Communication method and application apparatus

Abstract

Embodiments of the present application provide a communication method and an application apparatus. The method comprises: after a first communication apparatus receives a first model acquisition request from an OTT device, if a model requested by the first model acquisition request is permitted to be transmitted externally, the first communication apparatus transmits a first model acquisition response to the OTT device to indicate that the model can be transmitted. By using the embodiments of the present application, the OTT device can obtain the model requested by the OTT device, and the model is permitted to be transmitted externally, thereby improving the security of internal data within a communication network.

Description

Communication methods and application devices

[0001] This application claims priority to Chinese Patent Application No. 202510184553.5, filed on February 19, 2025, entitled "Communication Method and Application Device", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of artificial intelligence (AI) technology, and in particular to a communication method and application device. Background Technology

[0003] AI is a technology that simulates the human brain to perform complex calculations. With the improvement of data storage and computing power, artificial intelligence is being used more and more. The 3rd Generation Partnership Project (3GPP) proposes to apply artificial intelligence to wireless communication systems to improve network performance and user experience through intelligent data collection and analysis.

[0004] As the storage capacity of communication networks gradually increases, they can provide storage services, allowing some AI-based models to be stored on communication devices within the network. However, devices outside the communication network (such as over-the-top (OTT) devices) may need to access models stored within the network. Therefore, it is necessary to communicate with these external devices to transfer the models. Summary of the Invention

[0005] This application discloses a communication method and application device that can transmit models to OTT devices, and the transmitted models are allowed to be sent out, which can improve data security.

[0006] In a first aspect, embodiments of this application disclose a communication method applied to a first communication device. The first communication device may be a terminal device, access network device, core network device, or operation administration and maintenance (OAM) device, or it may be a component, part, or circuit within the terminal device, access network device, core network device, or OAM device, or it may be a chip or chip system applicable to the terminal device, access network device, core network device, or OAM device.

[0007] The following example uses a first communication device. The method includes: receiving a first model acquisition request from an OTT device, the first model acquisition request being used to request the acquisition of a model; and, if the model is permitted to be transmitted, sending a first model acquisition response to the OTT device, the first model acquisition response indicating that the model can be transmitted. In this way, the OTT device can acquire the requested model, and since the model is permitted to be transmitted, the security of data within the communication network can be improved.

[0008] In this application, the OTT device can directly send a first model acquisition request to the first communication device, or it can send the first model acquisition request through other communication devices; this is not limited here. The model can be stored in the first communication device, or it can be stored in a communication device other than the first communication device. It is understood that if the model is stored in the first communication device, the first communication device can send the stored model to the OTT device. If the model is stored in a communication device other than the first communication device, the communication device storing the model can send the model to the OTT device through the first communication device, or it can send the model to the OTT device through the first communication device and other communication devices, or it can send the model to the OTT device through other communication devices, or it can send the model directly to the OTT device. When the communication device storing the model cannot communicate directly with the OTT device, transmitting the model through the first communication device capable of communicating with the OTT device can improve the success rate of model transmission.

[0009] In this application, an OTT device can be an internet device that provides various application services to users via the internet. An OTT device can be a server from a communication equipment manufacturer, or a supplier of AI solutions (such as autonomous driving solutions, extended reality (XR) solutions), etc. An OTT device can be understood as a third-party application service platform, a communication device within an external network of the communication system. This external network can be an external network of the 3GPP communication network. The services provided by the OTT device can be internet television services, app stores, etc., without limitation.

[0010] In this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0011] In some feasible implementations, the first model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and model requirement information.

[0012] The request identifier for the first model acquisition request is used to identify the first model acquisition request, which helps in determining whether the received response is a response to the first model acquisition request. Model requirement information can be understood as the model attribute information required or expected by the OTT device.

[0013] In some feasible implementations, the model requirement information includes at least one of the following: expected model input information, expected model output information, expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0014] The expected model input information can be the input information of the model expected or required by the OTT device, and the expected model output information can be the output information of the model expected or required by the OTT device. The expected model functional information can be the functional information of the model required or expected by the OTT device, and the expected model operational information can be the operational information of the model expected or required by the OTT device (e.g., whether inference or training is possible). The model performance requirements can be the performance information of the model required or expected by the OTT device (e.g., accuracy not less than a given threshold). The model input information, model output information, model functional information, model operational information, model performance information, and model storage information can be found in the model attribute information described later, and will not be elaborated upon here.

[0015] Model format requirements indicate the format of model files required or expected by OTT devices. For example, model files can be .pt, .bin, .onnx, or .pth files. Model precision requirements indicate the precision of parameters in the model files required or expected by OTT devices, such as the parameter precision corresponding to formats like Float point32 (FP32), FP16, FP8, Int8, and brain float (BF)16. FP32 represents a 32-bit floating-point number, FP16 represents a 16-bit floating-point number, FP8 represents an 8-bit floating-point number, and Int8 represents an 8-bit integer. BF16 represents a 16-bit floating-point number, consisting of one sign bit, eight exponent bits, and seven decimal places. Its precision is lower than FP16, but its range is easily convertible between FP32 and FP16.

[0016] In some feasible implementations, the first model acquisition response includes at least one of the following: a request identifier of the first model acquisition request, a request completion indication, the model, a model identifier of the model, and the model attribute information; wherein the request completion indication is used to indicate whether the first model acquisition request has been completed.

[0017] In some feasible implementations, the model attribute information includes at least one of the following: the model identifier, the model input information, the model output information, the model function information, the model operation information, the model performance information, the model update information, the model source information, the model external information, the model quantization information, and the model storage information.

[0018] The model identifier is used to identify the model and can be used to determine the model and its other attribute information. The model identifier can be generated from the model type or the model creation time, etc., without limitation. The model's input information can be the model's input data, and the model's output information can be the model's output data.

[0019] The information of the input data and the information of the output data include one or more of the following: data type, data format, data precision, data storage location, and data storage volume (or the amount of storage space occupied by the data), which are not limited here. The data type can be a file type such as text, image, video, audio, or location. For example, if the input data is an image, the output data can be an image or text. Furthermore, the data type can be the content type of the data. For example, the input data can be communication information such as channel state information (CSI), and the output data can be the processing result of the communication information, such as compressed text information of CSI. Another example is that the input data is an image of the vehicle's external environment, and the output data can be the recognition result of the vehicle's external environment, such as text information or audio information of obstacles. Yet another example is that the input data is user-inputted voice, and the output data can be the recognition result of the user's intent. The data type can also be a format type corresponding to different data precisions, such as integer or floating-point. The data format can be the corresponding file format type, such as a GIF file, a TXT file, or a PTH file. Data precision can be represented by integer, floating-point, or limited by the number of decimal places.

[0020] The model's functional information can be a description of its function or purpose, indicating its capabilities, such as image compression, recognition, speech-to-text conversion, or image-to-text conversion, etc., without limitation. The model's operational information can be the type of operation, such as training or inference. Operational information can also be operational instructions, indicating how the model's functions are implemented, how to obtain information that meets the model's input requirements, and how to train or infer the model. The model's performance information indicates its performance evaluation metrics, such as accuracy.

[0021] Model update information may include, but is not limited to, one or more of the following: model update time, update version, and update functions. Model source information may include, but is not limited to, the model's storage location, the storage device storing the model, the storage device's identifier (e.g., the UE ID of the terminal device, the vendor identifier of the terminal device, etc.), the updating device for updating the model, and the identifier of the updating device. Model outbound information is used to indicate whether the model is allowed to be shared externally. When the model's outbound information indicates that it is allowed to be shared externally, it can be understood as an outbound authorization instruction; when the model's outbound information indicates that it is not allowed to be shared externally, it can be understood as an outbound denial instruction. Model quantization information may include, but is not limited to, one or more of the following: model quantization level, quantization type, and compression ratio. Model storage information is used to indicate the storage capacity required for the model, and can also be used to indicate the capacity required for model training or inference. This application does not limit the model attribute information.

[0022] Optionally, model attribute information may include intra-network attribute information and extra-network attribute information. Intra-network attribute information can be understood as the attribute information of the model that is allowed to be used by communication devices within the internal network of the communication system (such as a 3GPP communication system), while extra-network attribute information can be understood as the attribute information of the model that is allowed to be used by communication devices outside the communication system (such as OTT devices). For example, the model operations allowed to be used by communication devices within the internal network of the communication system are model training and inference, while the model operations allowed to be used by communication devices outside the communication system are inference, etc. This can improve the accuracy of model usage and data security. This application does not limit which attribute information is used by communication devices within the internal network of the communication system and which is used by communication devices outside the communication system; it can be determined based on whether the data used by the model is internal or external data, or based on whether the data used by the model is allowed to be transmitted externally, etc.

[0023] In some feasible implementations, the model attribute information obtained by the first model in the response may include all model attribute information, or it may include model attribute information that is allowed to be used by OTT devices. Model attribute information that is allowed to be used by OTT devices can be understood as model attribute information that is allowed to be used by communication devices on the external network of the communication system. For example, the model's input information specifically refers to input information that is allowed to be used by OTT devices, the model's output information specifically refers to output information that is allowed to be used by OTT devices, and the model's operation information specifically refers to operation information of functions that are allowed to be used by OTT devices, such as allowing OTT devices to perform inference on the model but not allowing them to train the model.

[0024] It is understandable that sending only the model attribute information that OTT devices are allowed to use, rather than all the model's attribute information, to the OTT device improves the efficiency and accuracy of the indication and avoids information leakage. If the first model acquisition response includes the model, that is, the first communication device directly sends the model to the OTT device. If the first model acquisition response does not include the model, the system can wait for the first communication device or another communication device to transmit the model to the OTT device.

[0025] In some feasible implementations, the method further includes: sending a second model acquisition response to the OTT device when the model is not allowed to be sent out, the second model acquisition response indicating that the model cannot be sent.

[0026] In some feasible implementations, the method further includes sending a second model acquisition request to a second communication device, the second model acquisition request being used to request the acquisition of a model.

[0027] The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information in the first model acquisition request. Alternatively, the second model acquisition request can include some of the information in the first model acquisition request. For example, the second model acquisition request might include the expected input information and / or expected output information from the first model acquisition request, but exclude other information. The first model acquisition request, in addition to the expected input and / or expected output information, may also include expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc. The description of the second model acquisition request can be referenced from that of the first model acquisition request, and the request identifier of the second model acquisition request can be the same as that of the first model acquisition request.

[0028] In some feasible implementations, the second communication device can be a different communication device from the first communication device, and the second communication device can communicate with the first communication device. For example, when the first communication device is a core network device or an OAM device, the second communication device can be an access network device or a terminal device; when the first communication device is a terminal device, the second communication device can be a core network device, an OAM device, or an access network device; when the first communication device is an access network device, the second communication device can be a core network device or an OAM device, etc.

[0029] It is understood that when the first communication device stores the model, it can send the model to the OTT device. When the first communication device does not store or register a model, it can send a second model retrieval request to the second communication device, either to instruct the communication device storing the model to send it, or to determine whether the model can be sent. The communication device storing the model can be the second communication device, or it can be a device other than the first or second communication device, such as a terminal device.

[0030] In some feasible implementations, the method further includes: determining whether the model can be allowed to be shared based on the first model acquisition request.

[0031] For example, the model is allowed to be transmitted if the expected model input information and / or expected model output information does not involve privacy data of the communication network, and is not allowed to be transmitted if the expected model input information and / or expected model output information involves privacy data of the communication network. The privacy data of the communication network can be user data, such as user location information. Alternatively, the privacy data of the communication network can be air interface information, such as reference signal receiving power (RSRP). This can prevent privacy data leakage and improve data security.

[0032] For example, if the expected model input information and / or expected model output information are deemed authorized information, then the model is permitted to transmit externally; if the expected model input information and / or expected model output information are not deemed authorized information, then the model is not permitted to transmit externally. The authorized information can be protocol-defined or vendor-defined, or configured wireless information that can be transmitted externally. This avoids the transmission of unauthorized information and provides data security.

[0033] In this application, the first communication device for determining whether a model can be distributed externally can be a network device such as an access network device, a core network device, or an OAM device. If the first communication device is not a network device, but rather a terminal device, a second model acquisition request can be sent to the network device, thereby determining whether the model can be distributed externally through the network device. Here, the network device can be the aforementioned second communication device, and the second model acquisition request can be described as described above, without further elaboration here. If the first communication device is a network device, a second model acquisition request can also be sent to another network device. For example, if the first communication device is an access network device, another network device can act as the second communication device, such as a core network device or an OAM device, thereby determining whether the model can be distributed externally through the second communication device.

[0034] In some feasible implementations, the method may further include: receiving a third model acquisition response from a second communication device, the third model acquisition response being used to indicate whether the model can be sent.

[0035] Thus, after the second communication device determines whether the model can be sent, it can send a third model acquisition response to the first communication device. If the third model acquisition response indicates that the model can be sent, the communication device storing the model can send the model to the OTT device. For example, if the communication device storing the model is the first communication device, the first communication device can directly send the model to the OTT device. If the communication device storing the model is the second communication device, the second communication device can directly send the model to the OTT device, or it can send the model to the OTT device through the first communication device. If the communication device storing the model is neither the first nor the second communication device, the communication device storing the model can directly send the model to the OTT device, or it can send the model to the OTT device through the first and / or the second communication device; this is not limited here. If the third model acquisition response indicates that the model cannot be sent, the first communication device, the second communication device, and the communication device storing the model may not send the model to the OTT device.

[0036] In some feasible implementations, the method may further include: receiving a model acquisition failure instruction from a second communication device, the instruction indicating that model acquisition has failed or that no suitable model exists. The lack of a suitable model could mean that a model registered in the second communication device does not meet the model requirements in the first model acquisition request, or that a model meeting the requirements in the first model acquisition request cannot be shared externally. Thus, even if a model is permitted to be shared externally, the OTT device will not receive the model corresponding to the first model acquisition request.

[0037] In some feasible implementations, the method may further include receiving a model from a second communication device. Thus, the second communication device can transmit the model to the OTT device via the first communication device. The model transmitted by the second communication device to the first communication device may be a model stored in the second communication device, or a model stored in a communication device other than the first and second communication devices.

[0038] In some feasible implementations, the method further includes: receiving a model registration request from a communication device storing a model, the model registration request being used to request registration of the model.

[0039] It is understood that after the first communication device registers a model, it can store the model's attribute information and the correspondence between the model's identifier and the device identifier of the communication device storing the model. This allows the communication device storing the model and the model's attribute information to be determined based on the model identifier. The model identifier can be assigned by the first communication device, for example, during the model registration process. Alternatively, the model identifier can be assigned by the device training or storing the model, etc., without limitation. After receiving a first model retrieval request, the first communication device can select the model corresponding to the first model retrieval request from the registered models and request the communication device storing that model to transmit the model, thus enabling model management.

[0040] For example, the communication device of the storage model is the UE, the first communication device is the BS, and the UE sends a model registration request to the BS. Another example is that the communication device of the storage model is the BS, the first communication device is the CN device, and the BS sends a model registration request to the CN device; or the communication device of the storage model is the UE, the first communication device is the CN device, and the UE sends a model registration request to the CN device, etc.

