WO2024067098A1 - 模型信息上报方法、设备、装置及存储介质 - Google Patents
模型信息上报方法、设备、装置及存储介质 Download PDFInfo
- Publication number
- WO2024067098A1 WO2024067098A1 PCT/CN2023/118607 CN2023118607W WO2024067098A1 WO 2024067098 A1 WO2024067098 A1 WO 2024067098A1 CN 2023118607 W CN2023118607 W CN 2023118607W WO 2024067098 A1 WO2024067098 A1 WO 2024067098A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- model
- information
- network device
- terminal
- signaling
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 138
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 66
- 238000010801 machine learning Methods 0.000 claims abstract description 20
- 230000011664 signaling Effects 0.000 claims description 294
- 230000006870 function Effects 0.000 claims description 113
- 238000005259 measurement Methods 0.000 claims description 92
- 230000005540 biological transmission Effects 0.000 claims description 67
- 238000012549 training Methods 0.000 claims description 60
- 230000004913 activation Effects 0.000 claims description 53
- 230000009849 deactivation Effects 0.000 claims description 45
- 230000004044 response Effects 0.000 claims description 33
- 238000004590 computer program Methods 0.000 claims description 27
- 238000013507 mapping Methods 0.000 claims description 18
- 238000001994 activation Methods 0.000 description 46
- 238000010586 diagram Methods 0.000 description 24
- 238000005516 engineering process Methods 0.000 description 10
- 238000007726 management method Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000000691 measurement method Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 5
- 101100233916 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR5 gene Proteins 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010295 mobile communication Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 101001121408 Homo sapiens L-amino-acid oxidase Proteins 0.000 description 1
- 101000827703 Homo sapiens Polyphosphoinositide phosphatase Proteins 0.000 description 1
- 102100026388 L-amino-acid oxidase Human genes 0.000 description 1
- 102100023591 Polyphosphoinositide phosphatase Human genes 0.000 description 1
- 101100012902 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) FIG2 gene Proteins 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 235000019527 sweetened beverage Nutrition 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
- H04L5/003—Arrangements for allocating sub-channels of the transmission path
- H04L5/0048—Allocation of pilot signals, i.e. of signals known to the receiver
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W60/00—Affiliation to network, e.g. registration; Terminating affiliation with the network, e.g. de-registration
Definitions
- the present disclosure relates to the field of wireless communication technology, and in particular to a model information reporting method, device, apparatus and storage medium.
- the base station needs to cyclically send Tx beams in different directions, and the terminal uses Rx beams to receive Tx beams and measures the Channel State Information Reference Signal (CSI-RS) or Synchronization Signal Block (SSB) signals sent on all Tx beams to select the beam with the best receiving performance (such as Reference Signal Received Power (RSRP)).
- CSI-RS Channel State Information Reference Signal
- SSB Synchronization Signal Block
- the base station will then use the downlink beam to transmit information to the terminal.
- RSRP Reference Signal Received Power
- the CSI-RS or SSB signal needs to be sent on all Tx beams, which consumes a lot of resources; and the terminal needs to measure the CSI-RS or SSB signals sent on all Tx beams separately, which makes the terminal implementation more complicated and has a large measurement overhead.
- the present invention provides a model information reporting method, equipment, device and storage medium to support AI/ML technology to predict downlink beams.
- the present disclosure provides a model information reporting method, applied to a terminal, comprising:
- first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, where the AI/ML models are used for downlink beam prediction;
- the first information is sent to a network device.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- sending the first information to the network device includes:
- the first signaling includes one or more of the following:
- sending the first information to the network device includes:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- terminal capability information is sent to the network device, where the terminal capability information includes the first information.
- sending the first information to the network device includes:
- UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- sending the first information to the network device includes:
- RRC request message sent by a network device, wherein the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- an RRC response message is sent to the network device, where the RRC response message includes the first information.
- sending the first information to the network device includes:
- the measurement configuration message includes one or more of a measurement configuration, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- RRC signaling, UCI signaling, or signaling for transmitting data information is sent to the network device, where the RRC signaling, UCI signaling, or signaling for transmitting data information includes the first information.
- sending the first information to the network device includes:
- model registration trigger message sent by a network device, wherein the model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- model registration information is sent to the network device, where the model registration information includes the first information.
- the method further includes:
- the method further comprises:
- RRC signaling or media access control layer-control unit MAC-CE sent by a network device, where the RRC signaling or MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication;
- a model activation or deactivation or fallback operation is performed.
- the method further includes:
- model identifier allocation message sent by the network device, wherein the model identifier allocation message includes a model identifier allocated by the network device;
- the model identifier allocated by the network device is associated with the model indicated in the first information.
- the present disclosure further provides a model information reporting method, which is applied to a network device, comprising:
- the first information sent by the receiving terminal is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- the receiving terminal sends the first information, including:
- the first signaling includes one or more of the following:
- the receiving terminal sends the first information, including:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- the receiving terminal sends the first information, including:
- the UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- the receiving terminal sends the first information, including:
- the RRC request message contains the application scenario, network One or more of device configuration, model function, and maximum number of models for feedback;
- An RRC response message sent by the terminal is received, where the RRC response message includes the first information.
- the receiving terminal sends the first information, including:
- the measurement configuration message includes one or more of measurement configuration, application scenario, network device configuration, model function, and maximum number of feedback models;
- Receive RRC signaling, UCI signaling, or signaling for transmitting data information sent by the terminal where the RRC signaling, UCI signaling, or signaling for transmitting data information includes the first information.
- the receiving terminal sends the first information, including:
- model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- Model registration information sent by the terminal is received, where the model registration information includes the first information.
- the method after receiving the first information sent by the terminal, the method further includes:
- An RRC configuration message is sent to the terminal according to the first information, where the RRC configuration message includes one or more model identifiers.
- the method further comprises:
- An RRC signaling or a media access control layer-control unit MAC-CE is sent to the terminal, wherein the RRC signaling or the MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication.
- the method after receiving the first information sent by the terminal, the method further includes:
- a model identifier allocation message is sent to the terminal according to the first information, wherein the model identifier allocation message includes the model identifier allocated by the network device.
- the present disclosure further provides a terminal, including a memory, a transceiver, and a processor;
- 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:
- first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, where the AI/ML models are used for downlink beam prediction;
- the first information is sent to a network device.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- sending the first information to the network device includes:
- the first signaling includes one or more of the following:
- sending the first information to the network device includes:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- terminal capability information is sent to the network device, where the terminal capability information includes the first information.
- sending the first information to the network device includes:
- UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- sending the first information to the network device includes:
- RRC request message sent by a network device, wherein the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- an RRC response message is sent to the network device, where the RRC response message includes the first information.
- sending the first information to the network device includes:
- the measurement configuration message includes one or more of a measurement configuration, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- RRC signaling, UCI signaling, or signaling for transmitting data information is sent to the network device, where the RRC signaling, UCI signaling, or signaling for transmitting data information includes the first information.
- sending the first information to the network device includes:
- model registration trigger message sent by a network device, wherein the model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- model registration information is sent to the network device, where the model registration information includes the first information.
- the operation after sending the first information to the network device, the operation further includes:
- the operations further include:
- RRC signaling or media access control layer-control unit MAC-CE sent by a network device, where the RRC signaling or MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication;
- a model activation or deactivation or fallback operation is performed.
- the operation after sending the first information to the network device, the operation further includes:
- model identifier allocation message sent by the network device, wherein the model identifier allocation message includes a model identifier allocated by the network device;
- the model identifier allocated by the network device is associated with the model indicated in the first information.
- the present disclosure further provides a network device, including a memory, a transceiver, and a processor;
- 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 information sent by the receiving terminal is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- the receiving terminal sends the first information, including:
- the first signaling includes one or more of the following:
- the receiving terminal sends the first information, including:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- the receiving terminal sends the first information, including:
- the UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- the receiving terminal sends the first information, including:
- the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- An RRC response message sent by the terminal is received, where the RRC response message includes the first information.
- the receiving terminal sends the first information, including:
- the measurement configuration message includes one or more of measurement configuration, application scenario, network device configuration, model function, and maximum number of feedback models;
- Receive RRC signaling, UCI signaling, or signaling for transmitting data information sent by the terminal where the RRC signaling, UCI signaling, or signaling for transmitting data information includes the first information.
- the receiving terminal sends the first information, including:
- model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- Model registration information sent by the terminal is received, where the model registration information includes the first information.
- the operation after receiving the first information sent by the terminal, the operation further includes:
- An RRC configuration message is sent to the terminal according to the first information, where the RRC configuration message includes one or more model identifiers.
- the operations further include:
- An RRC signaling or a media access control layer-control unit MAC-CE is sent to the terminal, wherein the RRC signaling or the MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication.
- the operation after receiving the first information sent by the terminal, the operation further includes:
- a model identifier allocation message is sent to the terminal according to the first information, wherein the model identifier allocation message includes the model identifier allocated by the network device.
- the present disclosure further provides a model information reporting device, including:
- a determining unit configured to determine first information, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, where the AI/ML model is used for downlink beam prediction;
- the first sending unit is used to send the first information to the network device.
- the present disclosure further provides a model information reporting device, including:
- the third receiving unit is used to receive first information sent by the terminal, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- the present disclosure also provides a computer-readable storage medium, which stores a computer program, and the computer program is used to enable a computer to execute the model information reporting method described in the first aspect, or execute the model information reporting method described in the second aspect.
- the present disclosure further provides a communication device, in which a computer program is stored, and the computer program is used to enable the communication device to execute the model information reporting method described in the first aspect, or execute the model information reporting method described in the second aspect.
- the present disclosure also provides a processor-readable storage medium, which stores a computer program, and the computer program is used to enable the processor to execute the model information reporting method described in the first aspect as described above, or execute the model information reporting method described in the second aspect as described above.
- the present disclosure further provides a chip product, in which a computer program is stored, and the computer program is used to enable the chip product to execute the model information reporting method described in the first aspect as described above, or execute the model information reporting method described in the second aspect as described above.
- the model information reporting method, device, apparatus and storage medium provided in the present invention enable a terminal to report relevant information of one or more AI/ML models to a network device through first information, so that the network device can subsequently activate the use of the model and configure the Tx beam sending pattern as needed, thereby improving the system transmission performance and better supporting the use of AI/ML technology for beam management and beam prediction.
- FIG1 is a flow chart of a method for reporting model information according to an embodiment of the present disclosure
- FIG2 is a second flow chart of the model information reporting method provided in an embodiment of the present disclosure.
- FIG3 is a schematic diagram of one implementation of the model information reporting method provided in an embodiment of the present disclosure.
- FIG4 is a second schematic diagram of the implementation of the model information reporting method provided in the embodiment of the present disclosure.
- FIG5 is a third schematic diagram of the implementation of the model information reporting method provided in the embodiment of the present disclosure.
- FIG6 is a fourth implementation diagram of the model information reporting method provided in an embodiment of the present disclosure.
- FIG. 7 is a fifth implementation diagram of the model information reporting method provided in an embodiment of the present disclosure.
- FIG8 is a schematic diagram of the structure of a terminal provided in an embodiment of the present disclosure.
- FIG9 is a schematic diagram of the structure of a network device provided in an embodiment of the present disclosure.
- FIG10 is a schematic diagram of a structure of a model information reporting device according to an embodiment of the present disclosure.
- FIG. 11 is a second schematic diagram of the structure of the model 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.
- One is to measure the RSRP of all downlink beam pairs (a beam pair consisting of a Tx beam of the base station and an Rx beam of the UE), and the beam pair corresponding to the maximum RSRP is the optimal beam pair, and the base station is informed of the beam pair.
- the base station will use the beam pair in the future.
- the Tx beam in the beam pair sends information to the UE, and the UE receives information using the Rx beam in the beam pair.
- the UE fixes or selects the best DL Rx beam, receives and measures the received power RSRP of all DL Tx beams sent by the base station, and the Tx beam corresponding to the largest RSRP is the optimal Tx beam, and informs the base station.
- the base station subsequently uses this Tx beam to send information to the UE.
- Another method is that the base station fixes or selects the best DL Tx beam, and the UE uses all DL Rx beams to receive and measure the received power RSRP of the DL Tx beam sent by the base station.
- the Rx beam corresponding to the maximum RSRP is the optimal Rx beam.
- the base station subsequently uses the Tx beam to send information to the UE, the UE uses the optimal Rx beam to receive.
- the UE needs to use all Rx beams to receive the CSI-RS/SSB of each Tx beam sent by the base station for measurement.
- the UE fixes or selects the best Rx beam and measures all Tx beams sent by the base station. It needs to measure 32 Tx beams to calculate the best Tx beam. The same problem also exists for the DL Rx beam measurement method.
- the present disclosure proposes to use artificial intelligence (AI) or machine learning (AI) technology to predict the downlink beam.
- AI artificial intelligence
- AI machine learning
- the present disclosure can also be used in other scenarios, such as measuring other wide beams that are easy to measure for sending SSB, or beams in other frequency bands, so as to predict the narrow beam that sends CSI-RS, or predict the best beam in the high-frequency beam; for example, using the previously measured beam measurement results, Predict the beam (pair) with the best reception performance among the Tx beams or beam pairs that the base station will send in the future; for example, in DL Rx beam prediction, the base station fixes or selects the best Tx beam, continuously sends the reference signal (RS), and the UE uses different DL Rx beams to receive and measure RS, and predicts the best Rx beam based on the measurement results.
- This can save RS transmission resources, UE measurement overhead, and reduce UE measurement delay.
- the present disclosure provides a clear solution for how the UE reports model information to the base station when using AI/ML technology for beam management and prediction.
- FIG. 1 is a flow chart of a method for reporting model information provided by an embodiment of the present disclosure. The method can be applied to a terminal. As shown in FIG. 1 , the method includes the following steps:
- Step 100 determine first information, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- Step 101 Send first information to a network device.
- the terminal in order to better support the use of AI/ML technology for beam management and beam prediction, can report first information to the network device (such as a base station), and the first information can be used to indicate one or more AI/ML models, which are AI/ML models that can be used for downlink beam prediction (hereinafter referred to as models).
- the downlink beam refers to the downlink beam used for transmitting information between the network device and the terminal, including a downlink transmit beam (Tx beam), a downlink receive beam (Rx beam), a downlink beam pair, etc.
- the network device After the network device receives the first information sent by the terminal, it can learn which models the terminal can use and related information of these models, and can subsequently perform model selection and activation processes based on this information when needed.
- the model indicated in the first information may be a model stored in the terminal itself, or may be a model stored in a third-party device, which is not limited here.
- the terminal can report the model information stored by the third-party device to the network device, and then download the corresponding model from the third-party device for use according to the instructions of the network device (such as configuration, activation or selection commands).
- the first information may include one or more of the following:
- the first information may include a model identifier, which may be a serial number or serial number of the model, or identification information representing the model type, or other identification information that can distinguish different models.
- a model identifier which may be a serial number or serial number of the model, or identification information representing the model type, or other identification information that can distinguish different models.
- the model identification may include one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- the dataset identifier used for model training, the RS configuration identifier used for model training, the beam description information identifier used for model training, and the number of input beams and the number of output beams of the model can all be understood as identification information representing the model type, which are specifically described as follows:
- a dataset is a dataset used for model training.
- the terminal can train different models through different datasets. If the dataset and the trained model are one-to-one corresponding, the dataset identifier can represent the model identifier.
- a dataset may include a dataset identifier and one or more dataset samples, where the dataset samples may include: multiple beam identifiers, and the measurement results of the beams corresponding to the beam identifiers, and one or more optimal beam identifiers (as output labels for AI/ML model training).
- RS configuration is the RS configuration used for model training.
- the terminal can measure the downlink reference signal based on the RS configuration and use the measurement results to train the model. If the RS configuration and the trained model are one-to-one corresponding, the RS configuration identifier can represent the model identifier.
- the RS configuration may include the RS configuration identifier, RS type, RS transmission resource, RS identifier, etc.
- Beam description information identifier The terminal can use the beam description information and the result of measuring the downlink reference signal to train the model. If the beam description information and the trained model are one-to-one corresponding, the beam description information identifier can represent the model identifier.
- the beam description information may include a beam identifier or an RS identifier, the network device antenna configuration of each beam identifier or RS identifier corresponding to the beam, the angle information of the beam, the width information of the beam, etc.
- the number of input beams and the number of output beams of the model may be a pair of ⁇ input beam number, output beam number ⁇ information, such as a pair of ⁇ input Tx beam number, output Tx beam number ⁇ information, or a pair of ⁇ input Rx beam number, output Rx beam number ⁇ information.
