Model data transmission method and communication device

Abstract

The embodiment of the present application provides a model data transmission method and a communication device, which relates to the field of communications, and can transmit ML model data to UE, providing support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks. The method includes: the first terminal device receives scheduling information of a first system information block SIB from an access network device, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; the first terminal device obtains the first SIB according to the scheduling information of the first SIB; the first terminal device sends a first message to the access network device according to the information of the first ML model in the first SIB, and the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of communications, and in particular to a model data transmission method and a communication device. Background Art

[0002] With the diversified development of wireless communication networks, artificial intelligence (AI) plays an increasingly important role in wireless communication networks. AI can also be called machine learning. For example, user equipment (UE) can complete specific tasks based on machine learning (ML) models. For example, in an autonomous driving scenario, UE (e.g., on-board equipment) can use ML models to predict vehicle driving information, including: vehicle driving direction, driving speed, etc. ML models can also be called AI models.

[0003] There are many types of ML models, and the size of ML model data is large, often reaching hundreds of megabytes or gigabytes. It is impossible for UE to pre-load all ML model data. However, existing technologies do not support the transmission of ML model data, which limits the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks. Summary of the invention

[0004] The embodiments of the present application provide a model data transmission method and a communication device, which can transmit ML model data to UE, and provide support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0005] In a first aspect, a model data transmission method is provided, wherein the first terminal device receives scheduling information of a first system information block SIB from an access network device, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; the first terminal device obtains the first SIB according to the scheduling information of the first SIB; the first terminal device sends a first message to the access network device according to the information of the first ML model in the first SIB, and the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

[0006] In the present application, a dedicated system information block (SIB) is provided for sending information of the ML model. The UE can request ML model data matching the UE capability information from the access network device according to the information in the SIB, so as to complete the artificial intelligence service using the ML model. It can be seen that the method provided in the embodiment of the present application can realize the sending of ML model data, and provide support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0007] In combination with the first aspect, in a first possible implementation manner of the first aspect, the first terminal device receives scheduling information of a first SIB from the access network device, including: the first terminal device receives SIB 1 from the access network device, and the SIB 1 includes the scheduling information of the first SIB.

[0008] In this application, a specific implementation of sending a dedicated SIB (for example, the first SIB described in the embodiment of this application) is provided, and the scheduling information of the dedicated SIB is sent through SIB 1, so that the terminal device can obtain the dedicated SIB, and thus obtain ML model data from the access network device according to the ML model information in the dedicated SIB. Among them, the ML model information is the information of the ML model.

[0009] In combination with the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the first terminal device obtains the first SIB according to the scheduling information of the first SIB, including: the first terminal device determines the receiving window of the first SIB according to the scheduling information of the first SIB; the first terminal device receives the first SIB in the receiving window of the first SIB.

[0010] In the present application, the access network device may broadcast the SIB related to the ML model. The terminal device may receive the SIB in the window (eg, the receiving window of the first SIB) in which the access network device broadcasts the SIB to obtain the ML model information.

[0011] In combination with the first aspect or the first or second possible implementation of the first aspect, in a third possible implementation of the first aspect, the first terminal device obtains the first SIB based on the scheduling information of the first SIB, including: sending a system information request to the access network device, the system information request including the identification information of the first SIB; receiving the first SIB from the access network device.

[0012] In this application, the terminal device can actively obtain SIB related to the ML model from the access network device.

[0013] In combination with the first possible implementation manner of the first aspect, in a fourth possible implementation manner of the first aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0014] In the present application, scheduling information of another dedicated SIB (SIB related to a dedicated bearer) can also be sent through SIB 1, so that the terminal device can establish a dedicated bearer that can be used to transmit ML model data according to the configuration information in the dedicated SIB, so as to use the ML model to complete the artificial intelligence business. It can be seen that the method provided in the embodiment of the present application can realize the transmission of ML model data, and provide support for the wide application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0015] In combination with the fourth possible implementation manner of the first aspect, in a fifth possible implementation manner of the first aspect, the method further includes: the first terminal device determines a receiving window of the second SIB according to scheduling information of the second SIB; and the first terminal device receives the second SIB in the receiving window of the second SIB.

[0016] In the present application, the access network device may broadcast the SIB related to the dedicated bearer. The terminal device may receive the SIB in the window (eg, the first SIB receiving window) in which the access network device broadcasts the SIB, and obtain the configuration information of the dedicated bearer.

[0017] In combination with the fourth or fifth possible implementation manner of the first aspect, in a sixth possible implementation manner of the first aspect, the method further includes: sending a system information request to the access network device, the system information request including identification information of the second SIB; and receiving the second SIB from the access network device.

[0018] In this application, the terminal device can actively obtain the SIB related to the dedicated bearer.

[0019] In combination with any one of the fourth to sixth possible implementations of the first aspect, in the seventh possible implementation of the first aspect, the method also includes: the first terminal device sends a second message to the access network device; the second message is used to characterize the need for the first terminal device to transmit ML model data to the second terminal device, so that the access network device sends configuration information of the first bearer to the second terminal device, and the second message includes an identifier of the second terminal device; the first terminal device sends the ML model data to the second terminal device through the first bearer.

[0020] In the present application, the terminal device can also inform the access network device of the ML model data transmission requirements through a second message. The access network device responds to the second message by sending dedicated bearer configuration information to other terminal devices so as to establish a bearer that can be used to transmit ML model data between the terminal devices.

[0021] In combination with any one of the fourth to seventh possible implementations of the first aspect, in an eighth possible implementation of the first aspect, the method further includes: the first terminal device sends configuration information of the first bearer to the second terminal device; and the first terminal device sends ML model data to the second terminal device through the first bearer.

[0022] In the present application, the terminal device also sends configuration information of a dedicated bearer to other terminal devices so that a bearer that can be used to transmit ML model data is established between the terminal devices.

[0023] In combination with the first aspect or any possible implementation manner above the first aspect, in the ninth possible implementation manner of the first aspect, the configuration information of the first bearer includes at least one of the following items: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0024] The present application also provides a specific implementation of bearer configuration information, so that the terminal device can establish a bearer according to the specific configuration information.

[0025] In combination with the first aspect or any possible implementation manner above the first aspect, in the tenth possible implementation manner of the first aspect, the information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the first SIB is the last segment of the information of the first ML model.

[0026] The present application also provides a specific implementation of the ML model information, so that the terminal device can request the access network device to send the model data of the ML model to the terminal device according to the specific ML model information.

[0027] In combination with the first aspect or any possible implementation manner above the first aspect, in an eleventh possible implementation manner of the first aspect, the access network device includes a distributed unit DU and a centralized unit CU; the first terminal device receives the scheduling information of the first SIB from the DU; or, the first terminal device receives the first SIB from the DU; or, the first terminal device receives the scheduling information of the second SIB from the DU; or, the first terminal device receives the second SIB from the DU.

[0028] The present application is also applicable to a separated access network device. Specifically, the DU can obtain the SIB related to the dedicated bearer or the SIB related to the ML model from the CU.

[0029] In combination with the first aspect or any one of the above possible implementation manners of the first aspect, in a twelfth possible implementation manner of the first aspect, the first message includes an identifier of the first ML model.

[0030] In a second aspect, a communication device is disclosed, including: a communication unit, used to receive scheduling information of a first system information block SIB from an access network device, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; a processing unit, used to obtain the first SIB according to the scheduling information of the first SIB; the communication unit is also used to send a first message to the access network device according to the information of the first ML model in the first SIB, and the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

[0031] In combination with the second aspect, in a first possible implementation manner of the second aspect, the communication unit is specifically used to receive SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

[0032] In combination with the second aspect or the first possible implementation method of the second aspect, the processing unit is specifically used to determine the receiving window of the first SIB according to the scheduling information of the first SIB; and receive the first SIB through the communication unit in the receiving window of the first SIB.

[0033] In combination with the second aspect or the first or second possible implementation of the second aspect, in a third possible implementation of the second aspect, the processing unit is specifically used to receive the first SIB from the access network device by sending a system information request to the access network device, wherein the system information request includes identification information of the first SIB.

[0034] In combination with the first possible implementation manner of the second aspect, in a fourth possible implementation manner of the second aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0035] In combination with the fourth possible implementation method of the second aspect, in a fifth possible implementation method of the second aspect, the processing unit is also used to determine a receiving window of the second SIB according to the scheduling information of the second SIB; and receive the second SIB in the receiving window of the second SIB through the communication unit.

[0036] In combination with the fourth or fifth possible implementation manner of the second aspect, in a sixth possible implementation manner of the second aspect, the processing unit is used to send a system information request to the access network device through the communication unit, wherein the system information request includes identification information of the second SIB; and receive the second SIB from the access network device.

[0037] In combination with any one of the fourth to sixth possible implementations of the second aspect, in the seventh possible implementation of the second aspect, the communication unit is also used to send a second message to the access network device; the second message is used to characterize the need for the first terminal device to transmit ML model data to the second terminal device, so that the access network device sends configuration information of the first bearer to the second terminal device, and the second message includes an identifier of the second terminal device; the communication unit is also used to send the ML model data to the second terminal device through the first bearer.

[0038] In combination with any one of the fourth to seventh possible implementations of the second aspect, in an eighth possible implementation of the second aspect, the communication unit is further used to send configuration information of the first bearer to the second terminal device, and to send ML model data to the second terminal device through the first bearer.

[0039] In combination with any one of the fourth to eighth possible implementation methods of the second aspect, in the ninth possible implementation method of the second aspect, the configuration information of the first bearer includes at least one of the following items: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0040] In combination with the second aspect or any one of the first to ninth possible implementations of the second aspect, in the tenth possible implementation of the second aspect, the information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the first SIB is the last segment of the information of the first ML model.

[0041] In combination with the second aspect or any one of the first to tenth possible implementations of the second aspect, in an eleventh possible implementation of the second aspect, the access network device includes a distributed unit DU and a centralized unit CU;

[0042] The communication unit receives scheduling information of the first SIB from the DU; or, the communication unit receives the first SIB from the DU; or, the communication unit receives scheduling information of the second SIB from the DU; or, the communication unit receives the second SIB from the DU.

[0043] In combination with the second aspect or any one of the first to eleventh possible implementations of the second aspect, in a twelfth possible implementation of the second aspect, the first message includes an identifier of the first ML model.

[0044] According to a third aspect, a method for establishing a bearer is provided, including: a first terminal device receives scheduling information of a first system information block SIB from an access network device, the first SIB includes configuration information of a first bearer, and the first bearer is used to transmit machine learning ML model data; the first terminal device obtains the first SIB according to the scheduling information of the first SIB; the first terminal device establishes the first bearer according to the configuration information of the first bearer.

[0045] In the present application, a dedicated system information block (SIB) is provided for sending configuration information of a dedicated bearer. The UE can establish a bearer that can be used to transmit ML model data based on the information in the SIB, so as to obtain ML model data and use the ML model to complete the artificial intelligence service. It can be seen that the method provided in the embodiment of the present application can realize the transmission of ML model data and provide support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0046] In combination with the third aspect, in a first possible implementation manner of the third aspect, the first terminal device receives scheduling information of a first SIB from the access network device, including: the first terminal device receives SIB 1 from the access network device, and the SIB 1 includes the scheduling information of the first SIB.

[0047] In combination with the third aspect or the first possible implementation method of the third aspect, in a second possible implementation method of the third aspect, the first terminal device obtains the first SIB according to the scheduling information of the first SIB, including: the first terminal device determines the receiving window of the first SIB according to the scheduling information of the first SIB; the first terminal device receives the first SIB in the receiving window of the first SIB.

[0048] In combination with the third aspect or the first or second possible implementation of the third aspect, in a third possible implementation of the third aspect, the first terminal device obtains the first SIB based on the scheduling information of the first SIB, including: sending a system information request to the access network device, the system information request including identification information of the first SIB; receiving the first SIB from the access network device.

[0049] In combination with the third aspect or any one of the first to third possible implementations of the third aspect, in a fourth possible implementation of the third aspect, the method further includes: the first terminal device sends a first message to the access network device; the first message is used to characterize the need for the first terminal device to transmit ML model data to the second terminal device, and the first message includes an identifier of the second terminal device; the first terminal device sends the ML model data to the second terminal device through the first bearer.