[0041] For example, the terminal device storing the model sends a model registration request to the network device. The network device can be an access network device, a core network device, etc. The network device can be a first communication device, or a communication device other than the first communication device, such as a second communication device. In this way, the network device can determine the terminal device storing the model and the attribute information of the model based on the model's model identifier. If the network device determines that it can send the model out, it can select a model from the registered models and request the terminal device storing that model to transmit the model, thus achieving model management.

[0042] In some feasible implementations, the model registration request may include, but is not limited to, model attribute information. Thus, the first communication device can select the model requested by the first model acquisition request based on the model attribute information, which helps improve the accuracy and efficiency of model selection.

[0043] In some feasible implementations, the method further includes receiving model feedback information from the OTT device.

[0044] In some feasible implementations, the method further includes sending model feedback information to the second communication device.

[0045] In this application, the model feedback information received by the first communication device from the OTT device can be referred to as the first model feedback information, and the model feedback information received by the second communication device from the first communication device can be referred to as the second model feedback information. The second model feedback information can be all or part of the information in the first model feedback information.

[0046] For example, the first communication device can forward all the first model feedback information to the second communication device. Thus, the first and second communication devices can perform the same or different analyses on the first model feedback information. The same analysis could involve not performing any additional operations if the model performance indicator indicates that the model's performance meets the requirements, not performing any additional operations if the re-request indicator indicates that no new model needs to be transmitted, and sending a model transmission command to the device storing the model if the re-request indicator indicates that a new model needs to be transmitted. The different analysis could involve the first communication device determining that it needs to forward the model feedback information to the second communication device if the model performance indicator indicates that the model's performance does not meet the requirements. The second communication device, in turn, determining to send a model switching command or an update command to the communication device storing the model if the model performance indicator indicates that the model's performance does not meet the requirements.

[0047] For example, the first communication device can analyze a portion of the model feedback, and the remaining model feedback information can be sent to the second communication device as second model feedback information, so that the second communication device can analyze it. In this way, the second communication device can analyze the model feedback information of the OTT device, and based on this, switch or update the model, which helps to improve the model's performance.

[0048] The model feedback information below can be either the first model feedback information or the second model feedback information.

[0049] In some feasible implementations, the model feedback information includes a model performance indicator, and the method further includes: when the model performance indicator indicates that the model's performance does not meet the model performance requirements, sending an update instruction or a switching instruction for the model and / or a model retransmission instruction to the communication device storing the model.

[0050] The update or switch instructions may include target metrics, such as the minimum accuracy of the model. Update instructions may also include model identifiers, etc. This application does not limit the content of update and switch instructions. The model retransmission instruction is used to instruct the retransmission of the model, which can be a retrained and updated model, or a reselected model, etc. Thus, in the case of updating the model, model performance can be improved. In the case of switching the model, it is beneficial to improve the performance and accuracy of the model used by OTT devices.

[0051] In some feasible implementations, the model feedback information includes a re-request indication, which indicates whether a new model should be transmitted.

[0052] It is understandable that when a re-request instruction is used to indicate that a new model should not be transmitted, no other additional operations need to be performed, such as not needing to respond to the OTT device again. When a re-request instruction is used to indicate that a new model should be transmitted, requiring a response from the OTT device to transmit the new model. For example, when a re-request instruction is used to indicate that a new model should be transmitted, the model needs to be reselected or retrained to switch or update the model, and a model acquisition response can also be sent to the OTT device. This acquisition response can be referred to the description of the first model acquisition response, and will not be repeated here. When a re-request instruction is used to indicate that a new model should not be transmitted, the model does not need to be reselected or retrained, and the model information does not need to be transmitted to the OTT device.

[0053] In some feasible implementations, the method further includes performing feedback analysis on the model feedback information. This helps to improve the model's performance.

[0054] In some feasible implementations, the method further includes selecting a model based on the first model acquisition request. For example, selecting a model based on the model requirement information in the first model acquisition request. This allows for the selection of a model that meets the requirements, thereby improving the accuracy of the transmitted model.

[0055] Furthermore, the first communication device can select a model based on factors such as the model's service requirements and / or the available resources for model transmission; this is not limited here. It is understood that the transmitted model's service requirements can be those specified by the OTT device, which can improve the accuracy of the transmitted model. When resources for model transmission are limited, a model with fewer parameters can be selected, thus avoiding model transmission failures. Alternatively, the model's parameters can be divided into multiple parts for separate transmission, reducing the transmission load on the first communication device. When multiple communication devices store the same model, each communication device can transmit a portion of the parameters separately, thereby reducing the transmission load on each communication device.

[0056] Secondly, this application discloses another communication method applied to an OTT device. The OTT device can be an internet device that provides various application services to users via the internet. The OTT device may be a server of a communication device manufacturer, or a communication device in an external network of a communication system, such as a supplier of AI solutions (e.g., autonomous driving solutions, XR solutions). The external network of the communication system can be a network outside the 3GPP communication network, such as the external network of the 3GPP communication system. The first and second communication devices can be communication devices within the 3GPP communication network, such as terminal devices and / or network devices in the internal network of the 3GPP communication system. The OTT device can be replaced by components, parts, or circuits within the OTT device, or it can be a chip or chip system applicable to the OTT device.

[0057] The following example uses an OTT device. The method includes: sending a first model acquisition request to a first communication device, the first model acquisition request being used to request the acquisition of a model; and receiving a first model acquisition response from the first communication device, the first model acquisition response being used to indicate whether the model can be sent.

[0058] In some feasible implementations, the method further includes: receiving a second model acquisition response from the first communication device, the second model acquisition response being used to indicate that the model cannot be sent.

[0059] In some feasible implementations, the method further includes sending model feedback information to the first communication device.

[0060] In some feasible implementations, the model feedback information includes at least one of the following: a model performance indicator, a re-request indicator, and a request identifier for the first model acquisition request; wherein the model performance indicator is used to indicate whether the performance of the model meets the model performance requirements, and the re-request indicator is used to indicate whether a new model should be transmitted.

[0061] In some feasible implementations, the method further includes: training or inferring the model based on the response obtained from the first model to obtain model feedback information. This helps to improve the performance of the model.

[0062] In some feasible implementations, the first model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and model requirement information.

[0063] In some feasible implementations, the model requirement information includes at least one of the following: expected model input information, expected model output information, expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0064] In some feasible implementations, the first model acquisition response includes at least one of the following: a request identifier of the first model acquisition request, a request completion indication, the model, a model identifier of the model, and the model attribute information; wherein the request completion indication is used to indicate whether the first model acquisition request has been completed.

[0065] In some feasible implementations, the model attribute information includes at least one of the following: the model identifier, the model input information, the model output information, the model function information, the model operation information, the model performance information, the model update information, the model source information, the model external information, the model quantization information, and the model storage information.

[0066] In some feasible implementations, the first communication device includes at least one of the following: access network equipment, core network equipment, operation, maintenance and management equipment (OAM), and terminal equipment.

[0067] In this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0068] It should be understood that the implementing entity of the second aspect is the OTT device. The specific content of the second aspect corresponds to that of the first aspect, and the corresponding features and beneficial effects achieved by the second aspect can be referred to the description of the first aspect. To avoid repetition, detailed descriptions are appropriately omitted here.

[0069] Thirdly, this application discloses another communication device applied to a second communication device. This second communication device can be a different communication device from the first communication device, and the second communication device is capable of communicating with the first communication device. The first communication device can be a terminal device, access network device, core network device, or operation, maintenance, and management (OAM) device, etc., or it can be a component, part, or circuit inside the terminal device, access network device, core network device, or OAM device, or it can be a chip or chip system that can be applied to the terminal device, access network device, core network device, or OAM device. That is, when the first communication device is a core network device or an OAM device, the second communication device can be an access network device or a terminal device; when the first communication device is a terminal device, the second communication device can be a core network device, an OAM device, or an access network device; when the first communication device is an access network device, the second communication device can be a core network device or an OAM device, etc.

[0070] In this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0071] The method includes: receiving a second model acquisition request from a first communication device, the second model acquisition request being used to request the acquisition of a model. Thus, if model transmission is permitted, the model can be sent or instructed to be sent via the second communication device.

[0072] In some feasible implementations, the method may further include: determining whether the model is permissible to be shared based on the second model acquisition request. This avoids the transmission of unauthorized information and provides data security.

[0073] In some feasible implementations, the method may further include sending a third model acquisition response to a first communication device, indicating whether the model is permitted to be transmitted. Thus, the first communication device can determine whether to transmit the model based on the third model acquisition response.

[0074] In some feasible implementations, the method may further include: selecting a model based on a second model acquisition request. This allows for the selection of a suitable model, thereby improving the accuracy of the transmitted model.

[0075] In some feasible implementations, the method may further include sending a model transmission instruction to a terminal device to instruct the transmission of the model. This method is applicable when model sharing is permitted, and the terminal device is a communication device storing the model. That is, when the first or second communication device determines that model sharing is permitted, it can instruct the terminal device storing the model to send the model to the OTT device. The terminal device may send the model directly to the OTT device, or it may send the model to the OTT device through other communication devices such as the first and / or second communication devices; this is not limited to these methods.

[0076] In some feasible implementations, the method may further include receiving a model from a terminal device. This method is applicable when model sharing is permitted, and the terminal device is a communication device that stores the model. That is, when the first or second communication device determines that model sharing is permitted, the second communication device can receive the model from the terminal device storing the model and send it to the OTT device. The second communication device can send the model directly to the OTT device, or it can send the model to the OTT device through the first communication device or other communication devices, etc., without limitation.

[0077] In some feasible implementations, the method may further include: sending the model to a first communication device. This method is applicable when model sharing is permitted, and the model may be stored in a second communication device or terminal device.

[0078] In some feasible implementations, the method may further include: sending a model acquisition failure instruction to the first communication device to indicate that model acquisition has failed. Thus, even if the model is permitted to be shared externally, the OTT device will not receive the model corresponding to the first model acquisition request.

[0079] In some feasible implementations, the method may further include: receiving model feedback information from the first communication device.

[0080] In some feasible implementations, the method may further include: providing feedback information to the model feedback information.

[0081] In this application, the model feedback information received by the first communication device from the OTT device can be referred to as the first model feedback information, and the model feedback information received by the second communication device from the first communication device can be referred to as the second model feedback information. The second model feedback information can be all or part of the information in the first model feedback information. Thus, the second communication device can analyze the model feedback information from the OTT device and, based on this, switch or update the model, thereby improving model performance.

[0082] In some feasible implementations, the method may further include: sending a model update instruction or a switching instruction to the terminal device, and / or a model retransmission instruction. The model update instruction indicates that the model be updated, and the model switching instruction indicates that the model be switched. The model retransmission instruction indicates that the model be retransmitted; this model could be a retrained, updated model, or a newly selected model, etc. Thus, in the case of updating the model, model performance can be improved. In the case of switching the model, it is beneficial to improve the performance and accuracy of the model used by the OTT device.

[0083] In some feasible implementations, the method may further include: receiving a model registration request from a terminal device to request model registration. In this way, the second communication device can manage the model.

[0084] In some feasible implementations, the second model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and model requirement information.

[0085] In some feasible implementations, the model requirement information includes at least one of the following: expected model input information, expected model output information, expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0086] The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information in the first model acquisition request. Alternatively, the second model acquisition request can include some of the information in the first model acquisition request. For example, the second model acquisition request might include the expected input information and / or expected output information from the first model acquisition request, but exclude other information. The first model acquisition request, in addition to the expected input and / or expected output information, may also include expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc. The description of the second model acquisition request can be referenced from that of the first model acquisition request, and the request identifier of the second model acquisition request can be the same as that of the first model acquisition request.

[0087] It should be understood that the third aspect is executed by the second communication device. The method executed by the second communication device can be referred to the description of the first aspect; to avoid repetition, a detailed description is omitted here.

[0088] Fourthly, this application discloses a communication device, including units, modules, or means for performing the steps of any of the first to third aspects or any implementation method therein. The modules, units, or means can be implemented by software, by hardware, or by a combination of software and hardware.

[0089] Fifthly, this application discloses another communication device, which includes a processor for executing computer programs or instructions, such that when the processor executes the computer programs or instructions, the methods provided in any of the first to third aspects or any feasible embodiments described above are implemented. Optionally, the communication device further includes a memory.

[0090] Optionally, the communication device may also include interface circuitry, wherein the processor is used to communicate with other devices or components via the interface circuitry.

[0091] Optionally, the communication device may also include a memory.

[0092] In conjunction with the fourth or fifth aspect, in some feasible examples, the communication device may be a first communication device, an OTT device, a second communication device, or a terminal device.

[0093] When the first communication device is a terminal device, the terminal device may store the model. When the first communication device is an access network device, a core network device, or an OAM device, the model may be stored in the terminal device, the first communication device, or the second communication device, etc., without limitation.

[0094] In this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0095] Sixthly, this application provides a communication system including a first communication device and a transfer device. When the first communication device operates in the communication system, it is used to perform the method provided in the first aspect or its feasible embodiments described above. When the transfer device operates in the communication system, it is used to perform the method provided in the second aspect or its feasible embodiments described above.

[0096] In some feasible examples, the communication device may further include a second communication device. When the second communication device operates in the communication system, it is used to perform the method provided in the third aspect or its feasible embodiments described above. Where neither the first nor the second communication device includes a terminal device, the communication device may further include a terminal device.

[0097] In a seventh aspect, this application provides a computer-readable storage medium storing a computer program or instructions that, when executed, cause the method of any one of the first to third aspects or any possible implementation thereof to be implemented.

[0098] Eighthly, this application provides a computer program product comprising a computer program or instructions that, when executed, cause the method of any one of the first to third aspects or any possible implementation thereof to be implemented.

[0099] Ninthly, this application provides a chip or chip system including at least one processor for calling and executing instructions stored in a memory, causing a communication device on which the chip or chip system is mounted to perform the method of any one of the first to third aspects or any possible implementation thereof.

[0100] Optionally, the chip also includes a communication interface for receiving or sending signals.

[0101] Optionally, the chip or chip system may also include memory.

[0102] In a tenth aspect, this application provides another chip, including: an input interface, an output interface, and a processing circuit. The input interface, the output interface, and the processing circuit are connected via internal connection paths. The processing circuit is used to execute the method of any of the above aspects or possible examples. Optionally, the chip also includes a memory. The input interface, the output interface, the processor, and the memory are connected via internal connection paths. The processor is used to execute code in the memory. When the code is executed, the processor is used to execute the method of any of the above aspects or possible examples.

[0103] In one aspect, this application provides a chip system including at least one processor and a communication interface, the communication interface and at least one processor being interconnected via a line, the at least one processor being used to run computer programs or instructions to perform the methods in any of the above aspects or possible examples.

[0104] It should be understood that the implementation and beneficial effects of the above-mentioned aspects can be mutually referenced.

[0105] Furthermore, in the process of executing any of the first to third aspects and any possible implementations of the method described above, the processes related to sending and / or receiving information can be understood as the process of the processor outputting information and / or the processor receiving input information. When outputting information, the processor can output the information to a transceiver (or communication interface or transmitting module) for transmission. After the information is output by the processor, it may require further processing before reaching the transceiver. Similarly, when the processor receives input information, the transceiver (or communication interface or transmitting module) receives the information and inputs it to the processor. Furthermore, after the transceiver receives the information, it may require further processing before being input to the processor.