- the terminal may train only one model for each ⁇ input beam number, output beam number ⁇ pair, and the ⁇ input beam number, output beam number ⁇ pair may represent the model.
- the first information may not include a model identifier.
- the method further includes:
- model identifier allocation message sent by a network device, wherein the model identifier allocation message includes a model identifier allocated by the network device;
- the model identifier allocated by the network device is associated with the model indicated in the first information.
- the network device may assign corresponding model identifiers to the models reported by the terminal according to the model information reported by the terminal.
- the network device may send a model identifier assignment message to the terminal, and the model identifier assignment message includes the model identifier assigned by the network device.
- the model identifier assignment message may also include model information corresponding to each model identifier (one or more items in the first information), or other indication information, which may be used to indicate the association relationship between the assigned model identifier and the model indicated by the first information, so that the terminal can associate the model identifier assigned by the network device with the model indicated in the first information according to the model identifier assignment message.
- the model identifier assignment message includes a model identifier assigned by a network device, but does not include explicit indication information indicating the association between the assigned model identifier and the model indicated by the first information.
- the terminal can, by default, associate the model identifier assigned by the network device with the model indicated in the first information in the order of the model information in the first information reported by it.
- the terminal can report the scenarios to which the model is applicable.
- the application scenarios may include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma (urban macro cell), Umi (urban micro cell), etc.
- the application scenario of the model reported by the terminal can be used by subsequent network devices to select and activate the terminal to use the appropriate model for beam prediction according to the application scenario.
- the terminal may report the network device configuration applicable to the model.
- the network device configuration may include: antenna configuration, beam configuration, reference signal configuration, etc. of the network device.
- the network device configuration applicable to the model reported by the terminal can be used by subsequent network devices to select and activate the terminal to use an appropriate model for beam prediction based on the network device configuration.
- the terminal may report the functions of the model, for example, the model functions may include: spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Spatial domain prediction refers to measuring a small number of beams, or beams of other reference signal types, and predicting the optimal beam among a large number of beams;
- frequency domain prediction refers to measuring a beam on frequency 1 and predicting the optimal beam among beams on another frequency 2;
- time domain prediction refers to measuring the beam at the present moment and predicting the optimal beam at the future moment.
- the terminal reports the model function, which can be used by subsequent network devices to select and activate the terminal to use the appropriate model for beam prediction based on the model function.
- the terminal can report the number of input beams of the model, such as the number of input Tx beams of the model, or the number of input Rx beams of the model, or the number of input beam pairs of the model, etc.
- the terminal reports the number of input beams of the model, which makes it easier for the network device to configure and send Tx beams after selecting the model used by the terminal, thereby improving the terminal prediction performance.
- the terminal can report the number of output beams of the model, such as the number of output Tx beams of the model, or the number of output Rx beams of the model, or the number of output beam pairs of the model, etc.
- the terminal reports the number of output beams of the model, so that the network device can subsequently schedule the terminal to provide feedback on the optimal beam output by the model, such as indicating the number of Tx beams fed back by the terminal.
- the terminal can report the downlink reference signal type applicable to the model.
- the types may include: CSI-RS, SSB, Phase-Tracking Reference Signal (PT-RS), Cell Reference Signal (CRS), Demodulation Reference Signal (DMRS), etc.
- the terminal can report the downlink beam sending order applicable to the model, so that the network equipment can configure and send the Tx beam after selecting the model used by the terminal, thereby improving the beam prediction performance.
- the terminal can report the downlink beam pattern applicable to the model, so that the network equipment can configure and send the Tx beam after selecting the model used by the terminal, thereby improving the beam prediction performance.
- the terminal can report the mapping relationship between the downlink beam identifier applicable to the model and the physical downlink beam, which can be used to enable the network device to understand the beam identifier (such as Tx beam ID) reported by the terminal after predicting the optimal beam, and send subsequent data information on its corresponding physical beam.
- the beam identifier such as Tx beam ID
- the terminal can report the identifier of the downlink transmit beam to which the model applies.
- the terminal can recommend which Tx beam the network device should send by reporting the identifier of the downlink transmit beam to which the model applies, so that the terminal can use different Rx beams to receive the reference signal sent by the Tx beam for measurement, and predict the optimal Rx beam based on the measurement results.
- the terminal can report the number of transmissions of the downlink transmit beam applicable to the model.
- the terminal can report the number of transmissions of the downlink transmit beam applicable to the model, the Tx beam sent by the recommended network device, and the number of times the reference signal is sent on the Tx beam, so that the terminal can use different Rx beams to receive the reference signal sent by the Tx beam for measurement, and predict the optimal Rx beam based on the measurement results.
- the model information reporting method provided in the embodiment of the present disclosure can report the relevant information of one or more AI/ML models to the network device through the first information, so that the network device can subsequently activate the use of the model and configure the sending pattern of the Tx beam as needed, thereby improving the transmission performance of the system and being able to Better support for beam management and beam prediction using AI/ML techniques.
- the terminal sending the first information to the network device may include:
- the first signaling may include one or more of the following:
- Terminal capability signaling.
- the terminal may report the first information via terminal capability information (UE capability Information).
- UE capability Information terminal capability information
- the UCI can be the UCI transmitted on the Physical Uplink Control Channel (PUCCH) or the Physical Uplink Shared Channel (PUSCH).
- PUCCH Physical Uplink Control Channel
- PUSCH Physical Uplink Shared Channel
- RRC Radio Resource Control
- the terminal may report the first information to the network device through the signaling for transmitting data information.
- the terminal may report the first information to the network device through the signaling for model registration.
- sending first information to a network device includes:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- terminal capability information is sent to the network device, where the terminal capability information includes first information.
- the network device can send a terminal capability query (UECapabilityEnquiry) message to the terminal to query the terminal for terminal capability information.
- the terminal capability query message may include one or more of the application scenario, network device configuration, model function, and the maximum number of feedback models.
- the maximum number of models fed back refers to the maximum number of models that the terminal can feed back.
- the maximum number of models fed back is 6, which means that the terminal can report information of up to 6 models.
- the terminal After receiving the terminal capability query message sent by the network device, the terminal can feed back information of the corresponding model in the terminal capability information (UECapabilityInformation) according to the terminal capability query message.
- UECapabilityInformation information of the corresponding model in the terminal capability information
- the terminal can upload the application scenario.
- Corresponding model information model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- sending first information to a network device includes:
- UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- the UCI is sent to the network device, where the UCI includes the first information.
- the network device can send a UCI feedback resource configuration message (for example, an RRC configuration message) to the terminal to configure the UCI feedback resources of the terminal.
- the UCI feedback resource configuration message may include one or more of the UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models.
- the terminal may feedback UCI on the configured UCI feedback resource according to the UCI feedback resource configuration message, which includes information of the corresponding model.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- sending first information to a network device includes:
- RRC request message sent by a network device, where the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- an RRC response message is sent to the network device, where the RRC response message includes first information.
- the network device may send an RRC request message to the terminal, for example, requesting the terminal to report model information.
- the RRC request message may include one or more of the application scenario, network device configuration, model function, and the maximum number of models for feedback.
- the terminal After receiving the RRC request message sent by the network device, the terminal can feed back information of the corresponding model through an uplink RRC message according to the RRC request message.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- sending first information to a network device includes:
- the measurement configuration message includes one or more of a measurement configuration, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- RRC signaling, UCI signaling, or signaling for transmitting data information is sent to the network device, where the RRC signaling, UCI signaling, or signaling for transmitting data information includes first information.
- the network device can send a measurement configuration message to the terminal, which may include one or more of the following: measurement configuration (such as Tx beam configuration, reference signal configuration, etc.), application scenario, network device configuration, model function, and maximum number of feedback models.
- measurement configuration such as Tx beam configuration, reference signal configuration, etc.
- application scenario such as Tx beam configuration, reference signal configuration, etc.
- network device configuration such as Tx beam configuration, reference signal configuration, etc.
- model function such as a maximum number of feedback models.
- the terminal After receiving the measurement configuration message sent by the network device, the terminal can feedback the information of the corresponding model according to the measurement configuration message.
- the terminal can feedback the information of one or more models using RRC signaling, UCI signaling or signaling for transmitting data information.
- the terminal can determine the application scenario or network device configuration based on the measurement configuration message, thereby selecting one or more models suitable for the existing application scenario or network device configuration, and reporting their information to the network device, thereby reducing the amount of model information reported by the terminal.
- sending first information to a network device includes:
- model registration trigger message sent by a network device, where the model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- model registration information is sent to the network device, where the model registration information includes first information.
- the network device can send a model registration trigger message to the terminal to trigger model registration.
- the model registration trigger message may include one or more of the application scenario, network device configuration, model function, and the maximum number of models for feedback.
- the terminal After receiving the model registration trigger message sent by the network device, the terminal can send model registration information to the network device according to the model registration trigger message, including information of the corresponding model.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- the method further includes:
- the network device after the network device receives the first information reported by the terminal, it can configure the model available to the terminal through RRC signaling (such as an RRC configuration message) according to the model information reported by the terminal, and the configuration information can include a model identifier.
- RRC signaling such as an RRC configuration message
- the terminal after the terminal receives the RRC configuration message sent by the network device, it can determine which model or models are used for downlink beam prediction.
- the network device may select a model corresponding to the current application scenario or network device configuration based on the current application scenario or network device configuration, and configure the model identifier of the corresponding model to the terminal.
- the network device can use RRC signaling to configure a group of models, and then activate one of the models through the Media Access Control-Control Element (MAC-CE) for optimal beam prediction; or RRC signaling only configures one model for optimal beam prediction; or RRC signaling configures a group of models, and the terminal selects one or more models for optimal beam prediction.
- MAC-CE Media Access Control-Control Element
- the method further comprises:
- RRC signaling or media access control layer-control unit MAC-CE sent by the network device, where the RRC signaling or MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication;
- the model activation, deactivation or fallback operation is performed.
- the network device may select to activate one or more models based on the current application scenario or network device configuration, and may specifically activate through RRC signaling or through MAC-CE.
- the activation signaling may include an identifier of the activation model and an activation indication.
- the network device when the model performance is poor (for example, compared with the performance of other models, or compared with the performance of legacy methods), the network device chooses to switch to other models. Either deactivate this model or fallback to the legacy method, which can be done through RRC activation/deactivation/fallback or MAC CE activation/deactivation/fallback.
- the activation/deactivation signaling includes the identifier of the activation/deactivation model and the activation/deactivation indication; the fallback signaling includes the fallback indication.
- the activation/deactivation/fallback indication may be a 1-bit indication.
- FIG. 2 is a second flow chart of a method for reporting model information provided by an embodiment of the present disclosure.
- the method can be applied to a network device (such as a base station). As shown in FIG. 2 , the method includes the following steps:
- Step 200 Receive first information sent by the terminal, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- the terminal in order to better support the use of AI/ML technology for beam management and beam prediction, the terminal can report first information to the network device, and the first information can be used to indicate one or more AI/ML models, which are AI/ML models that can be used for downlink beam prediction (hereinafter referred to as models).
- the downlink beam refers to the downlink beam used for transmitting information between the network device and the terminal, including a downlink transmit beam (Tx beam), a downlink receive beam (Rx beam), a downlink beam pair, etc.
- the network device After the network device receives the first information sent by the terminal, it can learn which models the terminal can use and related information of these models, and can subsequently perform model selection and activation processes based on this information when needed.
- the model indicated in the first information may be a model stored in the terminal itself, or may be a model stored in a third-party device, which is not limited here.
- the terminal can report the model information stored by the third-party device to the network device, and then download the corresponding model from the third-party device for use according to the instructions of the network device (such as configuration, activation or selection commands).
- the first information may include one or more of the following:
- the first information may include a model identifier, which may be a serial number or serial number of the model, or identification information representing the model type, or other identification information that can distinguish different models.
- a model identifier which may be a serial number or serial number of the model, or identification information representing the model type, or other identification information that can distinguish different models.
- the model identification may include one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- the dataset identifier used for model training, the RS configuration identifier used for model training, the beam description information identifier used for model training, and the number of input beams and the number of output beams of the model can all be understood as identification information representing the model type, which are specifically described as follows:
- a dataset is a dataset used for model training.
- the terminal can train different models through different datasets. If the dataset and the trained model are one-to-one corresponding, the dataset identifier can represent the model identifier.
- a dataset may include a dataset identifier and one or more dataset samples, where the dataset samples may include: multiple beam identifiers, and the measurement results of the beams corresponding to the beam identifiers, and one or more optimal beam identifiers (as output labels for AI/ML model training).
- RS configuration is the RS configuration used for model training.
- the terminal can measure the downlink reference signal based on the RS configuration and use the measurement results to train the model. If the RS configuration and the trained model are one-to-one corresponding, the RS configuration identifier can represent the model identifier.
- the RS configuration may include the RS configuration identifier, RS type, RS transmission resource, RS identifier, etc.
- Beam description information identifier The terminal can use the beam description information and the result of measuring the downlink reference signal to train the model. If the beam description information and the trained model are one-to-one corresponding, the beam description information identifier can represent the model identifier.
- the beam description information may include a beam identifier or an RS identifier, the network device antenna configuration of each beam identifier or RS identifier corresponding to the beam, the angle information of the beam, the width information of the beam, etc.
- Model input beam number and output beam number information may be ⁇ input beam number, output beam number ⁇ pair information, such as ⁇ input Tx beam number, output Tx beam number ⁇ pair information, or ⁇ input Rx beam number, output Rx beam number ⁇ pair information.
- the terminal may generate a corresponding ⁇ input beam number, output beam number ⁇ pair for each ⁇ input beam number, output beam number ⁇ pair. If only one model is trained, then the pair ⁇ input beam number, output beam number ⁇ can represent the model.
- the first information may not include a model identifier.
- the method further includes:
- a model identifier allocation message is sent to the terminal, where the model identifier allocation message includes a model identifier allocated by the network device.
- the network device may assign corresponding model identifiers to the models reported by the terminal according to the model information reported by the terminal.
- the network device may send a model identifier assignment message to the terminal, and the model identifier assignment message includes the model identifier assigned by the network device.
- the model identifier assignment message may also include model information corresponding to each model identifier (one or more items in the first information), or other indication information, which may be used to indicate the association relationship between the assigned model identifier and the model indicated by the first information, so that the terminal can associate the model identifier assigned by the network device with the model indicated in the first information according to the model identifier assignment message.
- the model identifier assignment message includes a model identifier assigned by a network device, but does not include explicit indication information indicating the association between the assigned model identifier and the model indicated by the first information.
- the terminal can, by default, associate the model identifier assigned by the network device with the model indicated in the first information in the order of the model information in the first information reported by it.
- the terminal can report the scenarios to which the model is applicable.
- the application scenarios may include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma, Umi, etc.
- the application scenario of the model reported by the terminal can be used by subsequent network devices to select and activate the terminal to use the appropriate model for beam prediction according to the application scenario.
- the terminal may report the network device configuration applicable to the model.
- the network device configuration may include: antenna configuration, beam configuration, reference signal configuration, etc. of the network device.
- the terminal reports the applicable network device configuration of the model, which can be used for subsequent network devices according to the network device Configuration selection activates the terminal to use the appropriate model for beam prediction.
- the terminal may report the functions of the model, for example, the model functions may include: spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Spatial domain prediction refers to measuring a small number of beams, or beams of other reference signal types, and predicting the optimal beam among a large number of beams;
- frequency domain prediction refers to measuring a beam on frequency 1 and predicting the optimal beam among beams on another frequency 2;
- time domain prediction refers to measuring the beam at the present moment and predicting the optimal beam at the future moment.
- the terminal reports the model function, which can be used by subsequent network devices to select and activate the terminal to use the appropriate model for beam prediction based on the model function.
- the terminal can report the number of input beams of the model, such as the number of input Tx beams of the model, or the number of input Rx beams of the model, or the number of input beam pairs of the model, etc.
- the terminal reports the number of input beams of the model, which makes it easier for the network device to configure and send Tx beams after selecting the model used by the terminal, thereby improving the terminal prediction performance.
- the terminal can report the number of output beams of the model, such as the number of output Tx beams of the model, or the number of output Rx beams of the model, or the number of output beam pairs of the model, etc.
- the terminal reports the number of output beams of the model, so that the network device can subsequently schedule the terminal to provide feedback on the optimal beam output by the model, such as indicating the number of Tx beams fed back by the terminal.
- the terminal may report a downlink reference signal type applicable to the model, and the downlink reference signal type may include: CSI-RS, SSB, PT-RS, CRS, DMRS, etc.
- the terminal can report the downlink beam sending order applicable to the model, so that the network equipment can configure and send the Tx beam after selecting the model used by the terminal, thereby improving the beam prediction performance.
