WO2024139524A1 - 辅助信息上报方法及装置 - Google Patents

辅助信息上报方法及装置 Download PDF

Info

Publication number
WO2024139524A1
WO2024139524A1 PCT/CN2023/124022 CN2023124022W WO2024139524A1 WO 2024139524 A1 WO2024139524 A1 WO 2024139524A1 CN 2023124022 W CN2023124022 W CN 2023124022W WO 2024139524 A1 WO2024139524 A1 WO 2024139524A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
auxiliary information
terminal
sub
model
Prior art date
Application number
PCT/CN2023/124022
Other languages
English (en)
French (fr)
Inventor
严雪
梁靖
曾二林
王达
费永强
Original Assignee
大唐移动通信设备有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 大唐移动通信设备有限公司 filed Critical 大唐移动通信设备有限公司
Publication of WO2024139524A1 publication Critical patent/WO2024139524A1/zh

Links

Definitions

  • First auxiliary information is sent to a network side device, where the first auxiliary information includes at least part of the information required to implement a model management process of artificial intelligence AI and/or machine learning ML.
  • the first configuration information includes one or more of the following:
  • the sending the first auxiliary information to the network side device includes one or more of the following:
  • the method before sending the first auxiliary information to the network side device, the method further includes one or more of the following:
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the terminal, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the network side device;
  • first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • the first auxiliary information includes:
  • the first auxiliary information is sent to the network side device.
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the terminal, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the network side device;
  • the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented in the terminal, or the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented in the network side device;
  • the sixth sub-auxiliary information is used to represent that the terminal expects to change the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to activate the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to deactivate the model of the AI and/or ML;
  • eighth sub-auxiliary information where the eighth sub-auxiliary information is used to indicate that the terminal expects to update the AI and/or ML model
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the first auxiliary information when the first auxiliary information includes sixth sub-auxiliary information, the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the seventh sub-auxiliary information includes one or more of the following:
  • the model identification information includes:
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • an embodiment of the present disclosure further provides a network side device, including a memory, a transceiver, and a processor, wherein:
  • a memory for storing a computer program; a transceiver for transmitting and receiving data under the control of the processor; and a processor for reading the computer program in the memory and performing the following operations:
  • the first auxiliary information sent by the receiving terminal includes at least part of the information required to implement the model management process of artificial intelligence AI and/or machine learning ML.
  • the operation before receiving the first auxiliary information sent by the terminal, the operation further includes:
  • first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the first auxiliary information includes one or more of the following:
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to characterize the terminal's expected model training The process is implemented in the terminal, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented in the network side device;
  • fifth auxiliary sub-information wherein the fifth auxiliary sub-information is used to indicate that the terminal expects the model update process to be implemented at the terminal, or the fifth auxiliary sub-information is used to indicate that the terminal expects the model update process to be implemented at the network side device;
  • the seventh sub-auxiliary information is used to characterize model information of the AI and/or ML expected or preferred by the terminal;
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • the first auxiliary information when the first auxiliary information includes eighth sub-auxiliary information, the first auxiliary information includes:
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the seventh sub-auxiliary information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • the model identification information includes:
  • the embodiment of the present disclosure further provides an auxiliary information reporting device, including:
  • the first configuration information includes one or more of the following:
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • Second indication information where the second indication information is used to indicate the reason why the terminal expects to change the AI and/or ML model, or the reason why the terminal expects to activate the AI and/or ML model, or the reason why the terminal expects to deactivate the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes eighth sub-auxiliary information, the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the seventh auxiliary information sub-information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • the model identification information includes:
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • an embodiment of the present disclosure further provides an auxiliary information reporting device, including:
  • the receiving module is used to receive first auxiliary information sent by the terminal, where the first auxiliary information includes at least part of the information required to implement the model management process of artificial intelligence AI and/or machine learning ML.
  • the device further includes:
  • the configuration information sending module is used to send first configuration information to the terminal before receiving the first auxiliary information sent by the terminal, where the first configuration information is the configuration information required by the terminal to report the first auxiliary information to the network side device.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the first auxiliary information includes one or more of the following:
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the terminal, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented at the network side device;
  • the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented in the terminal, or the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented in the network side device;
  • fifth auxiliary sub-information wherein the fifth auxiliary sub-information is used to indicate that the terminal expects the model update process to be implemented at the terminal, or the fifth auxiliary sub-information is used to indicate that the terminal expects the model update process to be implemented at the network side device;
  • the sixth sub-auxiliary information is used to represent that the terminal expects to change the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to activate the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to deactivate the model of the AI and/or ML;
  • the seventh sub-auxiliary information is used to characterize model information of the AI and/or ML expected or preferred by the terminal;
  • eighth sub-auxiliary information where the eighth sub-auxiliary information is used to indicate that the terminal expects to update the AI and/or ML model
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • the first auxiliary information when the first auxiliary information includes sixth auxiliary sub-information, the first auxiliary information includes:
  • Second indication information where the second indication information is used to indicate the reason why the terminal expects to change the AI and/or ML model, or the reason why the terminal expects to activate the AI and/or ML model, or the reason why the terminal expects to deactivate the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes eighth sub-auxiliary information, the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the seventh auxiliary sub-information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • the model identification information includes:
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • the auxiliary information reporting method and device provided in the embodiments of the present disclosure send first auxiliary information to a network side device through a terminal, and use the first auxiliary information to help implement the model management process of artificial intelligence AI and/or machine learning ML in a wireless communication system, thereby improving the practicality of AI and/or ML in the wireless communication system.
  • FIG1 is a flow chart of a method for reporting auxiliary information according to an embodiment of the present disclosure
  • FIG2 is a second flow chart of the auxiliary information reporting method provided in an embodiment of the present disclosure.
  • FIG3 is a schematic diagram of the structure of a network side device provided by an embodiment of the present disclosure.
  • FIG4 is a schematic diagram of the structure of a terminal provided by an embodiment of the present disclosure.
  • FIG5 is a schematic diagram of a structure of an auxiliary information reporting device provided in an embodiment of the present disclosure.
  • FIG. 6 is a second schematic diagram of the structure of the auxiliary information reporting device provided in an embodiment of the present disclosure.
  • the term "and/or” describes the association relationship of associated objects, indicating that three relationships may exist.
  • a and/or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone.
  • the character "/" generally indicates that the associated objects before and after are in an "or” relationship.
  • plurality in the embodiments of the present disclosure refers to two or more than two, and other quantifiers are similar thereto.
  • the disclosed embodiments provide an auxiliary information reporting method and apparatus to improve the practicality of artificial intelligence (AI) and/or machine learning (ML) in wireless communication systems.
  • AI artificial intelligence
  • ML machine learning
  • the method and the device are based on the same application concept. Since the method and the device solve the problem in a similar principle, the implementation of the device and the method can refer to each other, and the repeated parts will not be repeated.
  • the applicable systems can be global system of mobile communication (GSM) system, code division multiple access (CDMA) system, wideband code division multiple access (WCDMA) general packet radio service (GPRS) system, long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE 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 (NR) system, etc.
  • GSM global system of mobile communication
  • CDMA code division multiple access
  • WCDMA wideband code division multiple access
  • GPRS general packet radio service
  • LTE long term evolution
  • FDD LTE frequency division duplex
  • TDD LTE time division duplex
  • LTE-A long term evolution advanced
  • UMTS universal mobile telecommunication system
  • WiMAX worldwide interoperability for microwave access
  • NR new radio
  • EPS Evolved Packet System
  • 5GS 5G System
  • the terminal can report some auxiliary information to the network side.
  • the auxiliary information of the user equipment (UE) such as overheating will be reported to the network; for the deactivated secondary cell group (SCG), whether the UE wants the SCG to be deactivated will be reported to the network.
  • the UE provides the network side with its own situation through the auxiliary information reporting to assist the network side in making better decisions.
  • the embodiments of the present invention provide a method and device for reporting auxiliary information.
  • FIG. 1 is a flowchart of a method for reporting auxiliary information provided by an embodiment of the present disclosure. As shown in FIG. 1 , the method for reporting auxiliary information is applied to a terminal, that is, the execution subject is the terminal. The method includes:
  • Step 100 Send first auxiliary information to a network side device, where the first auxiliary information includes at least part of the information required to implement a model management process of artificial intelligence AI and/or machine learning ML.
  • the model management process of artificial intelligence AI and/or machine learning ML may include a process of optimizing the AI and/or ML model
  • model management process of artificial intelligence AI and/or machine learning ML may include any process involving AI and/or ML models, which is not limited in the embodiments of the present disclosure.
  • the terminal may send first auxiliary information to the network side device, where the first auxiliary information includes at least part of the information required to implement various processes of AI/ML life cycle management (LCM). That is, the terminal may send the first auxiliary information to the network side device, where the first auxiliary information may assist in the implementation of any one or more of the following:
  • Terminal capability management UE capability
  • the first auxiliary information can be sent to the network side through the terminal.
  • the first auxiliary information may include data for driving or triggering relevant functions of AI and/or ML, or information or instructions for driving or triggering the execution of relevant functions of AI and/or ML.
  • the sending of the first assistance information may be reported via a UEAssistanceInformation message
  • a Radio Resource Control (RRC) signaling may be introduced, which is dedicated to carrying the first auxiliary information.
  • RRC Radio Resource Control
  • the network side device may make a decision based on the first auxiliary information, obtain decision result information, and use the decision result information to indicate whether the expectations of the terminal are met; and send the decision result information to the terminal.
  • the network side device may decide whether to execute the relevant process corresponding to the first auxiliary information in combination with some other situations or information, and send decision result information to the terminal based on the decision result.
  • the process of the network-side device making a decision based on the first auxiliary information depends on the network implementation, which is not limited in the embodiments of the present disclosure.
  • the auxiliary information reporting method sents first auxiliary information to a network side device through a terminal, and uses the first auxiliary information to help implement the model management process of artificial intelligence AI and/or machine learning ML in a wireless communication system, thereby improving the practicality of AI and/or ML in the wireless communication system.
  • the method before sending the first auxiliary information to the network side device, the method further includes:
  • First configuration information sent by a network side device is received, where the first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • the network side device may first send first configuration information to the terminal, where the first configuration information is used to configure reporting of the first auxiliary information.
  • the terminal may send the first auxiliary information to the network side device based on the first configuration information.
  • the terminal reporting the auxiliary information does not necessarily have to be based on the network configuration, and may be based on the network configuration or not;
  • the terminal may send first auxiliary information to the network side device based on the first configuration information when receiving the first configuration information;
  • the terminal may also, if not receiving the first configuration information, information, and sends the first auxiliary information to the network side device.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the network side device may configure the first indication information and/or the first timer information (ProhibitTimer value) for the terminal;
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information of the data that the UE wishes to report the terminal to collect
  • the first timer information may control the terminal to report the auxiliary information of the data that the terminal wishes to report the terminal to collect for a maximum of a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information of the data that the terminal wishes to report the terminal to collect only once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report that the terminal expects the network side device to send data collected on the network side to the terminal, which is auxiliary information for the terminal to perform model training/model update, etc.
  • the first timer information controls the terminal to report at most a preset number of times before the timer expires that the terminal expects the network side device to send data collected on the network side to the terminal, which is auxiliary information for the terminal to perform model training/model update, etc.
  • the terminal is controlled to report at most once before the timer expires that the terminal expects the network side device to send data collected on the network side to the terminal, which is auxiliary information for the terminal to perform model training/model update, etc.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal hopes the model will be trained on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model will be trained on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model will be trained on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side
  • the first timer information may control the terminal to report only a preset number of auxiliary information that the terminal hopes the model to monitor on the terminal side/network side before the timer expires, such as controlling the terminal to report at most a preset number of auxiliary information that the terminal hopes the model to monitor on the terminal side/network side before the timer expires.
  • the terminal hopes that the model will monitor auxiliary information on the terminal side/network side by reporting only once.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal desires model change/switch/activation/deactivation
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal desires model change/switch/activation/deactivation at most a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information that the terminal desires model change/switch/activation/deactivation at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information of model recommendation/preference
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report only the auxiliary information of the model recommendation/preference for a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information of the model recommendation/preference only once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal hopes to update the model
  • the first timer information may control the terminal to report the auxiliary information that the terminal hopes to update the model at most a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information that the terminal hopes to update the model at most once before the timer expires.
  • the first indication information may explicitly or implicitly indicate that the terminal may/needs to report the auxiliary information that the terminal wishes to report based on AI and/or ML (AI/ML-based) or non-AI/ML-based
  • the first timer information ProhibitTimer value
  • Report multiple preset secondary terminals' auxiliary information that they want in AI/ML-based mode or non-AI/ML-based mode for example, control the terminal to report the auxiliary information that the terminal wants in AI/ML-based mode or non-AI/ML-based mode only once before the timer times out.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information of the AI/ML model that the terminal wishes to release
  • the first timer information may control the terminal to report the auxiliary information of the AI/ML model that the terminal wishes to release at most a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information of the AI/ML model that the terminal wishes to release at most once before the timer expires.
  • the network side device may configure one or more of the following information for the UE:
  • an indication (explicit or implicit) and/or a ProhibitTimer value, which is used to indicate that the UE can/needs to report auxiliary information of the data that the UE wishes to report to the UE Data collection;
  • This indication information is used to indicate that the UE can/needs to report that the UE wants the network side device to send the data after Data collection to the UE, which is auxiliary information for the UE to perform model training/model update, etc.;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants the model to be trained on the UE side/network side;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants the model to be updated on the UE side/network side;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report auxiliary information that the UE wants to change/switch/activate/deactivate the model;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report auxiliary information of the model recommendation/preference;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants the model to be updated;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE desires in an AI/ML-based manner or a non-AI/ML-based manner;
  • an indication information (explicit indication or implicit indication) and/or a ProhibitTimer value, where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants to release the AI/ML model.
  • the network side device may configure a ProhibitTimer for the terminal to prevent the UE from frequently reporting the first auxiliary information.
  • the sending the first auxiliary information to the network side device includes one or more of the following:
  • the first auxiliary information is sent to the network side device.
  • the terminal may perform at least one of the following:
  • the first auxiliary information is sent.
  • the method before sending the first auxiliary information to the network side device, the method further includes one or more of the following:
  • the terminal may perform at least one of the following:
  • the terminal may perform at least one of the following:
  • the sending terminal hopes that the network side device will send the data collected by the network side to the terminal, which is used as auxiliary information for the terminal to perform model training/model update, etc.
  • the terminal may perform at least one of the following when determining that the terminal wants the model to be trained on the terminal side/network side:
  • the terminal may perform at least one of the following when determining that the terminal wants the model to be monitored on the terminal side/network side:
  • the sending terminal wishes the model to monitor auxiliary information on the terminal side/network side.
  • the terminal may perform at least one of the following when determining that the terminal wants the model to be updated on the terminal side/network side:
  • the sending terminal wishes the model to update auxiliary information on the terminal side/network side.
  • the terminal may perform at least one of the following:
  • Reports cause values which can indicate, for example, poor model performance or model inapplicability.
  • the terminal may perform at least one of the following when determining that the terminal wants to report a model recommendation/preference:
  • the terminal may perform at least one of the following when determining that the terminal wants to update the model:
  • Reports cause values which can indicate, for example, poor model performance, poor model generalization, or model unsuitability.
  • the terminal may determine whether the terminal desires an AI/ML-based approach or a non-AI/ML-based approach. In this case, do at least one of the following:
  • Reports the cause value which can indicate, for example, poor AI/ML model performance, good AI/ML model performance, too much overhead, or overheating.
  • the terminal may perform at least one of the following when determining that the terminal wants to release the AI/ML model:
  • the sending terminal wishes to release auxiliary information of the AI/ML model
  • the cause value for example, it can indicate that the UE has insufficient memory or the overhead is too large.
  • the terminal may perform at least one of the following when determining that the network configures the UE to report the first auxiliary information:
  • the terminal may perform at least one of the following when determining that the first timer is not running:
  • the network side device can combine some other situations or information, such as depending on the network implementation, to decide whether to request the UE to report the data in the UE Data collection, whether to send the data set in the Data collection to the UE for UE model training/update, whether to instruct the UE to change/switch/activate/deactivate the model, etc., and then decide whether to send corresponding signaling to the UE based on the network decision result.
  • the first auxiliary information includes one or more of the following:
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the terminal side, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the network side device;
  • the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the terminal side, or the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the network side device;
  • the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the terminal side, or the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the network side device;
  • the sixth sub-auxiliary information is used to represent that the terminal expects to change the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to activate the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to deactivate the model of the AI and/or ML;
  • the seventh sub-auxiliary information is used to characterize model information of the AI and/or ML expected or preferred by the terminal;
  • eighth sub-auxiliary information where the eighth sub-auxiliary information is used to indicate that the terminal expects to update the AI and/or ML model
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • each sub-assistance information may correspond to its own first timer information ProhibitTimer, which is configured separately by the network side device, or determined based on default, or predefined by the protocol. Either pre-set or set in real time.
  • multiple sub-assistance information may correspond to a unified first timer information ProhibitTimer, which may be configured separately or uniformly by the network side device, or determined based on a default, or predefined by a protocol, or preset, or set in real time.
  • ProhibitTimer may be configured separately or uniformly by the network side device, or determined based on a default, or predefined by a protocol, or preset, or set in real time.
  • the data collected by the terminal may include data from the UE Data collection.
  • the data collected by the network side device may include the data after the network side Data collection.
  • the AI and/or ML model information expected or preferred by the terminal may include model recommendation/preference.
