or the RAN (radio access network) node in response to the first RRC control message; a second LPP (LTE Positioning Protocol) control message sent to the location management function (LMF) in response to the first LPP control message; or a second NAS (Network Access Stratum) control message sent to the core network in response to the first NAS control message.

[0110] Example 8. The method of any one of Examples 1 to 7, wherein the protocol to be used for transmission of the ML model between the user device and the storage location indicated by the storage information via the user plane comprises at least one of: File Transfer Protocol (FTP); Hypertext Transfer Protocol (HTTP); Transmission Control Protocol (TCP); or User Datagram Protocol (UDP).

[0111] Example 9. The method according to any one of Examples 1 to 8 further includes: receiving a capability request by the user equipment from the network node; and sending a capability response by the user equipment to the network node, wherein the capability response indicates that the user equipment has the ability to perform ML model transmission.

[0112] Example 10. A method according to any one of Examples 1 to 9, wherein the command includes a download command, the download command instructing: the user device to download the ML model from the storage location of the ML model indicated by the storage information to the user device; the method further includes: the user device downloading the ML model from the storage location of the ML model.

[0113] Example 11. The method of Example 10, wherein the control message includes an ML model metadata container, the ML model metadata container including at least a function identifier, the function identifier identifying: a function that the ML model is to be used to perform, wherein the method further comprises: using the downloaded ML model by the user device to perform the function specified by the function identifier.

[0114] Example 12. The method of any one of Examples 10 to 11, wherein the storage information indicates a storage location from which the ML model can be downloaded, including the network address of the host node storing the ML model, and at least one of a path for the ML model on the host node or a file name associated with the ML model.

[0115] Example 13. A method according to any one of Examples 10 to 12: wherein the command includes a download command, the download command instructing the user device to download the ML model from the storage location of the ML model; the method further includes: the user device providing a download completion indication or a download failure indication to the network node, the download completion indication indicating that the download of the ML model is completed by the user device, and the download failure indication indicating that the download of the ML model fails.

[0116] Example 14. A method according to any one of Examples 10 to 13: wherein the command includes a download command, the download command instructing the user equipment to download the ML model from the storage location of the ML model; wherein the control message further indicates the size of the ML model; the method further includes: determining, by the user equipment, whether there are sufficient storage resources at the user equipment to store and / or use the ML model; and if there are insufficient storage resources at the user equipment to store and / or use the ML model, sending, by the user equipment, a control message to the network node, the control message including: a failure indication indicating that the downloading of the ML model failed, including providing a failure reason indicating insufficient storage resources.

[0117] Example 15. A method according to any one of Examples 9 to 14, wherein the command includes a download command, the download command instructing the user device to download the ML model from the storage location of the ML model; and wherein the control message includes: a first checksum or hash for the ML model, the first checksum or hash can be used by the user device to verify the integrity of the ML model that can be downloaded by the user device.

[0118] Example 16. A method according to Example 15: wherein the performing of the transmission of the ML model includes: downloading the ML model by the user device from the storage location of the ML model; the method further includes: calculating a second checksum or hash of the downloaded ML model; comparing the second checksum or hash with the first checksum or hash to determine whether there is a match; determining whether the integrity of the downloaded ML model is verified based on whether the second checksum or hash matches the first checksum or hash; and sending, by the user device, to the network node any one of the following: a model download completion indication, confirming that the download of the ML model is complete if the integrity of the ML model is verified; or a model download failure indication, indicating that the download of the ML model failed if the integrity of the ML model is not verified, including providing a failure reason indicating the verification failure.

[0119] Example 17. The method of any one of Examples 1 to 9, wherein the command comprises an upload command instructing the user device to upload the ML model to the storage location of the ML model.

[0120] Example 18. The method of any one of Examples 1 to 9 and 17, wherein the storage information indicates a storage location to which the user device can upload the ML model, including a network address of the host node for storing the ML model, and at least one of a path for the ML model on the host node and a file name associated with the ML model.

