Method for area-based model updating in mobile communications

Abstract

Various solutions are described for area-based model updating of a device in mobile communications. The device can send area information to a network node to identify an area. The device can receive a model update message from a first network node according to the area. The device can update at least one model according to the model update message.

Description

[0001] Cross-referencing

[0002] This disclosure is part of a non-provisional application claiming priority to PCT application No. PCT / CN2024 / 082266, filed March 18, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure generally relates to mobile communications, and more specifically, to region-based model updates in mobile communications. Background Technology

[0004] Unless otherwise stated, the methods described in this section are not prior art as claimed in the following claims, and are not considered prior art by way of inclusion in this section.

[0005] In New Radio (NR) mobile communication systems, Artificial Intelligence (AI) and / or Machine Learning (ML) technologies have been introduced to enhance system functionality. In particular, AI and ML technologies have been widely applied across various industries, demonstrating significant improvements in operational efficiency. In the context of mobile communication systems, these technologies are likely to be increasingly integrated, with mobile devices gradually adopting AI / ML models to replace traditional algorithmic methods.

[0006] However, when deploying AI / ML models in mobile wireless communications, part of the challenge may lie in maintaining model accuracy and performance as devices traverse different geographical regions. More specifically, due to the inherent limitations of AI / ML model training and adaptation, AI / ML models optimized for a specific region may experience performance degradation when mobile devices operate in areas outside of where the model was initially trained or configured.

[0007] In some scenarios, one possible way to alleviate the above problems is to fine-tune model parameters or use online training methods to dynamically adjust the AI / ML model according to changes in the environment. However, such techniques may introduce a significant computational burden, making them impractical for devices with limited processing power, especially in real-time applications where AI-driven decisions need to be made in milliseconds.

[0008] Therefore, improving the flexibility and efficiency of AI / ML model updates has become a crucial issue in newly developed wireless communication networks. Consequently, it is necessary to provide appropriate solutions to enhance the flexibility and efficiency of AI / ML model updates. Summary of the Invention

[0009] The following abstract is for illustrative purposes only and is not intended to be limiting in any way. That is, the following abstract aims to introduce the concepts, key points, benefits, and advantages of the novel and non-obvious techniques described herein. Some embodiments will be further elaborated in the detailed description below. Therefore, the following abstract is not intended to identify the essential features of the claimed subject matter, nor is it intended to determine the scope of the claimed subject matter.

[0010] One objective of this disclosure is to provide a solution or scheme for problems related to region-based model update devices in mobile communications.

[0011] In one aspect, a method may include sending region information to a first network node by a device to identify a region. The method may also include receiving a model update message from the first network node by the device based on the region. The method may further include updating at least one model by the device based on the model update message.

[0012] In one aspect, a method may include receiving area information from user equipment (UE) by a device. The method may also include identifying an area by the device based on the area information. The method may further include sending a model update message to the UE to update at least one model based on the result of identifying the area.

[0013] In one aspect, a method may include determining multiple regions by means of a device. The method may also include determining multiple models associated with the multiple regions by means of the device. The method may further include determining model region information from network nodes based on the multiple models associated with the multiple regions by means of the device.

[0014] In one aspect, a method may include sending a model update request from a network node by a device. The method may also include receiving model region information from the network node by the device based on the model update request. The method may further include identifying regions by the device based on the model region information.

[0015] It is worth noting that although the content described herein may be set in the context of certain wireless access technologies, networks, and network topologies, such as Long-Term Evolution (LTE), LTE-Advanced, LTE-Advanced Pro, 5th Generation (5G), New Radio (NR), Internet-of-Things (IoT), Narrow Band Internet of Things (NB-IoT), Industrial Internet of Things (IIoT), and 6th Generation (6G), the proposed concepts, schemes, and any variations / derivatives thereof can be implemented, used, and implemented by other types of wireless access technologies, networks, and network topologies. Therefore, the scope of this disclosure is not limited to the examples described herein. Attached Figure Description

[0016] The accompanying drawings are intended to further understand this disclosure and are incorporated into and constitute a part of this disclosure. The drawings illustrate embodiments of this disclosure and, together with the description, serve to explain the principles of this disclosure. It will be understood that the drawings are not necessarily drawn to scale, as some components may be shown out of proportion to their actual dimensions in order to clearly illustrate the concepts of this disclosure.

[0017] Figure 1 This is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0018] Figure 2A-2C This is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0019] Figure 3A and 3B This is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0020] Figure 4A and 4B This is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0021] Figure 5A and 5B This is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0022] Figure 6A and 6BThis is a schematic diagram illustrating an example scenario under the implementation scheme described in this disclosure.

[0023] Figure 7 This is a block diagram of an example communication system according to an embodiment of the present disclosure.

[0024] Figure 8 This is an example flowchart of an implementation method based on the present disclosure.

[0025] Figure 9 This is an example flowchart of an implementation method based on the present disclosure.

[0026] Figure 10 This is an example flowchart of an implementation method based on the present disclosure.

[0027] Figure 11 This is an example flowchart of an implementation method based on the present disclosure. Detailed Implementation

[0028] Detailed embodiments and implementations of the subject matter of the claims of this application are disclosed herein. However, it should be understood that the disclosed embodiments and implementations are merely illustrative of the subject matter of the claims of this application and can be implemented in various forms. This disclosure can be implemented in many different forms and should not be construed as being limited to the exemplary embodiments and implementations set forth herein. Rather, these exemplary embodiments and implementations are provided to make the description of this disclosure exhaustive and complete, and to fully convey the scope of this disclosure to those skilled in the art. In the following description, details of well-known features and techniques may be omitted to avoid unnecessarily obscuring the presented embodiments and implementations.