  • the UE reports data that the UE wishes to report UE Data collection, and the UE is collecting data for AI/ML Data collection.
  • the auxiliary information reporting process may include:
  • Step 1 The network side device sends first configuration information to the UE.
  • the first configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to report the collected data for the Channel State Information (CSI) compression sub-use case, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • CSI Channel State Information
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to report the collected data, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted, that is, here is the CSI compression sub use case; the former indication information may display the configuration or may be implicitly expressed, and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to report the collected data for the CSI compression sub-use case to the network side device, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the first sub-assistance information indicates that the UE wishes to report the collected data for the CSI compression sub-use case to the network-side device;
  • the first sub-auxiliary information indicates that the UE wishes to report the collected data to the network device + the sixth indication information indicates which use case/sub-use case the collected data is for, i.e., the CSI compression sub-use case in this case;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 After the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to request the UE to report the collected data for the CSI compression sub use case based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to request the UE to report the collected data for the CSI compression sub use case, the network side device sends a signaling to the UE to make the request;
  • Step 5 The UE reports the data collected by the UE for the CSI compression sub-use case to the network-side device according to the request of the network-side device;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and auxiliary information of the network side device configuration
  • the ProhibitTimer value is for the whole, regardless of use case/sub use case.
  • the UE can send a whole auxiliary information to indicate that the UE now has collected data to report, regardless of which use case/sub use case the data is collected for.
  • the UE reports that the UE wants the network side device to send the data after the Data collection to the UE.
  • the UE needs the data set to complete the model training/model update and other tasks.
  • the CSI compression sub-use case is taken as an example, that is, the UE needs the data set to complete the CSI compression sub-use case model training/update and other tasks.
  • the auxiliary information reporting process may include:
  • Step 1 The network side device sends first configuration information to the UE.
  • the first configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes the network side device to send the data after the Data collection for the CSI compression sub use case to the UE, so as to be used for the UE to train/update the CSI compression sub use case model, the UE can report the auxiliary information to notify the network side device;
  • this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the network side device to send the data after Data collection to the UE for the UE to perform model training/update and other tasks, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted, that is, here is the CSI compression sub use case; the former indication information may display the configuration or may be implicitly expressed, and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants the network side device to send the data after the data collection for the CSI compression sub use case to the UE for the UE to train/update the CSI compression sub use case model, if the ProhibitTimer value is configured, the UE starts the ProhibitTimer, such as If the ProhibitTimer does not exist, the opening of the ProhibitTimer is skipped; and the auxiliary information is initialized to be reported to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the auxiliary information may include at least one of the following:
  • the second auxiliary sub-information indicates that the UE wishes the network-side device to send the data after the data collection for the CSI compression sub-use case to the UE, so as to enable the UE to train/update the CSI compression sub-use case model;
  • the second auxiliary sub-information indicates that the UE wishes the network-side device to send the data after the Data collection to the UE for the UE to perform model training/update and other tasks + the sixth indication information indicates which use case/sub use case this Data collection data is for (i.e., the CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 After the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to send the data of the Data collection for the CSI compression sub use case to the UE based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to send the data of the Data collection for the CSI compression sub use case to the UE, the network side device sends the data of the Data collection for the CSI compression sub use case to the UE;
  • Step 5 The UE receives the data of the Data collection for the CSI compression sub-use case sent by the network side device, and uses the received data to perform subsequent model training/update work for CSI compression;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is The indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE When the UE wants the network side device to send the data after Data collection for a certain/certain use cases or a certain/certain sub use cases to the UE for the UE to train/update a certain/certain use case or a certain/certain sub use case model, the UE can send a whole auxiliary information to indicate that the UE wants the network side device to send the data after Data collection to the UE for the UE to train/update the model, which is not for a specific use case/sub use case.
  • the UE reports that the UE wants the model to be trained on the UE side/network side; taking the CSI compression sub use case as an example, the auxiliary information reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to train the model for the CSI compression sub use case on the UE side/network side, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be trained on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same way as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to train the model for the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE also initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling may be introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the third sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be trained on the UE side/network side.
  • This indication information can be an enumeration type, i.e., ⁇ UE side, network side ⁇ ;
  • the third sub-auxiliary information indicates that the UE wants the model to be trained on the UE side/network side.
  • This indication information may be an enumeration type, i.e., ⁇ UE side, network side ⁇ + sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • the third sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be trained on the UE side/network side.
  • This indication information is two separate indication information, one indication information indicates that the UE wants the model to be trained on the UE side, and the other indication information indicates that the UE wants the model to be trained on the network side;
  • the third sub-auxiliary information indicates that the UE wants the model to be trained on the UE side/network side.
  • This indication information is two separate indication information, one indication information indicates that the UE wants the model to be trained on the UE side, and the other indication information indicates that the UE wants the model to be trained on the network side + the sixth indication information indicates which use case/sub use case is targeted (i.e., here is the CSI compression sub use case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network receives the first auxiliary information reported by the UE, the network-side device, in combination with some other situations or information, can decide whether the model training for the CSI compression sub-use case is performed on the UE side or the network side, depending on the network-side device implementation;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model training is carried out on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc.
  • Sub use cases that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE wants the model to be trained on the UE side/network side, not for a specific use case/sub use case.
  • the UE reports the UE's desired model monitoring on the UE side/network side; taking the CSI compression sub use case as an example, the following steps are included:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to monitor the model of the CSI compression sub use case on the UE side/network side, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be monitored on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to monitor the model of the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the fourth sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be monitored on the UE side/network side.
  • This indication information can be an enumeration type, i.e., ⁇ UE side, network side ⁇
  • the fourth sub-auxiliary information indicates that the UE wants the model to be monitored on the UE side/network side.
  • This indication information may be of an enumeration type, i.e., ⁇ UE side, network side ⁇ + sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be monitored on the UE side, and the other indication information indicating that the UE wants the model to be monitored on the network side;
  • This indication information is two separate indication information, one indication information indicating that the UE wishes the model to be monitored on the UE side, and the other indication information indicating that the UE wishes the model to be monitored on the network side + sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device, in combination with some other situations or information, may decide whether the model monitoring for the CSI compression sub use case is on the UE side or the network side, depending on the implementation of the network side device;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model monitoring is on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc.
  • Sub use cases that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE wants the model to be monitored on the UE side/network side, not for a specific use case/sub use case.
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to update the model for the CSI compression sub use case on the UE side/network side, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be updated on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case here); the former indication information may display the configuration or may be implicitly expressed (in the same way as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to update the model for the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the fifth auxiliary information indicates that the UE expects the model for the CSI compression sub use case.
  • this indication information can be an enumeration type, i.e. ⁇ UE side, network side ⁇
  • the fifth sub-auxiliary information indicates that the UE wants the model to be updated on the UE side/network side.
  • This indication information may be of an enumeration type, i.e., ⁇ UE side, network side ⁇ + the sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be updated on the UE side, and the other indication information indicating that the UE wants the model to be updated on the network side;
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be updated on the UE side, and the other indication information indicating that the UE wants the model to be updated on the network side + sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device, in combination with some other situations or information, may decide whether to update the model for the CSI compression sub use case on the UE side or the network side, depending on the implementation of the network side device;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model is updated on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is The indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE can send a whole auxiliary information to indicate that the UE wants the model to be updated on the UE side/network side, not for a specific use case/sub use case.
  • the first auxiliary information when the first auxiliary information includes sixth sub-auxiliary information, the first auxiliary information includes:
  • Second indication information where the second indication information is used to indicate the reason why the terminal expects to change the AI and/or ML model, or the reason why the terminal expects to activate the AI and/or ML model, or the reason why the terminal expects to deactivate the AI and/or ML model.
  • the UE reports that the UE wants to change/switch/activate/deactivate the model.
  • the reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wants to change/switch/activate/deactivate the model, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to change/switch/activate/deactivate the model, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 First timer information, such as a ProhibitTimer value.
  • ProhibitTimer is used to prevent the UE from frequently reporting this auxiliary information
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to change/switch/activate/deactivate the AI/ML model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the sixth sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants to change/switch/activate/deactivate the AI/ML model; this indication information is a whole 1-bit indication information;
  • the sixth sub-auxiliary information indicates that the UE wishes to change/switch/activate/deactivate the AI/ML model + the sixth indication information indicates which use case/sub use case is targeted (i.e., the CSI compression sub use case in this case);
  • second indication information such as a cause value, which may indicate that the model performance is poor or the model is not applicable
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to change/switch/activate/deactivate the AI/ML model based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to change/switch/activate/deactivate the AI/ML model, the network side device sends a command to the UE to instruct the UE to change/switch/activate/deactivate the AI/ML model;
  • Step 5 The UE changes/switches/activates/deactivates the AI/ML model according to the instructions of the network-side device;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases
  • sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the UE reports that the UE wants a model update.
  • the UE currently stores an AI/ML model for CSI compression.
  • the reporting process includes the following steps:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • Configuration information 1 first indication information, used to indicate that for the CSI compression sub-use case, when the UE wishes to update the model, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated. For example, when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate that it supports reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to update the model, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to update the model, if the ProhibitTimer value is configured, the UE starts the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips starting the ProhibitTimer; And initialize reporting of auxiliary information to the network side device, the reporting signaling may be a first auxiliary information message or introduce an RRC signaling to report the first auxiliary information message, and the first auxiliary information may include at least one of the following:
  • the eighth auxiliary information indicates that the UE wishes to update the model for the CSI compression sub-use case
  • the eighth sub-auxiliary information indicating that the UE wishes to update the model + the sixth indication information, indicating which use case/sub-use case is targeted (i.e., here is the CSI compression sub-use case);
  • third indication information such as a cause value, which may indicate, for example, that the model performance is poor, the model generalization ability is poor, or the model is not applicable;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE, and the network side device combines some other situations or information, and may decide whether to update the model depending on the implementation of the network side device;
  • Step 4 If the network side device decides to update the model, the network side device sends an instruction to the UE to instruct the UE to update the model;
  • Step 5 The UE updates the model according to the instructions of the network side device.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE wants to update the model of a certain/certain use case or a certain/certain sub use case, the UE can send an overall auxiliary information to indicate that the UE wants the model to be updated, not for a specific use case/sub use case.
  • the first Auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the UE reports whether the UE desires an AI/ML-based method or a non-AI/ML-based method.
  • the reporting process may include at least one of the following:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wishes to use the AI/ML-based method/non-AI/ML-based method, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to use the AI/ML-based method/non-AI/ML-based method, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to use the AI/ML-based method/non-AI/ML-based method, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the ninth auxiliary information indicates that the UE wishes to use the CSI compression sub use case.
  • This indication information can be an enumeration type, such as ⁇ AI/ML-based method, non-AI/ML-based method ⁇
  • the ninth sub-auxiliary information indicates that the UE wishes to use the AI/ML-based method/non-AI/ML-based method.
  • This indication information may be an enumeration type, such as ⁇ AI/ML-based method, non-AI/ML-based method ⁇ +the sixth indication information, indicating which use case/sub use case is targeted (i.e., here is the CSI compression sub use case);
  • the ninth sub-auxiliary information indicates that for the CSI compression sub use case, the UE wishes to use the AI/ML-based method or the non-AI/ML-based method.
  • This indication information is two separate indication information, one indicating information indicating that the UE wishes to use the AI/ML-based method, and the other indicating information indicating that the UE wishes to use the non-AI/ML-based method;
  • This indication information is two separate indication information, one indication information indicating that the UE wishes to use the AI/ML-based method, and the other indication information indicating that the UE wishes to use the non-AI/ML-based method + the sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • Fourth indication information such as a cause value, which may indicate, for example, poor AI/ML model performance, good AI/ML model performance, too much overhead, or overheating.
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE.
  • the network side device combines some other situations or information and may decide whether to use the AI/ML-based method or the non-AI/ML-based method depending on the implementation of the network side device.
  • Step 4 The network-side device notifies the UE of the decision result (i.e., whether to use the AI/ML-based method or the non-AI/ML-based method);
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, i.e.
  • the process can also be applied to use cases such as CSI feedback enhancement, beam management, and positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, and direct positioning, that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network-side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE can send an overall auxiliary information to indicate that the UE wants to use the AI/ML-based method or the non-AI/ML-based method, and it is not for a specific use case/sub use case.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the UE reports that the UE wants the model to be released.
  • the UE currently stores an AI/ML model for CSI compression.
  • the reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE hopes to release the model, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to release the model, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to release the model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the tenth sub-assistance information indicates that the UE wishes to release the model for the CSI compression sub-use case
  • the tenth sub-auxiliary information indicates that the UE wishes to release the model + the sixth indication information indicates which use case/sub-use case is targeted (i.e., the CSI compression sub-use case in this case);
  • Fifth indication information such as a cause value, which may indicate that the UE has insufficient memory or the overhead is too large.
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE, and the network side device combines some other situations or information, and may decide whether to release the model depending on the implementation of the network side device;
  • Step 4 If the network side device decides to release the model, the network side device sends an indication to the UE to notify the UE to release the model;
  • Step 5 The UE releases the model according to the instructions of the network side device.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE wants the model to be released, not for a specific use case/sub use case.
  • the seventh auxiliary information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • model information of the AI and/or ML expected or preferred by the terminal may include model identification information of the AI and/or ML model expected or preferred by the terminal;
  • model information of the AI and/or ML expected or preferred by the terminal may include model priority information of the AI or ML expected or preferred by the terminal;
  • the model priority indicated by the seventh auxiliary sub-information can be embodied by model identification information.
  • Other methods are also possible and are not limited here.
  • model identification information includes:
  • the model identification information can be: Model id, functionality id, configuration id, model parameters, model input/output, etc.
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wishes to report a model recommendation/preference, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to report a model recommendation/preference, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to change the AI/ML model from a model with a model ID of 1 to a model with a model ID of 3, or the UE wishes to notify the network side of the priority information of the AI/ML model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device, the reporting signaling may be a first auxiliary information message or introduce an RRC signaling to report the first auxiliary information message, and the first auxiliary information may include at least one of the following:
  • auxiliary sub-information such as a model priority information, indicating the model priority information desired by the UE for the CSI compression sub-use case.
  • Eighth sub-auxiliary information for example, including a model priority information and a sixth indication information, indicating the model priority information desired by the UE.
  • model identification information can be functionality id, configuration id, model parameters, model input/output, etc. in addition to the Model id here;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the auxiliary information sent by the UE.
  • the network side device combines some other situations or information and may decide whether to change the model, etc., depending on the network side device implementation;
  • Step 4 If the network side device decides to change the model, the network side device sends a command to the UE to notify the UE to change the model;
  • Step 5 The UE changes the model according to the instructions of the network side device
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases
  • sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • the above process can also be triggered before the AI/ML model is used on the UE side.
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • the first auxiliary information is applicable to any use case or sub use case of AI/ML, and may be per use case or per sub use case, or as a whole; for the case of per use case or per sub use case, the UE may report second indication information to indicate which use case or sub use case the UE auxiliary information is for;
  • the reporting process of the first auxiliary information in each embodiment of the present disclosure is applicable to any AI/ML use case/sub use case, that is, it can also be applicable to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as space-frequency domain CSI compression, CSI prediction, time domain beam prediction, space domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE When the UE wishes to report the first auxiliary information, no matter which use case/which use case or which sub use case/which sub use case collects data, the UE can send an overall auxiliary information to indicate that the UE now has collected data and wishes to report it, and it is not for a specific use case/sub use case.
  • the auxiliary information reporting method sents first auxiliary information to a network side device through a terminal, and uses the first auxiliary information to help implement the model management process of artificial intelligence AI and/or machine learning ML in a wireless communication system, thereby improving the practicality of AI and/or ML in the wireless communication system.
  • FIG2 is a flow chart of an auxiliary information reporting method provided by an embodiment of the present disclosure. As shown in FIG2 , the auxiliary information reporting method is applied to a network side device, that is, the execution subject is a network side device; the method includes:
  • Step 200 receiving first auxiliary information sent by a terminal, where the first auxiliary information includes at least part of the information required to implement a model management process of artificial intelligence AI and/or machine learning ML.
  • the model management process of artificial intelligence AI and/or machine learning ML may include a process of adjusting the model state of AI and/or ML;
  • the model management process of artificial intelligence AI and/or machine learning ML may include a process of optimizing the AI and/or ML model
  • model management process of artificial intelligence AI and/or machine learning ML may include AI and/or ML model application process
  • model management process of artificial intelligence AI and/or machine learning ML may include any process involving AI and/or ML models, which is not limited in the embodiments of the present disclosure.
  • model management process of artificial intelligence AI and/or machine learning ML includes one or more of the following LCM processes:
  • Terminal capability management UE capability