[0121] Example 19. A method according to any one of Examples 1 to 9, 17, and 18, wherein the command includes an upload command, the upload command instructing the user device to upload the ML model to the storage location of the ML model; the method further comprising: uploading, by the user device, the ML model to the storage location via the protocol indicated by the protocol information included in the control message.

[0122] Example 20. The method of Example 19, further comprising: calculating, by the user device, a checksum or hash of the uploaded ML model; and sending, by the user device, the checksum or hash of the uploaded ML model to the network node to allow the network node to determine the integrity of the uploaded ML model.

[0123] Example 21. The method of any one of Examples 1 to 9 and 17 to 20, further comprising: receiving, by the user device from the network node, any one of: an upload completion indication indicating that the upload of the ML model is completed by the user device and successfully received at the storage location, or an upload failure indication indicating that the upload of the ML model fails.

[0124] Example 22. The method according to Example 21 further includes: if the user device receives an upload failure indication for the ML model, performing the following: the user device re-uploading the ML model to the storage location via the protocol indicated by the protocol information; and the user device sending or re-sending a checksum or hash of the re-uploaded ML model to the network node to allow the network node to determine the integrity of the re-uploaded ML model.

[0125] Example 23. A device comprising: at least one processor; and at least one memory comprising computer program code; the at least one memory and the computer program code are configured to, together with the at least one processor, cause the device to at least perform a method according to any one of Examples 1 to 22.

[0126] Example 24. A non-transitory computer-readable storage medium comprising instructions stored thereon, which, when executed by at least one processor, are configured to cause a computing system to perform the method according to any one of Examples 1 to 22.

[0127] Example 25. An apparatus comprising components for performing the method according to any one of Examples 1 to 22.

[0128] Example 26. An apparatus comprising: at least one processor; and at least one memory comprising computer program code; the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to at least: receive, by a user equipment, a control message for controlling the transmission of a machine learning (ML) model from a network node, the control message comprising: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and perform, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0129] Example 27. A non-transitory computer-readable storage medium comprising instructions stored thereon, which, when executed by at least one processor, are configured to cause a computing system to: receive, by a user device, a control message for controlling the transmission of a machine learning (ML) model from a network node, the control message comprising: a command for the user device to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user device for transmission of the ML model via a user plane; and perform, by the user device, transmission of the ML model between the user device and the storage location indicated by the storage information via the protocol.

[0130] Example 28. An apparatus comprising: a component for receiving, by a user equipment, a control message for controlling transmission of a machine learning (ML) model from a network node, the control message comprising: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and a component for performing, by the user equipment, transmission of the ML model between the user equipment and the storage location indicated by the storage information via the protocol.

[0131] Figure 7is a block diagram of a wireless station or node (e.g., UE, user equipment, AP, BS, eNB, gNB, RAN node, network node, TRP, or other node) 1300 according to an example embodiment. The wireless station 1300 may include, for example, one or more (e.g., Figure 7 The wireless station also includes two RF (radio frequency) or wireless transceivers 1302A and 1302B, each of which includes a transmitter for transmitting signals and a receiver for receiving signals. The wireless station also includes a processor or control unit / entity (controller) 1304 for executing instructions or software and controlling the transmission and reception of signals, and a memory 1306 for storing data and / or instructions.

[0132] The processor 1304 may also make decisions or determinations, generate frames, packets, or messages for transmission, decode received frames or messages for further processing, and perform other tasks or functions described herein. For example, the processor 1304, which may be a baseband processor, may generate messages, packets, frames, or other signals for transmission via the wireless transceiver 1302 (1302A or 1302B). The processor 1304 may control the transmission of signals or messages over a wireless network, and may control the reception of signals or messages, etc., over a wireless network (e.g., after being down-converted by the wireless transceiver 1302). The processor 1304 may be programmable and capable of executing software or other instructions stored in a memory or on other computer media to perform the various tasks and functions described above, such as one or more of the tasks or methods described above. For example, the processor 1304 may be (or may include) hardware, programmable logic, a programmable processor executing software or firmware, and / or any combination thereof. For example, using other terminology, the processor 1304 and the transceiver 1302 may be considered together as a wireless transmitter / receiver system.

