WO2022161230A1 - 通信***中的模型更新方法、装置及存储介质 - Google Patents
通信***中的模型更新方法、装置及存储介质 Download PDFInfo
- Publication number
- WO2022161230A1 WO2022161230A1 PCT/CN2022/072837 CN2022072837W WO2022161230A1 WO 2022161230 A1 WO2022161230 A1 WO 2022161230A1 CN 2022072837 W CN2022072837 W CN 2022072837W WO 2022161230 A1 WO2022161230 A1 WO 2022161230A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- machine learning
- information
- communication environment
- access network
- network device
- Prior art date
Links
- 238000004891 communication Methods 0.000 title claims abstract description 891
- 238000000034 method Methods 0.000 title claims abstract description 155
- 238000010801 machine learning Methods 0.000 claims abstract description 900
- 238000013507 mapping Methods 0.000 claims abstract description 500
- 230000007613 environmental effect Effects 0.000 claims description 105
- 238000004590 computer program Methods 0.000 claims description 41
- 230000008859 change Effects 0.000 claims description 17
- 230000004044 response Effects 0.000 claims description 7
- 230000000694 effects Effects 0.000 abstract description 36
- 238000012545 processing Methods 0.000 description 16
- 238000012549 training Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 15
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 12
- 230000009286 beneficial effect Effects 0.000 description 11
- 238000005562 fading Methods 0.000 description 11
- 238000007726 management method Methods 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 5
- 230000007774 longterm Effects 0.000 description 4
- 239000000835 fiber Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
- H04L41/082—Configuration setting characterised by the conditions triggering a change of settings the condition being updates or upgrades of network functionality
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0806—Configuration setting for initial configuration or provisioning, e.g. plug-and-play
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
- H04L41/0816—Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W76/00—Connection management
- H04W76/20—Manipulation of established connections
- H04W76/27—Transitions between radio resource control [RRC] states
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W76/00—Connection management
- H04W76/10—Connection setup
- H04W76/19—Connection re-establishment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Definitions
- the present application relates to the field of communications, and in particular, to a model updating method, device and storage medium in a communication system.
- the machine learning model can be applied to channel estimation, channel prediction, channel state information feedback, etc. It can be seen that the model performance of the machine learning model affects the performance of the communication system.
- Machine learning models are usually learned from large amounts of training data. Because the communication environment is prone to change, the performance of the original machine learning model in the changed communication environment is general. It is necessary to obtain training data in the changed communication environment and retrain the machine learning model.
- the method of retraining the machine learning model multiple times is not efficient and cannot be applied to communication systems with high real-time requirements. Therefore, how to improve the update efficiency and effect of machine learning models in communication systems has become an urgent problem to be solved.
- the present application provides a model update method, device and storage medium in a communication system, which are used to solve the problem of how to improve the update efficiency and effect of a machine learning model in a communication system.
- the present application provides a method for updating a model in a communication system, which is applied to an access network device.
- the access network device is configured with a first mapping relationship set and multiple machine learning models associated with the first mapping relationship set.
- the first mapping relationship set includes the mapping relationship between multiple communication environment information and multiple machine learning models, and the method includes:
- the method further includes:
- the first mapping relationship set among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a first machine learning model set, including:
- a machine learning model is searched among multiple machine learning models to form a first machine learning model set.
- acquiring the communication environment information on the terminal side includes:
- the third information sent by the terminal is received, and in response to the third information, the second information is obtained, where the third information is used to instruct the access network device to obtain the second information.
- the method further includes:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines if the communication environment has changed; and/or,
- each environmental parameter in the communication environment information at the last moment on the terminal side is determined whether a communication environment occurs Variety.
- the method further includes:
- the fourth information includes the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal;
- the machine learning model is searched and the first machine learning model set is formed.
- the method further includes:
- the method further includes:
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the identification information of the machine learning model used when the terminal communicates with another access network device.
- the second The mapping relationship set is a subset of the mapping relationship associated with the terminal in the mapping relationship between the multiple communication environment information configured on the source access network device and the multiple machine learning models;
- the first mapping relationship set among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a first machine learning model set, including:
- a machine learning model suitable for the current communication environment is searched to form a first machine learning model set.
- the method further includes:
- the present application provides a method for updating a model of a communication system, which is applied to a core network device, where a third mapping relationship set and multiple machine learning models associated with the third mapping relationship set are preconfigured on the core network device,
- the third set of mapping relationships includes mapping relationships between multiple communication environment information and multiple machine learning models, and the method includes:
- the access network device sends seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the method further includes:
- mapping relationships among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a second set of machine learning models, including:
- a machine learning model is searched among the multiple machine learning models to form a second machine learning model set.
- Eighth information sent by the access network device is received, where the eighth information includes communication environment information on the access network device side and communication environment information on the terminal side.
- the method further includes:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines if the communication environment has changed; and/or,
- each environmental parameter in the communication environment information at the last moment on the terminal side is determined whether a communication environment occurs Variety.
- the method further includes:
- the ninth information is used to indicate that the communication environment changes, the terminal is switched between cells, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is reestablished, and the ninth information includes the following One or more pieces of information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device;
- a machine learning model suitable for the current communication environment is searched to form a second set of machine learning models, including:
- the ninth information it is determined that the communication environment has changed
- a machine learning model is searched to form a second machine learning model set.
- the method further includes:
- mapping relationship set sent by the multiple access network devices and the multiple machine learning models associated with the first mapping relationship set, a third mapping relationship set and multiple machine learning model.
- the present application provides a method for updating a model in a communication system, which is applied to a terminal, and the method includes:
- the machine learning models in the first machine learning model set are deployed according to the first information, or the machine learning models in the second machine learning model set are deployed according to the seventh information.
- the method before receiving the first information or the seventh information returned by the access network device, the method further includes:
- the method further includes:
- the fourth information is sent to the access network device, wherein, if the fourth information indicates that the communication environment on the terminal side has changed, the fourth information includes the communication environment information on the terminal side and / or the identification information of the machine learning model requested by the terminal.
- the terminal side is configured with a fourth mapping relationship set, and the fourth mapping relationship set includes the mapping relationship between multiple communication environment information associated with the terminal and multiple machine learning models, according to whether the communication environment on the terminal side occurs or not. change, and send fourth information to the access network device, including:
- the present application provides a model updating apparatus in a communication system, which is applied to an access network device, and the access network device is configured with a first mapping relationship set and multiple machine learning sets associated with the first mapping relationship set model, the first mapping relationship set includes mapping relationships between multiple communication environment information and multiple machine learning models, and the apparatus includes a memory, a transceiver, and a processor:
- a transceiver for sending and receiving data under the control of the processor
- a processor that reads the computer program in memory and performs the following operations:
- processor is also used to perform the following operations:
- the first mapping relationship set among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a first machine learning model set, including:
- a machine learning model is searched among multiple machine learning models to form a first machine learning model set.
- obtain the communication environment information on the terminal side including:
- the third information sent by the terminal is received, and in response to the third information, the second information is obtained, where the third information is used to instruct the access network device to obtain the second information.
- processor is also used to perform the following operations:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines if the communication environment has changed; and/or,
- each environmental parameter in the communication environment information at the last moment on the terminal side is determined whether a communication environment occurs Variety.
- processor is also used to perform the following operations:
- the fourth information includes the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal;
- the machine learning model is searched and the first machine learning model set is formed.
- processor is also used to perform the following operations:
- processor is also used to perform the following operations:
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the identification information of the machine learning model used when the terminal communicates with another access network device.
- the second The mapping relationship set is a subset of the mapping relationship associated with the terminal in the mapping relationship between the multiple communication environment information configured on the source access network device and the multiple machine learning models;
- the first mapping relationship set among the multiple machine learning models, find a machine learning model suitable for the current communication environment and form the first machine learning model set, including:
- a machine learning model suitable for the current communication environment is searched to form a first machine learning model set.
- receive the second information sent by the terminal including:
- the present application provides a model updating device for a communication system, which is applied to core network equipment, where a third mapping relationship set and multiple machine learning models associated with the third mapping relationship set are preconfigured on the core network equipment,
- the third set of mapping relationships includes mapping relationships between multiple pieces of communication environment information and multiple machine learning models
- the apparatus includes a memory, a transceiver, and a processor:
- a transceiver for sending and receiving data under the control of the processor
- a processor that reads the computer program in memory and performs the following operations:
- the access network device sends seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- processor is also used to perform the following operations:
- mapping relationships among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a second set of machine learning models, including:
- a machine learning model is searched among the multiple machine learning models to form a second machine learning model set.
- Eighth information sent by the access network device is received, where the eighth information includes communication environment information on the access network device side and communication environment information on the terminal side.
- processor is also used to perform the following operations:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines if the communication environment has changed; and/or,
- each environmental parameter in the communication environment information at the last moment on the terminal side is determined whether a communication environment occurs Variety.
- the processor is further configured to perform the following operations:
- the ninth information is used to indicate that the communication environment changes, the terminal is switched between cells, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is reestablished, and the ninth information includes the following One or more pieces of information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device;
- a machine learning model suitable for the current communication environment is searched to form a second set of machine learning models, including:
- the ninth information it is determined that the communication environment has changed
- a machine learning model is searched to form a second machine learning model set.
- processor is also used to perform the following operations:
- mapping relationship set sent by the multiple access network devices and the multiple machine learning models associated with the first mapping relationship set, a third mapping relationship set and multiple machine learning model.
- the present application provides a model updating device in a communication system, applied to a terminal, the device includes a memory, a transceiver and a processor:
- a transceiver for sending and receiving data under the control of the processor
- a processor that reads the computer program in memory and performs the following operations:
- the machine learning models in the first machine learning model set are deployed according to the first information, or the machine learning models in the second machine learning model set are deployed according to the seventh information.
- the processor before receiving the first information or the seventh information returned by the access network device, the processor is further configured to perform the following operations:
- processor is also used to perform the following operations:
- the information includes communication environment information on the terminal side and/or identification information of the machine learning model requested by the terminal.
- the terminal side is configured with a fourth mapping relationship set, and the fourth mapping relationship set includes the mapping relationship between multiple communication environment information associated with the terminal and multiple machine learning models, according to the communication environment on the terminal side. Whether there is a change, send fourth information to the access network device, including:
- the present application provides a model updating apparatus in a communication system, which is applied to an access network device, where a first mapping relationship set and multiple machines associated with the first mapping relationship set are configured on the access network device Learning model, the first mapping relationship set includes mapping relationships between multiple communication environment information and multiple machine learning models, and the device includes:
- a search unit configured to determine that the communication environment has changed, and according to the first set of mapping relationships, search for a machine learning model suitable for the current communication environment among multiple machine learning models and form a first set of machine learning models;
- a deployment unit configured to deploy the machine learning model in the first machine learning model set on the access network device, and/or, a sending unit, configured to send first information to the terminal, where the first information is used to indicate that the terminal is paired with The machine learning models in the first set of machine learning models are deployed.
- the device further includes:
- an acquisition unit configured to acquire the communication environment information on the access network device side and the communication environment information on the terminal side;
- the lookup unit is specifically used to:
- a machine learning model is searched among multiple machine learning models to form a first machine learning model set.
- the device further includes:
- a first receiving unit configured to receive second information sent by the terminal, where the second information includes communication environment information on the terminal side; or, receive third information sent by the terminal, and acquire the second information in response to the third information , where the third information is used to instruct the access network device to obtain the second information.
- the device further includes:
- the first determining unit is configured to, according to each environmental parameter in the communication environment information at the last moment on the access network device side, each environmental parameter in the communication environment information at the current moment on the access network side, and a plurality of environmental parameters and the The mapping relationship of preset thresholds is used to determine whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and the mapping relationship between multiple environmental parameters and preset thresholds to determine whether the communication environment has changed.
- the device further includes:
- the second receiving unit is configured to receive fourth information sent by the terminal, wherein, if the fourth information indicates that the communication environment on the terminal side has changed, the fourth information includes the communication environment information on the terminal side and/or the information requested by the terminal. Identification information of the machine learning model;
- a second determining unit configured to determine whether the communication environment has changed according to the fourth information
- the lookup unit is specifically used to:
- the machine learning model is searched and the first machine learning model set is formed.
- the device further includes:
- the third determining unit is configured to determine that the communication environment changes when the terminal switches between cells.
- the device further includes:
- a third receiving unit configured to receive fifth information sent by the source access network device of the terminal
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the identification information of the machine learning model used when the terminal communicates with another access network device.
- the second The mapping relationship set is a subset of the mapping relationship associated with the terminal in the mapping relationship between the multiple communication environment information configured on the source access network device and the multiple machine learning models;
- a first updating unit configured to update the first mapping relationship set according to the fifth information
- the fourth receiving unit is configured to receive all or part of the machine learning model associated with the second mapping relationship set sent by the source access network device if all or part of the machine learning model associated with the second mapping relationship set is not configured in the access network device. or part of a machine learning model;
- the second updating unit is configured to determine, according to the received all or part of the machine learning models associated with the second mapping relationship set and a plurality of machine learning models associated with the first mapping relationship set before updating, the A plurality of machine learning models associated with a mapping relationship set;
- the lookup unit is specifically used to:
- a machine learning model suitable for the current communication environment is searched to form a first machine learning model set.
- the device further includes:
- the fourth determining unit is configured to determine that the communication environment has changed if a message indicating RRC connection establishment or a message indicating RRC connection re-establishment sent by the terminal is received.
- the present application provides an apparatus for updating a model of a communication system, which is applied to core network equipment.
- the core network equipment is preconfigured with a third mapping relationship set and multiple machine learning models associated with the third mapping relationship set.