  • the terminal may send first auxiliary information to the network side device, where the first auxiliary information includes at least part of the information required to implement various processes of AI/ML life cycle management (LCM). That is, the terminal may send the first auxiliary information to the network side device, where the first auxiliary information may assist in the implementation of any one or more of the following:
  • Terminal capability management UE capability
  • the former is driven by data, while the latter is driven by events or time, in order to implement AI and/or ML in a wireless communication system
  • the related functions of ML can receive first auxiliary information sent by the terminal through the network side, and the first auxiliary information may include data for driving or triggering the related functions of AI and/or ML, or information or instructions for driving or triggering the execution of the related functions of AI and/or ML.
  • the sending of the first assistance information may be reported via a UEAssistanceInformation message
  • an RRC signaling may be introduced to be dedicated to carrying the first auxiliary information.
  • the network side device may make a decision based on the first auxiliary information, obtain decision result information, and use the decision result information to indicate whether the expectations of the terminal are met; and send the decision result information to the terminal.
  • the network side device may decide whether to execute the relevant process corresponding to the first auxiliary information in combination with some other situations or information, and send decision result information to the terminal based on the decision result.
  • the process of the network-side device making a decision based on the first auxiliary information depends on the network implementation, which is not limited in the embodiments of the present disclosure.
  • the auxiliary information reporting method provided by the embodiment of the present disclosure sends a first Auxiliary information:
  • the first auxiliary information helps the model management process of artificial intelligence AI and/or machine learning ML to be implemented in the wireless communication system, thereby improving the practicality of AI and/or ML in the wireless communication system.
  • the method before receiving the first auxiliary information sent by the terminal, the method further includes:
  • first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • the network side device may first send first configuration information to the terminal, where the first configuration information is used to configure reporting of the first auxiliary information.
  • the terminal may send the first auxiliary information to the network side device based on the first configuration information.
  • the terminal reporting the auxiliary information does not necessarily have to be based on the network configuration, and may be based on the network configuration or not;
  • the terminal may send first auxiliary information to the network side device based on the first configuration information when receiving the first configuration information;
  • the terminal may also send the first auxiliary information to the network side device based on the first configuration information without receiving the first configuration information.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the network side device may configure the first indication information and/or the first timer information (ProhibitTimer value) for the terminal;
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information of the data that the UE wishes to report the terminal to collect
  • the first timer information may control the terminal to report the auxiliary information of the data that the terminal wishes to report the terminal to collect for a maximum of a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information of the data that the terminal wishes to report the terminal to collect only once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report that the terminal hopes that the network side device sends the data collected by the network side to the terminal for the terminal to perform model training/ Auxiliary information of model update, etc.
  • the first timer information controls the terminal to report at most a preset number of times before the timer expires, and the terminal hopes that the network side device will send the data collected by the network side to the terminal, which is used for the terminal to perform model training/model update, etc.
  • the terminal is controlled to report at most once before the timer expires, and the terminal hopes that the network side device will send the data collected by the network side to the terminal, which is used for the terminal to perform model training/model update, etc.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal hopes the model will be trained on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model will be trained on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model will be trained on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model to monitor on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side at most a preset number of times before the timer expires.
  • the terminal is controlled to report the auxiliary information that the terminal hopes the model will be updated on the terminal side/network side at most once before the timer expires.
  • the first indication information may explicitly or implicitly indicate that the terminal may/needs to report auxiliary information that the terminal wishes to change/switch/activate/deactivate the model
  • the first timer information ProhibitTimer value
  • the control terminal may control the terminal to report only a preset number of times before the timer expires.
  • Auxiliary information of the desired model change/switch/activation/deactivation for example, the control terminal only reports the auxiliary information of the desired model change/switch/activation/deactivation once at most before the timer times out.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information of model recommendation/preference
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report only the auxiliary information of model recommendation/preference for a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information of model recommendation/preference only once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal hopes to update the model
  • the first timer information may control the terminal to report the auxiliary information that the terminal hopes to update the model at most a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information that the terminal hopes to update the model at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report auxiliary information that the terminal desires in an AI/ML-based manner or a non-AI/ML-based manner
  • the first timer information ProhibitTimer value
  • the terminal may control the terminal to report auxiliary information that the terminal desires in an AI/ML-based manner or a non-AI/ML-based manner at most for a preset number of times before the timer expires, for example, controlling the terminal to report auxiliary information that the terminal desires in an AI/ML-based manner or a non-AI/ML-based manner at most once before the timer expires.
  • the first indication information may explicitly indicate or implicitly indicate that the terminal can/needs to report the auxiliary information that the terminal wishes to release the AI/ML model
  • the first timer information may control the terminal to report the auxiliary information that the terminal wishes to release the AI/ML model at most for a preset number of times before the timer expires, for example, controlling the terminal to report the auxiliary information that the terminal wishes to release the AI/ML model at most once before the timer expires.
  • the network side device may configure one or more of the following information for the UE:
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report auxiliary information of the data that the UE wishes to report in the UE Data collection;
  • this indication information is used to indicate that the UE can/needs to report that the UE wants the network side device to send data after Data collection to the UE, which is auxiliary information for the UE to perform model training/model update, etc.;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report auxiliary information that the UE wants the model to be trained on the UE side/network side;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE hopes the model to monitor on the UE side/network side;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants the model to be updated on the UE side/network side;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report auxiliary information that the UE wants to change/switch/activate/deactivate the model;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report auxiliary information of the model recommendation/preference;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants the model to be updated;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value), where the indication information is used to indicate that the UE can/needs to report the auxiliary information in the AI/ML-based mode or the non-AI/ML-based mode that the UE desires;
  • a first indication information (explicit indication or implicit indication) and/or a first timer information (ProhibitTimer value) is configured, and the indication information is used to indicate that the UE can/needs to report the auxiliary information that the UE wants to release the AI/ML model.
  • the network side device can configure a ProhibitTimer for the terminal to prevent the UE from frequently logging on. Report the first auxiliary information.
  • the first auxiliary information includes one or more of the following:
  • first sub-assistance information where the first sub-assistance information is used to represent that the terminal expects to report data collected by the terminal
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the terminal side, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the network side device;
  • the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the terminal side, or the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the network side device;
  • the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the terminal side, or the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the network side device;
  • the sixth sub-auxiliary information is used to represent that the terminal expects to change the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to activate the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to deactivate the model of the AI and/or ML;
  • the seventh sub-auxiliary information is used to characterize model information of the AI and/or ML expected or preferred by the terminal;
  • eighth sub-auxiliary information where the eighth sub-auxiliary information is used to indicate that the terminal expects to update the AI and/or ML model
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • each sub-assistance information may respectively correspond to its own first timer information ProhibitTimer, which is separately configured by the network side device, or determined based on default, or predefined by the protocol, or preset, or set in real time.
  • ProhibitTimer which is separately configured by the network side device, or determined based on default, or predefined by the protocol, or preset, or set in real time.
  • multiple sub-assistance information may correspond to a unified first timer information ProhibitTimer, which may be configured separately or uniformly by the network side device, or determined based on a default, or predefined by a protocol, or preset, or set in real time.
  • ProhibitTimer may be configured separately or uniformly by the network side device, or determined based on a default, or predefined by a protocol, or preset, or set in real time.
  • the data collected by the terminal may include data from the UE Data collection.
  • the data collected by the network side device may include the data after the network side Data collection.
  • the AI and/or ML model information expected or preferred by the terminal may include model recommendation/preference.
  • the UE reports data that the UE wishes to report UE Data collection, and the UE is collecting data for AI/ML Data collection.
  • the auxiliary information reporting process may include:
  • Step 1 The network side device sends first configuration information to the UE.
  • the first configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to report the collected data for the Channel State Information (CSI) compression sub-use case, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • CSI Channel State Information
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to report the collected data, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted, that is, here is the CSI compression sub use case; the former indication information may display the configuration or may be implicitly expressed, and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to report the collected data for the CSI compression sub-use case to the network side device, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the first sub-assistance information indicates that the UE wishes to report the collected data for the CSI compression sub-use case to the network-side device;
  • the first sub-auxiliary information indicates that the UE wishes to report the collected data to the network device + the sixth indication information indicates which use case/sub-use case the collected data is for, i.e., the CSI compression sub-use case in this case;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 After the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to request the UE to report the collected data for the CSI compression sub use case based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to request the UE to report the collected data for the CSI compression sub use case, the network side device sends a signaling to the UE to make the request;
  • Step 5 The UE reports the data collected by the UE for the CSI compression sub-use case to the network-side device according to the request of the network-side device;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this
  • the configuration information and auxiliary information are per use case or per sub use case.
  • the configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE When the UE wants to report the collected data, no matter which use case/sub use case the data is collected for, the UE can send an overall auxiliary information to indicate that the UE now has collected data and wants to report it, not for a specific use case/sub use case.
  • the UE reports that the UE wants the network side device to send the data after the Data collection to the UE.
  • the UE needs the data set to complete the model training/model update and other tasks.
  • the CSI compression sub-use case is taken as an example, that is, the UE needs the data set to complete the CSI compression sub-use case model training/update and other tasks.
  • the auxiliary information reporting process may include:
  • Step 1 The network side device sends first configuration information to the UE.
  • the first configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes the network side device to send the data after the Data collection for the CSI compression sub use case to the UE, so as to be used for the UE to train/update the CSI compression sub use case model, the UE can report the auxiliary information to notify the network side device;
  • this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the network side device to send the data after Data collection to the UE for the UE to perform model training/update and other tasks, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted, that is, here is the CSI compression sub use case; the former indication information may display the configuration or may be implicitly expressed, and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants the network device to send data for the CSI compression sub use case The collected data is sent to the UE for training/updating the CSI compression sub use case model. If the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer. The UE also initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling may be introduced to report the first auxiliary information message.
  • the auxiliary information may include at least one of the following:
  • the second auxiliary sub-information indicates that the UE wishes the network-side device to send the data after the data collection for the CSI compression sub-use case to the UE, so as to enable the UE to train/update the CSI compression sub-use case model;
  • the second auxiliary sub-information indicates that the UE wishes the network-side device to send the data after the Data collection to the UE for the UE to perform model training/update and other tasks + the sixth indication information indicates which use case/sub use case this Data collection data is for (i.e., the CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 After the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to send the data of the Data collection for the CSI compression sub use case to the UE based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to send the data of the Data collection for the CSI compression sub use case to the UE, the network side device sends the data of the Data collection for the CSI compression sub use case to the UE;
  • Step 5 The UE receives the data of the Data collection for the CSI compression sub-use case sent by the network side device, and uses the received data to perform subsequent model training/update work for CSI compression;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE-assisted positioning, etc.
  • Position, direct positioning and other sub use cases that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network-side device are for the whole, without distinguishing between use case/sub use case.
  • the UE When the UE hopes that the network-side device will send the data after Data collection for a certain/certain use case or a certain/certain sub use case to the UE, so that the UE can perform certain/certain use case or certain/certain sub use case model training/update, etc., the UE can send an overall auxiliary information to indicate that the UE hopes that the network-side device will send the data after Data collection to the UE, so that the UE can perform model training/update, etc., and it is not for a special use case/sub use case.
  • the UE reports that the UE wants the model to be trained on the UE side/network side; taking the CSI compression sub use case as an example, the auxiliary information reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to train the model for the CSI compression sub use case on the UE side/network side, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be trained on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same way as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to train the model for the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message, and the first auxiliary information may include at least one of the following:
  • the third sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be trained on the UE side/network side.
  • This indication information can be an enumeration type, i.e., ⁇ UE side, network side ⁇ ;
  • the third sub-auxiliary information indicates that the UE wants the model to be trained on the UE side/network side.
  • This indication information may be an enumeration type, i.e., ⁇ UE side, network side ⁇ + sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • the third sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be trained on the UE side/network side.
  • This indication information is two separate indication information, one indication information indicates that the UE wants the model to be trained on the UE side, and the other indication information indicates that the UE wants the model to be trained on the network side;
  • the third sub-auxiliary information indicates that the UE wants the model to be trained on the UE side/network side.
  • This indication information is two separate indication information, one indication information indicates that the UE wants the model to be trained on the UE side, and the other indication information indicates that the UE wants the model to be trained on the network side + the sixth indication information indicates which use case/sub use case is targeted (i.e., here is the CSI compression sub use case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network receives the first auxiliary information reported by the UE, the network-side device, in combination with some other situations or information, can decide whether the model training for the CSI compression sub-use case is performed on the UE side or the network side, depending on the network-side device implementation;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model training is carried out on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to CSI feedback enhancement, beam management, positioning accuracy enhancement, etc.
  • Use case, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning and other sub use cases, that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE hopes the model training is on the UE side/network side, not for a specific use case/sub use case.
  • the UE reports the UE's desired model monitoring on the UE side/network side; taking the CSI compression sub use case as an example, the following steps are included:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to monitor the model of the CSI compression sub use case on the UE side/network side, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be monitored on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to monitor the model of the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message, and the first auxiliary information may include at least one of the following:
  • the fourth sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be monitored on the UE side/network side.
  • This indication information can be an enumeration type, i.e., ⁇ UE side, network side ⁇
  • the fourth sub-auxiliary information indicates that the UE wants the model to be monitored on the UE side/network side.
  • This indication information may be of an enumeration type, i.e., ⁇ UE side, network side ⁇ + sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be monitored on the UE side, and the other indication information indicating that the UE wants the model to be monitored on the network side;
  • This indication information is two separate indication information, one indication information indicating that the UE wishes the model to be monitored on the UE side, and the other indication information indicating that the UE wishes the model to be monitored on the network side + sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device, in combination with some other situations or information, may decide whether the model monitoring for the CSI compression sub use case is on the UE side or the network side, depending on the implementation of the network side device;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model monitoring is on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to CSI feedback enhancement, beam management, positioning accuracy enhancement, etc.
  • Use case, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning and other sub use cases, that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE wants the model to be monitored on the UE side/network side, not for a specific use case/sub use case.
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that when the UE wishes to update the model for the CSI compression sub use case on the UE side/network side, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wants the model to be updated on the UE side/network side, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case here); the former indication information may display the configuration or may be implicitly expressed (in the same way as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to update the model for the CSI compression sub use case on the UE side/network side, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE also initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be the first auxiliary information message or an RRC signaling may be introduced to report the first
  • the auxiliary information message may include at least one of the following:
  • the fifth sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants the model to be updated on the UE side/network side.
  • This indication information can be an enumeration type, i.e., ⁇ UE side, network side ⁇
  • the fifth sub-auxiliary information indicates that the UE wants the model to be updated on the UE side/network side.
  • This indication information may be of an enumeration type, i.e., ⁇ UE side, network side ⁇ + the sixth indication information, indicating which use case/sub use case is targeted (i.e., here, the CSI compression sub use case);
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be updated on the UE side, and the other indication information indicating that the UE wants the model to be updated on the network side;
  • This indication information is two separate indication information, one indication information indicating that the UE wants the model to be updated on the UE side, and the other indication information indicating that the UE wants the model to be updated on the network side + sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device, in combination with some other situations or information, may decide whether to update the model for the CSI compression sub use case on the UE side or the network side, depending on the implementation of the network side device;
  • Step 4 The network-side device notifies the UE of the decision result of the network-side device, that is, whether the model is updated on the UE side or the network side;
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or CSI prediction, time domain beam prediction, spatial domain beam prediction, UE-assisted positioning, etc.
  • Position, direct positioning and other sub use cases that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE hopes that the model is updated on the UE side/network side, not for a specific use case/sub use case.
  • the first auxiliary information when the first auxiliary information includes sixth sub-auxiliary information, the first auxiliary information includes:
  • Second indication information where the second indication information is used to indicate the reason why the terminal expects to change the AI and/or ML model, or the reason why the terminal expects to activate the AI and/or ML model, or the reason why the terminal expects to deactivate the AI and/or ML model.
  • the UE reports that the UE wants to change/switch/activate/deactivate the model.
  • the reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wants to change/switch/activate/deactivate the model, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 can also be an indication information indicating that when the UE wants to change/switch/activate/deactivate the model, the UE can report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (that is, here is the CSI compression sub use case); the former indication information can be displayed or implicitly indicated (the method can be the same as above), and the latter indication information Display configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to change/switch/activate/deactivate the AI/ML model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the sixth sub-auxiliary information indicates that for the CSI compression sub-use case, the UE wants to change/switch/activate/deactivate the AI/ML model; this indication information is a whole 1-bit indication information;
