WO2022082356A1

WO2022082356A1 - Communication method and apparatus

Info

Publication number: WO2022082356A1
Application number: PCT/CN2020/121875
Authority: WO
Inventors: 杨水根; 晋英豪; 周彧; 张亮亮
Original assignee: 华为技术有限公司
Priority date: 2020-10-19
Filing date: 2020-10-19
Publication date: 2022-04-28

Abstract

Provided in the embodiments of the present application are a communication method and apparatus. The method comprises: an access network device determining a resource location of a first paging message according to a machine learning (ML) capability classification identifier; the access network device sending, at the resource location, the first paging message to a terminal device, wherein the first paging message comprises the ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information, and correspondingly, the terminal device receiving the first paging message from the access network device, wherein the first paging message comprises the ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information; and the terminal device determining ML capability information of the terminal device according to the ML capability classification identifier. By using the embodiments of the present application, the overheads of resource transmission can be reduced.

Description

A communication method and device

technical field

The present application relates to the field of communication technologies, and in particular, to a communication method and device.

Background technique

With the continuous development of the diversification and intelligence of wireless communication networks, the traditional communication methods using higher and higher frequency spectrums, wider bandwidths, and more and more antennas can no longer satisfy people's expectations for existing The demand for wireless network communication, and with the explosive development of intelligent terminals and various applications, wireless communication network behavior and performance factors are more dynamic and unpredictable than in the past, and operating increasingly complex wireless communication networks at low cost and high efficiency is the current operation. a challenge facing business. Therefore, artificial intelligence (AI) and machine learning (ML) will assume more and more important tasks in wireless communication networks. AI refers to technology that presents human intelligence through computer programs. ML focuses on developing computer programs that can access data and use that data to learn on its own. In the subsequent description, without distinguishing between artificial intelligence and machine learning, ML models can also be called AI models.

As part of the wireless communication network, the terminal device also needs to introduce AI/ML and cooperate with the wireless network at the same time. For example, in wireless networks, AI/ML is introduced in the physical layer between terminal equipment and wireless communication network, media access control, wireless resource control, wireless resource management, operation and maintenance and other fields. For example, when training an ML model, the network device needs to upload training data from the terminal device; or, when training an ML model based on federated learning, the network device needs to send an initial ML model to the terminal device.

However, when AI/ML is introduced into a wireless communication network, a network device needs to send configuration information to a large number of terminal devices, so that the network devices can obtain training data; or, send an ML model to the terminal devices. Since the network device needs to send configuration information to each terminal device, this will result in a waste of transmission resources.

SUMMARY OF THE INVENTION

The embodiments of the present application disclose a communication method and apparatus, which can reduce transmission resource overhead.

A first aspect of the embodiments of the present application discloses a communication method, including: an access network device determines a resource location of a first paging message according to a machine learning ML capability classification identifier; The terminal device sends the first paging message, where the first paging message includes the ML capability classification identifier, where the ML capability classification identifier corresponds to a group of ML capability information.

In the above method, the access network device determines the resource location according to the ML capability classification identifier, and then sends the first paging message at the resource location, which can solve the problem of sending paging messages to a large number of terminal devices in the ML scenario , reduce the overhead of transmission resources, while in the prior art, when an access network device needs to send a paging message to multiple terminal devices, the paging message includes the identities of the multiple terminal devices. The way of paging messages will cause a huge waste of wireless resources, and even lead to serious congestion of wireless transmission, affecting the service quality of other normal services. By using the method of the embodiment of the present application, it is possible to avoid waste of wireless resources and serious congestion of wireless transmission.

In a possible implementation manner, before the access network device determines the resource location of the first paging message according to the machine learning ML capability classification identifier, the method further includes: the access network device receives a message from the network device. A second paging message, where the second paging message includes the ML capability classification identifier.

In another possible implementation manner, the second paging message includes a paging priority; the access network device sends the first paging message to the terminal device at the resource location, including: The access network device sends the first paging message to the terminal device according to the paging priority.

In the above method, the reliability of the first paging message can be ensured and the performance can be improved by means of the access network device sending the first paging message to the terminal device according to the paging priority.

In another possible implementation manner, the second paging message includes a paging area; the access network device sends the first paging message to the terminal device at the resource location, including: the The access network device sends the first paging message to the terminal device within the paging area.

A second aspect of the embodiments of the present application discloses a communication method, including: a terminal device receiving a first paging message from an access network device, where the first paging message includes a machine learning ML capability classification identifier, the ML capability The classification identifier corresponds to a group of ML capability information; the terminal device determines the ML capability information of the terminal device according to the ML capability classification identifier.

In the above method, by the way that the terminal device receives the first paging message from the access network device, the problem of sending paging messages to a large number of terminal devices in the ML scenario can be solved, and the overhead of transmission resources can be reduced, while the prior art Among them, when the access network device needs to send a paging message to multiple terminal devices, the paging message includes the identities of the multiple terminal devices, and the method of sending the paging message based on a specific terminal device will cause a huge amount of wireless resources. Waste, and even lead to serious congestion of wireless transmission, affecting the quality of service of other normal services. By using the method of the embodiment of the present application, it is possible to avoid waste of wireless resources and serious congestion of wireless transmission.

A third aspect of the embodiments of the present application discloses a communication method, including: an access network device receiving first configuration information from a network device, where the first configuration information includes a machine learning ML capability classification identifier, the ML capability classification identifier Corresponding to a group of ML capability information; the access network device determines, according to the ML capability classification identifier, a terminal device having the ML capability information corresponding to the ML capability classification identifier; the access network device sends the first 2. Configuration information, where the second configuration information is used to indicate the type of data collected by the terminal device.

In the above method, the communication access network device sends the second configuration information to the terminal device having the ML capability information corresponding to the ML capability classification identifier according to the ML capability classification identifier, instead of sending the second configuration information to each terminal device in the wireless communication network The second configuration information can solve the problem of sending configuration information to a large number of terminal devices in the ML scenario, reduce signaling overhead between the access network device and the terminal device, and avoid resource waste.

A fourth aspect of the embodiments of the present application discloses a communication method, including: an access network device sending machine learning ML capability mapping query information to a network device, where the ML capability mapping query information includes a second ML capability classification identifier, the ML capability mapping The capability mapping query information is used to request the second ML capability information corresponding to the second ML capability classification identifier; the access network device receives the ML capability mapping response information from the network device, and the ML capability mapping response information includes The second ML capability classification identifier corresponds to the second ML capability information; the access network device determines the correspondence between the second ML capability classification identifier and the second ML capability information according to the ML capability mapping response information.

In the above method, the access network device obtains the second ML capability information corresponding to the second ML capability classification identifier from the network device, instead of obtaining the second ML capability information from each terminal device in the wireless communication network. The overhead of wireless resources is reduced.

In another possible implementation manner, before the access network device sends the machine learning ML capability mapping query information to the network device, the method further includes: the access network device receives the second information from the terminal device. ML capability classification identifier.

A fifth aspect of the embodiments of the present application discloses a communication method, including: a network device receiving machine learning ML capability mapping query information from an access network device, where the ML capability mapping query information includes a second ML capability classification identifier, and the The ML capability mapping query information is used by the access network device to request the network device to provide second ML capability information corresponding to the second ML capability classification identifier; the network device sends the ML capability map to the access network device Response information, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.

In a possible implementation manner, before the network device receives the machine learning ML capability mapping query information from the access network device, the method further includes: the network device receiving the first ML capability information from the terminal device, The first ML capability information corresponds to a first ML capability classification identifier; the network device determines a second ML capability classification identifier according to the first ML capability classification identifier, and sends the second ML capability classification identifier to the terminal device.

In the above method, the network device assigns and sends the second ML capability classification identifier to the terminal device, so that the uniformity of the capability classification identifiers in the wireless communication network can be ensured.

A sixth aspect of the embodiments of the present application discloses a communication method, including: a terminal device sends first machine learning ML capability information to a network device, where the first ML capability information corresponds to a first ML capability classification identifier; the terminal device receives A second ML capability classification identifier from the network device.

In the above method, the network equipment can allocate the second ML capability classification identifier to the terminal device, so that the uniformity of the capability classification identifiers in the wireless communication network can be ensured.

A seventh aspect of the embodiments of the present application discloses a communication method, including: a core network device receives request information from a terminal device, where the request information includes one or more machine learning ML capability classification identifiers requested by the terminal device; The core network device determines the ML capability classification identifier that the terminal device is allowed to use, and sends the ML capability classification identifier that the terminal device is allowed to use to the terminal device and the access network device; the terminal device is allowed to use the ML capability classification identifier. The ML capability classification identifier is used for the access network device to perform access control or resource allocation to the terminal device.

In the above method, the way that the core network device determines the ML capability classification identifier that the terminal device is allowed to use, and sends the identifier to the terminal device and the access network device, can help the network operator to formulate flexible policies for the terminal device, such as , only allowing a terminal device to use a certain ML model in a specific cell.

An eighth aspect of the embodiments of the present application discloses a communication method, including: a terminal device sending request information to a core network device, where the request information includes one or more machine learning ML capability classification identifiers requested by the terminal device; The terminal device receives an ML capability classification identifier that is allowed to be used by the terminal device from the core network device, and the ML capability classification identifier that is permitted to be used by the terminal device is used by the access network device to access the terminal device. Control or resource allocation.

A ninth aspect of the embodiments of the present application discloses a communication method, including: an access network device receives a machine learning ML capability classification identifier that is allowed to be used by a terminal device from a core network device; Use the machine learning ML capability classification identifier to perform access control or resource allocation to the terminal device.

In the above method, the core network device determines the ML capability classification identifier that the terminal device is allowed to use, and sends the identifier to the terminal device and the access network device, so that the access network device can access the terminal device according to the identifier. Control and resource allocation, effective and rational use of resources.

A tenth aspect of the embodiments of the present application discloses a communication method, including: a second network device receiving a machine learning ML capability classification identifier from a first network device, the ML capability classification identifier corresponding to a group of ML capability information; The second network device determines the ML capability information of the terminal device according to the ML capability classification identifier.

In the above method, the method of obtaining the ML capability classification identifier from the first network device through the second network device, and then determining the ML capability information of the terminal device according to the ML capability classification identifier, does not need to be sent to each network device in the wireless communication network. The terminal device obtains the ML capability information, which saves the overhead of wireless resources.

In an optional manner, before the second network device receives the machine learning ML capability classification identifier from the first network device, the method further includes: the second network device sends to the first network device ML capability query information, where the ML capability query information is used to query the ML capability information of the terminal device.

An eleventh aspect of the embodiments of the present application discloses a communication method, including: a first network device sends a machine learning ML capability classification identifier to a second network device, where the ML capability classification identifier corresponds to a set of ML capability information; the ML capability classification identifier corresponds to a set of ML capability information; The capability classification identifier is used by the second network device to determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, before the first network device sends the machine learning ML capability classification identifier to the second network device, the method further includes: the first network device receives the information from the second network device. ML capability query information, where the ML capability query information is used to query the ML capability information of the terminal device.

In a possible implementation manner, before the first network device sends the machine learning ML capability classification identifier to the second network device, the method further includes: the first network device receives the ML capability from a terminal device Classification ID.