- the third mapping relationship set includes the mapping relationship between multiple communication environment information and multiple machine learning models, and the device includes:
- a search unit configured to determine that the communication environment has changed, and according to the third set of mapping relationships, search for a machine learning model suitable for the current communication environment among multiple machine learning models and form a second set of machine learning models;
- a sending unit configured to send sixth information including the second machine learning model set to the access network device, where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set, And/or, the access network device is instructed to send seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the device further includes:
- an acquisition unit configured to acquire the communication environment information on the access network device side and the communication environment information on the terminal side;
- the lookup unit is specifically used to:
- a machine learning model is searched among the multiple machine learning models to form a second machine learning model set.
- the device further includes:
- the first receiving unit is configured to receive eighth information sent by the access network device, where the eighth information includes communication environment information on the access network device side and communication environment information on the terminal side.
- the device further includes:
- the determining unit is used for determining according to each environmental parameter in the communication environment information at the last moment on the access network device side, each environmental parameter in the communication environment information at the current moment on the access network side, and a plurality of environmental parameters and presets
- the mapping relationship of the thresholds determines whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and other parameters.
- the mapping relationship between each environmental parameter and the preset threshold value determines whether the communication environment changes.
- the device further includes:
- a second receiving unit configured to receive ninth information sent by the access network device
- the ninth information is used to indicate that the communication environment changes, the terminal is switched between cells, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is reestablished, and the ninth information includes the following One or more pieces of information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device;
- the lookup unit is specifically used to:
- the ninth information it is determined that the communication environment has changed
- a machine learning model is searched to form a second machine learning model set.
- the device further includes:
- a third receiving unit configured to receive a first mapping relationship set sent by multiple access network devices and multiple machine learning models associated with the first mapping relationship set;
- a model processing unit configured to establish or update a third mapping relationship set and a third mapping relationship with the third mapping relationship set according to the first mapping relationship set sent by multiple access network devices and multiple machine learning models associated with the first mapping relationship set Sets associated multiple machine learning models.
- the present application provides a model updating device for a communication system, which is applied to a terminal, and the device includes:
- a receiving unit configured to receive first information or seventh information returned by the access network device, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set, and the seventh information is used to instruct the terminal
- the terminal deploys the machine learning model in the second machine learning model set
- the deployment unit is configured to deploy the machine learning model in the first machine learning model set according to the first information, or deploy the machine learning model in the second machine learning model set according to the seventh information.
- the device further includes a first sending unit for:
- the device further includes:
- the second sending unit is configured to send fourth information to the access network device according to whether the communication environment on the terminal side has changed, wherein if the fourth information indicates that the communication environment on the terminal side has occurred changes, the fourth information includes the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal.
- a fourth mapping relationship set is configured on one side of the terminal, and the fourth mapping relationship set includes mapping relationships between multiple communication environment information associated with the terminal and multiple machine learning models, and the second sending unit is specifically used for:
- the present application provides a processor-readable storage medium, where the processor-readable storage medium stores a computer program, and the computer program is used to cause a processor to execute the method described in the first aspect, the second aspect or the third aspect .
- the present application provides a computer program product comprising instructions that, when executed on a computer, cause the computer to perform the method as described in the first, second or third aspects above.
- the present application provides a communication system, including the terminal described in any of the above, the access network device described in any of the above, and the core network device described in any of the above.
- the model updating method, device and storage medium in a communication system provided by the present application, when the access network device determines that the communication environment has changed, according to the first mapping relationship set, the multiple Among the machine learning models, a machine learning model suitable for the current communication environment is searched and a first machine learning model set is formed.
- the access network device deploys the machine learning model in the first machine learning model set and/or the access network device instructs the terminal to deploy the machine learning model in the first machine learning model set. Therefore, it is possible to flexibly and quickly update the deployed machine learning model to a machine learning model adapted to the current communication environment on the access network device side and/or the terminal side according to changes in the communication environment, so as to improve the deployment efficiency of the communication system.
- the update efficiency and effect of the machine learning model thereby improving the performance of the communication system.
- FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application
- FIG. 2 is a flowchart of a model updating method in a communication system provided by an embodiment of the present application
- FIG. 3 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 4 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 5 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 6 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 7 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 8 is a flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 9 is a schematic diagram of a model updating apparatus in a communication system provided by an embodiment of the present application.
- FIG. 10 is a schematic diagram of a model updating apparatus in a communication system according to an embodiment of the application.
- FIG. 11 is a schematic diagram of a model updating apparatus in a communication system according to an embodiment of the application.
- FIG. 12 is a schematic diagram of a model updating apparatus in a communication system according to an embodiment of the application.
- FIG. 13 is a schematic diagram of a model updating apparatus in a communication system according to an embodiment of the application.
- FIG. 14 is a schematic diagram of a model updating apparatus in a communication system according to another embodiment of the present application.
- the term "and/or” describes the relationship between related objects, and means that there can be three relationships, for example, A and/or B, which can mean that A exists alone, A and B exist simultaneously, and B exists alone. a situation.
- the character "/” generally indicates that the associated objects are an "or” relationship.
- the term “plurality” refers to two or more than two, and other quantifiers are similar.
- applicable systems may be global system of mobile communication (GSM) system, code division multiple access (CDMA) system, wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA) general packet Wireless service (general packet radio service, GPRS) system, long term evolution (long term evolution, LTE) system, LTE frequency division duplex (frequency division duplex, FDD) system, LTE time division duplex (time division duplex, TDD) system, Long term evolution advanced (LTE-A) system, universal mobile telecommunication system (UMTS), worldwide interoperability for microwave access (WiMAX) system, 5G New Radio (New Radio, NR) system, etc.
- GSM global system of mobile communication
- CDMA code division multiple access
- WCDMA Wideband Code Division Multiple Access
- general packet Wireless service general packet Radio service
- GPRS general packet Wireless service
- LTE long term evolution
- LTE long term evolution
- FDD frequency division duplex
- TDD time division duplex
- LTE-A Long term evolution advanced
- UMTS universal mobile
- the communication system to which the technical solutions provided in the embodiments of the present application are applicable includes network equipment and terminals, and the network equipment may include access network equipment and core network equipment.
- the access network device may be, for example, a wireless access network device.
- Core network (Core Network, CN) equipment may refer to the access and mobility management function (AMF) entity, or may refer to the session management function (session management function, SMF) entity, the mobility management entity (mobility management function) management entity, MME) or other core network equipment.
- AMF access and mobility management function
- SMF session management function
- MME mobility management function management entity
- Access network (AN) equipment for example, a radio access network (RAN) device, which is a device that connects a terminal to a wireless network, which may be in the long term evolution (LTE)
- the evolved base station evolutional Node B, eNB or eNodeB
- a relay station or access point or a 5G base station (gNB) in the 5G network architecture (next generation system)
- gNB 5G base station
- Home evolved Node B Home evolved Node B, HeNB
- relay node relay node
- home base station femto
- pico base station pico base station
- the radio access network equipment may be a base station (such as a gNB) with a separate architecture of centralized unit (centralized unit, CU) and distributed unit (distributed unit, DU), and CU and During can also be separated geographically. layout.
- a base station such as a gNB
- CU central processing unit
- DU distributed unit
- Terminal Can be a device that provides voice and/or data connectivity to a user, a handheld device with wireless connectivity, or other processing device connected to a wireless modem, etc.
- the name of the terminal may be different.
- the terminal may be called user equipment (User Equipment, UE).
- a wireless terminal can communicate with one or more core networks (Core Network, CN) via a Radio Access Network (RAN), and the wireless terminal can be a mobile terminal, such as a mobile phone (or "cellular" phone) and computers with mobile terminals, which may be portable, pocket-sized, hand-held, computer built-in or vehicle mounted mobile devices, for example, which exchange language and/or data with the wireless access network.
- Core Network Core Network
- RAN Radio Access Network
- a wireless terminal may also be referred to as a system, subscriber unit, subscriber station, mobile station, mobile station, remote station, access point, A remote terminal (remote terminal), an access terminal (access terminal), a user terminal (user terminal), a user agent (user agent), and a user device (user device) are not limited in the embodiments of the present application.
- FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application.
- a communication system includes a network device and a terminal 110 , and the network device may include an access network device 120 and a core network device 130 .
- the terminal 110 is connected with the access network device 120 in a wireless manner
- the access network device 120 is connected with the core network device 130 in a wireless or wired manner.
- the core network device 130 and the access network device 120 may be independent and different physical devices, or the functions of the core network device 130 and the logical functions of the access network device 120 may be integrated on the same physical device, or It is a physical device that integrates part of the functions of the core network device 130 and part of the functions of the access network device 120 .
- the terminal 110 may be fixed-position or movable.
- FIG. 1 is just a schematic diagram, the communication system may also include other network devices, such as wireless repeater devices and wireless backhaul devices, which are not shown in FIG. 1 .
- the embodiments of the present application do not limit the number of core network devices 130, access network devices 120, and terminals 110 in the communication system.
- a machine learning model may be deployed on the core network device 130 and/or the terminal 110 , and the machine learning model may be used, for example, for channel estimation, Channel prediction, feedback of channel state information, etc., to take advantage of machine learning models to improve the accuracy of these communication operations.
- the model performance of a machine learning model depends on the training effect of the machine learning model, and the training effect of the machine learning model depends on the model architecture and training algorithm of the machine learning model itself, and on the training data of the machine learning model.
- the richer the training data the better the training effect of the machine learning model.
- a machine learning model obtained by training data in a certain communication environment is often only applicable to the communication environment. When the communication environment changes, the machine learning model will not be suitable for the changed communication environment.
- one method is to re-collect new training data and re-train the machine learning model after the communication environment changes.
- This method is less efficient and is not suitable for real-time communication systems; another method It adopts the strategy of transfer learning to reduce the amount of data required for model training and reduce the time of model training.
- transfer learning takes the machine learning model suitable for the original communication environment as the initial model, and according to the changes in the communication environment after the change. data to fine-tune this initial model.
- remove the last layer or several network layers of the machine learning model suitable for the original communication environment replace the removed network layer with a new network layer, and train the parameters of the new network layer on less training data.
- Transfer learning reduces the amount of data and time required for model training, but cannot meet the real-time connection requirements of communication systems.
- the embodiments of the present application provide a model updating method, device and storage medium in a communication system.
- the model updating method in the communication system provided by the embodiment of the present application, when the communication environment changes, according to the first mapping relationship set configured on the side of the access network device and multiple machines associated with the first mapping relationship set learning model, find a machine learning model suitable for the current communication environment and form a first machine learning model set, deploy the machine learning model in the first machine learning model set on the access network device side, and/or deploy on the terminal side
- the machine learning model in the first machine learning model set, wherein the first mapping relationship set includes the mapping relationship between multiple communication environment information and multiple machine learning models.
- the machine learning model deployed in the communication system can be flexibly updated to be suitable for current communication according to the pre-established mapping relationship between the communication environment information and the machine learning model and multiple pre-established machine learning models.
- the machine learning model of the environment improves the update efficiency and effect of the machine learning model deployed in the communication system, thereby improving the performance of the communication system.
- FIG. 2 is a schematic flowchart of a model updating method in a communication system according to an embodiment of the present application. As shown in FIG. 2, the method of this embodiment may include:
- the access network device searches for a machine learning model suitable for the current communication environment among multiple machine learning models according to the first mapping relationship set, and forms a first machine learning model set.
- the access network device is configured with a first mapping relationship set and multiple machine learning models associated with the first mapping relationship set, and the first mapping relationship set includes multiple pieces of communication environment information and the multiple machine learning models.
- the number of the first mapping relationship sets on the access network device may be one or more. If the number of the first mapping relationship sets is multiple, the access network device is configured with a set of mapping relationship sets that are respectively related to each first mapping relationship set. Connect multiple machine learning models.
- different communication environment information corresponds to different machine learning models
- one communication environment information may correspond to one or more machine learning models.
- One or more machine learning models corresponding to a piece of communication environment information are pre-trained based on the communication data in the communication environment described by the communication environment information, are suitable for the communication environment described by the communication environment information, and can be used in the communication environment described by the communication environment information.
- Communication environment information plays different roles in the communication environment described by the communication environment.
- the communication environment information A corresponds to the machine learning model a, the machine learning model b, and the machine learning model c respectively, wherein the machine learning model a is used in the communication environment described by the communication environment information A.
- the machine learning model b is used for channel prediction in the communication environment described by the communication environment information A
- the machine learning model c is used for the feedback of channel state information in the communication environment described by the communication environment information A.
- the access network device may search the first mapping relationship set for the communication environment information that matches the current communication environment, and obtain the communication environment information corresponding to the communication environment information that matches the current communication environment.
- Identification information for one or more machine learning models According to the identification information of the one or more machine learning models, the machine learning models are obtained from the multiple machine learning models corresponding to the first mapping relationship set, and the obtained one or more machine learning models are suitable for the current communication environment.
- a first set of machine learning models is formed by one or more machine learning models adapted to the current communication environment.
- the change of the communication environment includes the change of the communication environment on the side of the access network device and/or the change of the communication environment on the side of the terminal that communicates with the access network device.
- the identification information of different machine learning models is different, and the identification information of the machine learning model is, for example, a model serial number, a model label, etc. corresponding to the machine learning model.
- the machine learning model deployed on the access network device and/or the machine learning model deployed on the terminal is no longer suitable for the current communication environment. Therefore, after obtaining the first set of machine learning models, you can execute the S202 and/or S203, wherein the execution order of S202 and S203 is not limited here, S202 can be executed before S203, or can be executed after S203, and S202 and S203 can also be executed simultaneously.
