Data collection method, indication method for data collection, and apparatus

Abstract

A data collection method, an indication method for data collection, and an apparatus. The data collection method comprises: a terminal device receiving first information sent by a first network device or a second network device; and the terminal device collecting first data by using a first configuration matching the first information, wherein the first data is used for at least one of training a first model and supervising the first model, and the first model is an AI / ML model applied to the terminal device side.

Description

Data collection methods, data collection instruction methods and devices Technical Field

[0001] This application relates to the field of communications. Background Technology

[0002] In 3GPP Release 19, the application of Artificial Intelligence (AI) and Machine Learning (ML) technologies in communication systems is being studied. The AI / ML framework used in communication systems, like general AI / ML frameworks, includes data collection, model training, derivation, and performance supervision. The data collected, besides being used to train AI / ML models, can also be used to test and supervise the performance of AI / ML models. This is the most fundamental and crucial step in the AI / ML framework, and the content and quality of the collected data have a significant impact on the performance of the AI / ML model.

[0003] For communication systems, the entities used for training AI / ML models may differ depending on the entity applying the AI / ML model. For example, for AI / ML models applied to terminal devices, due to issues such as the size, computing power, storage, and power consumption of the terminal devices, training is typically not performed on the terminal devices themselves, but rather on related servers or devices. Therefore, after the terminal device collects the data for AI / ML model training, it needs to transfer this data to a server or device capable of performing AI / ML model training. Higher-level network devices in the communication system, such as the core network (CN) or operations, administration, and maintenance (OAM) system, may have the function of generating, training, and maintaining AI / ML models, or may have interfaces or pathways with servers or devices capable of AI / ML model training. Therefore, they can be responsible for managing AI / ML model training, for example, by collecting data from the terminal devices and directly training the AI / ML model, or by transferring this data to a server or device capable of performing AI / ML model training.

[0004] On the one hand, unlike existing terminal device reports, this data may only be used for AI / ML models on the terminal device side and does not require processing by the radio access network (RAN) network devices. On the other hand, for the protection of data privacy and security, the RAN network devices may not be allowed to read, process or use this data. Therefore, the content or process of receiving this data collected by the terminal device by the higher-level network devices may be invisible to the RAN network devices.

[0005] According to 3GPP's work progress, the standardization of data collection for AI / ML models on the terminal device side needs to meet the following requirements: the MNO (Mobile Network Operator) must have full control over the data collection and transmission process, and the MNO must be able to manage the transmission of data from the terminal device to the relevant server. Whether through the CN (Network Network) or OAM (Operational Information Center), data collected by the terminal device can be received through the user plane channel; therefore, this data collection process may be transparent to the RAN (Radio Network Equipment) network equipment.

[0006] It should be noted that the above introduction to the technical background is only for the purpose of providing a clear and complete explanation of the technical solutions of this application and facilitating understanding by those skilled in the art. It should not be assumed that these technical solutions are known to those skilled in the art simply because they have been described in the background section of this application. Summary of the Invention

[0007] To train AI / ML models or functions for use on terminal devices, these devices need to collect the necessary training data and transmit it to higher-level network equipment. The higher-level network equipment then executes or assists in training the AI / ML model. Based on the discussions at the 3GPP RAN106 plenary meeting, future research will focus on the impact on RAN-related aspects when terminal devices transmit collected data to higher-level network equipment via the user plane tunnel (UP tunnel), and the extent to which NG-RAN participates in controlling and configuring the data collected by the terminal devices.

[0008] Terminal devices participate in processes such as beam management, channel estimation, and signal measurement, which require configurations from network equipment in their Radio Access Network (RAN), such as gNBs. For example, terminal devices need to know the time-frequency resources and periods occupied by various reference signals transmitted by the cell in order to measure these reference signals, evaluate the received signal level, received signal quality, SINR, etc., corresponding to different reference signals, and ultimately determine the reported beam information, measurement results, and event triggers.

[0009] Using the above information, AI / ML technology can train models to obtain AI / ML models applicable to terminal devices. Therefore, the terminal device needs to send the above information, such as relevant configurations, measurement results, and evaluation reports, as training data to the server, device, or equipment used for AI / ML model training. For the terminal device to obtain this training data, relevant configurations from the RAN network equipment or the cell are required. Given that the entire data collection process is transparent to the RAN network equipment, the question of how the terminal device should obtain the resources or parameters used for data collection needs to be addressed.

[0010] To ensure consistency between the model used for inference on the terminal device and the scenario in which the model was trained, it is necessary to associate the model with the application scenario during training and to equip the terminal device and network device with new relevant information about this association. The terminal device can select a matching AI / ML model based on this information, and the network device can configure appropriate resources or parameters for the terminal device based on this information. However, currently, there is no discussion on how to determine the association between the model on the terminal device side and the adapted scenario (including corresponding resources), or how to notify the RAN network device.

[0011] Since the collected data is used for AI / ML model training rather than for regular reporting, the RAN network devices may not need to know the report results corresponding to the relevant configurations. The current resource configurations are usually indicated to the end devices by the RAN network devices along with the corresponding report configurations.

[0012] Measurement and evaluation of terminal devices are typically controlled by the RAN network equipment, including configuration parameters, measurement resources, and measurement reports. The results of these measurements and evaluations are used for training to obtain AI / ML models, which can then be applied to the terminal devices. However, the data collection and transmission process via the UP channel may be transparent to the RAN network equipment; therefore, how to implement data collection in this case requires specific definitions.

[0013] To address one or more of the aforementioned problems, embodiments of this application provide a data collection method, a data collection instruction method, and an apparatus.

[0014] According to one aspect of the embodiments of this application, a data collection apparatus is provided, the apparatus being applied to a terminal device, the apparatus comprising: a first receiving unit for receiving first information sent by a first network device or a second network device; and a collection unit for collecting first data using a first configuration matching the first information, the first data being used to train a first model and supervise at least one of the first model, the first model being an AI / ML model applied to the terminal device side.

[0015] According to another aspect of the embodiments of this application, a data collection instruction device is provided, the device being applied to a first network device, the device comprising: a second sending unit that sends first information to a terminal device and / or a second network device, the first information being at least related to the terminal device collecting first data, the first data being used for at least one of training a first model and supervising the first model, the first model being an AI / ML model applied to the terminal device side.

[0016] According to another aspect of the embodiments of this application, a data collection instruction device is provided, the device being applied to a second network device, the device comprising: a second receiving unit that receives first information from a first network device, or receives first information and / or configuration requirement information from a terminal device; a third sending unit that sends a first configuration matching the first information or the configuration requirement information to the terminal device; the first information or the configuration requirement information being at least related to the terminal device collecting first data, the first data being used for at least one of training a first model and supervising the first model, the first model being an AI / ML model applied to the terminal device side.

[0017] According to another aspect of the embodiments of this application, a communication system is provided, the communication system including at least one of a terminal device, a first network device, and a second network device, the terminal device including a data collection device according to the embodiments of this application, the first network device including a data collection instruction device according to another aspect of the embodiments of this application, and the second network device including a data collection instruction device according to yet another aspect of the embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a data collection method is provided, the method being applied to a terminal device, the method comprising: the terminal device receiving first information sent by a first network device or a second network device; the terminal device collecting first data using a first configuration matching the first information, the first data being used to train at least one of a first model and to supervise the first model, the first model being an AI / ML model applied to the terminal device side.

[0019] According to another aspect of the embodiments of this application, a data collection instruction method is provided, the method being applied to a first network device, the method comprising: the first network device sending first information to a terminal device and / or a second network device, the first information being at least related to the collection of first data by the terminal device, the first data being used for at least one of training a first model and supervising the first model, the first model being an AI / ML model applied to the terminal device side.

[0020] According to another aspect of the embodiments of this application, a data collection instruction method is provided, the method being applied to a second network device, the method comprising: the second network device receiving first information from a first network device, or the second network device receiving first information and / or configuration requirement information from a terminal device; the second network device sending a first configuration matching the first information or the configuration requirement information to the terminal device; the first information or the configuration requirement information being at least related to first data collected by the terminal device, the first data being used for at least one of training a first model and supervising the first model, the first model being an AI / ML model applied to the terminal device side.

[0021] According to another aspect of the embodiments of this application, a computer-readable program is provided, wherein when the program is executed in a data collection device or terminal device, the program causes the data collection device or terminal device to perform the data collection method described in the embodiments of this application.

[0022] According to another aspect of the embodiments of this application, a computer-readable program is provided, wherein when the program is executed in a data collection instruction device or a first network device or a second network device, the program causes the data collection instruction device or network device to perform the data collection instruction method described in the embodiments of this application.

[0023] According to another aspect of the embodiments of this application, a storage medium storing a computer-readable program is provided, wherein the computer-readable program causes a data collection device or terminal device to perform the data collection method described in the embodiments of this application.

[0024] According to another aspect of the embodiments of this application, a storage medium storing a computer-readable program is provided, wherein the computer-readable program causes a data collection instruction device or a first network device or a second network device to perform the data collection instruction method described in the embodiments of this application.

[0025] One of the beneficial effects of the embodiments of this application is that:

[0026] The terminal device receives first information sent by a first network device (e.g., a higher-layer network device) or as a second network device (e.g., a RAN network device), and collects first data for training and / or supervising AI / ML models using a first configuration that matches the first information. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0027] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0028] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0029] In addition, this application also provides a method for determining the relationship between information such as configuration and related information of AI / ML models or functions, so that all entities involved in AI / ML model training and use can jointly ensure the consistency of AI / ML model training and derivation stages based on the determined relationship.

[0030] Specific embodiments of this application are disclosed in detail with reference to the following description and accompanying drawings, indicating how the principles of this application can be adopted. It should be understood that the embodiments of this application are not limited in scope. Within the spirit and scope of the appended claims, embodiments of this application include many changes, modifications, and equivalents.