A twelfth aspect of the embodiments of the present application discloses a communication device, which may be an access network device or a chip in the access network device, and includes: a processing unit configured to determine a first search engine according to a machine learning ML capability classification identifier The resource location of the paging message; the transceiver unit is configured to send the first paging message to the terminal device at the resource location, where the first paging message includes the ML capability classification identifier, the ML capability classification identifier Corresponds to a set of ML capability information.

In a possible implementation manner, the transceiver unit is further configured to receive a second paging message from a network device, where the second paging message includes the ML capability classification identifier.

In another possible implementation manner, the second paging message includes a paging priority; the transceiver unit is further configured to send the first paging message to the terminal device according to the paging priority call message.

In another possible implementation manner, the second paging message includes a paging area; the transceiver unit is further configured to send the first paging message to the terminal device in the paging area.

Regarding the technical effects brought by the twelfth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the first aspect or corresponding implementation manners.

A thirteenth aspect of the embodiments of the present application discloses a communication device, which may be a terminal device or a chip in the terminal device, and includes: a transceiver unit configured to receive a first paging message from an access network device, the The first paging message includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information; the processing unit is configured to determine the ML capability information of the device according to the ML capability classification identifier.

Regarding the technical effects brought by the thirteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the second aspect or corresponding implementation manners.

A fourteenth aspect of the embodiments of the present application discloses a communication device, which may be an access network device or a chip in the access network device, and includes: a transceiver unit configured to receive first configuration information from the network device, and The first configuration information includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information; a processing unit is configured to determine the ML capability corresponding to the ML capability classification identifier according to the ML capability classification identifier information terminal equipment; the transceiver unit is further configured to send second configuration information to the terminal equipment, where the second configuration information is used to indicate the type of data collected by the terminal equipment.

Regarding the technical effects brought by the fourteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the third aspect or corresponding implementation manners.

A fifteenth aspect of the embodiments of the present application discloses a communication device, which may be an access network device or a chip in the access network device, and includes: a transceiver unit configured to send machine learning ML capability mapping query information to the network device , the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used to request the second ML capability information corresponding to the second ML capability classification identifier; the transceiver unit is further configured to Receive ML capability mapping response information from the network device, where the ML capability mapping response information includes a correspondence relationship between the second ML capability information corresponding to the second ML capability classification identifier; a processing unit configured to The mapping response information determines the correspondence between the second ML capability classification identifier and the second ML capability information.

In a possible implementation manner, the transceiver unit is further configured to receive the second ML capability classification identifier from the terminal device.

Regarding the technical effects brought by the fifteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the fourth aspect or corresponding implementation manners.

A sixteenth aspect of the embodiments of the present application discloses a communication device, which may be a network device or a chip in the network device, and includes: a processing unit configured to receive a machine learning ML capability map from an access network device according to a transceiver unit query information, the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used by the apparatus to request the network device to provide a second ML capability corresponding to the second ML capability classification identifier information; the transceiver unit is further configured to send ML capability mapping response information to the access network device, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.

In a possible implementation manner, the transceiver unit is further configured to receive first ML capability information from the terminal device, where the first ML capability information corresponds to the first ML capability classification identifier; the processing unit is further configured to use determining a second ML capability classification identifier according to the first ML capability classification identifier, and sending the second ML capability classification identifier to the terminal device.

Regarding the technical effects brought by the sixteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the fifth aspect or corresponding implementation manners.

A seventeenth aspect of the embodiments of the present application discloses a communication device, which may be a terminal device or a chip in the terminal device, and includes: a processing unit configured to send first machine learning ML capability information to a network device according to the transceiver unit, The first ML capability information corresponds to the first ML capability classification identifier; the transceiver unit is configured to receive the second ML capability classification identifier from the network device.

Regarding the technical effects brought by the seventeenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the sixth aspect or corresponding embodiments.

An eighteenth aspect of the embodiments of the present application discloses a communication device, which may be a core network device or a chip in the core network device, and includes: a processing unit configured to receive request information from a terminal device according to a transceiver unit, the The request information includes one or more machine learning ML capability classification identifiers requested to be used by the terminal device; the transceiver unit is configured to determine the ML capability classification identifiers that are allowed to be used by the terminal device, and report to the terminal device and access The network device sends the ML capability classification identifier that is allowed to be used by the terminal device; the ML capability classification identifier that is allowed to be used by the terminal device is used for the access network device to perform access control or resource allocation to the terminal device.

Regarding the technical effects brought by the eighteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the seventh aspect or corresponding embodiments.

A nineteenth aspect of the embodiments of the present application discloses a communication device, which may be a terminal device or a chip in the terminal device, and includes: a processing unit configured to send request information to a core network device according to a transceiver unit, where the request information including one or more machine learning ML capability classification identifiers requested to be used by the apparatus; the transceiver unit is configured to receive from the core network equipment ML capability classification identifiers that are allowed to be used by the apparatus, the apparatus allows the apparatus to use The used ML capability class identifier is used by the access network equipment to perform access control or resource allocation to the device.

Regarding the technical effects brought by the nineteenth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the eighth aspect or corresponding embodiments.

A twentieth aspect of an embodiment of the present application discloses a communication device, which may be an access network device or a chip in the access network device, and includes: a transceiver unit configured to receive a message from the core network device that is allowed to be used by the terminal device A machine learning ML capability classification identifier; a processing unit configured to perform access control or resource allocation to the terminal device according to the machine learning ML capability classification identifier that is allowed to be used by the terminal device.

Regarding the technical effects brought by the twentieth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the ninth aspect or corresponding embodiments.

A twenty-first aspect of the embodiments of the present application discloses a communication apparatus, which may be a second network device or a chip in the second network device, and includes: a transceiver unit configured to receive machine learning ML from the first network device A capability classification identifier, where the ML capability classification identifier corresponds to a group of ML capability information; a processing unit, configured to determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, the transceiver unit is further configured to send ML capability query information to the first network device, where the ML capability query information is used to query the ML capability information of the terminal device.

Regarding the technical effects brought by the twenty-first aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the tenth aspect or corresponding embodiments.

A twenty-second aspect of the embodiments of the present application discloses a communication device, which may be a first network device or a chip in the first network device, and includes: a processing unit configured to send a machine to the second network device according to the transceiver unit Learning the ML capability classification identifier, where the ML capability classification identifier corresponds to a group of ML capability information; the ML capability classification identifier is used by the second network device to determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, the transceiver unit is further configured to receive ML capability query information from the second network device, where the ML capability query information is used to query the ML capability information of the terminal device.

In another possible implementation manner, the transceiver unit is further configured to receive the ML capability classification identifier from the terminal device.

Regarding the technical effects brought by the twenty-second aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the eleventh aspect or corresponding embodiments.

A twenty-third aspect of an embodiment of the present application discloses a communication apparatus, which may be an access network device or a chip of the access network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Determine the resource location of the first paging message according to the machine learning ML capability classification identifier;

The transceiver sends the first paging message to the terminal device at the resource location, where the first paging message includes the ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information .

In a possible implementation manner, the processor is further configured to receive, through the transceiver, a second paging message from a network device, where the second paging message includes the ML capability classification identifier.

In another possible implementation manner, the second paging message includes a paging priority; the processor is further configured to send the first paging message to the terminal device according to the paging priority call message.

In another possible implementation manner, the second paging message includes a paging area; the processor is further configured to send the first paging area to the terminal device through the transceiver in the paging area A paging message.

Regarding the technical effects brought by the twenty-third aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the first aspect or corresponding embodiments.

A twenty-fourth aspect of an embodiment of the present application discloses a communication apparatus, which may be a terminal device or a chip of the terminal device, and includes at least one processor and a transceiver, wherein the at least one processor is configured to transmit and receive through the The processor communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

receiving, by the transceiver, a first paging message from an access network device, where the first paging message includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information;

Determine the ML capability information of the device according to the ML capability classification identifier.

Regarding the technical effects brought by the twenty-fourth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the second aspect or corresponding implementation manners.

A twenty-fifth aspect of an embodiment of the present application discloses a communication apparatus, which may be an access network device or a chip of the access network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Receive first configuration information from a network device through the transceiver, where the first configuration information includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information;

Determine the terminal device having the ML capability information corresponding to the ML capability classification identifier according to the ML capability classification identifier;

Send second configuration information to the terminal device through the transceiver, where the second configuration information is used to indicate the type of data collected by the terminal device.

Regarding the technical effects brought by the twenty-fifth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the third aspect or corresponding embodiments.

A twenty-sixth aspect of an embodiment of the present application discloses a communication apparatus, which may be an access network device or a chip of the access network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

The machine learning ML capability mapping query information is sent to the network device through the transceiver, where the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used to request the second ML capability classification identifier Corresponding second ML capability information;

Receive, by the transceiver, ML capability mapping response information from the network device, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier;

The correspondence between the second ML capability classification identifier and the second ML capability information is determined according to the ML capability mapping response information.

In a possible implementation manner, the processor is further configured to receive the second ML capability classification identifier from the terminal device through the transceiver.

Regarding the technical effects brought by the twenty-sixth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the fourth aspect or corresponding implementation manners.

A twenty-seventh aspect of an embodiment of the present application discloses a communication apparatus, which may be a network device or a chip of a network device, and includes at least one processor and a transceiver, wherein the at least one processor is configured to transmit and receive through the The processor communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Receive machine learning ML capability mapping query information from an access network device through the transceiver, where the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used for the access network device requesting the apparatus to provide second ML capability information corresponding to the second ML capability classification identifier;

Send ML capability mapping response information to the access network device through the transceiver, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.

In a possible implementation manner, the processor is further configured to receive first ML capability information from the terminal device through the transceiver, where the first ML capability information corresponds to the first ML capability classification identifier; The first ML capability classification identifier determines a second ML capability classification identifier, and sends the second ML capability classification identifier to the terminal device.

Regarding the technical effects brought by the twenty-seventh aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the fifth aspect or corresponding embodiments.

A twenty-eighth aspect of an embodiment of the present application discloses a communication apparatus, which may be a terminal device or a chip of a terminal device, and includes at least one processor and a transceiver, wherein the at least one processor is configured to transmit and receive through the The processor communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Send the first machine learning ML capability information to the network device through the transceiver, where the first ML capability information corresponds to the first ML capability classification identifier;

The second ML capability classification identification from the network device is received by the transceiver.

Regarding the technical effects brought by the twenty-eighth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the sixth aspect or corresponding implementation manners.

A twenty-ninth aspect of an embodiment of the present application discloses a communication apparatus, which may be a core network device or a chip of a core network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Receive request information from a terminal device through the transceiver, where the request information includes one or more machine learning ML capability classification identifiers requested to be used by the terminal device;

determining the ML capability classification identifier that the terminal device is allowed to use, and sending the ML capability classification identifier that the terminal device is allowed to use to the terminal device and the access network device through the transceiver; the allowing the terminal device to use The ML capability classification identifier is used by the access network device to perform access control or resource allocation to the terminal device.

Regarding the technical effects brought by the twenty-ninth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the seventh aspect or corresponding embodiments.