- the access network device deploys the machine learning model in the first machine learning model set.
- the machine learning model in the first machine learning model set is deployed on the access network device, so that the machine learning model suitable for the current communication environment is deployed on the access network device in time, and there is no need to redo the machine learning model.
- the training of the learning model effectively improves the update efficiency and effect of the machine learning model deployed on the access network equipment, improves the model performance of the machine learning model in the communication environment, and further improves the performance of the communication system.
- deploying the machine learning model in the first machine learning model set on the access network device means that the access network device initializes the machine learning model and runs it according to the model parameters of the machine learning model in the first machine learning model .
- the access network device sends first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set.
- the terminal deploys the machine learning model in the first machine learning model set according to the first information.
- first information may be sent to the terminal, where the first information includes model parameters of the machine learning models in the first machine learning model set, and is used to instruct the terminal to respond to the first machine learning model.
- the machine learning models in the learned models collection are deployed.
- the terminal deploys the machine learning model in the first machine learning model set, so that the machine learning model suitable for the current communication environment is deployed on the terminal in time without re-training the machine learning model, which is effective This greatly improves the update efficiency and effect of the machine learning model deployed on the terminal, improves the model performance of the machine learning model in the communication system, and further improves the performance of the communication system.
- deploying the machine learning model in the first machine learning model set on the terminal means that the terminal initializes and runs the machine learning model according to the model parameters of the machine learning model in the first machine learning model.
- the machine learning model suitable for the current communication environment and the machine learning model suitable for the current communication environment is deployed on the access network device and/or terminal, so that the access network device and/or terminal can be timely based on changes in the communication environment.
- the deployed machine learning model is updated to a machine learning model suitable for the changed communication environment, which improves the update efficiency and effect of the machine learning model deployed in the communication system, and improves the applicability and play of the machine learning model in the communication system. model performance, which in turn improves the performance of the communication system.
- the access network device may acquire the communication environment information on the access network device side and the communication environment information on the terminal side.
- the obtained access network device can be obtained according to the first mapping relationship set, The communication environment information on one side and the obtained communication environment information on the terminal side are searched for a machine learning model among multiple machine learning models, and a machine learning model suitable for the current communication environment is obtained.
- each communication environment information includes communication environment information on the access network device side and communication environment information on the terminal side.
- the communication environment information on the access network device side includes, for example, environment parameters such as antenna model parameters of the access network device and transmit power of the access network device.
- the communication environment information on the terminal side includes, for example, environment parameters such as antenna model parameters of the terminal, transmit power of the terminal, moving speed of the terminal, and location of the terminal. Therefore, the communication environment on the access network device side and the communication environment on the terminal side are comprehensively considered to improve the completeness of the communication environment in the first mapping relationship set, thereby improving the applicability and model effect of the machine learning model in the communication environment.
- the antenna model parameters of the base station include, for example, multiple antennas of the base station and Massive-MIMO of the base station.
- the antenna model parameters of the terminal include, for example, a single antenna of the terminal and multiple antennas of the terminal, the movement speed of the terminal includes, for example, low-speed movement, high-speed movement, and stationary, and the location of the terminal includes, for example, indoors and outdoors.
- each communication environment information may further include environment parameters shared by the access network device and the terminal, for example, the operating frequency range between the access network device and the terminal. , large-scale path loss parameters, communication scenarios, distance between access network equipment and terminals.
- Each communication environment information may also include fading parameters of the wireless communication channel, such as small-scale fading parameters, so as to further enrich the communication environment information and improve the completeness of the communication environment information.
- the operating frequency range includes, for example, Frequency Range 1 (Frequency Range 1, FR1), Frequency Range 2 (Frequency Range 2, FR2), etc.
- the frequency range of FR1 is, for example, 450 megahertz (MHz) to 6 gigahertz (GHz)
- the frequency range of FR2 is, for example, 24.25 GHz to 5.6 GHz.
- Large-scale road loss parameters include, for example, line-of-sight road loss and non-line-of-sight road loss lights.
- Small-scale fading parameters include, for example, flat fading, frequency selective fading, etc.
- Flat fading includes flat slow fading and flat fast fading
- frequency selective fading includes frequency selective slow fading and frequency selective fast fading.
- Communication scenarios include, for example, urban macro cells, urban micro cells, indoor hotspots, large-scale continuous coverage (mainly characterized by supporting high-speed movement through continuous wide-area coverage and covering vast areas such as suburban and rural areas), etc.
- the following table is a partial mapping relationship between multiple pieces of communication environment information and multiple machine learning models in the first mapping relationship set. As shown in the table below:
- the acquired communication environment information on the device side of the access network, and the acquired communication environment information on the terminal side in the process of searching for the machine learning model among the multiple machine learning models, the The obtained communication environment information on the access network device side and the obtained communication environment information on the terminal side are compared with a plurality of communication environment information in the first mapping relationship set to determine the obtained communication environment on the access network device side
- the information and the obtained communication environment information on the terminal side are the closest communication environment information in the first mapping relationship set, and the identification information of the machine learning model corresponding to the closest communication environment information is obtained.
- search the machine learning model corresponding to the acquired identification information of the machine learning model among the plurality of machine learning models associated with the first mapping relationship set that is, obtain a machine learning model suitable for the current communication environment.
- FIG. 3 is a schematic flowchart of a model updating method in a communication system provided by another embodiment of the present application. As shown in FIG. 3, the method of this embodiment may include:
- the terminal sends second information to an access network device, where the second information includes communication environment information on the terminal side.
- the terminal may periodically send the second information to the access network device, so as to periodically send the communication environment information on the terminal side to the access network device, or the terminal may also detect that a significant occurrence in its own communication environment occurs.
- the second information is sent to the access network device, for example, the second information is sent to the access network device when it is detected that the moving speed of the self has changed from high-speed movement to low-speed movement.
- the terminal may send third information to the access network device, and the access network device responds to the third information.
- the third information is used to instruct the access network device to obtain the second information, and the third information is, for example, a reference signal (Reference Signal, RS).
- the access network device acquires the communication environment information on the side of the access network device.
- the access network device searches for machine learning among multiple machine learning models according to the first mapping relationship set, the communication environment information on the access network device side, and the communication environment information on the terminal side model and constitute the first set of machine learning models.
- the access network device may determine whether the communication environment has changed according to the obtained communication environment information on the terminal side and/or the obtained communication environment information on the access network device side, so as to improve the judgment of whether the communication environment has changed. accuracy.
- the communication environment information on the terminal side and/or the obtained communication environment information on the access network device side it can be determined according to the communication environment information on the access network device side at the last moment.
- the environment information and the communication environment information of the access network equipment at the current moment to determine whether the environment information has changed; and/or the communication environment information of the terminal side at the last moment and the communication environment information of the terminal side at the current moment can be used to determine the communication environment. changes. Therefore, when the communication environment on the access network device side changes and/or the communication environment on the terminal side changes, it is determined that the communication environment changes.
- the access network device may The environment parameters in the communication environment information at the last moment on one side are compared with the environment parameters in the communication environment information of the access network device at the current moment. If there are different environmental parameters in the communication environment information of the device at the current moment, it is determined that the environment has changed.
- the environment information of the access network device at the last moment and the communication environment information of the access network device at the current moment it can be determined according to the communication environment information of the access network device.
- the value of the environmental parameter in the communication environment information on the access network device side at the previous moment is compared with the value of the environmental parameter in the communication environment information on the access network device side at the current moment. By comparison, a difference is obtained, and it is determined whether the difference is greater than a preset threshold corresponding to the environmental parameter. If there is a preset number of environmental parameters, the difference between the value in the communication environment information on the access network device side at the previous moment and the value in the communication environment information on the access network device side at the current moment is greater than the corresponding preset value. If the threshold is set, it is determined that the communication environment has changed. For example, if the difference between the transmission power of the access network device at the previous moment and the generation power of the access network device at the current moment is greater than a preset threshold corresponding to the transmission power, it is determined that the communication environment has changed.
- each environment in the communication environment information of the terminal side at the last moment can be determined.
- the parameters, each environmental parameter in the communication environment information at the current moment on the terminal side, and the mapping relationship between multiple environmental parameters and preset thresholds determine whether the communication environment has changed. For details, refer to the above and the description of the communication environment information of the access network device to determine whether the communication environment has changed, and details are not repeated here.
- the difference between the movement speed of the terminal at the last moment and the movement speed of the terminal at the current moment is greater than the preset threshold corresponding to the movement speed, and the distance between the terminal and the access network device at the previous moment is the same as the distance between the terminal and the connection at the current moment. If the difference between the distances between the network access devices is greater than the preset threshold value corresponding to the distance between the terminal and the access network device, it is determined that the communication environment has changed.
- the access network device deploys the machine learning model in the first machine learning model set.
- the access network device sends first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set.
- the terminal deploys the machine learning model in the first machine learning model set according to the first information.
- the access network device receives the second information sent by the terminal, and obtains the communication environment information on the terminal side, according to the first mapping relationship set, the communication environment information on the terminal side, and the communication environment information on the access network device side.
- Environment information find the machine learning model suitable for the current communication environment, and deploy the machine learning model suitable for the current communication environment on the access network equipment and/or terminals, so that the access network and the current communication environment can be changed in time.
- the machine learning model deployed on the terminal side is updated to a machine learning model suitable for the communication environment, which improves the update efficiency and effect of the deployment of the machine learning model deployed in the communication system, and improves the applicability and effectiveness of the machine learning model in the communication system.
- the access network device side in addition to determining whether the communication environment has changed according to the communication environment information on the access network device side and/or the communication environment information on the terminal side, when the terminal has During cell handover, it can be determined that the communication environment has changed, or if the access network device receives a message sent by the terminal indicating the establishment of a Radio Resource Control (RRC) connection (for example, when the terminal is just powered on, it is sent to the access network device.
- RRC Radio Resource Control
- the access network device can determine that the communication environment on the terminal side has changed, that is, determine that the communication environment has changed, so as to more comprehensively consider different situations of the communication environment change , to improve the accuracy of judging changes in the communication environment, thereby improving the deployment effect of the machine learning model.
- FIG. 4 is a schematic flowchart of a model updating method in a communication system provided by another embodiment of the present application. As shown in FIG. 4 , the method of this embodiment may include:
- this embodiment determines whether the communication environment on the terminal side has changed on the terminal, so as to improve the timeliness and accuracy of judging the change in the communication environment on the terminal side. .
- the terminal may acquire the communication environment information of the terminal side in real time or periodically, and determine the communication environment of the terminal side according to the communication environment information of the terminal side at the previous moment and the communication environment information of the terminal side at the current moment. changes.
- each environmental parameter in the communication environment information at the previous moment on the terminal side is compared with each environmental parameter in the communication environment at the previous moment on the terminal side. If different environmental parameters in the communication environment information at a moment exceed a preset number, it is determined that the communication environment on the terminal side has changed. Or, if there is a preset number of environmental parameters, the difference between the value in the communication environment information at the previous moment on the terminal side and the value in the communication environment information at the current moment on the terminal side exceeds the corresponding preset threshold, Then it is determined that the communication environment on the terminal side has changed.
- a preset number the difference between the value in the communication environment information at the previous moment on the terminal side and the value in the communication environment information at the current moment on the terminal side exceeds the corresponding preset threshold.
- the terminal sends fourth information to the access network device, where the fourth information is used to indicate whether the communication environment on the terminal side has changed, and the fourth information indicates whether the communication environment on the terminal side has changed.
- the communication environment has changed includes the fourth information indicating that the communication environment on the terminal side has changed or the fourth information indicating that the communication environment on the terminal side has not changed.
- the terminal after determining whether the communication environment on the terminal side has changed, the terminal can send the fourth information to the access network device, so that the access network device can directly know whether the communication environment on the terminal side has occurred according to the fourth information.
- the fourth information indicates that the communication environment on the terminal side changes
- the access network device does not need to judge whether the communication environment changes according to the communication environment information of the terminal and the communication environment information of the access network device.
- the fourth information indicates that the communication environment on the terminal side has changed, and the fourth information includes the communication environment information on the terminal side and/or the machine learning requested by the terminal.
- the identification information of the model so that the access network device returns the machine learning model suitable for the current communication environment to the terminal according to the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal.
- a fourth mapping relationship set is configured on the terminal side, and the fourth mapping relationship set includes mapping relationships between multiple communication environment information associated with the terminal and multiple machine learning models.
- the relationship set includes the correspondence between multiple pieces of communication environment information of the terminal and identification information of multiple machine learning models. Therefore, after the terminal obtains its own communication environment information and determines according to its own communication environment information that the communication environment on the terminal side has changed, the terminal can search for a communication environment related to the terminal in the fourth mapping relationship set according to the current communication environment information on the terminal side. Identification information of the machine learning model corresponding to the current communication environment information on the side. The found identification information of the machine learning model is the identification information of the machine learning model requested by the terminal.
- different communication environment information corresponds to different machine learning models
- one communication environment information may correspond to one or more machine learning models
- one or more machine learning models corresponding to the communication environment information are applicable. They play different roles in the communication environment of the communication environment information, for example, one machine learning model is used for channel estimation under the communication environment of the communication environment information, and the other is used for channel prediction.
- the access network device determines whether the communication environment changes according to the fourth information.
- the access network device can directly learn whether the communication environment on the terminal side has changed according to the fourth information. If the fourth information indicates that the communication environment on the terminal side has changed, the access network device determines that the communication environment has changed. If the fourth information indicates that the communication environment on the terminal side has not changed, the access network device may determine that the communication environment has not changed, or may continue to determine the communication environment based on whether the communication environment on the access network device side has changed. Whether or not there is a change, reference may be made to the foregoing embodiments for details, which will not be repeated.