[0133] In addition, reference Figure 7 , the controller (or processor) 1308 can execute software and instructions and can provide overall control for the station 1300 and can Figure 7 Other systems not shown provide controls, such as controlling input / output devices (e.g., display, keypad), and / or software that can execute one or more applications that can be provided on wireless station 1300, such as, for example, an email program, audio / video applications, a word processor, a voice over IP application, or other applications or software.

[0134] Additionally, a storage medium may be provided that includes stored instructions that, when executed by a controller or processor, may cause the processor 1304 or other controller or processor to perform one or more of the functions or tasks described above.

[0135] According to another example embodiment, the RF or wireless transceiver(s) 1302A / 1302B may receive signals or data, and / or transmit or send signals or data. The processor 1304 (and possibly the transceiver 1302A / 1302B) may control the RF or wireless transceiver 1302A or 1302B to receive, send, broadcast or send signals or data.

[0136] Embodiments of the various technologies described herein may be implemented in digital electronic circuit systems, or in computer hardware, firmware, software, or a combination thereof. The embodiments may be implemented as a computer program product, i.e., a computer program tangibly embodied in an information carrier, for example, in a machine-readable storage device or in a propagated signal, for execution by a data processing apparatus or for controlling the operation of a data processing apparatus, for example, a programmable processor, a computer, or multiple computers. The embodiments may also be provided on a computer-readable medium or a computer-readable storage medium, which may be a non-transitory medium. The embodiments of the various technologies may also include embodiments provided via transient signals or media, and / or downloadable programs and / or software embodiments via the Internet or (multiple) other networks (wired networks and / or wireless networks). In addition, the embodiments may be provided via machine type communication (MTC) or via the Internet of Things (IOT).

[0137] A computer program may be in source code form, object code form, or some intermediate form, and may be stored on some carrier, distribution medium, or computer-readable medium, which may be any entity or device capable of carrying the program. Examples of such carriers include recording media, computer memory, read-only memory, optical and / or electrical carrier signals, telecommunications signals, and software distribution packages. Depending on the required processing power, a computer program may be executed on a single electronic digital computer or distributed among multiple computers.

[0138] Furthermore, embodiments of the various techniques described herein may utilize cyber-physical systems (CPS) (systems that enable computing elements that control physical entities to collaborate). CPS may enable the implementation and utilization of a large number of interconnected ICT devices (sensors, actuators, processors, microcontrollers, etc.) embedded in physical objects at different locations. Mobile cyber-physical systems (where the physical system in question has inherent mobility) are a subcategory of cyber-physical systems. Examples of mobile physical systems include mobile robots and electronic devices that are transported by humans or animals. The popularity of smartphones has increased interest in the field of mobile cyber-physical systems. Therefore, various embodiments of the techniques described herein may be provided by one or more of these techniques.

[0139] Computer programs such as the above-mentioned computer program(s) may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form, including as stand-alone programs, or as modules, components, subroutines, or other units or portions thereof suitable for a computing environment. A computer program may be deployed to execute on one computer, or on multiple computers at one site, or on multiple computers distributed across multiple sites and interconnected by a communication network.

[0140] The method steps may be performed by one or more programmable processors executing a computer program or portion of a computer program to perform functions by operating on input data and generating output. The method steps may also be performed by, and apparatus may be implemented as, special purpose logic circuitry, such as an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).