A thirtieth aspect of the embodiments of the present application discloses a communication apparatus, which may be a terminal device or a chip of the terminal device, and includes at least one processor and a transceiver, wherein the at least one processor is configured to pass the transceiver In communication with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Sending request information to the core network device through the transceiver, where the request information includes one or more machine learning ML capability classification identifiers requested to be used by the device;

The ML capability classification identifier permitted to be used by the device from the core network device is received by the transceiver, where the ML capability classification identifier permitted to be used by the device is used by the access network device to perform access control on the device or resource allocation.

Regarding the technical effects brought by the thirtieth aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the eighth aspect or corresponding embodiments.

A thirty-first aspect of the embodiments of the present application discloses a communication apparatus, which may be an access network device or a chip of the access network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Receive, through the transceiver, a machine learning ML capability classification identifier that is allowed to be used by the terminal device from the core network device;

Perform access control or resource allocation to the terminal device according to the machine learning ML capability classification identifier that the terminal device is allowed to use.

Regarding the technical effects brought by the thirty-first aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the ninth aspect or corresponding embodiments.

A thirty-second aspect of an embodiment of the present application discloses a communication apparatus, which may be a second network device or a chip of the second network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

Receive, by the transceiver, a machine learning ML capability classification identifier from the first network device, where the ML capability classification identifier corresponds to a group of ML capability information;

Determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, the processor is further configured to send ML capability query information to the first network device, where the ML capability query information is used to query the ML capability information of the terminal device.

Regarding the technical effects brought by the thirty-second aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the tenth aspect or corresponding implementation manners.

A thirty-third aspect of an embodiment of the present application discloses a communication apparatus, which may be a first network device or a chip of the first network device, and includes at least one processor and a transceiver, wherein the at least one processor is used for The transceiver communicates with other devices, the memory is used to store a computer program, and the processor invokes the computer program to perform the following operations:

The machine learning ML capability classification identifier is sent to the second network device through the transceiver, where the ML capability classification identifier corresponds to a set of ML capability information; the ML capability classification identifier is used by the second network device to classify the ML capability according to the ML capability The classification identifier determines the ML capability information of the terminal device.

In a possible implementation manner, the processor is further configured to receive, through the transceiver, ML capability query information from the second network device, where the ML capability query information is used to query the terminal device's ML capability information.

In a possible implementation manner, the processor is further configured to receive the ML capability classification identifier from the terminal device through the transceiver.

Regarding the technical effects brought by the thirty-third aspect or possible implementation manners, reference may be made to the introduction to the technical effects of the eleventh aspect or corresponding embodiments.

A thirty-fourth aspect of an embodiment of the present application discloses a computer program, which, when the computer program is executed by a communication device, implements the method described in the possible implementation manner of any of the foregoing aspects.

A thirty-fifth aspect of an embodiment of the present application discloses a computer-readable storage medium, where a computer program or instruction is stored in the storage medium, and when the computer program or instruction is executed by a communication device, any of the foregoing aspects is implemented. The method described in possible implementations.

A thirty-sixth aspect of an embodiment of the present application discloses a chip system, the chip system includes at least one processor, a memory, and an interface circuit, and the memory, the interface circuit, and the at least one processor are interconnected through a line, Instructions are stored in the at least one memory; when the instructions are executed by the processor, the methods described in the possible implementation manners of any of the foregoing aspects are implemented.

Description of drawings

1 is a schematic structural diagram of a communication system provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of the architecture of a separate access network device provided by an embodiment of the present application;

3 is a schematic flowchart of a communication method provided by an embodiment of the present application;

FIG. 4 is a schematic flowchart of another communication method provided by an embodiment of the present application;

FIG. 5 is a schematic flowchart of another communication method provided by an embodiment of the present application;

6 is a schematic flowchart of another communication method provided by an embodiment of the present application;

7 is a schematic flowchart of another communication method provided by an embodiment of the present application;

8 is a schematic flowchart of another communication method provided by an embodiment of the present application;

9 is a schematic diagram of a protocol stack between a new terminal device and an access network device provided by an embodiment of the present application;

10 is a schematic diagram of a new HDAPb protocol stack provided by an embodiment of the present application;

11 is a schematic diagram of a new HDAPa protocol stack provided by an embodiment of the present application;

12 is a schematic diagram of a new protocol stack provided by an embodiment of the present application;

13 is a schematic diagram of a new HDAPc protocol stack provided by an embodiment of the present application;

FIG. 14 is a schematic structural diagram of a communication device provided by an embodiment of the present application;

FIG. 15 is a schematic structural diagram of another communication apparatus provided by an embodiment of the present application.

Detailed ways

The embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

Please refer to FIG. 1. FIG. 1 is a schematic structural diagram of a communication system 1000 provided by an embodiment of the present application. As shown in FIG. 1, the communication system includes a core network device 1001, a first access network device 1002, and a second access network device 1002. network equipment 1003 and terminal equipment 1004. The first access network device 1002 or the second access network device 1003 can communicate with the core network device 1001; the terminal device 1004 can communicate with the first access network device 1002 or the second access network device 1003, the terminal The device 1004 is also capable of communicating with the first access network device 1002 and the second access network device 1003 simultaneously, that is, multi-radio dual connectivity (MR-DC). In the MR-DC scenario, the first access network device 1002 may be the primary access network device, the second access network device 1003 may be the secondary access network device, the first access network device 1002 and the second access network device The device 1003 may be an access network device of a different communication standard, or may be an access network device of the same communication standard. It should be understood that the communication system 1000 to which the method of the embodiment of the present application can be applied may include more or less core network devices, access network devices or terminal devices. The interface between the core network device and the access network device is the NG interface, and the interface between the access network device and the access network device is the Xn interface. The methods in the embodiments of the present application may be applied to the communication system 1000 shown in FIG. 1 .

(1) Terminal equipment, also known as user equipment (UE), terminal, access terminal, subscriber unit, subscriber station, mobile station, remote station, remote terminal, mobile equipment, user terminal, wireless communication equipment, User agent or user device, etc. The terminal device can be a wireless terminal or a wired terminal. A wireless terminal can refer to a device with wireless transceiver function, which can be deployed on land, including indoor or outdoor, handheld or vehicle-mounted; it can also be deployed on water (such as ships). etc.); can also be deployed in the air (eg on airplanes, balloons, satellites, etc.). The terminal device may be a drone, an internet of things (IoT) device (for example, a sensor, an electricity meter, a water meter, etc.), a vehicle-to-everything (V2X) device, a wireless local area networks, WLAN) stations (station, ST), cellular phones, cordless phones, session initiation protocol (session initiation protocol, SIP) phones, wireless local loop (wireless local loop, WLL) stations, personal digital processing (personal digital assistant, PDA) devices, handheld devices with wireless communication capabilities, computing devices or other processing devices connected to wireless modems, in-vehicle devices, wearable devices (also known as wearable smart devices). The terminal may also be a terminal in a next-generation communication system, for example, a terminal in a fifth-generation mobile communication technology (5th generation mobile networks, 5G) communication system or a future evolved public land mobile network (public land mobile network, PLMN) terminal, terminal in a new radio (new radio, NR) system, etc., which are not limited here.

(2) The core network equipment, which can be an access and mobility management function (AMF), is mainly responsible for functions such as access control, mobility management, attachment and detachment, and gateway selection. The core network device can be a network data analytics function (NWDAF), which is mainly responsible for functions such as data collection and analysis. The core network equipment involved in this application is not limited to AMF and NWDAF.

(3) Access network equipment, also known as radio access network (RAN) equipment, is a device that connects terminal equipment to a wireless network, which can provide wireless resource management and service quality management for terminal equipment. , data encryption and compression. Exemplarily, the access network device may be in the following manner:

Next generation Node B (gNB): Provides NR control plane and/or user plane protocols and functions for terminal equipment;

Next generation evolutional Node B (ng-eNB): Provides the protocols and functions of the control plane and/or user plane of the evolved universal terrestrial radio access (E-UTRA) for terminal equipment ;

Centralized unit (CU): It mainly includes the radio resource control (RRC) layer of gNB, the service data adaptation protocol (SDAP) layer and the packet data convergence protocol (packet data convergence protocol). , PDCP) layer, or RRC layer and PDCP layer of ng-eNB;

Distributed unit (DU): It mainly includes the radio link control (RLC) layer, media access control (MAC) layer and physical layer of the gNB or ng-eNB;

Central unit-control plane (CU-CP): The control plane of the CU, mainly including the RRC layer in the gNB-CU or ng-eNB-CU, and the control plane in the PDCP layer;

Central unit-user plane (CU-UP): The user plane of the CU, mainly including the SDAP layer in the gNB-CU or ng-eNB-CU, and the user plane in the PDCP layer.

Data analysis and management (DAM): mainly responsible for data collection, machine learning (ML) model training, ML model generation, ML model update, ML model distribution and other functions.

Please refer to FIG. 2 , which is a schematic diagram of the architecture of a separate access network device. The access network equipment is divided into one CU and one or more DUs according to functions, wherein the CU and the DU are adjacent to each other through the F1 interface. Further, one CU may include one CU-CP and one or more CU-UPs. CU-CP and CU-UP can be connected through E1 interface, CU-CP and DU can be connected through F1 control plane interface (F1-C), CU-UP and DU can be connected through F1 user interface interface (F1-U) to connect. Further, the CU, DU or CU-CP can be connected to the DAM through the G1 interface, respectively. Optionally, the DAM can be used as an internal function of the CU, DU, or CU-CP, respectively. At this time, there is no G1 interface (or the G1 interface is an internal interface, which is invisible to the outside world).

It can be understood that the foregoing communication system shown in FIG. 2 is only for illustrating the technical solutions of the embodiments of the present application more clearly, and does not constitute a limitation on the technical solutions provided by the embodiments of the present application. For example, the communication system may also include other devices, such as network control devices. The network control device may be an operation management and maintenance (operation administration and maintenance, OAM) system, also called a network management system. The network control device may manage the aforementioned first access network device, second access network device, and core network device.

The following will introduce the technical terms involved in this application:

ML models, also known as artificial intelligence (AI) models. An ML model is a mathematical or signaling model composed of training data and expert knowledge to describe the characteristics of a given dataset statistically. ML models include supervised learning models, unsupervised learning models, reinforcement learning models, neural network models, and the like. Typically, ML is divided into a training part and an inference part. Among them, the training part refers to the process of learning and obtaining an ML model that performs a specific task based on a certain training data set. The inference part refers to the process in which the ML model calculates the input data and obtains the inference result.

In the NR system, a network device can send a paging message to terminal devices in an RRC idle state, an RRC connected state, or an RRC inactive state to initiate paging and data transmission to these terminal devices. Usually, when the network device needs to page the terminal device, it will send downlink control information (DCI) in one or more paging occasions (PO), indicating the resources that carry the paging message, the DCI It is a DCI scrambled by a paging-radio network temporary identity (P-RNTI) used for paging. Among them, PO is composed of periodic paging search space and control resource set (control resource set, CORESET). The terminal device will receive/detect/listen to DCI in one or more POs in a paging cycle to receive the paging message and determine whether the network device has initiated paging to itself.