- the access network device If the communication environment changes, the access network device according to the first mapping relationship set and the communication environment information on the access network device side, and according to the communication environment information on the terminal side and/or the machine learning model requested by the terminal.
- the identification information among the multiple machine learning models, searches for a machine learning model and forms a first machine learning model set.
- the access network device acquires the communication environment information on the side of the access network device. If the fourth information includes the communication environment information on the terminal side, the access network device may, according to the first mapping relationship set, the communication environment information on the access network device side and the communication environment information on the terminal side, communicate with the first Among the multiple machine learning models associated with the mapping relationship set, a machine learning model suitable for the current communication environment is searched, and one or more machine learning models suitable for the current communication environment constitute a first machine learning model set.
- the access network device may directly associate with the first mapping relationship set according to the identification information of the machine learning model requested by the terminal. Find one or more machine learning models corresponding to the identification information of the machine learning model requested by the terminal from among the plurality of cluster learning models. A first set of machine learning models is formed by the one or more machine learning models found.
- the access network device deploys the machine learning model in the first machine learning model set.
- the access network device sends first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set.
- the terminal deploys the machine learning model in the first machine learning model set according to the first information.
- the terminal determines on the terminal whether the communication environment on the side of the terminal changes, and the terminal sends fourth information to the access network device, and the access network device sets the fourth information according to the first mapping relationship and the access network device.
- One side of the communication environment information find the machine learning model suitable for the current communication environment, and deploy the machine learning model suitable for the current communication environment on the access network equipment and/or terminal, so as to improve the accuracy of the judgment of changes in the communication environment It can timely and accurately update the machine learning model deployed on the access network equipment and/or terminal to the machine learning model suitable for the communication environment according to the change of the communication environment, which improves the update efficiency of the machine learning model deployed in the communication system. And the effect improves the applicability of the machine learning model in the communication system and the performance of the model, thereby improving the performance of the communication system.
- FIG. 5 is a schematic flowchart of a model updating method in a communication system provided by another embodiment of the present application.
- FIG. 5 includes a terminal, an access network device, and a source access network device.
- the access network device can be understood as the target access network device of the terminal.
- the method of this embodiment may include:
- the terminal performs cell handover, and switches to the access network device.
- the access network device receives a handover request sent by the source access network device of the terminal.
- the terminal switches from the source access network device to the access network device, and the terminal performs cell handover.
- the access network device determines that the communication environment has changed.
- the access network device may determine that the communication environment changes when the terminal switches between cells.
- the source access network device sends fifth information to the access network device.
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the identification information of the machine learning model used when the terminal communicates with the source access network device.
- the source access network device is the access network device to which the terminal is connected before the cell handover.
- the second set of mapping relationships is a subset of mapping relationships associated with the terminal in the mapping relationships between multiple pieces of communication environment information configured on the source access network device and multiple machine learning models.
- mapping relationship between multiple communication environment information and multiple machine learning models on the source access network device obtain the mapping relationship between the communication environment information of the terminal at different times and multiple machine learning models, and combine the obtained mapping relationships Obtain a second set of mapping relationships; or, obtain a plurality of mapping relationships between communication environment information and a machine learning model collected when the terminal communicates with the source access network device, and combine the obtained mapping relationships to obtain a second set of mapping relationships; or, Obtain a mapping relationship between the communication environment information when the terminal communicates with the source access network device and the machine learning model used when the terminal communicates with the access network device, and combine the obtained mapping relationships to obtain a second mapping relationship set.
- the terminal when the terminal performs cell handover, the terminal switches from the source access network device to the current access network device, considering that the source access network device of the terminal stores the information about the communication between the terminal and the source access network device.
- the source access network device when the adopted machine learning model, communication environment information and other information, when the cell of the terminal changes, the source access network device sends fifth information to the current access network device.
- the access network device updates the first mapping relationship set.
- the access network device when the access network device receives the fifth information sent by the source access network device, according to the second mapping relationship set in the fifth information, the communication environment information when the terminal communicates with the source access network device, and / or the identification information of the machine learning model used when the terminal communicates with the source access network device, update the first mapping relationship set, so as to enrich the first mapping relationship set in a targeted manner, and further improve the search based on the first mapping relationship set Accuracy of machine learning models adapted to current environmental information.
- a possible implementation manner is: combining the communication environment information when the terminal communicates with the source access network device and multiple communication environment information in the first mapping relationship set to increase the first mapping relationship set.
- a mapping relationship sets the communication environment information associated with the terminal.
- Another possible implementation manner is as follows: the identification information of the machine learning model used when the terminal communicates with the source access network device and the identification information of the multiple machine learning models in the first mapping relationship set The identification information is combined to increase the identification information of the machine learning model associated with the terminal in the first mapping relationship set.
- Another possible implementation manner is: merging the second mapping relationship set with the first mapping relationship set to increase the communication environment information associated with the terminal in the first mapping relationship set Mapping relationship with machine learning model.
- the access network device may send to the source access network device a message requesting all or part of the machine learning models associated with the second set of mapping relationships, where the message may include all or part of the machine learning models associated with the second set of mapping relationships and receive all or part of the machine learning model associated with the second mapping relationship set returned by the source access network device.
- the identification information of the machine learning model is unified.
- the identification information of the machine learning model is not only unique on the current access network device, but also on multiple access network devices. Therefore, the access network device can collect the identification information of the machine learning model according to the second mapping relationship, and find out whether there is a corresponding identification information of the machine learning model in the second mapping relationship among the multiple machine learning models configured by itself. machine learning model. If the machine learning model corresponding to the identification information of one or more machine learning models in the second mapping relationship set cannot be found, it is determined that all or part of the machine learning models associated with the second mapping relationship set are not configured in the access network device .
- the access network device determines, according to the received all or part of the machine learning models associated with the second mapping relationship set and multiple machine learning models associated with the first mapping relationship set before the update, the first mapping relationship set after the update. Multiple machine learning models associated with a set of mapping relationships.
- the access network device after the access network device receives all or part of the machine learning models associated with the second mapping relationship set and sent by the source access network device, it can transfer all or part of the machine learning models associated with the second mapping relationship set
- the learning model is combined with multiple machine learning models associated with the first mapping relationship set before the update to obtain multiple machine learning models associated with the updated first mapping relationship set, thereby enriching the first mapping relationship in a targeted manner.
- the machine learning model associated with the relation set is
- the access network device searches for a machine learning model suitable for the current communication environment among the multiple machine learning models associated with the updated first mapping relationship set, and forms a first A collection of machine learning models.
- the identification information of the machine learning model adapted to the current communication environment information can be searched in the updated first mapping relationship set, and then Corresponding machine learning models are obtained from a plurality of machine learning models associated with the updated first mapping relationship set according to the identification information, and a first machine learning model set is constituted by the obtained one or more machine learning models.
- the fourth information sent by the terminal may also be received, and according to the identification information of the machine learning model requested by the terminal in the fourth information, the multiple machine learning models associated in the first mapping relationship set are obtained from one or more machine learning models obtained.
- the machine learning models constitute a first set of machine learning models.
- the access network device deploys the machine learning model in the first machine learning model set.
- the access network device sends first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set.
- the terminal deploys the machine learning model in the first machine learning model set according to the first information.
- the access network device updates the first mapping relationship set and multiple machine learning models associated with the first mapping set according to the fifth information sent by the source access network device.
- search for a machine learning model suitable for the current communication environment and perform on the access network device and/or terminal.
- the machine learning model of the current communication environment is deployed. Therefore, the machine learning model deployed on the access network equipment and/or terminal can be updated to a machine learning model suitable for the communication environment in a timely and accurate manner according to the change of the communication environment, which improves the update efficiency and efficiency of the machine learning model deployed in the communication system.
- the applicability and model performance of the machine learning model in the communication system are improved, thereby improving the performance of the communication system.
- a machine learning model suitable for the current communication environment can also be searched on the core network device to form a second machine learning model set, and the core network device sends the second machine learning model set to the access network device,
- the more powerful performance of the core network equipment is used to improve the search accuracy of the machine learning model, thereby improving the updating effect of the machine learning model.
- the second machine learning model set includes one or more machines that are suitable for the current communication environment and are found in multiple machine learning models associated with the third mapping relationship set according to the third mapping relationship set on the core network device.
- the learning model see the embodiment shown in FIG. 6 for details.
- FIG. 6 is a schematic flowchart of a model updating method in a communication system provided by another embodiment of the present application. As shown in FIG. 6 , the method of this embodiment may include:
- the core network device searches for a machine learning model suitable for the current communication environment among multiple machine learning models according to the third mapping relationship set, and forms a second machine learning model set.
- a third mapping relationship set and multiple machine learning models associated with the third mapping relationship set are configured on the core network device, and the third mapping relationship set includes multiple communication environment information and the multiple machine learning models. mapping relationship.
- the number of third mapping relationship sets on the access network device may be one or more. If the number of third mapping relationship sets is multiple, the access network device is configured with corresponding third mapping relationship sets respectively. Connect multiple machine learning models.
- different communication environment information corresponds to different machine learning models
- one communication environment information may correspond to one or more machine learning models.
- One or more machine learning models corresponding to a piece of communication environment information are pre-trained based on the communication data in the communication environment described by the communication environment information, are suitable for the communication environment described by the communication environment information, and can be used in the communication environment described by the communication environment information.
- Communication environment information plays different roles in the communication environment described by the communication environment. For example, a machine learning model corresponding to a communication environment information is used for channel estimation under the communication environment described by the communication environment information, and another machine learning model corresponding to the communication environment information is used for the channel estimation described by the communication environment information. Channel prediction in communication environments.
- the core network device may search for the identification information of one or more machine learning models corresponding to the current communication environment in the third mapping relationship set. According to the found identification information of the one or more machine learning models, the core network device obtains one or more machine learning models among the multiple machine learning models associated with the third mapping relationship set, and the obtained one or more machine learning models A machine learning model is a machine learning model suitable for the current communication environment. Therefore, a second set of machine learning models is formed by the acquired one or more machine learning models.
- the core network device may receive the first mapping relationship sent by multiple access network devices a set and a plurality of machine learning models associated with the first set of mapping relationships.
- the third mapping relationship is established or updated according to the first mapping relationship set sent by the multiple access network devices and the multiple machine learning models associated with the first mapping relationship set.
- the core network device may summarize the first mapping relationship set on multiple access network devices to obtain a third mapping relationship set, and analyze multiple access network devices associated with the first mapping relationship set on multiple access network devices.
- the machine learning models are aggregated to obtain a plurality of machine learning models associated with the third set of mapping relationships. For example, the same machine learning model is deduplicated, and identification information of different machine learning models is determined, so as to unify the identification information of the machine learning models on multiple access network devices.
- the core network device may receive the updated first mapping relationship set sent by the access network device and the updated machine learning model, update the third mapping relationship set and multiple machine learning models associated with the third mapping relationship set in time to improve the accuracy of the third mapping relationship set and the machine learning model on the core network device side , effectiveness.
- the access network device retrains one or more machine learning models according to the recently collected communication data every preset period, and adds the model identifier of the retrained machine learning model and the corresponding communication environment information to the first mapping relationship set and send the updated first mapping relationship set and machine learning model to the core network device.
- the core network device may periodically collect the first mapping relationship set on multiple access network devices and the multiple machine learning models associated with the first mapping relationship set, and actively update the third mapping relationship set in time. and a plurality of machine learning models associated with the third set of mapping relationships.
- the core network device may associate the third mapping relationship set and multiple machine learning models associated with the third mapping relationship set.
- the machine learning model is sent to multiple access network devices, or a part of the mapping relationship in the third mapping relationship set and multiple machine learning models associated with the partial mapping relationship are sent to multiple access network devices, and the access network device will receive
- the received mapping relationship is updated to the first mapping relationship set, and the multiple machine learning models associated with the received mapping relationship are updated to multiple machine learning models associated with the first mapping relationship set, thereby implementing different access networks.
- the core network device may periodically collect communication data and train a machine learning model according to the collected communication data, and according to the trained machine learning model and the communication environment information in the communication environment corresponding to the machine learning model , constructing and/or updating a third mapping relationship set and multiple machine learning models associated with the third mapping relationship set. Therefore, use the more powerful core network equipment to train the machine learning model and establish the mapping relationship between the machine learning model and the communication environment information, so as to improve the generation efficiency and effect of the machine learning model and the mapping relationship.
- the core network device sends sixth information including the second machine learning model set to the access network device, where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set, and/or, instructing the access network device to send seventh information to the terminal.
- the core network device may send sixth information including the second machine learning model set to the access network device. If the sixth information indicates that the access network device deploys the machine learning model in the second machine learning model set, perform S603. If the sixth information indicates that the access network device sends the seventh information to the terminal, perform S604. If the sixth information instructs the access network device to deploy the machine learning model in the second machine learning model set and the access network device sends the seventh information to the terminal, perform S603 and S604.
- the access network device deploys the machine learning model in the second machine learning model set.
- the access network device sends seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the terminal deploys the machine learning model in the second machine learning model set according to the seventh information.
- the core network device when determining that the communication environment changes, searches for a machine learning model suitable for the current communication environment, forms a second machine learning model set, and sends the second machine learning model set to the access network device to
- the machine learning models in the second machine learning model set are deployed in the communication system, so that more machine learning models and more mapping relationships can be configured on the core network device, so as to improve the search accuracy of the machine learning models, and further Improve the update effect of machine learning models in communication systems.