[0029] Overview

[0030] According to embodiments of this disclosure, various techniques, methods, schemes, and / or solutions related to region-based model updates in mobile communications are involved. According to this disclosure, multiple possible solutions can be implemented individually or in combination. That is, although these possible solutions are described individually below, two or more of them can be implemented in some combination.

[0031] Regarding this disclosure, certain communication mechanisms may be applicable to mobile communication networks (e.g., LTE networks, 5G / NR networks, IoT networks, or 6G networks). Specifically, a mobile communication network may include at least one user equipment (UE), a radio access network (RAN) node, and a data server. The UE may communicate with the RAN node or the data server through one or more base stations (BS). The RAN node may communicate with the data server.

[0032] More specifically, the data server can define different regions based on the network conditions associated with these regions. The data server can then determine different models (e.g., Artificial Intelligence (AI) and / or Machine Learning (ML) models) for different regions.

[0033] In certain scenarios, the data server can send information about defined regions and their associated models to the RAN node. The UE can report UE information (e.g., at least one of UE capabilities and region information) to the RAN node. Upon receiving the UE information, the RAN node can: (1) identify the region where the UE may be located, and (2) determine the model corresponding to that region. The RAN node can send a model update message to the UE so that the UE updates at least one model according to the model update message.

[0034] In certain scenarios, the UE can send a model update request to the data server. Upon receiving the model update request, the data server can send information about the defined region and its associated model to the UE. The UE can: (1) identify the region where the UE may be located, and (2) determine the model corresponding to that region. The UE can update at least one model based on the model corresponding to the identified region.

[0035] Therefore, since models (e.g., AI / ML models) can be region-based (i.e., different models can correspond to different regions), the UE can obtain the necessary information from the network and determine the model corresponding to the region where the UE may be located (e.g., for updating or application). Thus, the flexibility and efficiency of model updates can be significantly improved.

[0036] It should be noted that the model may include AI / ML models for enhancing the network system. Specifically, AI / ML models can be used to enhance Channel State Information (CSI) compression, Beam Management (BM), positioning, mobility, etc. Each AI / ML model can be determined (i.e., trained) by a data server and / or RAN node based on relevant network information (e.g., parameters and / or data) and the corresponding region. More specifically, the data server can define the region (e.g., by cell identifier, physical coordinates, or model input associated with the UE). The data server can then determine the network information for each region and train an AI / ML model for each region based on the corresponding network information.

[0037] For example, the data server defines a region 'P'. Then, the data server determines the CSI compression network information associated with region 'P' and trains a CSI compression AI / ML model 'p' based on the corresponding network information. Therefore, the UE within region 'P' uses model 'p' to determine CSI compression.

[0038] Figure 1 An example scenario 100 is illustrated under a scheme according to an embodiment of the present disclosure. Scenario 100 involves a UE, multiple BSs, RAN nodes, and a data server, which may be part of a wireless communication network (e.g., an LTE network, a 5G / NR network, an IoT network, or a 6G network). The UE can communicate with the RAN node through at least one BS. The RAN node can communicate with the data server. For ease of illustration, one UE, one RAN node, and one data server are described below. However, this is not intended to limit the network scenarios of the present disclosure.

[0039] In some embodiments, the RAN node and the data server may be separate network nodes. In these embodiments, the data server may determine model region information. Specifically, the data server may: (1) collect network data, (2) determine (i.e. define) different regions based on the network data, (3) determine different models (e.g., AI / ML models) for different regions, and (4) store model region information. The model region information may be associated with (or contain) the defined regions and their corresponding pre-determined models.

[0040] Figure 2A An example scenario 200A under the scheme according to an embodiment of this disclosure is illustrated. For example, a data server defines / determines regions 'A', 'B', and 'C' based on network conditions / information (such as the cell identifier of a BS). For region 'A', the data server trains model 'a' based on network parameters associated with region 'A' and determines that model 'a' is used in that region. For region 'B', the data server trains model 'b' based on network parameters associated with region 'B' and determines that model 'b' is used in that region. For region 'C', the data server trains model 'c' based on network parameters associated with region 'C' and determines that model 'c' is used in that region. In some cases, regions 'A', 'B', and 'C' may overlap (e.g., partially overlap). In some cases, regions 'A', 'B', and 'C' may not overlap.

[0041] In some embodiments, the RAN node and the data server may be the same network node (e.g., the RAN node may contain the data server). In these embodiments, the RAN node may determine model region information. Specifically, the RAN node may: (1) collect network data, (2) determine (i.e. define) different regions based on the network data, (3) determine different models (e.g., AI / ML models) for different regions, and (4) store the model region information in a local database. The model region information may be associated with (or contain) the defined regions and their corresponding pre-determined models.

[0042] Figure 2B An example scenario 200B is illustrated under a scheme according to an embodiment of this disclosure. For example, a data server defines / determines regions 'X', 'Y', and 'Z' based on physical coordinates. In this example, a region can be as follows: Figure 2B The diagram shows one or more network parameters (BS). For region 'X', the data server trains model 'x' based on the network parameters associated with region 'X' and determines that model 'x' is used in that region. For region 'Y', the data server trains model 'y' based on the network parameters associated with region 'Y' and determines that model 'y' is used in that region. For region 'Z', the data server trains model 'z' based on the network parameters associated with region 'Z' and determines that model 'z' is used in that region. In some cases, regions 'X', 'Y', and 'Z' may overlap. In other cases, regions 'X', 'Y', and 'Z' may not overlap.