[0050] In combination with the third aspect or any one of the first to fourth possible implementations of the third aspect, in a fifth possible implementation of the third aspect, the method further includes: the first terminal device sends configuration information of the first bearer to the second terminal device; the first terminal device sends ML model data to the second terminal device through the first bearer.

[0051] In combination with the first possible implementation manner of the third aspect, in a sixth possible implementation manner of the third aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model; the information of the first ML model includes at least one of the following: ML model identifier, ML model data size, ML model iteration number, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the second SIB is the last segment of the information of the first ML model.

[0052] In combination with the sixth possible implementation manner of the third aspect, in the seventh possible implementation manner of the third aspect, the first terminal device obtains the second SIB according to the scheduling information of the second SIB, including: the first terminal device determines the receiving window of the second SIB according to the scheduling information of the second SIB; the first terminal device receives the second SIB in the receiving window of the second SIB.

[0053] In combination with the sixth or seventh possible implementation manner of the third aspect, in the eighth possible implementation manner of the third aspect, the first terminal device obtains the second SIB based on the scheduling information of the second SIB, including: sending a system information request to the access network device, the system information request including the identification information of the second SIB; receiving the second SIB from the access network device.

[0054] In combination with the third aspect or any one of the first to eighth possible implementations of the third aspect, in the ninth possible implementation of the third aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0055] In combination with the third aspect or any one of the first to ninth possible implementations of the third aspect, in the tenth possible implementation of the third aspect, the access network device includes a distributed unit DU and a centralized unit CU; the first terminal device receives the scheduling information of the first SIB from the DU; or, the first terminal device receives the first SIB from the DU; or, the first terminal device receives the scheduling information of the second SIB from the DU; or, the first terminal device receives the second SIB from the DU.

[0056] In a fourth aspect, a communication device is provided, including: a communication unit, used to receive scheduling information of a first system information block SIB from an access network device, the first SIB including configuration information of a first bearer, and the first bearer is used to transmit machine learning ML model data; a processing unit, used to obtain the first SIB according to the scheduling information of the first SIB, and establish the first bearer according to the configuration information of the first bearer.

[0057] In the present application, a dedicated system information block (SIB) is provided for sending configuration information of a dedicated bearer. The UE can establish a bearer that can be used to transmit ML model data based on the information in the SIB, so as to obtain ML model data and use the ML model to complete the artificial intelligence service. It can be seen that the method provided in the embodiment of the present application can realize the transmission of ML model data and provide support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0058] In combination with the fourth aspect, in a first possible implementation manner of the fourth aspect, the communication unit is specifically used to receive SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

[0059] In combination with the fourth aspect or the first possible implementation of the fourth aspect, in the second possible implementation of the fourth aspect, the processing unit is specifically used to determine the receiving window of the first SIB according to the scheduling information of the first SIB; the communication unit is specifically used to receive the first SIB in the receiving window of the first SIB.

[0060] In combination with the fourth aspect or the first or second possible implementation manner of the fourth aspect, in a third possible implementation manner of the fourth aspect, the communication unit is specifically used to send a system information request to the access network device, wherein the system information request includes identification information of the first SIB; and receive the first SIB from the access network device.

[0061] In combination with the fourth aspect or any one of the first to third possible implementations of the fourth aspect, in a fourth possible implementation of the fourth aspect, the communication unit is also used to send a first message to the access network device; the first message is used to characterize a need for the first terminal device to transmit ML model data to a second terminal device, and the first message includes an identifier of the second terminal device; and the ML model data is sent to the second terminal device through the first bearer.

[0062] In combination with the fourth aspect or any one of the first to fourth possible implementations of the fourth aspect, in a fifth possible implementation of the fourth aspect, the communication unit is further used to send configuration information of the first bearer to the second terminal device, and send ML model data to the second terminal device through the first bearer.

[0063] In combination with the first possible implementation manner of the fourth aspect, in a sixth possible implementation manner of the fourth aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model; the information of the first ML model includes at least one of the following: ML model identifier, ML model data size, ML model iteration number, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the second SIB is the last segment of the information of the first ML model.

[0064] In combination with the sixth possible implementation manner of the fourth aspect, in the seventh possible implementation manner of the fourth aspect, the processing unit is specifically used to determine the receiving window of the second SIB according to the scheduling information of the second SIB; the communication unit is specifically used to receive the second SIB in the receiving window of the second SIB.

[0065] In combination with the sixth or seventh possible implementation manner of the fourth aspect, in the eighth possible implementation manner of the fourth aspect, the communication unit is specifically used to send a system information request to the access network device, wherein the system information request includes identification information of the second SIB; and receive the second SIB from the access network device.

[0066] In combination with the fourth aspect or any one of the first to eighth possible implementations of the fourth aspect, in the ninth possible implementation of the fourth aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0067] In combination with the fourth aspect or any one of the first to ninth possible implementations of the fourth aspect, in the tenth possible implementation of the fourth aspect, the access network device includes a distributed unit DU and a centralized unit CU; the communication unit is specifically used to receive the scheduling information of the first SIB from the DU; or, receive the first SIB from the DU; or, receive the scheduling information of the second SIB from the DU; or, receive the second SIB from the DU.

[0068] In a fifth aspect, a model data transmission method is provided, including: an access network device determines scheduling information of a first system information block SIB, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; the access network device sends the scheduling information of the first SIB.

[0069] In combination with the fifth aspect, in a first possible implementation manner of the fifth aspect, the method further includes: receiving a first message from a first terminal device, the first message being used to request the access network device to send model data of the first ML model to the first terminal device.

[0070] In combination with the fifth aspect or the first possible implementation of the fifth aspect, in a second possible implementation of the fifth aspect, the access network device sends the scheduling information of the first SIB, including: the access network device sends SIB1, and the SIB 1 includes the scheduling information of the first SIB.

[0071] In combination with the second possible implementation manner of the fifth aspect, in a third possible implementation manner of the fifth aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0072] In combination with the third possible implementation manner of the fifth aspect, in a fourth possible implementation manner of the fifth aspect, the method further includes: receiving a second message from the access network device, the second message being used to characterize a need for the first terminal device to transmit ML model data to a second terminal device, the second message including an identifier of the second terminal device; and sending configuration information of the first bearer to the second terminal device according to the second message.

[0073] In combination with the third or fourth possible implementation manner of the fifth aspect, in the fifth possible implementation manner of the fifth aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0074] In combination with the fifth aspect or any one of the above possible implementations of the fifth aspect, in a sixth possible implementation of the fifth aspect, the information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the first SIB is the last segment of the information of the first ML model.

[0075] In combination with the fifth aspect or any one of the above possible implementations of the fifth aspect, in the seventh possible implementation of the fifth aspect, the access network device includes a distributed unit DU and a centralized unit CU; the CU sends the scheduling information of the first SIB to the DU, and the DU sends the scheduling information of the first SIB; or, the CU sends the first SIB to the DU, and the DU sends the first SIB; or, the CU sends the scheduling information of the second SIB to the DU, and the DU sends the scheduling information of the second SIB; or, the CU sends the second SIB to the DU, and the DU sends the second SIB.

[0076] In a sixth aspect, a communication device is provided, which may be an access network device or a component in an access network device. Including: In a fifth aspect, a model data transmission method is provided, comprising: an access network device determines scheduling information of a first system information block SIB, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; the access network device sends the scheduling information of the first SIB.

[0077] In combination with the sixth aspect, in a first possible implementation manner of the sixth aspect, the communication unit is also used to receive a first message from a first terminal device, where the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

[0078] In combination with the sixth aspect or the first possible implementation manner of the sixth aspect, in a second possible implementation manner of the sixth aspect, the communication unit is specifically used to send SIB 1, where SIB 1 includes scheduling information of the first SIB.

[0079] In combination with the second possible implementation manner of the sixth aspect, in a third possible implementation manner of the sixth aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0080] In combination with the third possible implementation manner of the sixth aspect, in a fourth possible implementation manner of the sixth aspect, the communication unit is also used to receive a second message from the access network device, where the second message is used to characterize a need for the first terminal device to transmit ML model data to a second terminal device, and the second message includes an identifier of the second terminal device; and send configuration information of the first bearer to the second terminal device according to the second message.

[0081] In combination with the third or fourth possible implementation manner of the sixth aspect, in the fifth possible implementation manner of the sixth aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0082] In combination with the sixth aspect or any one of the above possible implementations of the sixth aspect, in a sixth possible implementation of the sixth aspect, the information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the first SIB is the last segment of the information of the first ML model.

[0083] In the seventh aspect, a communication method is provided, including: an access network device determines scheduling information of a first system information block SIB, the first SIB includes configuration information of a first bearer, and the first bearer is used to transmit machine learning ML model data; the access network device can also send the scheduling information of the first SIB, so that the first terminal device obtains the first SIB according to the scheduling information of the first SIB, so as to establish a first bearer according to the configuration information in the first SIB, which is used to transmit ML model data.

[0084] In combination with the seventh aspect, in a first possible implementation manner of the seventh aspect, the access network device sends the scheduling information of the first SIB, including: the access network device sends SIB 1, and the SIB 1 includes the scheduling information of the first SIB.

[0085] In combination with the seventh aspect or the first possible implementation manner of the seventh aspect, in the second possible implementation manner of the seventh aspect, the method also includes: the access network device receives a system information request sent by a first terminal device, and the system information request includes identification information of the first SIB; and sends the first SIB to the first terminal device in response to the system information request.

[0086] In another possible implementation, the access network device may also broadcast the first SIB, and the terminal device receives the first SIB in a corresponding receiving window.

[0087] In combination with the seventh aspect or any one of the first or second possible implementations of the seventh aspect, in a third possible implementation of the seventh aspect, the method further includes: receiving a first message sent by a first terminal device, the first message being used to characterize a requirement for the first terminal device to transmit ML model data to a second terminal device, the first message including an identifier of the second terminal device. In response to the first message, sending configuration information of a first bearer to the second terminal device, so that the second terminal device establishes a bearer for transmitting ML data according to the configuration information of the first bearer.

[0088] In combination with the seventh aspect or any possible implementation manner above the seventh aspect, in a fourth possible implementation manner of the seventh aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model; the information of the first ML model includes at least one of the following: ML model identifier, ML model data size, ML model iteration number, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the second SIB is the last segment of the information of the first ML model.

[0089] In combination with the seventh aspect or any one of the first to eighth possible implementations of the seventh aspect, in the fifth possible implementation of the seventh aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0090] In combination with the seventh aspect or any possible implementation manner above the seventh aspect, in a sixth possible implementation manner of the seventh aspect, the access network device includes a distributed unit DU and a centralized unit CU; the first terminal device receives the scheduling information of the first SIB from the DU; or, the first terminal device receives the first SIB from the DU; or, the first terminal device receives the scheduling information of the second SIB from the DU; or, the first terminal device receives the second SIB from the DU.

[0091] In an eighth aspect, a communication device is provided, which may be an access network device or a component in the access network device. The device comprises: a processing unit, configured to determine scheduling information of a first system information block SIB, wherein the first SIB comprises configuration information of a first bearer, and the first bearer is used to transmit machine learning ML model data; a communication unit, configured to send the scheduling information of the first SIB, so that a first terminal device obtains the first SIB according to the scheduling information of the first SIB, so as to establish a first bearer according to the configuration information in the first SIB, for transmitting ML model data.

[0092] In combination with the eighth aspect, in a first possible implementation manner of the eighth aspect, the communication unit is used to send SIB1, where SIB1 includes scheduling information of the first SIB.

[0093] In combination with the eighth aspect or the first possible implementation method of the eighth aspect, in the second possible implementation method of the eighth aspect, the communication unit is also used to receive a system information request sent by a first terminal device, wherein the system information request includes identification information of the first SIB; and send the first SIB to the first terminal device in response to the system information request.

[0094] In another possible implementation, the communication unit may further broadcast the first SIB, and the terminal device receives the first SIB in a corresponding receiving window.