- the core network device may acquire the communication environment information on the access network device side and the communication environment information on the terminal side.
- the machine learning model can be searched in the multiple machine learning models associated with the third mapping relationship set. The model is obtained to obtain a machine learning model suitable for the current communication environment, thereby improving the accuracy of the machine learning model search.
- the acquired communication environment information on the device side of the access network, and the acquired communication environment information on the terminal side search for a machine among multiple machine learning models associated with the third mapping relationship set
- search for a machine among multiple machine learning models associated with the third mapping relationship set For the implementation process of the learning model, refer to the first mapping relationship set, the acquired communication environment information on the device side of the access network, and the acquired communication environment information on the terminal side in the foregoing embodiment, in the first mapping relationship set related to the first mapping relationship set. Find the description of the machine learning model in the multiple machine learning models connected, and will not repeat them.
- FIG. 7 is a schematic flowchart of a model updating method in a communication system provided by another embodiment of the present application. As shown in FIG. 7 , the method of this embodiment may include:
- the access network device may periodically send the eighth information to the core network device, so as to periodically send the communication environment information on the access network device side and the communication environment information on the terminal side to the core network device.
- the access network device may send the eighth information to the core network device after receiving the communication environment information acquisition instruction of the core network device.
- the core network device searches for a machine learning model among multiple machine learning models according to the third mapping relationship set, the communication environment information on the access network device side, and the communication environment information on the terminal side A second set of machine learning models is formed.
- the implementation process of S702 may refer to the foregoing embodiments, which will not be repeated.
- the core network device may determine whether the communication environment has changed according to the obtained communication environment information on the terminal side and/or the obtained communication environment information on the access network device side, so as to improve the accuracy of the judgment of whether the communication environment has changed. sex.
- each environmental parameter in the communication environment information at the last moment on the access network device side determines whether the communication environment has changed, according to each environmental parameter in the communication environment information at the last moment on the access network device side, each environmental parameter in the communication environment information at the current moment on the access network side, and The mapping relationship between multiple environmental parameters and preset thresholds is used to determine whether the communication environment has changed; Each environmental parameter of the device, as well as the mapping relationship between multiple environmental parameters and preset thresholds, determine whether the communication environment has changed.
- the specific process for the core network device to determine whether the communication environment has changed may refer to the access network device to determine whether the communication environment is based on the acquired communication environment information on the terminal side and/or the acquired communication environment information on the access network device side. The description of the changes will not be repeated.
- the core network device sends sixth information including the second machine learning model set to the access network device, where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set, and/or, instructing the access network device to send seventh information to the terminal.
- the sixth information instructs the access network device to deploy the machine learning model in the second machine learning model set
- S704 is performed. If the sixth information indicates that the access network device sends the seventh information to the terminal, perform S705. If the sixth information instructs the access network device to deploy the machine learning model in the second machine learning model set and the access network device sends the seventh information to the terminal, perform S704 and S705.
- the access network device deploys the machine learning model in the second machine learning model set.
- the access network device sends seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the terminal deploys the machine learning model in the second machine learning model set according to the seventh information.
- the core network device receives the eighth information sent by the access network device, and obtains the communication environment information on the terminal side and the communication environment information on the access network device side.
- the terminal side The communication environment information and the communication environment information on the access network device side, find the machine learning model suitable for the current communication environment, and deploy the machine learning model suitable for the current communication environment on the access network device and/or terminal , so that the machine learning model deployed on the access network equipment and/or terminal can be updated to the machine learning model suitable for the communication environment in time according to the change of the communication environment, which improves the update efficiency and effect of the machine learning model deployed in the communication system,
- the applicability and model effect of the machine learning model in the communication system are improved, thereby improving the performance of the communication system.
- FIG. 8 is a schematic flowchart of a model updating method in a communication system according to another embodiment of the present application. As shown in FIG. 8 , the method of this embodiment may include:
- the ninth information is used to indicate that the communication environment changes, the terminal is switched between cells, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is reestablished, and the ninth information includes the following One or more pieces of information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device.
- the access network device may send ninth information to the core network device when receiving the fourth information sent by the terminal, and the fourth information indicates that the communication environment on the terminal side has changed, where the ninth information indicates the communication The environment changes.
- the access network device may send ninth information to the core network when it is determined that the communication environment changes based on the communication environment information on the terminal side and/or the communication environment information on the access network device side, and at this time The ninth information indicates that the communication environment has changed.
- the access network device may send the ninth information to the core network when it is determined that the communication environment has changed and the identification information of the machine learning model corresponding to the current communication environment cannot be found in the first mapping relationship set.
- the ninth information indicates that the communication environment has changed.
- the access network device may send ninth information to the core network device when it is determined that a cell handover occurs at the terminal, or when it receives a message indicating RRC connection establishment or an RRC connection reestablishment message sent by the terminal.
- Nine information indicates that the communication environment has changed, or indicates that the terminal has undergone cell handover, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is re-established.
- the access network device may also send the ninth information to the core network device when the RRC connection between itself and the terminal is successfully established or the RRC connection is successfully reestablished.
- the access network device determines that the communication environment has changed according to the ninth information.
- the access network device when the access network device receives the ninth information, it is determined that the communication environment has changed.
- the core network device searches for the machine learning model among the multiple machine learning models according to the third mapping relationship set and the ninth information, and forms a second machine learning model set.
- the core network device may, according to one or more of the communication environment information on the access network device side, the communication environment information on the terminal side, and the identification information of the machine learning model requested by the access network device, in the In the third mapping relationship set, the machine learning model is searched to form a second machine learning model set. For example, according to the communication environment information on the device side of the access network and the communication environment information on the terminal side, look up the model identifier of the machine learning model in the third mapping relationship set, and then find the model identifier of the machine learning model in the third mapping relationship according to the model identifier of the machine learning model. Among the multiple associated machine learning models, the machine learning model is acquired and a second machine learning model set is formed. For another example, according to the identification information of the machine learning model requested by the access network device, among the plurality of machine learning models associated with the third mapping relationship, the machine learning model is acquired to form a second machine learning model set.
- the core network device sends sixth information including the second machine learning model set to the access network device, where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set, and/or, instructing the access network device to send seventh information to the terminal.
- the sixth information indicates that the access network device deploys the machine learning model in the second machine learning model set
- S805 is performed. If the sixth information indicates that the access network device sends the seventh information to the terminal, perform S806. If the sixth information instructs the access network device to deploy the machine learning model in the second machine learning model set and the access network device sends the seventh information to the terminal, then perform S805 and S806.
- the access network device deploys the machine learning model in the second machine learning model set.
- the access network device sends seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the terminal deploys the machine learning model in the second machine learning model set according to the seventh information.
- the core network device receives the ninth information sent by the access network device, searches for a machine learning model suitable for the current communication environment according to the third mapping relationship set and the ninth information, and performs a search on the access network device and/or Or deploy the machine learning model suitable for the current communication environment on the terminal, so that the machine learning model deployed on the access network equipment and/or the terminal can be updated to the machine learning model suitable for the communication environment in time according to the change of the communication environment.
- the update efficiency and effect of the machine learning model deployed in the communication system are improved, the applicability and model effect of the machine learning model in the communication system are improved, and the performance of the communication system is improved.
- an embodiment of the present application provides a model updating apparatus in a communication system, which is applied to an access network device.
- the access network device is configured with a first mapping relationship set associated with the first mapping relationship set.
- the first mapping relationship set includes mapping relationships between multiple communication environment information and multiple machine learning models.
- the model updating apparatus in the communication system provided in this embodiment may be an access network device, and the model updating apparatus in the communication system may include: a transceiver 901 , a processor 902 , and a memory 903 .
- the transceiver 901 is used to receive and transmit data under the control of the processor 902 .
- the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 902 and various circuits of memory represented by memory 903 are linked together.
- the bus architecture may also link together various other circuits, such as peripherals, voltage regulators, and power management circuits, which are well known in the art and, therefore, will not be described further herein.
- the bus interface provides the interface.
- Transceiver 901 may be multiple elements, ie, including a transmitter and a receiver, providing means for communicating with various other devices over transmission media including wireless channels, wired channels, fiber optic cables, and the like.
- the processor 902 is responsible for managing the bus architecture and general processing, and the memory 903 may store data used by the processor 902 in performing operations.
- the processor 902 is responsible for managing the bus architecture and general processing, and the memory 903 may store data used by the processor 902 in performing operations.
- the processor 902 can be a central processing unit (central processing unit, CPU), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable Logic device (Complex Programmable Logic Device, CPLD), the processor can also use a multi-core architecture.
- CPU central processing unit
- ASIC Application Specific Integrated Circuit
- FPGA field programmable gate array
- CPLD Complex Programmable Logic Device
- the processor 902 invokes the computer program stored in the memory 903 to execute any one of the methods related to the access network device provided in the embodiments of the present application according to the obtained executable instructions.
- the processor and memory may also be physically separated.
- the processor 902 is configured to perform the following operations:
- a machine learning model suitable for the current communication environment is found to form a first machine learning model set; the first machine is deployed on the access network device. Learning the machine learning models in the model set, and/or sending first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning models in the first machine learning model set.
- processor 902 is further configured to perform the following operations:
- the first mapping relationship set among the multiple machine learning models, find a machine learning model suitable for the current communication environment and form a first machine learning model set, including:
- a machine learning model is searched among multiple machine learning models to form a first machine learning model set.
- obtain the communication environment information on the terminal side including:
- processor 902 is further configured to perform the following operations:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and multiple environmental parameters and predictions. Set the mapping relationship of thresholds to determine whether the communication environment has changed.
- processor 902 is further configured to perform the following operations:
- the fourth information includes the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal; According to the fourth information, it is determined whether the communication environment has changed.
- the machine learning model is searched and the first machine learning model set is formed.
- processor 902 is further configured to perform the following operations:
- processor 902 is further configured to perform the following operations:
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the terminal and another access network device
- the identification information of the machine learning model used during communication, and the second mapping relationship set is a subset of the mapping relationship associated with the terminal in the mapping relationship between multiple communication environment information configured on the source access network device and multiple machine learning models ;
- All or part of the machine learning models associated with the two sets of mapping relationships according to the received all or part of the machine learning models associated with the second set of mapping relationships and multiple machine learning models associated with the first set of mapping relationships before the update , determining multiple machine learning models associated with the updated first mapping relationship set;
- the first mapping relationship set among the multiple machine learning models, find a machine learning model suitable for the current communication environment and form a first machine learning model set, including:
- a machine learning model suitable for the current communication environment is searched to form a first machine learning model set.
- update the first mapping relationship set including:
- the second mapping relationship set and the first mapping relationship set are combined to obtain an updated first mapping relationship set.
- processor is also used to perform the following operations:
- an embodiment of the present application provides an apparatus for updating a model of a communication system, which is applied to core network equipment.
- the core network equipment is preconfigured with a third mapping relationship set and multiple sets of mapping relationships associated with the third mapping relationship set.
- the third set of mapping relationships includes mapping relationships between multiple pieces of communication environment information and multiple machine learning models.
- the model updating apparatus of the communication system in this embodiment may be a core network device, and the model updating apparatus of the communication system may include a transceiver 1001 , a processor 1002 and a memory 1003 .
- the bus architecture may include any number of interconnected buses and bridges, specifically, one or more processors represented by the processor 1002 and various circuits of the memory represented by the memory 1003 are linked together.
- the bus architecture may also link together various other circuits, such as peripherals, voltage regulators, and power management circuits, which are well known in the art and, therefore, will not be described further herein.
- the bus interface provides the interface.
- Transceiver 1001 may be multiple elements, ie, including a transmitter and a receiver, providing means for communicating with various other devices over transmission media including wireless channels, wired channels, fiber optic cables, and the like.
- the processor 1002 is responsible for managing the bus architecture and general processing, and the memory 1003 may store data used by the processor 1002 in performing operations.
- the processor 1002 may be a CPU, an ASIC, an FPGA or a CPLD, and the processor may also adopt a multi-core architecture.
- the processor 1002 invokes the computer program stored in the memory 1003 to execute any one of the methods related to the core network device provided in the embodiments of the present application according to the obtained executable instructions.
- the processor and memory may also be physically separated.
- processor 1002 is configured to perform the following operations:
- the sixth information of the learning model set where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set, and/or instruct the access network device to send the seventh information to the terminal , and the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- processor 1002 is further configured to perform the following operations:
- mapping relationships among multiple machine learning models, find a machine learning model suitable for the current communication environment and form a second set of machine learning models, including:
- a machine learning model is searched among the multiple machine learning models to form a second machine learning model set.
- Eighth information sent by the access network device is received, where the eighth information includes communication environment information on the access network device side and communication environment information on the terminal side.
- processor 1002 is further configured to perform the following operations:
- each environmental parameter in the communication environment information at the last moment on the access network device side determines whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and multiple environmental parameters and predictions. Set the mapping relationship of thresholds to determine whether the communication environment has changed.
- processor 1002 is further configured to perform the following operations:
- the ninth information includes one or more of the following information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device;
- a machine learning model suitable for the current communication environment is searched to form a second machine learning model set, including:
- the ninth information it is determined that the communication environment has changed
- a machine learning model is searched to form a second machine learning model set.
- processor 1002 is further configured to perform the following operations:
- the multiple machine learning models associated with the set are established or updated, and the third mapping relationship set and the multiple machine learning models associated with the third mapping relationship set are established or updated.
- an embodiment of the present application provides a model updating apparatus in a communication system, which is applied to a terminal.
- the model updating apparatus in the communication system provided in this embodiment may be a terminal, and the model updating apparatus in the communication system may include: a transceiver 1101 , a processor 1102 , and a memory 1103 .