[0106] Based on the above principles, for example, the information sent mentioned in the aforementioned method can be understood as information output by the processor. Similarly, the information received can be understood as information received by the processor from input.

[0107] Optionally, unless otherwise specified, or unless they contradict their actual function or internal logic in the relevant description, the operations of the processor, such as transmitting, sending, and receiving, can be more generally understood as processor output and receiving, input, and other operations.

[0108] Optionally, in the process of executing the method of any of the first or second aspects and any possible implementations described above, the processor may be a processor specifically designed to execute these methods, or it may be a processor that executes these methods by executing computer instructions stored in memory, such as a general-purpose processor. The memory may be a non-transitory memory, such as read-only memory (ROM), which may be integrated with the processor on the same chip or disposed on separate chips. This application does not limit the type of memory or the arrangement of the memory and processor. Attached Figure Description

[0109] The accompanying drawings used in the embodiments of this application are described below.

[0110] Figures 1A and 1B are schematic diagrams of the architecture of a communication system provided in an embodiment of this application;

[0111] Figure 2 is a functional architecture diagram of AI / ML provided in an embodiment of this application;

[0112] Figures 3 to 7 are interactive schematic diagrams of a communication method provided in an embodiment of this application;

[0113] Figure 8 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0114] Figure 9 is a schematic diagram of another communication device provided in an embodiment of this application;

[0115] Figure 10 is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0116] The technical solutions of this application embodiment can be applied to various communication systems, such as long term evolution (LTE) communication systems, new radio (NR) communication systems, LTE-A advanced (LTE-A) communication systems, device-to-device (D2D) communication systems, vehicle-to-everything (V2X) communication systems, machine-to-machine (M2M) communication systems, internet of things (IoT) communication systems, narrowband internet of things (NB-IoT) communication systems, integrated sensing and communication systems, frequency division duplex (FDD) communication systems, time division duplex (TDD) communication systems, non-terrestrial network (NTN) communication systems, wireless projection communication systems, integrated access and backhaul (IAB) communication systems, public land mobile network (PLMN) communication systems, and non-public networks. The network (NPN) communication system, as well as communication systems that evolve after 5G communication systems (such as 6G communication systems), or non-3rd generation partnership project (3GPP) communication systems, are not restricted.

[0117] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described below with reference to the accompanying drawings.

[0118] Please refer to Figure 1A, which is a schematic diagram of the architecture of a communication system provided in an embodiment of this application. As shown in Figure 1A, the communication system 1000 includes a radio access network (RAN) 100 and a core network (CN) 200. RAN 100 includes at least one RAN node (110a and 110b in Figure 1A, collectively referred to as 110) and at least one terminal device (120a-120j in Figure 1A, collectively referred to as 120). The terminal device is wirelessly connected to the access network device, and the access network device is wirelessly or wiredly connected to the core network. The core network device and the access network device can be independent and different physical devices, or the functions of the core network device and the logical functions of the access network device can be integrated on the same physical device, or a single physical device can integrate some of the functions of the core network device and some of the functions of the access network device. Terminal devices and access network devices can be interconnected via wired or wireless means. Figure 1A is a schematic diagram; this communication system may also include other devices, network elements, and networks.

[0119] In this application embodiment, a terminal device is an entity on the user side used to receive or transmit signals, providing voice and / or data to the user. A terminal device may be a terminal, user equipment (UE), access terminal, UE unit, UE station, mobile device, mobile station, mobile station, mobile terminal, mobile client, mobile unit, remote station, remote terminal, remote unit, wireless unit, wireless communication device, user agent, or user device, etc. The access terminal may be a cellular phone, cordless phone, session initiation protocol (SIP) phone, wireless local loop (WLL) station, personal digital assistant (PDA), handheld device with wireless communication capabilities, computing device or other processing device connected to a wireless modem, vehicle-mounted device, a terminal in a future 5G communication system, a terminal in a future evolved PLMN, or a terminal in a future NPN, etc.

[0120] As an example and not a limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0121] Terminal equipment can also be a communication module, satellite phone, or its components with satellite communication capabilities, or a satellite communication terminal, such as a very small aperture terminal (VSAT) (commonly referred to as a VSAT terminal), portable station, fixed station, vehicle-mounted or airborne satellite communication terminal, etc. It should be understood that a satellite communication terminal can serve as a micro base station to further provide data interfaces to accessed user equipment. Hereinafter, it will sometimes be simply referred to as a terminal.

[0122] In the embodiments of this application, the terminal device can be a terminal as a final product, such as the various terminal devices described above; it can also be a component or part with terminal functions; it can be a circuit or chip (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip or system-in-package (SIP) chip containing a modem core), a chip system, or a processor) that can be applied to the terminal to perform communication functions; or it can be a logic node, logic module, or software that can implement all or part of the terminal functions. The chip system can be composed of chips, or it can include chips and other discrete devices.

[0123] Terminal devices can communicate with each other using some kind of air interface technology (such as NR or LTE). Terminal devices can also communicate with network devices using some kind of air interface technology (such as NR or LTE).

[0124] In this embodiment, the access network device may be referred to as a RAN device, an access network node, or an access network apparatus, or simply an access network. The access network device is used to connect terminal devices to the network. That is, the access network provides access services to terminal devices so that they can access (or access) the network. The access network can support both wired and wireless access.

[0125] Optionally, the access network equipment consists of multiple AN / RAN nodes. AN / RAN nodes can include, but are not limited to: access points (APs), enhanced node Bs (eNBs), home evolved Node Bs (HNBs), baseband units (BBUs), next-generation node Bs (gNBs), transmission reception points (TRPs), transmission points (TPs), or other access nodes, such as wireless relay nodes or wireless backhaul nodes. AN / RAN nodes can be one or more antenna panels, or network nodes constituting gNBs or transmission points, such as BBUs or distributed units (DUs), or devices performing RAN functions in communication systems such as D2D, V2X, M2M, and U2U, such as roadside units (RSUs). AN / RAN nodes can be radio controllers in cloud radio access network (CRAN) scenarios, open RAN (O-RAN or ORAN), or access networks in future communication systems, etc., without any limitations.

[0126] In some deployments, a gNB may include a centralized unit (CU) and a dedicated unit (DU). The gNB may also include an active antenna unit (AAU). The CU implements some of the gNB's functions, and the DU implements others. For example, the CU handles non-real-time protocols and services, implementing radio resource control (RRC) and packet data convergence protocol (PDCP) layer functions. The DU handles physical layer protocols and real-time services, implementing radio link control (RLC), media access control (MAC), and physical (PHY) layer functions. The AAU implements some physical layer processing functions, radio frequency processing, and active antenna-related functions. RRC layer information is generated by the CU and is ultimately encapsulated by the DU's PHY layer to become PHY layer information, or it may be derived from PHY layer information. Therefore, in this architecture, higher-layer signaling, such as RRC layer signaling, can be considered as being sent by the DU, or by the DU+AAU. It is understood that network devices can be one or more of the following: CU nodes, DU nodes, and AAU nodes. Furthermore, a CU can be classified as a network device in the RAN or as a network device in the core network; this application does not limit this classification.

[0127] In some deployments, a gNB may include a radio unit (RU). The RU may be included in a radio frequency device or radio frequency unit, such as a remote radio unit (RRU), an AAU, or a remote radio head (RRH).

[0128] In different systems, CU (or CU-control plane (CP), CU-user plane (UP)), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0129] In different communication systems, core network equipment (hereinafter referred to as core network) can correspond to different devices. For example, in a 3G communication system, core network equipment can correspond to the Serving GPRS Support Node (SGSN) and / or the Gateway GPRS Support Node (GGSN); in a 4G communication system, core network equipment can correspond to the Mobility Management Entity (MME) and / or the Serving Gateway (S-GW); in a 5G communication system, core network equipment can correspond to policy control function (PCF) network elements, application function (AF) network elements, access and mobility management function (AMF) network elements, session management function (SMF) network elements, location management function (LMF) network elements, user plane function (UPF) network elements, network data analytics function (NWDAF), etc., as shown in Figure 1B.

[0130] Among them, the UPF network element is responsible for managing the transmission of user plane data and quality of service (QoS) control, traffic statistics and other functions. It can perform user data packet forwarding according to the routing rules of the session management network element, such as sending uplink data to the data network or other user plane network elements, and forwarding downlink data to other user plane network elements or (R)AN network elements.

[0131] The AMF (Access Default Mode) network element is responsible for user access management, security authentication, and mobility management. The LMF (Local Mode Default Mode) network element manages and controls location service requests from target terminals and processes location-related information. The SMF (Signal Management Default Mode) network element is responsible for session management, allocating and releasing resources for terminal device sessions. The PCF (Policy and Charging Rules Function) network element is responsible for user policy management. Similar to the Policy and Charging Rules Function (PCRF) network element in LTE, it is mainly responsible for policy authorization, quality of service (QoS), and generating charging rules, and distributing these rules to the UPF (Universal Programming Default Mode) network element via the SMF network element to complete the installation of the corresponding policies and rules. The AF (Application Default Mode) network element can be a third-party application control platform or the operator's own equipment. The AF network element is responsible for application management and can provide services to multiple application servers.

[0132] The NWDAF (Network Data Analyzer) element is used for analyzing network slicing-related data and can be extended to analyze various types of network data, including network operation data collected from network functions, statistical data related to terminal devices and networks obtained from Operation Administration and Maintenance (OAM) equipment, and application data obtained from third-party applications. The analysis results generated by NWDAF can also be output to network functions, OAM equipment, or third-party applications. NWDAF is also typically responsible for training AI models. AI models trained by NWDAF can be applied to network-specific domains such as mobility management, session management, and network automation, using AI methods to replace the numerical formula-based methods in traditional network functions.

[0133] OAM equipment, also known as OAM entities or OAM functions, refers to the network management work typically divided into three categories based on the actual needs of operator network operations: operation, administration, and maintenance, abbreviated as OAM. Operation mainly involves the analysis, prediction, planning, and configuration of daily network and services; maintenance mainly involves daily operational activities such as testing and fault management of the network and its services. OAM equipment can detect network operating status, optimize network connectivity and performance, improve network stability, and reduce network maintenance costs.

[0134] In the embodiments of this application, network elements may also be referred to as functional network elements, functional entities, nodes, devices, etc. A network element can be a network component implemented on dedicated hardware, a software instance running on dedicated hardware, or an instance of virtualized functionality on a suitable platform, such as a cloud platform. In different communication systems, the network elements such as AMF, SMF, PCF, and NWDAF shown in Figure 1B above may have other names, which are not limited in this application.

[0135] Optionally, referring to Figure 1B, the communication system may also include a data network device (not shown in Figure 1A). The data network device will be referred to as the data network below. The data network is used to provide business services to users. Generally, the client is a terminal, and the server is the data network. The data network provided by the data network can be a private network, such as a local area network (LAN). The data network can also be an external network not managed by the operator, such as the Internet. Alternatively, the data network can be a proprietary network jointly deployed by the operator, such as a network providing Internet Protocol Multimedia Subsystem (IMS) services.

[0136] In the embodiments of this application, the network device may be a network equipment as a final product, such as at least one of the above-mentioned access network equipment, core network equipment and data network equipment or network elements therein, or may include independent network elements, or may be a component or part with network equipment functions, or may be a communication chip (such as a processor, baseband chip or chip system, etc.) that can be applied in the network equipment.

[0137] The number and types of network devices and terminal devices included in the network architecture shown in Figures 1A and 1B are merely examples, and the embodiments of this application are not limited thereto. For example, it may also include more or fewer terminal devices communicating with the network devices. As another example, it may also include more or fewer network devices communicating with the terminal devices. For the sake of simplicity, they are not described one by one in the figures.

[0138] In this embodiment, the communication system may further include an OTT device, such as an OTT server. An OTT device can be an internet device that provides various application services to users via the internet. The OTT device may be a server belonging to a communication device manufacturer, or a communication device within an external network of the communication system, such as a supplier of AI solutions (e.g., autonomous driving solutions, XR solutions). The external network of the communication system may be an external network of the 3GPP communication network. The services provided by the OTT device may include internet television services, app stores, etc., and are not limited thereto.

[0139] The aforementioned terminal devices and network devices, as well as OAM devices and OTT devices, can all be referred to as communication devices. They can be general-purpose devices or special-purpose devices, and the embodiments of this application do not specifically limit them.

[0140] The network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0141] To facilitate understanding of the proposed solution, the functional framework for AI or machine learning (ML) is described below with reference to Figure 2. This functional framework is the AI / ML functional framework for the NR air interface in version R18 or R19 of protocol TS38.843, as shown in Figure 2. This functional framework includes data collection, model training, management, inference, and model storage functions.

[0142] The collected data can include monitoring data, training data, and inference data. Training data serves as the input for training AI / ML models. Inference data serves as the input for AI / ML inference functions, and monitoring data serves as the input for managing AI / ML models or functions.

[0143] The model training function is used to train, validate, and test AI / ML models. It can also be used to generate model performance metrics, which can be used as part of the model testing process. The model training function is also used to process training data, such as data preprocessing, cleaning, formatting, and transformation. After training or updating the model, it can be transferred or delivered to a device with model storage capabilities for model storage.

[0144] The management function is responsible for managing the operations of AI / ML models or functions, such as selection, deactivation, switching, and rollback, and also for performance monitoring. The management function is also responsible for making decisions to ensure correct inference operations are performed based on data received from the data acquisition and inference functions. Devices with management functions can send management instructions to devices with inference functions, which serve as the information required for managing the inference function's input. Devices with management functions can also request model transfers or delivery requests from devices with model storage functions to store the model. Performance feedback or retraining requests serve as the input information required for the model training function.

[0145] Devices with inference capabilities use inference data as input to provide outputs, such as inference outputs, to devices that apply AI / ML models or AI / ML functions. Inference outputs can be used to monitor the performance of AI / ML models or AI / ML functions. After storing models on devices with model storage capabilities, AI / ML models can be transferred or delivered to devices with inference capabilities.

[0146] This application proposes a communication method that can transmit models to OTT devices, and the transmitted models are allowed to be sent out, which can improve the security of data within the communication network.

[0147] The following descriptions will illustrate these methods through various embodiments. It should be understood that these methods can be used in combination. The technical solutions provided in this application are not limited to the processes described below. Furthermore, the scenario descriptions in this application are merely illustrative and do not limit the scope of the solutions described. The solutions in this application are applicable not only to the described scenarios but also to scenarios with similar problems.

[0148] The first communication device in this application embodiment can be a terminal device, access network device, core network device, or OAM device, etc., or it can be a component, part, or circuit inside the terminal device, access network device, core network device, or OAM device, or it can be a chip or chip system that can be applied in the terminal device, access network device, core network device, or OAM device. The second communication device in this application embodiment can be a different communication device from the first communication device, and the second communication device can communicate with the first communication device. For example, when the first communication device is a core network device or OAM device, the second communication device can be an access network device or a terminal device; when the first communication device is a terminal device, the second communication device can be a core network device, OAM device, or access network device; when the first communication device is an access network device, the second communication device can be a core network device or OAM device, etc. The OTT device in this application embodiment can refer to the foregoing, and these communication devices can be described with reference to Figure 1A or Figure 1B, and will not be repeated here.