[0031] Features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

[0032] It should be emphasized that the term "including / comprises / has" as used herein refers to the presence of a feature, whole, step, or component, but does not exclude the presence or addition of one or more other features, wholes, steps, or components. Attached Figure Description

[0033] The elements and features described in one drawing or embodiment of this application may be combined with elements and features shown in one or more other drawings or embodiments. Furthermore, in the drawings, similar reference numerals denote corresponding parts in several drawings and can be used to indicate corresponding parts used in more than one embodiment.

[0034] The accompanying drawings, which form part of the specification, are used to provide a further understanding of the embodiments of this application and illustrate the implementation methods of this application, together with the textual description, to explain the principles of this application. Obviously, the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any creative effort. In the drawings:

[0035] Figure 1 is a schematic diagram of a communication system according to an embodiment of this application;

[0036] Figure 2 is another schematic diagram of the communication system according to an embodiment of this application;

[0037] Figure 3 is a schematic diagram of terminal equipment evaluation;

[0038] Figure 4 is a schematic diagram of AI / ML model training;

[0039] Figure 5 is a schematic diagram of a data collection method according to an embodiment of this application;

[0040] Figure 6 is a schematic diagram of a data collection instruction method according to an embodiment of this application;

[0041] Figure 7 is a schematic diagram of a data collection instruction method according to an embodiment of this application;

[0042] Figure 8 is a schematic diagram of a data collection device according to an embodiment of this application;

[0043] Figure 9 is a schematic diagram of a data collection indication device according to an embodiment of this application;

[0044] Figure 10 is a schematic diagram of a data collection indication device according to an embodiment of this application;

[0045] Figure 11 is a schematic block diagram of the system configuration of a terminal device according to an embodiment of this application;

[0046] Figure 12 is a schematic block diagram of the system configuration of a network device according to an embodiment of this application;

[0047] Figure 13 is a schematic diagram of a data collection method according to an embodiment of this application. Detailed Implementation

[0048] Referring to the accompanying drawings, the foregoing and other features of this application will become apparent from the following description. Specific embodiments of this application are specifically disclosed in the description and drawings, illustrating partial implementations in which the principles of this application may be employed. It should be understood that this application is not limited to the described embodiments; rather, it includes all modifications, variations, and equivalents falling within the scope of the appended claims.

[0049] In the embodiments of this application, the terms "first," "second," etc., are used to distinguish different elements by name, but do not indicate the spatial arrangement or chronological order of these elements, and these elements should not be limited by these terms. The term "and / or" includes any one or more of the terms listed in association and all combinations thereof. The terms "comprising," "including," "having," etc., refer to the presence of the stated features, elements, components, or assemblies, but do not exclude the presence or addition of one or more other features, elements, components, or assemblies.

[0050] In the embodiments of this application, the singular forms "a," "the," etc., including the plural forms, should be broadly understood as "a kind" or "a class" rather than limited to the meaning of "an." Furthermore, the term "the" should be understood to include both the singular and plural forms, unless the context explicitly indicates otherwise. Additionally, the term "according to" should be understood as "at least partially based on…," and the term "based on" should be understood as "at least partially based on…," unless the context explicitly indicates otherwise.

[0051] In the embodiments of this application, the term "communication network" or "wireless communication network" may refer to a network that conforms to any of the following communication standards, such as Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), etc.

[0052] Furthermore, communication between devices in a communication system can be carried out according to communication protocols at any stage, including but not limited to the following communication protocols: 1G (generation), 2G, 2.5G, 2.75G, 3G, 4G, 4.5G, as well as 5G, 5G-Advanced, New Radio (NR), 6G, etc., and / or other currently known or future communication protocols.

[0053] In the embodiments of this application, the term "network device" refers, for example, to a device in a communication system that connects a terminal device to a communication network and provides services to that terminal device. Network devices may include, but are not limited to, the following devices: base station (BS), access point (AP), transmission reception point (TRP), broadcast transmitter, mobile management entity (MME), gateway, server, radio network controller (RNC), base station controller (BSC), OAM (Operations, Administration, and Maintenance), OAM domain nodes, core network, core network domain nodes, core network domain network elements, such as AMF, SMF, UPF, AUSF, UDM, UDR, NSSF, NEF, PCF, NRF, AF, DN, etc.

[0054] Base stations can include, but are not limited to: NodeBs (or NBs), evolved NodeBs (eNodeBs or eNBs), and 5G base stations (gNBs), Centralized Units (CUs), Distributed Units (DUs), etc. They may also include Remote Radio Heads (RRHs), Remote Radio Units (RRUs), relays, or low-power nodes (e.g., femeto, pico, etc.). The term "base station" can encompass some or all of their functions, and each base station can provide communication coverage to a specific geographic area. The term "cell" can refer to a base station and / or its coverage area, depending on the context in which the term is used.

[0055] In the embodiments of this application, the terms "User Equipment" (UE) or "Terminal Equipment" (TE) refer, for example, to a device that accesses a communication network and receives network services through a network device. A terminal device can be fixed or mobile, and may also be referred to as a mobile station (MS), terminal, subscriber station (SS), access terminal (AT), station, etc.

[0056] Terminal devices may include, but are not limited to, the following devices: cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, machine-type communication devices, laptops, cordless phones, smartphones, smartwatches, digital cameras, etc.

[0057] For example, in scenarios such as the Internet of Things (IoT), terminal devices can also be machines or devices for monitoring or measurement, such as including but not limited to: machine-type communication (MTC) terminals, vehicle communication terminals, device-to-device (D2D) terminals, machine-to-machine (M2M) terminals, and so on.

[0058] Furthermore, the terms "network side" or "network equipment side" refer to one side of the network, which can be a base station or include one or more network devices as described above. The terms "user side," "terminal side," or "terminal equipment side" refer to the side of the user or terminal, which can be a UE or include one or more terminal devices as described above. Unless otherwise specified, "equipment" can refer to either network equipment or terminal equipment.

[0059] The following examples illustrate the scenarios of embodiments of this application, but this application is not limited thereto.

[0060] Figure 1 is a schematic diagram of a communication system according to an embodiment of this application, illustrating the case of a terminal device and a network device as examples. As shown in Figure 1, the communication system 100 may include a first network device 101, a second network device 102, and a terminal device 103.

[0061] In this embodiment of the application, the first network device 101 may be a higher-level network device, such as a core network (CN) or OAM (Operations, Administration and Maintenance) or a node within the CN domain or a node within the OAM domain or a network element within the CN domain.

[0062] The second network device 102 can be a network device of the RAN, such as a base station.

[0063] For simplicity, Figure 1 is illustrated using only one terminal device and one second network device as an example, but the embodiments of this application are not limited thereto.

[0064] Figure 2 is another schematic diagram of the communication system according to an embodiment of this application, illustrating the case of a terminal device and a network device as examples, and the communication between the terminal device and the network device.

[0065] As shown in Figure 2, the network-side equipment includes an RU (radio frequency unit) for air interface transmission and reception, and a CU / DU for related radio access (RAN). It also includes core network-related processing units (CT / OAM / related processing units), edge computing-related processing units, and data units related to the core network or RAN side.

[0066] Figure 2 is merely an illustrative example; the relevant equipment units can be deployed in a centralized or distributed manner. Furthermore, the network-side equipment, excluding the RU, considers a hardware-software hybrid device based on a CPU / GPU architecture. The related signal processing extensively utilizes AI / ML models for implementation. The network equipment specifically refers to the equipment corresponding to the air interface-related RAN-side CU / DU units and their connected RU units, forming the traditional base station function, or the network equipment of the new architecture of 6G communication systems. The terminal-side equipment, in addition to the air interface-facing terminal equipment, may also include terminal-related server equipment and related data storage or data management units. The term "terminal-side equipment" here refers to its logical affiliation, not its actual physical location; for example, the terminal-side data unit may also exist within the network-side equipment. Similarly, the terminal-side equipment or the terminal equipment's signal processing unit also extensively utilizes AI / ML models for implementation.

[0067] As mentioned earlier, the terminal device participates in beam management, channel estimation, and signal measurement processes based on relevant configurations (such as resource parameters) from the RAN network equipment. Figure 3 is a schematic diagram of the terminal device performing the evaluation. As shown in Figure 3, the terminal device evaluates the received signal level, received signal quality, SINR, etc., corresponding to different reference signals based on resource parameters, such as the time-frequency resources and periods occupied by various reference signals transmitted by the cell, and finally determines the reported beam information, measurement results, event triggers, etc., such as RSRP, RSRQ, SINR, beam reports, CSI reports, and event reports.

[0068] The above information can be used to train an AI / ML model, resulting in an AI / ML model applicable to the terminal device. Figure 4 is a schematic diagram of AI / ML model training. As shown in Figure 4, the configuration, measurement results, and evaluation results are input into the model for training, resulting in a trained AI / ML model.

[0069] Various embodiments of the present application will now be described with reference to the accompanying drawings. These embodiments are merely exemplary and are not intended to limit the scope of the present application.

[0070] First aspect of the embodiments

[0071] This application provides a data collection method, which is applied to a terminal device, such as terminal device 103 in FIG1.

[0072] Figure 5 is a schematic diagram of a data collection method according to an embodiment of this application. As shown in Figure 5, the method includes:

[0073] 501: The terminal device receives first information sent by the first network device or the second network device;

[0074] 502: The terminal device uses a first configuration matching the first information to collect first data.

[0075] In this embodiment of the application, the first data is used to train at least one of the first model and to supervise the first model, wherein the first model is an AI / ML model applied to the terminal device side.

[0076] In this embodiment of the application, the first network device is a higher-level network device, such as a core network (CN), an OAM (Operations, Administration, and Maintenance) node, an OAM node, or a CN domain element.

[0077] The second network device is the RAN network device, for example, the second network device is a base station.

[0078] In this embodiment of the application, the first information is at least related to the first data collected by the terminal device. The first data is used for at least one of training a first model and supervising the first model. The first model is an AI / ML model applied to the terminal device side.

[0079] In the embodiments of this application, the AI / ML model can also be an AI / ML function, an AI / ML model corresponding to an AI / ML function, or an AI / ML model that can support or implement an AI / ML function.