  • the sixth sub-auxiliary information indicates that the UE wishes to change/switch/activate/deactivate the AI/ML model + the sixth indication information indicates which use case/sub use case is targeted (i.e., the CSI compression sub use case in this case);
  • second indication information such as a cause value, which may indicate that the model performance is poor or the model is not applicable
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 When the network side device receives the first auxiliary information reported by the UE, the network side device may decide whether to change/switch/activate/deactivate the AI/ML model based on some other situations or information, depending on the implementation of the network side device;
  • Step 4 If the network side device decides to change/switch/activate/deactivate the AI/ML model, the network side device sends a command to the UE to instruct the UE to change/switch/activate/deactivate the AI/ML model;
  • Step 5 The UE changes/switches/activates/deactivates the AI/ML model according to the instructions of the network-side device;
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases
  • sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the UE reports that the UE wants a model update.
  • the UE currently stores an AI/ML model for CSI compression.
  • the reporting process includes the following steps:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • Configuration information 1 first indication information, used to indicate that for the CSI compression sub-use case, when the UE wishes to update the model, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated. For example, when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate that it supports reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to update the model, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to update the model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the eighth auxiliary information indicates that the UE wishes to update the model for the CSI compression sub-use case
  • the eighth sub-auxiliary information indicating that the UE wishes to update the model + the sixth indication information, indicating which use case/sub-use case is targeted (i.e., here is the CSI compression sub-use case);
  • third indication information such as a cause value, which may indicate, for example, that the model performance is poor, the model generalization ability is poor, or the model is not applicable;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE, and the network side device combines some other situations or information, and may decide whether to update the model depending on the implementation of the network side device;
  • Step 4 If the network side device decides to update the model, the network side device sends an instruction to the UE to instruct the UE to update the model;
  • Step 5 The UE updates the model according to the instructions of the network side device.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE can send an overall auxiliary information to indicate that the UE wants the model to be updated. New, not targeting a specific use case/sub use case.
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the UE reports whether the UE desires an AI/ML-based method or a non-AI/ML-based method.
  • the reporting process may include at least one of the following:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wishes to use the AI/ML-based method/non-AI/ML-based method, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to use the AI/ML-based method/non-AI/ML-based method, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to use the AI/ML-based method/non-AI/ML-based method, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer. If the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer.
  • the UE also initializes the reporting of auxiliary information to the network side device.
  • the reporting signaling may be the first auxiliary information message or an RRC signaling may be introduced to report the first
  • the auxiliary information message may include at least one of the following:
  • the ninth sub-auxiliary information indicates that for the CSI compression sub use case, the UE wishes to use the AI/ML-based method/non-AI/ML-based method.
  • This indication information may be an enumeration type, such as ⁇ AI/ML-based method, non-AI/ML-based method ⁇
  • the ninth sub-auxiliary information indicates that the UE wishes to use the AI/ML-based method/non-AI/ML-based method.
  • This indication information may be an enumeration type, such as ⁇ AI/ML-based method, non-AI/ML-based method ⁇ +the sixth indication information, indicating which use case/sub use case is targeted (i.e., here is the CSI compression sub use case);
  • the ninth sub-auxiliary information indicates that for the CSI compression sub use case, the UE wishes to use the AI/ML-based method or the non-AI/ML-based method.
  • This indication information is two separate indication information, one indicating information indicating that the UE wishes to use the AI/ML-based method, and the other indicating information indicating that the UE wishes to use the non-AI/ML-based method;
  • This indication information is two separate indication information, one indication information indicating that the UE wishes to use the AI/ML-based method, and the other indication information indicating that the UE wishes to use the non-AI/ML-based method + the sixth indication information, indicating which use case/sub use case is targeted (i.e., CSI compression sub use case in this case);
  • Fourth indication information such as a cause value, which may indicate, for example, poor AI/ML model performance, good AI/ML model performance, too much overhead, or overheating.
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE.
  • the network side device combines some other situations or information and may decide whether to use the AI/ML-based method or the non-AI/ML-based method depending on the implementation of the network side device.
  • Step 4 The network-side device notifies the UE of the decision result (i.e., whether to use the AI/ML-based method or the non-AI/ML-based method);
  • Step 4 is an optional step.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE can send an overall auxiliary information to indicate that the UE wants to use an AI/ML-based method or a non-AI/ML-based method, not for a specific use case/sub use case.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the UE reports that the UE wants the model to be released.
  • the UE currently stores an AI/ML model for CSI compression.
  • the reporting process may include:
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE hopes to release the model, the UE can report the auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to release the model, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 First timer information, such as a ProhibitTimer value.
  • ProhibitTimer is used to prevent the UE from frequently reporting this auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wants to release the model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device.
  • the reporting signaling may be a first auxiliary information message or an RRC signaling is introduced to report the first auxiliary information message.
  • the first auxiliary information may include at least one of the following:
  • the tenth sub-assistance information indicates that the UE wishes to release the model for the CSI compression sub-use case
  • the tenth sub-auxiliary information indicates that the UE wants the model to be released + the sixth indication information indicates which use case/sub-use case is targeted (i.e., here is the CSI compression sub-use case);
  • Fifth indication information such as a cause value, which may indicate that the UE has insufficient memory or the overhead is too large.
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the first auxiliary information reported by the UE, and the network side device combines some other situations or information, and may decide whether to release the model depending on the implementation of the network side device;
  • Step 4 If the network side device decides to release the model, the network side device sends an indication to the UE to notify the UE to release the model;
  • Step 5 The UE releases the model according to the instructions of the network side device.
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is The indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use cases/sub use cases.
  • the UE wants to release the model of a certain/certain use cases or a certain/certain sub use cases, the UE can send an overall auxiliary information to indicate that the UE wants to release the model, not for a specific use case/sub use case.
  • the seventh auxiliary information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • model information of the AI and/or ML expected or preferred by the terminal may include model identification information of the AI and/or ML model expected or preferred by the terminal;
  • model information of the AI and/or ML expected or preferred by the terminal may include model priority information of the AI or ML expected or preferred by the terminal;
  • the model priority indicated by the seventh auxiliary sub-information can be embodied by model identification information.
  • Other methods are also possible and are not limited here.
  • model identification information includes:
  • the model identification information can be: Model id, functionality id, configuration id, model parameters, model input/output, etc.
  • the model identification information can be: Model id, functionality id, configuration id, model parameters, model input/output, etc.
  • Step 1 The network side device sends configuration information to the UE.
  • the configuration information includes one or more of the following information:
  • the first indication information is used to indicate that for the CSI compression sub use case, when the UE wishes to report a model recommendation/preference, the UE can report auxiliary information to notify the network side device; this indication information can be explicitly configured or implicitly indicated, such as when the network side device configures a ProhibitTimer value for reporting this auxiliary information, it can implicitly indicate support for reporting this auxiliary information;
  • Configuration information 1 may also be an indication information indicating that when the UE wishes to report a model recommendation/preference, the UE may report auxiliary information to notify the network side device + an indication information indicating which use case/sub use case is targeted (i.e., CSI compression sub use case is used here); the former indication information may display the configuration or may be implicitly expressed (in the same manner as above), and the latter indication information may display the configuration;
  • Configuration information 2 first timer information, such as a ProhibitTimer value, where the ProhibitTimer is used to prevent the UE from frequently reporting the auxiliary information;
  • step 1 may be configured to the UE in advance without time limit
  • step 1 is an optional step, that is, it is not a step that must be performed.
  • Step 2 When the UE wishes to change the AI/ML model from a model with a model ID of 1 to a model with a model ID of 3, or the UE wishes to notify the network side of the priority information of the AI/ML model, if the ProhibitTimer value is configured, the UE turns on the ProhibitTimer, and if the ProhibitTimer does not exist, the UE skips turning on the ProhibitTimer; and initializes reporting of auxiliary information to the network side device, the reporting signaling may be a first auxiliary information message or introduce an RRC signaling to report the first auxiliary information message, and the first auxiliary information may include at least one of the following:
  • auxiliary sub-information such as a model priority information, indicating the model priority information desired by the UE for the CSI compression sub-use case.
  • Eighth sub-auxiliary information for example, including a model priority information and a sixth indication information, indicating the model priority information desired by the UE.
  • model identification information can be functionality id, configuration id, model parameters, model input/output, etc. in addition to the Model id here;
  • the sixth indication information in step 1 and step 2 indicates which use case/sub use case is targeted, and the specific form may be an enumeration type, such as ⁇ CSI feedback enhancement, beam management, positioning accuracy enhancement ⁇ , or ⁇ CSI compression, time domain CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, direct positioning ⁇ ;
  • Step 3 The network side device receives the auxiliary information sent by the UE.
  • the network side device combines some other situations or information and may decide whether to change the model, etc., depending on the network side device implementation;
  • Step 4 If the network side device decides to change the model, the network side device sends a command to the UE to notify the UE to change the model;
  • Step 5 The UE changes the model according to the instructions of the network side device
  • the above process is applicable to any AI/ML use case/sub use case, that is, the above process can also be applied to use cases such as CSI feedback enhancement, beam management, and positioning accuracy enhancement, or sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, and direct positioning, that is, this configuration information and auxiliary information are per use case or per sub use case.
  • use cases such as CSI feedback enhancement, beam management, and positioning accuracy enhancement
  • sub use cases such as CSI prediction, time domain beam prediction, spatial domain beam prediction, UE assisted positioning, and direct positioning, that is, this configuration information and auxiliary information are per use case or per sub use case.
  • the above process may also be triggered before the UE side uses the AI/ML model.
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • the first auxiliary information is applicable to any use case or sub use case of AI/ML, and may be per use case or per sub use case, or as a whole; for the case of per use case or per sub use case, the UE may report second indication information to indicate which use case or sub use case the UE auxiliary information is for;
  • the reporting process of the first auxiliary information in each embodiment of the present disclosure is applicable to any AI/ML use case/sub use case, that is, it can also be applicable to use cases such as CSI feedback enhancement, beam management, positioning accuracy enhancement, or sub use cases such as space-frequency domain CSI compression, CSI prediction, time domain beam prediction, space domain beam prediction, UE assisted positioning, direct positioning, etc., that is, this configuration information and auxiliary information are per use case or per sub use case.
  • this configuration information and auxiliary information can also be for the whole, that is, the indication information and ProhibitTimer value configured by the network side device are for the whole, without distinguishing between use case/sub use case.
  • the UE When the UE wishes to report the first auxiliary information, no matter which use case/which use case or which sub use case/which sub use case collects data, the UE can send an overall auxiliary information to indicate that the UE now has collected data and wishes to report it, and it is not for a specific use case/sub use case.
  • the terminal device involved in the embodiments of the present disclosure may be a device that provides voice and/or data connectivity to users, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem.
  • the name of the terminal device may also be different.
  • the terminal device may be called a user equipment (UE).
  • the wireless terminal device can communicate with one or more core networks (CN) via a radio access network (RAN).
  • CN core networks
  • RAN radio access network
  • the wireless terminal device may be a mobile terminal device, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal device.
  • it may be a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device that exchanges language and/or data with a wireless access network.
  • Wireless terminal equipment may also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, an access point, a remote terminal device, an access terminal device, a user terminal device, a user agent, a user device, and the embodiments of the present disclosure are not limited thereto.
  • the network device involved in the embodiments of the present disclosure may be a base station, which may include multiple cells that provide services to the terminal.
  • the base station may also be called an access point, or may be a device in the access network that communicates with the wireless terminal device through one or more sectors on the air interface, or other names.
  • the network device can be used to interchange received air frames with Internet Protocol (IP) packets, acting as a router between the wireless terminal device and the rest of the access network, wherein the rest of the access network may include an Internet Protocol (IP) communication network.
  • IP Internet Protocol
  • the network device can also coordinate the attribute management of the air interface.
  • the network device involved in the embodiments of the present disclosure may be a network device (Base Transceiver Station, BTS) in the Global System for Mobile communications (GSM) or Code Division Multiple Access (CDMA), or a network device (NodeB) in Wide-band Code Division Multiple Access (WCDMA), or an evolved network device (evolutional Node B, eNB or e-NodeB) in the long term evolution (LTE) system, a 5G base station (gNB) in the 5G network architecture (next generation system), or a home evolved Node B (HeNB), a relay node, a home base station (femto), a pico base station (pico), etc., but is not limited in the embodiments of the present disclosure.
  • network devices may include centralized unit (CU) nodes and distributed unit (DU) nodes, and the centralized unit and the distributed unit may also be geographically separated.
  • FIG3 is a schematic diagram of the structure of a network side device provided by an embodiment of the present disclosure.
  • the network side device includes a memory 320, a transceiver 300, and a processor 310, wherein:
  • the memory 320 is used to store computer programs; the transceiver 300 is used to The processor 310 is used to read the computer program in the memory 320 and perform the following operations:
  • the first auxiliary information sent by the receiving terminal includes at least part of the information required to implement the model management process of artificial intelligence AI and/or machine learning ML.
  • the transceiver 300 is used to receive and send data under the control of the processor 310 .
  • the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 310 and various circuits of memory represented by memory 320 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 are therefore not further described herein.
  • the bus interface provides an interface.
  • the transceiver 300 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, which transmission medium includes a wireless channel, a wired channel, an optical cable, and other transmission media.
  • the processor 310 is responsible for managing the bus architecture and general processing, and the memory 320 may store data used by the processor 310 when performing operations.
  • Processor 310 can be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD).
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • CPLD complex programmable logic device
  • the processor can also adopt a multi-core architecture.
  • the processor 310 before receiving the first auxiliary information sent by the terminal, the processor 310 is further configured to:
  • first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the first auxiliary information includes one or more of the following:
  • the first sub-auxiliary information is used to indicate that the terminal expects to report the Data collected by the terminal;
  • second sub-auxiliary information where the second sub-auxiliary information is used to indicate that the terminal expects the network side device to send data collected by the network side device;
  • the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the terminal side, or the third sub-auxiliary information is used to represent that the terminal expects the model training process to be implemented on the network side device;
  • the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the terminal side, or the fourth sub-auxiliary information is used to represent that the terminal expects the model monitoring process to be implemented on the network side device;
  • the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the terminal side, or the fifth sub-auxiliary information is used to represent that the terminal expects the model update process to be implemented on the network side device;
  • the sixth sub-auxiliary information is used to represent that the terminal expects to change the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to activate the model of the AI and/or ML, or the sixth sub-auxiliary information is used to represent that the terminal expects to deactivate the model of the AI and/or ML;
  • the seventh sub-auxiliary information is used to characterize model information of the AI and/or ML expected or preferred by the terminal;
  • eighth sub-auxiliary information where the eighth sub-auxiliary information is used to indicate that the terminal expects to update the AI and/or ML model
  • Ninth sub-assistance information where the ninth sub-assistance information is used to indicate that the terminal expects to use an AI and/or ML-based method, or the ninth sub-assistance information is used to indicate that the terminal expects to use a non-AI and/or ML-based method;
  • the tenth sub-auxiliary information is used to characterize the model that the terminal expects to release the AI and/or ML.
  • the first auxiliary information when the first auxiliary information includes sixth sub-auxiliary information, the first auxiliary information includes:
  • the second indication information is used to indicate that the terminal expects to change the AI and/or ML model, or the reason why the terminal desires to activate the AI and/or ML model, or the reason why the terminal desires to deactivate the AI and/or ML model.
  • the first auxiliary information includes:
  • the third indication information is used to indicate the reason why the terminal expects to update the AI and/or ML model.
  • the first auxiliary information when the first auxiliary information includes ninth sub-auxiliary information, the first auxiliary information includes:
  • the fourth indication information is used to indicate the reason why the terminal expects to use the AI and/or ML-based method, or the fourth indication information is used to indicate the reason why the terminal expects to use the non-AI and/or ML-based method.
  • the first auxiliary information when the first auxiliary information includes tenth sub-auxiliary information, the first auxiliary information includes:
  • the fifth indication information is used to indicate the reason why the terminal expects to release the AI and/or ML model.
  • the seventh auxiliary information includes one or more of the following:
  • Model identification information of the AI and/or ML expected or preferred by the terminal
  • Model priority information of the AI and/or ML expected or preferred by the terminal is a model priority information of the AI and/or ML expected or preferred by the terminal.
  • model identification information includes:
  • the first auxiliary information includes:
  • sixth indication information where the sixth indication information is used to indicate one or more usage scenarios or usage sub-scenarios corresponding to the first auxiliary information.
  • the above-mentioned network side device provided in the embodiment of the present disclosure can implement all the method steps implemented by the method embodiment in which the execution subject is the network side device, and can achieve the same technical effect.
  • the parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.
  • FIG4 is a schematic diagram of the structure of a terminal provided by an embodiment of the present disclosure.
  • the terminal includes a memory 420, a transceiver 400, and a processor 410, wherein:
  • the memory 420 is used to store computer programs; the transceiver 400 is used to send and receive data under the control of the processor 410; the processor 410 is used to read the computer program in the memory 420 and perform the following operations:
  • First auxiliary information is sent to a network side device, where the first auxiliary information includes at least part of the information required to implement a model management process of artificial intelligence AI and/or machine learning ML.
  • the transceiver 400 is used to receive and send data under the control of the processor 410 .
  • the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 410 and various circuits of memory represented by memory 420 are linked together.
  • the bus architecture can also link various other circuits such as peripherals, voltage regulators, and power management circuits together, which are all well known in the art and are therefore not further described herein.
  • the bus interface provides an interface.
  • the transceiver 400 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, and these transmission media include transmission media such as wireless channels, wired channels, and optical cables.
  • the user interface 430 may also be an interface that can be connected to external and internal devices, and the connected devices include but are not limited to keypads, displays, speakers, microphones, joysticks, etc.
  • the processor 410 is responsible for managing the bus architecture and general processing, and the memory 420 can store data used by the processor 410 when performing operations.
  • the processor 410 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD), and the processor may also adopt 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 calls the computer program stored in the memory to execute any of the methods provided by the embodiments of the present disclosure according to the obtained executable instructions.
  • the processor and the memory can also be arranged physically separately.
  • the processor 410 before sending the first auxiliary information to the network side device, the processor 410 is further configured to:
  • First configuration information sent by a network side device is received, where the first configuration information is configuration information required for the terminal to report first auxiliary information to the network side device.
  • the first configuration information includes one or more of the following:
  • first indication information where the first indication information is used to indicate that the network-side device supports reporting of first auxiliary information
  • First timer information where the first timer is used to control the terminal to report first auxiliary information.
  • the sending the first auxiliary information to the network side device includes one or more of the following:
  • the first auxiliary information is sent to the network side device.
  • the processor 410 before sending the first auxiliary information to the network side device, the processor 410 is further configured to perform one or more of the following:
  • the first auxiliary information includes one or more of the following:

Abstract

本公开实施例提供一种辅助信息上报方法及装置,属于通信技术领域,该辅助信息上报方法包括:向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。

Description

辅助信息上报方法及装置
相关申请的交叉引用
本公开要求于2022年12月30日提交的申请号为202211735444.0,发明名称为“辅助信息上报方法及装置”的中国专利申请的优先权,其通过引用方式全部并入本文。
技术领域
本公开涉及通信技术领域,尤其涉及一种辅助信息上报方法及装置。
背景技术
随着科技的不断发展,人工智能(Artificial Intelligence,AI)和/或机器学习(Machine Learning,ML)技术也随之广泛地应用于生产生活中。
因此,可以将AI和/或ML应用在无线通信系统中。但是AI和/或ML的架构与无线通信系统的架构存在较多不同的地方,因此,如何在无线通信系统中实现AI和/或ML的相关功能或过程是不清楚的,导致AI和/或ML在无线通信系统中的实用性较差。
发明内容
本公开实施例提供一种辅助信息上报方法及装置,用以解决现有技术中AI和/或ML在无线通信系统中的实用性较差的缺陷,实现提高AI和/或ML在无线通信系统中的实用性。
第一方面,本公开实施例提供一种辅助信息上报方法,应用于终端,所述方法包括:
向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的辅助信息上报方法,在向所述网络侧 设备发送所述第一辅助信息之前,所述方法还包括:
接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述向网络侧设备发送第一辅助信息前,还包括以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端或网络侧实现;
确定所述终端期望模型监测过程在所述终端或网络侧实现;
确定所述终端期望模型更新过程在所述终端或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行;
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一辅助信 息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助 信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
第二方面,本公开实施例还提供一种辅助信息上报方法,应用于网络侧设备,所述方法包括:
接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的辅助信息上报方法,在接收终端发送的第一辅助信息之前,所述方法还包括:
向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测 过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助 信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,根据本公开一个实施例的辅助信息上报方法,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
第三方面,本公开实施例还提供一种终端,包括存储器,收发机,处理器,其中:
存储器,用于存储计算机程序;收发机,用于在所述处理器的控制下收发数据;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:
向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的终端,在向所述网络侧设备发送所述第一辅助信息之前,所述操作还包括:
接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的终端,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的终端,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,根据本公开一个实施例的终端,所述向网络侧设备发送第一辅助信息前,所述操作还包括以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端或网络侧实现;
确定所述终端期望模型监测过程在所述终端或网络侧实现;
确定所述终端期望模型更新过程在所述终端或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行。
可选地,根据本公开一个实施例的终端,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的终端,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的终端,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的终端,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的终端,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的终端,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的终端,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,根据本公开一个实施例的终端,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
第四方面,本公开实施例还提供一种网络侧设备,包括存储器,收发机,处理器,其中:
存储器,用于存储计算机程序;收发机,用于在所述处理器的控制下收发数据;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:
接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的网络侧设备,在接收终端发送的第一辅助信息之前,所述操作还包括:
向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的网络侧设备,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的网络侧设备,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练 过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的网络侧设备,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的网络侧设备,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的网络侧设备,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的网络侧设备,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的网络侧设备,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的网络侧设备,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,根据本公开一个实施例的网络侧设备,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
第五方面,本公开实施例还提供一种辅助信息上报装置,包括:
发送模块,用于向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述装置还包括:
配置信息接收模块,用于在向所述网络侧设备发送所述第一辅助信息之前,接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述发送模块用于以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述装置还包括:
执行模块,用于在所述向网络侧设备发送第一辅助信息前,执行以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端或网络侧实现;
确定所述终端期望模型监测过程在所述终端或网络侧实现;
确定所述终端期望模型更新过程在所述终端或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述 AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
第六方面,本公开实施例还提供一种辅助信息上报装置,包括:
接收模块,用于接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述装置还包括:
配置信息发送模块,用于在接收终端发送的第一辅助信息之前,向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,根据本公开一个实施例的辅助信息上报装置,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
本公开实施例提供的辅助信息上报方法及装置,通过终端向网络侧设备发送第一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
附图说明
为了更清楚地说明本公开实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本公开的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本公开实施例提供的辅助信息上报方法的流程示意图之一;
图2是本公开实施例提供的辅助信息上报方法的流程示意图之二;
图3是本公开实施例提供的一种网络侧设备的结构示意图;
图4是本公开实施例提供的一种终端的结构示意图;
图5是本公开实施例提供的辅助信息上报装置的结构示意图之一;
图6是本公开实施例提供的辅助信息上报装置的结构示意图之二。
具体实施方式
本公开实施例中术语“和/或”,描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。
本公开实施例中术语“多个”是指两个或两个以上,其它量词与之类似。
下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行 清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分实施例,并不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
本公开实施例提供了辅助信息上报方法方法及装置,用以提高人工智能(Artificial Intelligence,AI)和/或机器学习(Machine Learning,ML)在无线通信系统中的实用性。
其中,方法和装置是基于同一申请构思的,由于方法和装置解决问题的原理相似,因此装置和方法的实施可以相互参见,重复之处不再赘述。
本公开实施例提供的技术方案可以适用于多种系统,尤其是5G系统。例如适用的系统可以是全球移动通讯(global system of mobile communication,GSM)系统、码分多址(code division multiple access,CDMA)系统、宽带码分多址(Wideband Code Division Multiple Access,WCDMA)通用分组无线业务(general packet radio service,GPRS)系统、长期演进(long term evolution,LTE)系统、LTE频分双工(frequency division duplex,FDD)系统、LTE时分双工(time division duplex,TDD)系统、高级长期演进(long term evolution advanced,LTE-A)系统、通用移动系统(universal mobile telecommunication system,UMTS)、全球互联微波接入(worldwide interoperability for microwave access,WiMAX)系统、5G新空口(New Radio,NR)系统等。这多种系统中均包括终端设备和网络设备。系统中还可以包括核心网部分,例如演进的分组系统(Evloved Packet System,EPS)、5G系统(5GS)等。
针对不同的功能需求,终端可以上报一些辅助信息给网络侧,如在节能机制(power saving)中,过热信息(overheating)等用户设备(User Equipment,UE)辅助信息将上报给网络;对于去激活状态的(deactivated)辅小区组(Secondary Cell Group,SCG),UE是否希望SCG被去激活等辅助信息将上报给网络。UE通过辅助信息上报来提供UE自身的情况给网络侧,用于辅助网络侧更好的决策。
针对人工智能(Artificial Intelligence,AI)和/或机器学习(Machine  Learning,ML)的模型管理过程,本公开实施例提供了一种辅助信息上报方法及装置。
图1是本公开实施例提供的辅助信息上报方法的流程示意图之一,如图1所示,辅助信息上报方法应用于终端,即执行主体是终端,所述方法包括:
步骤100,向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括调整AI和/或ML的模型状态的过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括优化AI和/或ML的模型的过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括AI和/或ML的模型应用过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括任意涉及到AI和/或ML的模型的过程,本公开实施例对此不作限定。
可选地,人工智能AI和/或机器学习ML的模型管理过程,包括以下LCM过程的一项或多项:
数据收集Data collection;
模型训练Model training;
模型注册Model registration;
模型发布Model deployment;
模型配置Model configuration;
模型推断Model inference;
模型选择Model selection;
模型激活Model activation;
模型去激活Model deactivation;
模型切换Model switching;
模型释放Model fallback;
模型监测Model monitoring;
模型更新Model update;
模型传输Model transfer;
终端能力管理UE capability;
可选地,终端可以向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现AI/ML生命周期管理(Life-Cycle Management,LCM)的各个过程所需要的至少部分信息。即终端可以向网络侧设备发送第一辅助信息,所述第一辅助信息可以辅助以下任意一项或多项的实现:
数据收集Data collection;
模型训练Model training;
模型注册Model registration;
模型发布Model deployment;
模型配置Model configuration;
模型推断Model inference;
模型选择Model selection;
模型激活Model activation;
模型去激活Model deactivation;
模型切换Model switching;
模型释放Model fallback;
模型监测Model monitoring;
模型更新Model update;
模型传输Model transfer;或
终端能力管理UE capability。
可选地,由于AI和/或ML的架构与无线通信的架构存在差异,前者由数据驱动,后者由事件或时间驱动,为了在无线通信系统中实现AI和/或ML的相关功能,可以通过终端向网络侧发送第一辅助信息,第一辅助信息可以包括用于驱动或触发AI和/或ML的相关功能的数据,或者驱动或触发AI和/或ML的相关功能的执行的信息或指示。
可选地,第一辅助信息的发送可以通过UEAssistanceInformation消息来上报;
可选地,可以引入一条无线资源控制(Radio Resource Control,RRC)信令,专用于承载第一辅助信息。
可选地,网络侧设备收到第一辅助信息后,网络侧设备可以基于所述第一辅助信息进行决策,获得决策结果信息,所述决策结果信息用于表征是否满足所述终端的期望;并向终端发送所述决策结果信息。
可选地,网络侧设备收到第一辅助信息后,网络侧设备可以结合一些其它情况或信息,决策是否执行第一辅助信息对应的相关过程,依据决策结果,向终端发送决策结果信息。
可选地,网络侧设备基于第一辅助信息决策的过程取决于网络实现,本公开实施例对此不作限定。
本公开实施例提供的辅助信息上报方法,通过终端向网络侧设备发送第一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
可选地,在向所述网络侧设备发送所述第一辅助信息之前,所述方法还包括:
接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,网络侧设备可以首先向终端发送第一配置信息,第一配置信息用于配置第一辅助信息的上报,终端可以在收到第一配置信息后,基于第一配置信息,向网络侧设备发送第一辅助信息。
可选地,终端上报辅助信息不一定必须基于网络配置,可以基于网络配置也可以不基于网络配置;
可选地,终端可以在收到第一配置信息的情况下,基于第一配置信息,向网络侧设备发送第一辅助信息;
可选地,终端也可以在未收到第一配置信息的情况下,基于第一配置信 息,向网络侧设备发送第一辅助信息。
可选地,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,网络侧设备可以为终端配置第一指示信息和/或第一定时器信息(ProhibitTimer值);
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报UE希望上报终端收集的数据的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望上报终端收集的数据的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望上报终端收集的数据的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息,第一定时器信息(ProhibitTimer值)控制终端在定时器超时前最多只上报预设次终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型训练在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型训练在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型训练在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型监测在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型监测在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最 多只上报一次终端希望模型监测在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型监测在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型监测在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型监测在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型更新在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次上报终端希望模型更新在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型更新在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型变更/switch/激活/去激活的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型变更/switch/激活/去激活的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型变更/switch/激活/去激活的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报模型推荐/偏好(preference)的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次模型推荐/preference的辅助信息,比如控制终端在定时器超时前最多只上报一次模型推荐/preference的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型更新的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型更新的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型更新的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望基于AI和/或ML(AI/ML-based)方式或非AI/ML-based方式的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最 多只上报预设次终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望释放AI/ML模型的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望释放AI/ML模型的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望释放AI/ML模型的辅助信息。
在一个实施例中,网络侧设备可以为UE配置如下信息中的一种或多种:
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望上报UE Data collection的数据的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/模型更新等的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望模型训练在UE侧/网络侧的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望模型监测在UE侧/网络侧的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望模型更新在UE侧/网络侧的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望模型变更/switch/激活/去激活的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报模型推荐/preference的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望模型更新的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望AI/ML-based方式或非AI/ML-based方式的辅助信息;
配置一个指示信息(显性指示或隐含指示)和/或一个ProhibitTimer值,此指示信息用于指示UE可以/需要上报UE希望释放AI/ML模型的辅助信息。
可选地,网络侧设备可以为终端配置ProhibitTimer,防止UE频繁的上报第一辅助信息。
可选地,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,终端可以执行以下至少一项:
启动某一项第一辅助信息对应的ProhibitTimer,并在ProhibitTimer超时前,不再发送该第一辅助信息;
发送该第一辅助信息。
可选地,所述向网络侧设备发送第一辅助信息前,还包括以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端侧或网络侧实现;
确定所述终端期望模型监测过程在所述终端侧或网络侧实现;
确定所述终端期望模型更新过程在所述终端侧或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行;
可选地,终端可以在确定终端期望上报所述终端收集的数据的情况下,比如当UE有Data collection的数据,并希望将此数据上报给网络,则终端可以执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报UE希望上报终端收集的数据的辅助信息;
发送UE希望上报终端收集的数据的辅助信息。
可选地,终端可以在确定终端期望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新的情况下,比如当UE需要数据集进行模型训练/模型更新等,则终端可以执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息;
发送终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息。
可选地,终端可以在确定终端希望模型训练在终端侧/网络侧的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望模型训练在终端侧/网络侧的辅助信息;
发送终端希望模型训练在终端侧/网络侧的辅助信息。
可选地,终端可以在确定终端希望模型监测在终端侧/网络侧的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望模型监测在终端侧/网络侧的辅助信息;
发送终端希望模型监测在终端侧/网络侧的辅助信息。
可选地,终端可以在确定终端希望模型更新在终端侧/网络侧的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望模型更新在终端侧/网络侧的辅助信息;
发送终端希望模型更新在终端侧/网络侧的辅助信息。
可选地,终端可以在确定终端希望模型变更/switch/激活/去激活的情况下,比如当UE考虑到一些因素,希望模型变更/switch/激活/去激活时,则终端可以执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望模型变更/switch/激活/去激活的辅助信息;
发送终端希望模型变更/switch/激活/去激活的辅助信息;
上报cause值,比如可以指示模型性能较差或模型不适用等。
可选地,终端可以在确定终端希望上报模型推荐/preference的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报模型推荐/preference的辅助信息;
发送模型推荐/preference的辅助信息。
可选地,终端可以在确定终端希望模型更新的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望模型更新的辅助信息;
发送终端希望模型更新的辅助信息;
上报cause值,比如可以指示模型性能较差,模型泛化能力较差,或模型不适用等。
可选地,终端可以在确定终端希望AI/ML-based方式或非AI/ML-based 方式的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息;
发送终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息;
上报cause值,比如可以指示AI/ML模型性能较差,AI/ML模型性能较好,overhead太大,或overheating等。
可选地,终端可以在确定终端希望释放AI/ML模型的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报终端希望释放AI/ML模型的辅助信息;
发送终端希望释放AI/ML模型的辅助信息;
上报cause值,比如可以指示UE内存不足,或overhead太大等。
可选地,终端可以在确定网络配置UE上报第一辅助信息的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报第一辅助信息;
发送第一辅助信息。
可选地,终端可以在确定第一定时器没有运行的情况下,执行以下至少一项:
启动ProhibitTimer,并在ProhibitTimer超时前,不再上报第一辅助信息;
发送第一辅助信息。
可选地,网络侧设备收到上述UE上报的辅第一辅助信息后,网络侧设备可以结合一些其它情况或信息,比如取决于网络实现,决策是否请求UE上报UE Data collection的数据,是否发送Data collection的数据集给UE用于UE进行模型训练/更新,是否指示UE进行模型的变更/switch/激活/去激活等,而后依据网络决策结果,来决定是否发送相应的信令给UE。
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,每一个子辅助信息可以分别对应自己的第一定时器信息ProhibitTimer,由网络侧设备分开配置,或基于默认确定,或协议预定义, 或预设置,或实时设置。
可选地,多个子辅助信息可以对应统一的第一定时器信息ProhibitTimer,由网络侧设备分开或统一配置,或基于默认确定,或协议预定义,或预设置,或实时设置。
可选地,终端收集的数据可以包括UE Data collection的数据。
可选地,网络侧设备收集的数据可以包括网络侧Data collection后的数据。
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括模型推荐/preference。
在一个实施例中,UE上报UE希望上报UE Data collection的数据,UE正在进行AI/ML Data collection的数据收集,以CSI压缩sub use case为例,即UE正在进行针对CSI压缩sub use case的AI/ML Data collection工作,则辅助信息上报流程可以包括:
步骤1:网络侧设备发送第一配置信息给UE,第一配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望上报收集到的针对信道状态信息(Channel State Information,CSI)压缩sub use case的数据时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望上报收集到的数据时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;前一个指示信息可以显示配置,也可以隐式表示,后一个指示信息显示配置;
配置信息2:第一定时器信息,比如ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望上报收集到的针对CSI压缩使用子场景(sub use case)的数据给网络侧设备时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第一子辅助信息,指示UE希望上报收集到的针对CSI压缩sub use case的数据给网络侧设备;
(2)第一子辅助信息,指示UE希望上报收集到的数据给网络侧设备+第六指示信息,指示收集到的数据是针对的哪个use case/sub use case,即这里为CSI压缩sub use case;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域波束(beam)预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备收到此UE上报的第一辅助信息后,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否请求UE上报收集到的针对CSI压缩sub use case的数据;
步骤4:若网络侧设备决策请求UE上报收集到的针对CSI压缩sub use case的数据,网络侧设备发送信令给UE来请求;
步骤5:UE根据网络侧设备请求,上报UE收集到的针对CSI压缩sub use case的数据给网络侧设备;
可选地,上述过程对于任何AI/ML使用场景(use case)/使用子场景(sub use case)均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和 ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望上报收集到的数据,无论是针对哪个/哪些use case或者哪个/哪些sub use case收集的数据,UE可以发一个整体的辅助信息来指示UE现在有收集的数据希望上报,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望网络侧设备发送Data collection后的数据给UE,UE需要数据集来完成模型训练/模型更新等工作,这里以CSI压缩sub use case为例,即UE需要数据集来完成CSI压缩sub use case模型的训练/更新等工作,辅助信息上报流程可以包括:
步骤1:网络侧设备发送第一配置信息给UE,第一配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望网络侧设备发送针对CSI压缩sub use case的Data collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新等工作时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;前一个指示信息可以显示配置,也可以隐式表示,后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望网络侧设备发送针对CSI压缩sub use case的Data collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如 果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此辅助信息可以包括如下至少一项:
(1)第二子辅助信息,指示UE希望网络侧设备发送针对CSI压缩sub use case的Data collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作;
(2)第二子辅助信息,指示UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新等工作+第六指示信息,指示此Data collection数据是针对的哪个use case/sub use case(即这里为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息后,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否发送针对CSI压缩sub use case的Data collection的数据给UE;
步骤4:若网络侧设备决策发送针对CSI压缩sub use case的Data collection的数据给UE,网络侧设备发送针对CSI压缩sub use case的Data collection的数据给UE;
步骤5:UE接收网络侧设备发送的针对CSI压缩sub use case的Data collection的数据,并用收到的数据进行后续的针对CSI压缩的模型训练/更新工作;
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即 网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望网络侧设备发送针对某个/某些use case或某个/某些sub use case的Data collection后的数据给UE,用于UE进行某个/某些use case或某个/某些sub use case模型的训练/更新等工作,UE可以发一个整体的辅助信息来指示UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型训练在UE侧/网络侧;以CSI压缩sub use case为例,辅助信息上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型训练在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型训练在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型训练在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第三子辅助信息,指示针对CSI压缩sub use case,UE希望模型训练在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧};
(2)第三子辅助信息,指示UE希望模型训练在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第三子辅助信息,指示针对CSI压缩sub use case,UE希望模型训练在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型训练在UE侧,另一个指示信息指示UE希望模型训练在网络侧;