[0095] In combination with the eighth aspect or any one of the first or second possible implementations of the eighth aspect, in a third possible implementation of the eighth aspect, the communication unit is further configured to receive a first message sent by a first terminal device, the first message being used to characterize a requirement for the first terminal device to transmit ML model data to a second terminal device, the first message including an identifier of the second terminal device. In response to the first message, configuration information of a first bearer is sent to the second terminal device, so that the second terminal device establishes a bearer for transmitting ML data according to the configuration information of the first bearer.

[0096] In combination with the eighth aspect or any possible implementation manner above the eighth aspect, in a fourth possible implementation manner of the eighth aspect, the SIB 1 also includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model; the information of the first ML model includes at least one of the following: ML model identifier, ML model data size, ML model iteration number, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the second SIB is the last segment of the information of the first ML model.

[0097] In combination with the eighth aspect or any one of the first to eighth possible implementations of the eighth aspect, in the fifth possible implementation of the eighth aspect, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0098] In a ninth aspect, a communication device is provided, comprising at least one processor and a memory, wherein the at least one processor is coupled to the memory; the memory is used to store a computer program;

[0099] The at least one processor is used to execute the computer program stored in the memory so that the device performs the method described in the first aspect and any one of the implementations of the first aspect, or the method described in the third aspect and any one of the implementations of the third aspect.

[0100] In the tenth aspect, a computer-readable storage medium is provided, comprising: instructions are stored in the computer-readable storage medium; when the computer-readable storage medium is run on the communication device described in the above-mentioned second aspect and any one of the implementation methods of the second aspect, the communication device executes the communication method described in the above-mentioned first aspect and any one of the implementation methods of the first aspect.

[0101] In the eleventh aspect, a computer-readable storage medium is provided, comprising: instructions are stored in the computer-readable storage medium; when the computer-readable storage medium is run on the communication device described in the fourth aspect and any one of the implementations of the fourth aspect, the communication device executes the communication method described in the third aspect and any one of the implementations of the third aspect.

[0102] In a twelfth aspect, a wireless communication device is provided, the communication device including a processor, for example, applied in the communication device, for implementing the method described in the first aspect and any one of the implementations of the first aspect, the communication device may be, for example, a chip system. In a feasible implementation, the chip system also includes a memory, the memory is used to store program instructions and data necessary to implement the functions of the method described in the first aspect.

[0103] In a thirteenth aspect, a wireless communication device is provided, the communication device including a processor, for example, applied in the communication device, for implementing the functions or methods involved in the third aspect and any one of the implementations of the third aspect, the communication device may be, for example, a chip system. In a feasible implementation, the chip system also includes a memory, the memory is used to store program instructions and data necessary to implement the functions of the method described in the second aspect.

[0104] The chip system in the above aspects may be a system on chip (SOC) or a baseband chip, etc., wherein the baseband chip may include a processor, a channel encoder, a digital signal processor, a modem and an interface module, etc.

[0105] In a fourteenth aspect, a communication system is provided, comprising the communication device described in the second aspect, any possible implementation of the second aspect, the fourth aspect and any possible implementation of the fourth aspect.

[0106] In a possible implementation, the communication system further includes an access network device, and the access network device is used to send SIB 1 to the terminal device. SIB 1 may include scheduling information of SIBs related to the ML model and / or scheduling information of SIBs related to the dedicated bearer. The SIB related to the ML model includes model information of the ML model; the SIB related to the dedicated bearer includes configuration information of the dedicated bearer, and the dedicated bearer can be used to transmit ML model data. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] Figure 1 An architecture diagram of a communication system provided in an embodiment of the present application;

[0108] Figure 2 A schematic diagram of a split architecture provided in an embodiment of the present application;

[0109] Figure 3a A structural block diagram of a communication device provided in an embodiment of the present application;

[0110] Figure 3b Another structural block diagram of a network device provided in an embodiment of the present application;

[0111] Figure 4 A flowchart of a bearer establishment method provided in an embodiment of the present application;

[0112] Figure 5 A schematic diagram of a flow chart of a model data transmission method provided in an embodiment of the present application;

[0113] Figure 6 to Figure 12 A flow chart of a communication method provided in an embodiment of the present application;

[0114] Figure 13 to Figure 16 Another structural block diagram of a communication device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0115] The method provided in the embodiment of the present application is applicable to Figure 1 The communication system shown in FIG. Figure 1 As shown, the communication system includes a core network device 10, an access network device (the figure shows an access network device 20 and an access network device 30) and a terminal device 40. Among them, the access network device 20 and the access network device 30 can communicate with the core network device 10, and the terminal device 40 can communicate with the access network device 20 and the access network device 30. Figure 1 The communication system shown supports multi radiodual connectivity (MR-DC), and the terminal device 40 can communicate with the access network device 20 and the access network device 30 at the same time. In the MR-DC scenario, the access network device 20 can be the main access network device, and the access network device 30 can be the auxiliary access network device. The access network device 20 and the access network device 30 can support different communication standards, or they can support access network devices of the same communication standard.

[0116] Specifically, the core network device 10 may be an access and mobility management function (AMF), which is mainly responsible for functions such as access control, mobility management, attachment and detachment, and gateway selection. Alternatively, the core network device 10 may be a network data analysis function (NWDAF), which is mainly responsible for functions such as data collection and analysis. It should be noted that the core network device 10 is not limited to AMF and NWDAF, but may also be other devices.

[0117] The access network device 20 or the access network device 30 may be referred to as a radio access network (RAN) device. It is used to connect a terminal device to a wireless network and provide the terminal device with functions such as wireless resource management, quality of service management, data encryption and compression, etc. Exemplarily, the access network device 20 or the access network device 30 may have the following possibilities:

[0118] (1) gNB: Provides new radio (NR) control plane and / or user plane protocols and functions for terminal devices and accesses the 5th generation core (5GC);

[0119] (2) ng-eNB: Provides the protocols and functions of the control plane and / or user plane of evolved universal terrestrial radio access (E-UTRA) for terminal devices, and accesses the 5G core network (5GC);

[0120] (3) Central unit (CU): mainly includes the radio resource control (RRC), service data adaptation protocol (SDAP) and packet data convergence protocol (PDCP) layers of the gNB, or the RRC and PDCP layers of the ng-eNB;

[0121] (4) Distributed unit (DU): mainly includes the radio link control (RLC) layer, medium access control (MAC) layer and physical layer of the gNB or ng-eNB;

[0122] (5) Central unit-control plane (CU-CP): The control plane of the centralized unit mainly includes the RRC layer in the gNB-CU or ng-eNB-CU, and the control plane in the PDCP layer;

[0123] (6) Central unit-user plane (CU-UP): The user plane of the centralized unit mainly includes the SDAP layer in the gNB-CU or ng-eNB-CU and the user plane in the PDCP layer;

[0124] (7) Data analysis and management (DAM): Mainly responsible for data collection, ML model training, ML model generation, ML model update, ML model distribution and other functions.

[0125] The terminal device 40 may also be referred to as user equipment (UE), terminal, access terminal, user unit, user station, mobile station, remote station, remote terminal, mobile device, user terminal, wireless communication device, user agent or user device, etc. The terminal device may be a wireless terminal or a wired terminal. A wireless terminal may refer to a device with wireless transceiver function, which may be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; may also be deployed on water (such as ships, etc.); may also be deployed in the air (such as airplanes, balloons and satellites, etc.). The terminal device can be a drone, an Internet of Things (IoT) device (e.g., a sensor, an electric meter, a water meter, etc.), a vehicle-to-everything (V2X) device, a station (ST) in a wireless local area network (WLAN), a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA) device, a handheld device with wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device (also referred to as a wearable smart device). The terminal can also be a terminal in a next-generation communication system, for example, a terminal in a 5G communication system or a terminal in a future evolved public land mobile network (PLMN), a terminal in an NR communication system, etc., and the embodiments of the present application are not limited to this.

[0126] Figure 2 This is a schematic diagram of the architecture of a separate access network device. Figure 2, the access network equipment can be divided into one CU and one or more DUs according to the functions, wherein the CU and DU communicate with each other through the F1 interface. Optionally, the CU may include one CU-CP and one or more CU-UPs. The CU-CP and CU-UP may be connected through the E1 interface, the CU-CP and DU may be connected through the control plane interface (F1-C) of F1, and the CU-UP and DU may be connected through the user plane interface (F1-U) of F1. Furthermore, the CU, DU or CU-CP may be connected to the DAM through the G1 interface, respectively. Optionally, the DAM may be used as an internal function of the CU, DU or CU-CP, respectively, and the G1 interface does not exist in this case.

[0127] In an embodiment of the present application, a bearer for transmitting ML model data can be established, and a dedicated bearer can be used to efficiently transmit ML model data so as to complete artificial intelligence services using ML models. Alternatively, a dedicated system information block (SIB) is provided for sending information about ML models. The UE can request ML model data that matches the UE capability information from the access network device based on the information in the SIB so as to complete artificial intelligence services using the ML model. It can be seen that the method provided in the embodiment of the present application can realize the transmission of ML model data, and provide support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0128] First, the terms involved in the embodiments of the present application are explained:

[0129] (1) System information

[0130] In order to access the cell in the access network device normally, the UE needs to obtain the system information of the cell in the access network device, such as the information of the uplink and downlink cell bandwidth, the uplink and downlink time slot configuration, the relevant specific parameters of random access and transmission, etc. The system information is transmitted by the master information block (MIB) and the system information block (SIB), and is sent on different channels. The MIB contains very limited system information, such as the system frame number, the same-frequency cell selection indication, etc., and the system information contained in the MIB is sent on the broadcast channel (BCH). The SIB contains the main information of the system information and is sent on the downlink shared channel (DL-SCH). Depending on the different system information contained, SIB mainly has the following types:

[0131] SIB1: Contains information related to evaluating whether a UE is allowed to access a cell, as well as scheduling information that defines other system information. SIB1 also contains radio resource configuration information common to all UEs, and prohibition information for unified access control.

[0132] SIB2: Contains common information used for intra-frequency, inter-frequency and / or inter-system cell reselection.

[0133] SIB3: Contains neighbor cell related information used only for intra-frequency cell reselection.

[0134] SIB4: Contains relevant information used only for inter-frequency cell reselection.

[0135] SIB5: Contains relevant information used only for inter-system cell reselection.

[0136] SIB6: Contains the main notification of the earthquake and tsunami warning system (ETWS).

[0137] SIB7: Contains the second notification of ETWS.

[0138] SIB8: Contains commercial mobile alert service (CMAS) notifications.

[0139] SIB9: Contains information about the global positioning system (GPS) time and the universal coordinated time.

[0140] SIB10: Contains the human-readable network names of the non-public networks listed in SIB1.

[0141] SIB11: Contains information related to idle / inactive state measurements.

[0142] SIB12: Contains the communication configuration of the new wireless sidelink technology.

[0143] SIB13: Includes vehicle to everything (V2X) sidelink communications.

[0144] SIB14: Contains the configuration for V2X sidelink communication, which can be used in conjunction with the information contained in SIB13.

[0145] (2) ML Model

[0146] ML model is a machine learning model, which can also be called an artificial intelligence model. The ML model can be considered as an algorithm that realizes automatic "learning" of a computer. In an embodiment of the present application, the UE can use the ML model to implement specific business functions. For example: based on indicators such as the reference signal receiving power (RSRP), reference signal receiving quality (RSRQ) or signal to interference plus noise ratio (SINR) reported by the UE in the existing network and the resource utilization rate of the cell, the performance of the UE in the cell is predicted, for example, the throughput rate of the UE, and access (or switch to) the cell with the best performance is selected according to the prediction results. Alternatively, the UE uses the ML model for face recognition, prediction of vehicle driving information, etc.

[0147] (3) ML model data

[0148] In an embodiment of the present application, the ML model data may be an ML model file and / or ML model related data. Among them, the ML model file is used to record the information of the ML model, for example, the structural information of the ML model or the parameters of the ML model. Among them, the structural information of the ML model may indicate the input or output of the ML model, and may also indicate the network structure adopted by the ML model, for example: a convolutional neural network, a fully connected network, etc. The parameters of the ML model may be the weights, biases, gradient values, etc. of the network. Among them, the ML model related data may be any data related to the ML model, for example, training data, verification data, test data, model inference intermediate data, etc.