[0080] Due to limitations in the capabilities of terminal devices and the versatility of AI / ML models across terminal devices of the same type or attributes, training of AI / ML models on the terminal device side is usually not completed on a single terminal device, but rather on an entity specifically designed for training and generating AI / ML models applicable to the terminal device side, such as an AI / ML server.

[0081] The input and output of the AI / ML model applied to the terminal device side are determined by the entity that trains the AI / ML model, which can be called the AI / ML entity. The AI / ML entity obtains the data from the terminal device required to train the AI / ML model applied to the terminal device side through higher-level network devices.

[0082] Before an AI / ML entity begins acquiring data from a terminal device through higher-level network devices, the terminal device needs to collect the data required for the AI / ML model that the AI / ML entity wants to train. In other words, before the higher-level network devices trigger the terminal device to transmit or transfer the data required for training the AI / ML model, the terminal device needs to have the data required for training the AI / ML model, namely the first data or a portion of the first data.

[0083] The data can be data that already exists in the terminal device, or data that the terminal device collects according to instructions, such as data collected according to the instructions in the first information.

[0084] In this embodiment of the application, the first information includes at least one of the following:

[0085] Instruction information instructing the terminal device to collect the first data;

[0086] The type or attributes of the terminal device that collects this first data; for example, all terminal devices, the hardware type or attributes of the terminal device, the software type or attributes of the terminal device, the manufacturer of the terminal device, the version of the terminal device, and the capabilities of the terminal device.

[0087] The purpose of collecting the first data; for example, indicating that the purpose of collecting the data is to train the first model, or the purpose of collecting the data is to initially train the first model and / or retrain the first model, or the purpose of collecting the data is to supervise the first model;

[0088] Configuration information for sending the first data, the configuration information including at least one of the conditions for sending the first data and the configuration for sending the first data;

[0089] The relevant information of the first data, for example, includes relevant information about the input of the first model and / or relevant information about the output of the first model; or, for example, the relevant information of the first data indicates relevant information about the input of the first model and relevant information about the output of the first model, respectively.

[0090] The first association indicator may be indicated by at least one of the following: identifier, index, pointer, sequence number, number.

[0091] In this embodiment of the application, the first information is further used to indicate that at least two of the following pieces of information are associated:

[0092] The first association indication;

[0093] The relevant information of this first data;

[0094] The relevant information input to the first model;

[0095] The relevant information output by the first model;

[0096] The first model.

[0097] For example, the first model is an AI / ML model for beam management, and the first information includes at least one of the following:

[0098] The type or attributes of the terminal device that collects the first data, such as a terminal device that can support multiple beam management;

[0099] The relevant information of the first data includes, for example, the number of beams in each group of one or more groups (e.g., M groups), and / or the identifier or index or pointer of each beam, and / or the correlation information between beams within the same group, and / or the correlation information between beams in different groups, and / or the content indicating the first data, such as at least one of the beam's RSRP, RSRQ, SINR, and / or the optimal one or more (e.g. K) beams in each group;

[0100] Information related to the first model input, such as the number of beams in the first set of beams used for the first model input, and / or the identifier, index, or pointer of each beam in the first set of beams;

[0101] The relevant information output by the first model, such as the number of beams in the second set of beams that the first model is to predict, and / or the identifier, index or pointer of each beam in the second set of beams;

[0102] The correlation requirement or threshold between the first set of beams input to the first model and the second set of beams to be predicted by the first model. For example, the second set of beams may be the same as the first set of beams, or may contain the first set of beams, or may be a subset of the first set of beams, or may not contain the first set of beams.

[0103] For example, the first model is a predictive AI / ML model for mobility management, and the first information includes at least one of the following:

[0104] The type or attributes of the terminal device that collects the first data, such as a terminal device that supports measurements on two or more frequencies or bands, provides data collection for inter-frequency prediction models; or a terminal device whose channel environment changes are less than a specified threshold, provides data collection for predicting future measurement results;

[0105] Information related to the first data, such as one or more frequencies, and / or measurement object information, such as SSB, CSI-RS, SRS, and / or information indicating the content of the first data, such as at least one of layer 1 and / or layer 3 RSRP, RSRQ, SINR of the measured object, and / or measurement thresholds, and / or information related to the triggering event, such as the triggered event and the time information of the event.

[0106] The relevant information of the first model input includes, for example, the frequency or frequency band in which the input data is collected, and / or the measurement object information;

[0107] The relevant information output by the first model includes, for example, the frequency or band in which the output data was collected, and / or information related to the triggering event.

[0108] In this embodiment of the application, the first configuration includes at least one of the following:

[0109] Resource configuration, which is the resource configuration required by the terminal device to collect the first data;

[0110] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0111] A data collection instruction, which instructs the first configuration to collect the first data;

[0112] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0113] As described above, in step 501, the terminal device receives first information sent by the first network device or the second network device. For example, the first network device sends the first information to the terminal device, or the second network device forwards the first information from the first network device to the terminal device.

[0114] In this embodiment of the application, the first configuration is a configuration received by the terminal device from the second network device, or it is a configuration that the terminal device already has.

[0115] For example, the method further includes: the terminal device receiving the first configuration from the second network device.

[0116] In this embodiment of the application, when the terminal device receives the first configuration from the second network device, the method further includes:

[0117] The terminal device sends configuration request information to the second network device. The configuration request information is used to request the first configuration. For example, the configuration request information is related to the first information, and / or the configuration request information includes all or part of the first information.

[0118] For example, the method further includes: the terminal device using an existing configuration as the first configuration based on the first information. For example, the terminal device determines configuration requirement information based on the first information; the terminal device determines the first configuration among the existing configurations that matches the configuration requirement information based on the configuration requirement information.

[0119] The methods of this application embodiment will be specifically described below according to the different ways in which the terminal device receives the first information from the first network device or the second network device.

[0120] Method 1: The terminal device receives first information from the first network device;

[0121] In this scenario, the first network device sends first information to the second network device, the second network device sends third information to the relevant terminal device, and / or sends a first configuration to the terminal device based on the first information, and / or instructs the terminal device to collect data. The third information includes all or part of the information in the first information.

[0122] The first configuration is the configuration determined by the second network device based on the first information, including at least one of the following:

[0123] Resource configuration required for terminal devices to collect data.

[0124] The parameters on which the terminal device collects data are configured.

[0125] The first configuration is used for data collection instructions for the first model.

[0126] Reporting configuration for terminal device reporting results.

[0127] The first configuration includes a reporting configuration, instructing the terminal device that measurement and / or estimation results need to be reported to the second network device during the data collection process. The content to be reported and / or the resources used are indicated by the reporting configuration. Based on the received first information, the second network device determines a first configuration for data collection related to the first model. This first configuration does not include a reporting configuration, which instructs the terminal device that the measurement and / or estimation results performed according to the first configuration do not need to be reported to the second network device. This avoids the significant air interface signaling overhead associated with transmitting reports of measurement and / or estimation results during the data collection process.

[0128] For example, the first configuration includes CSI-ReportConfig; LTM-CSI-ReportConfig; CSI-MeasConfig; LTM-CSI-ResourceConfig; MeasConfig; MeasGapConfig; QuantityConfig, etc.

[0129] For example, in a beam management model that predicts a second set of beams using a first set of beams, the first information includes correlation requirement information, and / or quantity information, and / or identification information for the first and second sets of beams; the first configuration includes resource configuration, measurement configuration, and / or reporting configuration for reference signals of the first and second sets of beams that meet the first information requirements.

[0130] For example, in mobility management, a model predicts the results of radio resources on a second frequency based on the relevant measurement results of radio resources on a first frequency. The first information includes information about the first frequency and / or the second frequency, wherein, when the first frequency is not included, the first frequency is the frequency used by the current cell; the first configuration includes a measurement gap for measuring the first frequency and / or the second frequency, and / or conditions for starting the measurement, such as the time of starting the measurement, the event that needs to be met to start the measurement, and / or the relevant configuration of the event triggered by the measurement.

[0131] The terminal device collects data based on the first information and / or first configuration and / or existing resources or parameters indicated by the second network device, including performing relevant measurements or evaluations according to the first configuration and storing the relevant measurement or evaluation results, and / or reporting the relevant measurement or evaluation results to the second network device according to the reporting configuration indicated in the first configuration.

[0132] For example, if the second network device does not indicate the first configuration, or if the first configuration does not include resource and / or parameter configurations, the terminal device uses existing resource and / or parameter configurations to collect and store data that meets the first information requirements. Alternatively, if the terminal device can use existing resource and / or parameter configurations to collect some data that meets the first information requirements, the first configuration indicated by the second network device may only include the resource and / or parameter configurations that need to be added, used to collect the remaining data required by the first information requirements.

[0133] For example, in the measurement of AI / ML data collection for mobility management, the terminal device performs the measurement immediately after receiving the first configuration; or the measurement is performed according to measurement conditions specifically for data collection indicated by the first configuration, which includes a configuration of measurement events specifically for AI / ML data collection; or, if the first configuration does not include start measurement conditions, the measurement is performed using existing measurement conditions for mobility management. When the first configuration includes a reporting configuration, the terminal device also reports the measurement results to a second network device according to the reporting configuration.

[0134] In this embodiment of the application, by applying the above method one, the second network device can perform appropriate configuration and scheduling for the terminal device according to the first information indicated by the first network device, avoid conflicts between data collection and other services of the terminal device, and can completely control the behavior and content of the terminal device collecting data, preventing the leakage of relevant sensitive information of the second network device and possible malicious attacks.

[0135] Method 2: Direct instruction from the first network device or application;

[0136] For example, the first network device instructs the terminal device via a UP channel and / or NAS message.

[0137] Since the first network device directly instructs the first information to the terminal device without forwarding it through the second network device, or in other words, the instruction process of the first information is transparent to the second network device, the second network device cannot proactively provide the terminal device with the first configuration that matches the configuration requirement information. This configuration requirement information is determined by the terminal device based on the first information, and is the resource configuration required for data collection and / or the parameter configuration on which data collection is based.

[0138] For example, a terminal device receives first information directly indicated by a first network device or application, determines configuration requirement information based on the first information, and obtains a first configuration that matches the configuration requirement information.