[0043] Figure 2C An example scenario 200C under a scheme according to an embodiment of this disclosure is illustrated. For example, a data server defines / determines regions 'L', 'M', and 'N' based on model inputs associated with the UE (i.e., available model inputs on different UEs). In this example, the data server defines region 'L' for UEs with the same available model inputs (e.g., measurements for cell identifiers 1, 2, and 3). The data server defines region 'M' for UEs with the same available model inputs (e.g., measurements for cell identifiers 1 and 2). The data server defines region 'N' for UEs with the same available model inputs (e.g., measurements for cell identifiers 2 and 3). In some scenarios, the model inputs associated with the UE (i.e., available model inputs) include at least one of the following: Reference Signal Receiving Power (RSRP), CSI, Angle of Arrival (AoA), Timing Advance (TA), UE Capability, UE Speed, UE Location, etc.

[0044] For region 'L', the data server trains model 'l' based on the network parameters associated with region 'L' and determines that model 'l' is used in this region. For region 'M', the data server trains model 'm' based on the network parameters associated with region 'M' and determines that model 'm' is used in this region. For region 'N', the data server trains model 'n' based on the network parameters associated with region 'N' and determines that model 'n' is used in this region. In some cases, regions 'L', 'M', and 'N' may overlap. In other cases, regions 'L', 'M', and 'N' may not overlap.

[0045] Figure 3A An example scenario 300A is illustrated under an embodiment of this disclosure. In scenario 300A, the RAN node and the data server can be separate network nodes. In some embodiments, the data server can collect network data. The data server can define / determine regions based on the network data. The data server can then determine (e.g., train) different models for different regions. The data server can transmit the defined regions and model region information of the pre-determined models to the RAN node. The RAN node can obtain the model region information from the data server. The UE can pre-download the pre-determined models from either the RAN node or the data server.

[0046] In some cases, the UE can report its capabilities to the RAN node. The RAN node can then send area information configuration to the UE capabilities. Based on the area information configuration, the UE can send area information to the RAN node (e.g., information containing the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information, and (2) determine at least one model corresponding to that area based on the model area information. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the identified area.

[0047] Upon receiving a model update message, the UE can update at least one model based on the model update message. More specifically, upon receiving a model update message, the UE can determine at least one specific model from a pre-determined set of models. Subsequently, the UE can update the at least one model using the at least one specific model. In some cases, the UE can send a model update completion message to the RAN node to notify the RAN node that the model update is complete.

[0048] Figure 3BAn example scenario 300B is illustrated under an embodiment of this disclosure. In scenario 300B, the RAN node and the data server can be the same network node. In some embodiments, the RAN node can collect network data. The RAN node can define / determine regions based on the network data. The RAN node can then determine (e.g., train) different models for different regions. The RAN node can store the defined regions and model region information of the pre-determined models in a local database. The RAN node can retrieve the model region information from the local database. The UE can pre-download the pre-determined models from the RAN node.

[0049] In some cases, the UE can report a UE capability report to the RAN node. Subsequently, the RAN node can send area information configuration to the UE. Based on the area information configuration, the UE can send area information to the RAN node (e.g., the information includes the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information stored in its local database, and (2) determine at least one model corresponding to that area based on the model area information stored in its local database. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the area identification result.

[0050] Upon receiving a model update message, the UE can update at least one model based on the message. More specifically, upon receiving a model update message, the UE can determine at least one specific model from multiple predefined models. Subsequently, the UE can update the at least one model using that specific model. In some cases, the UE can send a model update completion message to the RAN node to notify the RAN node that the model update is complete.

[0051] Figure 4A An example scenario 400A is illustrated under a scheme according to an embodiment of this disclosure. In scenario 400A, the RAN node and the data server may be separate network nodes. In some embodiments, the data server may collect network data. The data server may define / determine regions based on the network data. The data server may then determine (e.g., train) different models for different regions. The data server may send the defined regions and model region information of the pre-determined models to the RAN node. The RAN node may obtain the model region information from the data server. The UE may not pre-download the pre-determined model.

[0052] In some cases, the UE can report a UE capability report to the RAN node. The RAN node can then send area information configuration to the UE. Based on the area information configuration, the UE can send area information to the RAN node (e.g., information containing the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information, and (2) determine at least one model corresponding to that area based on the model area information. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the result of identifying the area.

[0053] Upon receiving a model update message, the UE can update at least one model based on the message. More specifically, after receiving the model update message, the UE can send a model update request to the data server (e.g., to request an update of an in-use model). The data server can then send at least one specific model to the UE based on the model update request. After receiving (i.e., downloading) at least one specific model from the data server based on the model update request, the UE can update at least one model using that specific model. In some cases, the UE can send a model update completion message to the RAN node to notify the RAN node that the model update is complete.

[0054] Figure 4B An example scenario 400B is illustrated under a scheme according to an embodiment of this disclosure. In scenario 400B, the RAN node and the data server may be the same network node. In some embodiments, the RAN node may collect network data. The RAN node may define / determine regions based on the network data. The RAN node may then determine (e.g., train) different models for different regions. The RAN node may store the defined regions and model region information of the pre-determined models in a local database. The RAN node may retrieve the model region information from the local database. The UE may not pre-download the pre-determined models from the RAN node.

[0055] In some cases, the UE can report a UE capability report to the RAN node. The RAN node can then send area information configuration to the UE. Based on the area information configuration, the UE can send area information to the RAN node (e.g., information containing the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information, and (2) determine at least one model corresponding to that area based on the model area information stored in its local database. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the result of identifying the area.