(4)第三子辅助信息,指示UE希望模型训练在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型训练在UE侧,另一个指示信息指示UE希望模型训练在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型训练是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型训练在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或 per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型训练在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型训练在UE侧/网络侧,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型监测在UE侧/网络侧;以CSI压缩sub use case为例,包括如下步骤:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型监测在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型监测在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型监测在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第四子辅助信息,指示针对CSI压缩sub use case,UE希望模型监测在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}
(2)第四子辅助信息,指示UE希望模型监测在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第四子辅助信息,指示针对CSI压缩sub use case,UE希望模型监测在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型监测在UE侧,另一个指示信息指示UE希望模型监测在网络侧;
(4)第四子辅助信息,指示UE希望模型监测在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型监测在UE侧,另一个指示信息指示UE希望模型监测在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型监测是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型监测在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或 per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型监测在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型监测在UE侧/网络侧,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型更新在UE侧/网络侧;以CSI压缩sub use case为例,辅助信息上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型更新在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型更新在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型更新在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第五子辅助信息,指示针对CSI压缩sub use case,UE希望模型 更新在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}
(2)第五子辅助信息,指示UE希望模型更新在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第五子辅助信息,指示针对CSI压缩sub use case,UE希望模型更新在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型更新在UE侧,另一个指示信息指示UE希望模型更新在网络侧;
(4)第五子辅助信息,指示UE希望模型更新在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型更新在UE侧,另一个指示信息指示UE希望模型更新在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型更新是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型更新在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即 网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型更新在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型更新在UE侧/网络侧,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型变更/switch/激活/去激活,以CSI压缩sub use case为例,若UE侧存在5个AI/ML模型id=1,id=2,id=3,id=4,id=5用于CSI压缩,网络侧存在5个AI/ML模型id=6,id=7,id=8,id=9,id=10用于CSI解压缩。UE侧当前正在使用AI/ML模型id=1的模型进行CSI压缩工作,相对应的,网络侧使用AI/ML模型id=6的模型进行CSI解压缩工作,上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型变更/switch/激活/去激活时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型变更/switch/激活/去激活时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此
ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望将AI/ML模型变更/switch/激活/去激活时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第六子辅助信息,指示针对CSI压缩sub use case,UE希望将AI/ML模型变更/switch/激活/去激活;此指示信息是一个整体的1bit指示信息;
(2)第六子辅助信息,指示UE希望将AI/ML模型变更/switch/激活/去激活+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第二指示信息,比如一个cause值,比如可以指示模型性能较差或模型不适用等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否将AI/ML模型变更/switch/激活/去激活;
步骤4:若网络侧设备决策将AI/ML模型变更/switch/激活/去激活,网络侧设备发送指令给UE来指示UE将AI/ML模型变更/switch/激活/去激活;
步骤5:UE根据网络侧设备的指令将AI/ML模型变更/switch/激活/去激活;
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型更新,以CSI压缩sub use case为例,UE侧当前存储有用于CSI压缩的AI/ML模型,上报流程包括如下步骤:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
(1)配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型更新时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型更新时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
(2)配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望模型更新时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer; 并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第八子辅助信息,指示针对CSI压缩sub use case,UE希望模型更新;
(2)第八子辅助信息,指示UE希望模型更新+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第三指示信息,比如cause值,比如可以指示模型性能较差,模型泛化能力较差,或模型不适用等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否进行模型更新;
步骤4:若网络侧设备决策进行模型更新,网络侧设备发送指示给UE指示UE进行模型更新;
步骤5:UE根据网络侧设备指示进行模型更新。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望对某个/某些use case或某个/某些sub use case的模型进行更新,UE可以发一个整体的辅助信息来指示UE希望模型更新,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一 辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
在一个实施例中,UE上报UE希望AI/ML-based方式或非AI/ML-based方式,以CSI压缩sub use case为例,上报流程可以包括如下至少一项:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望使用AI/ML-based方式/非AI/ML-based方式时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望使用AI/ML-based方式/非AI/ML-based方式时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望使用AI/ML-based方式/非AI/ML-based方式时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第九子辅助信息,指示针对CSI压缩sub use case,UE希望使用 AI/ML-based方式/非AI/ML-based方式,此指示信息可以为枚举类型,如{AI/ML-based方式,非AI/ML-based方式}
(2)第九子辅助信息,指示UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息可以为枚举类型,如{AI/ML-based方式,非AI/ML-based方式}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第九子辅助信息,指示针对CSI压缩sub use case,UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望使用AI/ML-based方式,另一个指示信息指示UE希望使用非AI/ML-based方式;
(4)第九子辅助信息,指示UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望使用AI/ML-based方式,另一个指示信息指示UE希望使用非AI/ML-based方式+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(5)第四指示信息,比如一个cause值,比如可以指示AI/ML模型性能较差,AI/ML模型性能较好,overhead太大,或overheating等。
其中,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策使用AI/ML-based方式还是非AI/ML-based方式;
步骤4:网络侧设备将决策结果通知UE(即使用AI/ML-based方式还是非AI/ML-based方式);
步骤4是可选执行的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述 过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case使用AI/ML-based方式或非AI/ML-based方式,UE可以发一个整体的辅助信息来指示UE希望使用AI/ML-based方式或非AI/ML-based方式,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型释放,以CSI压缩sub use case为例,UE侧当前存储有用于CSI压缩的AI/ML模型,上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型释放时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型释放时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望模型释放时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第十子辅助信息,指示针对CSI压缩sub use case,UE希望模型释放;
(2)第十子辅助信息,指示UE希望模型释放+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第五指示信息,比如一个cause值,比如可以指示UE内存不足,或overhead太大等。
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否将模型释放;
步骤4:若网络侧设备决策将模型释放,网络侧设备发送指示给UE通知UE将模型释放;
步骤5:UE根据网络侧设备指示将模型释放。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望将某个/某些use case或某个/某些sub use case 的模型释放,UE可以发一个整体的辅助信息来指示UE希望模型释放,并不针对特别的use case/sub use case。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括终端期望或偏好的所述AI和/或ML的模型的模型标识信息;
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括终端期望或偏好的所述AI或ML的模型优先级信息;
比如,第七子辅助信息所指示的模型优先级可以以模型标识信息来体现模型优先级信息,例如,模型优先级信息可以是一个模型标识信息列表list,如model id=1,model id=5,model id=6等,这样体现的模型优先级信息为model id=1的模型优先级高于model id=5的模型高于model id=6的模型。也可以是其它方式,这里不限制。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,模型标识信息可以为:Model id,functionality id,configuration id,模型参数,模型输入/输出等。
在一个实施例中,UE上报模型推荐/preference;以CSI压缩sub use case为例,若UE侧存在5个AI/ML模型id=1,id=2,id=3,id=4,id=5用于CSI压缩,网络侧存在5个AI/ML模型id=6,id=7,id=8,id=9,id=10用于CSI解压缩。UE侧当前正在使用AI/ML模型id=1的模型进行CSI压缩工作,相对应的,网络侧使用AI/ML模型id=6的模型进行CSI解压缩工 作。
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望上报模型推荐/preference时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望上报模型推荐/preference时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望将AI/ML模型从模型id为1的模型变更到模型id为3的模型,或者UE希望将AI/ML模型的优先级信息通知网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第八子辅助信息,比如一个模型id值,指示针对CSI压缩sub use case,UE希望将当前的AI/ML模型变更为模型id=3的模型;
(2)第八子辅助信息,比如包括一个模型id值和第六指示信息,指示UE希望将当前的AI/ML模型变更为模型id=3的模型,指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;
(3)第八子辅助信息,比如一个模型优先级信息,指示针对CSI压缩sub use case,UE希望的模型优先级信息,此模型优先级可以是一个list的形式,如模型优先级为:id=3,id=1,id=2,id=5,id=6,id=4;
(4)第八子辅助信息,比如包括一个模型优先级信息和第六指示信息,指示UE希望的模型优先级信息,此模型优先级可以是一个list的形式,如模型优先级为:id=3,id=1,id=2,id=5,id=6,id=4+一个指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,模型标识信息除了此处的Model id,还可以为functionality id,configuration id,模型参数,模型输入/输出等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE发送的辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否进行模型变更等;
步骤4:若网络侧设备决策进行模型变更,网络侧设备发送指令给UE通知UE进行模型变更;
步骤5:UE根据网络侧设备指令进行模型变更;
可选的,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。
可选地,上述过程也可以在UE侧使用AI/ML模型之前触发。
可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
可选地,第一辅助信息对于任何一个AI/ML的use case或者sub use case均适用,可以是per use case的或者per sub use case的,或者整体的;对于per use case或per sub use case的情况,UE可以上报第二指示信息来指示此UE辅助信息针对的是哪个use case或哪个sub use case;
可选地,本公开各实施例中第一辅助信息的上报过程对于任何AI/ML use case/sub use case均适用,即也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者空频域CSI压缩,CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望上报第一辅助信息,无论是针对哪个/哪些use case或者哪个/哪些sub use case收集的数据,UE可以发一个整体的辅助信息来指示UE现在有收集的数据希望上报,并不针对特别的use case/sub use case。
本公开实施例提供的辅助信息上报方法,通过终端向网络侧设备发送第一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
图2是本公开实施例提供的辅助信息上报方法的流程示意图,如图2所示,辅助信息上报方法应用于网络侧设备,即执行主体为网络侧设备;所述方法包括:
步骤200,接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括调整AI和/或ML的模型状态的过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括优化AI和/或ML的模型的过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括AI 和/或ML的模型应用过程;
可选地,人工智能AI和/或机器学习ML的模型管理过程,可以包括任意涉及到AI和/或ML的模型的过程,本公开实施例对此不作限定。
可选地,人工智能AI和/或机器学习ML的模型管理过程,包括以下LCM过程的一项或多项:
数据收集Data collection;
模型训练Model training;
模型注册Model registration;
模型发布Model deployment;
模型配置Model configuration;
模型推断Model inference;
模型选择Model selection;
模型激活Model activation;
模型去激活Model deactivation;
模型切换Model switching;
模型释放Model fallback;
模型监测Model monitoring;
模型更新Model update;
模型传输Model transfer;
终端能力管理UE capability。
可选地,终端可以向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现AI/ML生命周期管理(Life-Cycle Management,LCM)的各个过程所需要的至少部分信息。即终端可以向网络侧设备发送第一辅助信息,所述第一辅助信息可以辅助以下任意一项或多项的实现:
数据收集Data collection;
模型训练Model training;
模型注册Model registration;
模型发布Model deployment;
模型配置Model configuration;
模型推断Model inference;
模型选择Model selection;
模型激活Model activation;
模型去激活Model deactivation;
模型切换Model switching;
模型释放Model fallback;
模型监测Model monitoring;
模型更新Model update;
模型传输Model transfer;或
终端能力管理UE capability。
可选地,由于AI和/或ML的架构与无线通信的架构存在差异,前者由数据驱动,后者由事件或时间驱动,为了在无线通信系统中实现AI和/或
ML的相关功能,可以通过网络侧接收终端发送的第一辅助信息,第一辅助信息可以包括用于驱动或触发AI和/或ML的相关功能的数据,或者驱动或触发AI和/或ML的相关功能的执行的信息或指示。
可选地,第一辅助信息的发送可以通过UEAssistanceInformation消息来上报;
可选地,可以引入一条RRC信令,专用于承载第一辅助信息。
可选地,网络侧设备收到第一辅助信息后,网络侧设备可以基于所述第一辅助信息进行决策,获得决策结果信息,所述决策结果信息用于表征是否满足所述终端的期望;并向终端发送所述决策结果信息。
可选地,网络侧设备收到第一辅助信息后,网络侧设备可以结合一些其它情况或信息,决策是否执行第一辅助信息对应的相关过程,依据决策结果,向终端发送决策结果信息。
可选地,网络侧设备基于第一辅助信息决策的过程取决于网络实现,本公开实施例对此不作限定。
本公开实施例提供的辅助信息上报方法,通过终端向网络侧设备发送第 一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
可选地,在接收终端发送的第一辅助信息之前,所述方法还包括:
向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,网络侧设备可以首先向终端发送第一配置信息,第一配置信息用于配置第一辅助信息的上报,终端可以在收到第一配置信息后,基于第一配置信息,向网络侧设备发送第一辅助信息。
可选地,终端上报辅助信息不一定必须基于网络配置,可以基于网络配置也可以不基于网络配置;
可选地,终端可以在收到第一配置信息的情况下,基于第一配置信息,向网络侧设备发送第一辅助信息;
可选地,终端也可以在未收到第一配置信息的情况下,基于第一配置信息,向网络侧设备发送第一辅助信息。
可选地,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,网络侧设备可以为终端配置第一指示信息和/或第一定时器信息(ProhibitTimer值);
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报UE希望上报终端收集的数据的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望上报终端收集的数据的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望上报终端收集的数据的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/ 模型更新等的辅助信息,第一定时器信息(ProhibitTimer值)控制终端在定时器超时前最多只上报预设次终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望网络侧设备发送网络侧收集的数据给终端,用于终端进行模型训练/模型更新等的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型训练在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型训练在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型训练在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型监测在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型监测在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型监测在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型监测在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型监测在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型监测在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型更新在终端侧/网络侧的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次上报终端希望模型更新在终端侧/网络侧的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型更新在终端侧/网络侧的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型变更/switch/激活/去激活的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希 望模型变更/switch/激活/去激活的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型变更/switch/激活/去激活的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报模型推荐/preference的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次模型推荐/preference的辅助信息,比如控制终端在定时器超时前最多只上报一次模型推荐/preference的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望模型更新的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望模型更新的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望模型更新的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望AI/ML-based方式或非AI/ML-based方式的辅助信息。
可选地,第一指示信息可以显性指示或隐含指示终端可以/需要上报终端希望释放AI/ML模型的辅助信息,第一定时器信息(ProhibitTimer值)可以控制终端在定时器超时前最多只上报预设次终端希望释放AI/ML模型的辅助信息,比如控制终端在定时器超时前最多只上报一次终端希望释放AI/ML模型的辅助信息。
在一个实施例中,网络侧设备可以为UE配置如下信息中的一种或多种:
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望上报UE Data collection的数据的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器 信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/模型更新等的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望模型训练在UE侧/网络侧的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望模型监测在UE侧/网络侧的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望模型更新在UE侧/网络侧的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望模型变更/switch/激活/去激活的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报模型推荐/preference的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望模型更新的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望AI/ML-based方式或非AI/ML-based方式的辅助信息;
配置一个第一指示信息(显性指示或隐含指示)和/或一个第一定时器信息(ProhibitTimer值),此指示信息用于指示UE可以/需要上报UE希望释放AI/ML模型的辅助信息。
可选地,网络侧设备可以为终端配置ProhibitTimer,防止UE频繁的上 报第一辅助信息。
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,每一个子辅助信息可以分别对应自己的第一定时器信息ProhibitTimer,由网络侧设备分开配置,或基于默认确定,或协议预定义,或预设置,或实时设置。
可选地,多个子辅助信息可以对应统一的第一定时器信息ProhibitTimer,由网络侧设备分开或统一配置,或基于默认确定,或协议预定义,或预设置,或实时设置。
可选地,终端收集的数据可以包括UE Data collection的数据。
可选地,网络侧设备收集的数据可以包括网络侧Data collection后的数据。
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括模型推荐/preference。
在一个实施例中,UE上报UE希望上报UE Data collection的数据,UE正在进行AI/ML Data collection的数据收集,以CSI压缩sub use case为例,即UE正在进行针对CSI压缩sub use case的AI/ML Data collection工作,则辅助信息上报流程可以包括:
步骤1:网络侧设备发送第一配置信息给UE,第一配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望上报收集到的针对信道状态信息(Channel State Information,CSI)压缩sub use case的数据时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望上报收集到的数据时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;前一个指示信息可以显示配置,也可以隐式表示,后一个指示信息显示配置;
配置信息2:第一定时器信息,比如ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望上报收集到的针对CSI压缩使用子场景(sub use case)的数据给网络侧设备时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第一子辅助信息,指示UE希望上报收集到的针对CSI压缩sub use case的数据给网络侧设备;
(2)第一子辅助信息,指示UE希望上报收集到的数据给网络侧设备+第六指示信息,指示收集到的数据是针对的哪个use case/sub use case,即这里为CSI压缩sub use case;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域波束(beam)预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备收到此UE上报的第一辅助信息后,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否请求UE上报收集到的针对CSI压缩sub use case的数据;
步骤4:若网络侧设备决策请求UE上报收集到的针对CSI压缩sub use case的数据,网络侧设备发送信令给UE来请求;
步骤5:UE根据网络侧设备请求,上报UE收集到的针对CSI压缩sub use case的数据给网络侧设备;
可选地,上述过程对于任何AI/ML使用场景(use case)/使用子场景(sub use case)均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这 个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望上报收集到的数据,无论是针对哪个/哪些use case或者哪个/哪些sub use case收集的数据,UE可以发一个整体的辅助信息来指示UE现在有收集的数据希望上报,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望网络侧设备发送Data collection后的数据给UE,UE需要数据集来完成模型训练/模型更新等工作,这里以CSI压缩sub use case为例,即UE需要数据集来完成CSI压缩sub use case模型的训练/更新等工作,辅助信息上报流程可以包括:
步骤1:网络侧设备发送第一配置信息给UE,第一配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望网络侧设备发送针对CSI压缩sub use case的Data collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新等工作时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;前一个指示信息可以显示配置,也可以隐式表示,后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望网络侧设备发送针对CSI压缩sub use case的Data  collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此辅助信息可以包括如下至少一项:
(1)第二子辅助信息,指示UE希望网络侧设备发送针对CSI压缩sub use case的Data collection后的数据给UE,用于UE进行CSI压缩sub use case模型的训练/更新等工作;
(2)第二子辅助信息,指示UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新等工作+第六指示信息,指示此Data collection数据是针对的哪个use case/sub use case(即这里为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息后,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否发送针对CSI压缩sub use case的Data collection的数据给UE;
步骤4:若网络侧设备决策发送针对CSI压缩sub use case的Data collection的数据给UE,网络侧设备发送针对CSI压缩sub use case的Data collection的数据给UE;
步骤5:UE接收网络侧设备发送的针对CSI压缩sub use case的Data collection的数据,并用收到的数据进行后续的针对CSI压缩的模型训练/更新工作;
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定 位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望网络侧设备发送针对某个/某些use case或某个/某些sub use case的Data collection后的数据给UE,用于UE进行某个/某些use case或某个/某些sub use case模型的训练/更新等工作,UE可以发一个整体的辅助信息来指示UE希望网络侧设备发送Data collection后的数据给UE,用于UE进行模型训练/更新,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型训练在UE侧/网络侧;以CSI压缩sub use case为例,辅助信息上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型训练在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型训练在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型训练在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧 设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第三子辅助信息,指示针对CSI压缩sub use case,UE希望模型训练在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧};