[0139] The methods for a terminal device to obtain the first configuration may include:

[0140] Method 2-1: The terminal device uses the existing resource configuration and / or parameter configuration as the first configuration, and uses the first configuration to collect data related to the first information.

[0141] For example, the terminal device determines configuration requirement information based on the first information. The terminal device already has a first configuration that matches the configuration requirement information; and, the terminal device reports the measurement or estimation results generated during the data collection process to the second network device according to the reporting configuration in the existing first configuration; or, if the reporting configuration in the existing first configuration indicates that the reporting of the measurement or estimation results generated during the data collection process of the AI / ML model on the terminal device side should be stopped, or if the reporting configuration is not effective, then the terminal device stops the measurement or estimation results performed according to the reporting configuration during the collection process, and resumes reporting the measurement or estimation results using the reporting configuration after the data collection process is completed.

[0142] For example, if the first information indicates that there are two beam sets, and the terminal device determines that it has the corresponding configuration of all beams in the two beam sets, such as CSI-ResourceConfig, it can perform related measurements and estimations on all beams in the two beam sets. Then, based on the existing corresponding configuration of all beams in the two beam sets, the terminal device performs beam measurements and estimations and collects the data indicated in the first information.

[0143] The method further includes, when the terminal device does not have a first configuration that matches the configuration requirement information, or the terminal device only has a first configuration that matches a portion of the configuration requirement information, the terminal device performs at least one of the following processes:

[0144] The terminal device sends configuration requirement information to the second network device, requesting the allocation of a first configuration that matches the configuration requirement information and is missing from the terminal device. This can be done, for example, through UAI, OtherConfig, or new uplink messages for requesting resource configuration and / or parameter configuration, such as RRC or MAC CE. The request includes configuration requirement information determined by the terminal device. For example, the first information indicates that there are two beam sets. The terminal device determines to measure or estimate the corresponding configuration of all beams in the two beam sets. The terminal device only has the configuration of one beam set or part of the beams. The terminal device requests the second network device to allocate the missing beam set or the configuration corresponding to the missing beam. The request includes the identifier of the missing beam set or the missing beam.

[0145] Send a message to the first network device indicating that there is no or missing first configuration, such as insufficient resources, and / or indicating that data collection related to the first information cannot be performed, such as the inability to perform the data collection; and / or the terminal device sends the message via a UP channel and / or a NAS message;

[0146] Wait for the second network device to instruct on the first configuration. After receiving the instruction on the first configuration from the network device, use the first configuration to collect data related to the first information.

[0147] Release the first piece of information and do not perform any data collection related to that first piece of information;

[0148] Method 2-2: The terminal device sends configuration requirement information to the second network device to request a first configuration that matches the configuration requirement information. The request includes at least one of the following:

[0149] The resource information required by this terminal device

[0150] The terminal device requires the following parameter information:

[0151] First association indication,

[0152] Relevant information from the first data,

[0153] The purpose of this request is, for example, for data collection of AI / ML models on the terminal device side.

[0154] The first association indication is also used to associate configuration requirement information and / or the first configuration. The terminal device can send this request via UAI, OtherConfig, or new uplink messages for requesting resource configuration and / or parameter configuration, such as RRC or MAC CE.

[0155] For example, when a terminal device sends configuration request information to a second network device to request the first configuration needed to collect data for training an AI / ML model with beam management, the request content includes information about one or more beams. This beam information may include beam correlation requirements, the number of beams, and specific beam identifiers or indices. For example, methods for indicating the information of one or more beams may include:

[0156] Method 2-2-1: Indicate information about the beams in the first beam set and / or the second beam set, wherein the beams in the first beam set and the beams in the second beam set are indicated respectively, the first beam set is the beam set about the AI / ML model input or that needs to be measured in reality, and the second beam set is the beam set about the AI / ML model output or that needs to be predicted by the AI / ML model;

[0157] Method 2-2-2: Indicates all relevant beam information for all AI / ML models, including information on all beams that are input to the AI / ML model or need to be measured, as well as information on all beams that are output to the AI / ML model or need to be predicted by the AI / ML model. The configuration requirements do not indicate which beams need to be measured or evaluated, or which beams the model should predict, when applying the AI / ML model for inference.

[0158] Method 2-2-3: Simply indicate that the terminal device does not have one or more beam information related to the AI / ML model in the first configuration;

[0159] Method 2-2-4: Indicate that there is no relevant information about a first configuration beam in the first beam set and / or the second beam set, respectively. The first beam set is about the AI / ML model input or the beam set that needs to be measured in reality, and the second beam set is about the AI / ML model output or the beam set that needs to be predicted by the AI / ML model.

[0160] Method 2-2-5: The terminal device of the AI / ML model does not have all the relevant beam information of the first configuration, and does not indicate in the configuration requirement information which beams are related to the AI / ML model input and which beams are related to the AI / ML model output.

[0161] For example, in the data collection of AI / ML models for mobility management, the request sent by the terminal device to the second network device includes at least one of the following:

[0162] Measurement gap

[0163] The frequency of measurement, and / or an indication of the frequency that the model input needs to be actually measured by the terminal device, and / or an indication of the frequency that the model will predict.

[0164] Measures skipped information, such as the percentage of skipped data and the skipped pattern.

[0165] Method 2-3: In addition to indicating the first information to the terminal device, the first network device also instructs the second network device to configure a first configuration for the terminal device that matches the first information.

[0166] For example, a first network device sends all or part of a first message to a second network device. This first message can be sent via NG-AP signaling. Upon receiving the first message, the second network device determines a first configuration matching the first message, sends the first configuration to the terminal device, and / or instructs the terminal device to collect data related to AI / ML models.

[0167] In this embodiment of the application, by applying method 2, the second network device can be prevented from participating in data collection processes or processing related data that are unrelated to it. This reduces the signaling overhead associated with the second network device and also prevents the possible extraction of sensitive user data on the second network device.

[0168] In this embodiment of the application, the method may further include determining the association relationship between information such as the first model, relevant information of the first model input, relevant information of the first model output, and the first configuration.

[0169] The terminal device uses a first model to deduce and obtain output information based on the input information, and performs predictions in the time domain, frequency domain, or spatial domain. The input information used by the terminal device to perform the derivation process and the output information obtained should match the input information and output information used during the training phase of the first model. Therefore, the entities related to the first model need to be able to associate the model with the model's input information and output information, and / or the configuration information corresponding to the input information and output information. That is, the relevant entities should determine (obtain / maintain) this association.

[0170] In this embodiment of the application, the determination of this association relationship can be triggered by the following conditions:

[0171] When preparing to collect data, initiate the process of determining the association; this data collection is used for the initial training of the first model. When the first model is abandoned after training, for example, if the training results do not meet expectations, initiate the process of deleting or releasing the association.

[0172] Once the first model has been successfully trained, the process of determining the association relationship is initiated.

[0173] This association can be established by introducing the aforementioned first association indicator, that is, by using an identifier, index, pointer, number, or sequence number to associate with at least one of the aforementioned associated contents. For example, when initially associating configuration information, in addition to indicating the configuration information, the first association indicator of the configuration information is also indicated. In subsequent use, the associated configuration information can be determined solely through the first association indicator.

[0174] For example, add the corresponding first association indicator to the following configuration, or when providing the following configuration, additionally configure the corresponding first association indicator:

[0175] CSI-ReportConfig; LTM-CSI-ReportConfig; CSI-MeasConfig; LTM-CSI-ResourceConfi-g; MeasConfig; MeasGapConfig; QuantityConfig.

[0176] For example, the relevant entities that determine this association include at least one of the following:

[0177] The execution entity is the entity that performs the derivation of the first model. For example, for the terminal device side model, the entity that performs the derivation of the first model is the terminal device.

[0178] A management entity is an entity that manages the first model. For example, the management entity is a first network device or AI / ML entity that has at least one of the functions of generating, training, and maintaining the first model; or, for example, the management entity is a first network device that is connected to an entity (i.e., an AI / ML entity) that implements at least one of the functions of generating, training, and maintaining the first model, and is able to assist the AI / ML entity in implementing at least one of the functions of generating, training, and maintaining the first model, for example, by initiating data collection and / or transferring the collected data to the AI / ML entity (e.g., a server used for generating, training, and maintaining the first model).

[0179] A configuration entity, which is an entity that provides relevant resources and / or parameter configurations for the process of training and using the first model, such as a second network device, i.e., a network device of the RAN (e.g., a base station), provides the configuration required for collecting first data during the data collection phase of training or supervising the first model; and provides the configuration required for the model input and / or indicates prediction information during the phase of deriving or predicting results using the first model.

[0180] For example, for a terminal device that acts as an execution entity, it determines at least two of the following associations:

[0181] The first model;

[0182] The relevant information input to the first model;

[0183] The relevant information output by the first model;

[0184] This first configuration;

[0185] This is the first piece of information;

[0186] The second network device, wherein the terminal device determines the association relationship, including:

[0187] The terminal device determines the association relationship based on the first association indication in the first information; or...

[0188] The terminal device establishes the association relationship based on at least one of the first information, the first configuration, and the first model.

[0189] For example, for a second network device that is a configuration entity, it determines at least two of the following associations:

[0190] The first model,

[0191] The relevant information input to the first model,

[0192] The relevant information output by the first model,

[0193] This first configuration,

[0194] This configuration requirement information,

[0195] This first piece of information,

[0196] The second network device,

[0197] The logical entity managed by the second network device.

[0198] The second network device establishes the association based on at least one of the first information and the configuration requirement information, or the second network device determines the association based on the first association instruction.

[0199] The first association indication is indicated by the first network device or terminal device, or the second network device determines the first association indication, which is used to associate at least one of the first model, the first information, the configuration requirement information, the first configuration, the second network device, and the logical entity managed by the second network device.

[0200] For example, the second network device determines the first association instruction based on the second association instruction, and the content to be associated by the second association instruction includes at least one of the first model, the first information, and the configuration requirement information.