Currently, in order for a terminal device to collect data, the network device sends configuration information to the terminal device, where the configuration information is used to indicate the specific type of data that the terminal device needs to collect. In addition, for a terminal device in an idle state (idle) or an inactive state (inactive), the network device initiates a paging process for the terminal device, so that the network device communicates with the terminal device, for example, the network sends a configuration to the terminal device. information or send the initial ML model to the end device. For a terminal device in a connected state (connected), the network device initiates a paging process for the terminal device, so that the network device can send new system information to the terminal device. Send configuration information or paging message to a specific terminal device through the above network device, for example, the configuration information or paging message contains the identity of the terminal device, for example, 5G s-temporary mobile subscriber identity (5gs-temporary mobile subscriber identity) subscription identifier, 5G-S-TMSI), or, when a network device needs to send a paging message to multiple terminal devices, the paging message contains the identities of multiple terminal devices. During ML model training, network devices usually require the participation of a large number of terminal devices. For example, the order of magnitude of terminal devices is 10^4. In this scenario, the existing method of sending configuration information or paging messages based on specific terminal devices will cause wireless The huge waste of resources even leads to serious congestion of wireless transmission and affects the service quality of other normal services. Therefore, in order to solve the problem of how to efficiently send configuration information or paging messages to a large number of terminal devices in the ML scenario, the embodiments of the present application propose the following solutions.

Please refer to Fig. 3, Fig. 3 is a kind of communication method provided by the embodiment of the present application, and this method includes but is not limited to the following steps:

Step S301: The network device sends a second paging message to the access network device.

Specifically, this step is an optional step, the network device may be a core network device or other access network device, and the second paging message includes the ML capability classification identifier. The second paging message may further include a paging priority and/or a paging area, the paging priority is used to indicate the priority of the first paging message, and the paging priority may be represented by a numerical value, for example, 1 indicates a high priority, 2 indicates a low priority, etc. Of course, the paging priority may also be expressed in other manners, which are not limited by the embodiments of this application. The paging area is used to indicate the area where the access network device sends the first paging message to the terminal device, for example, a tracking area identity (tracking area identity, TAI) list, and each TAI represents a tracking area.

Step S302: The access network device receives the second paging message from the network device.

Specifically, this step is an optional step. The network device may be a core network device or other access network device, and the second paging message includes the ML capability classification identifier. The second paging message may also include a paging priority and/or a paging area.

Step S303: The access network device determines the resource location of the first paging message according to the ML capability classification identifier.

Specifically, the resource location may be a paging frame (paging frame, PF) and/or PO, where the PF and PO to be detected by the terminal device are calculated and determined according to a predefined rule based on the ML capability classification identifier.

For example, the system frame number (system frame number, SFN) of the paging frame (paging frame, PF) where the PO is located in a discontinuous reception (DRX) cycle detected by the terminal device, and the PF corresponding to the SFN The index of the PO, etc., can be determined by the following formula:

The SFN of PF satisfies:

(SFN+PF_offset)mod T=(T div N)*(ML_category_ID mod N) (Formula 1);

The index is of the PO in this _SFN satisfies:

In the above formula, a PF is a radio frame, which may contain one or more POs, or a starting position of a PO. A PO is a set of physical downlink control channel (physical downlink control channel, PDCCH) listening moments, which may include multiple time slots (eg, subframes or OFDM symbols) for transmitting paging DCI. SFN represents the system frame number of the paging frame, PF_offset represents the offset of the PF, for example, 1, 2, 3, etc., T represents the DRX cycle, and N represents the total number of PFs included in one DRX cycle, for example, T, T /2, T/4, T/8, T/16, etc., N _s represents the number of POs contained in a PF, for example, 1, 2, 4, etc., mod represents the modulo operation, ML_category_ID represents the ML capability classification identifier, or The value obtained after the modulo operation is performed on the ML capability classification identifier, div represents the integer division operation, and floor(x) represents the rounding down of x. Wherein, PF_offset, T, N, N _s are related parameters of the first paging message based on the ML capability classification identification, which are used by the access network device and the terminal device to determine the resource location of the first paging message, PF_offset, T, N , N _s may also be related parameters based on paging messages of a specific terminal device, that is to say, terminal devices with the same type of capability information have the same values of PF_offset, T, N, and N _s parameters.

Specifically, the values of PF_offset, T, N, and _Ns may be pre-configured on the access network equipment and terminal equipment, or may be configured by the network control equipment to the access network equipment and terminal equipment, or may be the access network equipment Configured to the terminal device, in an example, the access network device sends system information to the terminal device, such as system information block type 1 (system information block type 1, SIB1) or other system information blocks (system information block, SIB) ), the system information includes the values of PF_offset, T, N, and _Ns ; in another example, the access network device sends a dedicated message, such as a radio resource control (RRC) message, to the terminal device, the The proprietary message includes the values of PF_offset, T, N, _Ns .

In an example, assuming that the ML capability classification identifier is ML_category_ID=448835805669362, the position of the paging frame is determined according to formula 1=(T div N)*(ML_category_ID mod N)

Then the position of the paging frame may appear at SFN=(128*i)+114, (where i=0 to N, but SFN<=1024). For example, the position of the paging frame may be 128, 242, 498, 626, 754, 868, 982. According to formula 2, the PO is determined at the position of 9 subframes of the paging frame.

Step S304: The access network device sends a first paging message to the terminal device at the resource location.

Specifically, the first paging message includes an ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information.

Specifically, the ML capability information includes one or more of ML model ID, ML model information, ML model data size, ML model iteration times, AI framework, computing power, data storage capability, ML model area, and ML model object. ML model ID, used to identify the ML model, for example, ML models include linear regression, logistic regression, decision tree, naive bayes, K-nearest neighbors (k- nearest neighbors), support vector machines, deep neural network, random forest, etc.; or, used to represent a unique ML model, such as ML model ID 1 for AlexNet model, ML Model ID 2 represents the 16-layer visual geometry group 16 (VGG16) model, and ML model ID 3 represents the ResNet-152 model. ML model information, such as the file of the ML model, contains specific model parameters, as well as the connection relationship between the various model parameters. The ML model data size represents the size of the data amount of the ML model data, for example, the ML model data size is 500 Mbytes. The number of ML model iterations represents the number of times the ML model is updated. The updating of the ML model is to update the parameters of the ML model by using the training data. One update to the ML model parameters is called an iteration. The number of ML model iterations may also be the number of ML model training rounds, that is, the number of rounds for updating the ML model. For example, the number of 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. AI frameworks are used to indicate the allowed frameworks for ML models, such as TensorFlow, IBM-Watson, Spark-MLib, MindSpore, etc. Each AI framework supports different ML model representation methods. Computing power, also known as computing power, is used to indicate or evaluate the ability of the terminal device to process data, such as the output speed when the terminal device calculates the hash function. For example, the number of floating point operations per second can be used. operations per second, FLOPS). The computing power of the terminal device is positively related to the speed of processing data. For example, the greater the computing power, the faster the data processing speed. The computing power of the terminal device is related to the hardware configuration performance of the terminal device itself, the smoothness of the operating system and other factors. . The data storage capacity is used to indicate the capacity of the terminal device to store data, for example, 1G bytes. The ML model area is used to indicate the scope of application of the ML model, which can be represented in multiple ways. In one example, the ML model area can be represented by TAI, and each TAI is used to represent a tracking area; in another example In another example, the ML model area can be represented by a cell identifier, and each cell identifier is used to uniquely represent a cell; in another example, the ML model area can be represented by a public lands mobile network (PLMN) identifier. Indicates that each PLMN identifier is used to represent a PLMN. ML model objects, which are used to indicate objects to which the ML model applies, such as quality of service, quality of user experience, key performance indicators, etc. The quality of service may be the guaranteed bit rate of the quality of service (quality of service, QOS) stream, the maximum bit rate of the stream, the packet delay buffer, the priority, etc.; the user experience quality may be the user experience score, for example, it may be Average opinion score between 1 and 5; key performance indicators can be throughput, capacity, latency, reliability, availability, etc.

In an example, it is assumed that the access network device determines the resource location as: the location of the paging frame PF may be 128, 242, 498, 626, 754, 868, 982, and the PO is at the 9 subframe location of the paging frame, Then the access network device sends the first paging message to the terminal device at the resource location.

In a possible implementation manner, the second paging message includes a paging priority, and the access network device sends the first paging message to the terminal device according to the paging priority. The paging priority is used to indicate the priority of the first paging message, and the paging priority can be represented by a numerical value, for example, 1 means high priority, 2 means low priority, etc. Of course, the paging priority There may also be other representations, which are not limited by the embodiments in this application.

In an example, it is assumed that the second paging message includes the paging priority of the first paging message sent to the terminal equipment included in the first ML capability classification identifier is 1, and the paging priority of the first paging message sent to the terminal device included in the first ML capability classification identifier is 1, and the paging priority of the first paging message sent to the terminal device included in the first ML capability classification identifier is The paging priority of the first paging message sent by the included terminal device is 2. After receiving the second paging message, the access network device first identifies the included terminal device to the first ML capability classification according to the paging priority. The first paging message is sent, and then the first paging message is sent to the terminal equipment included in the second ML capability classification identifier.

In a possible implementation manner, the second paging message includes a paging area, and the access network device sends the first paging message to the terminal device in the paging area. The paging area is used to indicate the area where the access network device sends the first paging message to the terminal device, for example, a tracking area identity (tracking area identity, TAI) list, and each TAI represents a tracking area.

Step S305: The terminal device receives the first paging message from the access network device at the resource location.

Specifically, for the method for the terminal device to determine the resource location, reference may be made to the description in step S303, which will not be repeated here. The first paging message includes an ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information.

Step S306: The terminal device determines the ML capability information of the terminal device according to the ML capability classification identifier.

Specifically, the ML capability classification identifier corresponds to a set of ML capability information. In an example, assuming that the ML capability information is an ML model, it can be understood that the terminal devices that support the same ML model have the same ML capability classification identifier, which supports the same The end devices of the ML model are a group. That is to say, an ML capability classification identifier indicates that the ML capability information of this group of terminal devices is the same, that is, the terminal device determines whether the terminal device belongs to the group indicated by the ML capability classification identifier according to the ML capability classification identifier, that is, Whether the terminal device has the ML capability information indicated by the ML capability classification identifier. For example, the ML capability classification ID is 1001, the ML capability information corresponding to the ML capability classification ID 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, the terminal device 1 supports the linear regression model and the supported AI framework is TensorFlow, and the terminal device 2 If the linear regression model is supported and the supported AI framework is TensorFlow, then terminal device 1 and terminal device 2 have the same ML capability classification ID as 1001.

In an example, it is assumed that the ML capability classification identifier is 1001, the ML capability information corresponding to the ML capability classification identifier 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, and the terminal device 3 determines according to the ML capability classification identifier 1001 that the linear regression model is not supported. The regression model and the AI framework is not TensorFlow, then the terminal device 3 determines according to the ML capability classification identifier that it does not belong to the group indicated by the ML capability classification identifier, that is, it does not have the ML capability information corresponding to the ML capability classification identifier, then the terminal device 3 discards the ML capability classification identifier. A paging message.