(2)第三子辅助信息,指示UE希望模型训练在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第三子辅助信息,指示针对CSI压缩sub use case,UE希望模型训练在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型训练在UE侧,另一个指示信息指示UE希望模型训练在网络侧;
(4)第三子辅助信息,指示UE希望模型训练在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型训练在UE侧,另一个指示信息指示UE希望模型训练在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型训练是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型训练在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等 use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型训练在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型训练在UE侧/网络侧,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型监测在UE侧/网络侧;以CSI压缩sub use case为例,包括如下步骤:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型监测在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型监测在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型监测在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧 设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第四子辅助信息,指示针对CSI压缩sub use case,UE希望模型监测在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}
(2)第四子辅助信息,指示UE希望模型监测在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第四子辅助信息,指示针对CSI压缩sub use case,UE希望模型监测在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型监测在UE侧,另一个指示信息指示UE希望模型监测在网络侧;
(4)第四子辅助信息,指示UE希望模型监测在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型监测在UE侧,另一个指示信息指示UE希望模型监测在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型监测是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型监测在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等 use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型监测在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型监测在UE侧/网络侧,并不针对特别的use case/sub use case。
在一个实施例中,UE上报UE希望模型更新在UE侧/网络侧;以CSI压缩sub use case为例,辅助信息上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示当UE希望针对CSI压缩sub use case的模型更新在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型更新在UE侧/网络侧,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望针对CSI压缩sub use case的模型更新在UE侧/网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一 辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第五子辅助信息,指示针对CSI压缩sub use case,UE希望模型更新在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}
(2)第五子辅助信息,指示UE希望模型更新在UE侧/网络侧,此指示信息可以为枚举类型,即{UE侧,网络侧}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第五子辅助信息,指示针对CSI压缩sub use case,UE希望模型更新在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型更新在UE侧,另一个指示信息指示UE希望模型更新在网络侧;
(4)第五子辅助信息,指示UE希望模型更新在UE侧/网络侧,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望模型更新在UE侧,另一个指示信息指示UE希望模型更新在网络侧+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策针对CSI压缩sub use case的模型更新是在UE侧还是网络侧;
步骤4:网络侧设备通知UE网络侧设备的决策结果,即模型更新在UE侧还是网络侧;
步骤4是可选的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定 位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case的模型更新在UE侧/网络侧,UE可以发一个整体的辅助信息来指示UE希望模型更新在UE侧/网络侧,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型变更/switch/激活/去激活,以CSI压缩sub use case为例,若UE侧存在5个AI/ML模型id=1,id=2,id=3,id=4,id=5用于CSI压缩,网络侧存在5个AI/ML模型id=6,id=7,id=8,id=9,id=10用于CSI解压缩。UE侧当前正在使用AI/ML模型id=1的模型进行CSI压缩工作,相对应的,网络侧使用AI/ML模型id=6的模型进行CSI解压缩工作,上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型变更/switch/激活/去激活时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型变更/switch/激活/去激活时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息 显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望将AI/ML模型变更/switch/激活/去激活时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第六子辅助信息,指示针对CSI压缩sub use case,UE希望将AI/ML模型变更/switch/激活/去激活;此指示信息是一个整体的1bit指示信息;
(2)第六子辅助信息,指示UE希望将AI/ML模型变更/switch/激活/去激活+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第二指示信息,比如一个cause值,比如可以指示模型性能较差或模型不适用等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:当网络侧设备收到UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否将AI/ML模型变更/switch/激活/去激活;
步骤4:若网络侧设备决策将AI/ML模型变更/switch/激活/去激活,网络侧设备发送指令给UE来指示UE将AI/ML模型变更/switch/激活/去激活;
步骤5:UE根据网络侧设备的指令将AI/ML模型变更/switch/激活/去激活;
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型更新,以CSI压缩sub use case为例,UE侧当前存储有用于CSI压缩的AI/ML模型,上报流程包括如下步骤:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
(1)配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型更新时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型更新时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
(2)配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望模型更新时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第八子辅助信息,指示针对CSI压缩sub use case,UE希望模型更新;
(2)第八子辅助信息,指示UE希望模型更新+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第三指示信息,比如cause值,比如可以指示模型性能较差,模型泛化能力较差,或模型不适用等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否进行模型更新;
步骤4:若网络侧设备决策进行模型更新,网络侧设备发送指示给UE指示UE进行模型更新;
步骤5:UE根据网络侧设备指示进行模型更新。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望对某个/某些use case或某个/某些sub use case的模型进行更新,UE可以发一个整体的辅助信息来指示UE希望模型更 新,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
在一个实施例中,UE上报UE希望AI/ML-based方式或非AI/ML-based方式,以CSI压缩sub use case为例,上报流程可以包括如下至少一项:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望使用AI/ML-based方式/非AI/ML-based方式时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望使用AI/ML-based方式/非AI/ML-based方式时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望使用AI/ML-based方式/非AI/ML-based方式时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一 辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第九子辅助信息,指示针对CSI压缩sub use case,UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息可以为枚举类型,如{AI/ML-based方式,非AI/ML-based方式}
(2)第九子辅助信息,指示UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息可以为枚举类型,如{AI/ML-based方式,非AI/ML-based方式}+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第九子辅助信息,指示针对CSI压缩sub use case,UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望使用AI/ML-based方式,另一个指示信息指示UE希望使用非AI/ML-based方式;
(4)第九子辅助信息,指示UE希望使用AI/ML-based方式/非AI/ML-based方式,此指示信息是两个分别单独的指示信息,一个指示信息指示UE希望使用AI/ML-based方式,另一个指示信息指示UE希望使用非AI/ML-based方式+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(5)第四指示信息,比如一个cause值,比如可以指示AI/ML模型性能较差,AI/ML模型性能较好,overhead太大,或overheating等。
其中,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策使用AI/ML-based方式还是非AI/ML-based方式;
步骤4:网络侧设备将决策结果通知UE(即使用AI/ML-based方式还是非AI/ML-based方式);
步骤4是可选执行的步骤。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望某个/某些use case或某个/某些sub use case使用AI/ML-based方式或非AI/ML-based方式,UE可以发一个整体的辅助信息来指示UE希望使用AI/ML-based方式或非AI/ML-based方式,并不针对特别的use case/sub use case。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
在一个实施例中,UE上报UE希望模型释放,以CSI压缩sub use case为例,UE侧当前存储有用于CSI压缩的AI/ML模型,上报流程可以包括:
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望模型释放时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望模型释放时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此 ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望模型释放时,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第十子辅助信息,指示针对CSI压缩sub use case,UE希望模型释放;
(2)第十子辅助信息,指示UE希望模型释放+第六指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
(3)第五指示信息,比如一个cause值,比如可以指示UE内存不足,或overhead太大等。
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE上报的第一辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否将模型释放;
步骤4:若网络侧设备决策将模型释放,网络侧设备发送指示给UE通知UE将模型释放;
步骤5:UE根据网络侧设备指示将模型释放。
可选地,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即 网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望将某个/某些use case或某个/某些sub use case的模型释放,UE可以发一个整体的辅助信息来指示UE希望模型释放,并不针对特别的use case/sub use case。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括终端期望或偏好的所述AI和/或ML的模型的模型标识信息;
可选地,终端期望或偏好的所述AI和/或ML的模型信息可以包括终端期望或偏好的所述AI或ML的模型优先级信息;
比如,第七子辅助信息所指示的模型优先级可以以模型标识信息来体现模型优先级信息,例如,模型优先级信息可以是一个模型标识信息列表list,如model id=1,model id=5,model id=6等,这样体现的模型优先级信息为model id=1的模型优先级高于model id=5的模型高于model id=6的模型。也可以是其它方式,这里不限制。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,模型标识信息可以为:Model id,functionality id,configuration id,模型参数,模型输入/输出等。
可选地,模型标识信息可以为:Model id,functionality id,configuration id,模型参数,模型输入/输出等。
在一个实施例中,UE上报模型推荐/preference;以CSI压缩sub use  case为例,若UE侧存在5个AI/ML模型id=1,id=2,id=3,id=4,id=5用于CSI压缩,网络侧存在5个AI/ML模型id=6,id=7,id=8,id=9,id=10用于CSI解压缩。UE侧当前正在使用AI/ML模型id=1的模型进行CSI压缩工作,相对应的,网络侧使用AI/ML模型id=6的模型进行CSI解压缩工作。
步骤1:网络侧设备发送配置信息给UE,配置信息包括如下信息中的一种或多种:
配置信息1:第一指示信息,用于指示针对CSI压缩sub use case,当UE希望上报模型推荐/preference时,UE可以上报辅助信息通知网络侧设备;此指示信息可以显示配置,也可以隐式表示,如当网络侧设备为此辅助信息的上报配置了一个ProhibitTimer值时,可以隐式的表示支持此辅助信息的上报;
配置信息1也可以为一个指示信息指示当UE希望上报模型推荐/preference时,UE可以上报辅助信息通知网络侧设备+一个指示信息指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);前一个指示信息可以显示配置,也可以隐式表示(方式可同上),后一个指示信息显示配置;
配置信息2:第一定时器信息,比如一个ProhibitTimer值,此ProhibitTimer用于避免UE频繁的上报此辅助信息;
可选地,步骤1可以提前配置给UE,没有时间限制;
可选地,步骤1为可选执行的步骤,即不为必须要执行的步骤。
步骤2:当UE希望将AI/ML模型从模型id为1的模型变更到模型id为3的模型,或者UE希望将AI/ML模型的优先级信息通知网络侧,如果配置了ProhibitTimer值,则UE开启ProhibitTimer,如果不存在ProhibitTimer,则跳过开启ProhibitTimer;并初始化上报辅助信息给网络侧设备,上报信令可以为第一辅助信息消息或引入一条RRC信令来上报第一辅助信息消息,此第一辅助信息可以包括如下至少一项:
(1)第八子辅助信息,比如一个模型id值,指示针对CSI压缩sub use  case,UE希望将当前的AI/ML模型变更为模型id=3的模型;
(2)第八子辅助信息,比如包括一个模型id值和第六指示信息,指示UE希望将当前的AI/ML模型变更为模型id=3的模型,指示针对的use case/sub use case是哪个,即此处为CSI压缩sub use case;
(3)第八子辅助信息,比如一个模型优先级信息,指示针对CSI压缩sub use case,UE希望的模型优先级信息,此模型优先级可以是一个list的形式,如模型优先级为:id=3,id=1,id=2,id=5,id=6,id=4;
(4)第八子辅助信息,比如包括一个模型优先级信息和第六指示信息,指示UE希望的模型优先级信息,此模型优先级可以是一个list的形式,如模型优先级为:id=3,id=1,id=2,id=5,id=6,id=4+一个指示信息,指示针对的use case/sub use case是哪个(即此处为CSI压缩sub use case);
可选地,模型标识信息除了此处的Model id,还可以为functionality id,configuration id,模型参数,模型输入/输出等;
可选地,步骤1和步骤2中的第六指示信息,指示针对的use case/sub use case是哪个,具体形式可以为枚举类型,如{CSI feedback enhancement,波束管理,定位精确度增强},或者{CSI压缩,时域CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位};
步骤3:网络侧设备接收UE发送的辅助信息,网络侧设备结合一些其它情况或信息,可以取决于网络侧设备实现,决策是否进行模型变更等;
步骤4:若网络侧设备决策进行模型变更,网络侧设备发送指令给UE通知UE进行模型变更;
步骤5:UE根据网络侧设备指令进行模型变更;
可选的,上述过程对于任何AI/ML use case/sub use case均适用,即上述过程也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。
可选地,上述过程也可以在UE侧使用AI/ML模型之前触发。可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
可选地,第一辅助信息对于任何一个AI/ML的use case或者sub use case均适用,可以是per use case的或者per sub use case的,或者整体的;对于per use case或per sub use case的情况,UE可以上报第二指示信息来指示此UE辅助信息针对的是哪个use case或哪个sub use case;
可选地,本公开各实施例中第一辅助信息的上报过程对于任何AI/ML use case/sub use case均适用,即也可以适用于CSI feedback enhancement,波束管理,定位精确度增强等use case,或者空频域CSI压缩,CSI预测,时域beam预测,空域beam预测,UE辅助定位,直接定位等sub use case,即这个配置信息和辅助信息是per use case或per sub use case的。或者这个配置信息和辅助信息也可以是针对整体的,即网络侧设备配置的指示信息和ProhibitTimer值是针对整体的,不区分use case/sub use case。当UE希望上报第一辅助信息,无论是针对哪个/哪些use case或者哪个/哪些sub use case收集的数据,UE可以发一个整体的辅助信息来指示UE现在有收集的数据希望上报,并不针对特别的use case/sub use case。
本公开实施例涉及的终端设备,可以是指向用户提供语音和/或数据连通性的设备,具有无线连接功能的手持式设备、或连接到无线调制解调器的其他处理设备等。在不同的系统中,终端设备的名称可能也不相同,例如在5G系统中,终端设备可以称为用户设备(User Equipment,UE)。无线终端设备可以经无线接入网(Radio Access Network,RAN)与一个或多个核心网(Core Network,CN)进行通信,无线终端设备可以是移动终端设备,如移动电话(或称为“蜂窝”电话)和具有移动终端设备的计算机,例如,可以是便携式、袖珍式、手持式、计算机内置的或者车载的移动装置,它们与无线接入网交换语言和/或数据。例如,个人通信业务(Personal Communication Service,PCS)电话、无绳电话、会话发起协议(Session  Initiated Protocol,SIP)话机、无线本地环路(Wireless Local Loop,WLL)站、个人数字助理(Personal Digital Assistant,PDA)等设备。无线终端设备也可以称为系统、订户单元(subscriber unit)、订户站(subscriber station),移动站(mobile station)、移动台(mobile)、远程站(remote station)、接入点(access point)、远程终端设备(remote terminal)、接入终端设备(access terminal)、用户终端设备(user terminal)、用户代理(user agent)、用户装置(user device),本公开实施例中并不限定。
本公开实施例涉及的网络设备,可以是基站,该基站可以包括多个为终端提供服务的小区。根据具体应用场合不同,基站又可以称为接入点,或者可以是接入网中在空中接口上通过一个或多个扇区与无线终端设备通信的设备,或者其它名称。网络设备可用于将收到的空中帧与网际协议(Internet Protocol,IP)分组进行相互更换,作为无线终端设备与接入网的其余部分之间的路由器,其中接入网的其余部分可包括网际协议(IP)通信网络。网络设备还可协调对空中接口的属性管理。例如,本公开实施例涉及的网络设备可以是全球移动通信系统(Global System for Mobile communications,GSM)或码分多址接入(Code Division Multiple Access,CDMA)中的网络设备(Base Transceiver Station,BTS),也可以是带宽码分多址接入(Wide-band Code Division Multiple Access,WCDMA)中的网络设备(NodeB),还可以是长期演进(long term evolution,LTE)系统中的演进型网络设备(evolutional Node B,eNB或e-NodeB)、5G网络架构(next generation system)中的5G基站(gNB),也可以是家庭演进基站(Home evolved Node B,HeNB)、中继节点(relay node)、家庭基站(femto)、微微基站(pico)等,本公开实施例中并不限定。在一些网络结构中,网络设备可以包括集中单元(centralized unit,CU)节点和分布单元(distributed unit,DU)节点,集中单元和分布单元也可以地理上分开布置。
图3是本公开实施例提供的一种网络侧设备的结构示意图,如图3所示,所述网络侧设备包括存储器320,收发机300,处理器310,其中:
存储器320,用于存储计算机程序;收发机300,用于在所述处理器 310的控制下收发数据;处理器310,用于读取所述存储器320中的计算机程序并执行以下操作:
接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
具体地,收发机300,用于在处理器310的控制下接收和发送数据。
其中,在图3中,总线架构可以包括任意数量的互联的总线和桥,具体由处理器310代表的一个或多个处理器和存储器320代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起,这些都是本领域所公知的,因此,本文不再对其进行进一步描述。总线接口提供接口。收发机300可以是多个元件,即包括发送机和接收机,提供用于在传输介质上与各种其他装置通信的单元,这些传输介质包括无线信道、有线信道、光缆等传输介质。处理器310负责管理总线架构和通常的处理,存储器320可以存储处理器310在执行操作时所使用的数据。
处理器310可以是中央处理器(Central Processing Unit,CPU)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或复杂可编程逻辑器件(Complex Programmable Logic Device,CPLD),处理器也可以采用多核架构。
可选地,在接收终端发送的第一辅助信息之前,处理器310还用于:
向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述 终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI 和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
在此需要说明的是,本公开实施例提供的上述网络侧设备,能够实现上述执行主体为网络侧设备的方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
图4是本公开实施例提供的一种终端的结构示意图,如图4所示,所述终端包括存储器420,收发机400,处理器410,其中:
存储器420,用于存储计算机程序;收发机400,用于在所述处理器410的控制下收发数据;处理器410,用于读取所述存储器420中的计算机程序并执行以下操作:
向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
具体地,收发机400,用于在处理器410的控制下接收和发送数据。
其中,在图4中,总线架构可以包括任意数量的互联的总线和桥,具体由处理器410代表的一个或多个处理器和存储器420代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起,这些都是本领域所公知的,因此,本文不再对其进行进一步描述。总线接口提供接口。收发机400可以是多个元件,即包括发送机和接收机,提供用于在传输介质上与各种其他装置通信的单元,这些传输介质包括无线信道、有线信道、光缆等传输介质。针对不同的用户设备,用户接口430还可以是能够外接内接需要设备的接口,连接的设备包括但不限于小键盘、显示器、扬声器、麦克风、操纵杆等。
处理器410负责管理总线架构和通常的处理,存储器420可以存储处理器410在执行操作时所使用的数据。
可选的,处理器410可以是中央处理器(Central Processing Unit,CPU)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或复杂可编程逻辑器件(Complex Programmable Logic Device,CPLD),处理器也可以采用多核架构。
处理器通过调用存储器存储的计算机程序,用于按照获得的可执行指令执行本公开实施例提供的任一所述方法。处理器与存储器也可以物理上分开布置。
可选地,在向所述网络侧设备发送所述第一辅助信息之前,所述处理器410还用于:
接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,所述向网络侧设备发送第一辅助信息前,处理器410还用于以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端侧或网络侧实现;
确定所述终端期望模型监测过程在所述终端侧或网络侧实现;
确定所述终端期望模型更新过程在所述终端侧或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行;
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
在此需要说明的是,本公开实施例提供的上述终端,能够实现上述执行主体为终端的方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
图5是本公开实施例提供的辅助信息上报装置的结构示意图之一,如图5所示,辅助信息上报装置500包括:
发送模块510,用于向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
本公开实施例提供的辅助信息上报装置,通过终端向网络侧设备发送第一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
可选地,所述装置还包括:
配置信息接收模块,用于在向所述网络侧设备发送所述第一辅助信息之前,接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,所述第一配置信息,包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,所述发送模块用于以下一项或多项:
启动第一定时器;
向网络侧设备发送第一辅助信息。
可选地,所述装置还包括:
执行模块,用于在所述向网络侧设备发送第一辅助信息前,执行以下一项或多项:
确定所述终端期望上报所述终端收集的数据;
确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
确定所述终端期望模型训练过程在所述终端侧或网络侧实现;
确定所述终端期望模型监测过程在所述终端侧或网络侧实现;
确定所述终端期望模型更新过程在所述终端侧或网络侧实现;
确定所述终端期望变更所述AI和/或ML的模型;
确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
确定所述终端期望更新所述AI和/或ML的模型;
确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
确定所述终端期望释放所述AI和/或ML的模型;
确定网络配置UE上报第一辅助信息;
确定第一定时器没有运行;
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述 模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入信息;
模型输出信息。
可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
图6是本公开实施例提供的辅助信息上报装置的结构示意图之二,如图6所示,辅助信息上报装置600包括:
接收模块610,用于接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
本公开实施例提供的辅助信息上报装置,通过终端向网络侧设备发送第一辅助信息,通过第一辅助信息帮助人工智能AI和/或机器学习ML的模型管理过程在无线通信系统中实现,提高AI和/或ML在无线通信系统中的实用性。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
可选地,所述装置还包括:
配置信息发送模块,用于在接收终端发送的第一辅助信息之前,向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
可选地,所述第一配置信息包括以下一项或多项:
第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
可选地,所述第一辅助信息包括以下一项或多项:
第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端侧实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端侧实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端侧实现,或所述第五子辅助信息用于表征所述终端期望所述 模型更新过程在所述网络侧设备实现;
第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
可选地,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
可选地,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
可选地,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
可选地,所述第七子辅助信息包括以下一项或多项:
所述终端期望或偏好的所述AI和/或ML的模型标识信息;
所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
可选地,所述模型标识信息,包括:
模型标识;
功能标识;
配置标识;
模型参数;
模型输入;
模型输出信息。
可选地,所述第一辅助信息包括:
第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
需要说明的是,本公开实施例中对单元的划分是示意性的,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。另外,在本公开各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个处理器可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)或处理器(processor)执行本公开各个实施例所 述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
在此需要说明的是,本公开实施例提供的上述装置,能够实现上述方法实施例所实现的所有方法步骤,且能够达到相同的技术效果,在此不再对本实施例中与方法实施例相同的部分及有益效果进行具体赘述。
另一方面,本公开实施例还提供一种处理器可读存储介质,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上述各实施例提供的方法。
所述处理器可读存储介质可以是处理器能够存取的任何可用介质或数据存储设备,包括但不限于磁性存储器(例如软盘、硬盘、磁带、磁光盘(MO)等)、光学存储器(例如CD、DVD、BD、HVD等)、以及半导体存储器(例如ROM、EPROM、EEPROM、非易失性存储器(NAND FLASH)、固态硬盘(SSD))等。
本领域内的技术人员应明白,本公开的实施例可提供为方法、系统、或计算机程序产品。因此,本公开可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本公开可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本公开是参照根据本公开实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机可执行指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机可执行指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些处理器可执行指令也可存储在能引导计算机或其他可编程数据处理 设备以特定方式工作的处理器可读存储器中,使得存储在该处理器可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些处理器可执行指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然,本领域的技术人员可以对本公开进行各种改动和变型而不脱离本公开的精神和范围。这样,倘若本公开的这些修改和变型属于本公开权利要求及其等同技术的范围之内,则本公开也意图包含这些改动和变型在内。