[0149] The terminal device described in the embodiment of the present application can be Figure 3a The communication device 310 is used to implement the above. Figure 3a FIG. 3 is a schematic diagram of the hardware structure of a communication device 310 provided in an embodiment of the present application. The communication device 310 includes a processor 3101 and at least one communication interface ( Figure 3a The communication interface 3103 is used as an example for illustration only), and optionally, a memory 3102 is also included. The processor 3101, the memory 3102 and the communication interface 3103 are connected to each other.

[0150] Processor 3101 can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application.

[0151] The communication interface 3103 uses any transceiver or other device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0152] The memory 3102 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently or be connected to the processor. The memory may also be integrated with the processor.

[0153] The memory 3102 is used to store computer-executable instructions for executing the solution of the present application, and the execution is controlled by the processor 3101. The processor 3101 is used to execute the computer-executable instructions stored in the memory 3102, thereby realizing the intention processing method provided in the following embodiments of the present application.

[0154] Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application code, which is not specifically limited in the embodiments of the present application.

[0155] In a specific implementation, as an embodiment, the processor 3101 may include one or more CPUs, such as Figure 3a CPU0 and CPU1 in.

[0156] In a specific implementation, as an embodiment, the communication device 310 may include multiple processors, such as Figure 3a3101 and processor 3106 in the embodiment of the present invention. Each of these processors may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0157] In a specific implementation, as an embodiment, the communication device 310 may further include an output device 3104 and an input device 3105. The output device 3104 communicates with the processor 3101 and may display information in a variety of ways. For example, the output device 3104 may be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector. The input device 3105 communicates with the processor 3101 and may receive user input in a variety of ways. For example, the input device 3105 may be a mouse, a keyboard, a touch screen device, or a sensor device.

[0158] The communication device 310 can be a general-purpose device or a dedicated device. In a specific implementation, the communication device 310 can be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, an embedded device or a computer with a plurality of Figure 3a The embodiment of the present application does not limit the type of the communication device 310.

[0159] It should be noted that the communication device 310 may be a terminal device, a functional component or assembly implemented on the terminal, or a communication chip, such as a baseband chip. When the communication device 310 is a terminal device, the communication interface may be a radio frequency module. When the communication device 310 is a communication chip, the communication interface 3103 may be an input / output interface circuit of the chip, and the input / output interface circuit is used to read and output baseband signals.

[0160] Figure 3b The communication device 320 may be the access network device described in the embodiment of the present application.

[0161] The communication device includes at least one processor 3201, at least one transceiver 3203, at least one network interface 3204 and one or more antennas 3205. Optionally, at least one memory 3202 is also included. The processor 3201, the memory 3202, the transceiver 3203 and the network interface 3204 are connected, for example, through a bus. The antenna 3205 is connected to the transceiver 3203. The network interface 3204 is used for the communication device to be connected to other communication devices through a communication link, for example, the communication device is connected to a core network element through an S1 interface. In an embodiment of the present application, the connection may include various interfaces, transmission lines or buses, etc., which are not limited in this embodiment.

[0162] The processor in the embodiment of the present application, such as processor 3201, may include at least one of the following types: a general-purpose central processing unit (CPU), a digital signal processor (DSP), a microprocessor, an application-specific integrated circuit (ASIC), a microcontroller unit (MCU), a field programmable gate array (FPGA), or an integrated circuit for implementing logical operations. For example, processor 3201 may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. At least one processor 3201 may be integrated in one chip or located on multiple different chips.

[0163] The memory in the embodiment of the present application, such as memory 3202, may include at least one of the following types: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or EEPROM. In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0164] The memory 3202 may be independent and connected to the processor 3201. Optionally, the memory 3202 may be integrated with the processor 3201, for example, integrated into a chip. The memory 3202 may store program codes for executing the technical solutions of the embodiments of the present application, and the execution is controlled by the processor 3201. The various types of computer program codes executed may also be regarded as drivers of the processor 3201. For example, the processor 3201 is used to execute the computer program codes stored in the memory 3202, thereby realizing the technical solutions in the embodiments of the present application.

[0165] The transceiver 3203 can be used to support the reception or transmission of radio frequency signals between the communication device and the terminal device, and the transceiver 3203 can be connected to the antenna 3205. Specifically, one or more antennas 3205 can receive radio frequency signals, and the transceiver 3203 can be used to receive the radio frequency signals from the antennas, convert the radio frequency signals into digital baseband signals or digital intermediate frequency signals, and provide the digital baseband signals or digital intermediate frequency signals to the processor 3201, so that the processor 3201 further processes the digital baseband signals or digital intermediate frequency signals, such as demodulation and decoding. In addition, the transceiver 3203 can be used to receive modulated digital baseband signals or digital intermediate frequency signals from the processor 3201, convert the modulated digital baseband signals or digital intermediate frequency signals into radio frequency signals, and send the radio frequency signals through one or more antennas 3205. Specifically, the transceiver 3203 can selectively perform one or more stages of down-mixing and analog-to-digital conversion processing on the RF signal to obtain a digital baseband signal or a digital intermediate frequency signal, and the order of the down-mixing and analog-to-digital conversion processing is adjustable. The transceiver 3203 can selectively perform one or more stages of up-mixing and digital-to-analog conversion processing on the modulated digital baseband signal or digital intermediate frequency signal to obtain a RF signal, and the order of the up-mixing and digital-to-analog conversion processing is adjustable. Digital baseband signals and digital intermediate frequency signals can be collectively referred to as digital signals. A transceiver can be referred to as a transceiver circuit, a transceiver unit, a transceiver device, a transmitting circuit, a transmitting unit, or a transmitting device, etc.

[0166] It should be noted that the communication device 320 may be a complete communication device, a component or assembly that implements the functions of the communication device, or a communication chip. When the communication device 320 is a communication chip, the transceiver 3203 may be an interface circuit of the chip, which is used to read and output baseband signals.

[0167] The present application embodiment provides a bearer establishment method, such as Figure 4 As shown, the method comprises the following steps:

[0168] 401. An access network device sends scheduling information of a first SIB, where the first SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0169] The ML model is a machine learning model, and may also be referred to as an artificial intelligence model. The ML model data may be an ML model file and / or ML model related data. The scheduling information of the first SIB may indicate a transmission period of the first SIB.

[0170] In a possible implementation, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0171] Among them, the configuration index is used to uniquely indicate a set of configuration information. For example, the configuration index of the first bearer is used to indicate the configuration of the bearer dedicated to transmitting ML model data, which is different from the configuration index of the signaling radio bearer (SRB) and the data radio bearer (DRB). The sequence number size of the first bearer is used to indicate the length of the bearer dedicated to transmitting ML model data. The size of the sequence number is not limited. For example, it can be 12 bits, 18 bits, etc. The sequence number size of the first bearer can be the length of the uplink bearer dedicated to transmitting ML model data, and / or the length of the downlink bearer dedicated to transmitting ML model data. The discard time is used to characterize the time when the terminal device discards or releases the first bearer. The header compression information is used to indicate the compression information of the first bearer. For example, the header compression information is the value of the maximum context identifier. When the ML model data is transmitted through the first bearer, the ML model data is compressed according to the value of the maximum context identifier.

[0172] In a possible implementation, the access network device may send the scheduling information of the first SIB by broadcasting. For example, the access network device may send the scheduling information of the first SIB by SIB 1. For example, the access network device may broadcast SIB 1, and the SIB 1 includes the scheduling information of the first SIB.

[0173] 402. The first terminal device receives scheduling information of the first SIB from the access network device, and obtains the first SIB according to the scheduling information of the first SIB.

[0174] In a specific implementation, the first terminal device may receive SIB 1 from the access network device, and obtain scheduling information of the first SIB from SIB 1.

[0175] In a possible implementation, the access network device sends the first SIB by broadcasting. The first terminal device can determine at which time domain positions the access network device broadcasts the first SIB based on the scheduling information of the first SIB, thereby determining the receiving window of the first SIB, and can also receive the first SIB in the receiving window of the first SIB. For example, the first terminal device can start to attempt decoding in the first subframe of the receiving window until the first SIB is successfully received.

[0176] It should be noted that the receiving window is the time domain position determined by the first terminal device according to the scheduling information of the first SIB, which may be the time domain position at which the access network device sends the first SIB. The terminal device may receive the first SIB at the time domain position at which the access network device sends the first SIB.

[0177] In another possible implementation, after the terminal device receives the scheduling information of the first SIB, the terminal device may also request the first SIB from the access network device. For example, the first terminal device sends a system information request to the access network device, and the system information request includes the identification information of the first SIB. After receiving the system information request, the access network device sends the first SIB to the first terminal device by broadcast or unicast according to the identification information of the first SIB. The first terminal device can receive the first SIB sent by the access network device.

[0178] 403. The first terminal device establishes the first bearer according to the configuration information of the first bearer.

[0179] The first terminal device can establish a bearer dedicated to transmitting ML model data according to the configuration index of the first bearer, that is, allocate time-frequency resources dedicated to transmitting ML model data. The first terminal device can set sequence numbers for messages in the bearer dedicated to transmitting ML model data according to the sequence number size of the first bearer, for example, setting the message sequence numbers related to uplink ML model data, and / or the message sequence numbers related to downlink ML model data. The first terminal device can release the bearer dedicated to transmitting ML model data according to the discard time, that is, release the time-frequency resources dedicated to transmitting ML model data when the discard time arrives. The first terminal device can compress the data messages in the bearer dedicated to transmitting ML model data according to the header compression information of the first bearer.

[0180] Optionally, after the first terminal device establishes the first bearer, the ML model data may be transmitted through the first bearer.

[0181] In one possible implementation, the first terminal device communicates with the second terminal device through an access network device. When the first terminal device has a transmission requirement for ML model data, the access network device can send configuration information of a dedicated bearer for ML model data (for example, the first bearer described in an embodiment of the present application) to the second terminal device, so that the second terminal device can configure a dedicated bearer, and the first terminal device can send ML model data to the second terminal device through the dedicated bearer.

[0182] For example, the first terminal device sends a first message to the access network device; the first message is used to indicate the need for the first terminal device to transmit ML model data to the second terminal device, and the first message includes an identifier of the second terminal device. The first message may be a ML model data transmission request.

[0183] After receiving the first message, the access network device learns the transmission requirements of the first terminal device and can send the configuration information of the first bearer to the second terminal device. After receiving the configuration information of the first bearer, the second terminal device can establish the first bearer with the first terminal device according to the configuration information.

[0184] Further, the first terminal device may send ML model data to the second terminal device through the first bearer.

[0185] In another possible implementation, the first terminal device and the second terminal device may communicate directly, for example, the first terminal device and the second terminal device may communicate through PC 5. When the first terminal device has a transmission requirement for ML model data, the first terminal device may send configuration information of a dedicated bearer for ML model data (for example, the first bearer described in the embodiment of the present application) to the second terminal device, so that the second terminal device may configure a dedicated bearer, and the first terminal device may send ML model data to the second terminal device through the dedicated bearer.

[0186] For example, the first terminal device sends configuration information of the first bearer to the second terminal device.

[0187] The second terminal device receives the configuration information of the first bearer, and establishes the first bearer according to the configuration information.

[0188] Further, the first terminal device sends ML model data to the second terminal device through the first bearer.

[0189] Optionally, the access network device may also send scheduling information of a SIB related to the ML model. The SIB related to the ML model may include information of the ML model and uniquely corresponds to the ML model. For example, SIB 20 corresponds to the AlexNet model, and SIB 20 includes information of the AlexNet model.

[0190] The information of the ML model may be at least one of the following information: an ML model identifier, an ML model data size, an ML model iteration number, and segmentation information.

[0191] It should be noted that the ML model identifier is used to uniquely identify an ML model. For example, ML model identifier 1 represents the AlexNet model, ML model identifier 2 represents the 16-layer visual geometry group 16 (VGG16) model, and ML model identifier 3 represents the ResNet-152 model.

[0192] The ML model data size represents the size of the data volume of the ML model data. For example, the ML model data size is 500 bytes.