- the terminal can report the downlink beam pattern applicable to the model, so that the network After selecting the model used by the terminal, the device configures and sends the Tx beam to improve the beam prediction performance.
- the terminal can report the mapping relationship between the downlink beam identifier applicable to the model and the physical downlink beam, which can be used to enable the network device to understand the beam identifier (such as Tx beam ID) reported by the terminal after predicting the optimal beam, and send subsequent data information on its corresponding physical beam.
- the beam identifier such as Tx beam ID
- the terminal can report the identifier of the downlink transmit beam to which the model applies.
- the terminal can recommend which Tx beam the network device should send by reporting the identifier of the downlink transmit beam to which the model applies, so that the terminal can use different Rx beams to receive the reference signal sent by the Tx beam for measurement, and predict the optimal Rx beam based on the measurement results.
- the terminal can report the number of transmissions of the downlink transmission beam applicable to the model.
- the terminal can report the number of transmissions of the downlink transmission beam applicable to the model, the Tx beam sent by the recommended network device, and the number of times the reference signal is sent on the Tx beam, so that the terminal can use different Rx beams to receive the reference signal sent by the Tx beam for measurement, and predict the optimal Rx beam based on the measurement results.
- the model information reporting method provided in the embodiment of the present disclosure enables a network device to receive relevant information of one or more AI/ML models sent by a terminal, so that the use of the model can be activated and the sending pattern of the Tx beam can be configured as needed, thereby improving the transmission performance of the system and better supporting the use of AI/ML technology for beam management and beam prediction.
- the network device receiving the first information sent by the terminal may include:
- the first signaling may include one or more of the following:
- Terminal capability signaling For example, the terminal may report the first information via terminal capability information.
- the UCI may be the UCI transmitted on the PUCCH or PUSCH.
- receiving the first information sent by the terminal includes:
- the terminal capability query message includes one or more of the following: application scenario, network device configuration, model function, and maximum number of models fed back;
- Terminal capability information sent by a terminal is received, where the terminal capability information includes first information.
- the network device can send a terminal capability query (UECapabilityEnquiry) message to the terminal to query the terminal for terminal capability information.
- the terminal capability query message may include one or more of the application scenario, network device configuration, model function, and the maximum number of feedback models.
- the terminal After receiving the terminal capability query message sent by the network device, the terminal can feed back information of the corresponding model in the terminal capability information (UECapabilityInformation) according to the terminal capability query message.
- UECapabilityInformation information of the corresponding model in the terminal capability information
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- receiving the first information sent by the terminal includes:
- the UCI feedback resource configuration message includes one or more of the UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models;
- a UCI sent by a receiving terminal is received, where the UCI includes first information.
- the network device can send a UCI feedback resource configuration message (for example, an RRC configuration message) to the terminal to configure the UCI feedback resources of the terminal.
- the UCI feedback resource configuration message may include one or more of the UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models.
- the terminal may feedback UCI on the configured UCI feedback resource according to the UCI feedback resource configuration message, which includes information of the corresponding model.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the terminal The number of model information reported by the client.
- receiving the first information sent by the terminal includes:
- the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- An RRC response message sent by a receiving terminal is included in the RRC response message.
- the network device may send an RRC request message to the terminal, for example, requesting the terminal to report model information.
- the RRC request message may include one or more of the application scenario, network device configuration, model function, and the maximum number of models for feedback.
- the terminal After receiving the RRC request message sent by the network device, the terminal can feed back information of the corresponding model through an uplink RRC message according to the RRC request message.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- receiving the first information sent by the terminal includes:
- the measurement configuration message includes one or more of measurement configuration, application scenario, network device configuration, model function, and maximum number of feedback models;
- the receiving terminal sends RRC signaling, UCI signaling, or signaling used to transmit data information, where the RRC signaling, UCI signaling, or signaling used to transmit data information includes first information.
- the network device can send a measurement configuration message to the terminal, which may include one or more of the following: measurement configuration (such as Tx beam configuration, reference signal configuration, etc.), application scenario, network device configuration, model function, and maximum number of feedback models.
- measurement configuration such as Tx beam configuration, reference signal configuration, etc.
- application scenario such as Tx beam configuration, reference signal configuration, etc.
- network device configuration such as Tx beam configuration, reference signal configuration, etc.
- model function such as a maximum number of feedback models.
- the terminal After receiving the measurement configuration message sent by the network device, the terminal can feedback the information of the corresponding model according to the measurement configuration message.
- the terminal can feedback the information of one or more models using RRC signaling, UCI signaling or signaling for transmitting data information.
- the terminal can determine the application scenario or network device configuration based on the measurement configuration message, thereby selecting one or more models suitable for the existing application scenario or network device configuration, and reporting their information to the network device, thereby reducing the amount of model information reported by the terminal.
- receiving the first information sent by the terminal includes:
- model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- the model registration information sent by the terminal is received, where the model registration information includes the first information.
- the network device can send a model registration trigger message to the terminal to trigger model registration.
- the model registration trigger message may include one or more of the application scenario, network device configuration, model function, and the maximum number of models for feedback.
- the terminal After receiving the model registration trigger message sent by the network device, the terminal can send model registration information to the network device according to the model registration trigger message, including information of the corresponding model.
- the terminal can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the terminal.
- the method after receiving the first information sent by the terminal, the method further includes:
- an RRC configuration message is sent to the terminal, where the RRC configuration message includes one or more model identifiers.
- the network device after the network device receives the first information reported by the terminal, it can configure the model available to the terminal through RRC signaling (such as an RRC configuration message) according to the model information reported by the terminal, and the configuration information can include a model identifier.
- RRC signaling such as an RRC configuration message
- the terminal after the terminal receives the RRC configuration message sent by the network device, it can determine which model or models are used for downlink beam prediction.
- the network device may select a model corresponding to the current application scenario or network device configuration based on the current application scenario or network device configuration, and configure the model identifier of the corresponding model to the terminal.
- the network device may use RRC signaling to configure a group of models, and then activate one of the models through MAC-CE for optimal beam prediction; or RRC signaling only configures one model for optimal beam prediction; or RRC signaling configures a group of models, and the terminal selects one or more models for optimal beam prediction.
- the method further comprises:
- An RRC signaling or a media access control layer-control unit MAC-CE is sent to the terminal, where the RRC signaling or the MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication.
- the network device may select to activate one or more models based on the current application scenario or network device configuration, and may specifically activate through RRC signaling or through MAC-CE.
- the activation signaling may include an identifier of the activation model and an activation indication.
- the network device when the performance of the model is not good (for example, compared with the performance of other models, or compared with the performance of legacy methods), the network device chooses to switch to other models, or deactivate the model, or fallback to the legacy method, which can be activated/deactivated/fallback through RRC, or activated/deactivated/fallback through MAC CE.
- the activation/deactivation signaling includes the identifier of the activation/deactivation model and the activation/deactivation indication; the fallback signaling includes the fallback indication.
- the activation/deactivation/fallback indication may be a 1-bit indication.
- Example 1 UE reports model information through UE Capability signaling.
- FIG3 is one of the implementation diagrams of the model information reporting method provided by the embodiment of the present disclosure. As shown in FIG3 , the specific steps are as follows:
- Step 1 The network device sends a UECapabilityEnquiry message to query the UE for UE capability information.
- the UECapabilityEnquiry message includes one or more of: application scenario, network device configuration, model function, and maximum number of models to be fed back.
- Application scenarios include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma, Umi, etc.
- Network equipment configuration includes: network equipment antenna configuration, beam configuration, reference signal configuration, etc.
- the model functions mainly include spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Maximum number of models for feedback refers to the maximum number of models that the UE can feedback. For example, the maximum number of models for feedback is 6, which means that the UE can report information of up to 6 models.
- Step 2 The UE feeds back the corresponding model information in the UECapabilityInformation information according to the UECapabilityEnquiry message sent by the base station.
- the UECapabilityEnquiry message includes an application scenario
- the UE can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the UE.
- Step 3 The network device configures the available models for the UE through RRC based on the model information reported by the UE.
- the configuration information includes the model ID.
- the network device selects the corresponding model based on the model information reported by the UE, as well as the existing scenario and configuration, and configures it to the UE.
- the network device selects a model corresponding to an existing application scenario or network device configuration based on the application scenario or network device configuration.
- the network device can use RRC to configure a group of models, and then activate one of the models through MAC CE for optimal beam prediction; or RRC only configures one model for optimal beam prediction; or RRC configures a group of models, and UE selects one or more models for optimal beam prediction.
- Step 4 The network device configures the relevant downlink reference signal and sends the Tx beam according to the selected or activated model information, such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc., and configures the UE feedback information according to the number of Tx beams output by the model.
- model information such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc.
- Step 5 The UE feeds back the Tx beam output by the model according to the network equipment configuration, for example, the best beam ID (optionally, adding the RSRP feedback of the best beam), or the N best Top-N best beam IDs (optionally, adding the RSRP feedback of the Top-N best beams), or N beam IDs (the best beam is included in them, optionally, adding the RSRP feedback of N beams).
- the best beam ID optionally, adding the RSRP feedback of the best beam
- the N best Top-N best beam IDs optionally, adding the RSRP feedback of the Top-N best beams
- N beam IDs the best beam is included in them, optionally, adding the RSRP feedback of N beams.
- Step 6 When the performance of the model is not good (for example, compared with the performance of other models or the performance of the legacy method), the network device chooses to switch to other models, or deactivate this model, or fallback to the legacy method, which can be activated/deactivated/fallbacked through RRC. Or activate/deactivate/fallback through MAC CE.
- the activation/deactivation/fallback signaling includes the ID of the activation/deactivation model or the model number (which can save bits) and a 1-bit activation/deactivation/fallback indication.
- Embodiment 2 The UE sends model information on the feedback resource through UCI.
- FIG. 4 is a second implementation diagram of the model information reporting method provided in the embodiment of the present disclosure. As shown in FIG. 4 , the specific steps are as follows:
- Step 1 The network sends a UCI feedback resource configuration message to configure UCI feedback resources to the UE.
- the UCI feedback resource configuration message includes: one or more of: UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models.
- Application scenarios include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma, Umi, etc.
- Network equipment configuration includes: network equipment antenna configuration, beam configuration, reference signal configuration, etc.
- the model functions mainly include spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Maximum number of models for feedback refers to the maximum number of models that the UE can feedback. For example, the maximum number of models for feedback is 6, which means that the UE can report information of up to 6 models.
- Step 2 The UE feeds back UCI on the resource according to the UCI feedback resource configuration message sent by the network device, which includes information of the corresponding model.
- the UE can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the UE.
- Step 3 The network device configures the available models for the UE through RRC based on the model information reported by the UE.
- the configuration information includes the model ID.
- Step 4 The network device configures the relevant downlink reference signal and sends the Tx beam according to the selected or activated model information, such as the number of Tx beams, the order of Tx beam transmission, and the pattern of Tx beam transmission. It also configures the UE according to the number of Tx beams output by the model. Feedback information.
- model information such as the number of Tx beams, the order of Tx beam transmission, and the pattern of Tx beam transmission. It also configures the UE according to the number of Tx beams output by the model. Feedback information.
- Step 5 The UE feeds back the Tx beam output by the model according to the network equipment configuration, for example, the best beam ID (optionally, adding the RSRP feedback of the best beam), or the N best Top-N best beam IDs (optionally, adding the RSRP feedback of the Top-N best beams), or N beam IDs (the best beam is included in them, optionally, adding the RSRP feedback of N beams).
- the best beam ID optionally, adding the RSRP feedback of the best beam
- the N best Top-N best beam IDs optionally, adding the RSRP feedback of the Top-N best beams
- N beam IDs the best beam is included in them, optionally, adding the RSRP feedback of N beams.
- Step 6 When the performance of this model is not good (for example, compared with the performance of other models or the performance of the legacy method), the network device chooses to switch to other models, or deactivate this model, or fallback to the legacy method, which can be done through RRC activation/deactivation/fallback, or through MAC CE activation/deactivation/fallback.
- Embodiment 3 UE feeds back model information via RRC signaling.
- FIG5 is a third implementation diagram of the model information reporting method provided in the embodiment of the present disclosure. As shown in FIG5 , the specific steps are as follows:
- Step 1 The network sends an RRC request message to request the UE to report model information.
- the RRC request message includes one or more of the following: application scenario, network equipment configuration, model function, and the maximum number of models for feedback.
- Application scenarios include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma, Umi, etc.
- Network equipment configuration includes: network equipment antenna configuration, beam configuration, reference signal configuration, etc.
- the model functions mainly include spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Maximum number of models for feedback refers to the maximum number of models that the UE can feedback. For example, the maximum number of models for feedback is 6, which means that the UE can report information of up to 6 models.
- Step 2 UE feeds back the corresponding model information through the uplink RRC message based on the RRC request message sent by the network device.
- the UE can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the UE.
- Step 3 The network device configures the available models for the UE through RRC based on the model information reported by the UE.
- the configuration information includes the model ID.
- Step 4 The network device configures the relevant downlink reference signal and sends the Tx beam according to the selected or activated model information, such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc., and configures the UE feedback information according to the number of Tx beams output by the model.
- model information such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc.
- Step 5 The UE feeds back the Tx beam output by the model according to the network equipment configuration, for example, the best beam ID (optionally, adding the RSRP feedback of the best beam), or the N best Top-N best beam IDs (optionally, adding the RSRP feedback of the Top-N best beams), or N beam IDs (the best beam is included in them, optionally, adding the RSRP feedback of N beams).
- the best beam ID optionally, adding the RSRP feedback of the best beam
- the N best Top-N best beam IDs optionally, adding the RSRP feedback of the Top-N best beams
- N beam IDs the best beam is included in them, optionally, adding the RSRP feedback of N beams.
- Step 6 When the performance of this model is not good (for example, compared with the performance of other models or the performance of the legacy method), the network device chooses to switch to other models, or deactivate this model, or fallback to the legacy method, which can be done through RRC activation/deactivation/fallback, or through MAC CE activation/deactivation/fallback.
- Embodiment 4 The UE feeds back model information according to the measurement configuration message.
- FIG6 is a fourth implementation diagram of the model information reporting method provided in the embodiment of the present disclosure. As shown in FIG6 , the specific steps are as follows:
- Step 1 The network sends a measurement configuration message, including Tx beam configuration, reference signal configuration, etc.
- Step 2 The UE feeds back the corresponding model information according to the measurement configuration message.
- the UE determines the application scenario or network device configuration according to the measurement configuration information, thereby selecting one or more models applicable to the existing application scenario or network device configuration, and reporting their information to the network device, thereby reducing the amount of model information reported by the UE.
- RRC signaling and UCI signaling can be used to transmit data information and feedback the One or more models.
- Step 3 The network device configures the available models for the UE through RRC based on the model information reported by the UE.
- the configuration information includes the model ID.
- Step 4 The network device configures the relevant downlink reference signal and sends the Tx beam according to the selected or activated model information, such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc., and configures the UE feedback information according to the number of Tx beams output by the model.
- model information such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc.
- Step 5 The UE feeds back the Tx beam output by the model according to the network equipment configuration, for example, the best beam ID (optionally, adding the RSRP feedback of the best beam), or the N best Top-N best beam IDs (optionally, adding the RSRP feedback of the Top-N best beams), or N beam IDs (the best beam is included in them, optionally, adding the RSRP feedback of N beams).
- the best beam ID optionally, adding the RSRP feedback of the best beam
- the N best Top-N best beam IDs optionally, adding the RSRP feedback of the Top-N best beams
- N beam IDs the best beam is included in them, optionally, adding the RSRP feedback of N beams.
- Step 6 When the performance of this model is not good (for example, compared with the performance of other models or the performance of the legacy method), the network device chooses to switch to other models, or deactivate this model, or fallback to the legacy method, which can be done through RRC activation/deactivation/fallback, or through MAC CE activation/deactivation/fallback.
- Embodiment 5 The UE feeds back model information via signaling for model registration.
- FIG. 7 is a fifth implementation diagram of the model information reporting method provided in the embodiment of the present disclosure. As shown in FIG. 7 , the specific steps are as follows:
- Step 1 The network sends a model registration trigger message to trigger the UE to register the model.
- the model registration trigger message includes: one or more of: application scenario, network device configuration, model function, and maximum number of models fed back.
- Application scenarios include: urban, rural, indoor, outdoor, highway, high-speed rail, Uma, Umi, etc.
- Network equipment configuration includes: network equipment antenna configuration, beam configuration, reference signal configuration, etc.
- the model functions mainly include spatial domain prediction, frequency domain prediction, time domain prediction, downlink beam pair prediction, downlink Tx beam prediction, downlink Rx beam prediction, etc.