[0056] Upon receiving a model update message, the UE can update at least one model based on the message. More specifically, after receiving the model update message, the UE can send a model update request to the RAN node (e.g., to request an update of an in-use model). The RAN node can then send at least one specific model to the UE based on the model update request. After receiving (i.e., downloading) at least one specific model from the data server based on the model update request, the UE can update at least one model using that specific model. In some cases, the UE can send a model update completion message to the RAN node to notify it that the model update is complete.

[0057] Figure 5A An example scenario 500A is illustrated under a scheme according to an embodiment of this disclosure. In scenario 500A, the RAN node and the data server may be separate network nodes. In some embodiments, the data server may collect network data. The data server may define / determine regions based on the network data. The data server may then determine (e.g., train) different models for different regions. The data server may send the defined regions and model region information of the pre-determined models to the RAN node. The RAN node may obtain the model region information from the data server.

[0058] In some cases, the RAN node can send area information configuration to the UE. Based on the area information configuration, the UE can send area information to the RAN node (e.g., the information includes the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information, and (2) determine at least one model corresponding to that area based on the model area information. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the result of identifying the area.

[0059] Upon receiving a model update message, the UE can update at least one model based on the message. More specifically, after receiving the model update message, the UE can send a model update request to the data server (e.g., to request a candidate model update). The data server can then send at least one candidate model to the UE based on the model update request. After receiving at least one candidate model from the data server based on the model update request, the UE can update at least one model using that candidate model. In some cases, the UE can send a model update completion message to the RAN node to notify the RAN node that the model update is complete.

[0060] Figure 5B An example scenario 500B is illustrated under a scheme according to an embodiment of this disclosure. In scenario 500B, the RAN node and the data server can be the same network node. The RAN node can collect network data. The RAN node can define / determine regions based on the network data. The RAN node can then determine (e.g., train) different models for different regions. The RAN node can store the defined regions and model region information of the pre-determined models in a local database. The RAN node can retrieve the model region information from the local database.

[0061] In some cases, the RAN node may send area information configuration to the UE. Based on this area information configuration, the UE can send area information to the RAN node (e.g., the information includes the cell identifier of the connected base station, the UE's physical location information, or model inputs associated with the UE). Upon receiving the area information, the RAN node can: (1) identify the area where the UE may be located based on the area information and model area information, and (2) determine at least one model corresponding to the area based on the model area information. The RAN node can send a model update message (e.g., a model update command) to the UE. The UE can receive the model update message from the RAN node based on the result of identifying the area.

[0062] Upon receiving a model update message, the UE can update at least one model based on the message. More specifically, after receiving the model update message, the UE can send a model update request to the RAN node (e.g., to request a candidate model update). The RAN node can then send at least one candidate model to the UE based on the model update request. After receiving at least one candidate model from the RAN node based on the model update request, the UE can update the at least one model using that candidate model. In some cases, the UE can send a model update completion message to the RAN node to notify it that the model update is complete.

[0063] Figure 6AAn example scenario 600A is illustrated under a scheme according to an embodiment of this disclosure. In scenario 600A, the RAN node and the data server may be separate network nodes. In some embodiments, the data server may collect network data. The data server may define / determine regions based on the network data. The data server may then determine (e.g., train) different models for different regions. The UE may pre-download the pre-determined models from the data server.

[0064] In some cases, the UE can send a model update request to the data server. The data server can send the defined region and the model region information of the pre-determined model to the UE based on the model update request. The UE can receive the model region information from the data server based on the model update request. According to the region information configuration, the UE can: (1) identify the region where the UE may be located based on the UE's location information (e.g., the information includes the cell identifier of the connected base station or the UE's physical location information or the model input associated with the UE) and the model region information, and (2) determine at least one model corresponding to the region based on the model region information.

[0065] The UE can then update at least one model based on the result of identifying the region. More specifically, based on the defined region and the model region information of the pre-determined model, the UE can determine at least one specific model corresponding to the determined region, and update at least one model using the at least one specific model.

[0066] Figure 6B An example scenario 600B is illustrated under a scheme according to an embodiment of this disclosure. In scenario 600B, the RAN node and the data server may be separate network nodes. In some embodiments, the data server may collect network data. The data server may define / determine regions based on the network data. The data server may then determine (e.g., train) different models for different regions.

[0067] In some cases, the UE can send a model update request to the data server. The data server can send the defined region and the model region information of the pre-determined model to the UE based on the model update request. The UE can receive the model region information from the data server based on the model update request. According to the region information configuration, the UE can: (1) identify the region where the UE may be located based on the UE's location information (e.g., the information includes the cell identifier of the connected base station or the UE's physical location information or the model input associated with the UE) and the model region information, and (2) determine at least one model corresponding to the region based on the model region information.

[0068] The UE can then send a candidate model update request to the data server. Upon receiving the candidate model update request, the data server can send at least one candidate model to the UE. The UE can then update at least one model using this candidate model.

[0069] The aforementioned region-based AI / ML model update operations can provide a general framework and process for realizing AI-based application scenarios, where the AI ​​training and / or inference process may depend on the characteristics of a specific region. In some cases, AI / ML schemes can be applied to UEs to predict Radio Resource Management (RRM) measurements. Available AI model inputs can include previously obtained measurements (e.g., RSRP) from the serving cell and neighboring cells. AI / ML model outputs can include predicted measurements, which can be based on time, spatial, or frequency domain predictions. In some cases, the AI / ML model can be pre-trained using measurements obtained from a specific region. As the UE moves between different regions (e.g., when a handover to a different serving cell occurs), the measurements available to the UE change and may not be suitable as input to the original AI / ML model. The aforementioned region-based AI / ML model update operations can facilitate the adaptation of AI / ML models to support region-based AI / ML application scenarios.