[0193] The number of ML model iterations represents the number of times the ML model is updated. To update the ML model is to update the parameters of the ML model using the training data. One update of the ML model parameters is called an iteration. The number of ML model iterations can also be the number of ML model training rounds, that is, the number of rounds of updating the ML model. For example, if the training data is 1000 samples, the number of ML model training rounds is 10, and the number of samples input for each training is 20, then the number of ML model iterations is 10*(1000 / 20)=500.

[0194] The segmentation information is used to indicate whether the ML model information in the SIB is the last segment of the information of the first ML model. It can be understood that when the information of the ML model is large, the information of the ML model can be divided into multiple segments, and the segmented information is carried by multiple SIBs. If the segmentation information obtained by the terminal device in the SIB indicates that the ML model information in the current SIB is the last segment, the terminal device can determine that all the information of the ML model has been successfully received.

[0195] In a possible implementation, the access network device sends scheduling information of the SIB related to the ML model by broadcasting.

[0196] In an example, the SIB related to the ML model is a second SIB. The SIB 1 sent by the access network device further includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model.

[0197] The information of the first ML model includes at least one of the following: a model identifier of the first ML model, a model data size of the first ML model, a number of iterations of the first ML model, and segmentation information. The segmentation information is used to indicate whether the information in the first SIB is the last segment of the information of the first ML model.

[0198] In a possible implementation, the access network device sends the second SIB by broadcasting. The first terminal device can determine at which time domain positions the access network device broadcasts the second SIB based on the scheduling information of the first SIB, thereby determining the receiving window of the second SIB, and can also receive the second SIB in the receiving window of the second SIB. For example, the first terminal device can start to attempt decoding in the first subframe of the receiving window until the second SIB is successfully received.

[0199] In another possible implementation, after the terminal device receives the scheduling information of the second SIB, the terminal device may also request the second SIB from the access network device.

[0200] For example, the first terminal device sends a system information request to the access network device, and the system information request includes identification information of the second SIB. After receiving the system information request, the access network device sends the second SIB to the first terminal device by broadcast or unicast according to the identification information of the second SIB. The first terminal device can receive the second SIB.

[0201] It should be noted that Figure 4 The method shown is also applicable to the embodiment of the present application Figure 2 The system shown in FIG. 1 includes a distributed unit DU and a centralized unit CU. The DU is responsible for sending the scheduling information of the first SIB, the scheduling information of the second SIB, the first SIB and the second SIB.

[0202] For example, the first terminal device receives scheduling information of the first SIB from the DU;

[0203] Alternatively, the first terminal device receives the first SIB from the DU;

[0204] Alternatively, the first terminal device receives scheduling information of the second SIB from the DU;

[0205] Alternatively, the first terminal device receives the second SIB from the DU.

[0206] Embodiments of the present application Figure 4 In the provided method, the access network device can send the configuration information of the bearer through the SIB, so that the terminal device can establish a bearer dedicated to the ML model data according to the configuration information, and transmit the ML model data through the bearer. The terminal device does not need to pre-load too much model data, and can obtain ML model data through a dedicated bearer according to actual needs, providing support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0207] like Figure 5 As shown, the embodiment of the present application also provides a model data transmission method, such as Figure 5 As shown, the method comprises the following steps:

[0208] 501. An access network device sends scheduling information of a first SIB, where the first SIB corresponds to a first ML model and includes information of the first ML model.

[0209] In an embodiment of the present application, the access network device may also send scheduling information of a SIB related to the ML model. The SIB related to the ML model may include information of the ML model and uniquely corresponds to the ML model. For example, SIB 20 corresponds to the AlexNet model, and SIB 20 includes information of the AlexNet model.

[0210] The information of the ML model may be at least one of the following information: an ML model identifier, an ML model data size, an ML model iteration number, and segmentation information. The explanation of the above information is referred to in the full text and will not be repeated here.

[0211] In a possible implementation, the access network device sends scheduling information of the SIB related to the ML model by broadcasting.

[0212] In an example, the SIB related to the ML model is a first SIB. The SIB 1 sent by the access network device includes: scheduling information of the first SIB, the first SIB corresponds to the first ML model, and the first SIB includes information of the first ML model.

[0213] The information of the first ML model includes at least one of the following: a model identifier of the first ML model, a model data size of the first ML model, a number of iterations of the first ML model, and segmentation information. The segmentation information is used to indicate whether the information in the second SIB is the last segment of the information of the first ML model.

[0214] 502. The first terminal device receives scheduling information of the first SIB from the access network device, and obtains the first SIB according to the scheduling information of the first SIB.

[0215] In a possible implementation, the access network device sends the first SIB by broadcasting. The first terminal device can determine at which time domain positions the access network device broadcasts the first SIB based on the scheduling information of the first SIB, thereby determining the receiving window of the first SIB, and can also receive the first SIB in the receiving window of the first SIB. For example, the first terminal device can start to attempt decoding in the first subframe of the receiving window until the first SIB is successfully received.

[0216] In another possible implementation, after the terminal device receives the scheduling information of the first SIB, the terminal device may also request the first SIB from the access network device.

[0217] For example, the first terminal device sends a system information request to the access network device, and the system information request includes identification information of the first SIB. After receiving the system information request, the access network device sends the first SIB to the first terminal device by broadcast or unicast according to the identification information of the first SIB. The first terminal device can receive the first SIB.

[0218] 503. The first terminal device sends a first message to an access network device according to information of the first ML model, where the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

[0219] In a specific implementation, the first terminal device can determine the ML model data that needs to be obtained from the access network device based on its own capabilities and the information of the first ML model. For example, the terminal device can determine the size of the ML model file based on the information of the first ML model, and based on its own computing power and storage space, determine that it supports the first ML model, and then request the model data of the first ML model from the access network device.

[0220] For example, the first SIB corresponds to the AlexNet model, and the identifier of the AlexNet model is model identifier 1. According to the information in the first SIB, it is determined that the computing power required by the AlexNet model is 1.4G FLOPS (i.e., 1.4G floating-point operations per second), and the model data size is 240MB. The computing power of the first terminal device is 10G FLOPS, and the model identifier 1 is sent to the access network device, requesting the access network device to send the model data of the AlexNet model.

[0221] In one possible implementation, the first message includes an identifier of the first ML model.

[0222] Optionally, the access network device sends scheduling information of the SIB related to the dedicated bearer by broadcasting. The dedicated bearer can be used to transmit ML model data. The SIB related to the dedicated bearer includes configuration information of the dedicated bearer. The configuration information of the bearer includes at least one of the following: a configuration index, a sequence number size of the bearer, a discard time of the bearer, and header compression information of the bearer.

[0223] For example, the dedicated bearer is called the first bearer, and the SIB related to the dedicated bearer is called the second SIB. The SIB 1 sent by the access network device also includes: scheduling information of the second SIB, the second SIB includes configuration information of the first bearer, and the first bearer is used to transmit ML model data. The scheduling information of the second SIB can indicate the transmission period of the second SIB.

[0224] Specifically, the second SIB includes configuration information of the first bearer. The configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer. The explanation of the configuration information is referred to above and will not be repeated here.

[0225] In a specific implementation, the first terminal device may receive SIB 1 from the access network device, and obtain scheduling information of the second SIB from SIB 1.

[0226] In a possible implementation, the access network device sends the second SIB by broadcasting. The first terminal device can determine at which time domain positions the access network device broadcasts the second SIB according to the scheduling information of the second SIB, thereby determining the receiving window of the second SIB, and can also receive the second SIB in the receiving window of the second SIB. For example, the first terminal device can start to attempt decoding in the first subframe of the receiving window until the second SIB is successfully received.

[0227] It should be noted that the receiving window is the time domain position determined by the first terminal device according to the scheduling information of the second SIB, which may be the time domain position at which the access network device sends the second SIB. The terminal device may receive the second SIB at the time domain position at which the access network device sends the second SIB.

[0228] In another possible implementation, after the terminal device receives the scheduling information of the second SIB, the terminal device may also request the second SIB from the access network device. For example, the first terminal device sends a system information request to the access network device, and the system information request includes the identification information of the second SIB. After receiving the system information request, the access network device sends the second SIB to the first terminal device by broadcast or unicast according to the identification information of the second SIB. The first terminal device can receive the second SIB sent by the access network device.

[0229] In one possible implementation, the first terminal device communicates with the second terminal device through an access network device. When the first terminal device has a transmission requirement for ML model data, the access network device can send configuration information of a dedicated bearer for ML model data (for example, the first bearer described in an embodiment of the present application) to the second terminal device, so that the second terminal device can configure a dedicated bearer, and the first terminal device can send ML model data to the second terminal device through the dedicated bearer.

[0230] For example, the first terminal device sends a second message to the access network device; the second message is used to indicate the need for the first terminal device to transmit ML model data to the second terminal device, and the second message includes an identifier of the second terminal device. The second message may be a ML model data transmission request.

[0231] After receiving the second message, the access network device learns the transmission requirements of the first terminal device and can send the configuration information of the first bearer to the second terminal device.

[0232] After receiving the configuration information of the first bearer, the second terminal device may establish the first bearer with the first terminal device according to the configuration information. Further, the first terminal device sends the ML model data to the second terminal device through the first bearer.

[0233] In another possible implementation, the first terminal device and the second terminal device may communicate directly, for example, the first terminal device and the second terminal device may communicate through PC 5. When the first terminal device has a transmission requirement for ML model data, the first terminal device may send configuration information of a dedicated bearer for ML model data (for example, the first bearer described in the embodiment of the present application) to the second terminal device, so that the second terminal device may configure a dedicated bearer, and the first terminal device may send ML model data to the second terminal device through the dedicated bearer.

[0234] For example, the first terminal device sends configuration information of the first bearer to the second terminal device.

[0235] The second terminal device receives the configuration information of the first bearer, and establishes the first bearer according to the configuration information.

[0236] Further, the first terminal device sends ML model data to the second terminal device through the first bearer.

[0237] It should be noted that Figure 5 The method shown is also applicable to the embodiment of the present application Figure 2 The system shown in FIG. 1 includes a distributed unit DU and a centralized unit CU. The DU is responsible for sending the scheduling information of the first SIB, the scheduling information of the second SIB, the first SIB and the second SIB.

[0238] For example, the first terminal device receives scheduling information of the first SIB from the DU;

[0239] Alternatively, the first terminal device receives the first SIB from the DU;

[0240] Alternatively, the first terminal device receives scheduling information of the second SIB from the DU;

[0241] Alternatively, the first terminal device receives the second SIB from the DU.

[0242] Embodiments of the present application Figure 5 In the provided method, the SIB related to the ML model can be configured, and the access network device can send the SIB related to the ML model, so that the terminal device can determine which ML model model data the access network device can send based on the received SIB. Further, the ML model data can be requested from the access network device based on its own capabilities. The terminal device does not need to pre-load too much ML model data, and can request ML model data based on SIB and actual needs, providing support for the widespread application of artificial intelligence (or machine learning) technology in wireless communication networks.

[0243] The embodiment of the present application also provides a communication method, where an access network device can send configuration information of a dedicated bearer to a terminal device through system information (e.g., SIB). The dedicated bearer can be used to transmit ML model data, and hereinafter referred to as a packet data convergence protocol-computing radio bearer (PDCP-CRB). Figure 6 As shown, the method comprises the following steps:

[0244] 601. An access network device broadcasts SIB 1, where SIB 1 includes scheduling information of SIB x. SIB x is related to PDCP-CRB.

[0245] Specifically, SRB can only transmit control signaling, while ML model data is usually large, for example, it can reach hundreds of megabytes, so SRB cannot be used to transmit ML model data. DRB is used to transmit user data from the data network, and ML model data does not all come from the data network, so DRB cannot be used to transmit ML model data. PDCP-CRB is different from SRB and DRB, and can be used to transmit ML model data. The SIB x here is the first SIB described in the embodiment of the present application.

[0246] In a possible implementation, the SIB related to the PDCP-CRB includes configuration information of the PDCP-CRB. For example, SIB x includes configuration information of the PDCP-CRB. For the convenience of description, the SIB related to the PDCP-CRB is referred to as PDCP-CRB-SIB, and SIB x may be PDCP-CRB-SIB.