- Maximum number of models for feedback refers to the maximum number of models that the UE can feedback. For example, the maximum number of models for feedback is 6, which means that the UE can report information of up to 6 models.
- Step 2 The UE feeds back the corresponding model information through the model registration information according to the model registration trigger message sent by the network device.
- the UE can upload the model information corresponding to the application scenario, and the model information of other application scenarios does not need to be uploaded, thereby reducing the amount of model information reported by the UE.
- Step 3 The network device configures the available models for the UE through RRC based on the model information reported by the UE.
- the configuration information includes the model ID.
- Step 4 The network device configures the relevant downlink reference signal and sends the Tx beam according to the selected or activated model information, such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc., and configures the UE feedback information according to the number of Tx beams output by the model.
- model information such as the number of Tx beams, the order of Tx beam transmission, the Tx beam transmission pattern, etc.
- Step 5 The UE feeds back the Tx beam output by the model according to the network equipment configuration, for example, the best beam ID (optionally, adding the RSRP feedback of the best beam), or the N best Top-N best beam IDs (optionally, adding the RSRP feedback of the Top-N best beams), or N beam IDs (the best beam is included in them, optionally, adding the RSRP feedback of N beams).
- the best beam ID optionally, adding the RSRP feedback of the best beam
- the N best Top-N best beam IDs optionally, adding the RSRP feedback of the Top-N best beams
- N beam IDs the best beam is included in them, optionally, adding the RSRP feedback of N beams.
- Step 6 When the performance of this model is not good (for example, compared with the performance of other models or the performance of the legacy method), the network device chooses to switch to other models, or deactivate this model, or fallback to the legacy method, which can be done through RRC activation/deactivation/fallback, or through MAC CE activation/deactivation/fallback.
- the methods and devices provided in the various embodiments of the present disclosure are based on the same application concept. Since the methods and devices solve problems based on similar principles, the implementation of the devices and methods can refer to each other, and the repeated parts will not be repeated.
- FIG8 is a schematic diagram of the structure of a terminal provided in an embodiment of the present disclosure.
- the terminal includes The embodiment includes a memory 820, a transceiver 810 and a processor 800; wherein the processor 800 and the memory 820 may also be arranged physically separately.
- the memory 820 is used to store computer programs; the transceiver 810 is used to send and receive data under the control of the processor 800.
- the transceiver 810 is used to receive and send data under the control of the processor 800 .
- the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 800 and various circuits of memory represented by memory 820 are linked together.
- the bus architecture can also link various other circuits such as peripherals, regulators, and power management circuits together, which are all well known in the art, and therefore, the present disclosure will not further describe them.
- the bus interface provides an interface.
- the transceiver 810 can 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 830 can 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 800 is responsible for managing the bus architecture and general processing, and the memory 820 can store data used by the processor 800 when performing operations.
- the processor 800 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).
- the processor can also adopt a multi-core architecture.
- the processor 800 calls the computer program stored in the memory 820 to execute any of the methods provided in the embodiments of the present disclosure according to the obtained executable instructions, for example: determining first information, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction; sending the first information to the network device.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- sending first information to a network device includes:
- the first signaling includes one or more of the following:
- sending first information to a network device includes:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- terminal capability information is sent to the network device, where the terminal capability information includes first information.
- sending first information to a network device includes:
- UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- the UCI is sent to the network device, where the UCI includes the first information.
- sending first information to a network device includes:
- RRC request message sent by a network device, where the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- an RRC response message is sent to the network device, where the RRC response message includes first information.
- sending first information to a network device includes:
- the measurement configuration message includes one or more of a measurement configuration, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- RRC signaling, UCI signaling, or signaling for transmitting data information is sent to the network device, where the RRC signaling, UCI signaling, or signaling for transmitting data information includes first information.
- sending first information to a network device includes:
- model registration trigger message sent by a network device, where the model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- model registration information is sent to the network device, where the model registration information includes first information.
- the method further includes:
- the method further comprises:
- Receive RRC signaling or media access control layer-control unit MAC-CE sent by the network device includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication;
- the model activation, deactivation or fallback operation is performed.
- the method further includes:
- model identifier allocation message sent by a network device, wherein the model identifier allocation message includes a model identifier allocated by the network device;
- the model identifier allocated by the network device is associated with the model indicated in the first information.
- FIG9 is a schematic diagram of the structure of a network device provided in an embodiment of the present disclosure.
- the network device includes a memory 920, a transceiver 910, and a processor 900; wherein the processor 900 and the memory 920 may also be arranged physically separately.
- the memory 920 is used to store computer programs; the transceiver 910 is used to send and receive data under the control of the processor 900.
- the transceiver 910 is used to receive and send data under the control of the processor 900 .
- the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 900 and various circuits of memory represented by memory 920 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 in this disclosure.
- the bus interface provides an interface.
- the transceiver 910 may be a plurality of components, including a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, which may include a wireless channel, a wired channel, an optical cable, and other transmission media.
- the processor 900 is responsible for managing the bus architecture and general processing, and the memory 920 can store data used by the processor 900 when performing operations.
- the processor 900 may be a CPU, an ASIC, an FPGA or a CPLD, and the processor may also adopt a multi-core architecture.
- the processor 900 is configured to execute any of the methods provided in the embodiments of the present disclosure according to the obtained executable instructions by calling the computer program stored in the memory 920, for example: receiving first information sent by the terminal, the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, AI/ The ML model is used for downlink beam prediction.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- receiving the first information sent by the terminal includes:
- the first signaling includes one or more of the following:
- receiving the first information sent by the terminal includes:
- the terminal capability query message includes one or more of the following: application scenario, network device configuration, model function, and maximum number of models fed back;
- Terminal capability information sent by a terminal is received, where the terminal capability information includes first information.
- receiving the first information sent by the terminal includes:
- the UCI feedback resource configuration message includes one or more of the UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models;
- a UCI sent by a receiving terminal is received, where the UCI includes first information.
- receiving the first information sent by the terminal includes:
- the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- An RRC response message sent by a receiving terminal is included in the RRC response message.
- receiving the first information sent by the terminal includes:
- the measurement configuration message includes one or more of measurement configuration, application scenario, network device configuration, model function, and maximum number of feedback models;
- the receiving terminal sends RRC signaling, UCI signaling, or signaling used to transmit data information, where the RRC signaling, UCI signaling, or signaling used to transmit data information includes first information.
- receiving the first information sent by the terminal includes:
- model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- the model registration information sent by the terminal is received, where the model registration information includes the first information.
- the method after receiving the first information sent by the terminal, the method further includes:
- an RRC configuration message is sent to the terminal, where the RRC configuration message includes one or more model identifiers.
- the method further comprises:
- RRC signaling or media access control layer-control unit MAC-CE Send RRC signaling or media access control layer-control unit MAC-CE to the terminal, where the RRC signaling or MAC-CE contains one of the model identifier, activation indication, deactivation indication, and fallback indication, or Multiple items.
- the method after receiving the first information sent by the terminal, the method further includes:
- a model identifier allocation message is sent to the terminal, where the model identifier allocation message includes a model identifier allocated by the network device.
- FIG. 10 is a schematic diagram of a structure of a model information reporting device provided in an embodiment of the present disclosure. As shown in FIG. 10 , the device includes:
- a determining unit 1000 is used to determine first information, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction;
- the first sending unit 1010 is configured to send first information to a network device.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- sending first information to a network device includes:
- the first signaling includes one or more of the following:
- sending first information to a network device includes:
- the terminal capability query message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- terminal capability information is sent to the network device, where the terminal capability information includes first information.
- sending first information to a network device includes:
- UCI feedback resource configuration message includes one or more of a UCI feedback resource time-frequency position, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- the UCI is sent to the network device, where the UCI includes the first information.
- sending first information to a network device includes:
- RRC request message sent by a network device, where the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- an RRC response message is sent to the network device, where the RRC response message includes first information.
- sending first information to a network device includes:
- the measurement configuration message includes one or more of a measurement configuration, an application scenario, a network device configuration, a model function, and a maximum number of feedback models;
- RRC signaling, UCI signaling, or signaling for transmitting data information is sent to the network device, where the RRC signaling, UCI signaling, or signaling for transmitting data information includes first information.
- sending first information to a network device includes:
- model registration trigger message sent by a network device, where the model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- model registration information is sent to the network device, where the model registration information includes first information.
- the apparatus further includes a first receiving unit, configured to:
- the apparatus further includes a second receiving unit, configured to:
- RRC signaling or media access control layer-control unit MAC-CE sent by the network device, where the RRC signaling or MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication;
- the model activation, deactivation or fallback operation is performed.
- the apparatus further comprises an associating unit, configured to:
- model identifier allocation message sent by a network device, wherein the model identifier allocation message includes a model identifier allocated by the network device;
- the model identifier allocated by the network device is associated with the model indicated in the first information.
- FIG. 11 is a second structural diagram of a model information reporting device provided in an embodiment of the present disclosure. As shown in FIG. 11 , the device includes:
- the third receiving unit 1100 is used to receive first information sent by the terminal, where the first information is used to indicate one or more artificial intelligence or machine learning AI/ML models, and the AI/ML model is used for downlink beam prediction.
- the first information includes one or more of the following:
- the model identification includes one or more of the following:
- serial number or serial number of the model The serial number or serial number of the model
- receiving the first information sent by the terminal includes:
- the first signaling includes one or more of the following:
- receiving the first information sent by the terminal includes:
- the terminal capability query message includes one or more of the following: application scenario, network device configuration, model function, and maximum number of models fed back;
- Terminal capability information sent by a terminal is received, where the terminal capability information includes first information.
- receiving the first information sent by the terminal includes:
- the UCI feedback resource configuration message includes one or more of the UCI feedback resource time-frequency position, application scenario, network device configuration, model function, and maximum number of feedback models;
- a UCI sent by a receiving terminal is received, where the UCI includes first information.
- receiving the first information sent by the terminal includes:
- the RRC request message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models for feedback;
- An RRC response message sent by a receiving terminal is included in the RRC response message.
- receiving the first information sent by the terminal includes:
- the measurement configuration message includes one or more of measurement configuration, application scenario, network device configuration, model function, and maximum number of feedback models;
- the receiving terminal sends RRC signaling, UCI signaling, or signaling used to transmit data information, where the RRC signaling, UCI signaling, or signaling used to transmit data information includes first information.
- receiving the first information sent by the terminal includes:
- model registration trigger message includes one or more of an application scenario, a network device configuration, a model function, and a maximum number of models fed back;
- the model registration information sent by the terminal is received, where the model registration information includes the first information.
- the apparatus further comprises:
- the second sending unit is used to send an RRC configuration message to the terminal according to the first information, where the RRC configuration message includes one or more model identifiers.
- the apparatus further comprises:
- the third sending unit is used to send RRC signaling or media access control layer-control unit MAC-CE to the terminal, and the RRC signaling or MAC-CE includes one or more of a model identifier, an activation indication, a deactivation indication, and a fallback indication.
- the apparatus further comprises a dispensing unit for:
- a model identifier allocation message is sent to the terminal, where the model identifier allocation message includes a model identifier allocated by the network device.
- each functional unit in each embodiment of the present disclosure may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
- the above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
- the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium.
- the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product.
- the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present disclosure.
- the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.
- an embodiment of the present disclosure further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is used to enable a computer to execute the model information reporting method provided by the above embodiments.
- the computer-readable storage medium may be any available medium or data storage device that can be accessed by a computer, including but not limited to magnetic storage devices (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs) etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state drive (SSD)), etc.
- magnetic storage devices such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs) etc.
- optical storage such as CD, DVD, BD, HVD, etc.
- semiconductor storage such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state drive (SSD)
- 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
- the system can also include core network parts, such as the Evolved Packet
- the terminal involved in the embodiments of the present disclosure may be a device that provides voice and/or data connectivity to a user, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem.
- the name of the terminal may also be different.
- the terminal may be called a user equipment (UE).
- a wireless terminal device may communicate with one or more core networks (CN) via a radio access network (RAN).
- the wireless terminal device may be a mobile terminal device, such as a mobile phone (or a "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 radio 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 Remote station, access point, remote terminal equipment, access terminal equipment, user terminal equipment, user agent, and user device are not limited in the embodiments of the present disclosure.
- 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.
- Network devices and terminals can each use one or more antennas for multiple input multiple output (MIMO) transmission.
- MIMO transmission can be single user MIMO (SU-MIMO) or multi-user MIMO (MU-MIMO).
- MIMO transmission can be 2D-MIMO, 3D-MIMO, FD-MIMO or massive-MIMO, or it can be diversity transmission, precoded transmission or beamforming transmission, etc.
- the embodiments of the present disclosure may be provided as methods, systems, or Computer program product. Therefore, the present disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.
- each process and/or box in the flowchart and/or block diagram, as well as the combination of the process and/or box in the flowchart and/or block diagram can be implemented by computer executable instructions.
- These computer executable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.
- processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the processor-readable memory produce a product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
- processor-executable instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows in the flowchart and/or one or more blocks in the block diagram.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本公开提供一种模型信息上报方法、设备、装置及存储介质,该方法包括:终端确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;向网络设备发送所述第一信息。
Description
相关申请的交叉引用
本申请要求于2022年09月30日提交的申请号为202211215756.9,发明名称为“模型信息上报方法、设备、装置及存储介质”的中国专利申请的优先权,其通过引用方式全部并入本文。
本公开涉及无线通信技术领域,尤其涉及一种模型信息上报方法、设备、装置及存储介质。
在新无线(New Radio,NR)下行波束管理场景中,确定最优下行波束的方式,需要基站循环在不同方向发送Tx beam(发送波束),终端使用Rx beam(接收波束)接收Tx beam,并测量所有Tx beam上发送的信道状态信息参考信号(Channel State Information Reference Signal,CSI-RS)或同步信号块(Synchronization Signal Block,SSB)信号,选出接收性能(如参考信号接收功率(Reference Signal Received Power,RSRP))最好的波束(best beam),用于后续基站使用该下行波束给终端传输信息。
现有技术的实现方案,CSI-RS或SSB信号需要在所有Tx beam上发送,资源消耗较大;且终端需要分别测量所有Tx beam上发送的CSI-RS或SSB信号,终端实现较为复杂,并且测量开销较大。
发明内容
针对现有技术存在的问题,本公开提供一种模型信息上报方法、设备、装置及存储介质,用以支持AI/ML技术预测下行波束。
第一方面,本公开提供一种模型信息上报方法,应用于终端,包括:
确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;
向网络设备发送所述第一信息。
在一些实施例中,所述第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,所述模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
向网络设备发送第一信令,所述第一信令中包含所述第一信息;
其中,所述第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述终端能力查询消息,向所述网络设备发送终端能力信息,所述终端能力信息中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述UCI反馈资源配置消息,向所述网络设备发送UCI,所述UCI中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述RRC请求消息,向所述网络设备发送RRC响应消息,所述RRC响应消息中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述测量配置消息,向所述网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述模型注册触发消息,向所述网络设备发送模型注册信息,所述模型注册信息中包含所述第一信息。
在一些实施例中,向网络设备发送所述第一信息之后,所述方法还包括:
接收网络设备发送的RRC配置消息,所述RRC配置消息中包含一个或多个模型标识;
根据所述RRC配置消息,确定用于下行波束预测的AI/ML模型。
在一些实施例中,所述方法还包括:
接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;
根据所述RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
在一些实施例中,向网络设备发送所述第一信息之后,所述方法还包括:
接收所述网络设备发送的模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识;
根据所述模型标识分配消息,将所述网络设备分配的模型标识与所述第一信息中指示的模型相关联。
第二方面,本公开还提供一种模型信息上报方法,应用于网络设备,包括:
接收终端发送的第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
在一些实施例中,所述第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,所述模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
接收终端发送的第一信令,所述第一信令中包含所述第一信息;
其中,所述第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的终端能力信息,所述终端能力信息中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的UCI,所述UCI中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送RRC请求消息,所述RRC请求消息中包含应用场景、网络
设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的RRC响应消息,所述RRC响应消息中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的模型注册信息,所述模型注册信息中包含所述第一信息。
在一些实施例中,接收终端发送的第一信息之后,所述方法还包括:
根据所述第一信息,向所述终端发送RRC配置消息,所述RRC配置消息中包含一个或多个模型标识。
在一些实施例中,所述方法还包括:
向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
在一些实施例中,接收终端发送的第一信息之后,所述方法还包括:
根据所述第一信息,向所述终端发送模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识。
第三方面,本公开还提供一种终端,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:
确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;
向网络设备发送所述第一信息。
在一些实施例中,所述第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,所述模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
向网络设备发送第一信令,所述第一信令中包含所述第一信息;
其中,所述第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述终端能力查询消息,向所述网络设备发送终端能力信息,所述终端能力信息中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述UCI反馈资源配置消息,向所述网络设备发送UCI,所述UCI中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述RRC请求消息,向所述网络设备发送RRC响应消息,所述RRC响应消息中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述测量配置消息,向所述网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
在一些实施例中,所述向网络设备发送所述第一信息,包括:
接收网络设备发送的模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据所述模型注册触发消息,向所述网络设备发送模型注册信息,所述模型注册信息中包含所述第一信息。
在一些实施例中,向网络设备发送所述第一信息之后,所述操作还包括:
接收网络设备发送的RRC配置消息,所述RRC配置消息中包含一个或多个模型标识;
根据所述RRC配置消息,确定用于下行波束预测的AI/ML模型。
在一些实施例中,所述操作还包括:
接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;
根据所述RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
在一些实施例中,向网络设备发送所述第一信息之后,所述操作还包括:
接收所述网络设备发送的模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识;
根据所述模型标识分配消息,将所述网络设备分配的模型标识与所述第一信息中指示的模型相关联。
第四方面,本公开还提供一种网络设备,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信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
在一些实施例中,所述第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,所述模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
接收终端发送的第一信令,所述第一信令中包含所述第一信息;
其中,所述第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的终端能力信息,所述终端能力信息中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的UCI,所述UCI中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的RRC响应消息,所述RRC响应消息中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
在一些实施例中,所述接收终端发送的第一信息,包括:
向终端发送模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收所述终端发送的模型注册信息,所述模型注册信息中包含所述第一信息。
在一些实施例中,接收终端发送的第一信息之后,所述操作还包括:
根据所述第一信息,向所述终端发送RRC配置消息,所述RRC配置消息中包含一个或多个模型标识。
在一些实施例中,所述操作还包括:
向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
在一些实施例中,接收终端发送的第一信息之后,所述操作还包括:
根据所述第一信息,向所述终端发送模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识。
第五方面,本公开还提供一种模型信息上报装置,包括:
确定单元,用于确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;
第一发送单元,用于向网络设备发送所述第一信息。
第六方面,本公开还提供一种模型信息上报装置,包括:
第三接收单元,用于接收终端发送的第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
第七方面,本公开还提供一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序用于使计算机执行如上所述第一方面所述的模型信息上报方法,或执行如上所述第二方面所述的模型信息上报方法。
第八方面,本公开还提供一种通信设备,所述通信设备中存储有计算机程序,所述计算机程序用于使通信设备执行如上所述第一方面所述的模型信息上报方法,或执行如上所述第二方面所述的模型信息上报方法。
第九方面,本公开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处理器执行如上所述第一方面所述的模型信息上报方法,或执行如上所述第二方面所述的模型信息上报方法。
第十方面,本公开还提供一种芯片产品,所述芯片产品中存储有计算机程序,所述计算机程序用于使芯片产品执行如上所述第一方面所述的模型信息上报方法,或执行如上所述第二方面所述的模型信息上报方法。
本公开提供的模型信息上报方法、设备、装置及存储介质,终端可以通过第一信息向网络设备上报一个或多个AI/ML模型的相关信息,从而使得网络设备后续可以按需激活模型的使用和配置Tx beam的发送图样,提升系统传输性能,能够更好地支持利用AI/ML技术进行波束管理和波束预测。
为了更清楚地说明本公开实施例或相关技术中的技术方案,下面将对实施例或相关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为本公开实施例提供的模型信息上报方法的流程示意图之一;
图2为本公开实施例提供的模型信息上报方法的流程示意图之二;
图3为本公开实施例提供的模型信息上报方法的实施示意图之一;
图4为本公开实施例提供的模型信息上报方法的实施示意图之二;
图5为本公开实施例提供的模型信息上报方法的实施示意图之三;
图6为本公开实施例提供的模型信息上报方法的实施示意图之四;
图7为本公开实施例提供的模型信息上报方法的实施示意图之五;
图8为本公开实施例提供的终端的结构示意图;
图9为本公开实施例提供的网络设备的结构示意图;
图10为本公开实施例提供的模型信息上报装置的结构示意图之一;
图11为本公开实施例提供的模型信息上报装置的结构示意图之二。
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。
为了便于更加清晰地理解本公开各实施例的技术方案,首先对本公开各实施例相关的一些技术内容进行介绍。
1、三种计算下行(Downlink,DL)最优波束(best beam)的方式。
一种是测量所有下行波束对(DL beam pair)(基站的一个Tx beam和UE的一个Rx beam组成的一个beam pair)的RSRP,得到最大的RSRP对应的beam pair即为最优beam pair,并告知基站,后续基站采用该beam pair中
的Tx beam给UE发送信息,UE采用该beam pair中的Rx beam接收信息。
另一种是UE固定或选择一个最好的DL Rx beam,接收测量基站发送的所有DL Tx beam的接收功率RSRP,得到最大的RSRP对应的Tx beam即为最优Tx beam,并告知基站,后续基站采用该Tx beam给UE发送信息。
再一种是基站固定或选择一个最好的DL Tx beam,UE采用所有DL Rx beam接收测量基站发送的该DL Tx beam的接收功率RSRP,得到最大的RSRP对应的Rx beam即为最优Rx beam,后续基站采用该Tx beam给UE发送信息时,UE采用该最优Rx beam接收。
如上所述,对于DL beam pair测量方式,UE需要使用全部的Rx beam接收基站发送的每个Tx beam的CSI-RS/SSB进行测量。例如,UE有4个Rx beam,基站有32个Tx beam,如果UE采用全部Rx beam分别接收每个Tx beam,则需要测量4*32=128个CSI-RS/SSB,即每个Rx beam都需要测量所有的Tx beam,才能计算出,最好的beam pair。对于DL Tx beam测量方式,UE固定或选择一个最好的Rx beam,测量基站发送的所有Tx beam,需要测量32个Tx beam,才能计算出,最好的Tx beam。对于DL Rx beam测量方式,也有同样的问题。
考虑到上述计算方式需要测量全部的Rx beam和Tx beam,用于测量的参考信号占用传输资源较大,UE测量复杂度高,测量消耗较大,UE测量时延较高。本公开提出利用人工智能(Artificial Intelligence,AI)或机器学习(Machine Learning,AI)技术对下行波束进行预测,对于DL beam pair测量方式,只需要测量一部分DL beam pair,例如只发送32个Tx beam中的8个,则UE只需要测量4*8=32个beam pair上的CSI-RS/SSB,就可以准确预测出32个Tx beam和4个Rx beam组成的128个beam pair中接收性能最好的beam pair;对于DL Tx beam测量方式,UE固定或选择一个最好的Rx beam,只需要测量8个Tx beam,就可以准确预测出32个Tx beam中接收性能最好的Tx beam。当然本公开还可以用于其他场景,例如测量其他便于测量的发送SSB的宽beam,或其他频段的beam,从而预测发送CSI-RS的窄beam,或者预测高频率的beam中的best beam;又例如利用以前测量过的beam测量结果,
预测出基站未来发送的Tx beam或beam pair中接收性能最好的beam(pair);又例如,DL Rx beam预测,基站固定或选择一个最好的Tx beam,连续发送参考信号(Reference Signal,RS),UE使用不同的DL Rx beam接收测量RS,通过测量结果,预测出最好的Rx beam。从而可以节省RS发送资源、UE测量开销和降低UE测量时延。
对于利用AI/ML技术进行波束管理和预测时,UE如何将模型信息上报给基站,本公开提供了明确的解决方案。
图1为本公开实施例提供的模型信息上报方法的流程示意图之一,该方法可应用于终端,如图1所示,该方法包括如下步骤:
步骤100、确定第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/ML模型用于下行波束预测。
步骤101、向网络设备发送第一信息。
具体地,本公开实施例中,为了更好地支持利用AI/ML技术进行波束管理和波束预测,终端可以向网络设备(例如基站)上报第一信息,该第一信息可用于指示一个或多个AI/ML模型,这些AI/ML模型是可用于下行波束预测的AI/ML模型(以下也简称模型)。其中下行波束指的是网络设备与终端之间传输信息所使用的下行波束,包括下行发送波束(Tx beam)、下行接收波束(Rx beam)、下行波束对(beam pair)等。
网络设备接收到终端发送的第一信息之后,便可以获知终端可以使用哪些模型,以及这些模型的相关信息,后续可以在需要的时候根据这些信息进行模型选择激活等流程。
在一些实施例中,第一信息中指示的模型可以是终端自身存储的模型,或者也可以是第三方设备存储的模型,此处不做限定。
对于第三方设备存储模型的情形,终端可以将第三方设备存储的模型信息上报给网络设备,后续根据网络设备的指示(例如配置、激活或选择命令),再从第三方设备下载相应的模型来使用。
在一些实施例中,第一信息中可以包含以下一项或多项:
(1)模型标识。
具体地,第一信息中可以包含模型标识,该模型标识可以是模型的序号或编号,也可以是代表模型类型的标识信息,或者其他可以区分不同模型的标识信息等。
在一些实施例中,模型标识可以包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
其中,用于模型训练的数据集标识、用于模型训练的RS配置标识、用于模型训练的波束描述信息标识、模型的输入波束数量与输出波束数量信息都可以理解为代表模型类型的标识信息,具体说明如下:
数据集标识:数据集是用于模型训练的数据集,终端可以通过不同的数据集训练不同的模型,若数据集和训练的模型是一一对应的,则数据集标识可以代表模型标识。例如,数据集中可以包括数据集标识以及一个或多个数据集样本,其中数据集样本可以包括:多个波束标识,以及该波束标识对应的波束的测量结果,一个或多个最优波束的标识(作为AI/ML模型训练的输出标签)。
RS配置标识:RS配置是用于模型训练的RS配置,终端可以基于RS配置,对下行参考信号进行测量,使用测量结果训练模型,如果RS配置和训练的模型是一一对应的,则RS配置标识可以代表模型标识。例如,RS配置中可以包括RS配置标识,RS种类,RS发送资源,RS标识等。
波束描述信息标识:终端可以使用波束描述信息,以及对下行参考信号进行测量的结果,训练模型,如果波束描述信息和训练的模型是一一对应的,则波束描述信息标识可以代表模型标识。例如,波束描述信息可以包括波束标识或RS标识、各波束标识或RS标识对应的波束的网络设备天线配置、波束的角度信息、波束的宽度信息等。
模型的输入波束数量与输出波束数量信息:一种可能的实现方式中,模
型的输入波束数量与输出波束数量信息可以是{输入波束数,输出波束数}对信息,比如{输入Tx beam数,输出Tx beam数}对信息,或者{输入Rx beam数,输出Rx beam数}对信息。终端可以针对每个{输入波束数,输出波束数}对只训练一个模型,那么{输入波束数,输出波束数}对可以表征模型。
在一些实施例中,一种实施方式中,第一信息中也可以不包含模型标识,终端向网络设备发送第一信息之后,该方法还包括:
接收网络设备发送的模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识;
根据模型标识分配消息,将网络设备分配的模型标识与第一信息中指示的模型相关联。
具体地,第一信息中不包含模型标识的情况下,网络设备在接收到终端发送的第一信息之后,可以根据终端上报的这些模型信息,给终端上报的这些模型分配相应的模型标识。
确定分配的模型标识之后,网络设备可以向终端发送模型标识分配消息,该模型标识分配消息中包含网络设备分配的模型标识。在一些实施例中,该模型标识分配消息中还可以包含每个模型标识对应的模型信息(第一信息中的一项或多项),或者其他指示信息,这些指示信息可用于指示所分配的模型标识与第一信息所指示的模型之间的关联关系,从而终端可以根据模型标识分配消息,将网络设备分配的模型标识与第一信息中指示的模型相关联。
一种可能的实现方式中,模型标识分配消息中包含网络设备分配的模型标识,而不包含显式的指示信息指示所分配的模型标识与第一信息所指示的模型之间的关联关系,终端可以在接收到网络设备发送的模型标识分配消息之后,默认按照其上报的第一信息中各个模型信息的顺序,依次将网络设备分配的模型标识与第一信息中指示的模型相关联。
(2)模型的应用场景。
具体地,终端可以上报模型适用的场景,例如,应用场景可以包括:城市,农村,室内,室外,高速公路,高铁,Uma(城市宏小区),Umi(城市微小区)等等。
终端上报模型的应用场景,可用于后续网络设备根据应用场景选择激活终端使用合适的模型进行波束预测。
(3)模型适用的网络设备配置。
具体地,终端可以上报模型适用的网络设备配置,例如,网络设备配置可以包括:网络设备的天线配置,波束配置,参考信号配置等等。
终端上报模型适用的网络设备配置,可用于后续网络设备根据网络设备配置选择激活终端使用合适的模型进行波束预测。
(4)模型功能。
具体地,终端可以上报模型的功能,例如,模型功能可以包括:空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等等。空域预测是指测量少量的波束,或者其他参考信号类型的波束,预测大量波束中的最优波束;频域预测是指测量频率1上的波束,预测另一个频率2上的波束中的最优波束;时域预测是指测量现在时刻的波束,预测未来时刻的最优波束。
终端上报模型的功能,可用于后续网络设备根据模型功能选择激活终端使用合适的模型进行波束预测。
(5)模型的输入波束数量。
具体地,终端可以上报模型的输入波束数量,比如模型的输入Tx beam数,或者模型的输入Rx beam数,或者模型的输入beam pair数等。
终端上报模型的输入波束数量,便于让网络设备在选择终端使用的模型后,进行配置和发送Tx beam,提升终端预测性能。
(6)模型的输出波束数量。
具体地,终端可以上报模型的输出波束数量,比如模型的输出Tx beam数,或者模型的输出Rx beam数,或者模型的输出beam pair数等。
终端上报模型的输出波束数量,便于让网络设备后续调度该终端对模型输出的最优波束进行反馈,例如指示终端反馈的Tx beam数目。
(7)模型适用的下行参考信号RS类型。
具体地,终端可以上报模型适用的下行参考信号类型,下行参考信号类
型可以包括:CSI-RS,SSB,相位跟踪参考信号(Phase-Tracking Reference Signal,PT-RS),小区参考信号(Cell Reference Signal,CRS),解调参考信号(Demodulation Reference Signal,DMRS)等等。
(8)模型适用的下行波束发送顺序。
具体地,终端可以上报模型适用的下行波束发送顺序,便于让网络设备在选择终端使用的模型后,进行配置和发送Tx beam,提升波束预测性能。
(9)模型适用的下行波束图样。