[0070] Example Implementation

[0071] Figure 7 An example communication system 700 according to an embodiment of this disclosure is shown, including an example communication device 710, an example network device 720, and an example network device 730. The communication device 710, network device 720, and network device 730 can each perform various functions to implement the schemes, techniques, processes, and methods described herein related to region-based model updates in mobile communications, including the aforementioned scenarios / schemes and processes 800, 900, 1000, and 1100 described below.

[0072] The communication device 710 may be part of an electronic device, which may be a UE (User Equipment), such as a portable or mobile device, a wearable device, a mobile communication device, or a computing device. For example, the communication device 710 may be implemented in a smartphone, smartwatch, personal digital assistant, digital camera, or computing device (such as a tablet, laptop, or mobile phone). The communication device 710 may also be part of a machine-type device, which may be an Internet of Things (IoT), Narrowband Internet of Things (NB-IoT), or Industrial Internet of Things (IIoT) device, such as a non-movable or fixed device, a home device, a wired communication device, or a computing device. For example, the communication device 710 may be implemented in a smart thermostat, a smart refrigerator, a smart door lock, a wireless speaker, or a home control center. Alternatively, the communication device 710 may be implemented as one or more integrated circuit (IC) chips, such as, but not limited to, one or more single-core processors, one or more multi-core processors, one or more reduced-instruction-set computing (RISC) processors, or one or more complex-instruction-set computing (CISC) processors. The communication device 710 may include... Figure 7 The communication device 710 may also include at least some of the components shown, such as processor 712. It may also include one or more other components unrelated to the present disclosure (e.g., internal power supply, display device, and / or user interface device), therefore, these components of the communication device 710 are not included in... Figure 7 This is shown in the text and not described below, in order to simplify and refine the content.

[0073] Network devices 720 / 730 may be part of a network device, which may be a network node, such as a satellite, base station, cell, router, gateway, or data processing center. For example, network devices 720 / 730 may be implemented in a RAN node and / or eNodeB in an LTE network, in a RAN node and / or gNB in ​​a 5G / NR, IoT, NB-IoT, or IIoT network, or in a RAN node, satellite, and / or base station in a 6G network. For example, network devices 720 / 730 may be implemented as a data processing center in a network. Alternatively, network devices 720 / 730 may be implemented as one or more IC chips, such as, but not limited to, one or more single-core processors, one or more multi-core processors, or one or more RISC or CISC processors. Network devices 720 / 730 may include... Figure 7 The network device 720 / 730 may also include at least some of the components shown, such as processors 722 / 732. The network device 720 / 730 may also include one or more other components unrelated to the present disclosure (e.g., internal power supply, display device, and / or user interface device), therefore, these components of the network device 720 / 730 are not... Figure 7 This is shown in the text and not described below, in order to simplify and refine the content.

[0074] In one aspect, each of processors 712, 722, and 732 may be implemented as one or more single-core processors, one or more multi-core processors, or one or more CISC processors. That is, although the singular term "processor" is used herein to refer to processors 712, 722, and 732, each of processors 712, 722, and 732 may comprise multiple processors in some embodiments and a single processor in others, depending on the different implementations of this disclosure. In another aspect, each of processors 712, 722, and 732 may be implemented in hardware (and optionally firmware) and may include, for example, but not limited to, one or more transistors, one or more diodes, one or more capacitors, one or more resistors, one or more inductors, one or more memristors, and / or one or more transformers, these electronic components being configured and arranged to achieve the specific purposes of this disclosure. In other words, in at least some embodiments, each of processors 712, 722 and 732 is a dedicated machine specifically designed, arranged and configured to perform a particular task, including region-based model updates in devices (e.g., represented by communication device 710) and networks (e.g., represented by network devices 720 / 730), consistent with various embodiments of this disclosure.

[0075] In some embodiments, the communication device 710 may further include a transceiver 716 connected to the processor 712, capable of wirelessly transmitting and receiving data. In other words, the processor 712 can transmit and receive data, such as configurations, messages, signals, information, and instructions, through the transceiver 716. In some embodiments, the communication device 710 may further include a memory 714 connected to the processor 712, accessible by the processor 712 and storing data therein. In some embodiments, the network device 720 may further include a transceiver 726 connected to the processor 722, capable of wirelessly transmitting and receiving data. In other words, the processor 722 can transmit and receive data, such as configurations, messages, signals, information, and instructions, through the transceiver 726. In some embodiments, the network device 720 may further include a memory 724 connected to the processor 722, accessible by the processor 722 and storing data therein. In some embodiments, the network device 730 may further include a transceiver 736 connected to the processor 732, capable of wirelessly transmitting and receiving data. In other words, processor 732 can send and receive data, such as configurations, messages, signals, information, and instructions, via transceiver 736. In some embodiments, network device 730 may also include memory 734 connected to processor 732, which can be accessed by processor 732 and stores data therein. Therefore, communication device 710, network device 720, and network device 730 can perform wireless communication via transceiver 716, transceiver 726, and transceiver 736, respectively. For ease of understanding, the following description of the operation, functions, and capabilities of communication device 710 and network devices 720 and 730 is performed in the context of a mobile communication environment, where communication device 710 is implemented as a communication device or UE, and network devices 720 / 730 are implemented as network nodes of a communication network, such as RAN nodes or data centers.