[0247] Among them, the configuration information of PDCP-CRB includes the configuration index of PDCP-CRB, and the configuration index of PDCP-CRB is used to uniquely distinguish the configuration information of PDCP-CRB. Optionally, the configuration information of PDCP-CRB may also include at least the following information: PDCP-CRB sequence number size, PDCP-CRB discard time, and header compression information. PDCP-CRB discard time: used to indicate the time when the terminal device discards or releases the PDCP-CRB; header compression information is used to indicate the compression information of PDCP-CRB.

[0248] Optionally, the scheduling information of SIB x may be the transmission period of SIB x. For example, SIB 1 includes the transmission period of SIB x. The transmission period may be in units of radio frames. For example, a transmission period of "rf8" may indicate that the transmission period of SIB x is 8 radio frames. The length of a radio frame is 10 ms, that is, the transmission period of SIB x is 80 ms.

[0249] 602. After receiving SIB 1, the terminal device obtains SIB x according to the scheduling information of SIB x.

[0250] In a specific implementation, the terminal device may determine whether the system information includes PDCP-CRB-SIB after receiving SIB 1. For example, if SIB 1 includes scheduling information of PDCP-CRB-SIB, it is determined that the system information includes PDCP-CRB-SIB.

[0251] refer to Figure 6 , the terminal device can obtain SIB x in two ways:

[0252] Method 1: The access network device sends PDCP-CRB-SIB by broadcasting (i.e. Figure 6 SIB x) of the process shown, please refer to Figure 6 As shown in step 602a in .

[0253] Step 602a: The terminal device starts to attempt decoding at the starting subframe of the PDCP-CRB-SIB receiving window until the PDCP-CRB-SIB is successfully received.

[0254] Specifically, the terminal device can determine in which windows the access network device sends PDCP-CRB-SIB according to the scheduling information of SIB x in SIB 1, thereby determining the receiving window of PDCP-CRB-SIB. Optionally, the receiving window of PDCP-CRB-SIB can be the time domain position of PDCP-CRB-SIB.

[0255] Method 2: Send PDCP-CRB-SIB according to the needs of the terminal device. For details, refer to Figure 6 As shown in step 602b and step 602c in .

[0256] Step 602b: The terminal device sends a system information request to the access network device, including an identifier of SIB x.

[0257] Specifically, when the terminal device requests PDCP-CRB-SIB from the access network device, the system information request includes "PDCP-CRB-SIB". The identifier of PDCP-CRB-SIB can be "PDCP-CRB-SIB" or other parameters, for example, "SIBy", which is not limited in the embodiments of the present application.

[0258] Step 602c: After receiving the system information request sent by the terminal device, the access network device sends a PDCP-CRB-SIB to the terminal device by broadcast or unicast.

[0259] 603. The terminal device establishes a PDCP-CRB according to the configuration information in SIB x.

[0260] Specifically, the terminal device configures the access network device to perform ML model transmission bearer according to the configuration information in the PDCP-CRB-SIB, and the implementation mainly includes the following two methods:

[0261] (1) If the terminal device itself does not have an established PDCP-CRB, the terminal device establishes a PDCP-CRB for transmitting the ML model / data, and configures the PDCP-CRB according to the information in the received PDCP-CRB-SIB, such as the sequence number size, discard time, robust header compression, etc.

[0262] (2) If the terminal device already has an established PDCP-CRB, the terminal device updates the configuration of the PDCP-CRB according to the information in the received PDCP-CRB-SIB, for example, reconfiguring the sequence number size, discard time, robust header compression, etc. of the PDCP-CRB.

[0263] 604. The terminal device receives ML model data from the access network device through PDCP-CRB.

[0264] The ML model data may be an ML model file and / or ML model related data.

[0265] Figure 6 In the method shown, the access network device sends PDCP-CRB configuration information to the terminal device by means of system information, for example, by sending PDCP-CRB-SIB to the terminal device. The terminal device configures the bearer of ML model data transmission with the access network device according to the information in PDCP-CRB-SIB, thereby realizing efficient transmission of ML model data.

[0266] The embodiment of the present application also provides a communication method, where an access network device can send ML model information to a terminal device through system information (e.g., SIB). The SIB carrying the ML model information can be called a SIB related to the ML model, hereinafter referred to as ML-SIB. Figure 7 As shown, the method comprises the following steps:

[0267] 701. The access network device broadcasts SIB 1, where SIB 1 includes scheduling information of SIB x. SIB x is related to the ML model.

[0268] It should be noted that the access network device may store model data of multiple ML models, and the information of different ML models is different, and each ML model (or model information) corresponds to a specific SIB. SIB 1 may include scheduling information of multiple SIBs related to the ML model. For the convenience of description, the SIB related to the ML model is hereinafter referred to as ML-SIB, and SIB x may be ML-SIB. SIB x here is the first SIB described in the embodiment of the present application.

[0269] For example, ML-SIB-1 corresponds to the AlexNet model, ML-SIB-2 corresponds to the 16-layer Visual Geometry Group (VGG16) model, and ML-SIB-3 corresponds to the 152-layer residual network (ResNet-152) model. The correspondence between the SIB and the ML model can be pre-set, and the terminal device can determine that the SIB includes information about the ML model related to the SIB when receiving the SIB. For example, when the terminal device receives ML-SIB-1, it knows that ML-SIB-1 contains information about the AlexNet model.

[0270] In a possible implementation, the ML-SIB includes information about the ML model. The information about the ML model includes at least one of the following:

[0271] ML model ID: used to uniquely identify an ML model. The ML model ID can be assigned by the access network device. For example, ML model ID 1 represents the AlexNet model, ML model ID 2 represents the VGG16 model, and ML model ID 3 represents the ResNet-152 model.

[0272] ML model iteration number: used to represent the number of times the parameters of the ML model are updated.

[0273] ML model size: used to characterize the size of the ML model data.

[0274] Optionally, the information of each ML model may further include ML model object information, wherein the ML model object information is used to indicate the object on which the ML model acts.

[0275] In one possible implementation, the ML model may be applied to: quality of service, quality of user experience, initial buffering, rebuffering frequency, stall ratio, key performance indicators, etc.

[0276] Among them, the quality of service can be the guaranteed bit rate of the quality of service (QOS) flow, the maximum bit rate of the flow, the packet delay buffer, the priority, etc.; the user experience quality can be the user experience score, for example, it can be a mean opinion score between 1 and 5; the initial buffering can be the time from the terminal device initiating video playback to the actual playback; the rebuffering frequency can be the ratio of the number of stall events divided by the time window; the stall ratio can be the ratio of the sum of the duration of the stall events divided by the time window, and the key performance indicators can be throughput, capacity, delay, reliability, availability, etc.

[0277] In a possible implementation, each action object may also be configured with a corresponding weight, and the terminal device allocates corresponding resources to the action object according to the weight of the action object. The resources allocated to the action object may be computing power.

[0278] Optionally, the auxiliary information of each ML model may also include segment information. The segment information includes a segment type and / or a segment sequence number. The segment type is used to indicate whether the ML model information contained in the ML-SIB is the last segment. For example, lastSegment indicates the last segment, and notLastSegment indicates not the last segment.

[0279] The segment sequence number is used to indicate the sequence number of the ML-SIB. The sequence number of the ML-SIB can indicate which segment of the entire ML model information the ML model information contained in the ML-SIB belongs to. For example, the segment sequence number 0 represents the first segment of the ML model information, the segment sequence number 1 represents the second segment of the ML model information, and so on.

[0280] Optionally, the scheduling information of SIB x may be the transmission period of SIB x. For example, SIB 1 includes the transmission period of SIB x. The transmission period may be in units of radio frames. For example, a transmission period of "rf8" may indicate that the transmission period of SIB x is 8 radio frames. The length of a radio frame is 10 ms, that is, the transmission period of SIB x is 80 ms.

[0281] 702. After receiving SIB 1, the terminal device obtains SIB x according to the scheduling information of SIB x.

[0282] In a specific implementation, after receiving SIB 1, the terminal device can determine which ML model-related SIBs are included in the system information.

[0283] For example, if SIB 1 includes scheduling information of ML-SIB, it is determined that the system information includes ML-SIB. For example, if SIB1 includes scheduling information of ML-SIB-1 and ML-SIB-2, it is determined that the system information includes ML-SIB-1 and ML-SIB-2, that is, it includes SIBs related to the AlexNet model and SIBs related to the VGG16 model.

[0284] refer to Figure 7 , the terminal device can obtain SIB x in two ways:

[0285] Method 1: The access network device sends ML-SIB by broadcasting (i.e. Figure 7 SIB x) in the process shown, please refer to Figure 7 As shown in step 702a in .

[0286] Step 702a: The terminal device starts to attempt decoding at the start subframe of the ML-SIB receiving window until the ML-SIB is successfully received.

[0287] Specifically, the terminal device can determine in which windows the access network device sends the ML-SIB according to the scheduling information of SIB x in SIB 1, thereby determining the receiving window of the ML-SIB. Optionally, the receiving window of the ML-SIB can be the time domain position of the ML-SIB.

[0288] Method 2: Send ML-SIB according to the needs of the terminal device. For details, refer to Figure 7 As shown in step 702b and step 702c in.

[0289] Step 702b: The terminal device sends a system information request to the access network device, including the identifier of SIB x.

[0290] Specifically, when the terminal device requests ML-SIB from the access network device, the system information request includes "ML-SIB-x". The identifier of ML-SIB can be "ML-SIB-x" or other parameters, for example, "SIB y", which is not limited in the embodiments of the present application.

[0291] For example, the system information request includes "ML-SIB-1", that is, requesting the access network device for the SIB related to the AlexNet model.

[0292] Step 702c: After receiving the system information request sent by the terminal device, the access network device sends the ML-SIB to the terminal device by broadcast or unicast.

[0293] 703. The terminal device determines the model data that needs to be requested from the access network device according to the model information in SIB x.

[0294] Specifically, the terminal device determines the ML model data that the access network device needs to send to the terminal device based on the ML model information in the ML-SIB and its own capabilities (such as computing power, storage space size, etc.).

[0295] For example, the terminal device determines the specific ML model identifier based on its own computing power and storage space. For example, the computing power required for the AlexNet model is 1.4G FLOPS (i.e., 1.4G floating-point operations per second), and the model size is 240MB; the computing power required for the ResNet-152 model is 22.6G FLOPS, and the model size is 240MB. If the UE's computing power is 10G FLOPS, the UE selects the AlexNet model, i.e., selects model identifier 1.

[0296] 704. The terminal device sends a request message to the access network device to request ML model data.

[0297] Specifically, the request message includes a specific ML model identifier determined by the terminal device according to its own computing capability and storage space, for example, including "model identifier 1".

[0298] 705. The access network device sends the ML model data to the terminal device.

[0299] Specifically, the access network device determines the ML model data requested by the terminal device according to the model identifier in the request message.

[0300] In a possible implementation, SIB 1 may also include scheduling information of PDCP-CRB-SIB. The terminal device may also obtain PDCP-CRB-SIB according to the scheduling information of PDCP-CRB-SIB, and PDCP-CRB-SIB includes configuration information of PDCP-CRB.

[0301] The terminal device can establish a bearer for transmitting ML model data with the access network device according to the configuration information of the PDCP-CRB, and receive the ML model data from the access network device through the bearer.

[0302] Figure 7 In the method shown, the access network device sends information of the ML model to the terminal device via system information. The terminal device determines the ML model data that the access network device needs to send to the terminal device based on the information in the ML-SIB and its own capabilities, thereby reducing the ML model collaborative interaction between the terminal device and the access network device, especially in the scenario where the access network device transmits the ML model to multiple terminal devices, which can significantly reduce the air interface signaling overhead.

[0303] The present application also provides a communication method. Figure 6 , Figure 7 The method is different from that of FIG. 1 , in which the core network device can send ML model control information to the access network device to indicate the access network device how to send ML-SIB. For example, it can indicate whether to send ML-SIB by broadcasting. Figure 8 As shown, the method comprises the following steps:

[0304] 801. A core network device sends control information to an access network device to indicate a sending method of an ML-SIB.

[0305] Specifically, the control information includes an identifier of an ML-SIB. Optionally, the control information includes identifiers of multiple different ML-SIBs.