具体地,终端可以上报模型适用的下行波束图样(pattern),便于让网络设备在选择终端使用的模型后,进行配置和发送Tx beam,提升波束预测性能。
(10)模型适用的下行波束标识与物理的下行波束之间的映射关系。
具体地,终端可以上报模型适用的下行波束标识与物理的下行波束之间的映射关系,可用于让网络设备理解终端预测最优波束后上报的波束标识(如Tx beam ID),并在其对应的物理波束上发送后续数据信息。
(11)模型适用的下行发送波束的标识。
具体地,终端可以上报模型适用的下行发送波束的标识。例如,对于下行Rx beam预测场景,终端可以通过上报模型适用的下行发送波束的标识,建议网络设备发送哪个Tx beam,以便于终端使用不同的Rx beam接收该Tx beam发送的参考信号进行测量,并根据测量结果预测出最优Rx beam。
(12)模型适用的下行发送波束的发送次数。
具体地,终端可以上报模型适用的下行发送波束的发送次数。例如,对于下行Rx beam预测场景,终端可以通过上报模型适用的下行发送波束的发送次数,建议网络设备发送的Tx beam,以及在该Tx beam上发送几次参考信号,以便于终端使用不同的Rx beam接收该Tx beam发送的参考信号进行测量,并根据测量结果预测出最优Rx beam。
本公开实施例提供的模型信息上报方法,终端可以通过第一信息向网络设备上报一个或多个AI/ML模型的相关信息,从而使得网络设备后续可以按需激活模型的使用和配置Tx beam的发送图样,提升系统传输性能,能够更
好地支持利用AI/ML技术进行波束管理和波束预测。
在一些实施例中,终端向网络设备发送第一信息,可以包括:
向网络设备发送第一信令,第一信令中包含第一信息;
其中,第一信令可以包括以下一种或多种:
(1)终端能力(UE capability)信令。例如,终端可以通过终端能力信息(UE capability Information)上报第一信息。
(2)上行控制信息(Uplink Control Information,UCI)信令。该UCI可以是物理上行控制信道(Physical Uplink Control Channel,PUCCH)或物理上行共享信道(Physical Uplink Shared Channel,PUSCH)上传输的UCI。
(3)无线资源控制(Radio Resource Control,RRC)信令。
(4)用于传输数据信息的信令。终端可以在进行数据信息传输时,通过用于传输数据信息的信令携带第一信息上报给网络设备。
(5)用于模型注册的信令。终端可以在进行模型注册时,通过用于模型注册的信令携带第一信息上报给网络设备。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据终端能力查询消息,向网络设备发送终端能力信息,终端能力信息中包含第一信息。
具体地,网络设备可以向终端发送终端能力查询(UECapabilityEnquiry)消息,向终端查询终端能力信息,该终端能力查询消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
其中,最大反馈的模型数目是指终端可以反馈的最大模型个数,例如最大反馈的模型数目为6,即终端最多能上报6个模型的信息。
终端在接收到网络设备发送的终端能力查询消息之后,可以根据该终端能力查询消息,在终端能力信息(UECapabilityInformation)中反馈相应的模型的信息。
例如,终端能力查询消息中包含应用场景,则终端可以上传该应用场景
对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据UCI反馈资源配置消息,向网络设备发送UCI,UCI中包含第一信息。
具体地,网络设备可以向终端发送UCI反馈资源配置消息(例如可以是RRC配置消息),配置终端的UCI反馈资源,该UCI反馈资源配置消息中可包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的UCI反馈资源配置消息之后,可以根据该UCI反馈资源配置消息,在配置的UCI反馈资源上反馈UCI,其中包括相应的模型的信息。
例如,UCI反馈资源配置消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据RRC请求消息,向网络设备发送RRC响应消息,RRC响应消息中包含第一信息。
具体地,网络设备可以向终端发送RRC请求(request)消息,例如向终端申请上报模型信息,该RRC请求消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的RRC请求消息之后,可以根据该RRC请求消息,通过上行RRC消息,反馈相应的模型的信息。
例如,RRC请求消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据测量配置消息,向网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
具体地,网络设备可以向终端发送测量配置消息,该测量配置消息中可包含测量配置(如Tx beam的配置,参考信号配置等)、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的测量配置消息之后,可以根据该测量配置消息,反馈相应的模型的信息。在一些实施例中,终端可以使用RRC信令、UCI信令或用于传输数据信息的信令,反馈一个或多个模型的信息。
例如,终端可以根据测量配置消息,判断应用场景或网络设备配置,从而选择一个或多个适用于现有应用场景或网络设备配置的模型,将其信息上报给网络设备,从而减少终端上报模型信息的数量。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据模型注册触发消息,向网络设备发送模型注册信息,模型注册信息中包含第一信息。
具体地,网络设备可以向终端发送模型注册触发消息,用于触发模型注册,该模型注册触发消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的模型注册触发消息之后,可以根据该模型注册触发消息,向网络设备发送模型注册信息,其中包括相应的模型的信息。
例如,模型注册触发消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,向网络设备发送第一信息之后,该方法还包括:
接收网络设备发送的RRC配置消息,RRC配置消息中包含一个或多个模型标识;
根据RRC配置消息,确定用于下行波束预测的AI/ML模型。
具体地,网络设备接收到终端上报的第一信息之后,可以根据终端上报的模型信息,通过RRC信令(如RRC配置消息)配置终端可用的模型,配置信息中可以包含模型标识。从而终端在接收到网络设备发送的RRC配置消息之后,可以据此确定哪个或哪些模型用于进行下行波束预测。
例如,网络设备可以基于当前的应用场景或网络设备配置,选择该应用场景或网络设备配置对应的模型,并将相应模型的模型标识配置给终端。
在一些实施例中,网络设备可以使用RRC信令配置一组模型,然后通过媒体接入控制层-控制单元(Media Access Control-Control Element,MAC-CE)激活其中一个模型,用于最优波束预测;或者RRC信令只配置1个模型,用于最优波束预测;或者RRC信令配置一组模型,终端选择其中一个或多个模型,用于最优波束预测。
在一些实施例中,该方法还包括:
接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;
根据RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
具体地,网络设备可以基于当前的应用场景或网络设备配置,选择激活一个或多个模型,具体可以通过RRC信令激活或者通过MAC-CE激活。激活信令中可以包括该激活模型的标识,以及激活指示。
一种实施方式中,当该模型性能不好时(比如和其他模型性能相比较,或者和legacy(之前的)方法性能相比较),网络设备选择切换到其他模型,
或者去激活这个模型,或者fallback(回退)到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
激活/去激活信令中包括该激活/去激活模型的标识,以及激活/去激活指示;fallback信令中包括fallback指示。其中,激活/去激活/fallback指示可以是1比特指示。
图2为本公开实施例提供的模型信息上报方法的流程示意图之二,该方法可应用于网络设备(例如基站),如图2所示,该方法包括如下步骤:
步骤200、接收终端发送的第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/ML模型用于下行波束预测。
具体地,本公开实施例中,为了更好地支持利用AI/ML技术进行波束管理和波束预测,终端可以向网络设备上报第一信息,该第一信息可用于指示一个或多个AI/ML模型,这些AI/ML模型是可用于下行波束预测的AI/ML模型(以下也简称模型)。其中下行波束指的是网络设备与终端之间传输信息所使用的下行波束,包括下行发送波束(Tx beam)、下行接收波束(Rx beam)、下行波束对(beam pair)等。
网络设备接收到终端发送的第一信息之后,便可以获知终端可以使用哪些模型,以及这些模型的相关信息,后续可以在需要的时候根据这些信息进行模型选择激活等流程。
在一些实施例中,第一信息中指示的模型可以是终端自身存储的模型,或者也可以是第三方设备存储的模型,此处不做限定。
对于第三方设备存储模型的情形,终端可以将第三方设备存储的模型信息上报给网络设备,后续根据网络设备的指示(例如配置、激活或选择命令),再从第三方设备下载相应的模型来使用。
在一些实施例中,第一信息中可以包含以下一项或多项:
(1)模型标识。
具体地,第一信息中可以包含模型标识,该模型标识可以是模型的序号或编号,也可以是代表模型类型的标识信息,或者其他可以区分不同模型的标识信息等。
在一些实施例中,模型标识可以包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
其中,用于模型训练的数据集标识、用于模型训练的RS配置标识、用于模型训练的波束描述信息标识、模型的输入波束数量与输出波束数量信息都可以理解为代表模型类型的标识信息,具体说明如下:
数据集标识:数据集是用于模型训练的数据集,终端可以通过不同的数据集训练不同的模型,若数据集和训练的模型是一一对应的,则数据集标识可以代表模型标识。例如,数据集中可以包括数据集标识以及一个或多个数据集样本,其中数据集样本可以包括:多个波束标识,以及该波束标识对应的波束的测量结果,一个或多个最优波束的标识(作为AI/ML模型训练的输出标签)。
RS配置标识:RS配置是用于模型训练的RS配置,终端可以基于RS配置,对下行参考信号进行测量,使用测量结果训练模型,如果RS配置和训练的模型是一一对应的,则RS配置标识可以代表模型标识。例如,RS配置中可以包括RS配置标识,RS种类,RS发送资源,RS标识等。
波束描述信息标识:终端可以使用波束描述信息,以及对下行参考信号进行测量的结果,训练模型,如果波束描述信息和训练的模型是一一对应的,则波束描述信息标识可以代表模型标识。例如,波束描述信息可以包括波束标识或RS标识、各波束标识或RS标识对应的波束的网络设备天线配置、波束的角度信息、波束的宽度信息等。
模型的输入波束数量与输出波束数量信息:一种可能的实现方式中,模型的输入波束数量与输出波束数量信息可以是{输入波束数,输出波束数}对信息,比如{输入Tx beam数,输出Tx beam数}对信息,或者{输入Rx beam数,输出Rx beam数}对信息。终端可以针对每个{输入波束数,输出波束数}
对只训练一个模型,那么{输入波束数,输出波束数}对可以表征模型。
在一些实施例中,一种实施方式中,第一信息中也可以不包含模型标识,网络设备接收终端发送的第一信息之后,该方法还包括:
根据第一信息,向终端发送模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识。
具体地,第一信息中不包含模型标识的情况下,网络设备在接收到终端发送的第一信息之后,可以根据终端上报的这些模型信息,给终端上报的这些模型分配相应的模型标识。
确定分配的模型标识之后,网络设备可以向终端发送模型标识分配消息,该模型标识分配消息中包含网络设备分配的模型标识。在一些实施例中,该模型标识分配消息中还可以包含每个模型标识对应的模型信息(第一信息中的一项或多项),或者其他指示信息,这些指示信息可用于指示所分配的模型标识与第一信息所指示的模型之间的关联关系,从而终端可以根据模型标识分配消息,将网络设备分配的模型标识与第一信息中指示的模型相关联。
一种可能的实现方式中,模型标识分配消息中包含网络设备分配的模型标识,而不包含显式的指示信息指示所分配的模型标识与第一信息所指示的模型之间的关联关系,终端可以在接收到网络设备发送的模型标识分配消息之后,默认按照其上报的第一信息中各个模型信息的顺序,依次将网络设备分配的模型标识与第一信息中指示的模型相关联。
(2)模型的应用场景。
具体地,终端可以上报模型适用的场景,例如,应用场景可以包括:城市,农村,室内,室外,高速公路,高铁,Uma,Umi等等。
终端上报模型的应用场景,可用于后续网络设备根据应用场景选择激活终端使用合适的模型进行波束预测。
(3)模型适用的网络设备配置。
具体地,终端可以上报模型适用的网络设备配置,例如,网络设备配置可以包括:网络设备的天线配置,波束配置,参考信号配置等等。
终端上报模型适用的网络设备配置,可用于后续网络设备根据网络设备
配置选择激活终端使用合适的模型进行波束预测。
(4)模型功能。
具体地,终端可以上报模型的功能,例如,模型功能可以包括:空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等等。空域预测是指测量少量的波束,或者其他参考信号类型的波束,预测大量波束中的最优波束;频域预测是指测量频率1上的波束,预测另一个频率2上的波束中的最优波束;时域预测是指测量现在时刻的波束,预测未来时刻的最优波束。
终端上报模型的功能,可用于后续网络设备根据模型功能选择激活终端使用合适的模型进行波束预测。
(5)模型的输入波束数量。
具体地,终端可以上报模型的输入波束数量,比如模型的输入Tx beam数,或者模型的输入Rx beam数,或者模型的输入beam pair数等。
终端上报模型的输入波束数量,便于让网络设备在选择终端使用的模型后,进行配置和发送Tx beam,提升终端预测性能。
(6)模型的输出波束数量。
具体地,终端可以上报模型的输出波束数量,比如模型的输出Tx beam数,或者模型的输出Rx beam数,或者模型的输出beam pair数等。
终端上报模型的输出波束数量,便于让网络设备后续调度该终端对模型输出的最优波束进行反馈,例如指示终端反馈的Tx beam数目。
(7)模型适用的下行参考信号RS类型。
具体地,终端可以上报模型适用的下行参考信号类型,下行参考信号类型可以包括:CSI-RS,SSB,PT-RS,CRS,DMRS等等。
(8)模型适用的下行波束发送顺序。
具体地,终端可以上报模型适用的下行波束发送顺序,便于让网络设备在选择终端使用的模型后,进行配置和发送Tx beam,提升波束预测性能。
(9)模型适用的下行波束图样。
具体地,终端可以上报模型适用的下行波束图样(pattern),便于让网络
设备在选择终端使用的模型后,进行配置和发送Tx beam,提升波束预测性能。
(10)模型适用的下行波束标识与物理的下行波束之间的映射关系。
具体地,终端可以上报模型适用的下行波束标识与物理的下行波束之间的映射关系,可用于让网络设备理解终端预测最优波束后上报的波束标识(如Tx beam ID),并在其对应的物理波束上发送后续数据信息。
(11)模型适用的下行发送波束的标识。
具体地,终端可以上报模型适用的下行发送波束的标识。例如,对于下行Rx beam预测场景,终端可以通过上报模型适用的下行发送波束的标识,建议网络设备发送哪个Tx beam,以便于终端使用不同的Rx beam接收该Tx beam发送的参考信号进行测量,并根据测量结果预测出最优Rx beam。
(12)模型适用的下行发送波束的发送次数。
具体地,终端可以上报模型适用的下行发送波束的发送次数。例如,对于下行Rx beam预测场景,终端可以通过上报模型适用的下行发送波束的发送次数,建议网络设备发送的Tx beam,以及在该Tx beam上发送几次参考信号,以便于终端使用不同的Rx beam接收该Tx beam发送的参考信号进行测量,并根据测量结果预测出最优Rx beam。
本公开实施例提供的模型信息上报方法,网络设备可以接收终端发送的一个或多个AI/ML模型的相关信息,从而后续可以按需激活模型的使用和配置Tx beam的发送图样,提升系统传输性能,能够更好地支持利用AI/ML技术进行波束管理和波束预测。
在一些实施例中,网络设备接收终端发送的第一信息,可以包括:
接收终端发送的第一信令,第一信令中包含第一信息;
其中,第一信令可以包括以下一种或多种:
(1)终端能力信令。例如,终端可以通过终端能力信息上报第一信息。
(2)UCI信令。该UCI可以是PUCCH或PUSCH上传输的UCI。
(3)RRC信令。
(4)用于传输数据信息的信令。
(5)用于模型注册的信令。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的终端能力信息,终端能力信息中包含第一信息。
具体地,网络设备可以向终端发送终端能力查询(UECapabilityEnquiry)消息,向终端查询终端能力信息,该终端能力查询消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的终端能力查询消息之后,可以根据该终端能力查询消息,在终端能力信息(UECapabilityInformation)中反馈相应的模型的信息。
例如,终端能力查询消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的UCI,UCI中包含第一信息。
具体地,网络设备可以向终端发送UCI反馈资源配置消息(例如可以是RRC配置消息),配置终端的UCI反馈资源,该UCI反馈资源配置消息中可包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的UCI反馈资源配置消息之后,可以根据该UCI反馈资源配置消息,在配置的UCI反馈资源上反馈UCI,其中包括相应的模型的信息。
例如,UCI反馈资源配置消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终
端上报模型信息的数量。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC响应消息,RRC响应消息中包含第一信息。
具体地,网络设备可以向终端发送RRC请求(request)消息,例如向终端申请上报模型信息,该RRC请求消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的RRC请求消息之后,可以根据该RRC请求消息,通过上行RRC消息,反馈相应的模型的信息。
例如,RRC请求消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
具体地,网络设备可以向终端发送测量配置消息,该测量配置消息中可包含测量配置(如Tx beam的配置,参考信号配置等)、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的测量配置消息之后,可以根据该测量配置消息,反馈相应的模型的信息。在一些实施例中,终端可以使用RRC信令、UCI信令或用于传输数据信息的信令,反馈一个或多个模型的信息。
例如,终端可以根据测量配置消息,判断应用场景或网络设备配置,从而选择一个或多个适用于现有应用场景或网络设备配置的模型,将其信息上报给网络设备,从而减少终端上报模型信息的数量。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的模型注册信息,模型注册信息中包含第一信息。
具体地,网络设备可以向终端发送模型注册触发消息,用于触发模型注册,该模型注册触发消息中可包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项。
终端在接收到网络设备发送的模型注册触发消息之后,可以根据该模型注册触发消息,向网络设备发送模型注册信息,其中包括相应的模型的信息。
例如,模型注册触发消息中包含应用场景,则终端可以上传该应用场景对应的模型信息,其他应用场景的模型信息可以不用上传,从而减少终端上报模型信息的数量。
在一些实施例中,接收终端发送的第一信息之后,该方法还包括:
根据第一信息,向终端发送RRC配置消息,RRC配置消息中包含一个或多个模型标识。
具体地,网络设备接收到终端上报的第一信息之后,可以根据终端上报的模型信息,通过RRC信令(如RRC配置消息)配置终端可用的模型,配置信息中可以包含模型标识。从而终端在接收到网络设备发送的RRC配置消息之后,可以据此确定哪个或哪些模型用于进行下行波束预测。
例如,网络设备可以基于当前的应用场景或网络设备配置,选择该应用场景或网络设备配置对应的模型,并将相应模型的模型标识配置给终端。
在一些实施例中,网络设备可以使用RRC信令配置一组模型,然后通过MAC-CE激活其中一个模型,用于最优波束预测;或者RRC信令只配置1个模型,用于最优波束预测;或者RRC信令配置一组模型,终端选择其中一个或多个模型,用于最优波束预测。
在一些实施例中,该方法还包括:
向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
具体地,网络设备可以基于当前的应用场景或网络设备配置,选择激活一个或多个模型,具体可以通过RRC信令激活或者通过MAC-CE激活。激活信令中可以包括该激活模型的标识,以及激活指示。
一种实施方式中,当该模型性能不好时(比如和其他模型性能相比较,或者和legacy(之前的)方法性能相比较),网络设备选择切换到其他模型,或者去激活这个模型,或者fallback(回退)到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
激活/去激活信令中包括该激活/去激活模型的标识,以及激活/去激活指示;fallback信令中包括fallback指示。其中,激活/去激活/fallback指示可以是1比特指示。
本公开各实施例提供的方法是基于同一申请构思的,因此各方法的实施可以相互参见,重复之处不再赘述。
以下通过具体应用场景的实施例对本公开各上述实施例提供的方法进行举例说明。
实施例1:UE通过UE Capability信令上报模型(model)信息。
图3为本公开实施例提供的模型信息上报方法的实施示意图之一,如图3所示,其具体步骤如下:
步骤一:网络设备发送UECapabilityEnquiry消息,向UE查询UE能力信息。
其中,UECapabilityEnquiry消息包括:应用场景,网络设备配置,model功能,最大反馈的model数目中的一项或多项。
应用场景包括:城市,农村,室内,室外,高速公路,高铁,Uma,Umi等等。
网络设备配置包括:网络设备天线配置,波束配置,参考信号配置等等。
model功能,主要包括空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等。
最大反馈的model数目:是指UE可以反馈的最大model个数,例如最大反馈的model数目为6,即UE最多能上报6个model的信息。
步骤二:UE根据基站发送的UECapabilityEnquiry消息,在UECapabilityInformation信息中反馈相应的model的信息。
例如,UECapabilityEnquiry消息中,包含应用场景,则UE可以上传该应用场景对应的model信息,其他应用场景的model信息可以不用上传,从而减少UE上报model信息的数量。
具体反馈的model信息内容请见上述各方法实施例描述的第一信息。
步骤三:网络设备根据UE上报的model信息,通过RRC配置UE可用的model,配置信息包括model ID。
网络设备根据UE上报的model信息,以及现有的场景和配置,选择对应的model,并配置给UE。
例如,网络设备基于现有的应用场景或网络设备配置,选择该应用场景或网络设备配置对应的model。
网络设备可以使用RRC配置一组model,然后通过MAC CE激活其中一个model,用于最优波束预测;或者RRC只配置1个model,用于最优波束预测;或者RRC配置一组model,UE选择其中一个或多个model,用于最优波束预测。
步骤四:网络设备根据选择或激活的model信息,例如输入Tx beam数目,Tx beam发送顺序,Tx beam发送pattern等信息,配置相关下行参考信号,并进行Tx beam发送,并且根据model输出的Tx beam数目,配置UE的反馈信息。
步骤五:UE根据网络设备配置,反馈model输出的Tx beam,例如反馈最好的best beam ID(可选的,增加best beam的RSRP反馈),或者反馈N个最好的Top-N best beam ID(可选的,增加Top-N best beam的RSRP反馈),或者反馈N个beam ID(最好的beam包括在其中,可选的,增加N个beam的RSRP反馈)。