[0201] For example, the second association indication is an association indication received by the second network device from the first network device or the terminal device;

[0202] For example, the first association instruction is also used to associate at least one of the second association instruction and the content associated with the second association instruction.

[0203] The following section explains in detail the methods for determining this association relationship for different entities.

[0204] Method 1: Determined based on management entity configuration. The management entity (e.g., the first network device) assigns a first association instruction, and simultaneously instructs the execution entity (e.g., the terminal device) and / or the configuration entity (e.g., the second network device) on the assigned first association instruction and the content to be associated with the first association instruction. The execution entity and / or configuration entity can then determine the content associated with the first association based on the assigned first association instruction.

[0205] The configuration entity determines the first configuration related to the first associated content, such as resource configuration and / or parameter configuration and / or report configuration, according to the instructions of the management entity. The determined first configuration is also used as the content to be associated with the first instruction. That is, the determined first configuration is associated with the first association instruction indicated by the management entity and the content already associated with the first association instruction. For example, the configuration entity can determine the corresponding first configuration and the content associated with the first association instruction indicated by the management entity based on the first association instruction.

[0206] Method 2: Determined by the execution entity. The management entity instructs the execution entity to collect data and first information related to the data to be collected. The collected data is used to train the first model. The execution entity determines the configuration requirement information related to the first information based on the first information instructed by the management entity. The execution entity assigns a first association instruction to associate the following first association content: the first information instructed by the management entity and / or the determined configuration requirement information. The execution entity sends the assigned first association instruction and / or the associated first association content to the configuration entity and / or the management entity. For example, a terminal device sends information to a second network device, including the assigned first association instruction and the configuration requirement information associated with the instruction, and / or the terminal device sends or provides feedback on the assigned first association instruction to the first network device.

[0207] Method 2 further includes a configuration entity determining a relevant first configuration based on configuration requirement information received from the execution entity, and also using the first configuration as the first association content associated with the first association indication, which is a first association indication received from the execution entity along with the configuration requirement information.

[0208] Method 3: Determined by a configuration entity, such as a second network device.

[0209] Method 3-1: The configuration entity receives relevant information about the model from the management entity or execution entity, such as first information related to data collection and / or configuration requirement information related to that first information. The configuration entity determines a first configuration based on the relevant information about the model. The configuration entity assigns a first association instruction, using the first configuration and / or the relevant information about the model as the content associated with the first association instruction, and sends or provides feedback on the assigned first association instruction and / or the content associated with the first association instruction to the configuration entity and / or the management entity.

[0210] The method also includes the configuration entity receiving model-related information that is indicated by the management entity, the configuration entity sending or feeding back the assigned first association instruction and / or the content associated with the first association instruction to the management entity, and the management entity instructing the first association instruction and / or the content associated with the first association instruction to the execution entity;

[0211] The method also includes the configuration entity receiving model-related information that is instructed by the execution entity, the configuration entity sending or feeding back the assigned first association instruction and / or the content associated with the first association instruction to the execution entity, and the execution entity reporting the first association instruction and / or the content associated with the first association instruction to the management entity.

[0212] Method 3-2: The configuration entity receives relevant information about the model indicated by the management entity or the execution entity, such as first information related to data collection and / or configuration requirement information related to the first information, which also includes a second association instruction, such as relevant information assigned by the management entity or the execution entity to associate with the first model indicated by the management entity or the execution entity, such as the identifier, index, pointer, sequence number or number of the first model.

[0213] The configuration entity determines a first configuration based on the relevant information of the received first model. The configuration entity assigns a third association indication, which is used to associate the first configuration with the configuration entity or a logical entity managed by the configuration entity, such as a base station or a base station component or a cell managed by the base station. For example, the third association indication includes identification information of a second network device, such as a base station identifier (Gnb-id), a CU identifier (CU-ID), a DU identifier (DU-ID / CU-DU-ID), or a cell identifier (cell-id).

[0214] The configuration entity determines the first association instruction based on the second association instruction and the third association instruction. For example, the first association instruction consists of the second association instruction and the third association instruction, or the first association instruction includes the second association instruction and the third association instruction.

[0215] The configuration entity uses the relevant information of the first configuration and / or the first model as the first associated content associated with the first association instruction, and sends or feeds back the assigned first association instruction and / or the content associated with the first association instruction to the configuration entity and / or the management entity, in the same way as method 3-1.

[0216] In method 3, the first association indication is determined by the configuration entity, which may include information or attributes of the configuration entity, such that the first model may be a specific model of the configuration entity or a logical entity managed by the configuration entity, such as a model of a specific base station or a specific cell.

[0217] In this embodiment, the association can also be determined through configuration instructions. For example, indicating or configuring the first associated content in the same message or the same structure or information unit within the message indicates that the first associated content is associated. For example, the content in the first information configured by the management entity is associated, and the management entity associates the content in the first information with the model related to the first information; the execution entity associates the first information and / or related configuration requirement information and / or related first configuration; the configuration entity associates the first information and / or related configuration requirement information and / or related first configuration. The determination process is similar to the method described above for determining the association through the first association instruction.

[0218] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0219] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0220] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0221] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0222] Second aspect of the embodiments

[0223] This application provides a data collection instruction method, which is applied to a first network device and corresponds to the data collection method applied to a terminal device described in the first aspect embodiment. The same or corresponding content can be referred to the description in the first aspect embodiment.

[0224] This method is applied to a first network device, such as the first network device 101 in Figure 1.

[0225] Figure 6 is a schematic diagram of a data collection instruction method according to an embodiment of this application. As shown in Figure 6, the method includes:

[0226] 601: The first network device sends first information to the terminal device and / or the second network device.

[0227] In this embodiment of the application, the first information is at least related to the first data collected by the terminal device. The first data is used for at least one of training a first model and supervising the first model. The first model is an AI / ML model applied to the terminal device side.

[0228] In this embodiment of the application, the first information includes at least one of the following:

[0229] Instruction information instructing the terminal device to collect the first data;

[0230] The type or attributes of the terminal device that collects this first data;

[0231] The purpose of collecting this first data;

[0232] Configuration information for sending the first data, the configuration information including at least one of the conditions for sending the first data and the configuration for sending the first data;

[0233] The relevant information of this first data;

[0234] First related indication.

[0235] In this embodiment of the application, the relevant information of the first data includes at least one of the following:

[0236] The relevant information input to the first model;

[0237] The relevant information output by the first model;

[0238] In this embodiment of the application, the relevant information of the first data respectively indicates relevant information about the input of the first model and relevant information about the output of the first model.

[0239] In this embodiment of the application, the first association indication is indicated by at least one of the following: identifier, index, pointer, serial number, number.

[0240] In this embodiment of the application, the first information is further used to indicate that at least two of the following pieces of information are associated:

[0241] The first association indication;

[0242] The relevant information of this first data;

[0243] The relevant information input to the first model;

[0244] The relevant information output by the first model;

[0245] The first model.

[0246] In this embodiment of the application, the first information includes the first association indication.

[0247] The first association indication is used by the terminal device to determine at least two of the following association relationships:

[0248] The first model;

[0249] The relevant information input to the first model;

[0250] The relevant information output by the first model;

[0251] The first configuration is matched with the first information;

[0252] This is the first piece of information;

[0253] The second network device.

[0254] In this embodiment of the application, the method further includes:

[0255] The first network device sends a second message to the second network device.

[0256] The second information is used at least to instruct the second network device to send a first configuration matching the first information to the terminal device, the first configuration including at least one of the following:

[0257] Resource configuration, which is the resource configuration required by the terminal device to collect the first data;

[0258] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0259] A data collection instruction, which instructs the first configuration to collect the first data;

[0260] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0261] In the embodiments of this application, the specific implementation of the above operations can be referred to the relevant descriptions in the embodiments of the first aspect, which will not be repeated here.

[0262] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0263] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0264] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0265] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0266] Third aspect of the embodiments

[0267] This application provides a data collection instruction method, which is applied to a second network device and corresponds to the data collection method applied to a terminal device described in the first aspect embodiment. The same or corresponding content can be referred to the description in the first aspect embodiment.

[0268] This method is applied to a second network device, such as the second network device 102 in Figure 1.

[0269] Figure 7 is a schematic diagram of a data collection instruction method according to an embodiment of this application. As shown in Figure 7, the method includes:

[0270] 701: The second network device receives first information from the first network device, or the second network device receives first information and / or configuration requirement information from the terminal device;

[0271] 702: The second network device sends a first configuration that matches the first information or the configuration requirement information to the terminal device;

[0272] In this embodiment of the application, the first information or the configuration requirement information is at least related to the first data collected by the terminal device. The first data is used for at least one of training the first model and supervising the first model. The first model is an AI / ML model applied to the terminal device side.

[0273] In this embodiment of the application, the method further includes:

[0274] The second network device sends a third message to the terminal device.

[0275] The third information includes all or part of the information in the first information, and / or the third information includes at least the first association indication.

[0276] In this embodiment of the application, the first information includes at least one of the following:

[0277] Instruction information instructing the terminal device to collect the first data;

[0278] The type or attributes of the terminal device that collects this first data;

[0279] The purpose of collecting this first data;

[0280] Configuration information for sending the first data, the configuration information including at least one of the conditions for sending the first data and the configuration for sending the first data;

[0281] The relevant information of this first data;

[0282] First related indication.

[0283] In this embodiment of the application, the relevant information of the first data includes at least one of the following:

[0284] The relevant information input to the first model;

[0285] The relevant information output by the first model;

[0286] In this embodiment of the application, the relevant information of the first data respectively indicates relevant information about the input of the first model and relevant information about the output of the first model.

[0287] In this embodiment of the application, the first association indication is indicated by at least one of the following: identifier, index, pointer, serial number, number.

[0288] In this embodiment of the application, the first information is further used to indicate that at least two of the following pieces of information are associated:

[0289] The first association indication;

[0290] The relevant information of this first data;

[0291] The relevant information input to the first model;

[0292] The relevant information output by the first model;

[0293] The first model.

[0294] In this embodiment of the application, the first information includes a first association indication.