Claims (74)

  1. 一种辅助信息上报方法,应用于终端,所述方法包括:
    向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  2. 根据权利要求1所述的辅助信息上报方法,其中,在向所述网络侧设备发送所述第一辅助信息之前,所述方法还包括:
    接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
  3. 根据权利要求2所述的辅助信息上报方法,其中,所述第一配置信息,包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  4. 根据权利要求1所述的辅助信息上报方法,其中,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
    启动第一定时器;
    向网络侧设备发送第一辅助信息。
  5. 根据权利要求4所述的辅助信息上报方法,其中,所述向网络侧设备发送第一辅助信息前,还包括以下一项或多项:
    确定所述终端期望上报所述终端收集的数据;
    确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    确定所述终端期望模型训练过程在所述终端或网络侧实现;
    确定所述终端期望模型监测过程在所述终端或网络侧实现;
    确定所述终端期望模型更新过程在所述终端或网络侧实现;
    确定所述终端期望变更所述AI和/或ML的模型;
    确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
    确定所述终端期望更新所述AI和/或ML的模型;
    确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML 的方式;
    确定所述终端期望释放所述AI和/或ML的模型;
    确定网络配置UE上报第一辅助信息;
    确定第一定时器没有运行。
  6. 根据权利要求1-5任一项所述的辅助信息上报方法,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于 AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
  7. 根据权利要求6所述的辅助信息上报方法,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
  8. 根据权利要求6所述的辅助信息上报方法,其中,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  9. 根据权利要求6所述的辅助信息上报方法,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  10. 根据权利要求6所述的辅助信息上报方法,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  11. 根据权利要求6所述的辅助信息上报方法,其中,所述第七子辅助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  12. 根据权利要求11所述的辅助信息上报方法,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入信息;
    模型输出信息。
  13. 根据权利要求1-12任一项所述的辅助信息上报方法,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  14. 一种辅助信息上报方法,应用于网络侧设备,所述方法包括:
    接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  15. 根据权利要求14所述的辅助信息上报方法,其中,在接收终端发送的第一辅助信息之前,所述方法还包括:
    向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
  16. 根据权利要求15所述的辅助信息上报方法,其中,所述第一配置信息包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  17. 根据权利要求14-16任一项所述的辅助信息上报方法,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
  18. 根据权利要求17所述的辅助信息上报方法,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
  19. 根据权利要求17所述的辅助信息上报方法,其中,在所述第一辅 助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  20. 根据权利要求17所述的辅助信息上报方法,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  21. 根据权利要求17所述的辅助信息上报方法,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  22. 根据权利要求17所述的辅助信息上报方法,其中,所述第七子辅助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  23. 根据权利要求22所述的辅助信息上报方法,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入;
    模型输出信息。
  24. 根据权利要求14-23任一项所述的辅助信息上报方法,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  25. 一种终端,包括存储器,收发机,处理器:
    存储器,用于存储计算机程序;收发机,用于在所述处理器的控制下收发数据;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:
    向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  26. 根据权利要求25所述的终端,其中,在向所述网络侧设备发送所述第一辅助信息之前,所述操作还包括:
    接收网络侧设备发送的第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
  27. 根据权利要求26所述的终端,其中,所述第一配置信息,包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  28. 根据权利要求25所述的终端,其中,所述向网络侧设备发送第一辅助信息,包括以下一项或多项:
    启动第一定时器;
    向网络侧设备发送第一辅助信息。
  29. 根据权利要求28所述的终端,其中,所述向网络侧设备发送第一辅助信息前,所述操作还包括以下一项或多项:
    确定所述终端期望上报所述终端收集的数据;
    确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    确定所述终端期望模型训练过程在所述终端或网络侧实现;
    确定所述终端期望模型监测过程在所述终端或网络侧实现;
    确定所述终端期望模型更新过程在所述终端或网络侧实现;
    确定所述终端期望变更所述AI和/或ML的模型;
    确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
    确定所述终端期望更新所述AI和/或ML的模型;
    确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
    确定所述终端期望释放所述AI和/或ML的模型;
    确定网络配置UE上报第一辅助信息;
    确定第一定时器没有运行。
  30. 根据权利要求25-29任一项任一项所述的终端,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
  31. 根据权利要求30所述的终端,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
  32. 根据权利要求30所述的终端,其中,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  33. 根据权利要求30所述的终端,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  34. 根据权利要求30所述的终端,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  35. 根据权利要求30所述的终端,其中,所述第七子辅助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  36. 根据权利要求35所述的终端,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入信息;
    模型输出信息。
  37. 根据权利要求25-36任一项所述的终端,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  38. 一种网络侧设备,包括存储器,收发机,处理器:
    存储器,用于存储计算机程序;收发机,用于在所述处理器的控制下收发数据;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:
    接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  39. 根据权利要求38所述的网络侧设备,其中,在接收终端发送的第一辅助信息之前,所述操作还包括:
    向终端发送第一配置信息,所述第一配置信息为所述终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
  40. 根据权利要求39所述的网络侧设备,其中,所述第一配置信息包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  41. 根据权利要求38-40任一项所述的网络侧设备,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
  42. 根据权利要求41所述的网络侧设备,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因, 或所述终端期望去激活所述AI和/或ML模型的原因。
  43. 根据权利要求41所述的网络侧设备,其中,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  44. 根据权利要求41所述的网络侧设备,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  45. 根据权利要求41所述的网络侧设备,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  46. 根据权利要求41所述的网络侧设备,其中,所述第七子辅助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  47. 根据权利要求46所述的网络侧设备,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入;
    模型输出信息。
  48. 根据权利要求38-47任一项所述的网络侧设备,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  49. 一种辅助信息上报装置,包括:
    发送模块,用于向网络侧设备发送第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  50. 根据权利要求49所述的辅助信息上报装置,其中,所述装置还包括:
    配置信息接收模块,用于在向所述网络侧设备发送所述第一辅助信息之前,接收网络侧设备发送的第一配置信息,所述第一配置信息为终端向所述网络侧设备上报第一辅助信息所需要的配置信息。
  51. 根据权利要求50所述的辅助信息上报装置,其中,所述第一配置信息,包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  52. 根据权利要求49所述的辅助信息上报装置,其中,所述发送模块用于以下一项或多项:
    启动第一定时器;
    向网络侧设备发送第一辅助信息。
  53. 根据权利要求52所述的辅助信息上报装置,其中,所述装置还包括:
    执行模块,用于在所述向网络侧设备发送第一辅助信息前,执行以下一项或多项:
    确定终端期望上报终端收集的数据;
    确定所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    确定所述终端期望模型训练过程在所述终端或网络侧实现;
    确定所述终端期望模型监测过程在所述终端或网络侧实现;
    确定所述终端期望模型更新过程在所述终端或网络侧实现;
    确定所述终端期望变更所述AI和/或ML的模型;
    确定上报所述终端期望或偏好的所述AI和/或ML的模型信息;
    确定所述终端期望更新所述AI和/或ML的模型;
    确定所述终端期望使用基于AI和/或ML的方式或基于非AI和/或ML的方式;
    确定所述终端期望释放所述AI和/或ML的模型;
    确定网络配置UE上报第一辅助信息;
    确定第一定时器没有运行。
  54. 根据权利要求49-53任一项所述的辅助信息上报装置,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望所述网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述AI和/或ML的模型。
  55. 根据权利要求54所述的辅助信息上报装置,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
  56. 根据权利要求54所述的辅助信息上报装置,其中,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  57. 根据权利要求54所述的辅助信息上报装置,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  58. 根据权利要求54所述的辅助信息上报装置,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  59. 根据权利要求54所述的辅助信息上报装置,其中,所述第七子辅 助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  60. 根据权利要求59所述的辅助信息上报装置,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入信息;
    模型输出信息。
  61. 根据权利要求49-60任一项所述的辅助信息上报装置,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  62. 一种辅助信息上报装置,包括:
    接收模块,用于接收终端发送的第一辅助信息,所述第一辅助信息包括实现人工智能AI和/或机器学习ML的模型管理过程所需要的至少部分信息。
  63. 根据权利要求62所述的辅助信息上报装置,其中,所述装置还包括:
    配置信息发送模块,用于在接收终端发送的第一辅助信息之前,向终端发送第一配置信息,所述第一配置信息为所述终端向网络侧设备上报第一辅助信息所需要的配置信息。
  64. 根据权利要求63所述的辅助信息上报装置,其中,所述第一配置信息包括以下一项或多项:
    第一指示信息,所述第一指示信息用于指示所述网络侧设备支持第一辅助信息的上报;
    第一定时器信息,所述第一定时器用于控制终端上报第一辅助信息。
  65. 根据权利要求62-64任一项所述的辅助信息上报装置,其中,所述第一辅助信息包括以下一项或多项:
    第一子辅助信息,所述第一子辅助信息用于表征所述终端期望上报所述终端收集的数据;
    第二子辅助信息,所述第二子辅助信息用于表征所述终端期望网络侧设备发送所述网络侧设备收集的数据;
    第三子辅助信息,所述第三子辅助信息用于表征所述终端期望模型训练过程在所述终端实现,或所述第三子辅助信息用于表征所述终端期望所述模型训练过程在所述网络侧设备实现;
    第四子辅助信息,所述第四子辅助信息用于表征所述终端期望模型监测过程在所述终端实现,或所述第四子辅助信息用于表征所述终端期望所述模型监测过程在所述网络侧设备实现;
    第五子辅助信息,所述第五子辅助信息用于表征所述终端期望模型更新过程在所述终端实现,或所述第五子辅助信息用于表征所述终端期望所述模型更新过程在所述网络侧设备实现;
    第六子辅助信息,所述第六子辅助信息用于表征所述终端期望变更所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望激活所述AI和/或ML的模型,或所述第六子辅助信息用于表征所述终端期望去激活所述AI和/或ML的模型;
    第七子辅助信息,所述第七子辅助信息用于表征所述终端期望或偏好的所述AI和/或ML的模型信息;
    第八子辅助信息,所述第八子辅助信息用于表征所述终端期望更新所述AI和/或ML的模型;
    第九子辅助信息,所述第九子辅助信息用于表征所述终端期望使用基于AI和/或ML的方式,或所述第九子辅助信息用于表征所述终端期望使用基于非AI和/或ML的方式;
    第十子辅助信息,所述第十子辅助信息用于表征所述终端期望释放所述 AI和/或ML的模型。
  66. 根据权利要求65所述的辅助信息上报装置,其中,在所述第一辅助信息包括第六子辅助信息的情况下,所述第一辅助信息包括:
    第二指示信息,所述第二指示信息用于指示所述终端期望变更所述AI和/或ML模型的原因,或所述终端期望激活所述AI和/或ML模型的原因,或所述终端期望去激活所述AI和/或ML模型的原因。
  67. 根据权利要求65所述的辅助信息上报装置,其中,在所述第一辅助信息包括第八子辅助信息的情况下,所述第一辅助信息包括:
    第三指示信息,所述第三指示信息用于指示所述终端期望更新所述AI和/或ML模型的原因。
  68. 根据权利要求65所述的辅助信息上报装置,其中,在所述第一辅助信息包括第九子辅助信息的情况下,所述第一辅助信息包括:
    第四指示信息,所述第四指示信息用于指示所述终端期望使用基于AI和/或ML方式的原因,或所述第四指示信息用于指示所述终端期望使用基于非AI和/或ML方式的原因。
  69. 根据权利要求65所述的辅助信息上报装置,其中,在所述第一辅助信息包括第十子辅助信息的情况下,所述第一辅助信息包括:
    第五指示信息,所述第五指示信息用于指示所述终端期望释放所述AI和/或ML模型的原因。
  70. 根据权利要求65所述的辅助信息上报装置,其中,所述第七子辅助信息包括以下一项或多项:
    所述终端期望或偏好的所述AI和/或ML的模型标识信息;
    所述终端期望或偏好的所述AI和/或ML的模型优先级信息。
  71. 根据权利要求70所述的辅助信息上报装置,其中,所述模型标识信息,包括:
    模型标识;
    功能标识;
    配置标识;
    模型参数;
    模型输入;
    模型输出信息。
  72. 根据权利要求62-71任一项所述的辅助信息上报装置,其中,所述第一辅助信息包括:
    第六指示信息,所述第六指示信息用于指示所述第一辅助信息对应的一项或多项使用场景或使用子场景。
  73. 一种处理器可读存储介质,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13任一项所述的方法。
  74. 一种处理器可读存储介质,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14至24任一项所述的方法。
PCT/CN2023/124022 2022-12-30 2023-10-11 辅助信息上报方法及装置 WO2024139524A1 (zh)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211735444.0 2022-12-30

Publications (1)

Publication Number Publication Date
WO2024139524A1 true WO2024139524A1 (zh) 2024-07-04

Family

ID=

Similar Documents

Publication Publication Date Title
CN110505073B (zh) 一种移动边缘计算方法及装置
WO2024139524A1 (zh) 辅助信息上报方法及装置
US20240223474A1 (en) Method for reporting qoe measurement reports, and device, apparatus and storage medium
WO2022171154A1 (zh) 数据分析方法、装置、电子设备及存储介质
WO2024093606A1 (zh) 底层触发移动性的处理方法和装置
WO2024120277A1 (zh) 节能控制方法、装置及存储介质
WO2024125421A1 (zh) 一种模型同步方法、装置、设备及存储介质
WO2024067322A1 (zh) 候选小区配置的变更方法、装置、终端和网络设备
CN118282880A (zh) 辅助信息上报方法及装置
WO2024022405A1 (zh) 一种测量对齐的方法、终端及网络侧设备
WO2023207606A1 (zh) 成功切换报告的处理方法、装置及存储介质
WO2023202437A1 (zh) 多播业务的传输方法及装置
WO2024093897A1 (zh) 通信方法及装置
WO2024093406A1 (zh) 候选小区的切换方法和装置
WO2023005657A1 (zh) 信息发送方法、处理方法、终端、网络设备和存储介质
WO2024027383A1 (zh) 数据处理方法及装置
WO2024066399A1 (zh) 进入连接态的方法、装置、终端和网络设备
WO2024027450A1 (zh) 保持辅节点不变的方法及装置
WO2024093619A1 (zh) Ai/ml模型监测方法、设备、装置及存储介质
WO2024109393A1 (zh) 服务质量参数调整方法、装置及存储介质
WO2024022023A1 (zh) QoE测量配置的指示方法及装置
WO2024093656A1 (zh) 一种无线资源管理rrm测量的配置方法及装置
WO2024093750A1 (zh) 信息发送方法、设备、装置及存储介质
WO2024016969A1 (zh) L1/l2移动性命令的处理方法、装置及存储介质
WO2024120194A1 (zh) 算网融合方法、装置及存储介质