[0141] For example, processors suitable for executing a computer program include both general-purpose and special-purpose microprocessors, and any one or more processors of any kind of digital computer, chip, or chipset. Typically, the processor will receive instructions and data from a read-only memory or a random access memory, or both. Elements of a computer may include at least one processor for executing instructions and one or more memory devices for storing instructions and data. Typically, a computer may also include, or be operatively coupled to, receive data from or transfer data to, or both, one or more mass storage devices (e.g., magnetic, magneto-optical, or optical disks) for storing data. Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile memory, including, for example, semiconductor memory devices such as EPROM, EEPROM, and flash memory devices; magnetic disks, such as internal hard disks or removable disks; magneto-optical disks; and CD ROM and DVD-ROM disks. The processor and memory may be supplemented by, or incorporated in, special-purpose logic circuitry.

[0142] To provide interaction with a user, embodiments may be implemented on a computer having a display device (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) display) for displaying information to the user, and a user interface (e.g., a keyboard and a pointing device, e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, feedback provided to the user may be any form of sensory feedback, such as visual feedback, auditory feedback, or tactile feedback; and input from the user may be received in any form, including sound, voice, or tactile input.

[0143] The embodiments may be implemented in a computing system that includes a back-end component, such as a data server, a middleware component, such as an application server, a front-end component, such as a client computer having a graphical user interface or a web browser through which a user can interact with the embodiments, or any combination of such back-end, middleware, or front-end components. The components may be interconnected by any form or medium of digital data communication, such as a communication network. Examples of communication networks include local area networks (LANs) and wide area networks (WANs), such as the Internet.

[0144] While certain features of the described embodiments have been illustrated as described herein, those skilled in the art will now be able to devise numerous modifications, substitutions, changes, and equivalents. It should therefore be understood that the appended claims are intended to cover all such modifications and variations that fall within the true spirit of the various embodiments.

Claims

1. A method comprising: receiving, by a user equipment from a network node, a control message for controlling transmission of a machine learning (ML) model, the control message comprising: a command for the user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; and Transmission of the ML model between the user device and the storage location indicated by the storage information is performed by the user device via the protocol.

2. The method of claim 1 , wherein the storage information indicating the storage location of the ML model comprises: The network address of the host node; as well as At least one of the following: The path for the ML model on the host node; or file name.

3. The method according to any one of claims 1 to 2, wherein the control message further includes an ML model metadata container, wherein the ML model metadata container includes: At least one of the following ML model metadata: an ML model identifier, which uniquely identifies the ML model in the user device or in the network; A function identifier that identifies the function that the ML model will be used to perform; or A region indication identifies a region or one or more cells for which the ML model is valid or to be used.

4. The method according to any one of claims 1 to 3, wherein the ML model comprises at least one of the following: Complete ML model; part of an ML model; or The increment or change of the ML model indicating the change or difference of the ML model.

5. The method according to any one of claims 1 to 4, wherein the network node comprises at least one of the following: gNB; eNB; Base station or access point; Centralized Unit (CU) and / or Distributed Unit (DU); Radio Access Network (RAN) nodes; Access and Mobility Function (AMF); core network or core network node; or Location Management Function (LMF).

6. The method according to any one of claims 1 to 5, wherein the control message comprises at least one of the following: a first radio resource control (RRC) control message received from a gNB or a RAN (Radio Access Network) node; a first LPP (LTE Positioning Protocol) control message received from a Location Management Function (LMF); or A first NAS (Network Access Stratum) control message is received from the core network.

7. The method according to claim 6, further comprising: Sending at least one of the following messages regarding the transmission of the ML model: a second radio resource control (RRC) control message sent to the gNB or the RAN (Radio Access Network) node in response to the first RRC control message; a second LPP (LTE Positioning Protocol) control message sent to the Location Management Function (LMF) in response to the first LPP control message; or A second NAS (Network Access Stratum) control message is sent to the core network in response to the first NAS control message.

8. The method according to any one of claims 1 to 7, wherein the protocol to be used for transmission of the ML model between the user equipment and the storage location indicated by the storage information via the user plane comprises at least one of the following: File Transfer Protocol (FTP); Hypertext Transfer Protocol (HTTP); Transmission Control Protocol (TCP); or User Datagram Protocol (UDP).