In an example, it is assumed that the ML capability classification identifier is 1001, the ML capability information corresponding to the ML capability classification identifier 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, and the terminal device 4 determines that the linear regression is supported according to the ML capability classification identifier 1001. model and the AI framework is TensorFlow, then the terminal device 4 determines according to the ML capability classification identifier to belong to the group indicated by the ML capability classification identifier, that is, has the ML capability information corresponding to the ML capability classification identifier, then the terminal device 4 starts a further process, for example , if the terminal device 4 is in an idle state, the terminal device 4 initiates a random access process to access the network; if the terminal device 4 is in an inactive state, the terminal device 4 initiates an RRC connection recovery process to resume the suspended RRC connects or performs update based on the RAN notification area; if the terminal device 4 is in the connected state, the terminal device 4 acquires new system information.

Please refer to FIG. 4. FIG. 4 is a communication method provided by an embodiment of the present application. The method includes but is not limited to the following steps:

Step S401: The network device sends first configuration information to the access network device.

Specifically, the network device may be a core network device, a network control device, or other access network device, and the first configuration information includes an ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information. The related explanation that the ML capability classification identifier corresponds to a set of ML capability information can refer to step S306, which will not be repeated here.

Specifically, the first configuration information may be a paging configuration, which is used by the access network device to initiate a paging process for the terminal device; the first configuration information may also be a minimization of drive tests (MDT) measurement The configuration is used for the access network device to configure the related information of the MDT measurement to the terminal device, for example, the related parameters collected by the M1 measurement. MDT measurements include logged MDT (logged MDT) and immediate MDT (immediate MDT). Among them, the logged MDT refers to the MDT performed by the terminal device in the RRC idle state or inactive state. The terminal equipment stores the measurement results, and the terminal equipment reports the measurement results to the access network equipment when switching to the RRC connection state. The logged MDT measurement may measure at least one of: random access channel failure measurement, signal strength measurement, connection establishment failure measurement, radio link failure measurement, and the like. The immediate MDT refers to the MDT performed by the terminal device in the RRC connection state. Once the configured MDT reporting conditions are met, the terminal device reports the measurement result to the access network device. The instant MDT measurement may measure at least one of the following: data volume measurement of the terminal, throughput rate measurement, packet transmission delay measurement, packet loss rate measurement, processing delay measurement, and the like. M1 measurement refers to downlink signal quality measurement on the serving cell and/or the same-frequency neighbor cell, and/or the inter-frequency neighbor cell, and/or the inter-system neighbor cell.

Step S402: The access network device receives the first configuration information from the network device.

Specifically, the first configuration information includes an ML capability classification identifier, and the first configuration information may also be a paging configuration or an MDT configuration.

Step S403: The access network device determines, according to the ML capability classification identifier, a terminal device having ML capability information corresponding to the ML capability classification identifier.

Specifically, the ML capability classification identifier corresponds to a set of ML capability information, and the terminal devices having the ML capability information corresponding to the same ML capability classification identifier have the same ML capability classification identifier. In an example, it is assumed that the ML capability classification identifier 1001 corresponds to the ML capability The information is that the linear regression model is supported and the supported AI framework is TensorFlow. Assuming that terminal device 1 supports the linear regression model and the supported AI framework is TensorFlow, and terminal device 2 supports the linear regression model and the supported AI framework is TensorFlow, then terminal device 1 and The terminal equipment 2 has the same ML capability classification identifier.

In an example, the terminal device may determine the terminal device having the ML capability information corresponding to the ML capability classification identifier according to the following three manners.

The first way: the access network device is pre-configured with the corresponding relationship between the ML capability classification identifier and the terminal device. According to the corresponding relationship, the access network device can determine the terminal device corresponding to the ML capability classification identifier. For example, assuming that the ML capability classification ID is 1001, the ML capability information corresponding to the ML capability classification ID is that the linear regression model is supported and the supported AI framework is TensorFlow, the terminal device 1 supports the linear regression model and the supported AI framework is TensorFlow, and the terminal device 2 If the linear regression model is supported and the supported AI framework is TensorFlow, the corresponding relationship between the ML capability classification identifier and the terminal device can be: ML capability classification identifier 1001 corresponds to terminal device 1 and terminal device 2.

The second way: the terminal device actively reports its own ML capability classification identifier to the access network device. For example, the terminal device reports the ML capability classification identifier of the terminal device to the access network device through a UE assistance information (UE assistance information) message.

The third way: the access network device sends a request message to the terminal device, and the request message is used to request the terminal device to report its own ML capability classification identifier to the access network device. Correspondingly, after the terminal device receives the request message, Report its own ML capability classification identifier to the access network device. For example, the access network device sends a UE capability enquiry (UE capability enquiry) message to the terminal device, which is used to request to obtain the ML capability classification identifier of the terminal device. Further, after receiving the UE capability enquiry message, the terminal device sends an The network device sends a UE capability information (UE capability information) message, where the UE capability information message contains the ML capability classification identifier of the terminal device.

Step S404: The access network device sends the second configuration information to the above-mentioned terminal device.

Specifically, the second configuration information is used to indicate the type of data collected by the terminal device. The second configuration information may be an MDT measurement configuration, and is used for the access network device to configure related information of the MDT measurement to the terminal device, for example, the related parameters collected by the M1 measurement.

In an example, it is assumed that the access network device determines that the terminal device having the ML capability information corresponding to the ML capability classification identifier is terminal device 1 according to the pre-configured correspondence between the ML capability classification identifier and the terminal device, then the access network device sends the The terminal device 1 sends the second configuration information.

In an example, it is assumed that the access network device determines that the terminal devices having the ML capability information corresponding to the ML capability classification identifier are terminal device 1 and terminal device 2 according to the pre-configured correspondence between the ML capability classification identifier and the terminal device, then connect The network access device sends the second configuration information to the terminal device 1 and the terminal device 2 .

In the above method, the access network device sends the second configuration information to the terminal device having the ML capability information corresponding to the ML capability classification identifier according to the ML capability classification identifier, instead of sending the first configuration information to each terminal device in the wireless communication network 2. Configuration information. In this way, the problem of sending configuration information to a large number of terminal devices in the ML scenario can be solved, the signaling overhead between the access network device and the terminal device can be reduced, and the waste of resources can be avoided.

Please refer to FIG. 5. FIG. 5 is another communication method provided by an embodiment of the present application. The method includes but is not limited to the following steps:

Step S501: The terminal device sends the first machine learning ML capability information to the network device.

Specifically, this step is an optional step. The network device may be a core network device or a network control device. The first ML capability information corresponds to the first ML capability classification identifier, and the first ML capability information includes ML model ID, ML model information, ML model data size, ML model iteration times, AI framework, computing power, data storage capability, ML One or more of a model area and an ML model object. For details, reference may be made to the description in step S304, which will not be repeated here.

Optionally, the terminal device may also send a first ML capability classification identifier to the network device, where the first ML classification identifier is specified by the manufacturer of the terminal device or specified by the PLMN.

In an example, assuming that the first ML capability information of the terminal device is that the terminal device supports a linear regression model and the supported AI framework is TensorFlow, then the terminal device sends the first ML capability information to the network device, where the first ML capability information is Linear regression models are supported and the supported AI framework is TensorFlow.

Step S502: The network device receives the first ML capability information from the terminal device.

Specifically, this step is an optional step. The first ML capability information corresponds to the first ML capability classification identifier.

In an example, assuming that the first ML capability information supports a linear regression model and the supported AI framework is TensorFlow, the network device receives the first ML capability information from the terminal device, and the first ML capability information supports linear regression The model and supported AI framework is TensorFlow. Since the first ML capability information corresponds to the first ML capability classification identifier, the network device may determine the first ML capability classification identifier according to the first ML capability information. Assuming that the first ML capability information supports a linear regression model and the supported AI framework is TensorFlow, and the first ML capability classification identifier corresponding to the first ML capability information is 0001, the network device can determine the first ML capability according to the first ML capability information. The ML capability classification is identified as 0001.

Step S503: The network device determines the second ML capability classification identifier according to the first ML capability classification identifier.

Specifically, this step is an optional step. The second ML capability classification identifier may be the same as the first ML capability classification identifier, or may be different from the first ML capability classification identifier. For example, when the first ML capability classification identifier is an ML capability classification identifier specified by the manufacturer of the terminal device, the second ML capability classification identifier is an ML capability classification identifier specified by a PLMN allocated by the network device to the terminal device; when the When the first ML capability classification identifier is an ML capability classification identifier specified by a PLMN, the second ML capability classification identifier is an ML capability classification identifier specified by a new PLMN allocated by the network device to the terminal device.

In an example, it is assumed that the first ML capability classification identifier is specified by a terminal equipment manufacturer, the first ML capability classification identifier is 0001, and the first ML capability information corresponding to the first ML capability classification identifier 0001 is support The linear regression model and the supported AI framework is TensorFlow, the network device assigns a second ML capability classification identifier specified by PLMN to the terminal device, the second ML capability classification identifier is 1001, and the second ML capability classification identifier corresponds to the second ML capability classification identifier. The ML capability information is that the linear regression model is supported and the supported AI framework is TensorFlow.

Step S504: The network device sends the second ML capability classification identifier to the terminal device.

Specifically, this step is an optional step.

Step S505: The terminal device receives the second ML capability classification identifier from the network device.

Specifically, this step is an optional step.

Step S506: The terminal device sends the second ML capability classification identifier to the access network device.

Specifically, the manner in which the terminal device sends the second ML capability classification identifier to the access network device may include the following two:

The first way: the terminal device actively reports its own second ML capability classification identifier to the access network device. For example, the terminal device reports the second ML capability classification identifier of the terminal device to the access network device through a UE assistance information (UE assistance information) message.

The second way: the access network device sends a request message to the terminal device, and the request message is used to request the terminal device to report its second ML capability classification identifier to the access network device. Correspondingly, the terminal device receives the request message After that, report its own second ML capability classification identifier to the access network device. For example, the access network device sends a UE capability enquiry (UE capability enquiry) message to the terminal device, which is used to request to obtain the second ML capability classification identifier of the terminal device. Further, after receiving the UE capability enquiry message, the terminal device sends an The access network device sends a UE capability information (UE capability information) message, where the UE capability information message includes the second ML capability classification identifier of the terminal device.

Step S507: The access network device receives the second ML capability classification identifier from the terminal device.

Step S508: The access network device sends the ML capability mapping query information to the network device.

Specifically, the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used to request the network device to provide the second ML capability information corresponding to the second ML capability classification identifier.

In an example, assuming that the second ML capability classification identifier is 1001, the access network device sends ML capability mapping query information to the network device, where the ML capability mapping query information includes the second ML capability classification identifier as 1001.

Step S509: The network device receives the ML capability mapping query information from the access network device.

Specifically, the ML capability mapping query information includes the second ML capability classification identifier.

Step S510: The network device sends the ML capability mapping response information to the access network device.

Specifically, the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier. When the ML capability mapping response information includes multiple pieces of second ML capability information, the ML capability mapping response information may further include a second ML capability classification identifier corresponding to each second ML capability information.