[0149] In the embodiments of this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0150] Please refer to Figure 3, which is an interactive schematic diagram of a communication method provided in an embodiment of this application. The communication devices involved in this method may include a first communication device, a second communication device, and an OTT device, etc. The first and second communication devices can be communication devices within a 3GPP communication network, and the OTT device can be a communication device outside of a 3GPP communication network. This method includes, but is not limited to, the following steps S301 and S302, wherein:

[0151] Step S301: The OTT device sends a first model acquisition request to the first communication device. The first model acquisition request is used to request the acquisition of a model.

[0152] Accordingly, the first communication device receives the first model acquisition request from the OTT device. That is, the OTT device requests to acquire the model. The OTT device can send the first model acquisition request directly to the first communication device, or it can send the first model acquisition request through other communication devices; this is not limited here.

[0153] In this embodiment, the model can be stored in a first communication device, or in a communication device other than the first communication device, such as a second communication device or a terminal device. It is understood that when the model is stored in the first communication device, the first communication device can send the stored model to an OTT device. When the model is stored in a communication device other than the first communication device, the communication device storing the model can send the model to the OTT device through the first communication device, or through the first communication device and other communication devices, or through other communication devices, or directly to the OTT device.

[0154] For example, if the first communication device is a CN device, OAM device, or BS, and the model is stored in the first communication device, then the first communication device can send the model to the OTT device. Alternatively, if the first communication device is a CN device or OAM device, and the model is stored in the UE, the UE can forward the model to the first communication device via the BS; that is, the communication device storing the model sends the model to the OTT device through both the first communication device and the BS. Or, if the first communication device is a UE, and the model is stored in the UE, then the UE can send the model to the OTT device. It can be understood that when the communication device storing the model cannot directly communicate with the OTT device, transmitting the model through the first communication device, which can communicate with the OTT device, can improve the success rate of model transmission.

[0155] The communication device for storing the model can be a first communication device, such as a CN device, an OAM device, or a BS device, and the model can be stored in the first communication device; or, the first communication device can be a terminal device, and the model can be stored in the first communication device. Alternatively, the communication device for storing the model may not be the first communication device, meaning the model is stored in a communication device other than the first communication device. For example, if the first communication device is a CN device or an OAM device, the model can be stored in a terminal device.

[0156] In some feasible implementations, the method may further include: the communication device storing the model sending a model registration request to the first communication device. Correspondingly, the first communication device receives the model registration request from the communication device storing the model.

[0157] For example, the communication device of the storage model is the UE, the first communication device is the BS, and the UE sends a model registration request to the BS. Another example is that the communication device of the storage model is the BS, the first communication device is the CN device, and the BS sends a model registration request to the CN device; or the communication device of the storage model is the UE, the first communication device is the CN device, and the UE sends a model registration request to the CN device, etc.

[0158] The model registration request is used to request the registration of a model. Registering a model can involve storing model attribute information, as well as the correspondence between the model's identifier and the device identifier of the communication device storing the model. This allows the communication device storing the model and the model's attribute information to be determined based on the model identifier. The model identifier can be assigned by the first communication device, for example, during the model registration process. Alternatively, the model identifier can be assigned by the device training or storing the model, etc., without limitation. After receiving the first model acquisition request, the first communication device can select the model corresponding to the first model acquisition request from the registered models and request the communication device storing that model to transmit the model, thus enabling model management.

[0159] For example, the terminal device storing the model sends a model registration request to the network device. The network device can be an access network device, a core network device, etc. The network device can be a first communication device, or a communication device other than the first communication device, such as a second communication device. In this way, the network device can determine the terminal device storing the model and the attribute information of the model based on the model's model identifier. If the network device determines that it can send the model out, it can select a model from the registered models and request the terminal device storing that model to transmit the model, thus achieving model management.

[0160] In some feasible implementations, the model registration request may include, but is not limited to, model attribute information.

[0161] In the embodiments of this application, model attribute information is used to indicate the attribute information of the model, such as at least one of the following: model identifier, model input information, model output information, model function information, model operation information, model performance information, model update information, model source information, model external information, model quantization level, model storage information, etc.

[0162] The model identifier is used to identify the model and can be used to determine the model and its other attribute information. The model identifier can be generated from the model type or the model generation time, etc., without limitation here. The model input information can be the information of the model's input data, and the model output information can be the information of the model's output data. The input and output data information can include one or more of the following: data type, data format, data precision, data storage location, data storage volume (or the amount of storage space occupied by the data), etc., without limitation here.

[0163] Data types can be file types such as text, images, videos, audio, and location. For example, if the input data is an image, the output data can be an image or text. Furthermore, data types can be content types. For instance, if the input data is communication information such as channel state information (CSI), the output data can be the processing result of that communication information, such as compressed text information of CSI. Another example is an image of the vehicle's external environment as input, and the output data can be the recognition result of the external environment, such as text or audio information about obstacles. Yet another example is user-inputted speech, and the output data can be the recognition result of the user's intent. Data types can also be format types corresponding to different data precisions, such as integers or floating-point numbers. Data formats can be the corresponding file format types, such as GIF files, TXT files, and PTH files. Data precision can be represented by integers or floating-point numbers, or it can be limited by the number of decimal places.

[0164] The model's functional information can be a description of its function or purpose, indicating its capabilities, such as image compression, recognition, speech-to-text conversion, or image-to-text conversion, etc., without limitation. The model's operational information can be the type of operation, such as training or inference. Operational information can also be operational instructions, indicating how the model's functions are implemented, how to obtain information that meets the model's input requirements, and how to train or infer the model. The model's performance information indicates its performance evaluation metrics, such as accuracy.

[0165] Model update information may include, but is not limited to, one or more of the following: model update time, update version, and update functions. Model source information may include, but is not limited to, the model's storage location, the storage device storing the model, the storage device's identifier (e.g., the UE ID of the terminal device, the vendor identifier of the terminal device, etc.), the updating device for updating the model, and the identifier of the updating device. Model outbound information is used to indicate whether the model is allowed to be shared externally. When the model's outbound information indicates that it is allowed to be shared externally, it can be understood as an outbound authorization instruction; when the model's outbound information indicates that it is not allowed to be shared externally, it can be understood as an outbound denial instruction. Model quantization information may include, but is not limited to, one or more of the following: model quantization level, quantization type, and compression ratio. Model storage information is used to indicate the storage capacity required for the model, and can also be used to indicate the capacity required for model training or inference. This application does not limit the model attribute information.

[0166] Optionally, model attribute information may include intra-network attribute information and extra-network attribute information. Intra-network attribute information can be understood as the attribute information of the model that is allowed to be used by communication devices within the internal network of the communication system (such as a 3GPP communication system), while extra-network attribute information can be understood as the attribute information of the model that is allowed to be used by communication devices outside the communication system (such as OTT devices). For example, the model operations allowed to be used by communication devices within the internal network of the communication system are model training and inference, while the model operations allowed to be used by communication devices outside the communication system are inference, etc. This can improve the accuracy of model usage and data security. This application does not limit which attribute information is used by communication devices within the internal network of the communication system and which is used by communication devices outside the communication system; it can be determined based on whether the data used by the model is internal or external data, or based on whether the data used by the model is allowed to be transmitted externally, etc.

[0167] This application does not limit the content of the first model acquisition request. In some feasible implementations, the first model acquisition request may include, but is not limited to, at least one of the following: the request identifier of the first model acquisition request and the model requirement information.

[0168] The request identifier for the first model acquisition request is used to identify the first model acquisition request, which helps to determine whether the received response is a response to the model acquisition request. The request identifier for the first model acquisition request can be generated by the device identifier of the OTT device, or by the model identifier, or by the time of the first model acquisition request, the device identifier storing the model, etc., without limitation here.

[0169] Model requirement information can be understood as the model attribute information required or expected by OTT devices. In some feasible implementations, model requirement information includes, but is not limited to, at least one of the following: expected model input information, expected model output information, expected model function information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage requirements, etc.

[0170] The expected model input information can be the input information of the model expected or required by the OTT device, and the expected model output information can be the output information of the model expected or required by the OTT device. The expected model functional information can be the functional information of the model required or expected by the OTT device, and the expected model operational information can be the operational information of the model expected or required by the OTT device (e.g., whether inference or training is possible). The model performance requirements can be the performance information of the model required or expected by the OTT device (e.g., accuracy not less than a given threshold), and the model storage requirements can be the storage information of the model required or expected by the OTT device. The descriptions of the model's input information, output information, functional information, operational information, performance information, and storage information are similar to those described in the preceding section on model attribute information and will not be repeated here.

[0171] Model format requirements indicate the format of model files required or expected by OTT devices. For example, model files can be .pt, .bin, .onnx, or .pth files. Model precision requirements indicate the precision of parameters in the model files required or expected by OTT devices, such as the parameter precision corresponding to formats like Float point32 (FP32), FP16, FP8, Int8, and brain float (BF)16. FP32 represents a 32-bit floating-point number, FP16 represents a 16-bit floating-point number, FP8 represents an 8-bit floating-point number, and Int8 represents an 8-bit integer. BF16 represents a 16-bit floating-point number, consisting of one sign bit, eight exponent bits, and seven decimal places. Its precision is lower than FP16, but its range is easily convertible between FP32 and FP16.

[0172] In some feasible implementations, the first communication device selects a model based on a first model acquisition request. For example, the model is selected based on the model requirement information in the first model acquisition request. In this way, a model that meets the requirements can be selected, thereby improving the accuracy of the transmitted model.

[0173] For example, the first communication device is a BS or a UE, which can select a model with symbol requirements based on the stored model attribute information and model requirement information.

[0174] Furthermore, the first communication device can select a model based on factors such as the model's service requirements and / or the available resources for model transmission; this is not limited here. It is understood that the transmitted model's service requirements can be those specified by the OTT device, which can improve the accuracy of the transmitted model. When resources for model transmission are limited, a model with fewer parameters can be selected, thus avoiding model transmission failures. Alternatively, the model's parameters can be divided into multiple parts for separate transmission, reducing the transmission load on the first communication device. When multiple communication devices store the same model, each communication device can transmit a portion of the parameters separately, thereby reducing the transmission load on each communication device.

[0175] In other feasible implementations, the second communication device selects a model. The method for selecting the model can refer to the method used by the first communication device, whereby the second communication device selects a model based on a request to obtain a second model; this will not be elaborated further here.

[0176] In some feasible implementations, the method further includes: the first communication device sending a second model acquisition request to the second communication device. Correspondingly, the second communication device receives the second model acquisition request from the first communication device.

[0177] The second model acquisition request is used to request the acquisition of a model. The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information from the first model acquisition request. Alternatively, the second model acquisition request can include only some information from the first model acquisition request. For example, the second model acquisition request might include the expected input information and / or expected output information from the first model acquisition request, but exclude other information. The first model acquisition request, in addition to the expected input and / or expected output information, may also include expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc., beyond the expected input and / or expected output information. The second model acquisition request can refer to the description of the first model acquisition request, and the request identifier of the second model acquisition request can be the same as the request identifier of the first model acquisition request. It is understood that when the second communication device selects a model that matches the model acquisition request based on the model attribute information, the accuracy and efficiency of model selection can be improved.

[0178] This application does not limit the choice between the first and second communication devices for selecting the model. The communication device selecting the model can be a communication device that has registered the model, or a communication device that stores the model, etc. For example, if the first communication device is a CN device or an OAM device, and the second communication device is a BS that has registered the model, then the BS can select the model, i.e., the second communication device selects the model. As another example, if the first communication device is a UE, and the UE stores the model, then the UE can select the model; or, if the first communication device is a CN device, an OAM device, or a BS, and the CN device, OAM device, or BS stores the model, then the CN device, OAM device, or BS can select the model, i.e., the first communication device can select the model.

[0179] Optionally, after the first or second communication device selects a model, if the selected communication device has the model stored, it can send the model to the OTT device. If the selected communication device does not have the model stored, it can send a model transmission command to the communication device that stores the model to instruct the model to be transmitted. The communication device can send the model directly to the OTT device, or it can transmit the model through the first and / or second communication devices, etc., without limitation.

[0180] Step S302: When the model is permitted to be sent out, the first communication device sends a first model acquisition response to the OTT device. The first model acquisition response is used to indicate that the model can be sent.

[0181] Accordingly, the OTT device receives the first model to obtain the response.

[0182] In some feasible implementations, the first model acquisition response may include a request identifier of the first model acquisition request, a request completion indication, the model, the model identifier of the model, and model attribute information.

[0183] The request identifier and model attribute information of the first model acquisition request, as well as the model identifier, can be referred to in the description of the content of the first model acquisition request, and will not be repeated here. In some feasible implementations, the model attribute information in the first model acquisition response may include all model attribute information, or it may be model attribute information that is allowed to be used by OTT devices, etc. The model attribute information that is allowed to be used by OTT devices can be understood as the model attribute information that is allowed to be used by communication devices of the external network of the communication system. For example, the model's input information is specifically the input information that is allowed to be used by OTT devices, the model's output information is specifically the output information that is allowed to be used by OTT devices, and the model's operation information is specifically the operation information of the functions that are allowed to be used by OTT devices, etc.

[0184] It is understandable that sending only the model attribute information that OTT devices are allowed to use, rather than all the model's attribute information, to the OTT device improves the efficiency and accuracy of the indication and avoids information leakage. If the first model acquisition response includes the model, that is, the first communication device directly sends the model to the OTT device. If the first model acquisition response does not include the model, the system can wait for the first communication device or another communication device to transmit the model to the OTT device.

[0185] The request completion indicator indicates whether the first model acquisition request has been completed. For example, a request completion indicator of 0 indicates that the first model acquisition request has been completed, and a request completion indicator of 1 indicates that the first model acquisition request has not been completed. As another example, a request completion indicator of 1 indicates that the first model acquisition request has been completed, and a request completion indicator of 0 indicates that the first model acquisition request has not been completed. A completed first model acquisition request means that the model requested for transmission has been successfully transmitted. An incomplete first model acquisition request means that not all of the models requested for transmission have been transmitted, or it is uncertain whether to transmit any models.

[0186] In this embodiment, "model permission to be shared" means that the model is allowed to be sent to the external network of the communication system. In practice, this can mean that the model's input and / or output information is allowed to be sent to the external network of the communication system, such as to an OTT device. In some feasible implementations, the method may further include: a first communication device determining whether the model is permission to be shared based on a first model acquisition request. If yes, step S302 is executed, i.e., a first model acquisition response is sent to the OTT device. If no, a second model acquisition response is sent to the OTT device if the model is not permission to be shared. Accordingly, the OTT device receives the second model acquisition response sent by the first communication device.

[0187] The second model acquisition response indicates that the model cannot be sent. For example, a model acquisition response of 0 indicates that the model cannot be sent (i.e., the second model acquisition response); a model acquisition response of 1 indicates that the model can be sent (i.e., the first model acquisition response). Again, for example, a model acquisition response of 1 indicates that the model cannot be sent (i.e., the second model acquisition response); a model acquisition response of 0 indicates that the model can be sent (i.e., the first model acquisition response). The second model acquisition response may also include, but is not limited to, the request identifier of the first model acquisition request, etc., which are not limited here.