9. The method according to any one of claims 1 to 8, further comprising: Receiving, by the user equipment, a capability request from the network node; The user equipment sends a capability response to the network node, where the capability response indicates that the user equipment has the capability to perform ML model transmission.

10. The method according to any one of claims 1 to 9, wherein the command comprises a download command, the download command instructing the user device to download the ML model from the storage location of the ML model indicated by the storage information; The method further comprises: The ML model is downloaded by the user device from the storage location of the ML model.

11. The method according to claim 10, wherein the control message comprises an ML model metadata container, the ML model metadata container comprising at least a function identifier, the function identifier identifying a function to be performed by the ML model, wherein the method further comprises: The function specified by the function identifier is executed by the user device using the downloaded ML model.

12. The method according to any one of claims 10 to 11, wherein the storage information indicates: a storage location from which the ML model is downloaded, including the network address of the host node storing the ML model, and at least one of a path for the ML model on the host node or a file name associated with the ML model.

13. The method according to any one of claims 10 to 12, wherein the command comprises a download command, the download command instructing the user device to download the ML model from the storage location of the ML model; The method further comprises: The user equipment provides a download completion indication or a download failure indication to the network node, wherein the download completion indication indicates that the download of the ML model is completed by the user equipment, and the download failure indication indicates that the download of the ML model fails.

14. The method according to any one of claims 10 to 13, wherein the command comprises a download command, the download command instructing the user device to download the ML model from the storage location of the ML model; wherein the control message further indicates the size of the ML model; The method further comprises: determining, by the user device, whether there are sufficient storage resources at the user device to store and / or use the ML model; If there are insufficient storage resources at the user equipment to store and / or use the ML model, a control message is sent by the user equipment to the network node, the control message including: a failure indication indicating that downloading of the ML model failed, including providing a failure reason indicating insufficient storage resources.

15. The method according to any one of claims 9 to 14, wherein the command comprises a download command, the download command instructing the user device to download the ML model from the storage location of the ML model; The control message includes: A first checksum or hash for the ML model, the first checksum or hash being used by the user device for verifying the integrity of the ML model to be downloaded by the user device.

16. The method according to claim 15, Wherein said performing said transmitting of said ML model comprises: downloading, by the user device, the ML model from the storage location of the ML model; The method further comprises: calculating a second checksum or hash of the downloaded ML model; comparing the second checksum or hash to the first checksum or hash to determine if there is a match; determining whether the integrity of the downloaded ML model is verified based on whether the second checksum or hash matches the first checksum or hash; and Sending, by the user equipment, any one of the following to the network node: A model download completion indication, if the integrity of the ML model is verified, confirming that the download of the ML model is complete; or A model download failure indication indicates that the download of the ML model has failed if the integrity of the ML model has not been verified, including providing a failure reason indicating the verification failure. 17 . The method according to claim 1 , wherein the command comprises an upload command instructing the user device to upload the ML model to the storage location of the ML model.

18. The method according to any one of claims 1 to 9 and 17, wherein the storage information indicates a storage location to which the user device is to upload the ML model, including a network address of the host node for storing the ML model, and at least one of a path for the ML model on the host node and a file name associated with the ML model.

19. The method according to any one of claims 1 to 9, 17 and 18, wherein the command comprises an upload command, the upload command instructing the user device to upload the ML model to the storage location of the ML model; The method further comprises: The ML model is uploaded to the storage location by the user equipment via the protocol indicated by the protocol information included in the control message.

20. The method according to claim 19, further comprising: Calculating, by the user device, a checksum or hash of the uploaded ML model; as well as The checksum or hash of the uploaded ML model is sent by the user equipment to the network node to allow the network node to determine the integrity of the uploaded ML model.

21. The method according to any one of claims 1 to 9 and 17 to 20, further comprising: Receiving, by the user equipment from the network node, any one of: an upload completion indication indicating that the uploading of the ML model is completed by the user equipment and successfully received at the storage location, or an upload failure indication indicating that the uploading of the ML model fails.