[0295] The first association indication is used by the terminal device to determine at least two of the following association relationships:

[0296] The first model;

[0297] The relevant information input to the first model;

[0298] The relevant information output by the first model;

[0299] This first configuration;

[0300] This is the first piece of information;

[0301] The second network device.

[0302] In this embodiment of the application, the first configuration includes at least one of the following:

[0303] Resource configuration, which is the resource configuration required for the terminal device to collect the first data;

[0304] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0305] A data collection instruction, which instructs the first configuration to collect the first data;

[0306] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0307] In this embodiment of the application, the method further includes:

[0308] The second network device determines at least two of the following associations:

[0309] The first model,

[0310] The relevant information input to the first model,

[0311] The relevant information output by the first model,

[0312] This first configuration,

[0313] This configuration requirement information,

[0314] This first piece of information,

[0315] The second network device,

[0316] The logical entity managed by the second network device.

[0317] The second network device establishes the association based on at least one of the first information and the configuration requirement information, or the second network device determines the association based on the first association instruction.

[0318] The first association indication is indicated by the first network device or terminal device, or

[0319] The second network device determines the first association indication.

[0320] The first association indication is used to associate at least one of the first model, the first information, the configuration requirement information, the first configuration, the second network device, and the logical entity managed by the second network device.

[0321] In this embodiment of the application, the second network device determines the first association indication, including:

[0322] The second network device determines the first association instruction based on the second association instruction.

[0323] The second association instruction is used to associate at least one of the first model, the first information, and the configuration requirement information.

[0324] In this embodiment, the second association indication is an association indication received by the second network device from the first network device or the terminal device.

[0325] In this embodiment of the application, the first association instruction is also used to associate at least one of the second association instruction and the content associated with the second association instruction.

[0326] In the embodiments of this application, the specific implementation of the above operations can be referred to the relevant descriptions in the embodiments of the first aspect, which will not be repeated here.

[0327] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0328] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0329] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0330] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0331] Fourth aspect of the embodiment

[0332] This application provides a data collection device installed on a terminal device. Since the principle by which this device solves the problem is similar to the method in the first aspect embodiment, its specific implementation can refer to the implementation of the method described in the first aspect embodiment; the same or related details will not be repeated.

[0333] Figure 8 is a schematic diagram of a data collection device according to an embodiment of this application. As shown in Figure 8, the data collection device 800 includes:

[0334] The first receiving unit 801 receives first information sent by the first network device or the second network device;

[0335] Collection unit 802, which uses a first configuration matching the first information, collects first data.

[0336] In this embodiment of the application, the first data is used to train at least one of the first model and to supervise the first model, wherein the first model is an AI / ML model applied to the terminal device side.

[0337] In this embodiment of the application, the first information includes at least one of the following:

[0338] Instruction information instructing the terminal device to collect the first data;

[0339] The type or attributes of the terminal device that collects this first data;

[0340] The purpose of collecting this first data;

[0341] Configuration information for sending the first data, the configuration information including at least one of the conditions for sending the first data and the configuration for sending the first data;

[0342] The relevant information of this first data;

[0343] First related indication.

[0344] In this embodiment of the application, the relevant information of the first data includes at least one of the following:

[0345] The relevant information input to the first model;

[0346] The relevant information output by the first model;

[0347] In this embodiment of the application, the relevant information of the first data respectively indicates relevant information about the input of the first model and relevant information about the output of the first model.

[0348] In this embodiment of the application, the first association indication is indicated by at least one of the following: identifier, index, pointer, serial number, number.

[0349] In this embodiment of the application, the first information is further used to indicate that at least two of the following pieces of information are associated:

[0350] The first association indication;

[0351] The relevant information of this first data;

[0352] The relevant information input to the first model;

[0353] The relevant information output by the first model;

[0354] The first model.

[0355] In this embodiment of the application, the first configuration includes at least one of the following:

[0356] Resource configuration, which is the resource configuration required for the terminal device to collect the first data;

[0357] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0358] A data collection instruction, which instructs the first configuration to collect the first data;

[0359] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0360] In this embodiment of the application, the terminal device receives the first configuration from the second network device; and / or the terminal device uses an existing configuration as the first configuration based on the first information.

[0361] In this embodiment of the application, when the terminal device receives the first configuration from the second network device, the apparatus further includes:

[0362] The first sending unit sends configuration request information to the second network device, the configuration request information being used to request the first configuration.

[0363] The configuration requirement information is related to the first information, and / or the configuration requirement information includes all or part of the first information.

[0364] In this embodiment of the application, the terminal device uses the existing configuration as the first configuration based on the first information, which includes: the terminal device determining configuration requirement information based on the first information; and the terminal device determining the first configuration that matches the configuration requirement information among the existing configurations based on the configuration requirement information.

[0365] In this embodiment of the application, the device further includes:

[0366] The first determining unit determines at least two of the following relationships:

[0367] The first model;

[0368] The relevant information input to the first model;

[0369] The relevant information output by the first model;

[0370] This first configuration;

[0371] This is the first piece of information;

[0372] The second network device, wherein the terminal device determines the association relationship, including:

[0373] The terminal device determines the association relationship based on the first association indication in the first information; or...

[0374] The terminal device establishes the association relationship based on at least one of the first information, the first configuration, and the first model.

[0375] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0376] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0377] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0378] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0379] Fifth aspect of the embodiment

[0380] This application provides a data collection instruction device applied to a first network device. Since the principle by which this device solves the problem is similar to the method in the second aspect of the embodiment, its specific implementation can refer to the implementation of the method described in the second aspect of the embodiment; the same or related parts will not be repeated.

[0381] Figure 9 is a schematic diagram of a data collection instruction device according to an embodiment of this application. As shown in Figure 9, the data collection instruction device 900 includes:

[0382] The second sending unit 901 sends the first information to the terminal device and / or the second network device.

[0383] In this embodiment of the application, the first information is at least related to the first data collected by the terminal device. The first data is used for at least one of training a first model and supervising the first model. The first model is an AI / ML model applied to the terminal device side.

[0384] In this embodiment of the application, the first information includes at least one of the following:

[0385] Instruction information instructing the terminal device to collect the first data;

[0386] The type or attributes of the terminal device that collects this first data;

[0387] The purpose of collecting this first data;

[0388] The configuration information for sending this first data.

[0389] The configuration information includes at least one of the conditions for sending the first data and the configuration for sending the first data;

[0390] The relevant information of this first data;

[0391] First related indication.

[0392] In this embodiment of the application, the relevant information of the first data includes at least one of the following:

[0393] The relevant information input to the first model;

[0394] The relevant information output by the first model;

[0395] In this embodiment of the application, the relevant information of the first data respectively indicates relevant information about the input of the first model and relevant information about the output of the first model.

[0396] In this embodiment of the application, the first association indication is indicated by at least one of the following: identifier, index, pointer, serial number, number.

[0397] In this embodiment of the application, the first information is further used to indicate that at least two of the following pieces of information are associated:

[0398] The first association indication;

[0399] The relevant information of this first data;

[0400] The relevant information input to the first model;

[0401] The relevant information output by the first model;

[0402] The first model.

[0403] In this embodiment of the application, the first information includes the first association indication, which is used by the terminal device to determine at least two of the following association relationships:

[0404] The first model;

[0405] The relevant information input to the first model;

[0406] The relevant information output by the first model;

[0407] The first configuration is matched with the first information;

[0408] This is the first piece of information;

[0409] The second network device.

[0410] In this embodiment of the application, the second sending unit 901 also sends second information to the second network device.

[0411] The second information is used at least to instruct the second network device to send a first configuration matching the first information to the terminal device, the first configuration including at least one of the following:

[0412] Resource configuration, which is the resource configuration required for the terminal device to collect the first data;

[0413] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0414] A data collection instruction, which instructs the first configuration to collect the first data;

[0415] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0416] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0417] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0418] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0419] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0420] Implementation of the sixth aspect

[0421] This application provides a data collection indication device applied to a second network device. Since the principle by which this device solves the problem is similar to the method in the third aspect embodiment, its specific implementation can refer to the implementation of the method described in the third aspect embodiment; the same or related details will not be repeated.

[0422] Figure 10 is a schematic diagram of a data collection instruction device according to an embodiment of this application. As shown in Figure 10, the data collection instruction device 1000 includes:

[0423] The second receiving unit 1001 receives first information from the first network device, or receives first information and / or configuration requirement information from the terminal device;

[0424] The third sending unit 1002 sends a first configuration that matches the first information or the configuration requirement information to the terminal device.

[0425] In this embodiment of the application, the first information or the configuration requirement information is at least related to the first data collected by the terminal device. The first data is used for at least one of training the first model and supervising the first model. The first model is an AI / ML model applied to the terminal device side.

[0426] In this embodiment of the application, the third sending unit also sends third information to the terminal device.

[0427] The third information includes all or part of the information in the first information, and / or the third information includes at least the first association indication.

[0428] In this embodiment of the application, the first information includes at least one of the following:

[0429] Instruction information instructing the terminal device to collect the first data;

[0430] The type or attributes of the terminal device that collects this first data;

[0431] The purpose of collecting this first data;

[0432] Configuration information for sending the first data, the configuration information including at least one of the conditions for sending the first data and the configuration for sending the first data;

[0433] The relevant information of this first data;

[0434] First related indication.

[0435] In this embodiment of the application, the relevant information of the first data includes at least one of the following:

[0436] The relevant information input to the first model;

[0437] The relevant information output by the first model;

[0438] And / or,

[0439] The relevant information in the first data indicates information about the input of the first model and information about the output of the first model, respectively.

[0440] And / or,

[0441] The first association indication is indicated by at least one of the following: identifier, index, pointer, sequence number, number.

[0442] And / or,

[0443] This first piece of information is also used to indicate that at least two of the following pieces of information are related:

[0444] The first association indication;

[0445] The relevant information of this first data;

[0446] The relevant information input to the first model;

[0447] The relevant information output by the first model;

[0448] The first model.

[0449] In this embodiment of the application, the first information includes a first association indication.