[0188] This application does not limit how the determination of whether a model can be permitted to be transmitted is made based on the first model acquisition request. Optionally, the determination can be made based on the expected model input information and / or expected model output information in the first model acquisition request. For example, the model can be permitted to be transmitted if the expected model input information and / or expected model output information does not involve privacy data of the communication network, and the model cannot be permitted to be transmitted if the expected model input information and / or expected model output information involves privacy data of the communication network. The privacy data of the communication network can be user data, such as user location information. The privacy data of the communication network can also be air interface information, such as reference signal receiving power (RSRP). This can prevent privacy data leakage and improve data security.

[0189] For example, if the expected model input information and / or expected model output information are deemed authorized information, then the model is permitted to transmit externally; if the expected model input information and / or expected model output information are not deemed authorized information, then the model is not permitted to transmit externally. The authorized information can be protocol-defined or vendor-defined, or configured wireless information that can be transmitted externally. This avoids the transmission of unauthorized information and provides data security.

[0190] In this embodiment, the first communication device for determining whether a model can be distributed externally can be a network device such as an access network device, a core network device, or an OAM device. If the first communication device is not a network device, but rather a terminal device, a second model acquisition request can be sent to the network device, thereby determining whether the model can be distributed externally through the network device. Here, the network device can be the aforementioned second communication device, and the second model acquisition request can be referred to the foregoing, and will not be repeated here. If the first communication device is a network device, a second model acquisition request can also be sent to another network device. For example, if the first communication device is an access network device, another network device can act as the second communication device, such as a core network device or an OAM device, thereby determining whether the model can be distributed externally through the second communication device.

[0191] In some feasible implementations, the method may further include: the first communication device sending a third model acquisition response to the second communication device. Correspondingly, the second communication device receives the third model acquisition response from the first communication device.

[0192] The third model acquisition response indicates whether the model can be sent; it can be understood as a permission judgment result. For example, a third model acquisition response of 0 indicates that the model can be sent, and a third model acquisition response of 1 indicates that the model cannot be sent. As another example, a third model acquisition response of 1 indicates that the model can be sent, and a third model acquisition response of 0 indicates that the model cannot be sent.

[0193] In this embodiment, the second communication device can be a communication device that stores the model, or it can be a communication device that does not store the model. This method can be applied to situations where the first communication device does not store the model. In this case, the first communication device can obtain a response through the third model and inform the second communication device whether the model can be sent.

[0194] For example, if the first communication device is a CN device or an OAM device, the second communication device is a BS, and the communication device storing the model is a UE, then the first communication device can send a third model acquisition response to the BS to indicate whether the model can be sent. If the third model acquisition response indicates that the model can be sent, a model transmission instruction can be sent to the UE to instruct the UE to send the model. The UE can send the model through the BS, and the BS can then send the model through the first communication device. The first communication device can then send the model to an OTT device. Alternatively, the model can be sent directly; for example, the UE can directly send the model to the OTT device, and the BS can directly send the model to the OTT device. If the third model acquisition response indicates that the model cannot be sent, the UE may choose not to send the model.

[0195] If the CN device or OAM device determines whether the model can be transmitted externally, and if the first communication device is not a CN device or OAM device, then in some feasible implementations, the method may further include: the CN device or OAM device sending a third model acquisition response to the first communication device. Alternatively, it can be described as a second communication device sending a third model acquisition response to the first communication device.

[0196] The first communication device can be a terminal device or an access network device. If the first communication device is a terminal device, the second communication device can be a network device such as an access network device, a core network device, or an OAM device. If the first communication device is an access network device, the second communication device can be a network device other than an access network device, such as a core network device or an OAM device.

[0197] For example, if the first communication device is a UE and the communication device storing the model is also a UE, then the CN device or OAM device can send a third model acquisition response to the UE to indicate whether the model can be sent. If the third model acquisition response indicates that the model can be sent, the UE can send the model and / or the first model acquisition response to the OTT device. Alternatively, if the third model acquisition response indicates that the model cannot be sent, the UE can send a second model acquisition response to the OTT device without sending the model.

[0198] For example, if the first communication device is a BS (Browser Base Station), and the communication device storing the model is also a BS, then the CN (Network Address Translation) device or the OAM (Online Availability) device can send a third model retrieval response to the BS to indicate whether the model can be sent. If the third model retrieval response indicates that the model can be sent, the BS can send the model and / or the first model retrieval response to the OTT (Online Telephone / Telephone) device. Alternatively, if the third model retrieval response indicates that the model cannot be sent, the BS can send a second model retrieval response to the OTT device without sending the model.

[0199] It is understood that when the first communication device stores a model, it can send the model to the OTT device. When the first communication device does not store a model, it can send a third model acquisition response via a second communication device, such that if the third model acquisition response indicates that a model can be sent, the communication device storing the model can send the model to the first communication device. This communication device can send the model directly to the OTT device, or it can send the model to the OTT device via the second communication device and / or the first communication device; this is not limited here. The first model acquisition response or the second model acquisition response can be sent by the first communication device to the OTT device, or it can be sent by the second communication device after receiving the third model acquisition response.

[0200] In some feasible implementations, the method further includes: the first communication device sending a second model acquisition request to the second communication device. Correspondingly, the second communication device receives the second model acquisition request from the first communication device. The second model acquisition request is used to request the acquisition of a model, as described above, and will not be repeated here.

[0201] It is understood that when the first communication device stores the model, it can send the model to the OTT device. When the first communication device does not store or register a model, it can send a second model retrieval request to the second communication device, either to instruct the communication device storing the model to send it, or to determine whether the model can be sent. The communication device storing the model can be the second communication device, or it can be a device other than the first or second communication device, such as a terminal device.

[0202] In some feasible implementations, the method may further include: the second communication device sending a third model acquisition response to the first communication device. Correspondingly, the first communication device receives the third model acquisition response from the second communication device.

[0203] The third model acquisition response indicates whether the model can be sent. Thus, after the second communication device determines whether the model can be sent, it can send a third model acquisition response to the first communication device. When the third model acquisition response indicates that the model can be sent, the communication device storing the model can send the model to the OTT device. For example, if the communication device storing the model is the first communication device, it can directly send the model to the OTT device. If the communication device storing the model is the second communication device, it can directly send the model to the OTT device, or it can send the model to the OTT device through the first communication device. If the communication device storing the model is neither the first nor the second communication device, it can directly send the model to the OTT device, or it can send the model to the OTT device through the first and / or the second communication device; this is not limited here. When the third model acquisition response indicates that the model cannot be sent, the first communication device, the second communication device, and the communication device storing the model may not send the model to the OTT device.

[0204] In some feasible implementations, the method may further include: the second communication device sending a model transmission command to the terminal device. Correspondingly, the terminal device receives the model transmission command from the second communication device.

[0205] The model transmission instruction is used to instruct the transmission of the model. This method is applicable when model transmission is permitted, and the terminal device is a communication device storing the model. That is, when the first or second communication device determines that the model can be transmitted, it can instruct the terminal device storing the model to send the model to the OTT device. The terminal device can send the model directly to the OTT device, or it can send the model to the OTT device through other communication devices such as the first and / or second communication devices, etc., without limitation.

[0206] In some feasible implementations, the method may further include: the terminal device sending a model to the second communication device. Correspondingly, the second communication device receives the model from the terminal device.

[0207] This method is applicable when model sharing is permitted, and the terminal device is a communication device that stores the model. That is, when the first or second communication device determines that model sharing is permitted, the second communication device can receive the model from the terminal device storing the model and send it to the OTT device. The second communication device can send the model directly to the OTT device, or it can send the model to the OTT device through the first communication device or other communication devices, etc., without limitation.

[0208] In some feasible implementations, the method may further include: the second communication device sending a model acquisition failure instruction to the first communication device. Correspondingly, the first communication device receives the model acquisition failure instruction from the second communication device. The model acquisition failure instruction indicates that model acquisition has failed, or that no suitable model exists. The lack of a suitable model could be due to a registered model in the second communication device not meeting the model requirements in the first model acquisition request, or it could be due to a model meeting the model requirements in the first model acquisition request not being allowed to be shared. Thus, even if the model is allowed to be shared, the OTT device will not receive the model corresponding to the first model acquisition request.

[0209] In some feasible implementations, the method may further include: the second communication device sending a model to the first communication device. Correspondingly, the first communication device receives the model from the second communication device.

[0210] This method is applicable when model sharing is permitted, and the model can be stored in a second communication device or terminal device. Thus, the second communication device can send the model to an OTT device via the first communication device. The model sent from the second communication device to the first communication device can be a model stored in the second communication device, or a model stored in a communication device other than the first and second communication devices.

[0211] In the method shown in Figure 3, after the first communication device receives the first model acquisition request from the OTT device, if the first model acquisition request requires that the acquired model be allowed to be shared externally, it can send a first model acquisition response to the OTT device to indicate that the model can be sent. In this way, the OTT device can obtain the requested model, and since the model is allowed to be shared externally, the security of data within the communication network can be improved.

[0212] In some feasible implementations, the method may further include: the OTT device training or inferring the model based on the response obtained from the first model, and obtaining model feedback information. For example, the model is trained or inferred based on the model's attribute information in the response obtained from the first model, such as external attribute information, to obtain training results or inference results; the model's performance is determined based on the training results or inference results, and it can also be determined whether the model needs to be retransmitted. It can be understood that the OTT device obtaining model feedback information based on the model attribute information in the response obtained from the first model is beneficial to improving the model's performance.

[0213] In some feasible implementations, the method may further include: the OTT device sending model feedback information to the first communication device. Correspondingly, the first communication device receives the model feedback information from the OTT device.

[0214] The model feedback information is used to indicate the model's output. This application does not limit the content of the model feedback information; for example, it may include a request identifier for the first model acquisition request. The request identifier for the first model acquisition request can be referred to the foregoing and will not be repeated here.

[0215] In some feasible implementations, model feedback information may include model performance indicators.

[0216] The model performance indicator is used to indicate whether the model's performance meets the model performance requirements. For example, a model performance indicator of 0 indicates that the model's performance meets the requirements, while a model performance indicator of 1 indicates that the model's performance does not meet the requirements. The model performance requirements are as described above and will not be repeated here.

[0217] If the model performance indicator is used to indicate that the model's performance meets the model performance requirements, no other additional operations need to be performed. If the communication device storing the model is a first communication device, and the model's performance does not meet the model performance requirements, no other additional operations need to be performed, or the model may be retrained or a new model may be selected. For example, if the first communication device is a UE, BS, CN device, or OAM device, after receiving model feedback information from an OTT device, the first communication device may not perform any additional operations, or it may retrain or select a new model.

[0218] In some feasible embodiments, if the communication device storing the model is not the first communication device, the method may further include: when a model performance indicator indicates that the model's performance does not meet the model performance requirements, the first communication device sends a model update instruction or switching instruction to the communication device storing the model, or the first communication device sends model feedback information to the second communication device. Correspondingly, the communication device storing the model receives the model update instruction or switching instruction from the first communication device. Alternatively, the second communication device receives the model feedback information from the first communication device.

[0219] In this method, the communication device for storing the model can be any communication device other than the first communication device. For example, if the first communication device is a core network device or an OAM device, the communication device for storing the model can be an access network device or a terminal device; if the first communication device is an access network device, the communication device for storing the model can be a terminal device, a core network device, an OAM device, etc. The model update instruction is used to instruct the model to be updated and can be called a model update instruction. The model switching instruction is used to instruct the model to be switched and can be called a model switching instruction. The update instruction or switching instruction can include target metrics, such as the minimum accuracy of the model. The update instruction can also include the model identifier, etc. This application does not limit the content of the update instruction and the switching instruction. Thus, if the model's performance does not meet the model performance requirements, the model can be retrained or a new model can be transmitted so that the OTT device can obtain a model that meets the model performance requirements.

[0220] The communication device storing the model can receive model update or switching instructions sent by the first communication device through the second communication device, or it can directly receive model update or switching instructions sent by the first communication device, etc., without limitation. For example, the first communication device is a CN device or an OAM device, and the UE stores the model. After receiving model feedback information from the OTT device, the first communication device can send model update or switching instructions to the UE through the BS. As another example, the first communication device is a CN device or an OAM device, and the BS stores the model. After receiving model feedback information from the OTT device, the first communication device can send model update or switching instructions to the BS.

[0221] In some feasible implementations, the model feedback information may include a re-request indication. This re-request indication is used to indicate whether a new model should be transmitted. For example, a re-request indication of 0 indicates the transmission of a new model, and a re-request indication of 1 indicates no transmission of a new model. As another example, a re-request indication of 1 indicates the transmission of a new model, and a re-request indication of 0 indicates no transmission of a new model.

[0222] It is understood that when a re-request instruction is used to indicate that a new model should not be transmitted, no other additional operations need to be performed, such as not needing to respond to the OTT device again. When a re-request instruction is used to indicate that a new model should be transmitted, a response from the OTT device is required to transmit the new model. For example, when a re-request instruction is used to indicate that a new model should be transmitted, the model needs to be reselected or retrained to switch or update the model, and a model acquisition response may also be transmitted to the OTT device. This acquisition response can be referred to the description of the first model acquisition response, which will not be repeated here. When a re-request instruction is used to indicate that a new model should not be transmitted, the model does not need to be reselected or retrained, and the model information does not need to be transmitted to the OTT device.

[0223] The model feedback information described above is merely an example, and this application does not limit it.

[0224] In some feasible implementations, the method may further include: the first communication device analyzing the model feedback information.

[0225] For example, the model being fed back can be determined based on the request identifier in the model feedback information. Another example is determining whether the model's performance meets the performance requirements based on the model performance indication in the model feedback information. If the model's performance meets the requirements, no additional operations are performed. If the model's performance does not meet the requirements, the model feedback information can be forwarded to a second communication device or the communication device storing the model, or a model switching or update command can be sent to the second communication device or the communication device storing the model. Yet another example is determining whether a new model needs to be transmitted based on the re-request indication in the model feedback information. If a new model does not need to be transmitted, no additional operations are performed. If a new model needs to be transmitted, the model can be switched or reselected, or a model retrieval response can be sent to the OTT device. In this way, model information can be obtained based on the model feedback information, and it can be determined whether to respond to it.

[0226] The process by which the first communication device analyzes the model feedback information can be understood as a feedback analysis process of the first communication device, which helps improve the model's performance. When the second communication device receives the model feedback information from the first communication device, the second communication device can also analyze the model feedback information.

[0227] In this application, the model feedback information received by the first communication device from the OTT device can be referred to as the first model feedback information, and the model feedback information received by the second communication device from the first communication device can be referred to as the second model feedback information. The second model feedback information can be all or part of the information in the first model feedback information.

[0228] For example, the first communication device can forward all the first model feedback information to the second communication device. Thus, the first and second communication devices can perform the same or different analyses on the first model feedback information. The same analysis could involve not performing any additional operations if the model performance indicator indicates that the model's performance meets the requirements, not performing any additional operations if the re-request indicator indicates that no new model needs to be transmitted, and sending a model transmission command to the device storing the model if the re-request indicator indicates that a new model needs to be transmitted. The different analysis could involve the first communication device determining that it needs to forward the model feedback information to the second communication device if the model performance indicator indicates that the model's performance does not meet the requirements. The second communication device, in turn, determining to send a model switching command or an update command to the communication device storing the model if the model performance indicator indicates that the model's performance does not meet the requirements.

[0229] For example, the first communication device can analyze a portion of the model feedback, and the remaining model feedback information can be sent to the second communication device as second model feedback information, so that the second communication device can analyze it. In this way, the second communication device can analyze the model feedback information of the OTT device, and based on this, switch or update the model, which helps to improve the model's performance.