[0306] For example, the control information includes ML-SIB-1, ML-SIB-2, ML-SIB-3, etc. ML-SIB-1 corresponds to the AlexNet model, ML-SIB-2 corresponds to the VGG16 model, and ML-SIB-3 corresponds to ResNet-152. The correspondence between the model identifier and the ML model can be preset.

[0307] Optionally, the control information may also include the following information:

[0308] (1) Broadcast priority: used to indicate the broadcast priority of different ML-SIBs. For example, priority 1 represents the highest priority, priority 2 represents the second highest priority, and so on.

[0309] The access network device considers the broadcast priority of the ML-SIB when broadcasting the ML-SIB. For example, the ML-SIB with a higher broadcast priority is broadcasted first, and the ML-SIBs with the same broadcast priority have the same processing, broadcast or not broadcast.

[0310] (2) Broadcast indication: used to indicate starting or stopping the broadcast of ML-SIB. For example, when the broadcast indication is start, the access network device will start broadcasting ML-SIB; when the broadcast indication is stop, the access network device will stop broadcasting ML-SIB.

[0311] Steps 802 to 806 are the same as steps 701 to 705 described above.

[0312] Optional, Figure 8 The method shown includes step 807 .

[0313] 807. When the access network device fails to broadcast the ML-SIB, feedback information is sent to the core network device.

[0314] The feedback information may include an identifier of the ML-SIB that failed to be broadcast. When the core network device receives the feedback information, it may determine that the ML-SIB corresponding to the identifier in the feedback information cannot be broadcast, and the ML-SIB may be sent in a unicast manner.

[0315] Figure 8 In the method shown, the core network device can send control information to the access network device to instruct the access network device whether to send ML-SIB by broadcasting, which is beneficial for the access network device to determine the ML-SIB sending method and send ML-SIB in the correct manner to ensure the security of ML model data.

[0316] The present application also provides a communication method suitable for Figure 2The DU obtains the ML-SIB from other devices (e.g., CU, CU-CP, or DAM), and sends the ML-SIB to the terminal device. Fig. 9 As shown, the method comprises the following steps:

[0317] 901. DU obtains ML-SIB from a first device.

[0318] The first device may be a CU, a CU-CP or a DAM.

[0319] Optionally, in step 901, the DU may also obtain control information from the first device.

[0320] Among them, the explanation of ML-SIB and control information refers to the relevant description of the embodiment mentioned above, and will not be repeated here.

[0321] Fig. 9 The method shown includes steps 902 to 906, and steps 902 to 906 are the same as steps 701 to 705 described above.

[0322] Optional, Fig. 9 The method shown further includes step 907. Specifically, step 907 is: the DU returns feedback information to the first device.

[0323] When the DU fails to send the ML-SIB, the DU sends ML feedback information to the CU / CU-CP, and the feedback information includes the identifier of the ML-SIB that failed to be sent.

[0324] It should be noted that the CU / CU-CP may not store the ML-SIB and / or control information. The CU / CU-CP may first obtain the ML-SIB and / or control information from the DAM, and then send the ML-SIB and / or control information to the DU.

[0325] In this scenario, in step 907, after receiving the feedback information sent by the DU, the CU / CU-CP sends the feedback information to the DAM.

[0326] The present application also provides a communication method, which is also applicable to Figure 2 The DU obtains the PDCP-CRB-SIB from other devices (e.g., CU, CU-CP, or DAM), and sends the PDCP-CRB-SIB to the terminal device. Fig.10 As shown, the method comprises the following steps:

[0327] 1001. DU obtains PDCP-CRB-SIB from the first device.

[0328] The first device may be a CU, a CU-CP or a DAM. The explanation of PDCP-CRB-SIB refers to the relevant description of the above-mentioned embodiment, which will not be repeated here.

[0329] Optionally, in step 1001, the DU may also obtain control information from the first device. The control information is used to indicate the transmission mode of the PDCP-CRB-SIB. For example, indicating that the transmission mode of the PDCP-CRB-SIB is unicast or broadcast.

[0330] Fig.10 The method shown includes steps 1002 to 1005, and steps 1002 to 1005 are the same as steps 601 to 604 described above.

[0331] It should be noted that the CU / CU-CP may not store the PDCP-CRB-SIB. The CU / CU-CP may first obtain the PDCP-CRB-SIB from the DAM and then send the PDCP-CRB-SIB to the DU.

[0332] Optional, Fig.10 The method shown also includes step 1006. Specifically, step 1006 is: the DU returns feedback information to the first device.

[0333] When the DU fails to broadcast the PDCP-CRB-SIB, the DU sends feedback information to the CU / CU-CP, where the feedback information includes the identifier of the PDCP-CRB-SIB that failed to broadcast.

[0334] The embodiment of the present application also provides a communication method, whereby a terminal device can transmit ML model data to other terminal devices. Specifically, the configuration information of the PDCP-CRB can be forwarded by the access network device so as to transmit the ML model data via the PDCP-CRB. Fig.11 As shown, the method comprises the following steps:

[0335] Steps 1101 to 1104 are the same as steps 601 to 604 described above.

[0336] The difference is that the terminal device in steps 601 to 604 is replaced by terminal device 1.

[0337] Step 1105: Terminal device 1 sends an ML model data transmission request to the access network device.

[0338] Specifically, the ML model data transmission request may include a destination ID. The destination ID is used to indicate the target of the ML model data transmission process of the terminal device 1, and the destination ID may be an identifier of a device receiving the ML model data from the terminal device 1. For example, the destination ID may be an identifier of the terminal device 2.

[0339] In a possible implementation, the identifier of the terminal device 2 is a bit string or an octet string.

[0340] Optionally, the ML model data transmission request further includes quality of service information corresponding to the destination ID. The quality of service information corresponding to the destination ID may be a quality of service parameter for transmitting the ML model data, for example, the quality of service parameter is a maximum stream bit rate, a guaranteed stream bit rate, a priority, a packet delay budget, a packet error rate, etc.

[0341] Optionally, the ML model data transmission request further includes a communication mode corresponding to the destination ID. The communication mode corresponding to the destination ID may be a transmission mode of the ML model data transmission, such as unicast, multicast, broadcast, etc.

[0342] Step 1106: The access network device sends the ML model data transmission configuration to the terminal device 2, including the configuration information of the PDCP-CRB.

[0343] Among them, the PDCP-CRB configuration information is as described above and will not be repeated here.

[0344] Optionally, according to the communication mode corresponding to the destination ID in the ML model data transmission request, the access network device may send the PDCP-CRB configuration information to the terminal device 2 in the following manner:

[0345] (a) Unicast: The access network device sends the PDCP-CRB configuration information to the terminal device 2 separately, for example, sending the PDCP-CRB configuration information to the terminal device 2 through the radio resource control reconfiguration protocol.

[0346] (b) Broadcast: The access network device broadcasts the PDCP-CRB-SIB, which protects the configuration information of the PDCP-CRB. After receiving the PDCP-CRB-SIB, the terminal device 2 configures the PDCP-CRB.

[0347] (c) Multicast: The access network device sends a multicast message to the group to which the terminal device 2 belongs. The multicast message includes the configuration information of the PDCP-CRB.

[0348] Step 1107: Terminal device 1 and terminal device 2 establish a PDCP-CRB, and transmit ML model data via the PDCP-CRB.

[0349] Fig.11In the method shown, terminal device 1 sends an ML model data transmission request to the access network device, triggering the access network device to send the PDCP-CRB configuration to terminal device 2, so that terminal device 1 and terminal device 2 can transmit ML models / data to each other. In addition, the access network device sends the PDCP-CRB configuration to terminal device 2 only after receiving the ML model data transmission request from terminal device 1, thereby saving air interface signaling overhead.

[0350] The embodiment of the present application also provides a communication method, which is applicable to a scenario where there is a direct communication interface between terminal devices, and the terminal device can transmit ML model data to other terminal devices. Specifically, PDCP-CRB configuration information can be sent to other terminal devices so as to transmit ML model data through PDCP-CRB. Fig.12 As shown, the method comprises the following steps:

[0351] Step 1201: Terminal device 1 sends ML model data transmission configuration to terminal device 2, including configuration information of PDCP-CRB.

[0352] Among them, the PDCP-CRB configuration information refers to the above description and may include PDCP-CRB configuration index, sequence number size, discard time, robust header compression information, etc.

[0353] It should be noted that the terminal device 1 can also Figure 6 The method shown obtains PDCP-CRB-SIB, and obtains the configuration information of PDCP-CRB in PDCP-CRB-SIB.

[0354] Step 1202: Terminal device 2 establishes a PDCP-CRB with terminal device 1 according to the configuration information of PDCP-CRB.

[0355] It should be noted that if the terminal device 2 successfully configures the PDCP-CRB, it sends an ML model data transmission configuration completion message to the terminal device 1. The ML model data transmission configuration completion message is used to indicate that the PDCP-CRB is successfully configured.

[0356] If the terminal device 2 fails to configure the PDCP-CRB, it sends an ML model data transmission configuration failure message to the terminal device 1, where the ML model data transmission configuration failure message is used to indicate that the PDCP-CRB configuration fails.

[0357] Step 1203: Terminal device 1 and terminal device 2 transmit ML model data via PDCP-CRB.

[0358] Fig.12In the provided method, ML model data transmission configuration can be sent between terminal devices through a direct communication interface (e.g., PC5 interface), so that ML model data can be directly transmitted between terminal devices without the participation of access network devices, thereby reducing the processing load of the access network devices.

[0359] In the case of dividing each functional module into corresponding functional modules, Fig.13 A possible structural diagram of the communication device involved in the above embodiments is shown. Fig.13 The communication device shown may be the terminal device described in the embodiment of the present application, or may be a component in the terminal device that implements the above method, or may be a chip used in the terminal device. The chip may be a SOC or a baseband chip with communication functions, etc. Fig.13 As shown, the communication device includes a processing unit 1301 and a communication unit 1302. The processing unit may be one or more processors, and the communication unit may be a transceiver or a communication interface.

[0360] The processing unit 1301 may be used to support the terminal device to perform internal processing such as message generation or message parsing, for example, to support the first device to generate a first message. The processing unit 1301 may also support the terminal device to perform steps 402, 403, 502, 603, 703, etc., and / or other processes for the technology described herein.

[0361] The communication unit 1302 is used to support communication between the terminal device and other communication devices, for example, to support interaction between the terminal device and other terminal devices, or interaction between the terminal device and an access network device, for example, to support the terminal device to execute step 503, etc., and / or other processes for the technology described in this document.

[0362] It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0363] like Fig.14 As shown, the communication device may further include a storage unit 1303, and the storage unit 1303 is used to store program codes and / or data of the communication device.

[0364] The processing unit 1301 may include at least one processor, the communication unit 1302 may be a transceiver or a communication interface, and the storage unit 1303 may include a memory.

[0365] It should be noted that, in the above-mentioned embodiments of the communication device, each unit may also be referred to as a module, a component, a circuit, etc.

[0366] In an example, the communication unit 1302 is used to receive scheduling information of a first system information block SIB from an access network device, where the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model.

[0367] The processing unit 1301 is used to obtain the first SIB according to the scheduling information of the first SIB; the communication unit is also used to send a first message to the access network device according to the information of the first ML model in the first SIB, and the first message is used to request the access network device to send the model data of the first ML model to the first terminal device.

[0368] The communication unit 1302 is specifically configured to receive SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

[0369] The processing unit 1301 is specifically configured to determine a receiving window of the first SIB according to the scheduling information of the first SIB; and receive the first SIB through the communication unit in the receiving window of the first SIB.

[0370] The processing unit 1301 is specifically configured to receive the first SIB from the access network device by sending a system information request to the access network device, where the system information request includes identification information of the first SIB.

[0371] Optionally, the SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

[0372] The processing unit 1301 is further configured to determine a receiving window of the second SIB according to the scheduling information of the second SIB; and receive the second SIB in the receiving window of the second SIB through the communication unit.

[0373] The processing unit 1301 is configured to send a system information request to the access network device through the communication unit, where the system information request includes identification information of the second SIB; and receive the second SIB from the access network device.