[0076] In some embodiments, each of memories 714, 724, and 734 may include a random-access memory (RAM), such as dynamic RAM (DRAM), static RAM (SRAM), thyristor RAM (T-RAM), and / or zero-capacitor RAM (Z-RAM). Alternatively, or additionally, each of memories 714, 724, and 734 may include a read-only memory (ROM), such as a mask ROM, a programmable ROM (PROM), an erasable programmable ROM (EPROM), and / or an electrically erasable programmable ROM (EEPROM). Alternatively, or additionally, each of memory 714, memory 724 and memory 734 may contain a non-volatile random-access memory (NVRAM), such as flash memory, solid-state memory, ferroelectric RAM (FeRAM), magnetoresistive RAM (MRAM) and / or phase-change memory.

[0077] Example Process

[0078] Figure 8 An example flow 800 of an embodiment of this disclosure is illustrated. Flow 800 may be (whether partially or entirely) an example implementation of the above-described scenarios / solutions, all related to region-based model updates in mobile communications according to this disclosure. Flow 800 may represent one aspect of the functional implementation of communication device 710. Flow 800 may include one or more operations, actions, or functions as shown in blocks 810 to 830. Although shown as separate blocks, the individual blocks of flow 800 may be divided into more blocks, merged into fewer blocks, or omitted depending on the desired implementation. Furthermore, each block of flow 800 may be arranged according to... Figure 8 The process 800 may be executed in the order shown, or in a different order. Process 800 may be implemented by communication device 710 or any suitable communication device (e.g., UE) or machine-type device. For illustrative purposes only and without limitation, process 800 is described below in the context of communication device 710. Process 800 may begin at block 810.

[0079] In block 810, process 800 may involve the processor 712 of communication device 710 sending area information to a first network node to identify an area. Process 800 may proceed from block 810 to block 820.

[0080] In block 820, process 800 may involve the processor 712 of communication device 710 receiving model update messages from the first network node according to the region. Process 800 may proceed from block 820 to block 830.

[0081] In block 830, process 800 may involve the processor 712 of communication device 710 updating at least one model according to the model update message.

[0082] In some implementations, process 800 may further involve processor 712 sending a capability report to a first network node. Process 800 may further involve processor 712 receiving area information configuration from the first network node based on the capability report. This area information configuration can be sent accordingly.

[0083] In some implementations, process 800 may further involve processor 712 determining at least one specific model from a plurality of predefined models based on the model update message. Process 800 may further involve processor 712 updating the at least one model using the at least one specific model.

[0084] In some implementations, process 800 may further involve processor 712 sending an in-use model update request to a second network node based on the model update message. Process 800 may further involve processor 712 receiving at least one specific model from the second network node based on the in-use model update request. Process 800 may further involve processor 712 updating the at least one model using the at least one specific model. The first network node and the second network node may be the same network node or different network nodes.

[0085] In some implementations, process 800 may further involve processor 712 receiving area information configuration from a first network node. This area information can be sent according to the area information configuration.

[0086] In some implementations, process 800 may further involve processor 712 sending a candidate model update request to a second network node according to the model update message. Process 800 may further involve processor 712 receiving at least one candidate model from the second network node according to the candidate model update request. Process 800 may further involve processor 712 updating the at least one model using the at least one candidate model. The first network node and the second network node may be the same network node or different network nodes.

[0087] Figure 9 An example flow 900 of an embodiment of this disclosure is illustrated. Flow 900 may be (whether partially or entirely) an example implementation of the above-described scenarios / solutions, all related to region-based model updates in mobile communications according to this disclosure. Flow 900 may represent one aspect of the functional implementation of network device 720. Flow 900 may include one or more operations, actions, or functions as shown in blocks 910 to 930. Although shown as separate blocks, the individual blocks of flow 900 may be divided into more blocks, merged into fewer blocks, or omitted depending on the desired implementation. Furthermore, each block of flow 900 may be arranged according to... Figure 9 The process 900 may be executed in the order shown, or in a different order. Process 900 may be implemented by network device 720 or any suitable network device (e.g., RAN node) or machine-type device. For illustrative purposes only and without limitation, process 900 is described below in the context of network device 720. Process 900 may begin at block 910.

[0088] In block 910, process 900 may involve the processor 722 of network device 720 receiving area information from the UE. Process 900 can proceed from block 910 to block 920.

[0089] In block 920, process 900 may involve the processor 722 of network device 720 identifying the region of the UE based on the region information. Process 900 may then proceed from block 920 to block 930.

[0090] In block 930, process 900 may involve the processor 722 of network device 720 sending a model update message to the UE based on the result of identifying the region to update at least one model.

[0091] In some implementations, process 900 may further involve processor 722 obtaining model region information from a local database or network node. The region can be identified based on this region information and the model region information.

[0092] In some implementations, the model region information can be obtained from a local database. Process 900 may further involve processor 722 determining multiple regions. Process 900 may further involve processor 722 determining multiple models associated with the multiple regions. Process 900 may further involve processor 722 storing the model region information in the local database based on the multiple models associated with the multiple regions.

[0093] In some implementations, process 900 may further involve processor 722 receiving a capability report from the UE. Process 900 may further involve processor 722 sending a region information configuration to the UE based on the capability report. The region information can be received according to the region information configuration.

[0094] In some implementations, the model update message can be used to update at least one model using at least one specific model. This at least one specific model can be determined from a plurality of predefined models, obtained from a local database, or obtained from a network node.

[0095] In some implementations, process 900 may further involve processor 722 sending area information configuration to the UE. The area information can be received according to the area information configuration.