[0230] In some feasible implementations, the method may further include: the second communication device sending a model update instruction or switching instruction, and / or a model retransmission instruction to the terminal device. Correspondingly, the terminal device receives the model update instruction or switching instruction, and / or the model retransmission instruction from the second communication device.

[0231] The model update command instructs users to update the model, while the model switch command instructs users to switch models. The model retransmission command instructs users to retransmit the model, which can be a retrained and updated model or a newly selected model. Thus, updating the model improves its performance. Switching the model enhances the performance and accuracy of the models used by OTT devices.

[0232] When the first or second communication device determines that a new model needs to be transmitted, the communication device storing the new model can send a fourth model retrieval response and / or the model to the OTT device. This communication device can send the fourth model retrieval response and / or the model directly to the OTT device, or it can send the fourth model retrieval response and / or the model through the first and / or second communication devices, etc., without limitation. The description of the fourth model retrieval response can be referred to the description of the first model retrieval response, and will not be repeated here.

[0233] The communication methods provided in this application are described below with reference to different first and second communication devices, and these communication methods should be included in the method shown in Figure 3. Please refer to the communication methods described in Figures 4 to 7 respectively. In Figure 4, the first communication device can be a core network device or an OAM device, and the second communication device can be an access network device. Also in Figure 4, the model can be stored in the terminal device, but not in network devices such as access network devices, core network devices, or OAM devices. The method shown in Figure 4 is applicable to scenarios where the terminal device cannot directly communicate with the OTT device and needs to interact with the OTT device through network devices such as access network devices, core network devices, or OAM devices.

[0234] As shown in Figure 4, the communication method may include, but is not limited to, the following steps:

[0235] Optionally, in step S400, the terminal device sends a model registration request to the access network device.

[0236] Accordingly, the access network device receives the model registration request from the terminal device. The model registration request indicates a request to register a model. The model registration request may include, but is not limited to, model attribute information, as described above and will not be repeated here. After receiving the model registration request, the access network device may perform steps not shown in Figure 4, such as storing model attribute information and assigning a model identifier to the model.

[0237] Step S401: The OTT device sends a first model acquisition request to the core network device or OAM device.

[0238] Accordingly, the core network device or OAM device receives the first model acquisition request from the OTT device. The first model acquisition request is used to request the acquisition of a model, and can be referred to the description in step S301, which will not be repeated here. The first model acquisition request may include, but is not limited to, a request identifier for the model acquisition request, model requirement information, etc.

[0239] Optionally, in step S402, the core network device or OAM device determines whether the model can be allowed to be sent out based on the first model acquisition request.

[0240] Whether the model can be shared externally can be determined using the aforementioned method, and will not be repeated here. Optionally, if the core network device or OAM device determines that the model can be shared externally, the method may further include: step S403, whereby the core network device or OAM device sends a second model acquisition request to the access network device. Correspondingly, the access network device receives the second model acquisition request from the core network device or OAM device.

[0241] In this embodiment, the second model acquisition request is used to request the acquisition of a model, as described in step S301, and will not be repeated here. For example, the second model acquisition request may include a request identifier for the model acquisition request and / or model requirement information, etc., which will not be repeated here. The second model acquisition request may be the same as the first model acquisition request, or may be a part of the content of the first model acquisition request.

[0242] If the core network device or OAM device determines that the model cannot be transmitted, the method may optionally further include: step S404, whereby the core network device or OAM device sends a second model acquisition response to the OTT device. Correspondingly, the OTT device receives the second model acquisition response from the core network device or OAM device. The second model acquisition response indicates that the model cannot be transmitted.

[0243] Optionally, in step S405, the access network device selects a model based on the second model acquisition request. That is, the second communication device selects a model. In this way, a model that meets the requirements can be selected, thereby improving the accuracy of the transmission model.

[0244] After selecting a model, the method may optionally include: step S406, whereby the access network device sends a model transmission instruction to the terminal device. That is, the second communication device sends a model transmission instruction to the terminal device. Correspondingly, the terminal device receives the model transmission instruction from the access network device. If no suitable model is found, the method may optionally include: step S407, whereby the access network device sends a model acquisition failure instruction to the core network device or OAM device. Correspondingly, the core network device or OAM device receives the model acquisition failure instruction from the access network device.

[0245] Optionally, after the core network device or OAM device receives the model acquisition failure instruction, the method may further include a step not shown in Figure 4: the core network device or OAM device sends the model acquisition failure instruction to the OTT device. Correspondingly, the OTT device receives the model acquisition failure instruction from the core network device or OAM device.

[0246] The model transmission command is used to indicate the transmission model. The model acquisition failure command indicates that model acquisition failed, or that no suitable model is available. The lack of a suitable model could be due to either a registered model in the access network device not meeting the model requirements in the model acquisition request, or a model meeting the requirements not being able to be transmitted. Therefore, after determining that no suitable model is available, a model acquisition failure command can be sent to the OTT device.

[0247] Optionally, after the terminal device receives the model transmission instruction, it can execute step S408, in which the terminal device sends the model to the access network device. Correspondingly, the access network device receives the model from the terminal device.

[0248] Optionally, after step S408, the method may further include: step S409, whereby the access network device sends a model to the core network device or the OAM device. Correspondingly, the core network device or the OAM device receives the model from the access network device.

[0249] If the OAM device determines to the OTT device that the model can be transmitted externally, step S410 can be executed, whereby the core network device or the OAM device sends a first model acquisition response to the OTT device. Correspondingly, the OTT device receives the first model acquisition response from the core network device or the OAM device.

[0250] The first model acquisition response is used to indicate that the model can be sent. Referring to the first model acquisition response mentioned above, in Figure 4, the first model acquisition response includes at least the model, and may also include the request identifier of the first model acquisition request, the request completion indication, model attribute information, etc., which will not be elaborated here.

[0251] Optionally, in step S411, the OTT device trains or infers the model to obtain the first model feedback information.

[0252] Optionally, in step S412, the OTT device sends the first model feedback information to the core network device or OAM device. Correspondingly, the core network device or OAM device receives the first model feedback information from the OTT device.

[0253] Optionally, in step S413, the core network equipment or OAM equipment analyzes the feedback information of the first model to obtain the feedback information of the second model.

[0254] Optionally, in step S414, the core network device or OAM device sends the second model feedback information to the access network device. Correspondingly, the access network device receives the second model feedback information from the core network device or OAM device.

[0255] Optionally, in step S415, the access network device analyzes the feedback information of the second model to obtain the feedback information of the third model.

[0256] The first, second, and third model feedback information mentioned above can be referred to in the foregoing description of model feedback information, and may include at least one or more of the following: request identifier, model performance indicator, and re-request indicator. Any two of the first, second, and third model feedback information may be the same or different, and this is not limited here. For example, the steps for analyzing the first model feedback information can refer to the foregoing description of the analysis of model feedback information by the first communication device, and the steps for analyzing the second model feedback information can refer to the foregoing description of the analysis of model feedback information by the second communication device.

[0257] In the case where the model performance indicator in the third model feedback information indicates that the model performance does not meet the model performance requirements, and / or the re-request indicator indicates that a new model should be transmitted, optionally, step S416 is executed: the access network device sends a model update instruction or switching instruction, and / or a model retransmission instruction to the terminal device. Correspondingly, the terminal device receives the model update instruction or switching instruction, and / or the model retransmission instruction from the access network device.

[0258] The model retransmission instruction is used to instruct the model to be transmitted again. This model can be an updated model obtained through retraining, or a newly selected model, etc. Optionally, before sending the model switching instruction, the method may further include: the access network device can determine the switched model to be transmitted based on third model feedback information and / or a second model acquisition request, etc.

[0259] In this application, the updated or switched model can be referred to as the new model. Optionally, after the terminal device updates or switches the model, step S417 can be executed, in which the terminal device sends the new model to the access network device. Correspondingly, the access network device receives the model from the terminal device.

[0260] Optionally, after step S417, step S418 can be executed, in which the access network device sends a new model to the core network device or OAM device. Correspondingly, the core network device or OAM device receives the model from the access network device.

[0261] Optionally, in step S419, the core network device or OAM device sends a fourth model acquisition response to the OTT device. Correspondingly, the OTT device receives the fourth model acquisition response from the core network device or OAM device.

[0262] The fourth model acquisition response shown in Figure 4 can be used to indicate that a model can be sent, and the fourth model acquisition response includes a new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here. It can be understood that after step S419, the OTT device can train or infer the model, and if the model performance does not meet the model performance requirements, iteratively execute the steps not shown in Figure 4. These steps are as described in steps S411 to S419 above, and will not be repeated here, or the fourth model acquisition response may indicate that the model is not sent. In some feasible implementations, if the model performance indication is used to indicate that the model performance meets the model performance requirements, the steps after step S412 may not be executed, such as steps S415 to S419.

[0263] In the method shown in Figure 4, after receiving a first model acquisition request from an OTT device, the first communication device first determines whether the requested model can be shared externally. If the model can be sent, a model transmission command is sent to the communication device storing the model (i.e., the terminal device) via the second communication device, causing the terminal device to send the model to the OTT device through both the first and second communication devices. In this way, the OTT device can obtain the requested model, and since the model is permitted to be shared externally, the security of data within the communication network can be improved. The second communication device can also be used to select a model. If the model performance does not meet the requirements, the model can be transmitted again; this model can be a retrained model or a newly selected model, which helps to improve model performance.

[0264] Figure 4 uses one terminal device as an example, but in reality, it can correspond to one or more terminal devices. For example, the terminal devices for steps S417 and S408 can be the same or different. Step S400 can be executed individually by one or more terminal devices, and the execution time can be the same or different, etc., which is not limited here.

[0265] The steps performed by the access network device in Figure 4 can also be performed by the core network device or the OAM device. In other words, the functions of the access network device can be merged into those of the core network device or the OAM device, such as steps S400 and S405. Thus, the terminal device sends a model registration request to the core network device or the OAM device, which can then select a model based on the first model acquisition request.

[0266] Alternatively, the steps performed by the core network device or OAM device in Figure 4 can be executed by the BS. That is, the functions of the core network device or OAM device can be merged into the access network device, such as step S402. In this way, the access network device determines whether the model can be transmitted externally based on the first model acquisition request. If transmission is permitted, the access network device selects a model based on the first model acquisition request and can also send a model transmission instruction to the terminal device, enabling the terminal device to send the model to the OTT device through the access network device. This model can be carried in the first model acquisition response. If transmission is not permitted, the access network device sends a second model acquisition response to the OTT device.

[0267] In Figure 5, the first communication device can be a terminal device, and the second communication device can be an access network device, a core network device, or an OAM device, and the model can be stored in the terminal device. The method shown in Figure 5 is applicable to scenarios where the terminal device can communicate directly with the OTT device. Optionally, the OTT device and the terminal device can belong to the same manufacturer, so the OTT device can directly send a first model acquisition request to the terminal device manufactured by its manufacturer. Alternatively, the OTT device can be a device that has pre-applied for model storage services with the terminal device, without applying for model storage services from the network side, so the OTT device can store the model through the terminal device. As shown in Figure 5, this communication method may include, but is not limited to, the following steps:

[0268] Step S501: The OTT device sends a first model acquisition request to the terminal device. Correspondingly, the terminal device receives the first model acquisition request from the OTT device.

[0269] Step S502: The terminal device sends a second model acquisition request to the core network device, OAM device, or access network device. Correspondingly, the core network device, OAM device, or access network device receives the second model acquisition request from the terminal device.

[0270] The first and second model acquisition requests described above are both used to request model acquisition, as previously mentioned, and will not be repeated here. The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information from the first model acquisition request. Alternatively, the second model acquisition request can include some information from the first model acquisition request. For example, the second model acquisition request may include the expected input information and / or expected output information from the first model acquisition request. The first model acquisition request may also include, in addition to the expected input information and / or expected output information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc., for the expected model.

[0271] Optionally, in step S503, the core network device, OAM device, or access network device determines whether the model can be allowed to be sent out based on the second model acquisition request.

[0272] In other words, the second communication device determines whether the model can be transmitted. The method for determining whether the model can be transmitted can be referred to the previous method and will not be repeated here. The third model acquisition response, indicating whether the model can be transmitted, can be sent in step S504, that is, the core network device, OAM device, or access network device sends the third model acquisition response to the terminal device. Correspondingly, the terminal device receives the third model acquisition response from the core network device, OAM device, or access network device. The third model acquisition response indicates whether the model can be transmitted. If not, step S505 can be executed, and the terminal device sends a second model acquisition response to the OTT device. Correspondingly, the OTT device receives the second model acquisition response from the terminal device. The second model acquisition response indicates that the model cannot be transmitted. If yes or no, step S506 can be executed, and the terminal device selects a model according to the first model acquisition request. That is, the first communication device selects a model.

[0273] If model selection fails, such as when no suitable model is available or the suitable model cannot be sent, the following steps (not shown in Figure 5) can be executed: the terminal device sends a model acquisition failure command to the OTT device to indicate that model acquisition has failed. Correspondingly, the OTT device receives the model acquisition failure command from the terminal device. Optionally, after model selection, step S507 can be executed: the terminal device sends a first model acquisition response to the OTT device. Correspondingly, the OTT device receives the first model acquisition response from the terminal device. The first model acquisition response indicates that a model can be sent. Referring to the aforementioned first model acquisition response, in Figure 5, this first model acquisition response includes at least the model, and may also include a request identifier for the first model acquisition request, a request completion indication, model attribute information, etc., which will not be elaborated further here.

[0274] Optionally, in step S508, the OTT device trains or infers the model to obtain the first model feedback information.

[0275] Optionally, in step S509, the OTT device sends first model feedback information to the terminal device. Correspondingly, the terminal device receives the first model feedback information from the OTT device.

[0276] Optionally, in step S510, the terminal device analyzes the feedback information from the first model to obtain the feedback information from the second model.

[0277] The first and second model feedback information mentioned above can refer to the description of model feedback information above, and may include at least one or more of the following: request identifier, model performance indicator, and re-request indicator. The first and second model feedback information may be the same or different, and this is not limited here.

[0278] In the case where the model performance indicator in the second model feedback information is used to indicate that the model performance does not meet the model performance requirements, and / or the re-request indicator is used to indicate the transmission of a new model, optionally, step S511 is performed, and the terminal device updates the model or switches the model.

[0279] Optionally, the method may further include: step S512, whereby the terminal device sends a fourth model acquisition response to the OTT device. Correspondingly, the OTT device receives the fourth model acquisition response from the terminal device.

[0280] The fourth model acquisition response shown in Figure 5 can be used to indicate that a model can be sent, and the fourth model acquisition response includes a new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here. It can be understood that after step S512, the OTT device can train or infer the model, and iteratively execute the steps not shown in Figure 5 if the model performance does not meet the model performance requirements. These steps may be as shown in steps S508 to S512, or the fourth model acquisition response may indicate that the model is not sent. In some feasible implementations, if the model performance indication is used to indicate that the model performance meets the model performance requirements, the steps after step S509 may not be executed.

[0281] Figure 5 also uses a single terminal device as an example, but in reality, it can correspond to one or more terminal devices.