[0450] The first association indication is used by the terminal device to determine at least two of the following association relationships:

[0451] The first model;

[0452] The relevant information input to the first model;

[0453] The relevant information output by the first model;

[0454] This first configuration;

[0455] This is the first piece of information;

[0456] The second network device.

[0457] In this embodiment of the application, the first configuration includes at least one of the following:

[0458] Resource configuration, which is the resource configuration required by the terminal device to collect the first data;

[0459] The parameter configuration is the parameter configuration on which the terminal device collects the first data;

[0460] A data collection instruction, which instructs the first configuration to collect the first data;

[0461] The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of measurements and / or evaluations performed by the terminal device based on the resource configuration and / or the parameter configuration.

[0462] In this embodiment of the application, the device further includes:

[0463] The second determining unit determines the relationship between at least two of the following:

[0464] The first model,

[0465] The relevant information input to the first model,

[0466] The relevant information output by the first model,

[0467] This first configuration,

[0468] This configuration requirement information,

[0469] This first piece of information,

[0470] The second network device,

[0471] The logical entity managed by the second network device.

[0472] The second network device establishes the association based on at least one of the first information and the configuration requirement information, or the second network device determines the association based on the first association instruction.

[0473] The first association indication is indicated by the first network device or terminal device, or

[0474] The second determining unit is the first associated indication.

[0475] The first association indication is used to associate at least one of the first model, the first information, the configuration requirement information, the first configuration, the second network device, and the logical entity managed by the second network device.

[0476] In this embodiment of the application, the second determining unit determines the first association indication based on the second association indication.

[0477] The second association instruction is used to associate at least one of the first model, the first information, and the configuration requirement information.

[0478] In this embodiment of the application, the second association indication is an association indication received by the second network device from the first network device or the terminal device, and / or,

[0479] The first association instruction is also used to associate at least one of the second association instruction and the content associated with the second association instruction.

[0480] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0481] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0482] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0483] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0484] Seventh aspect of the embodiment

[0485] This application provides a terminal device that includes a data collection apparatus according to an embodiment of the fourth aspect.

[0486] Figure 11 is a schematic block diagram of the system configuration of a terminal device according to an embodiment of this application. As shown in Figure 11, the terminal device 1100 may include a processor 1110 and a memory 1120; the memory 1120 is coupled to the processor 1110. It is worth noting that this figure is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunications functions or other functions.

[0487] In one embodiment, the functionality of the data collection device can be integrated into the processor 1110.

[0488] The processor 1110 is configured to: receive first information sent by a first network device or a second network device; the terminal device collects first data using a first configuration matching the first information, the first data being used to train at least one of a first model and to supervise the first model, the first model being an AI / ML model applied to the terminal device side.

[0489] In another embodiment, the data collection device can be configured separately from the processor 1110. For example, the data collection device can be configured as a chip connected to the processor 1110, and the functions of the data collection device can be realized through the control of the processor 1110.

[0490] As shown in Figure 11, the terminal device 1100 may further include: a communication module 1130, an input unit 1140, a display 1150, and a power supply 1160. It is worth noting that the terminal device 1100 does not necessarily include all the components shown in Figure 11; furthermore, the terminal device 1100 may also include components not shown in Figure 11, which can be found in related technologies.

[0491] As shown in Figure 11, the processor 1110, sometimes also referred to as a controller or operation control, may include a microprocessor or other processor device and / or logic device. The processor 1110 receives input and controls the operation of various components of the terminal device 1100.

[0492] The memory 1120 may be, for example, one or more of a cache, flash memory, hard drive, removable medium, volatile memory, non-volatile memory, or other suitable devices. It can store various types of data, and also stores programs for executing related information. The processor 1110 can execute the program stored in the memory 1120 to perform information storage or processing, etc. The functions of other components are similar to those in existing systems and will not be described further here. The components of the terminal device 1100 can be implemented using dedicated hardware, firmware, software, or a combination thereof without departing from the scope of the invention.

[0493] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0494] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0495] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0496] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0497] Eighth aspect of the embodiment

[0498] This application provides a network device that includes a data collection instruction device according to the embodiments of the fifth or sixth aspect. That is, the network device is a first network device or a second network device.

[0499] Figure 12 is a schematic block diagram of the system configuration of a network device according to an embodiment of this application. As shown in Figure 12, the network device 1200 may include a processor 1210 and a memory 1220; the memory 1220 is coupled to the processor 1210. The memory 1220 can store various data; in addition, it also stores an information processing program 1230, and executes the program 1230 under the control of the processor 1210 to receive various information sent by terminal devices and send various information to terminal devices.

[0500] In one embodiment, the function of the data collection indication device can be integrated into the processor 1210.

[0501] For the first network device, the processor 1210 can be configured to: send first information to a terminal device and / or a second network device, the first information being at least related to first data collected by the terminal device, the first data being used for training a first model and supervising at least one of the first model, the first model being an AI / ML model applied to the terminal device side.

[0502] For the second network device, the processor 1210 can be configured to: the second network device receive first information from the first network device, or the second network device receives first information and / or configuration requirement information from the terminal device; the second network device sends a first configuration matching the first information or the configuration requirement information to the terminal device; the first information or the configuration requirement information is at least related to first data collected by the terminal device, the first data being used for at least one of training a first model and supervising the first model, the first model being an AI / ML model applied to the terminal device side.

[0503] In another embodiment, the data collection indicator can be configured separately from the processor 1210. For example, the data collection indicator can be configured as a chip connected to the processor 1210, and the function of the data collection indicator can be realized through the control of the processor 1210.

[0504] In addition, as shown in Figure 12, network device 1200 may also include a transceiver 1240 and an antenna 1250, etc.; the functions of the above components are similar to those in the prior art, and will not be described in detail here. It is worth noting that network device 1200 does not necessarily include all the components shown in Figure 12; furthermore, network device 1200 may also include components not shown in Figure 12, which can be referred to in the prior art.

[0505] As can be seen from the above embodiments, the terminal device receives first information sent by a first network device (e.g., a higher-level network device, such as the core network, OAM, etc.) or as a second network device (e.g., a RAN network device), and uses a first configuration matching the first information to collect first data for training and / or supervising AI / ML models. Thus, the terminal device can flexibly use control plane signaling and user plane channels to initiate the data collection process.

[0506] When the RAN network device needs to be aware of the data collection behavior of the terminal device, the higher-level network device can notify the RAN network device to make relevant configurations through signaling. The detailed configuration and related data collection actions are controlled by the RAN network device, thereby preventing the terminal device from reporting some sensitive information related to the RAN network device to third-party applications or servers without permission, and reducing security risks.

[0507] For AI / ML model training that collects sensitive information from terminal devices, the relevant information of the AI / ML model can be indicated through the user plane channel to prevent RAN network devices from obtaining the relevant information and harming the interests of the model provider.

[0508] In addition, this application also provides a method for determining the relationship between AI / ML model-related information and the configuration information required by the AI / ML model during its lifecycle, so that all entities involved in the training and use of the AI / ML model can jointly ensure the consistency of the AI / ML model training and use stages based on the determined relationship.

[0509] Ninth aspect of the embodiment

[0510] This application provides a communication system including a terminal device according to the seventh aspect embodiment and / or a first network device and / or a second network device according to the eighth aspect embodiment. Specific details can be found in the descriptions of the seventh and eighth aspect embodiments.

[0511] For example, the structure of the communication system can be seen in FIG1. ​​As shown in FIG1, the communication system 100 includes a first network device 101, a second network device 102 and a terminal device 103. The terminal device 103 may be the same as the terminal device described in the seventh aspect embodiment, and / or the first network device 101 and the second network device 102 may be the same as the first network device and the second network device described in the eighth aspect embodiment. Repeated content will not be described again.

[0512] Tenth aspect embodiment

[0513] This application provides a data collection method, which is applied to a terminal device, such as terminal device 103 in FIG1.

[0514] Figure 13 is a schematic diagram of a data collection method according to an embodiment of this application. As shown in Figure 13, the method includes:

[0515] 1301: The terminal device receives an operation command from the application on the terminal device;

[0516] 1302: The terminal device uses a first configuration matching the operation command to collect first data.

[0517] In this embodiment of the application, the first data is used to train at least one of the first model and to supervise the first model, wherein the first model is an AI / ML model applied to the terminal device side.

[0518] In this embodiment of the application, the method further includes:

[0519] The terminal device receives the first configuration from the second network device; and / or

[0520] According to the operation instruction, the terminal device uses the existing configuration as the first configuration.

[0521] In this embodiment of the application, when the terminal device receives the first configuration from the second network device, the method further includes:

[0522] The terminal device sends configuration request information to the second network device, which is used to request the first configuration.

[0523] In this embodiment of the application, the method further includes:

[0524] The terminal device determines at least two of the following associations:

[0525] The first model;

[0526] The relevant information input to the first model;

[0527] The relevant information output by the first model;

[0528] This first configuration;

[0529] The operation instructions in this application;

[0530] The second network device.

[0531] For content that is the same as or similar to the embodiments of the first aspect, please refer to the description in the embodiments of the first aspect, which will not be repeated here.

[0532] The apparatus and methods described above in this application can be implemented in hardware or in combination with software. This application relates to a computer-readable program that, when executed by a logic component, enables the logic component to implement the apparatus or components described above, or to implement the various methods or steps described above. Logic components include, for example, field-programmable logic devices (FPGAs), microprocessors, and processors used in computers. This application also relates to storage media for storing the above programs, such as hard disks, magnetic disks, optical disks, DVDs, and flash memory.

[0533] The methods / apparatus described in conjunction with the embodiments of this application can be directly embodied in hardware, software modules executed by a processor, or a combination of both. For example, one or more and / or one or more combinations of functional block diagrams shown in FIG8 can correspond to various software modules in a computer program flow, or to various hardware modules. These software modules can correspond to the various steps shown in FIG5, respectively. These hardware modules can, for example, be implemented by embedding these software modules using a Field Programmable Gate Array (FPGA).