- the transceiver 1101 is used to receive and transmit data under the control of the processor 1102 .
- the bus architecture may include any number of interconnected buses and bridges, specifically, one or more processors represented by processor 1102 and various circuits of memory represented by memory 1103 are linked together.
- the bus architecture may also link together various other circuits, such as peripherals, voltage regulators, and power management circuits, which are well known in the art and, therefore, will not be described further herein.
- the bus interface provides the interface.
- Transceiver 1101 may be multiple elements, ie, including transmitters and receivers, providing means for communicating with various other devices over transmission media including wireless channels, wired channels, fiber optic cables, and the like.
- the processor 1102 is responsible for managing the bus architecture and general processing, and the memory 1103 may store data used by the processor 1102 in performing operations.
- the model updating device in the communication system may also include a user interface 1104.
- the user interface 1104 may also be an interface that can be externally connected to the required equipment, and the connected equipment includes but is not limited to a keypad, a display , speakers, microphones, joysticks, etc.
- the processor 1102 is responsible for managing the bus architecture and general processing, and the memory 1103 may store data used by the processor 1102 in performing operations.
- the processor 1102 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (Field-Programmable Gate Array, FPGA) or a complex programmable Logic device (Complex Programmable Logic Device, CPLD), the processor can also use a multi-core architecture.
- CPU central processing unit
- ASIC application specific integrated circuit
- FPGA field programmable gate array
- CPLD complex programmable Logic Device
- the processor 1102 invokes the computer program stored in the memory 1103 to execute any one of the methods related to the terminal provided by the embodiments of the present application according to the obtained executable instructions.
- the processor and memory may also be physically separated.
- processor 1102 is configured to perform the following operations:
- Receive the first information or seventh information returned by the access network device where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set, and the seventh information is used to instruct the terminal to deploy the second machine learning model Deploying the machine learning models in the learning model set; deploying the machine learning models in the first machine learning model set according to the first information, or deploying the machine learning models in the second machine learning model set according to the seventh information .
- the processor 1102 before receiving the first information or the seventh information returned by the access network device, the processor 1102 is further configured to perform the following operations:
- the second information includes communication environment information on the terminal side; or send third information to the access network device, where the third information is used to instruct the access network device to obtain the first information 2.
- processor 1102 is further configured to perform the following operations:
- the fourth information includes the communication environment information on the terminal side and/or the identification information of the machine learning model requested by the terminal.
- the terminal side is configured with a fourth mapping relationship set, and the fourth mapping relationship set includes the mapping relationship between multiple communication environment information associated with the terminal and multiple machine learning models, according to whether the communication environment on the terminal side occurs or not. change, and send fourth information to the access network device, including:
- the identification information of the machine learning model requested by the terminal is determined in the fourth mapping relationship set according to the communication environment information on the terminal side; and the fourth information is sent to the access network device, wherein , and the fourth information includes identification information of the machine learning model requested by the terminal.
- an embodiment of the present application provides a model updating apparatus in a communication system, which is applied to access network equipment, where a first mapping relationship set and a first mapping relationship set related to the first mapping relationship set are configured on the access network equipment.
- the plurality of machine learning models are connected, and the first mapping relationship set includes the mapping relationship between the plurality of communication environment information and the plurality of machine learning models.
- the model updating apparatus in the communication system in this embodiment may be an access network device, and the model updating apparatus in the communication system includes a search unit 1201 , a deployment unit 1202 and/or a sending unit 1203 .
- a search unit 1201 is used to determine that the communication environment has changed, and according to the first set of mapping relationships, among multiple machine learning models, search for a machine learning model suitable for the current communication environment and form a first set of machine learning models;
- a deployment unit 1202 configured to deploy the machine learning model in the first machine learning model set on the access network device;
- the sending unit 1203 is configured to send first information to the terminal, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set.
- the device further includes:
- an acquisition unit configured to acquire the communication environment information on the access network device side and the communication environment information on the terminal side;
- the search unit 1202 is specifically used for:
- a machine learning model is searched among multiple machine learning models to form a first machine learning model set.
- the device further includes:
- a first receiving unit configured to receive second information sent by the terminal, where the second information includes communication environment information on the terminal side; or, receive third information sent by the terminal, and acquire the second information in response to the third information , where the third information is used to instruct the access network device to obtain the second information.
- the device further includes:
- the first determining unit is configured to, according to each environmental parameter in the communication environment information at the last moment on the access network device side, each environmental parameter in the communication environment information at the current moment on the access network side, and a plurality of environmental parameters and the The mapping relationship of preset thresholds is used to determine whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and the mapping relationship between multiple environmental parameters and preset thresholds to determine whether the communication environment has changed.
- the device further includes:
- the second receiving unit is configured to receive fourth information sent by the terminal, wherein, if the fourth information indicates that the communication environment on the terminal side has changed, the fourth information includes the communication environment information on the terminal side and/or the information requested by the terminal. Identification information of the machine learning model;
- a second determining unit configured to determine whether the communication environment has changed according to the fourth information
- the search unit 1202 is specifically used for:
- the machine learning model is searched and the first machine learning model set is formed.
- the device further includes:
- the third determining unit is configured to determine that the communication environment changes when the terminal switches between cells.
- the device further includes:
- a third receiving unit configured to receive the fifth information sent by the source access network device of the terminal
- the fifth information includes the second mapping relationship set, the communication environment information when the terminal communicates with the source access network device, and/or the identification information of the machine learning model used when the terminal communicates with another access network device.
- the second The mapping relationship set is a subset of the mapping relationship associated with the terminal in the mapping relationship between the multiple communication environment information configured on the source access network device and the multiple machine learning models;
- a first updating unit configured to update the first mapping relationship set according to the fifth information, and determine that the communication environment has changed
- the fourth receiving unit is configured to receive all or part of the machine learning model associated with the second mapping relationship set sent by the source access network device if all or part of the machine learning model associated with the second mapping relationship set is not configured in the access network device. or part of a machine learning model;
- the second updating unit is configured to determine, according to the received all or part of the machine learning models associated with the second mapping relationship set and a plurality of machine learning models associated with the first mapping relationship set before updating, the A plurality of machine learning models associated with a mapping relationship set;
- the search unit 1202 is specifically used for:
- a machine learning model suitable for the current communication environment is searched to form a first machine learning model set.
- the first updating unit is specifically configured to: combine the second mapping relationship set with the first mapping relationship set to obtain an updated first mapping relationship set.
- the device further includes:
- the fourth determining unit is configured to determine that the communication environment has changed if a message indicating RRC connection establishment or a message indicating RRC connection re-establishment sent by the terminal is received.
- an embodiment of the present application provides a model updating device in a communication system.
- the model updating device in the communication system in this embodiment may be core network equipment, and the model in the communication system
- the updating apparatus includes: a searching unit 1301 and a sending unit 1302 .
- a search unit 1301 is used to determine that the communication environment has changed, and according to the third set of mapping relationships, among multiple machine learning models, search for a machine learning model suitable for the current communication environment and form a second set of machine learning models;
- a sending unit 1302 configured to send sixth information including the second machine learning model set to the access network device, where the sixth information is used to instruct the access network device to deploy the machine learning model in the second machine learning model set , and/or instruct the access network device to send seventh information to the terminal, where the seventh information is used to instruct the terminal to deploy the machine learning model in the second machine learning model set.
- the device further includes:
- an acquisition unit configured to acquire the communication environment information on the access network device side and the communication environment information on the terminal side;
- the search unit 1301 is specifically used for:
- a machine learning model is searched among the multiple machine learning models to form a second machine learning model set.
- the device further includes:
- the first receiving unit is configured to receive eighth information sent by the access network device, where the eighth information includes communication environment information on the access network device side and communication environment information on the terminal side.
- the device further includes:
- the determining unit is used for determining according to each environmental parameter in the communication environment information at the last moment on the access network device side, each environmental parameter in the communication environment information at the current moment on the access network side, and a plurality of environmental parameters and presets
- the mapping relationship of the thresholds determines whether the communication environment has changed; and/or, according to each environmental parameter in the communication environment information at the last moment on the terminal side, each environmental parameter in the communication environment information at the current moment on the terminal side, and other parameters.
- the mapping relationship between each environmental parameter and the preset threshold value determines whether the communication environment changes.
- the device further includes:
- a second receiving unit configured to receive ninth information sent by the access network device
- the ninth information is used to indicate that the communication environment changes, the terminal is switched between cells, the RRC connection between the access network device and the terminal is established, or the RRC connection between the access network device and the terminal is reestablished, and the ninth information includes the following One or more pieces of information: communication environment information on the access network device side, communication environment information on the terminal side, and identification information of the machine learning model requested by the access network device;
- the search unit 1301 is specifically used for:
- the ninth information it is determined that the communication environment has changed
- a machine learning model is searched to form a second machine learning model set.
- the device further includes:
- a third receiving unit configured to receive a first mapping relationship set sent by multiple access network devices and multiple machine learning models associated with the first mapping relationship set;
- a model processing unit configured to establish or update a third mapping relationship set and a third mapping relationship with the third mapping relationship set according to the first mapping relationship set sent by multiple access network devices and multiple machine learning models associated with the first mapping relationship set Sets associated multiple machine learning models.
- an embodiment of the present application provides a model updating device in a communication system.
- the model updating device in the communication system in this embodiment may be a terminal, and the model updating device in the communication system includes : the receiving unit 1401 and the deploying unit 1402 .
- the receiving unit 1401 is configured to receive first information or seventh information returned by the access network device, where the first information is used to instruct the terminal to deploy the machine learning model in the first machine learning model set, and the seventh information is used to instructing the terminal to deploy the machine learning model in the second machine learning model set;
- the deployment unit 1402 is configured to deploy the machine learning model in the first machine learning model set according to the first information, or deploy the machine learning model in the second machine learning model set according to the seventh information.
- the device further includes a first sending unit for:
- the device further includes:
- the second sending unit is configured to send fourth information to the access network device according to whether the communication environment on the terminal side has changed, wherein, if the fourth information indicates that the communication environment on the terminal side has changed, the fourth information includes the terminal Communication environment information on one side and/or identification information of the machine learning model requested by the terminal.
- a fourth mapping relationship set is configured on one side of the terminal, and the fourth mapping relationship set includes mapping relationships between multiple communication environment information associated with the terminal and multiple machine learning models, and the second sending unit is specifically used for:
- each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
- the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
- the integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a processor-readable storage medium.
- the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, and the computer software products are stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods described in the various embodiments of the present application.
- the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes .
- an embodiment of the present application provides a processor-readable storage medium, where the processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the relevant access provided by the embodiment of the present application.
- the processor-readable storage medium stores a computer program
- the computer program is used to enable the processor to execute the relevant access provided by the embodiment of the present application.
- an embodiment of the present application provides a processor-readable storage medium, where the processor-readable storage medium stores a computer program, and the computer program is used to cause the processor to execute the relevant core network provided by the embodiment of the present application.
- the embodiments of the present application provide a processor-readable storage medium, where the processor-readable storage medium stores a computer program, and the computer program is used to make the processor execute any one of the related terminals provided by the embodiments of the present application.
- the described method enables the processor to implement all the method steps implemented by the terminal in the above method embodiments, and can achieve the same technical effects, and the same parts and beneficial effects as the method embodiments in this embodiment will not be performed here. Describe in detail.
- the processor-readable storage medium can be any available medium or data storage device that can be accessed by a processor, including, but not limited to, magnetic storage (eg, floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.), optical storage (eg, CD, DVD, BD, HVD, etc.), and semiconductor memory (eg, ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state disk (SSD)), etc.
- magnetic storage eg, floppy disk, hard disk, magnetic tape, magneto-optical disk (MO), etc.
- optical storage eg, CD, DVD, BD, HVD, etc.
- semiconductor memory eg, ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state disk (SSD)
- an embodiment of the present application further provides a computer program product containing instructions, the computer program is stored in a storage medium, and at least one processor can read the computer program from the storage medium
- a computer program when the at least one processor executes the computer program, it can implement all the method steps implemented by the access network device in the above method embodiments, and can achieve the same technical effect, and the reference to this embodiment will not be discussed here.
- the same parts and beneficial effects of the method embodiments will be described in detail.
- an embodiment of the present application further provides a computer program product containing instructions, the computer program is stored in a storage medium, and at least one processor can read the computer from the storage medium Program, when the at least one processor executes the computer program, it can implement all the method steps implemented by the core network device in the above method embodiments, and can achieve the same technical effect, and the method in this embodiment will not be implemented here.
- the same parts and beneficial effects of the examples will be described in detail.
- an embodiment of the present application further provides a computer program product containing instructions, the computer program is stored in a storage medium, and at least one processor can read the computer program from the storage medium, When the at least one processor executes the computer program, all the method steps implemented by the terminal in the above method embodiments can be implemented, and the same technical effect can be achieved, and the parts in this embodiment that are the same as those in the method embodiments will not be described here. And the beneficial effects will be described in detail.
- the embodiments of the present application further provide a communication system, including core network equipment, access network equipment, and terminals.
- the core network equipment is the core network equipment described in the foregoing apparatus embodiments, which can perform all the method steps performed by the core network equipment in the foregoing method embodiments, and can achieve the same technical effect.
- the access network equipment is the access network equipment described in the foregoing apparatus embodiments, and can execute all the method steps performed by the access network equipment in the foregoing method embodiments, and can achieve the same technical effect.
- the terminal is the terminal described in the foregoing apparatus embodiments, and can execute all the method steps executed by the terminal in the foregoing method embodiments, and can achieve the same technical effect. The same parts and beneficial effects in this embodiment as in the method embodiment will not be described in detail here.