22. The method according to claim 21, further comprising: If the user equipment receives an indication of a failure to upload the ML model, the user equipment performs the following: Re-uploading the ML model to the storage location by the user device via the protocol indicated by the protocol information; A checksum or hash of the re-uploaded ML model is sent or re-sent by the user equipment to the network node to allow the network node to determine the integrity of the re-uploaded ML model.

23. An apparatus comprising: at least one processor; as well as at least one memory including computer program code; The at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to at least: receiving, from a network node, a control message for controlling transmission of a machine learning (ML) model, the control message comprising: a command for the apparatus to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the apparatus for transmission of the ML model via a user plane; as well as Transmitting the ML model between the device and the storage location indicated by the storage information is performed via the protocol.

24. The apparatus of claim 23, wherein the storage information indicating the storage location of the ML model comprises: The network address of the host node; as well as At least one of the following: The path for the ML model on the host node; or file name.

25. The apparatus according to any one of claims 23 to 24, wherein the control message further comprises an ML model metadata container, wherein the ML model metadata container comprises: At least one of the following ML model metadata: an ML model identifier, which uniquely identifies the ML model in the device or in a network; A function identifier that identifies the function that the ML model will be used to perform; or A region indication identifies a region or one or more cells for which the ML model is valid or to be used.

26. The apparatus of any one of claims 23 to 25, wherein the ML model comprises at least one of: Complete ML model; part of an ML model; or The increment or change of the ML model indicating the change or difference of the ML model.

27. The apparatus according to any one of claims 23 to 26, wherein the network node comprises at least one of the following: gNB; eNB; Base station or access point; Centralized Unit (CU) and / or Distributed Unit (DU); Radio Access Network (RAN) nodes; Access and Mobility Function (AMF); core network or core network node; or Location Management Function (LMF).

28. The apparatus according to any one of claims 23 to 27, wherein the control message comprises at least one of the following: a first radio resource control (RRC) control message received from a gNB or a RAN (Radio Access Network) node; a first LPP (LTE Positioning Protocol) control message received from a Location Management Function (LMF); or A first NAS (Network Access Stratum) control message is received from the core network.

29. The apparatus of claim 28, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: send at least one of the following messages regarding the transmission of the ML model: a second radio resource control (RRC) control message sent to the gNB or the RAN (Radio Access Network) node in response to the first RRC control message; a second LPP (LTE Positioning Protocol) control message sent to the Location Management Function (LMF) in response to the first LPP control message; or A second NAS (Network Access Stratum) control message is sent to the core network in response to the first NAS control message.

30. The apparatus according to any one of claims 23 to 29, wherein the protocol to be used for transmission of the ML model between the apparatus and the storage location indicated by the storage information via the user plane comprises at least one of the following: File Transfer Protocol (FTP); Hypertext Transfer Protocol (HTTP); Transmission Control Protocol (TCP); or User Datagram Protocol (UDP).

31. The apparatus according to any one of claims 23 to 30, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: receiving a capability request from the network node; A capability response is sent to the network node, where the capability response indicates that the device has the capability to perform ML model transmission.

32. The device according to any one of claims 23 to 31, wherein the command comprises a download command, the download command instructing the apparatus to download the ML model from the storage location of the ML model indicated by the storage information; wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: The ML model is downloaded from the storage location of the ML model.

33. The apparatus of claim 32, wherein the control message comprises an ML model metadata container, the ML model metadata container comprising at least a function identifier, the function identifier identifying a function to be performed by the ML model, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: The function specified by the function identifier is executed using the downloaded ML model.

34. The apparatus according to any one of claims 32 to 33, wherein the storage information indicates a storage location from which the ML model is downloaded, including the network address of the host node storing the ML model, and at least one of a path for the ML model on the host node or a file name associated with the ML model.