[0534] The software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. A storage medium can be coupled to the processor, enabling the processor to read information from and write information to the storage medium; or the storage medium can be an integral part of the processor. The processor and storage medium can reside in an ASIC. The software module can be stored in the memory of a mobile terminal or in a memory card that can be inserted into the mobile terminal. For example, if the device (such as a mobile terminal) uses a high-capacity MEGA-SIM card or a high-capacity flash memory device, the software module can be stored in the MEGA-SIM card or the high-capacity flash memory device.

[0535] One or more and / or one or more combinations of functional blocks described in Figure 8 can be implemented as a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, or any suitable combination thereof for performing the functions described herein. One or more and / or one or more combinations of functional blocks described in Figure 8 can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

[0536] The present application has been described above with reference to specific embodiments. However, those skilled in the art should understand that these descriptions are exemplary and not intended to limit the scope of protection of the present application. Those skilled in the art can make various modifications and variations to the present application based on its spirit and principles, and these modifications and variations are also within the scope of the present application.

[0537] According to the various embodiments disclosed in this application, the following notes are also disclosed:

[0538] 1. A data collection method, the method being applied to a terminal device, the method comprising:

[0539] The terminal device receives operation instructions from the application on the terminal device;

[0540] The terminal device uses a first configuration matching the operation command to collect first data.

[0541] The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

[0542] 2. The method according to Appendix 1, wherein the method further comprises:

[0543] The terminal device receives the first configuration from the second network device; and / or

[0544] The terminal device uses the existing configuration as the first configuration according to the operation instruction.

[0545] 3. The method according to Appendix 2, wherein, for the case where the terminal device receives the first configuration from the second network device, the method further includes:

[0546] The terminal device sends configuration request information to the second network device, and the configuration request information is used to request the first configuration.

[0547] 4. The method according to Appendix 1, wherein the method further comprises:

[0548] The terminal device determines at least two of the following associations:

[0549] The first model;

[0550] The relevant information input to the first model;

[0551] The relevant information output by the first model;

[0552] The first configuration;

[0553] The operation instructions in the application;

[0554] The second network device.

[0555] 5. A data collection method, the method being applied to a terminal device, the method comprising:

[0556] The terminal device receives the first information sent by the higher-level network device or the RAN network device;

[0557] The terminal device uses a first configuration that matches the first information to collect first data.

[0558] The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

[0559] 6. A method for instructing data collection, the method being applied to a higher-level network device, the method comprising:

[0560] The higher-level network device sends the first information to the terminal device and / or the RAN network device.

[0561] The first information is at least related to the first data collected by the terminal device.

[0562] The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

[0563] 7. A data collection instruction method, the method being applied to a network device of a RAN, the method comprising:

[0564] The RAN network device receives first information from a higher-level network device, or the RAN network device receives first information and / or configuration requirement information from a terminal device;

[0565] The second network device sends a first configuration that matches the first information or the configuration requirement information to the terminal device;

[0566] The first information or the configuration requirement information is at least related to the first data collected by the terminal device.

[0567] The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

Claims

1. A data collection device, the device being applied to a terminal device, the device comprising: The first receiving unit receives first information sent by the first network device or the second network device; The collection unit, using a first configuration matching the first information, collects first data. The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

2. The apparatus according to claim 1, wherein, The first information includes at least one of the following: Instruction information instructing the terminal device to collect the first data; The type or attributes of the terminal device that collects the first data; The purpose of collecting the first data; The configuration information for sending the first data includes at least one of the conditions for sending the first data and the configuration for sending the first data. Relevant information about the first data; First related indication.

3. The apparatus according to claim 2, wherein, The relevant information of the first data includes at least one of the following: The relevant information input to the first model; The relevant information output by the first model; And / or, The relevant information in the first data indicates information about the input of the first model and information about the output of the first model, respectively. And / or, The first association indication is indicated by at least one of the following: identifier, index, pointer, sequence number, number. And / or, The first information is also used to indicate that at least two of the following pieces of information are related: First association indication; Relevant information about the first data; The relevant information input to the first model; The relevant information output by the first model; The first model.

4. The apparatus according to claim 1, wherein, The first configuration includes at least one of the following: Resource configuration, wherein the resource configuration is the resource configuration required by the terminal device to collect the first data; The parameter configuration is the parameter configuration on which the terminal device collects the first data; Data collection instruction, wherein the data collection instruction is used to instruct the first configuration to collect the first data; The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of the measurement and / or evaluation performed by the terminal device based on the resource configuration and / or the parameter configuration.

5. The apparatus according to claim 1, wherein, The terminal device receives the first configuration from the second network device; and / or The terminal device uses the existing configuration as the first configuration based on the first information.

6. The apparatus according to claim 5, wherein, In the case where the terminal device receives the first configuration from the second network device, the apparatus further includes: The first sending unit sends configuration request information to the second network device, the configuration request information being used to request the first configuration. The configuration requirement information is related to the first information, and / or the configuration requirement information includes all or part of the first information. And / or, The terminal device uses the existing configuration as the first configuration based on the first information, including: The terminal device determines the configuration requirements based on the first information; The terminal device determines the first configuration that matches the configuration requirement information from the existing configurations based on the configuration requirement information.

7. The apparatus according to claim 1, wherein, The device further includes: The first determining unit determines at least two of the following relationships: The first model; The relevant information input to the first model; The relevant information output by the first model; The first configuration; The first information; The second network device, Wherein, the terminal device determines the association relationship, including: The terminal device determines the association relationship based on the first association indication in the first information; or... The terminal device establishes the association relationship based on at least one of the first information, the first configuration, and the first model.

8. A data collection indication device, the device being applied to a first network device, the device comprising: The second sending unit sends the first information to the terminal device and / or the second network device. The first information is at least related to the first data collected by the terminal device. The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

9. The apparatus according to claim 8, wherein, The first information includes at least one of the following: Instruction information instructing the terminal device to collect the first data; The type or attributes of the terminal device that collects the first data; The purpose of collecting the first data; Send the configuration information of the first data. The configuration information includes at least one of the conditions for sending the first data and the configuration for sending the first data; Relevant information about the first data; First related indication.

10. The apparatus according to claim 9, wherein, The relevant information of the first data includes at least one of the following: The relevant information input to the first model; The relevant information output by the first model; And / or, The relevant information in the first data indicates information about the input of the first model and information about the output of the first model, respectively. And / or, The first association indication is indicated by at least one of the following: identifier, index, pointer, sequence number, number. And / or, The first information is also used to indicate that at least two of the following pieces of information are related: First association indication; Relevant information about the first data; The relevant information input to the first model; The relevant information output by the first model; The first model.

11. The apparatus according to claim 9, wherein, The first information includes the first association indication. The first association indication is used by the terminal device to determine at least two of the following association relationships: The first model; The relevant information input to the first model; The relevant information output by the first model; A first configuration, which matches the first information; The first information; The second network device.

12. The apparatus according to claim 8, wherein, The second sending unit also sends second information to the second network device. The second information is used at least to instruct the second network device to send a first configuration matching the first information to the terminal device, the first configuration including at least one of the following: Resource configuration, wherein the resource configuration is the resource configuration required by the terminal device to collect the first data; The parameter configuration is the parameter configuration on which the terminal device collects the first data; Data collection instruction, wherein the data collection instruction is used to instruct the first configuration to collect the first data; The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of the measurement and / or evaluation performed by the terminal device based on the resource configuration and / or the parameter configuration.

13. A data collection indication device, the device being applied to a second network device, the device comprising: The second receiving unit receives first information from the first network device, or receives first information and / or configuration requirement information from the terminal device; The third sending unit sends a first configuration that matches the first information or the configuration requirement information to the terminal device. The first information or the configuration requirement information is at least related to the first data collected by the terminal device. The first data is used to train a first model and supervise at least one of the first models, wherein the first model is an AI / ML model applied to the terminal device side.

14. The apparatus according to claim 13, wherein, The third sending unit also sends third information to the terminal device. The third information includes all or part of the information in the first information, and / or the third information includes at least the first association indication.

15. The apparatus according to claim 13, wherein, The first information includes at least one of the following: Instruction information instructing the terminal device to collect the first data; The type or attributes of the terminal device that collects the first data; The purpose of collecting the first data; The configuration information for sending the first data includes at least one of the conditions for sending the first data and the configuration for sending the first data. Relevant information about the first data; First related indication.

16. The apparatus according to claim 15, wherein, The relevant information of the first data includes at least one of the following: The relevant information input to the first model; The relevant information output by the first model; And / or, The relevant information in the first data indicates information about the input of the first model and information about the output of the first model, respectively. And / or, The first association indication is indicated by at least one of the following: identifier, index, pointer, sequence number, number. And / or, The first information is also used to indicate that at least two of the following pieces of information are related: First association indication; Relevant information about the first data; The relevant information input to the first model; The relevant information output by the first model; The first model.

17. The apparatus according to claim 13, wherein, The first information includes a first association indication. The first association indication is used by the terminal device to determine at least two of the following association relationships: The first model; The relevant information input to the first model; The relevant information output by the first model; The first configuration; The first information; The second network device, And / or, the first configuration includes at least one of the following: Resource configuration, wherein the resource configuration is the resource configuration required by the terminal device to collect the first data; The parameter configuration is the parameter configuration on which the terminal device collects the first data; Data collection instruction, wherein the data collection instruction is used to instruct the first configuration to collect the first data; The reporting configuration is the configuration by which the terminal device reports to the first network device and / or the second network device, and the report is related to the results of the measurement and / or evaluation performed by the terminal device based on the resource configuration and / or the parameter configuration.

18. The apparatus according to claim 14, wherein, The device further includes: The second determining unit determines the first association indication. The first association indication is used to associate at least one of the first model, the first information, the configuration requirement information, the first configuration, the second network device, and the logical entity managed by the second network device.

19. The apparatus according to claim 18, wherein, The second determining unit determines the first association indication based on the second association indication. The second association indication is used to associate at least one of the first model, the first information, and the configuration requirement information.

20. The apparatus according to claim 19, wherein, The second association indication is an association indication received by the second network device from the first network device or the terminal device, and / or, The first association instruction is also used to associate at least one of the second association instruction and the content associated with the second association instruction.

Images