步骤六:当该model性能不好时(比如和其他model性能相比较,或者和legacy方法性能相比较),网络设备选择切换到其他model,或者去激活这个model,或者fallback到legacy方法,可以通过RRC激活/去激活/fallback,
或者通过MAC CE激活/去激活/fallback。
激活/去激活/fallback信令包括该激活/去激活model的ID或者是model序号(可以节省比特数),以及1比特激活/去激活/fallback指示。
实施例2:UE通过UCI在该反馈资源上发送model信息。
图4为本公开实施例提供的模型信息上报方法的实施示意图之二,如图4所示,其具体步骤如下:
步骤一:网络发送UCI反馈资源配置消息,向UE配置UCI反馈资源。
其中UCI反馈资源配置消息包括:UCI反馈资源时频位置,应用场景,网络设备配置,model功能,最大反馈的model数目中的一项或多项。
应用场景包括:城市,农村,室内,室外,高速公路,高铁,Uma,Umi等等。
网络设备配置包括:网络设备天线配置,波束配置,参考信号配置等等。
model功能,主要包括空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等。
最大反馈的model数目:是指UE可以反馈的最大model个数,例如最大反馈的model数目为6,即UE最多能上报6个model的信息。
步骤二:UE根据网络设备发送的UCI反馈资源配置消息,在该资源上反馈UCI,其中包括相应的model的信息。
例如,UCI反馈资源配置消息中,包含应用场景,则UE可以上传该应用场景对应的model信息,其他应用场景的model信息可以不用上传,从而减少UE上报model信息的数量。
具体反馈的model信息内容请见上述各方法实施例描述的第一信息。
以下步骤三到步骤六同实施例一。
步骤三:网络设备根据UE上报的model信息,通过RRC配置UE可用的model,配置信息包括model ID。
步骤四:网络设备根据选择或激活的model信息,例如输入Tx beam数目,Tx beam发送顺序,Tx beam发送pattern等信息,配置相关下行参考信号,并进行Tx beam发送,并且根据model输出的Tx beam数目,配置UE
的反馈信息。
步骤五:UE根据网络设备配置,反馈model输出的Tx beam,例如反馈最好的best beam ID(可选的,增加best beam的RSRP反馈),或者反馈N个最好的Top-N best beam ID(可选的,增加Top-N best beam的RSRP反馈),或者反馈N个beam ID(最好的beam包括在其中,可选的,增加N个beam的RSRP反馈)。
步骤六:当该model性能不好时(比如和其他model性能相比较,或者和legacy方法性能相比较),网络设备选择切换到其他model,或者去激活这个model,或者fallback到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
实施例3:UE通过RRC信令反馈model信息。
图5为本公开实施例提供的模型信息上报方法的实施示意图之三,如图5所示,其具体步骤如下:
步骤一:网络发送RRC request消息,向UE申请上报model信息。
其中RRC request消息包括:应用场景,网络设备配置,model功能,最大反馈的model数目中的一项或多项。
应用场景包括:城市,农村,室内,室外,高速公路,高铁,Uma,Umi等等。
网络设备配置包括:网络设备天线配置,波束配置,参考信号配置等等。
model功能,主要包括空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等。
最大反馈的model数目:是指UE可以反馈的最大model个数,例如最大反馈的model数目为6,即UE最多能上报6个model的信息。
步骤二:UE根据网络设备发送的RRC request消息,通过上行RRC消息,反馈相应的model的信息。
例如,RRC request消息中,包含应用场景,则UE可以上传该应用场景对应的model信息,其他应用场景的model信息可以不用上传,从而减少UE上报model信息的数量。
具体反馈的model信息内容请见上述各方法实施例描述的第一信息。
以下步骤三到步骤六同实施例一。
步骤三:网络设备根据UE上报的model信息,通过RRC配置UE可用的model,配置信息包括model ID。
步骤四:网络设备根据选择或激活的model信息,例如输入Tx beam数目,Tx beam发送顺序,Tx beam发送pattern等信息,配置相关下行参考信号,并进行Tx beam发送,并且根据model输出的Tx beam数目,配置UE的反馈信息。
步骤五:UE根据网络设备配置,反馈model输出的Tx beam,例如反馈最好的best beam ID(可选的,增加best beam的RSRP反馈),或者反馈N个最好的Top-N best beam ID(可选的,增加Top-N best beam的RSRP反馈),或者反馈N个beam ID(最好的beam包括在其中,可选的,增加N个beam的RSRP反馈)。
步骤六:当该model性能不好时(比如和其他model性能相比较,或者和legacy方法性能相比较),网络设备选择切换到其他model,或者去激活这个model,或者fallback到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
实施例4:UE根据测量配置消息,反馈model信息。
图6为本公开实施例提供的模型信息上报方法的实施示意图之四,如图6所示,其具体步骤如下:
步骤一:网络发送测量配置消息,包括Tx beam的配置,参考信号配置等。
步骤二:UE根据测量配置消息,反馈相应的model的信息。
例如,UE根据测量配置信息,判断应用场景或网络设备配置,从而选择一个或多个适用于现有应用场景或网络设备配置的model,将其信息上报给网络设备,从而减少UE上报model信息的数量。
具体反馈的model信息内容请见上述各方法实施例描述的第一信息。
具体可以使用RRC信令,UCI信令,用于传输数据信息的信令,反馈该
一个或多个model。
以下步骤三到步骤六同实施例一。
步骤三:网络设备根据UE上报的model信息,通过RRC配置UE可用的model,配置信息包括model ID。
步骤四:网络设备根据选择或激活的model信息,例如输入Tx beam数目,Tx beam发送顺序,Tx beam发送pattern等信息,配置相关下行参考信号,并进行Tx beam发送,并且根据model输出的Tx beam数目,配置UE的反馈信息。
步骤五:UE根据网络设备配置,反馈model输出的Tx beam,例如反馈最好的best beam ID(可选的,增加best beam的RSRP反馈),或者反馈N个最好的Top-N best beam ID(可选的,增加Top-N best beam的RSRP反馈),或者反馈N个beam ID(最好的beam包括在其中,可选的,增加N个beam的RSRP反馈)。
步骤六:当该model性能不好时(比如和其他model性能相比较,或者和legacy方法性能相比较),网络设备选择切换到其他model,或者去激活这个model,或者fallback到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
实施例5:UE通过用于模型注册的信令反馈model信息。
图7为本公开实施例提供的模型信息上报方法的实施示意图之五,如图7所示,其具体步骤如下:
步骤一:网络发送模型注册触发消息,触发UE注册模型。
其中模型注册触发消息包括:应用场景,网络设备配置,model功能,最大反馈的model数目中的一项或多项。
应用场景包括:城市,农村,室内,室外,高速公路,高铁,Uma,Umi等等。
网络设备配置包括:网络设备天线配置,波束配置,参考信号配置等等。
model功能,主要包括空域预测,频域预测,时域预测,下行beam pair预测,下行Tx beam预测,下行Rx beam预测等。
最大反馈的model数目:是指UE可以反馈的最大model个数,例如最大反馈的model数目为6,即UE最多能上报6个model的信息。
步骤二:UE根据网络设备发送的模型注册触发消息,通过模型注册信息,反馈相应的model的信息。
例如,模型注册触发消息中,包含应用场景,则UE可以上传该应用场景对应的model信息,其他应用场景的model信息可以不用上传,从而减少UE上报model信息的数量。
具体反馈的model信息内容请见上述各方法实施例描述的第一信息。
以下步骤三到步骤六同实施例一。
步骤三:网络设备根据UE上报的model信息,通过RRC配置UE可用的model,配置信息包括model ID。
步骤四:网络设备根据选择或激活的model信息,例如输入Tx beam数目,Tx beam发送顺序,Tx beam发送pattern等信息,配置相关下行参考信号,并进行Tx beam发送,并且根据model输出的Tx beam数目,配置UE的反馈信息。
步骤五:UE根据网络设备配置,反馈model输出的Tx beam,例如反馈最好的best beam ID(可选的,增加best beam的RSRP反馈),或者反馈N个最好的Top-N best beam ID(可选的,增加Top-N best beam的RSRP反馈),或者反馈N个beam ID(最好的beam包括在其中,可选的,增加N个beam的RSRP反馈)。
步骤六:当该model性能不好时(比如和其他model性能相比较,或者和legacy方法性能相比较),网络设备选择切换到其他model,或者去激活这个model,或者fallback到legacy方法,可以通过RRC激活/去激活/fallback,或者通过MAC CE激活/去激活/fallback。
本公开各实施例提供的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。
图8为本公开实施例提供的终端的结构示意图,如图8所示,该终端包
括存储器820,收发机810和处理器800;其中,处理器800与存储器820也可以物理上分开布置。
存储器820,用于存储计算机程序;收发机810,用于在处理器800的控制下收发数据。
具体地,收发机810用于在处理器800的控制下接收和发送数据。
其中,在图8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800代表的一个或多个处理器和存储器820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,本公开不再对其进行进一步描述。总线接口提供接口。收发机810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830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800负责管理总线架构和通常的处理,存储器820可以存储处理器800在执行操作时所使用的数据。
处理器800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800通过调用存储器820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,例如:确定第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/ML模型用于下行波束预测;向网络设备发送第一信息。
在一些实施例中,第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,向网络设备发送第一信息,包括:
向网络设备发送第一信令,第一信令中包含第一信息;
其中,第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据终端能力查询消息,向网络设备发送终端能力信息,终端能力信息中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据UCI反馈资源配置消息,向网络设备发送UCI,UCI中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据RRC请求消息,向网络设备发送RRC响应消息,RRC响应消息中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据测量配置消息,向网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据模型注册触发消息,向网络设备发送模型注册信息,模型注册信息中包含第一信息。
在一些实施例中,向网络设备发送第一信息之后,该方法还包括:
接收网络设备发送的RRC配置消息,RRC配置消息中包含一个或多个模型标识;
根据RRC配置消息,确定用于下行波束预测的AI/ML模型。
在一些实施例中,该方法还包括:
接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,
RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;
根据RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
在一些实施例中,向网络设备发送第一信息之后,该方法还包括:
接收网络设备发送的模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识;
根据模型标识分配消息,将网络设备分配的模型标识与第一信息中指示的模型相关联。
图9为本公开实施例提供的网络设备的结构示意图,如图9所示,该网络设备包括存储器920,收发机910和处理器900;其中,处理器900与存储器920也可以物理上分开布置。
存储器920,用于存储计算机程序;收发机910,用于在处理器900的控制下收发数据。
具体地,收发机910用于在处理器900的控制下接收和发送数据。
其中,在图9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900代表的一个或多个处理器和存储器920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,本公开不再对其进行进一步描述。总线接口提供接口。收发机910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900负责管理总线架构和通常的处理,存储器920可以存储处理器900在执行操作时所使用的数据。
处理器900可以是CPU、ASIC、FPGA或CPLD,处理器也可以采用多核架构。
处理器900通过调用存储器920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,例如:接收终端发送的第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/
ML模型用于下行波束预测。
在一些实施例中,第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,接收终端发送的第一信息,包括:
接收终端发送的第一信令,第一信令中包含第一信息;
其中,第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的终端能力信息,终端能力信息中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的UCI,UCI中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC响应消息,RRC响应消息中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的模型注册信息,模型注册信息中包含第一信息。
在一些实施例中,接收终端发送的第一信息之后,该方法还包括:
根据第一信息,向终端发送RRC配置消息,RRC配置消息中包含一个或多个模型标识。
在一些实施例中,该方法还包括:
向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或
多项。
在一些实施例中,接收终端发送的第一信息之后,该方法还包括:
根据第一信息,向终端发送模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识。
在此需要说明的是,本公开实施例提供的上述终端和网络设备,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10为本公开实施例提供的模型信息上报装置的结构示意图之一,如图10所示,该装置包括:
确定单元1000,用于确定第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/ML模型用于下行波束预测;
第一发送单元1010,用于向网络设备发送第一信息。
在一些实施例中,第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,向网络设备发送第一信息,包括:
向网络设备发送第一信令,第一信令中包含第一信息;
其中,第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据终端能力查询消息,向网络设备发送终端能力信息,终端能力信息中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据UCI反馈资源配置消息,向网络设备发送UCI,UCI中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据RRC请求消息,向网络设备发送RRC响应消息,RRC响应消息中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据测量配置消息,向网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
在一些实施例中,向网络设备发送第一信息,包括:
接收网络设备发送的模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
根据模型注册触发消息,向网络设备发送模型注册信息,模型注册信息中包含第一信息。
在一些实施例中,该装置还包括第一接收单元,用于:
接收网络设备发送的RRC配置消息,RRC配置消息中包含一个或多个模型标识;
根据RRC配置消息,确定用于下行波束预测的AI/ML模型。
在一些实施例中,该装置还包括第二接收单元,用于:
接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;
根据RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
在一些实施例中,该装置还包括关联单元,用于:
接收网络设备发送的模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识;
根据模型标识分配消息,将网络设备分配的模型标识与第一信息中指示的模型相关联。
图11为本公开实施例提供的模型信息上报装置的结构示意图之二,如图11所示,该装置包括:
第三接收单元1100,用于接收终端发送的第一信息,第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,AI/ML模型用于下行波束预测。
在一些实施例中,第一信息中包含以下一项或多项:
模型标识;
模型的应用场景;
模型适用的网络设备配置;
模型功能;
模型的输入波束数量;
模型的输出波束数量;
模型适用的下行参考信号RS类型;
模型适用的下行波束发送顺序;
模型适用的下行波束图样;
模型适用的下行波束标识与物理的下行波束之间的映射关系;
模型适用的下行发送波束的标识;
模型适用的下行发送波束的发送次数。
在一些实施例中,模型标识包括以下一种或多种:
模型的序号或编号;
用于模型训练的数据集标识;
用于模型训练的RS配置标识;
用于模型训练的波束描述信息标识;
模型的输入波束数量与输出波束数量信息。
在一些实施例中,接收终端发送的第一信息,包括:
接收终端发送的第一信令,第一信令中包含第一信息;
其中,第一信令包括以下一种或多种:
终端能力信令;
上行控制信息UCI信令;
无线资源控制RRC信令;
用于传输数据信息的信令;
用于模型注册的信令。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送终端能力查询消息,终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的终端能力信息,终端能力信息中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送UCI反馈资源配置消息,UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的UCI,UCI中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送RRC请求消息,RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC响应消息,RRC响应消息中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送测量配置消息,测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的RRC信令、UCI信令或用于传输数据信息的信令,RRC信令、UCI信令或用于传输数据信息的信令中包含第一信息。
在一些实施例中,接收终端发送的第一信息,包括:
向终端发送模型注册触发消息,模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;
接收终端发送的模型注册信息,模型注册信息中包含第一信息。
在一些实施例中,该装置还包括:
第二发送单元,用于根据第一信息,向终端发送RRC配置消息,RRC配置消息中包含一个或多个模型标识。
在一些实施例中,该装置还包括:
第三发送单元,用于向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
在一些实施例中,该装置还包括分配单元,用于:
根据第一信息,向终端发送模型标识分配消息,模型标识分配消息中包含网络设备分配的模型标识。
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计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序用于使计算机执行上述各实施例提供的模型信息上报方法。
在此需要说明的是,本公开实施例提供的计算机可读存储介质,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。
所述计算机可读存储介质可以是计算机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。
本公开实施例涉及的终端,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,终端的名称可能也不相同,例如在5G系统中,终端可以称为用户设备(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。
网络设备与终端之间可以各自使用一或多根天线进行多输入多输出(Multi Input Multi Output,MIMO)传输,MIMO传输可以是单用户MIMO(Single User MIMO,SU-MIMO)或多用户MIMO(Multiple User MIMO,MU-MIMO)。根据根天线组合的形态和数量,MIMO传输可以是2D-MIMO、3D-MIMO、FD-MIMO或massive-MIMO,也可以是分集传输或预编码传输或波束赋形传输等。
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 (73)
- 一种模型信息上报方法,应用于终端,包括:确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;向网络设备发送所述第一信息。
- 根据权利要求1所述的模型信息上报方法,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求2所述的模型信息上报方法,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求1至3任一项所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:向网络设备发送第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求4所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述终端能力查询消息,向所述网络设备发送终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求4所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述UCI反馈资源配置消息,向所述网络设备发送UCI,所述UCI中包含所述第一信息。