35. The device according to any one of claims 32 to 34, wherein the command comprises a download command, the download command instructing the apparatus to download the ML model from the storage location of the ML model; The at least one memory and the computer program code are configured to, together with the at least one processor, cause the apparatus to: provide a download completion indication or a download failure indication to the network node, wherein the download completion indication indicates that the download of the ML model is completed by the apparatus, and the download failure indication indicates that the download of the ML model fails.

36. The device according to any one of claims 32 to 35, wherein the command comprises a download command, the download command instructing the apparatus to download the ML model from the storage location of the ML model; wherein the control message further indicates the size of the ML model; wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: determining whether there are sufficient storage resources at the device to store and / or use the ML model; If there are insufficient storage resources at the device to store and / or use the ML model, a control message is sent to the network node, the control message comprising: A failure indication indicating that downloading of the ML model failed, including providing a failure reason indicating insufficient storage resources.

37. The device according to any one of claims 31 to 36, wherein the command comprises a download command, the download command instructing the apparatus to download the ML model from the storage location of the ML model; The control message includes: A first checksum or hash for the ML model, the first checksum or hash used by the device to verify the integrity of the ML model to be downloaded by the device.

38. The device according to claim 37, wherein Wherein said performing said transmitting of said ML model comprises: The at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: downloading the ML model from the storage location of the ML model; wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: calculating a second checksum or hash of the downloaded ML model; comparing the second checksum or hash to the first checksum or hash to determine if there is a match; determining whether the integrity of the downloaded ML model is verified based on whether the second checksum or hash matches the first checksum or hash; and Send any of the following to the network node: A model download completion indication, if the integrity of the ML model is verified, confirming that the download of the ML model is complete; or A model download failure indication indicates that the download of the ML model has failed if the integrity of the ML model has not been verified, including providing a failure reason indicating the verification failure.

39. The apparatus according to any one of claims 23 to 31, wherein the command comprises an upload command instructing the apparatus to upload the ML model to the storage location of the ML model.

40. The apparatus according to any one of claims 23 to 31 and 39, wherein the storage information indicates a storage location to which the apparatus is to upload the ML model, comprising a network address of the host node for storing the ML model, and at least one of a path for the ML model on the host node and a file name associated with the ML model.

41. The device according to any one of claims 23 to 31, 39 and 40, wherein the command comprises an upload command, the upload command instructing the apparatus to upload the ML model to the storage location of the ML model; wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: Uploading the ML model to the storage location via the protocol indicated by the protocol information included in the control message.

42. The apparatus of claim 41 , wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: Calculating a checksum or hash of the uploaded ML model; and The checksum or hash of the uploaded ML model is sent to the network node to allow the network node to determine the integrity of the uploaded ML model.

43. The apparatus of any one of claims 23 to 31 and 39 to 42, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: Receiving from the network node any one of: an upload completion indication indicating that the upload of the ML model was completed by the device and successfully received at the storage location, or an upload failure indication indicating that the upload of the ML model failed.

44. The apparatus of claim 43, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: if the apparatus receives an upload failure indication for the ML model, perform the following: Re-uploading the ML model to the storage location via the protocol indicated by the protocol information; A checksum or hash of the re-uploaded ML model is sent or re-sent to the network node to allow the network node to determine the integrity of the re-uploaded ML model.

45. A non-transitory computer-readable storage medium comprising instructions stored thereon, the instructions, when executed by at least one processor, being configured to cause a computing system to: Receiving a control message for controlling transmission of a machine learning (ML) model from a network node, the control message comprising: a command for a user equipment to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the user equipment for transmission of the ML model via a user plane; as well as Transmission of the ML model between the user device and the storage location indicated by the storage information is performed via the protocol.

46. ​​An apparatus comprising: means for receiving a control message for controlling transmission of a machine learning (ML) model from a network node, the control message comprising: a command for the apparatus to download or upload the ML model, storage information indicating a storage location of the ML model, and protocol information indicating a protocol to be used by the apparatus for transmission of the ML model via a user plane; and means for performing transfer of the ML model between the device and the storage location indicated by the storage information via the protocol.

Images