- the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, optical storage, and the like.
- processor-executable instructions may also be stored in a processor-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the processor-readable memory result in the manufacture of means including the instructions product, the instruction means implements the functions specified in the flow or flow of the flowchart and/or the block or blocks of the block diagram.
- processor-executable instructions can also be loaded onto a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process that Execution of the instructions provides steps for implementing the functions specified in the flowchart or blocks and/or the block or blocks of the block diagrams.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
Description
Claims (40)
- 一种通信***中的模型更新方法,应用于接入网设备,其特征在于,所述接入网设备上配置有第一映射关系集及与所述第一映射关系集相关联的多个机器学习模型,所述第一映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述方法包括:确定通信环境发生变化,根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合;在所述接入网设备上部署所述第一机器学习模型集合中的机器学习模型,和/或,向终端发送第一信息,其中,所述第一信息用于指示所述终端对所述第一机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求1所述的方法,其特征在于,所述方法还包括:获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息;所述根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合,包括:根据所述第一映射关系集、所述接入网设备一侧的通信环境信息以及所述终端一侧的通信环境信息,在所述多个机器学习模型中查找机器学习模型并构成所述第一机器学习模型集合。
- 根据权利要求2所述的方法,其特征在于,所述获取所述终端一侧的通信环境信息包括:接收所述终端发送的第二信息,其中,所述第二信息包括所述终端一侧的通信环境信息;或者,接收所述终端发送的第三信息,并响应于所述第三信息,获取所述第二信息,其中,所述第三信息用于指示所述接入网设备获取所述第二信息。
- 根据权利要求2或3所述的方法,其特征在于,所述方法还包括:根据所述接入网设备一侧上一时刻的通信环境信息中的各环境参数、所述接入网一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化;和/或,根据所述终端一侧上一时刻的通信环境信息中的各环境参数、所述终端一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化。
- 根据权利要求1所述的方法,其特征在于,所述方法还包括:接收所述终端发送的第四信息,其中,若所述第四信息中指示所述终端一侧的通信环境发生变化,则所述第四信息包括所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息;根据所述第四信息,确定通信环境是否发生变化;如果通信环境发生变化,则根据所述第一映射关系集和所述接入网设备一侧的通信环境信息,以及根据所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息,在所述多个机器学习模型中,查找机器学习模型并构成所述第一机器 学习模型集合。
- 根据权利要求1所述的方法,其特征在于,所述方法还包括:当所述终端发生小区切换时,确定通信环境发生变化。
- 根据权利要求6所述的方法,其特征在于,所述方法还包括:接收所述终端的源接入网设备发送的第五信息;其中,所述第五信息包括第二映射关系集、所述终端与所述源接入网设备通信时的通信环境信息和/或所述终端与所述源接入网设备通信时所采用的机器学习模型的标识信息,所述第二映射关系集为所述源接入网设备上配置的多个通信环境信息与多个机器学习模型的映射关系中与所述终端相关联的映射关系子集;根据所述第五信息,对所述第一映射关系集进行更新;如果所述接入网设备中未配置与所述第二映射关系集相关联的全部或部分机器学习模型,则接收所述源接入网设备发送的与所述第二映射关系集相关联的全部或部分机器学习模型;根据接收的与所述第二映射关系集相关联的全部或部分机器学习模型和与更新前的所述第一映射关系集相关联的多个机器学习模型,确定与更新后的所述第一映射关系集相关联的多个机器学习模型;所述根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合,包括:根据更新后的所述第一映射关系集,在与更新后的所述第一映射关系集相关联的多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成所述第一机器学习模型集合。
- 根据权利要求1-3中任一项所述的方法,其特征在于,所述方法还包括:如果接收到所述终端发送的指示无线资源控制RRC连接建立的消息或者指示RRC连接重建的消息,则确定通信环境发生变化。
- 一种通信***的模型更新方法,应用于核心网设备,其特征在于,所述核心网设备上预先配置有第三映射关系集及与所述第三映射关系集相关联的多个机器学习模型,所述第三映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述方法包括:确定通信环境发生变化,根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合;向接入网设备发送包括所述第二机器学习模型集合的第六信息,其中,所述第六信息用于指示所述接入网设备对所述第二机器学习模型集合中的机器学习模型进行部署,和/或,指示所述接入网设备向终端发送第七信息,所述第七信息用于指示所述终端对所述第二机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求9所述的方法,其特征在于,所述方法还包括:获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息;所述根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合,包括:根据所述第三映射关系集、所述接入网设备一侧的通信环境信息以及所述终端一 侧的通信环境信息,在所述多个机器学习模型中查找机器学习模型并构成所述第二机器学习模型集合。
- 根据权利要求10所述的方法,其特征在于,所述获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息,包括:接收所述接入网设备发送的第八信息,所述第八信息包括所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息。
- 根据权利要求10或11所述的方法,其特征在于,所述方法还包括:根据所述接入网设备一侧上一时刻的通信环境信息中的各环境参数、所述接入网一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化;和/或,根据所述终端一侧上一时刻的通信环境信息中的各环境参数、所述终端一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化。
- 根据权利要求9所述的方法,其特征在于,所述方法还包括:接收所述接入网设备发送的第九信息;其中,所述第九信息用于指示通信环境发生变化、所述终端发生小区切换、所述接入网设备与所述终端之间的RRC连接建立、或者所述接入网设备与所述终端之间的RRC连接重建,所述第九信息包括如下一种或多种信息:所述接入网设备一侧的通信环境信息、所述终端一侧的通信环境信息、所述接入网设备请求的机器学习模型的标识信息;所述确定通信环境发生变化,根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合,包括:根据所述第九信息,确定通信环境发生变化;根据所述第三映射关系集和所述第九信息,在所述多个机器学习模型中,查找机器学习模型并构成所述第二机器学习模型集合。
- 根据权利要求9-11中任一项所述的方法,其特征在于,所述方法还包括:接收多个接入网设备发送的第一映射关系集和与所述第一映射关系集相关联的多个机器学习模型;根据所述多个接入网设备发送的第一映射关系集和与所述第一映射关系集相关联的多个机器学习模型,建立或更新所述第三映射关系集及与所述第三映射关系集相关联的多个机器学习模型。
- 一种通信***中的模型更新方法,应用于终端,其特征在于,所述方法包括:接收接入网设备返回的第一信息或第七信息,其中,所述第一信息用于指示所述终端对第一机器学习模型集合中的机器学习模型,所述第七信息用于指示所述终端对第二机器学习模型集合中的机器学习模型进行部署;根据所述第一信息对所述第一机器学习模型集合中的机器学习模型进行部署,或者,根据所述第七信息对所述第二机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求15所述的方法,其特征在于,所述接收接入网设备返回的第一信息或第七信息之前,所述方法还包括:向所述接入网设备发送第二信息,其中,所述第二信息包括所述终端一侧的通信环境信息;或者,向所述接入网设备发送第三信息,其中,所述第三信息用于指示所述接入网设备获取所述第二信息。
- 根据权利要求15或16所述的方法,其特征在于,所述方法还包括:确定所述终端一侧的通信环境是否发生变化;根据所述终端一侧的通信环境是否发生变化,向所述接入网设备发送第四信息,其中,如果所述第四信息指示所述终端一侧的通信环境发生变化,则所述第四信息包括所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息。
- 根据权利要求17所述的方法,其特征在于,所述终端一侧配置有第四映射关系集,所述第四映射关系集包括与所述终端相关联的多个通信环境信息与多个机器学习模型的映射关系,所述根据所述终端一侧的通信环境是否发生变化,向所述接入网设备发送第四信息,包括:如果确定所述终端一侧的通信环境发生变化,则根据所述终端一侧的通信环境信息,在所述第四映射关系集中,确定所述终端请求的机器学习模型的标识信息;向所述接入网设备发送第四信息,其中,所述第四信息包括所述终端请求的机器学习模型的标识信息。
- 一种通信***中的模型更新装置,应用于接入网设备,其特征在于,所述接入网设备上配置有第一映射关系集及与所述第一映射关系集相关联的多个机器学习模型,所述第一映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述装置包括存储器、收发机和处理器:所述存储器,用于存储计算机程序;所述收发机,用于在所述处理器的控制下收发数据;所述处理器,用于读取所述存储器中的计算机程序并执行以下操作:确定通信环境发生变化,根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合;在所述接入网设备上部署所述第一机器学习模型集合中的机器学习模型,和/或,向终端发送第一信息,其中,所述第一信息用于指示所述终端对所述第一机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求19所述的装置,其特征在于,所述处理器还用于执行以下操作:获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息;所述根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合,包括:根据所述第一映射关系集、所述接入网设备一侧的通信环境信息、以及所述终端一侧的通信环境信息,在所述多个机器学习模型中查找机器学习模型并构成所述第一机器学习模型集合。
- 根据权利要求20所述的装置,其特征在于,所述获取所述终端一侧的通信环境信息,包括:接收所述终端发送的第二信息,其中,所述第二信息包括所述终端一侧的通信环 境信息;或者,接收所述终端发送的第三信息,并响应于所述第三信息,获取所述第二信息,其中,所述第三信息用于指示所述接入网设备获取所述第二信息。
- 根据权利要求20或21所述的装置,其特征在于,所述处理器还用于执行以下操作:根据所述接入网设备一侧上一时刻的通信环境信息中的各环境参数、所述接入网一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化;和/或,根据所述终端一侧上一时刻的通信环境信息中的各环境参数、所述终端一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化。
- 根据权利要求19所述的装置,其特征在于,所述处理器还用于执行以下操作:接收所述终端发送的第四信息,其中,若所述第四信息中指示所述终端一侧的通信环境发生变化,则所述第四信息包括所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息;根据所述第四信息,确定通信环境是否发生变化;如果通信环境发生变化,则根据所述第一映射关系集和所述接入网设备一侧的通信环境信息,以及根据所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息,在所述多个机器学习模型中,查找机器学习模型并构成所述第一机器学习模型集合。
- 根据权利要求19所述的装置,其特征在于,所述处理器还用于执行以下操作:当所述终端发生小区切换时,确定通信环境发生变化。
- 根据权利要求24所述的装置,其特征在于,所述处理器还用于执行以下操作:接收所述终端的源接入网设备发送的第五信息;其中,所述第五信息包括第二映射关系集、所述终端与所述源接入网设备通信时的通信环境信息和/或所述终端与所述源接入网设备通信时所采用的机器学习模型的标识信息,所述第二映射关系集为所述源接入网设备上配置的多个通信环境信息与多个机器学习模型的映射关系中与所述终端相关联的映射关系子集;根据所述第五信息,对所述第一映射关系集进行更新;如果所述接入网设备中未配置与所述第二映射关系集相关联的全部或部分机器学习模型,则接收所述源接入网设备发送的与所述第二映射关系集相关联的全部或部分机器学习模型;根据接收的与所述第二映射关系集相关联的全部或部分机器学习模型和与更新前的所述第一映射关系集相关联的多个机器学习模型,确定与更新后的所述第一映射关系集相关联的多个机器学习模型;所述根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合,包括:根据更新后的所述第一映射关系集,在与更新后的所述第一映射关系集相关联的多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成所述第一机器 学习模型集合。
- 根据权利要求19-21中任一项所述的装置,其特征在于,所述处理器还用于执行以下操作:如果接收到所述终端发送的指示RRC连接建立的消息或者指示RRC连接重建的消息,则确定通信环境发生变化。
- 一种通信***的模型更新装置,应用于核心网设备,其特征在于,所述核心网设备上预先配置有第三映射关系集及与所述第三映射关系集相关联的多个机器学习模型,所述第三映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述装置包括存储器、收发机和处理器:所述存储器,用于存储计算机程序;所述收发机,用于在所述处理器的控制下收发数据;所述处理器,用于读取所述存储器中的计算机程序并执行以下操作:确定通信环境发生变化,根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合;向接入网设备发送包括所述第二机器学习模型集合的第六信息,其中,所述第六信息用于指示所述接入网设备对所述第二机器学习模型集合中的机器学习模型进行部署,和/或,指示所述接入网设备向终端发送第七信息,所述第七信息用于指示所述终端对所述第二机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求27所述的装置,其特征在于,所述处理器还用于执行以下操作:获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息;所述根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合,包括:根据所述第三映射关系集、所述接入网设备一侧的通信环境信息以及所述终端一侧的通信环境信息,在所述多个机器学习模型中查找机器学习模型并构成所述第二机器学习模型集合。
- 根据权利要求28所述的装置,其特征在于,所述获取所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息,包括:接收所述接入网设备发送的第八信息,所述第八信息包括所述接入网设备一侧的通信环境信息和所述终端一侧的通信环境信息。
- 根据权利要求28或29所述的装置,其特征在于,所述处理器还用于执行以下操作:根据所述接入网设备一侧上一时刻的通信环境信息中的各环境参数、所述接入网一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化;和/或,根据所述终端一侧上一时刻的通信环境信息中的各环境参数、所述终端一侧当前时刻的通信环境信息中的各环境参数、以及多个环境参数与预设阈值的映射关系,确定通信环境是否发生变化。
- 根据权利要求27所述的装置,其特征在于,所述处理器还用于执行以下操作:接收所述接入网设备发送的第九信息;其中,所述第九信息用于指示通信环境发生变化、所述终端发生小区切换、所述接入网设备与所述终端之间的RRC连接建立、或者所述接入网设备与所述终端之间的RRC连接重建,所述第九信息包括如下一种或多种信息:所述接入网设备一侧的通信环境信息、所述终端一侧的通信环境信息、所述接入网设备请求的机器学习模型的标识信息;所述确定通信环境发生变化,根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合,包括:根据所述第九信息,确定通信环境发生变化;根据所述第三映射关系集和所述第九信息,在所述多个机器学习模型中,查找机器学习模型并构成所述第二机器学习模型集合。
- 根据权利要求27-29中任一项所述的装置,其特征在于,所述处理器还用于执行以下操作:接收多个接入网设备发送的第一映射关系集和与所述第一映射关系集相关联的多个机器学习模型;根据所述多个接入网设备发送的第一映射关系集和与所述第一映射关系集相关联的多个机器学习模型,建立或更新所述第三映射关系集及与所述第三映射关系集相关联的多个机器学习模型。
- 一种通信***中的模型更新装置,应用于终端,其特征在于,所述装置包括存储器、收发机和处理器:所述存储器,用于存储计算机程序;所述收发机,用于在所述处理器的控制下收发数据;所述处理器,用于读取所述存储器中的计算机程序并执行以下操作:接收接入网设备返回的第一信息或第七信息,其中,所述第一信息用于指示所述终端对第一机器学习模型集合中的机器学习模型或对第二机器学习模型集合中的机器学习模型进行部署;根据所述第一信息对所述第一机器学习模型集合中的机器学习模型进行部署,或者,根据所述第七信息对所述第二机器学习模型集合中的机器学习模型进行部署。