[0374] The communication unit 1302 is also used to send a second message to the access network device; the second message is used to represent the need for the first terminal device to transmit ML model data to the second terminal device, and the second message includes an identifier of the second terminal device; the communication unit is also used to send the ML model data to the second terminal device through the first bearer.

[0375] The communication unit 1302 is further configured to send configuration information of the first bearer to the second terminal device, and send ML model data to the second terminal device through the first bearer.

[0376] Optionally, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0377] Optionally, the information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the first SIB is the last segment of the information of the first ML model.

[0378] Optionally, the first message includes an identifier of the first ML model.

[0379] In another possible implementation, the communication unit 1302 is used to receive scheduling information of a first system information block SIB from an access network device, where the first SIB includes configuration information of a first bearer, and the first bearer is used to transmit machine learning ML model data.

[0380] The processing unit 1301 is configured to obtain the first SIB according to the scheduling information of the first SIB, and establish the first bearer according to the configuration information of the first bearer.

[0381] The communication unit 1302 is specifically configured to receive SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

[0382] The processing unit 1301 is specifically configured to determine a receiving window of the first SIB according to the scheduling information of the first SIB. The communication unit 1302 is specifically configured to receive the first SIB in the receiving window of the first SIB.

[0383] The communication unit 1302 is specifically configured to send a system information request to the access network device, where the system information request includes identification information of the first SIB; and receive the first SIB from the access network device.

[0384] The communication unit 1302 is also used to send a first message to the access network device; the first message is used to represent the need for the first terminal device to transmit ML model data to the second terminal device, and the first message includes an identifier of the second terminal device; and the ML model data is sent to the second terminal device through the first bearer.

[0385] The communication unit 1302 is further used to send configuration information of the first bearer to the second terminal device, and send ML model data to the second terminal device through the first bearer.

[0386] Optionally, the SIB 1 also includes: scheduling information of a second SIB, the second SIB corresponds to the first ML model, and the second SIB includes information of the first ML model; the information of the first ML model includes at least one of the following: ML model identifier, ML model data size, ML model iteration number, and segmentation information; wherein the segmentation information is used to indicate whether the ML model information in the second SIB is the last segment of the information of the first ML model.

[0387] The processing unit 1301 is specifically used to determine a receiving window of the second SIB according to the scheduling information of the second SIB; the communication unit is specifically used to receive the second SIB in the receiving window of the second SIB.

[0388] The communication unit 1302 is specifically configured to send a system information request to the access network device, where the system information request includes identification information of the second SIB; and receive the second SIB from the access network device.

[0389] Optionally, the configuration information of the first bearer includes at least one of the following: a configuration index of the first bearer, a sequence number size of the first bearer, a discard time of the first bearer, and header compression information of the first bearer.

[0390] In the case of dividing each functional module into corresponding functional modules, Fig.15 A possible structural diagram of the communication device involved in the above embodiments is shown. Fig.15 The communication device shown may be the access network device described in the embodiment of the present application, or may be a component in the access network device that implements the above method, or may be a chip used in the access network device. The chip may be a SOC or a baseband chip with communication functions, etc. Fig.15 As shown, the communication device includes a processing unit 1401 and a communication unit 1402. The processing unit may be one or more processors, and the communication unit may be a transceiver or a communication interface.

[0391] The processing unit 1401 may be used, for example, to support the access network device to perform internal processing such as message generation or message parsing, for example, to support the access network device to generate SIB 1, and / or other processes for the technology described herein.

[0392] The communication unit 1402 is used to support communication between the access network device and other communication devices, for example, to support interaction between the access network device and the terminal device, to support the second device to execute step 401, step 501, step 601, etc., and / or other processes for the technology described in this article.

[0393] It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module and will not be repeated here.

[0394] like Fig.16 As shown, the communication device may further include a storage unit 1403, and the storage unit 1403 is used to store program codes and / or data of the communication device.

[0395] The processing unit 1401 may include at least one processor, the communication unit 1402 may be a transceiver or a communication interface, and the storage unit 1403 may include a memory.

[0396] It should be noted that, in the above-mentioned embodiments of the communication device, each unit may also be referred to as a module, a component, a circuit, etc.

[0397] The present application embodiment provides a computer-readable storage medium, in which instructions are stored; the instructions are used to execute Figures 4 to 12 The method shown.

[0398] The embodiment of the present application provides a computer program product including instructions, which, when executed on a communication device, enables the communication device to execute the following Figures 4 to 12 The method shown.

[0399] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the communication device can be divided into different functional modules to complete all or part of the functions described above.

[0400] The processor in the embodiment of the present application may include but is not limited to at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor and other types of computing devices that run software, each of which may include one or more cores for executing software instructions to perform operations or processing. The processor may be a separate semiconductor chip, or it may be integrated into a semiconductor chip together with other circuits. For example, it may form a SoC (system on chip) with other circuits (such as a codec circuit, a hardware acceleration circuit, or various buses and interface circuits), or it may be integrated into the ASIC as a built-in processor of an ASIC. The ASIC with the integrated processor may be packaged separately or with other circuits. In addition to the core for executing software instructions for operations or processing, the processor may further include necessary hardware accelerators, such as field programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0401] The memory in the embodiments of the present application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or EEPROM. In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0402] In the present application, "at least one" refers to one or more. "Multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or its similar expressions refers to any combination of these items, including any combination of single items 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, c can be single or multiple. In addition, in order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the differences.

[0403] In the several embodiments provided in the present application, it should be understood that the disclosed database access device and method can be implemented in other ways. For example, the above-described database access device embodiment is only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the database access device or unit can be electrical, mechanical or other forms.

[0404] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0405] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0406] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks or optical disks.

[0407] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A model data transmission method, characterized in that: include: The first terminal device receives scheduling information of a first system information block SIB from an access network device, where the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; The first terminal device acquires the first SIB according to the scheduling information of the first SIB; The first terminal device sends a first message to the access network device according to the information of the first ML model in the first SIB, where the first message is used to request the access network device to send model data of the first ML model to the first terminal device; The access network equipment includes a distributed unit DU and a centralized unit CU; The first terminal device receives scheduling information of the first SIB from the DU; or The first terminal device receives the first SIB from the DU; or, The first terminal device receives scheduling information of a second SIB from the DU; or, The first terminal device receives the second SIB from the DU.

2. The method according to claim 1, characterized in that The first terminal device receives scheduling information of a first SIB from the access network device, including: The first terminal device receives SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

3. The method according to claim 1 or 2, characterized in that: The first terminal device acquires the first SIB according to the scheduling information of the first SIB, including: The first terminal device determines a receiving window of the first SIB according to the scheduling information of the first SIB; The first terminal device receives the first SIB in a reception window of the first SIB.

4. The method according to claim 2, characterized in that: The SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

5. The method according to claim 4, characterized in that The method further comprises: The first terminal device determines a receiving window of the second SIB according to the scheduling information of the second SIB; The first terminal device receives the second SIB in a reception window of the second SIB.

6. The method according to claim 4 or 5, characterized in that: The method further comprises: The first terminal device sends a second message to the access network device; the second message is used to indicate a requirement of the first terminal device to transmit ML model data to the second terminal device, so that the access network device sends configuration information of the first bearer to the second terminal device, and the second message includes an identifier of the second terminal device; The first terminal device sends ML model data to the second terminal device through the first bearer.

7. The method according to claim 4 or 5, characterized in that: The method further comprises: The first terminal device sends configuration information of the first bearer to the second terminal device; The first terminal device sends ML model data to the second terminal device through the first bearer.

8. The method according to claim 4 or 5, characterized in that: The configuration information of the first bearer includes at least one of the following: The configuration index of the first bearer, the sequence number size of the first bearer, the discard time of the first bearer, and the header compression information of the first bearer.

9. The method according to claim 1 or 2 or 4 or 5, characterized in that: The information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the first SIB is the last segment of the information of the first ML model.

10. A model data transmission method, characterized in that: include: The access network device determines scheduling information of a first system information block SIB, where the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; The access network device sends scheduling information of the first SIB; The access network equipment includes a distributed unit DU and a centralized unit CU; The CU sends the scheduling information of the first SIB to the DU, and the DU sends the scheduling information of the first SIB; or, The CU sends the first SIB to the DU, and the DU sends the first SIB; or, The CU sends scheduling information of the second SIB to the DU, and the DU sends scheduling information of the second SIB; or, The CU sends the second SIB to the DU, and the DU sends the second SIB.

11. The method according to claim 10, characterized in that The method further comprises: A first message is received from a first terminal device, where the first message is used to request the access network device to send model data of the first ML model to the first terminal device.

12. The method according to claim 10 or 11, characterized in that: The access network device sending the scheduling information of the first SIB includes: The access network device sends SIB 1, where SIB 1 includes scheduling information of the first SIB.

13. The method according to claim 12, characterized in that The SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

14. The method according to claim 13, characterized in that The method further comprises: receiving a second message from the access network device, where the second message is used to indicate a requirement of the first terminal device to transmit ML model data to the second terminal device, and the second message includes an identifier of the second terminal device; The configuration information of the first bearer is sent to the second terminal device according to the second message.

15. The method according to claim 13, characterized in that The configuration information of the first bearer includes at least one of the following: The configuration index of the first bearer, the sequence number size of the first bearer, the discard time of the first bearer, and the header compression information of the first bearer.

16. The method according to claim 10 or 11, characterized in that The information of the first ML model includes at least one of the following: an ML model identifier, an ML model data size, an ML model iteration count, and segmentation information; wherein the segmentation information is used to indicate whether the model information in the first SIB is the last segment of the information of the first ML model.

17. A communication device, characterized in that: include: A communication unit, configured to receive scheduling information of a first system information block SIB from an access network device, where the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; A processing unit, configured to obtain the first SIB according to the scheduling information of the first SIB; The communication unit is further used to send a first message to the access network device according to the information of the first ML model in the first SIB, where the first message is used to request the access network device to send model data of the first ML model to the first terminal device; The access network equipment includes a distributed unit DU and a centralized unit CU; The communication unit receives scheduling information of the first SIB from the DU; or The communication unit receives the first SIB from the DU; or, The communication unit receives scheduling information of a second SIB from the DU; or, The communication unit receives the second SIB from the DU.

18. The device according to claim 17, characterized in that The communication unit is specifically configured to receive SIB 1 from the access network device, where SIB 1 includes scheduling information of the first SIB.

19. The device according to claim 18, characterized in that The SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

20. A communication device, characterized in that: include: Distributed unit DU and centralized unit CU, The CU sends scheduling information of a first system information block SIB to the DU, and the DU sends scheduling information of the first SIB, the first SIB corresponds to a first machine learning ML model, and the first SIB includes information of the first ML model; or, The CU sends the first SIB to the DU, and the DU sends the first SIB; or, The CU sends scheduling information of the second SIB to the DU, and the DU sends scheduling information of the second SIB; or, The CU sends the second SIB to the DU, and the DU sends the second SIB.

21. The device according to claim 20, characterized in that The DU is further configured to receive a first message from a first terminal device, wherein the first message is configured to request the communication apparatus to send model data of the first ML model to the first terminal device.

22. The device according to claim 20 or 21, characterized in that The DU is specifically used to send SIB 1, where SIB 1 includes scheduling information of the first SIB.

23. The device according to claim 22, characterized in that The SIB 1 also includes: scheduling information of a second SIB, the second SIB includes configuration information of a first bearer, and the first bearer is used to transmit ML model data.

24. A communication device, characterized in that: comprising a processor coupled to a memory; Memory for storing computer programs; A processor, configured to execute the computer program stored in the memory, so that the apparatus performs the method according to any one of claims 1 to 16.

25. A computer-readable storage medium, characterized in that: A program or instruction is stored, and when the program or instruction is executed by a processor, the method according to any one of claims 1 to 16 is executed.

26. A computer program product, characterized in that The computer program product comprises instructions, which, when executed by a processor, cause the method according to any one of claims 1 to 16 to be performed.

27. A chip, characterized in that: The chip includes a processor and an interface circuit, wherein the interface circuit is coupled to the processor, and the processor is configured to run a computer program or instruction so that the method according to any one of claims 1 to 16 is executed.

Images