In an example, it is assumed that when the ML capability mapping response information includes a second ML capability information, and the second ML capability information corresponding to the second ML capability classification identifier 1001 is a linear regression model supported and the supported AI framework is TensorFlow, then the network The device sends ML capability mapping response information to the access network device, where the ML capability mapping response information includes that the second ML capability information is that the linear regression model is supported and the supported AI framework is TensorFlow.

In an example, it is assumed that when the ML capability mapping response information includes two pieces of second ML capability information, the second ML capability information corresponding to the second ML capability classification identifier 1001 is a linear regression model supported and the supported AI framework is TensorFlow, the first The second ML capability information corresponding to the second ML capability classification identifier 1002 is that the random forest model is supported and the supported AI framework is Spark-MLib, then the network device sends the ML capability mapping response information to the access network device, and the ML capability mapping response information includes: The second ML capability information corresponding to the second ML capability classification identifier 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, and the second ML capability information corresponding to the second ML capability classification identifier 1002 is AI that supports the random forest model and is supported The framework is Spark-MLib.

Step S511: The access network device receives the ML capability mapping response information from the network device.

Specifically, the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.

Step S512: The access network device determines the correspondence between the second ML capability classification identifier and the second ML capability information according to the ML capability mapping response information.

In an example, it is assumed that the ML capability mapping response information includes a second ML capability information, the second ML capability information is that the linear regression model is supported and the supported AI framework is TensorFlow, then the access network device maps the response information according to the ML capability It is determined that the second ML capability information corresponding to the second ML capability classification identifier 1001 supports the linear regression model and the supported AI framework is TensorFlow.

In yet another example, it is assumed that the ML capability mapping response information includes two pieces of second ML capability information, and the ML capability mapping response information includes that the second ML capability information corresponding to the second ML capability classification identifier 1001 supports the linear regression model and supports the The AI framework is TensorFlow, the second ML capability information corresponding to the second ML capability classification identifier 1002 is that the random forest model is supported and the supported AI framework is Spark-MLib, then the access network device determines the second ML based on the ML capability mapping response information The second ML capability information corresponding to the capability classification identifier 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, and the second ML capability information corresponding to the second ML capability classification identifier 1002 is that the random forest model is supported and the supported AI framework is Spark -MLib.

Please refer to FIG. 6. FIG. 6 is another communication method provided by an embodiment of the present application. The method includes but is not limited to the following steps:

Step S601: The terminal device sends request information to the core network device.

Specifically, the request information includes one or more ML capability classification identifiers requested to be used by the terminal device. For example, the request information is used for the terminal device to initiate initial registration with the core network, or mobility update registration, or periodic registration update, or emergency registration.

In an example, it is assumed that the two ML capability class identifiers requested by the terminal device are 1001 and 1002, and the terminal device sends request information to the core network device, where the request information includes two ML capability class identifiers 1001 and 1002.

Step S602: The core network device receives the request information from the terminal device.

Specifically, the request information includes one or more ML capability classification identifiers requested to be used by the terminal device.

Step S603: The core network device determines the ML capability classification identifier that the terminal device is allowed to use.

Specifically, the ML capability classification identifier allowed to be used by the terminal device is used for the access network device to perform access control or resource allocation to the terminal device. Optionally, the determination by the core network device of the ML capability classification identifier that the terminal device is allowed to use may be implemented in the following manner: the core network device authenticates the terminal device, and determines the authority of the ML-related tasks that the terminal device can participate in (for example, only allowing the terminal device to use a certain ML model in a specific cell) to determine the ML capability class identifiers that the terminal device is allowed to use.

In an example, it is assumed that the request information includes two ML capability classification identifiers 1001 and 1002, the ML capability information corresponding to the ML capability classification identifier 1001 is that the linear regression model is supported and the supported AI framework is TensorFlow, and the ML capability classification identifier 1002 corresponds to The ML capability information supports the random forest model and the supported AI framework is Spark-MLib. When the core network device authenticates the terminal device, it is determined that only the terminal device is allowed to use the linear regression model in a specific cell. Since the ML capability classification identifier 1001 corresponds to If the ML capability information supports the linear regression model and the supported AI framework is TensorFlow, the terminal device determines that the ML capability classification identifier allowed for the terminal device is 1001.

Step S604: The core network device sends the ML capability classification identifier that the terminal device is allowed to use to the terminal device.

Step S605: The terminal device receives the ML capability classification identifier that is allowed to be used by the terminal device from the core network device.

In an example, it is assumed that the ML capability class identifier allowed for the terminal device to use is 1001, and the terminal device receives the ML capability class identifier 1001 that the terminal device allows to use from the core network device.

Step S606: The core network device sends the ML capability classification identifier that the terminal device is allowed to use to the access network device.

Specifically, when the core network device sends the ML capability classification identifier that the terminal device is allowed to use to the access network, it may be implemented in the following manner: In an example, the core network device sends an initial context establishment request message to the access network device, the initial The context establishment request message includes the ML capability classification identifier that the terminal device is allowed to use; in another example, the core network device sends a UE context modification request message to the access network device, and the UE context modification request message includes the ML that the terminal device is allowed to use. capability classification identifier; in another example, the core network device sends a handover request message to the access network device, the handover request message includes the ML capability classification identifier that the terminal device is allowed to use; in another example, the core network device sends a handover request message to the access network device. The access network device sends a downlink non-access stratum (non-access stratum, NAS) transmission message, where the downlink NAS transmission message includes the ML capability classification identifier that the terminal device is allowed to use; in another example, the core network device sends the access The network device sends a path transfer request confirmation message, where the path transfer request confirmation message includes the ML capability classification identifier that the terminal device is allowed to use.

Step S607: The access network device receives the ML capability classification identifier that is allowed to be used by the terminal device from the core network device.

Step S608: The access network device performs access control or resource allocation to the terminal device according to the ML capability classification identifier that is allowed to be used by the terminal device.

In an example, the access network device determines the ML capability classification identifiers allowed by each terminal device in the wireless communication network, and allocates resources and sends configuration information only to the terminal devices in the ML capability classification identifiers that are allowed to be used by the terminal device.

In another example, the access network device determines the ML capability classification identifiers allowed by each terminal device in the wireless communication network, and only accesses the terminal devices in the ML capability classification identifiers that are allowed to be used by the terminal devices.

Please refer to FIG. 7. FIG. 7 is another communication method provided by an embodiment of the present application. The method includes but is not limited to the following steps:

Step S701: The terminal device sends the ML capability classification identifier to the first network device.

Specifically, this step is an optional step, and the first network device may be an access network device.

Step S702: The first network device receives the ML capability classification identifier from the terminal device.

Specifically, this step is an optional step.

Step S703: The second network device sends the ML capability query information to the first network device.

Specifically, this step is an optional step. The second network device may be an access network device. Optionally, the first network device may be an access network device connected before the terminal device moves, and the second network device may be an access network device connected after the terminal device moves. Optionally, the first network device may also be a primary access network device in the MR-DC scenario, and the second network device may also be a secondary access network device in the MR-DC scenario. Optionally, the terminal device experiences RRC interruption, failure or suspension in the area served by the first access network device, then enters the area served by the second access network device, and initiates RRC connection recovery or RRC to the second network device When the connection is re-established, the first network device may be the first access network device, and the second network device may be the second access network device. The ML capability query information is used to query the ML capability information of the terminal device.

Step S704: The first network device receives the ML capability query information from the second network device.

Specifically, this step is an optional step. The ML capability query information is used by the second network device to query the ML capability information of the terminal device.

Step S705: The first network device sends the ML capability classification identifier to the second network device.

Specifically, the ML capability classification identifier corresponds to a group of ML capability information, which can be referred to as described in step S306, which will not be repeated here.

In an example, when the first network device is an access network device and the second network device is an access network device, the first network device may send a handover request message to the second network device, where the handover request message includes the terminal device or, the first network device may send an auxiliary node addition request message to the second network device, and the auxiliary node addition request message includes the ML capability classification identifier of the terminal device; or, the first network device may send an auxiliary node addition request message to the second network device. The network device sends a UE context recovery response message, where the UE context recovery response message includes the ML capability classification identifier of the terminal device.

Step S706: The second network device receives the ML capability classification identifier from the first network device.

Step S707: The second network device determines the ML capability information of the terminal device according to the ML capability classification identifier.

Specifically, the ML capability information includes ML model ID, ML model information, ML model data size, ML model iteration times, artificial intelligence (artificial intelligence, AI) framework, computing power, data storage capability, ML model area, and ML model objects. one or more of the. Reference may be made to the description in step S304, which is not repeated here.

In an example, it is assumed that the ML capability classification identifier is 1001, the ML capability classification identifier 1001 corresponds to the ML capability information that supports a linear regression model and the supported AI framework is TensorFlow, and the second network device determines the terminal device’s status according to the ML capability classification identifier 1001 The ML capability information is that the linear regression model is supported and the supported AI framework is TensorFlow.

The ML capability classification identifiers, the first ML capability classification identifiers and the second ML capability classification identifiers described in the above-mentioned Figures 3, 4, 5, 6 and 7 all refer to the ML capability classification identifiers of the terminal equipment. The capability information, the first ML capability information, and the second ML capability information all refer to the ML capability information of the terminal device.

The ML capability classification identifier described in the embodiment of FIG. 8 refers to the ML capability classification identifier of the network device, and the ML capability information corresponding to the ML capability classification identifier refers to the ML capability information of the network device.

Please refer to FIG. 8. FIG. 8 is another communication method provided by an embodiment of the present application. The method includes but is not limited to the following steps:

Step S801: The first network device sends a third ML capability classification identifier to the second network device.

Specifically, the first network device may be an access network device or a core network device, and the second network device may be an access network device or a core network device. The third ML capability classification identifier refers to the third ML capability classification identifier of the first network device, and the third ML capability classification identifier corresponds to a group of third ML capability information. For details, reference may be made to step S306, which will not be repeated here. The third ML capability information includes ML model ID, ML model information, ML model data size, ML model iteration times, artificial intelligence (AI) framework, computing power, data storage capability, ML model area, and ML model objects. For one or more items, refer to step S304 for details, and details are not repeated here.

In an example, assuming that the third ML capability classification identifier is 3001, the first network device sends the third ML capability classification identifier to the second network device, where the third ML capability classification identifier is 3001.

When the first network device is a core network device and the second network device is an access network device, in an example, the first network device sends a next-generation interface setup response message to the second network device, where the next-generation interface setup response message is includes a third ML capability classification identifier; in another example, the first network device sends an AMF configuration update message to the second network device, where the AMF configuration update message includes the third ML capability classification identifier.

When the first network device is an access network device and the second network device is a core network device, in an example, the first network device sends a next-generation interface establishment request message to the second network device, and the next-generation interface establishment request message In another example, the first network device sends a RAN configuration update message to the second network device, and the RAN configuration update message includes the third ML capability classification identifier.