[0282] In the method shown in Figure 5, after receiving a first model acquisition request from an OTT device, the first communication device sends a second model acquisition request to the second communication device, so that the second communication device can determine whether the model requested in the first model acquisition request can be sent. If the model can be sent, the second communication device sends a third model acquisition response to the first communication device, instructing the first communication device storing the model to send the model, so that the terminal device sends the model to the OTT device. In this way, the OTT device can obtain the requested model, and the model is allowed to be sent out, which can improve the security of data within the communication network. The terminal device can also be used to select a model. If the model performance does not meet the model performance requirements, the model can be transmitted again; this model can be a retrained model or a newly selected model, which helps to improve model performance.

[0283] In Figure 6, the first communication device (BS) can be an access network device, and the second communication device can be a core network device or an OAM device. The model can be stored in the access network device. The method shown in Figure 6 is applicable to scenarios where the access network device can communicate directly with the OTT device. The access network device can provide services such as model storage services, model selection, and model generation for obtaining responses. As shown in Figure 6, this communication method may include, but is not limited to, the following steps:

[0284] Step S601: The OTT device sends a first model acquisition request to the access network device. Correspondingly, the access network device receives the first model acquisition request from the OTT device.

[0285] Step S602: The access network device sends a second model acquisition request to the core network device or OAM device. Correspondingly, the core network device or OAM device receives the second model acquisition request from the access network device.

[0286] The first and second model acquisition requests described above are both used to request model acquisition, as previously mentioned, and will not be repeated here. The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information in the first model acquisition request. Alternatively, the second model acquisition request can include only some of the information in the first model acquisition request. For example, the second model acquisition request might include the expected input and / or expected output information from the first model acquisition request, but exclude other information. The first model acquisition request, in addition to the expected input and / or expected output information, may also include expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc., beyond the expected input and / or expected output information. The request identifier for the second model acquisition request can be the same as the request identifier for the first model acquisition request.

[0287] Optionally, in step S603, the core network device or OAM device determines whether the model can be allowed to be sent out based on the second model acquisition request.

[0288] In other words, the second communication device determines whether the model can be transmitted. The method for determining whether the model can be transmitted can be referred to the previous method and will not be repeated here. The third model acquisition response, indicating whether the model can be transmitted, can be sent in step S604, that is, the core network device or OAM device sends a third model acquisition response to the access network device. Correspondingly, the access network device receives the third model acquisition response from the core network device or OAM device. The third model acquisition response indicates whether the model can be transmitted. If not, step S605 can be executed, and the access network device sends a second model acquisition response to the OTT device. Correspondingly, the OTT device receives the second model acquisition response from the access network device. The second model acquisition response indicates that the model cannot be transmitted. If yes or yes, step S606 can be executed, and the access network device selects a model according to the first model acquisition request. That is, the first communication device selects a model.

[0289] If model selection fails, such as when no suitable model is available or the suitable model cannot be sent, the following steps (not shown in Figure 6) can be executed: the access network device sends a model acquisition failure command to the OTT device to indicate that model acquisition has failed. Correspondingly, the OTT device receives the model acquisition failure command from the access network device. Optionally, after model selection, step S607 can be executed: the access network device sends a first model acquisition response to the OTT device. Correspondingly, the OTT device receives the first model acquisition response from the access network device. The first model acquisition response indicates that a model can be sent. Referring to the aforementioned first model acquisition response, in Figure 6, this first model acquisition response includes at least the model, and may also include a request identifier for the first model acquisition request, a request completion indication, model attribute information, etc., which will not be elaborated further here.

[0290] Optionally, in step S608, the OTT device trains or infers the model to obtain feedback information from the first model.

[0291] Optionally, in step S609, the OTT device sends first model feedback information to the access network device. Correspondingly, the access network device receives the first model feedback information from the OTT device.

[0292] Optionally, in step S610, the access network device analyzes the feedback information of the first model to obtain the feedback information of the second model.

[0293] The first and second model feedback information mentioned above can refer to the description of model feedback information above, and may include at least one or more of the following: request identifier, model performance indicator, and re-request indicator. The first and second model feedback information may be the same or different, and this is not limited here.

[0294] In the case where the model performance indicator in the second model feedback information is used to indicate that the model performance does not meet the model performance requirements, and / or the re-request indicator is used to indicate the transmission of a new model, optionally, step S611 is performed, and the access network device updates the model or switches the model.

[0295] Optionally, the method may further include: step S612, whereby the access network device sends a fourth model acquisition response to the OTT device. Correspondingly, the OTT device receives the fourth model acquisition response from the access network device.

[0296] The fourth model acquisition response shown in Figure 6 can be used to indicate that a model can be sent, and the fourth model acquisition response includes the new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here. It can be understood that after step S612, the OTT device can train or infer the model, and if the performance of the new model does not meet the model performance requirements, iteratively execute the steps not shown in Figure 6. These steps may be as shown in steps S608 to S612, or the fourth model acquisition response may indicate that the model is not sent. In some feasible implementations, if the model performance indication is used to indicate that the model performance meets the model performance requirements, the steps after step S609 may not be executed.

[0297] In the method shown in Figure 6, after receiving a first model acquisition request from an OTT device, the first communication device sends a second model acquisition request to the second communication device, so that the second communication device can determine whether the model requested in the first model acquisition request can be sent. If the model can be sent, the second communication device sends a third model acquisition response to the first communication device, instructing the first communication device storing the model to send the model, so that the terminal device sends the model to the OTT device. In this way, the OTT device can obtain the requested model, and the model is allowed to be sent out, which can improve the security of data within the communication network. The access network device can also be used to select a model. If the model performance does not meet the model performance requirements, the model can be transmitted again; this model can be a retrained model or a newly selected model, which helps to improve model performance.

[0298] In Figure 7, the first communication device can be an access network device, a core network device, or an OAM device. A second communication device is not required, and the model can be stored in the first communication device. This first communication device can provide model storage services, model selection, and services for generating models and obtaining responses. As shown in Figure 7, this communication method may include, but is not limited to, the following steps:

[0299] Step S701: The OTT device sends a first model acquisition request to the first communication device. Correspondingly, the first communication device receives the first model acquisition request from the OTT device.

[0300] The first model retrieval request is used to request the retrieval of the model, as described above, and will not be repeated here.

[0301] Optionally, in step S702, the first communication device determines whether the model can be sent out based on the first model acquisition request.

[0302] The method for determining whether a model can be shared externally can be referred to the previous method and will not be repeated here. If the model is not allowed to be shared externally, the step not shown in Figure 7 can be executed: the first communication device sends a second model acquisition response to the OTT device. Correspondingly, the OTT device receives the second model acquisition response from the first communication device. The second model acquisition response indicates that the model cannot be sent. If the model is allowed to be shared externally, step S703 can be executed: the first communication device selects a model based on the first model acquisition request.

[0303] If model selection fails, such as due to the absence of a suitable model or the inability to send a suitable model, the following steps (not shown in Figure 7) can be executed: the first communication device sends a model acquisition failure instruction to the OTT device, indicating that model acquisition has failed. Correspondingly, the OTT device receives the model acquisition failure instruction from the first communication device. After model selection, step S704 can be executed: the first communication device sends a first model acquisition response to the OTT device. Correspondingly, the OTT device receives the first model acquisition response from the first communication device. The first model acquisition response indicates that a model can be sent. Referring to the aforementioned first model acquisition response, in Figure 7, this first model acquisition response includes at least the model, and may also include a request identifier for the first model acquisition request, a request completion indication, model attribute information, etc., which will not be elaborated further here.

[0304] Optionally, in step S705, the OTT device trains or infers the model to obtain the first model feedback information.

[0305] Optionally, in step S706, the OTT device sends first model feedback information to the first communication device. Correspondingly, the first communication device receives the first model feedback information from the OTT device.

[0306] Optionally, in step S707, the first communication device analyzes the feedback information of the first model to obtain the feedback information of the second model.

[0307] The first and second model feedback information mentioned above can refer to the description of model feedback information above, and may include at least one or more of the following: request identifier, model performance indicator, and re-request indicator. The first and second model feedback information may be the same or different, and this is not limited here.

[0308] In the case where the model performance indicator in the second model feedback information or the first model feedback information is used to indicate that the model performance does not meet the model performance requirements, and / or the re-request indicator is used to indicate the transmission of a new model, optionally, step S708 is performed, whereby the first communication device updates the model or switches the model.

[0309] Optionally, the method may further include: step S709, whereby the first communication device sends a fourth model acquisition response to the OTT device. Correspondingly, the OTT device receives the fourth model acquisition response from the first communication device.

[0310] The fourth model acquisition response shown in Figure 7 can be used to indicate that a model can be sent, and the fourth model acquisition response includes a new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here. It can be understood that after step S709, the OTT device can train or infer the model, and if the model performance does not meet the model performance requirements, iteratively execute the steps not shown in Figure 7. These steps may be as shown in steps S705 to S709, or the fourth model acquisition response may indicate that the model is not sent. In some feasible implementations, if the model performance indication is used to indicate that the model performance meets the model performance requirements, the steps after step S706 may not be executed.

[0311] In the method shown in Figure 7, after receiving a first model acquisition request from an OTT device, the first communication device determines whether the requested model can be sent. If the model can be sent, it can be sent to the OTT device. In this way, the OTT device can obtain the requested model, and since the model is permitted to be sent externally, the security of data within the communication network can be improved. The first communication device can also be used to select a model. If the model performance does not meet the requirements, the model can be transmitted again; this model can be a retrained model or a newly selected model, which helps to improve model performance.

[0312] The methods of the embodiments of this application have been described in detail above, and the apparatus of the embodiments of this application is provided below.

[0313] Please refer to Figure 8, which is a schematic diagram of a communication device provided in an embodiment of this application. The communication device may include a transceiver unit 801 and a processing unit 802. The transceiver unit 801 may be a device with signal input (receiving) or output (transmitting) capabilities, used for transmitting signals to other devices or other components within a device. The processing unit 802 may be a device with processing capabilities, and may include one or more processors for executing instructions (or code or programs), such as processing communication protocols and communication data.

[0314] The communication device can be a first communication device, an OTT device, a second communication device, or a terminal device. The first communication device can be an access network device, a core network device, an OAM device, or a terminal device, etc. The second communication device can be a different communication device from the first communication device, and the second communication device can communicate with the first communication device. When the first communication device is a core network device or an OAM device, the second communication device can be an access network device or a terminal device; when the first communication device is a terminal device, the second communication device can be a core network device, an OAM device, or an access network device; when the first communication device is an access network device, the second communication device can be a core network device or an OAM device, etc.

[0315] In the embodiments of this application, the functions implemented by the terminal device can be implemented either by components, parts, or circuits within the terminal device, or by chips or chip systems applicable to the terminal device. Similarly, the functions implemented by the access network device can be implemented either by components, parts, or circuits within the access network device, or by chips or chip systems applicable to the access network device. The functions implemented by the core network device can be implemented either by components, parts, or circuits within the core network device, or by chips or chip systems applicable to the core network device. The functions implemented by the OAM device can be implemented either by components, parts, or circuits within the OAM device, or by chips or chip systems applicable to the OAM device.

[0316] In one embodiment, when the communication device can be a first communication device, wherein:

[0317] The transceiver unit 801 is used to receive a first model acquisition request from an OTT device, wherein the first model acquisition request is used to request the acquisition of a model;

[0318] Processing unit 802 is used to determine whether the model can be allowed to be sent out based on the first model acquisition request;

[0319] The transceiver unit 801 is further configured to send a first model acquisition response to the OTT device when the model is permitted to be transmitted, the first model acquisition response being used to indicate that the model can be transmitted.

[0320] Optionally, the transceiver unit 801 is further configured to send a second model acquisition response to the OTT device if the model is not allowed to be transmitted, the second model acquisition response being used to indicate that the model cannot be transmitted.

[0321] Optionally, the processing unit 802 is used to determine whether the model can be allowed to be sent out based on the first model acquisition request.

[0322] Optionally, the transceiver unit 801 is further configured to send a second model acquisition request to the second communication device, the second model acquisition request being used to request the acquisition of the model.

[0323] Optionally, the transceiver unit 801 is further configured to receive a third model acquisition response from the second communication device, the third model acquisition response being used to indicate whether the model can be sent.

[0324] Optionally, the transceiver unit 801 is further configured to receive a model acquisition failure instruction from the second communication device, the model acquisition failure instruction being used to indicate that the model acquisition has failed.

[0325] Optionally, the transceiver unit 801 is also used to receive a model of the second communication device.

[0326] Optionally, the transceiver unit 801 is further configured to receive a model registration request from a communication device storing a model, the model registration request being used to request the registration of the model.

[0327] Optionally, the transceiver unit 801 is also used to receive model feedback information from the OTT device.

[0328] Optionally, the transceiver unit 801 is also used to send model feedback information to the second communication device.

[0329] Optionally, the model feedback information includes a model performance indicator, and the transceiver unit 801 is further configured to send an update instruction or a switching instruction for the model to the communication device storing the model when the model performance indicator indicates that the model's performance does not meet the model performance requirements.

[0330] Optionally, the model feedback information includes a re-request indication, which indicates whether a new model should be transferred.

[0331] Optionally, the processing unit 802 is also used to perform feedback analysis on the model feedback information.

[0332] Optionally, the processing unit 802 is further configured to select the model according to the first model acquisition request.

[0333] Optionally, the first model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and the model requirement information.

[0334] Optionally, the model requirement information includes at least one of the following: expected model input information, expected model output information, expected model function information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0335] Optionally, the first model acquisition response includes at least one of the following: a request identifier of the first model acquisition request, a request completion indication, the model, a model identifier of the model, and model attribute information; wherein, the request completion indication is used to indicate whether the first model acquisition request has been completed.

[0336] Optionally, the model attribute information includes at least one of the following: the model identifier, the model input information, the model output information, the model function information, the model operation information, the model performance information, the model update information, the model source information, the model external information, the model quantization information, and the model storage information.

[0337] Optionally, the transceiver unit 801 is further configured to send a fourth model acquisition response to the OTT device. The fourth model acquisition response may indicate that a model can be sent, and may include the new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here.

[0338] In one embodiment, when the communication device can be an OTT device, wherein:

[0339] The transceiver unit 801 is used to send a first model acquisition request to the first communication device, wherein the first model acquisition request is used to request the acquisition of a model;

[0340] The transceiver unit 801 is also configured to receive a first model acquisition response from the first communication device, wherein the first model acquisition response is used to indicate whether the model can be sent.

[0341] The processing unit 802 is used to train or infer the model based on the response obtained from the first model to obtain model feedback information.

[0342] Optionally, the transceiver unit 801 is further configured to receive a second model acquisition response from the first communication device, the second model acquisition response being used to indicate that the model cannot be sent.

[0343] Optionally, the transceiver unit 801 is also used to send model feedback information to the first communication device.

[0344] Optionally, the model feedback information includes at least one of the following: model performance indicator, re-request indicator, and request identifier of the first model acquisition request; wherein, the model performance indicator is used to indicate whether the performance of the model meets the model performance requirements, and the re-request indicator is used to indicate whether a new model should be transmitted.

[0345] Optionally, the first model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and the model requirement information.