- 根据权利要求4所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述RRC请求消息,向所述网络设备发送RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求4所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述测量配置消息,向所述网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求4所述的模型信息上报方法,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述模型注册触发消息,向所述网络设备发送模型注册信息,所述模型注册信息中包含所述第一信息。
- 根据权利要求1或2所述的模型信息上报方法,其中,向网络设备发送所述第一信息之后,所述方法还包括:接收网络设备发送的RRC配置消息,所述RRC配置消息中包含一个或多个模型标识;根据所述RRC配置消息,确定用于下行波束预测的AI/ML模型。
- 根据权利要求1或2所述的模型信息上报方法,其中,所述方法还包括:接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;根据所述RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
- 根据权利要求1或2所述的模型信息上报方法,其中,向网络设备发送所述第一信息之后,所述方法还包括:接收所述网络设备发送的模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识;根据所述模型标识分配消息,将所述网络设备分配的模型标识与所述第一信息中指示的模型相关联。
- 一种模型信息上报方法,应用于网络设备,包括:接收终端发送的第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
- 根据权利要求13所述的模型信息上报方法,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求14所述的模型信息上报方法,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求13至15任一项所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:接收终端发送的第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求16所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:向终端发送终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求16所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:向终端发送UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的UCI,所述UCI中包含所述第一信息。
- 根据权利要求16所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:向终端发送RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求16所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:向终端发送测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC信令、UCI信令或用于传输数据信息的信令, 所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求16所述的模型信息上报方法,其中,所述接收终端发送的第一信息,包括:向终端发送模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的模型注册信息,所述模型注册信息中包含所述第一信息。
- 根据权利要求13或14所述的模型信息上报方法,其中,接收终端发送的第一信息之后,所述方法还包括:根据所述第一信息,向所述终端发送RRC配置消息,所述RRC配置消息中包含一个或多个模型标识。
- 根据权利要求13或14所述的模型信息上报方法,其中,所述方法还包括:向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
- 根据权利要求13或14所述的模型信息上报方法,其中,接收终端发送的第一信息之后,所述方法还包括:根据所述第一信息,向所述终端发送模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识。
- 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:确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;向网络设备发送所述第一信息。
- 根据权利要求25所述的终端,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求26所述的终端,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求25至27任一项所述的终端,其中,所述向网络设备发送所述第一信息,包括:向网络设备发送第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求28所述的终端,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述终端能力查询消息,向所述网络设备发送终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求28所述的终端,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述UCI反馈资源配置消息,向所述网络设备发送UCI,所述UCI中包含所述第一信息。
- 根据权利要求28所述的终端,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述RRC请求消息,向所述网络设备发送RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求28所述的终端,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述测量配置消息,向所述网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求28所述的终端,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述模型注册触发消息,向所述网络设备发送模型注册信息,所述模型注册信息中包含所述第一信息。
- 根据权利要求25或26所述的终端,其中,向网络设备发送所述第一信息之后,所述操作还包括:接收网络设备发送的RRC配置消息,所述RRC配置消息中包含一个或多个模型标识;根据所述RRC配置消息,确定用于下行波束预测的AI/ML模型。
- 根据权利要求25或26所述的终端,其中,所述操作还包括:接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;根据所述RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
- 根据权利要求25或26所述的终端,其中,向网络设备发送所述第一信息之后,所述操作还包括:接收所述网络设备发送的模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识;根据所述模型标识分配消息,将所述网络设备分配的模型标识与所述第一信息中指示的模型相关联。
- 一种网络设备,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信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
- 根据权利要求37所述的网络设备,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求38所述的网络设备,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求37至39任一项所述的网络设备,其中,所述接收终端发送的第一信息,包括:接收终端发送的第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求40所述的网络设备,其中,所述接收终端发送的第一 信息,包括:向终端发送终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求40所述的网络设备,其中,所述接收终端发送的第一信息,包括:向终端发送UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的UCI,所述UCI中包含所述第一信息。
- 根据权利要求40所述的网络设备,其中,所述接收终端发送的第一信息,包括:向终端发送RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求40所述的网络设备,其中,所述接收终端发送的第一信息,包括:向终端发送测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求40所述的网络设备,其中,所述接收终端发送的第一信息,包括:向终端发送模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的模型注册信息,所述模型注册信息中包含所述第一 信息。
- 根据权利要求37或38所述的网络设备,其中,接收终端发送的第一信息之后,所述操作还包括:根据所述第一信息,向所述终端发送RRC配置消息,所述RRC配置消息中包含一个或多个模型标识。
- 根据权利要求37或38所述的网络设备,其中,所述操作还包括:向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项。
- 根据权利要求37或38所述的网络设备,其中,接收终端发送的第一信息之后,所述操作还包括:根据所述第一信息,向所述终端发送模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识。
- 一种模型信息上报装置,包括:确定单元,用于确定第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测;第一发送单元,用于向网络设备发送所述第一信息。
- 根据权利要求49所述的模型信息上报装置,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求50所述的模型信息上报装置,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求49至51任一项所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:向网络设备发送第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求52所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述终端能力查询消息,向所述网络设备发送终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求52所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的UCI反馈资源配置消息,所述UCI反馈资源配置消 息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述UCI反馈资源配置消息,向所述网络设备发送UCI,所述UCI中包含所述第一信息。
- 根据权利要求52所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述RRC请求消息,向所述网络设备发送RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求52所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述测量配置消息,向所述网络设备发送RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求52所述的模型信息上报装置,其中,所述向网络设备发送所述第一信息,包括:接收网络设备发送的模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;根据所述模型注册触发消息,向所述网络设备发送模型注册信息,所述模型注册信息中包含所述第一信息。
- 根据权利要求49或50所述的模型信息上报装置,其中,所述装置还包括第一接收单元,用于:接收网络设备发送的RRC配置消息,所述RRC配置消息中包含一个或多个模型标识;根据所述RRC配置消息,确定用于下行波束预测的AI/ML模型。
- 根据权利要求49或50所述的模型信息上报装置,其中,所述装置还包括第二接收单元,用于:接收网络设备发送的RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活指示、回退指示中的一项或多项;根据所述RRC信令或MAC-CE,执行模型激活或去激活或回退操作。
- 根据权利要求49或50所述的模型信息上报装置,其中,所述装置还包括关联单元,用于:接收所述网络设备发送的模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识;根据所述模型标识分配消息,将所述网络设备分配的模型标识与所述第一信息中指示的模型相关联。
- 一种模型信息上报装置,包括:第三接收单元,用于接收终端发送的第一信息,所述第一信息用于指示一个或多个人工智能或机器学习AI/ML模型,所述AI/ML模型用于下行波束预测。
- 根据权利要求61所述的模型信息上报装置,其中,所述第一信息中包含以下一项或多项:模型标识;模型的应用场景;模型适用的网络设备配置;模型功能;模型的输入波束数量;模型的输出波束数量;模型适用的下行参考信号RS类型;模型适用的下行波束发送顺序;模型适用的下行波束图样;模型适用的下行波束标识与物理的下行波束之间的映射关系;模型适用的下行发送波束的标识;模型适用的下行发送波束的发送次数。
- 根据权利要求62所述的模型信息上报装置,其中,所述模型标识包括以下一种或多种:模型的序号或编号;用于模型训练的数据集标识;用于模型训练的RS配置标识;用于模型训练的波束描述信息标识;模型的输入波束数量与输出波束数量信息。
- 根据权利要求61至63任一项所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:接收终端发送的第一信令,所述第一信令中包含所述第一信息;其中,所述第一信令包括以下一种或多种:终端能力信令;上行控制信息UCI信令;无线资源控制RRC信令;用于传输数据信息的信令;用于模型注册的信令。
- 根据权利要求64所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:向终端发送终端能力查询消息,所述终端能力查询消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的终端能力信息,所述终端能力信息中包含所述第一信息。
- 根据权利要求64所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:向终端发送UCI反馈资源配置消息,所述UCI反馈资源配置消息中包含UCI反馈资源时频位置、应用场景、网络设备配置、模型功能、最大反馈的 模型数目中的一项或多项;接收所述终端发送的UCI,所述UCI中包含所述第一信息。
- 根据权利要求64所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:向终端发送RRC请求消息,所述RRC请求消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC响应消息,所述RRC响应消息中包含所述第一信息。
- 根据权利要求64所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:向终端发送测量配置消息,所述测量配置消息中包含测量配置、应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的RRC信令、UCI信令或用于传输数据信息的信令,所述RRC信令、UCI信令或用于传输数据信息的信令中包含所述第一信息。
- 根据权利要求64所述的模型信息上报装置,其中,所述接收终端发送的第一信息,包括:向终端发送模型注册触发消息,所述模型注册触发消息中包含应用场景、网络设备配置、模型功能、最大反馈的模型数目中的一项或多项;接收所述终端发送的模型注册信息,所述模型注册信息中包含所述第一信息。
- 根据权利要求61或62所述的模型信息上报装置,其中,所述装置还包括:第二发送单元,用于根据所述第一信息,向所述终端发送RRC配置消息,所述RRC配置消息中包含一个或多个模型标识。
- 根据权利要求61或62所述的模型信息上报装置,其中,所述装置还包括:第三发送单元,用于向终端发送RRC信令或媒体接入控制层-控制单元MAC-CE,所述RRC信令或MAC-CE中包含模型标识、激活指示、去激活 指示、回退指示中的一项或多项。
- 根据权利要求61或62所述的模型信息上报装置,其中,所述装置还包括分配单元,用于:根据所述第一信息,向所述终端发送模型标识分配消息,所述模型标识分配消息中包含所述网络设备分配的模型标识。
- 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序用于使计算机执行权利要求1至12任一项所述的方法,或执行权利要求13至24任一项所述的方法。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211215756.9A CN117858117A (zh) | 2022-09-30 | 2022-09-30 | 模型信息上报方法、设备、装置及存储介质 |
CN202211215756.9 | 2022-09-30 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024067098A1 true WO2024067098A1 (zh) | 2024-04-04 |
Family
ID=90476084
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2023/118607 WO2024067098A1 (zh) | 2022-09-30 | 2023-09-13 | 模型信息上报方法、设备、装置及存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN117858117A (zh) |
WO (1) | WO2024067098A1 (zh) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021211703A1 (en) * | 2020-04-16 | 2021-10-21 | Qualcomm Incorporated | Machine learning model selection in beamformed communications |
US20210345134A1 (en) * | 2018-10-19 | 2021-11-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Handling of machine learning to improve performance of a wireless communications network |
CN114144977A (zh) * | 2019-08-06 | 2022-03-04 | 华为技术有限公司 | 波束赋形方法、装置、无线接入网设备及可读存储介质 |
CN114938712A (zh) * | 2022-04-13 | 2022-08-23 | 北京小米移动软件有限公司 | 波束选择方法和装置 |
-
2022
- 2022-09-30 CN CN202211215756.9A patent/CN117858117A/zh active Pending
-
2023
- 2023-09-13 WO PCT/CN2023/118607 patent/WO2024067098A1/zh unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20210345134A1 (en) * | 2018-10-19 | 2021-11-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Handling of machine learning to improve performance of a wireless communications network |
CN114144977A (zh) * | 2019-08-06 | 2022-03-04 | 华为技术有限公司 | 波束赋形方法、装置、无线接入网设备及可读存储介质 |
WO2021211703A1 (en) * | 2020-04-16 | 2021-10-21 | Qualcomm Incorporated | Machine learning model selection in beamformed communications |
CN114938712A (zh) * | 2022-04-13 | 2022-08-23 | 北京小米移动软件有限公司 | 波束选择方法和装置 |
Also Published As
Publication number | Publication date |
---|---|
CN117858117A (zh) | 2024-04-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022152068A1 (zh) | 资源确定方法、通信设备和存储介质 | |
TW202241106A (zh) | 傳輸配置指示tci狀態的確定方法、裝置及終端設備 | |
WO2022042294A1 (zh) | 波束指示方法、网络设备、终端、装置及存储介质 | |
WO2023202658A1 (zh) | 信息处理方法、网络侧设备及终端 | |
WO2024067098A1 (zh) | 模型信息上报方法、设备、装置及存储介质 | |
WO2023207744A1 (zh) | 资源调度方法、设备、装置及存储介质 | |
WO2024169462A1 (zh) | 模型监控方法、装置、终端及网络侧设备 | |
WO2024032308A1 (zh) | Ta传输的方法及装置 | |
WO2024067199A1 (zh) | 资源协调方法、装置及存储介质 | |
TWI857573B (zh) | 資訊傳輸方法、裝置、網路設備及終端 | |
WO2023208046A1 (zh) | 资源选择方法、设备、装置及存储介质 | |
WO2023231767A1 (zh) | 定时提前值传输方法、装置及存储介质 | |
WO2024027642A1 (zh) | 一种信息传输方法、装置及设备 | |
WO2024208358A1 (zh) | 定位参考信号发送方法、设备、装置及存储介质 | |
WO2023202693A1 (zh) | 一种信息传输方法、装置、网络设备及终端 | |
WO2024046092A1 (zh) | 目标小区的选择方法、配置方法、信息传输方法及设备 | |
WO2023202694A1 (zh) | 信号传输方法、装置、终端及网络设备 | |
WO2024067067A1 (zh) | 下行波束预测方法、设备、装置及存储介质 | |
WO2023207668A1 (zh) | 一种物理上行共享信道的传输方法、设备及存储介质 | |
WO2024139264A1 (zh) | 波束指示方法、装置及设备 | |
WO2023202323A1 (zh) | 一种信息处理方法、装置及设备 | |
WO2024140078A1 (zh) | 连续条件小区接入方法、装置及存储介质 | |
WO2024164948A1 (zh) | 测量prs的方法、设备、装置及存储介质 | |
WO2023185987A1 (zh) | 传输处理方法、网络设备、终端、装置及存储介质 | |
WO2024046272A1 (zh) | 信息传输方法、装置及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23870335 Country of ref document: EP Kind code of ref document: A1 |