When the first network device is an access network device and the second network device is an access network device, in an example, the first network device sends an XN interface establishment request message to the second network device, where the XN interface establishment request message includes a third ML capability classification identifier; in yet another example, the first network device sends an XN interface setup response message to the second network device, where the XN interface setup response message includes a third ML capability classification identifier; in yet another example , the first network device sends a next-generation radio access network node configuration update confirmation message to the second network device, where the next-generation radio access network node configuration update confirmation message includes a third ML capability classification identifier.

When the first network device is a DU and the second network device is a CU or a CU-CP, in an example, the first network device sends an F1 establishment request message to the second network device, where the F1 establishment request message includes a third ML capability classification identifier; in another example, the first network device sends a DU configuration update message to the second network device, where the DU configuration update message includes a third ML capability classification identifier.

When the first network device is a CU-UP and the second network device is a CU-CP, in an example, the first network device sends a CU-UP E1 establishment request message to the second network device, the CU-UP E1 establishment request message The message includes a third ML capability classification identifier; in another example, the first network device sends a CU-CP E1 establishment response message to the second network device, and the CU-CP E1 establishment response message includes the third ML capability classification identifier ; In another example, the first network device sends a CU-UP configuration update message to the second network device, where the CU-UP configuration update message includes a third ML capability classification identifier.

Step S802: The second network device receives the third ML capability classification identifier from the first network device.

Specifically, the third MLN capability classification identifier corresponds to a group of third ML capability information.

Step S803: The second network device sends a fourth ML capability classification identifier to the first network device.

Specifically, this step is an optional step. The fourth ML capability classification identifier refers to the ML capability classification identifier of the second network device, the fourth ML capability classification identifier corresponds to a set of fourth ML capability information, and the fourth ML capability information refers to the ML capability information of the second network device.

Step S804: The first network device receives the fourth ML capability classification identifier sent from the second network device.

Specifically, this step is an optional step.

In the above method, the network device and the network device exchange their own ML capability classification identifiers to avoid signaling overhead when one of the network devices does not support the ML capability information.

In the above embodiments, the terminal device and the access network device, and the terminal device and the network device may send related information based on the existing protocol stack, for example, send related information between the terminal device and the access network device based on an RRC message. Terminal equipment and network equipment can also use the new protocol stack to send related information. For example, a new protocol stack between terminal equipment and access network equipment is shown in Figure 9, based on a new computing bearer packet data convergence protocol-computing radio bearer (packet data convergence protocol computing radio bearer, PDCP-CRB) , the terminal equipment and the access network equipment use the DAP protocol to send relevant information, wherein the CRB is a kind of different from the existing signaling radio bearer (signaling radio bearer, SRB), data radio bearer (DRB) The new bearer, PDCP-CRB realizes ML-related data transmission, orderly transmission, encryption and decryption, repeatability detection, etc. DAP protocol realizes segmentation, sorting, and integrity protection of ML-related data between terminal equipment and access network equipment , encryption and decryption functions.

Related information can be sent between the access network device and the access network device based on an existing protocol stack, or a new protocol stack can be used to send related information, as shown in FIG. 10 . HDAPb supports data transmission (eg, data segmentation, data sequencing) between access network devices, and data security (eg, data integrity protection, data encryption, data decryption) and other functions. HDAPb uses the service provided by the Xn application protocol (Xn application protocol, XnAP), that is, the HDAPb message is carried in the XnAP message.

Related information can be sent between the access network device and the core network device based on an existing protocol stack, or a new protocol stack can be used to send related information, as shown in FIG. 11 . The high data analytics protocol annex (HDAPa) supports data transmission between access network equipment and core network equipment (eg, data segmentation, data sequencing), and data security (eg, data integrity protection, data encryption, data decryption) and other functions. HDAPa uses the services provided by the next generation application protocol (NGAP), that is, HDAPa messages are carried in NGAP messages.

The terminal device and the core network device can send the relevant information based on the existing protocol stack, or can use the new protocol stack to send the relevant information, as shown in FIG. 12 . The high-level data analysis protocol (HDAP) is used to send relevant information between the terminal equipment and the core network equipment, so as to realize the functions of segmentation, sorting, integrity protection, encryption and decryption of relevant information.

DU and CU/CU-CP can send related information based on the existing protocol stack, or can use a new protocol stack to send related information, as shown in Figure 13. The high data analytics protocol type c (HDAPc) protocol supports data transmission between DU and CU/CU-CP (such as data segmentation, data ordering), and data security (such as data integrity protection, data encryption, data decryption) and other functions. HDAPc messages may be carried in F1AP messages.

It can be understood that, in order to implement the functions in the foregoing embodiments, the network device and the terminal device include corresponding hardware structures and/or software modules for performing each function. Those skilled in the art should easily realize that the units and method steps of each example described in conjunction with the embodiments disclosed in the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is performed by hardware or computer software-driven hardware depends on the specific application scenarios and design constraints of the technical solution.

FIG. 14 and FIG. 15 are schematic structural diagrams of possible communication apparatuses provided by embodiments of the present application. These communication apparatuses can be used to implement the functions of the terminal equipment or the network equipment in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments. In the embodiment of the present application, the communication device may be the core network device 1001, the first access network device 1002, the second access network device 1003, or the terminal device 1004 as shown in FIG. 1, or may be applied to the terminal A module (such as a chip) of a device or network device.

As shown in FIG. 14 , the communication device 1400 includes a processing unit 1401 and a transceiver unit 1402 . The communication apparatus 1400 is configured to implement the functions of the terminal device or the network device in the method embodiment shown in FIG. 3 , FIG. 4 , FIG. 5 , FIG. 6 , FIG. 7 or FIG. 8 .

When the communication apparatus 1400 is used to implement the function of the access network device in the method embodiment shown in FIG. 3 : the processing unit 1401 is configured to determine the resource location of the first paging message according to the machine learning ML capability classification identifier; the transceiver unit 1402 is used to send the first paging message to the terminal device at the resource location, where the first paging message includes the ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information.

In a possible implementation manner, the transceiver unit 1402 is further configured to receive a second paging message from a network device, where the second paging message includes the ML capability classification identifier.

In another possible implementation manner, the second paging message includes a paging priority; the transceiver unit 1402 is further configured to send the first paging priority to the terminal device according to the paging priority paging message.

In another possible implementation manner, the second paging message includes a paging area; the transceiver unit 1402 is further configured to send the first paging message to the terminal device in the paging area .

When the communication apparatus 1400 is used to implement the function of the terminal device in the method embodiment shown in FIG. 3 : the transceiver unit 1402 is configured to receive a first paging message from an access network device, where the first paging message includes a machine Learning the ML capability classification identifier, where the ML capability classification identifier corresponds to a group of ML capability information; the processing unit 1401 is configured to determine the ML capability information of the device according to the ML capability classification identifier.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 3 , and details are not repeated here.

When the communication apparatus 1400 is used to implement the function of the access network device in the method embodiment shown in FIG. 4 : the transceiver unit 1402 is configured to receive first configuration information from the network device, where the first configuration information includes machine learning ML A capability classification identifier, the ML capability classification identifier corresponds to a group of ML capability information; the processing unit 1401 is configured to determine, according to the ML capability classification identifier, a terminal device having the ML capability information corresponding to the ML capability classification identifier; Unit 1402 is further configured to send second configuration information to the terminal device, where the second configuration information is used to indicate the type of data collected by the terminal device.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 4 , and details are not repeated here.

When the communication apparatus 1400 is used to implement the function of the access network device in the method embodiment shown in FIG. 5 : the transceiver unit 1402 is configured to send machine learning ML capability mapping query information to the network device, where the ML capability mapping query information includes The second ML capability classification identifier, the ML capability mapping query information is used to request the second ML capability information corresponding to the second ML capability classification identifier; the transceiver unit 1402 is further configured to receive the ML from the network device capability mapping response information, the ML capability mapping response information includes the correspondence relationship of the second ML capability information corresponding to the second ML capability classification identifier; the processing unit 1401 is configured to determine the first ML capability mapping response information according to the ML capability mapping response information The correspondence between the second ML capability classification identifier and the second ML capability information.

In a possible implementation manner, the transceiver unit 1402 is further configured to receive the second ML capability classification identifier from the terminal device.

When the communication apparatus 1400 is used to implement the function of the network device in the method embodiment shown in FIG. 5 : the processing unit 1401 is configured to receive the machine learning ML capability mapping query information from the access network device according to the transceiver unit 1802 , the ML The capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used by the apparatus to request the network device to provide the second ML capability information corresponding to the second ML capability classification identifier; the transceiver unit Step 1402 is further configured to send ML capability mapping response information to the access network device, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.

In a possible implementation manner, the transceiver unit 1402 is further configured to receive first ML capability information from the terminal device, where the first ML capability information corresponds to the first ML capability classification identifier; the processing unit is further configured to The second ML capability classification identifier is determined according to the first ML capability classification identifier, and the second ML capability classification identifier is sent to the terminal device.

When the communication apparatus 1400 is used to implement the function of the terminal device in the method embodiment shown in FIG. 5 : the processing unit 1401 is configured to send the first machine learning ML capability information to the network device according to the transceiver unit 1402, the first ML capability The information corresponds to the first ML capability classification identifier; the transceiver unit 1402 is configured to receive the second ML capability classification identifier from the network device.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 5 , and details are not repeated here.

When the communication apparatus 1400 is used to implement the function of the core network device in the method embodiment shown in FIG. 6 : the processing unit 1401 is configured to receive request information from the terminal device according to the transceiver unit, where the request information includes the request from the terminal device One or more machine learning ML capability classification identifiers used; the transceiver unit 1402 is configured to determine the ML capability classification identifiers that are allowed to be used by the terminal device, and send the terminal device and the access network device to allow the terminal device to use The ML capability classification identifier used by the device; the ML capability classification identifier allowed to be used by the terminal device is used for the access network device to perform access control or resource allocation to the terminal device.

When the communication apparatus 1400 is used to implement the function of the terminal device in the method embodiment shown in FIG. 6 : the processing unit 1401 is configured to send request information to the core network device according to the transceiver unit, where the request information includes the information requested by the apparatus to be used. One or more machine learning ML capability classification identifiers; the transceiver unit 1402 is configured to receive an ML capability classification identifier that is allowed to be used by the device from the core network equipment, and the ML capability classification identifier that is allowed to be used by the device It is used for the access network equipment to perform access control or resource allocation to the apparatus.

When the communication apparatus 1400 is used to implement the function of the access network device in the method embodiment shown in FIG. 6: the transceiver unit 1402 is configured to receive a machine learning ML capability classification identifier that is allowed to be used by the terminal device from the core network device;

The processing unit 1401 is configured to perform access control or resource allocation to the terminal device according to the machine learning ML capability classification identifier that is allowed to be used by the terminal device.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 6 , and details are not repeated here.

When the communication apparatus 1400 is used to implement the function of the second network device in the method embodiment shown in FIG. 7 : the transceiver unit 1402 is configured to receive a machine learning ML capability classification identifier from the first network device, the ML capability classification identifier Corresponding to a group of ML capability information; the processing unit 1401 is configured to determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, the transceiver unit 1402 is further configured to send ML capability query information to the first network device, where the ML capability query information is used to query the ML capability information of the terminal device.