[0096] In some implementations, the model update message can be used to update the at least one model with at least one candidate model, and the at least one candidate model can be obtained from a local database or from a network node.

[0097] Figure 10 An example flow 1000 of an embodiment of this disclosure is illustrated. Flow 1000 may be (whether partially or entirely) an example implementation of the above-described scenario / solution, for region-based model updates in mobile communications according to this disclosure. Flow 1000 may represent one aspect of a feature implementation of network device 730. Flow 1000 may include one or more operations, actions, or functions as shown in blocks 1010 to 1030. Although shown as discrete blocks, the individual blocks of flow 1000 may be divided into more blocks, merged into fewer blocks, or omitted, depending on the desired implementation. Furthermore, the blocks of flow 1000 may be arranged according to... Figure 10 The process can be executed in the order shown, or in a different order. Process 1000 can be implemented by network device 730 or any suitable network device (e.g., a data server) or machine-type device. For illustrative purposes only and without limitation, process 1000 is described below in the context of network device 730. Process 1000 may begin at block 1010.

[0098] In block 1010, process 1000 may involve the processor 732 of network device 730 determining multiple regions. Process 1000 can proceed from block 1010 to block 1020.

[0099] In block 1020, process 1000 may involve the processor 732 of network device 730 determining multiple models associated with the multiple regions. Process 1000 may proceed from block 1020 to block 1030.

[0100] In block 1030, process 1000 may involve the processor 732 of network device 730 sending model region information to network nodes based on the multiple models associated with the multiple regions.

[0101] In some implementations, these multiple regions can be determined based on cell identifiers, physical coordinates, or model inputs associated with the UE.

[0102] In some implementations, process 1000 may further involve processor 732 receiving an in-use model update request from the UE. Process 1000 may further involve processor 732 sending at least one specific model to the UE based on the in-use model update request, so as to update at least one model using the at least one specific model.

[0103] In some implementations, process 1000 may further involve processor 732 receiving a candidate model update request from the UE. Process 1000 may further involve processor 732 sending at least one candidate model to the UE according to the candidate model update request, so as to update the at least one model with the at least one candidate model.

[0104] In some implementations, the multiple models associated with the multiple regions include machine learning models.

[0105] Figure 11 An example flow 1100 of an embodiment of this disclosure is illustrated. Flow 1100 may be (whether partially or entirely) an example implementation of the above-described scenario / solution, for region-based model updates in mobile communications according to this disclosure. Flow 1100 may represent one aspect of a characteristic implementation of communication device 710. Flow 1100 may include one or more operations, actions, or functions as shown in blocks 1110 to 1130. Although shown as discrete blocks, the individual blocks of flow 1100 may be divided into more blocks, merged into fewer blocks, or omitted, depending on the desired implementation. Furthermore, the blocks of flow 1100 may be arranged according to... Figure 11 The process can be executed in the order shown, or in a different order. Process 1100 can be implemented by communication device 710 or any suitable communication device (e.g., UE) or machine-type device. For illustrative purposes only and without limitation, process 1100 is described below in the context of communication device 710. Process 1100 may begin from block 1110.

[0106] In block 1110, process 1100 may involve the processor 712 of communication device 710 sending a model update request to a network node. Process 1100 can then proceed from block 1110 to block 1120.

[0107] In block 1120, process 1100 may involve the processor 712 of communication device 710 receiving model region information from the network node according to the model update request. Process 1100 can then proceed from block 1120 to block 1130.

[0108] In block 1130, process 1100 may involve the processor 712 of communication device 710 identifying the region based on the model region information.

[0109] In some implementations, process 1100 may involve processor 712 updating at least one model from a plurality of predefined models based on the result of identifying the region.

[0110] In some implementations, process 1100 may involve processor 712 sending a candidate model update request to a network node based on the result of identifying the region. Process 1100 may involve processor 712 receiving at least one candidate model from the network node according to the candidate model update request. Process 1100 may involve processor 712 updating the at least one model using the at least one candidate model.

[0111] Additional notes

[0112] The subject of this description sometimes shows different components contained within or connected to other components. It should be understood that such architectures are merely examples, and many other architectures can actually be implemented to achieve the same functionality. Conceptually, any arrangement of components to achieve the same functionality is effectively “associated” to achieve the desired function. Therefore, any two components combined in this document to achieve a specific function can be considered “associated” to achieve the desired function, regardless of the architecture or intermediate components. Similarly, any two such associated components can also be considered “operationally connected” or “operationally coupled” to achieve the desired function, and any two components that can be suchly associated can also be considered “operationally coupled” to achieve the desired function. Specific examples of operational coupling include, but are not limited to, physically matable and / or physically interactive components and / or wirelessly interactive and / or logically interactive components.

[0113] Furthermore, regarding the use of almost all plural and / or singular terms in this document, those skilled in the art can appropriately convert from plural to singular and / or from singular to plural depending on the context and / or application. For clarity, various singular / plural permutations are explicitly listed herein.