- 根据权利要求33所述的装置,其特征在于,所述接收接入网设备返回的第一信息或第七信息之前,所述处理器还用于执行以下操作:向所述接入网设备发送第二信息,其中,所述第二信息包括所述终端一侧的通信环境信息;或者,向所述接入网设备发送第三信息,其中,所述第三信息用于指示所述接入网设备获取所述第二信息。
- 根据权利要求33或34所述的装置,其特征在于,所述处理器还用于执行以下操作:确定所述终端一侧的通信环境是否发生变化;根据所述终端一侧的通信环境是否发生变化,向所述接入网设备发送第四信息,其中,如果所述第四信息指示所述终端一侧的通信环境发生变化,则所述第四信息包括所述终端一侧的通信环境信息和/或所述终端请求的机器学习模型的标识信息。
- 根据权利要求35所述的装置,其特征在于,所述终端一侧配置有第四映射关系集,所述第四映射关系集包括与所述终端相关联的多个通信环境信息与多个机器学习模型的映射关系,所述根据所述终端一侧的通信环境是否发生变化,向所述接入网设备发送第四信息,包括:如果确定所述终端一侧的通信环境发生变化,则根据所述终端一侧的通信环境信息,在所述第四映射关系集中,确定所述终端请求的机器学习模型的标识信息;向所述接入网设备发送第四信息,其中,所述第四信息用于至少所述终端一侧的通信环境发生变化,且所述第四信息包括所述终端请求的机器学习模型的标识信息。
- 一种通信***中的模型更新装置,应用于接入网设备,其特征在于,所述接入网设备上配置有第一映射关系集及与所述第一映射关系集相关联的多个机器学习模型,所述第一映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述装置包括:查找单元,用于确定通信环境发生变化,根据所述第一映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第一机器学习模型集合;部署单元,用于在所述接入网设备上部署所述第一机器学习模型集合中的机器学习模型,和/或,发送单元,用于向终端发送第一信息,其中,所述第一信息用于指示所述终端对所述第一机器学习模型集合中的机器学习模型进行部署。
- 一种通信***的模型更新装置,应用于核心网设备,其特征在于,所述核心网设备上预先配置有第三映射关系集及与所述第三映射关系集相关联的多个机器学习模型,所述第三映射关系集包括多个通信环境信息与所述多个机器学习模型的映射关系,所述装置包括:查找单元,用于确定通信环境发生变化,根据所述第三映射关系集,在所述多个机器学习模型中,查找适用于当前通信环境的机器学习模型并构成第二机器学习模型集合;发送单元,用于向接入网设备发送包括所述第二机器学习模型集合的第六信息,其中,所述第六信息用于指示所述接入网设备对所述第二机器学习模型集合中的机器学习模型进行部署,和/或,指示所述接入网设备向终端发送第七信息,所述第七信息用于指示所述终端对所述第二机器学习模型集合中的机器学习模型进行部署。
- 一种通信***中的模型更新装置,应用于终端,其特征在于,所述装置包括:接收单元,用于接收接入网设备返回的第一信息或第七信息,其中,所述第一信息用于指示所述终端对适用于当前通信环境的第一机器学习模型集合中的机器学习模型进行部署,所述第七信息用于指示所述终端对第二机器学习模型集合中的机器学习模型进行部署;部署单元,用于根据所述第一信息对所述第一机器学习模型集合中的机器学习模型进行部署,或者,根据所述第七信息对所述第二机器学习模型集合中的机器学习模型进行部署。
- 一种处理器可读存储介质,其特征在于,所述处理器可读存储介质存储有计算机程序,所述计算机程序用于使所述处理器执行权利要求1至8中任一项所述的方 法、或者权利要求9-14中任一项所述的方法、或者权利要求15-18中任一项所述的方法。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP22745105.1A EP4287574A4 (en) | 2021-02-01 | 2022-01-19 | METHOD AND APPARATUS FOR UPDATING MODEL IN COMMUNICATION SYSTEM, AND STORAGE MEDIUM |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110138849.5A CN114844785B (zh) | 2021-02-01 | 2021-02-01 | 通信***中的模型更新方法、装置及存储介质 |
CN202110138849.5 | 2021-02-01 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022161230A1 true WO2022161230A1 (zh) | 2022-08-04 |
Family
ID=82561000
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2022/072837 WO2022161230A1 (zh) | 2021-02-01 | 2022-01-19 | 通信***中的模型更新方法、装置及存储介质 |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP4287574A4 (zh) |
CN (1) | CN114844785B (zh) |
WO (1) | WO2022161230A1 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024149288A1 (zh) * | 2023-01-12 | 2024-07-18 | 维沃移动通信有限公司 | Ai模型分发、接收方法、终端及网络侧设备 |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024040617A1 (en) * | 2022-08-26 | 2024-02-29 | Qualcomm Incorporated | Ml model generalization and specification |
WO2024087119A1 (zh) * | 2022-10-27 | 2024-05-02 | Oppo广东移动通信有限公司 | 模型控制方法、装置、设备及介质 |
CN118302998A (zh) * | 2022-11-03 | 2024-07-05 | 北京小米移动软件有限公司 | 模型的标识方法及装置 |
WO2024092755A1 (en) * | 2022-11-04 | 2024-05-10 | Huawei Technologies Co., Ltd. | Management of machine learning models in communication systems |
CN118211670A (zh) * | 2022-12-16 | 2024-06-18 | 大唐移动通信设备有限公司 | 一种模型同步方法、装置、设备及存储介质 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101146312A (zh) * | 2007-10-17 | 2008-03-19 | 中兴通讯股份有限公司 | 一种无线传播模型的自动选择方法 |
CN108197714A (zh) * | 2018-01-30 | 2018-06-22 | 北京小米移动软件有限公司 | 利用机器学习模型进行工作模式判决的方法及装置 |
US10165105B1 (en) * | 2017-09-19 | 2018-12-25 | Qualcomm Incorporated | Method and system for user equipment communication mode selection |
CN109446524A (zh) * | 2018-10-25 | 2019-03-08 | 第四范式(北京)技术有限公司 | 一种语音质检方法及装置 |
CN111340232A (zh) * | 2020-02-17 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | 在线预测服务的部署方法、装置、电子设备及存储介质 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102007044671B4 (de) * | 2007-09-18 | 2013-02-21 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Verfahren zur Schätzung von Parametern eines in einer dynamischen Mehrwegeumgebung empfangenen GNSS-Navigationssignals |
CN107705359B (zh) * | 2017-10-13 | 2020-07-07 | 北京邮电大学 | 利用三维视觉重建技术的信道建模方法和装置 |
US11544617B2 (en) * | 2018-04-23 | 2023-01-03 | At&T Intellectual Property I, L.P. | Network-based machine learning microservice platform |
CN109150775B (zh) * | 2018-08-14 | 2020-03-17 | 西安交通大学 | 自适应噪声环境动态变化的鲁棒性在线信道状态估计方法 |
US20210345134A1 (en) * | 2018-10-19 | 2021-11-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Handling of machine learning to improve performance of a wireless communications network |
WO2020139181A1 (en) * | 2018-12-28 | 2020-07-02 | Telefonaktiebolaget Lm Ericsson (Publ) | A wireless device, a network node and methods therein for updating a first instance of a machine learning model |
GB201903172D0 (en) * | 2019-03-08 | 2019-04-24 | Kortical Ltd | Methods of deploying machine learning models |
CN112166445A (zh) * | 2019-04-16 | 2021-01-01 | 华为技术有限公司 | 基于区块链网络的联合学习方法和联合学习设备 |
CN112054863B (zh) * | 2019-06-06 | 2021-12-21 | 华为技术有限公司 | 一种通信方法及装置 |
-
2021
- 2021-02-01 CN CN202110138849.5A patent/CN114844785B/zh active Active
-
2022
- 2022-01-19 WO PCT/CN2022/072837 patent/WO2022161230A1/zh active Application Filing
- 2022-01-19 EP EP22745105.1A patent/EP4287574A4/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101146312A (zh) * | 2007-10-17 | 2008-03-19 | 中兴通讯股份有限公司 | 一种无线传播模型的自动选择方法 |
US10165105B1 (en) * | 2017-09-19 | 2018-12-25 | Qualcomm Incorporated | Method and system for user equipment communication mode selection |
CN108197714A (zh) * | 2018-01-30 | 2018-06-22 | 北京小米移动软件有限公司 | 利用机器学习模型进行工作模式判决的方法及装置 |
CN109446524A (zh) * | 2018-10-25 | 2019-03-08 | 第四范式(北京)技术有限公司 | 一种语音质检方法及装置 |
CN111340232A (zh) * | 2020-02-17 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | 在线预测服务的部署方法、装置、电子设备及存储介质 |
Non-Patent Citations (2)
Title |
---|
HOWARD H. YANG, ZHAO ZHONGYUAN, TONY Q. S. QUEK: "Enabling Intelligence at Network Edge: An Overview of Federated Learning", SINGAPORE UNIVERSITY OF TECHNOLOGY AND DESIGN, SINGAPORE 487372, SINGAPORE%BEIJING UNIVERSITY OF POST AND TELECOMMUNICATION, BEIJING 100876, CHINA, 30 June 2020 (2020-06-30), pages 2 - 10, XP055953723, Retrieved from the Internet <URL:https://res-www.zte.com.cn/mediares/magazine/publication/com_en/article/202002/202002002.pdf> [retrieved on 20220822], DOI: 10.12142/ZTECOM.202002002 * |
See also references of EP4287574A4 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024149288A1 (zh) * | 2023-01-12 | 2024-07-18 | 维沃移动通信有限公司 | Ai模型分发、接收方法、终端及网络侧设备 |
Also Published As
Publication number | Publication date |
---|---|
CN114844785B (zh) | 2024-02-06 |
EP4287574A4 (en) | 2024-07-03 |
CN114844785A (zh) | 2022-08-02 |
EP4287574A1 (en) | 2023-12-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022161230A1 (zh) | 通信***中的模型更新方法、装置及存储介质 | |
US11510215B2 (en) | Electronic device and method for radio resource management (RRM) measurement relaxation | |
US9107123B2 (en) | Systems and methods for limiting mobile device measurements for cell reselection and handover | |
US9578560B1 (en) | Systems and methods for managing network connections | |
WO2013138225A1 (en) | Mobility parameter adjustment and mobility state estimation in heterogeneous networks | |
JP7495524B2 (ja) | ローカルエリア機械学習無線リソース管理のための測定構成 | |
US9749912B2 (en) | Measurement configuration for heterogenous networks | |
US10257755B2 (en) | Handover of a user equipment having offloaded user data traffic between two small cells controlled by different macro cells | |
WO2022237812A1 (zh) | 条件重配方法、设备、装置及存储介质 | |
WO2018059151A1 (zh) | 资源管理指示方法及装置 | |
US20230023603A1 (en) | Measurement object indication method and apparatus | |
Sun et al. | A universal predictive mobility management scheme for urban ultra-dense networks with control/data plane separation | |
CN103959859B (zh) | 用于选择性地重置和转移无线通信信道的方法和装置 | |
CN103188750A (zh) | 切换方法、节点和*** | |
Huang et al. | A handover scheme for LTE wireless networks under the assistance of GPS | |
Aboud et al. | Markov chain based predictive model for efficient handover management in vehicle-to-infrastructure communications | |
WO2022209234A1 (ja) | Ranノード、ue及び方法 | |
CN114143839B (zh) | Ng链路断链时的处理方法、网络设备、装置及存储介质 | |
CN116097752A (zh) | 切换方法、装置、设备及存储介质 | |
WO2021248391A1 (zh) | 无线通信的方法和终端设备 | |
CN106688269B (zh) | 用于确定无线设备是否是由于负载平衡原因而切换到目标小区的合适候选者的无线电网络节点和方法 | |
WO2020038442A1 (zh) | 一种信息传递的方法和装置 | |
US20240244499A1 (en) | Methods and apparatus for providing mobility state information | |
WO2024152734A1 (zh) | 锚点终端确定方法、装置及存储介质 | |
WO2024125509A1 (zh) | 模型更新方法、装置、设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22745105 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18263703 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022745105 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022745105 Country of ref document: EP Effective date: 20230901 |