When the communication apparatus 1400 is used to implement the function of the first network device in the method embodiment shown in FIG. 7: the processing unit 1401 is configured to send the machine learning ML capability classification identifier to the second network device according to the transceiver unit 1402, the ML The capability classification identifier corresponds to a group of ML capability information; the ML capability classification identifier is used by the second network device to determine the ML capability information of the terminal device according to the ML capability classification identifier.

In a possible implementation manner, the transceiver unit 1402 is further configured to receive ML capability query information from the second network device, where the ML capability query information is used to query the ML capability information of the terminal device.

In another possible implementation manner, the transceiver unit 1402 is further configured to receive the ML capability classification identifier from the terminal device.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 7 , and details are not repeated here.

When the communication apparatus 1400 is used to implement the function of the first network device in the method embodiment shown in FIG. 8: the processing unit 1401 is configured to send a third ML capability classification identifier to the second network device through the transceiver unit 1402, the first The three ML capability classification identifiers correspond to a set of third ML capability information; the transceiver unit 1402 is further configured to receive a fourth ML capability classification identifier from the second network device, where the fourth ML capability classification identifier corresponds to a set of fourth ML capability classification identifiers. ML capability information.

When the communication apparatus 1400 is used to implement the function of the second network device in the method embodiment shown in FIG. 8 : the processing unit 1401 is configured to receive the third ML capability classification identifier from the second network device through the transceiver unit 1402, the said The third ML capability classification identifier corresponds to a group of third ML capability information; the transceiver unit 1402 is further configured to send a fourth ML capability classification identifier to the second network device, where the fourth ML capability classification identifier corresponds to a group of fourth ML capability information.

More detailed descriptions about the above-mentioned processing unit 1401 and the transceiver unit 1402 can be obtained directly by referring to the relevant descriptions in the method embodiment shown in FIG. 8 , and details are not repeated here.

As shown in FIG. 15 , the communication apparatus 1500 includes a processor 1501 and a transceiver 1503 . The processor 1501 and the transceiver 1503 are coupled to each other. It can be understood that the transceiver 1503 can be a transceiver or an input-output interface. Optionally, the communication device 1500 may further include a memory 1502 for storing instructions executed by the processor 1501 or input data required by the processor 1501 to execute the instructions or data generated after the processor 1501 executes the instructions.

When the communication apparatus 1500 is used to implement the method shown in FIG. 3 , FIG. 4 , FIG. 5 , FIG. 6 , FIG. 7 or FIG. 8 , the processor 1501 is used to implement the functions of the above processing unit 1401 , and the transceiver 1503 is used to implement the above mentioned functions. Function of the transceiver unit 1402.

When the above communication device is a chip applied to a terminal device, the terminal device chip implements the functions of the terminal device in the above method embodiments. The terminal device chip receives information from other modules (such as a radio frequency module or an antenna) in the terminal device, and the information is sent by the network device to the terminal device; or, the terminal device chip sends information to other modules (such as a radio frequency module or an antenna) in the terminal device antenna) to send information, the information is sent by the terminal equipment to the network equipment.

When the above communication device is a chip applied to a network device, the network device chip implements the functions of the network device in the above method embodiments. The network device chip receives information from other modules (such as a radio frequency module or an antenna) in the network device, and the information is sent by the terminal device to the network device; or, the network device chip sends information to other modules in the network device (such as a radio frequency module or an antenna). antenna) to send information, the information is sent by the network equipment to the terminal equipment.

It can be understood that the processor in the embodiments of the present application may be a central processing unit (Central Processing Unit, CPU), and may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application-specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (Field Programmable Gate Array, FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

The method steps in the embodiments of the present application may be implemented in a hardware manner, or may be implemented in a manner in which a processor executes software instructions. Software instructions may be composed of corresponding software modules, and software modules may be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory memory, registers, hard disk, removable hard disk, CD-ROM or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and storage medium may reside in an ASIC. Alternatively, the ASIC may be located in a network device or in an end device. Of course, the processor and the storage medium may also exist in the network device or the terminal device as discrete components.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are executed in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, network equipment, user equipment, or other programmable apparatus. The computer program or instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instructions may be downloaded from a website site, computer, A server or data center transmits by wire or wireless to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server, data center, or the like that integrates one or more available media. The usable media may be magnetic media, such as floppy disks, hard disks, magnetic tapes; optical media, such as digital video discs; and semiconductor media, such as solid-state drives.

In the various embodiments of the present application, if there is no special description or logical conflict, the terms and/or descriptions between different embodiments are consistent and can be referred to each other, and the technical features in different embodiments are based on their inherent Logical relationships can be combined to form new embodiments.

In this application, "at least one" means one or more, and "plurality" means two or more. "And/or", which describes the relationship between the associated objects, means that there can be three relationships, for example, A and/or B, which can mean: A exists alone, A and B exist at the same time, and B exists alone, where A, B can be singular or plural. In the text description of this application, the character "/" generally indicates that the related objects are a kind of "or" relationship; in the formula of this application, the character "/" indicates that the related objects are a kind of "division" Relationship.

It can be understood that, the various numbers and numbers involved in the embodiments of the present application are only for the convenience of description, and are not used to limit the scope of the embodiments of the present application. The size of the sequence numbers of the above processes does not imply the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic.

Claims

A communication method, comprising:

The access network device determines the resource location of the first paging message according to the machine learning ML capability classification identifier;

The access network device sends the first paging message to the terminal device at the resource location, where the first paging message includes the ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capabilities information.
The method according to claim 1, wherein before the access network device determines the resource location of the first paging message according to the machine learning ML capability classification identifier, the method further comprises:

The access network device receives a second paging message from the network device, where the second paging message includes the ML capability classification identifier.
The method according to claim 2, wherein the second paging message includes a paging priority; the access network device sends the first paging message to the terminal device at the resource location, include:

The access network device sends the first paging message to the terminal device according to the paging priority.
The method according to claim 2 or 3, wherein the second paging message includes a paging area; and the access network device sends the first paging message to the terminal device at the resource location ,include:

The access network device sends the first paging message to the terminal device within the paging area.
A communication method, comprising:

The terminal device receives a first paging message from an access network device, where the first paging message includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information;

The terminal device determines the ML capability information of the terminal device according to the ML capability classification identifier.
A communication method, comprising:

The access network device receives first configuration information from the network device, where the first configuration information includes a machine learning ML capability classification identifier, and the ML capability classification identifier corresponds to a group of ML capability information;

The access network device determines, according to the ML capability classification identifier, a terminal device having ML capability information corresponding to the ML capability classification identifier;

The access network device sends second configuration information to the terminal device, where the second configuration information is used to indicate the type of data collected by the terminal device.
A communication method, comprising:

The access network device sends machine learning ML capability mapping query information to the network device, where the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used to request that the second ML capability classification identifier corresponds to the second ML capability information;

The access network device receives ML capability mapping response information from the network device, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier;

The access network device determines the correspondence between the second ML capability classification identifier and the second ML capability information according to the ML capability mapping response information.
The method according to claim 7, wherein before the access network device sends the machine learning ML capability mapping query information to the network device, the method further comprises:

The access network device receives the second ML capability classification identifier from the terminal device.
A communication method, comprising:

The network device receives machine learning ML capability mapping query information from the access network device, the ML capability mapping query information includes a second ML capability classification identifier, and the ML capability mapping query information is used by the access network device to request the The network device provides second ML capability information corresponding to the second ML capability classification identifier;

The network device sends ML capability mapping response information to the access network device, where the ML capability mapping response information includes second ML capability information corresponding to the second ML capability classification identifier.
The method according to claim 9, wherein before the network device receives the machine learning ML capability mapping query information from the access network device, the method further comprises:

receiving, by the network device, first ML capability information from the terminal device, where the first ML capability information corresponds to the first ML capability classification identifier;

The network device determines a second ML capability classification identifier according to the first ML capability classification identifier, and sends the second ML capability classification identifier to the terminal device.
A communication method, comprising:

The terminal device sends first machine learning ML capability information to the network device, where the first ML capability information corresponds to the first ML capability classification identifier;

The terminal device receives the second ML capability classification identifier from the network device.
A communication method, comprising:

The core network device receives request information from the terminal device, where the request information includes one or more machine learning ML capability classification identifiers requested by the terminal device;

The core network device determines the ML capability classification identifier that the terminal device is allowed to use, and sends the ML capability classification identifier that the terminal device is allowed to use to the terminal device and the access network device; the terminal device is allowed to use the ML capability classification identifier. The ML capability classification identifier is used for the access network device to perform access control or resource allocation to the terminal device.
A communication method, comprising:

The terminal device sends request information to the core network device, where the request information includes one or more machine learning ML capability classification identifiers requested by the terminal device;

The terminal device receives from the core network device an ML capability classification identifier that is permitted to be used by the terminal device, and the ML capability classification identifier that is permitted to be used by the terminal device is used by the access network device to connect to the terminal device. access control or resource allocation.
A communication method, comprising:

The access network device receives the machine learning ML capability classification identifier that the terminal device is allowed to use from the core network device;

The access network device performs access control or resource allocation on the terminal device according to the machine learning ML capability classification identifier that is allowed to be used by the terminal device.
A communication method, comprising:

The second network device receives the machine learning ML capability classification identifier from the first network device, where the ML capability classification identifier corresponds to a group of ML capability information;

The second network device determines the ML capability information of the terminal device according to the ML capability classification identifier.
The method according to claim 15, wherein before the second network device receives the machine learning ML capability classification identifier from the first network device, the method further comprises:

The second network device sends ML capability query information to the first network device, where the ML capability query information is used to query the ML capability information of the terminal device.
A communication method, comprising:

The first network device sends a machine learning ML capability classification identifier to the second network device, where the ML capability classification identifier corresponds to a group of ML capability information; the ML capability classification identifier is used by the second network device to classify according to the ML capability The identifier determines the ML capability information of the terminal device.
The method according to claim 17, wherein before the first network device sends the machine learning ML capability classification identifier to the second network device, the method further comprises:

The first network device receives ML capability query information from the second network device, where the ML capability query information is used to query the ML capability information of the terminal device.
The method according to claim 17 or 18, wherein before the first network device sends the machine learning ML capability classification identifier to the second network device, the method further comprises:

The first network device receives the ML capability classification identifier from the terminal device.
A communication device, characterized in that it comprises a device for performing any one of claims 1 to 4, 5, 6, 7 to 8, 9 to 10, 11, 12, 14, 14, 15 to 16 or 17-19 A unit of the method described in item.
A communication device, characterized in that it includes a processor and a transceiver, and the transceiver is configured to receive signals from other communication devices other than the communication device and transmit to the processor or transmit signals from the processor. The signal is sent to other communication devices than the communication device, and the processor is used to implement the claims 1 to 4, 5, 6, 7 to 8, 9 to 10, 11, 12 by means of a logic circuit or executing a computer program , 14, 14, 15 to 16 or the method of any of 17-19.
A computer-readable storage medium, characterized in that, the computer-readable storage medium stores a computer program, and when the computer program is executed, the computer program as claimed in claims 1 to 4, 5, 6, 7 to 8, and 9 is implemented. to the method of any one of 10, 11, 12, 14, 14, 15 to 16, or 17-19.