[0346] Optionally, the first model requirement information includes at least one of the following: expected model input information, expected model output information, expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0347] Optionally, the first model acquisition response includes at least one of the following: a request identifier of the first model acquisition request, a request completion indication, the model, a model identifier of the model, and model attribute information; wherein, the request completion indication is used to indicate whether the first model acquisition request has been completed.

[0348] Optionally, the model attribute information includes at least one of the following: the model identifier, the model input information, the model output information, the model function information, the model operation information, the model performance information, the model update information, the model source information, the model external information, the model quantization information, and the model storage information.

[0349] Optionally, the transceiver unit 801 is further configured to receive a fourth model acquisition response from the first communication device, indicating that a model can be sent, and the fourth model acquisition response includes a new model. Optionally, the fourth model acquisition response may also include a request identifier of the first model acquisition request, a request completion indication, model attribute information, etc., which are not limited here.

[0350] In one embodiment, when the communication device can be a second communication device, wherein:

[0351] The transceiver unit 801 is used to receive a second model acquisition request from the first communication device, the second model acquisition request being used to request the acquisition of a model.

[0352] Optionally, the processing unit 802 is used to determine whether the model can be sent out based on the second model acquisition request.

[0353] Optionally, the transceiver unit 801 is also used to send a third model acquisition response to the first communication device, indicating whether the model is allowed to be transmitted.

[0354] Optionally, the processing unit 802 is used to select a model based on the second model acquisition request.

[0355] Optionally, the transceiver unit 801 is also used to send a model transmission instruction to the terminal device to instruct the transmission model.

[0356] Optionally, the transceiver unit 801 is also used to receive the model of the terminal device.

[0357] Optionally, the transceiver unit 801 is also used to send a model to the first communication device.

[0358] Optionally, the transceiver unit 801 is also configured to send a model acquisition failure instruction to the first communication device to indicate that the model acquisition has failed.

[0359] Optionally, the transceiver unit 801 is also used to receive model feedback information from the first communication device.

[0360] Optionally, the processing unit 802 is used to provide feedback information to the model feedback information.

[0361] Optionally, the transceiver unit 801 is also used to send model update instructions or switching instructions, and / or model retransmission instructions to the terminal device. The model update instruction is used to indicate updating the model, and the model switching instruction is used to indicate switching the model. The model retransmission instruction is used to indicate retransmission of the model, which can be the updated model obtained through retraining, or a newly selected model, etc.

[0362] Optionally, the transceiver unit 801 is also used to receive a model registration request from the terminal device for requesting model registration.

[0363] Optionally, the second model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and the model requirement information.

[0364] Optionally, the model requirement information includes at least one of the following: expected model input information, expected model output information, expected model function information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

[0365] The second model acquisition request can be the same as the first model acquisition request, meaning it can include all the information in the first model acquisition request. Alternatively, the second model acquisition request can include some of the information in the first model acquisition request. For example, the second model acquisition request might include the expected input information and / or expected output information from the first model acquisition request, but exclude other information. The first model acquisition request, in addition to the expected input and / or expected output information, may also include expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, model storage size requirements, etc. The description of the second model acquisition request can be referenced from that of the first model acquisition request, and the request identifier of the second model acquisition request can be the same as that of the first model acquisition request.

[0366] It should be noted that the implementation of each unit can also correspond to the description of any of the method embodiments in Figures 3 to 7. When the first communication device is a terminal device, the terminal device can store the model, and the transceiver unit 801 and processing unit 802 of the terminal device can refer to the description of the first communication device. When the first communication device is an access network device, core network device, or OAM device, the model can be stored in the terminal device, the first communication device, or the second communication device, etc., without limitation.

[0367] Please refer to Figure 9, which is a schematic diagram of another communication device provided in an embodiment of this application. As shown in Figure 9, the communication device may include a processor 111. The processor 111 may also be referred to as a processing unit, which can implement certain control functions. When the processor 111 is running, it causes the communication device to execute any of the methods described in Figures 3 to 7 in the embodiments of this application.

[0368] The communication device shown in Figure 9 may further include a storage medium 112, which may also be referred to as a storage unit or a memory. Instructions 114 are stored on the storage medium 112. These instructions 114 can be executed on the processor 111, causing the communication device to perform any of the methods described in Figures 3 to 7 of the embodiments of this application.

[0369] Optionally, the processor 111 may include instructions 113 that can be executed on the processor 111 to cause the communication device to perform any of the methods described in Figures 3 to 7 of the embodiments of this application.

[0370] The communication device may be a first communication device, a second communication device, an OTT device, or a terminal device, etc., used to implement the method described in the method embodiments. However, the scope of the device described in this application is not limited thereto; the communication device may be a standalone device or part of a larger device. For example, the communication device may be:

[0371] (1) An independent integrated circuit IC, or chip, or chip system or subsystem;

[0372] (2) A collection of one or more ICs, optionally, the collection of ICs may include a storage component for storing data and / or instructions;

[0373] (3) ASIC, such as modems;

[0374] (4) Modules that can be embedded in other devices.

[0375] Please refer to Figure 10, which is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For ease of explanation, Figure 10 only shows the main components of the terminal device. As shown in Figure 10, the terminal device includes a processor, a memory, a control circuit, an antenna, and input / output devices. The processor is mainly used to process communication protocols and communication data, control the entire terminal device, execute software programs, and process the data of the software programs. The memory is mainly used to store software programs and data. The radio frequency circuit is mainly used for the conversion between baseband signals and radio frequency signals and the processing of radio frequency signals. The antenna is mainly used for transmitting and receiving radio frequency signals in the form of electromagnetic waves. Input / output devices, such as touch screens, displays, and keyboards, are mainly used to receive user input data and output data to the user.

[0376] When the terminal device is powered on, the processor can read the software program from the storage unit, parse and execute the instructions of the software program, and process the data of the software program. When data needs to be transmitted wirelessly, the processor performs baseband processing on the data to be transmitted and outputs the baseband signal to the radio frequency (RF) circuit. The RF circuit processes the baseband signal to obtain the RF signal and transmits the RF signal outward in the form of electromagnetic waves through the antenna. When data is sent to the terminal device, the RF circuit receives the RF signal through the antenna. This RF signal is further converted into a baseband signal and output to the processor. The processor converts the baseband signal back into data and processes the data.

[0377] For ease of explanation, Figure 10 shows only one memory and processor. In actual terminal devices, multiple processors and memories may exist. Memory may also be referred to as storage medium or storage device, etc., and the embodiments of this application do not limit this.

[0378] In one embodiment, the antenna is used to perform the operations performed by the transceiver unit 801 in the above embodiments. The processor can be used to perform the operations performed by the processing unit 802 in the above embodiments.

[0379] This application also provides a computer-readable storage medium storing instructions that, when executed by a computer or processor, can implement the relevant steps in the communication method provided in the above-described method embodiments.

[0380] This application also provides a computer program product including instructions that, when executed by a computer or processor, cause one or more steps of any of the above-described communication methods to be performed. If the constituent modules of the aforementioned devices are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0381] This application provides a chip or chip system including at least one processor for calling and executing instructions stored in a memory, causing a communication device with the chip installed to perform any of the methods described above.

[0382] This application also provides another chip, including a processor and a memory, wherein the processor is used to call and execute instructions stored in the memory, causing a communication device with the chip installed to perform any of the methods described above.

[0383] This application embodiment also provides another chip, including: an input interface, an output interface, and a processing circuit. The input interface, the output interface, and the processing circuit are connected via internal connection paths. The processing circuit is used to execute any of the methods described above. Optionally, the chip also includes a memory. The input interface, the output interface, the processor, and the memory are connected via internal connection paths. The processor is used to execute code in the memory. When the code is executed, the processor is used to execute any of the methods described above.

[0384] This application also provides another chip system, including at least one processor and a communication interface, wherein the communication interface and at least one processor are interconnected via a line, and the at least one processor is used to run computer programs or instructions to perform any of the methods described above. This chip system may be composed of chips, or may include chips and other discrete devices.

[0385] This application also provides a communication system, which includes a first communication device and an OTT device, as detailed in any of the methods shown in Figures 3 to 7. The system may also include a second communication device and / or a terminal device, etc., which are not limited herein.

[0386] It should be understood that the memory mentioned in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be a hard disk drive (HDD), a solid-state drive (SSD), ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be RAM, which is used as an external cache. Memory is any other medium capable of carrying or storing desired program code having an instruction or data structure form and accessible by a computer, but is not limited thereto. The memory in the embodiments of this application can also be a circuit or any other device capable of implementing a storage function for storing program instructions and / or data.

[0387] It should also be understood that the processor mentioned in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor, or any conventional processor, etc.

[0388] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, the memory (storage module) is integrated into the processor.

[0389] It should be noted that the memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.

[0390] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments provided herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0391] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0392] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0393] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0394] The steps in the methods of this application can be adjusted, combined, or deleted according to actual needs. Each step in each embodiment can be partially performed (for example, the communication device may not perform the steps performed by the communication device in the above embodiments). The execution order of different steps can be changed. The embodiments described herein can be combined with other embodiments, different embodiments can be combined with each other, and different steps of different embodiments herein can be combined.

[0395] The modules / units in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0396] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments.

[0397] In this application, it may refer to a communication protocol or specification, such as the 3GPP communication protocol.

[0398] In this application, unless otherwise specified, "at least one" means "one or more".

[0399] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the embodiments of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0400] In the embodiments of this application, "including" can refer to a relationship of inclusion or an equality relationship. For example, A includes B, which could mean that A includes B and may also include other content, or that A and B are the same content.

[0401] In the description of this application, unless otherwise stated, " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B can mean A or B. "And / or" in this application is merely a description of the relationship between the related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Furthermore, in the description of this application, unless otherwise stated, "multiple" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0402] In this application, the words "exemplarily" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design that is described as "exemplarily" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0403] In the description of this application, "instruction" can include direct and indirect instructions, as well as explicit and implicit instructions. The information indicated by a certain piece of information is called the information to be instructed. In the specific implementation process, there are many ways to instruct the information to be instructed. For example, the information to be instructed can be directly instructed, such as by instructing the information itself or its index. Alternatively, the information to be instructed can be indirectly indicated by instructing other information, where there is a relationship between the indicated other information and the information to be instructed. Another example is that only a part of the information to be instructed can be indicated, while the other parts are known or pre-agreed upon. Furthermore, the instruction of specific information can be achieved by using a pre-agreed (such as an agreement) arrangement of various pieces of information, thereby reducing the instruction overhead to some extent.

[0404] It is understood that in the description of this application, "when," "if," and "if" all refer to the device making a corresponding action under certain objective circumstances, and are not limited to a specific time, nor do they require the device to make a judgment action when it is implemented, nor do they imply any other limitations.

[0405] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

Claims

1. A communication method, characterized in that, Applied to a first communication device, the method includes: Receive a first model acquisition request from an OTT device, the first model acquisition request being used to request the acquisition of a model; If the model is permitted to be transmitted, a first model acquisition response is sent to the OTT device, the first model acquisition response indicating that the model can be transmitted.

2. The method according to claim 1, characterized in that, Also includes: If the model is not allowed to be sent out, a second model acquisition response is sent to the OTT device, which indicates that the model cannot be sent.

3. The method according to claim 1 or 2, characterized in that, Also includes: A second model acquisition request is sent to a second communication device, the second model acquisition request being used to request the acquisition of the model.

4. The method according to any one of claims 1 to 3, characterized in that, Also includes: Determine whether the model can be allowed to be sent out based on the first model acquisition request.

5. The method according to claim 3, characterized in that, Also includes: The third model acquisition response of the second communication device is received, and the third model acquisition response is used to indicate whether the model can be sent.

6. The method according to claim 3 or 5, characterized in that, Also includes: The system receives a model acquisition failure instruction from the second communication device, the model acquisition failure instruction indicating that the model acquisition has failed.

7. The method according to claim 3 or 5, characterized in that, Also includes: The model received by the second communication device.

8. The method according to any one of claims 1 to 7, characterized in that, Also includes: Receive model feedback information from the OTT device.

9. The method according to any one of claims 3 to 8, characterized in that, Also includes: Send model feedback information to the second communication device.

10. The method according to claim 8 or 9, characterized in that, The model feedback information includes model performance indicators, and the method further includes: When the model performance indicator indicates that the model's performance does not meet the model performance requirements, an update instruction or a switching instruction for the model is sent to the communication device storing the model.

11. The method according to any one of claims 8 to 10, characterized in that, The model feedback information includes a re-request indication, which indicates whether a new model should be transmitted.

12. The method according to any one of claims 8 to 11, characterized in that, Also includes: The feedback information from the model is analyzed.

13. The method according to any one of claims 1 to 12, characterized in that, Also includes: Select the model based on the first model acquisition request.

14. A communication method, characterized in that, Applied to a passing device, the method includes: Send a first model acquisition request to the first communication device, wherein the first model acquisition request is used to request the acquisition of a model; The system receives a first model acquisition response from the first communication device, which indicates whether the model can be sent.

15. The method according to claim 14, characterized in that, Also includes: The system receives a second model acquisition response from the first communication device, the second model acquisition response indicating that the model cannot be sent.

16. The method according to claim 14 or 15, characterized in that, Also includes: Send model feedback information to the first communication device.

17. The method according to claim 16, characterized in that, The model feedback information includes at least one of the following: model performance indicator, re-request indicator, and request identifier of the first model acquisition request; The model performance indicator is used to indicate whether the performance of the model meets the model performance requirements, and the re-request indicator is used to indicate whether to transfer a new model.

18. The method according to claim 16 or 17, characterized in that, Also includes: The model is trained or inferred based on the response obtained from the first model to obtain the model feedback information.

19. The method according to any one of claims 1 to 18, characterized in that, The first model acquisition request includes at least one of the following: the request identifier of the first model acquisition request and the model requirement information.

20. The method according to claim 19, characterized in that, The model requirements information includes at least one of the following: expected model input information, expected model output information, expected model functional information, expected model operation information, model performance requirements, model format requirements, model accuracy requirements, and model storage size requirements.

21. The method according to any one of claims 1 to 20, characterized in that, The first model obtains a response including at least one of the following: the request identifier of the first model obtains the request, the request completion indication, the model, the model identifier of the model, and the model attribute information; The request completion indication is used to indicate whether the first model has completed the request.

22. The method according to claim 21, characterized in that, The model attribute information includes at least one of the following: the model identifier, the model input information, the model output information, the model function information, the model operation information, the model performance information, the model update information, the model source information, the model external information, the model quantization information, and the model storage information.

23. The method according to any one of claims 1 to 22, characterized in that, The first communication device includes at least one of the following: access network equipment, core network equipment, operation and maintenance management equipment, and terminal equipment.

24. A communication device, characterized in that, Includes units for performing the method as described in any one of claims 1 to 23.

25. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions that, when executed, cause the method according to any one of claims 1 to 23 to be implemented.

26. A computer program product, characterized in that, Includes a computer program or instructions that, when run, cause the method according to any one of claims 1 to 23 to be performed.

27. A chip or chip system, characterized in that, It includes at least one processor for retrieving and executing instructions stored in a memory, causing a communication device equipped with a chip or chip system to perform the method as described in any one of claims 1 to 23.

28. A communication system, characterized in that, The device includes a first communication device and a transfer device, the first communication device being used to perform the method according to any one of claims 1 to 13, or any one of claims 19 to 23, and the transfer device being used to perform the method according to any one of claims 14 to 23.

Images