[0114] Furthermore, those skilled in the art will understand that terms commonly used herein, particularly in appended claims, such as the body portion of appended claims, are generally considered "open-ended" terms. For example, the word "comprising" should be interpreted as "including but not limited to," the word "having" should be interpreted as "having at least," and the word "including" should be interpreted as "including but not limited to," etc. Those skilled in the art will also further understand that if a specific quantity is introduced in a claim intentionally, that intention will be explicitly stated in the claim; if no such statement is made, then that intention does not exist. For example, for ease of understanding, the appended claims described below may contain the introductory phrases "at least one" and "one or more" to introduce the content of the claim. However, the use of such phrases should not be construed as limiting any particular claim containing that content to containing only one instance of that content, even if the same claim contains both the introductory phrases "one or more" or "at least one" and indefinite articles such as "a" or "one," for example, "a" and / or "one" should be interpreted as "at least one" or "one or more"; the same applies to definite articles used to introduce the content of the claim. Furthermore, even if a specific quantity is explicitly stated in the claims, those skilled in the art will recognize that such a statement should be interpreted as at least the stated quantity. For example, simply stating "two items" without further modification implies at least two items, or two or more items. Additionally, when using conventions such as "at least one A, B, and C," such structures should generally be interpreted in the way that those skilled in the art understand the convention. For example, "a system having at least one A, B, and C" includes, but is not limited to, systems with only A, only B, only C, A and B, A and C, B and C, and systems where A, B, and C coexist. Similarly, when using conventions such as "at least one A, B, or C," such structures should generally be interpreted in the way that those skilled in the art understand the convention. For example, "a system having at least one A, B, or C" includes, but is not limited to, systems with only A, only B, only C, A and B, A and C, B and C, and systems where A, B, and C coexist. Those skilled in the art will further understand that virtually any separable words and / or phrases appearing in the specification, claims, or drawings, presenting two or more alternative terms, should be understood to include the possibility of including one term, any two terms, or all of the terms. For example, the phrase “A or B” should be understood as including the possibility of “A” or “B” or “A and B”.

[0115] As can be seen from the foregoing, various embodiments of this disclosure have been described for illustrative purposes, and various modifications can be made without departing from the scope and spirit of this disclosure. Therefore, the various embodiments disclosed herein are not intended to be limiting, and the true scope and spirit are defined by the following claims.

Claims

1. A method comprising: The device's processor sends region information to the first network node to identify the region; The processor receives the model update message from the first network node based on the region. as well as The processor updates at least one model based on the model update message.

2. The method as described in claim 1, wherein, Updating the at least one model based on the model update message further includes: The processor determines at least one specific model from multiple predefined models based on the model update message; and The processor updates the at least one model using the at least one specific model.

3. The method as described in claim 1, wherein, Updating the at least one model based on the model update message further includes: The processor sends an update request for the in-use model to the second network node based on the model update message; The processor receives at least one specific model from the second network node based on the in-use model update request; and The processor updates the at least one model using the at least one specific model. The first network node and the second network node may be the same network node or different network nodes.

4. The method of claim 1, wherein, Further includes: The processor receives area information from the first network node for configuration. The information about this region is sent based on the configuration of this region information.

5. The method of claim 1, wherein, Updating the at least one model based on the model update message further includes: The processor sends a candidate model update request to the second network node based on the model update message; The processor receives at least one candidate model from the second network node according to the candidate model update request; and The processor updates the at least one model using the at least one candidate model. The first network node and the second network node may be the same network node or different network nodes.

6. A method comprising: The device's processor receives area information from the user equipment; The processor identifies the region of the user equipment based on the region information; as well as The processor sends a model update message to the user equipment based on the results of identifying the region to update at least one model.

7. The method of claim 6, wherein, Further includes: The processor obtains model region information from a local database or network node. The region is identified based on the region information and the model region information.

8. The method of claim 7, wherein, The model region information is obtained from the local database, and the method further includes: The processor determines multiple regions; The processor determines multiple models associated with the multiple regions; and The processor stores the model region information in the local database based on the multiple models associated with the multiple regions.

9. The method of claim 6, wherein, Further includes: The processor receives capability reports from the user equipment; and The processor sends the area information configuration to the user equipment based on the capability report. The information about this region is received based on the configuration of this region information.

10. The method of claim 6, wherein, The model update message is used to update the at least one model with at least one specific model, wherein the at least one specific model is determined from a plurality of predefined models, obtained from a local database, or obtained from a network node.

11. The method of claim 6, wherein, Further includes: The processor sends the area information configuration to the user equipment. The information about this region is received based on the configuration of this region information.

12. The method of claim 6, wherein, The model update message is used to update the at least one model with at least one candidate model, wherein the at least one candidate model is obtained from a local database or from a network node.

13. A method comprising: The device's processor determines multiple regions; The processor determines multiple models associated with these multiple regions; as well as The processor sends model region information to the network nodes based on the multiple models associated with the multiple regions.

14. The method of claim 13, wherein, These multiple regions are determined based on cell identifiers, physical coordinates, or model inputs associated with user equipment.

15. The method of claim 13, wherein, Further includes: The processor receives the in-use model update request from the user equipment; and The processor sends at least one specific model to the user equipment based on the in-use model update request, so as to update at least one model through the at least one specific model.

16. The method of claim 13, wherein, Further includes: The processor receives candidate model update requests from the user equipment. as well as The processor sends at least one candidate model to the user equipment according to the candidate model update request, so as to update the at least one model through the at least one candidate model.

17. The method of claim 13, wherein, The multiple models associated with these multiple regions include machine learning models.

18. A method comprising: The device's processor sends a model update request to the network node; The processor receives model region information from the network node based on the model update request; as well as The processor identifies the region based on the model region information.

19. The method of claim 18, wherein, Also includes: The processor updates at least one model from multiple predefined models based on the results of identifying the region.

20. The method of claim 18, wherein, Also includes: Based on the results of identifying the region, the processor sends a candidate model update request to the network node; The processor receives at least one candidate model from the network node based on the candidate model update request. as well as The processor updates the at least one model using the at least one candidate model.

Images