[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

WO2024046202A1 - 发送方法、用户设备、网络侧设备及可读存储介质 - Google Patents

发送方法、用户设备、网络侧设备及可读存储介质 Download PDF

Info

Publication number
WO2024046202A1
WO2024046202A1 PCT/CN2023/114677 CN2023114677W WO2024046202A1 WO 2024046202 A1 WO2024046202 A1 WO 2024046202A1 CN 2023114677 W CN2023114677 W CN 2023114677W WO 2024046202 A1 WO2024046202 A1 WO 2024046202A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
request
model
target
resource
Prior art date
Application number
PCT/CN2023/114677
Other languages
English (en)
French (fr)
Inventor
施源
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2024046202A1 publication Critical patent/WO2024046202A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • 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
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information

Definitions

  • This application belongs to the field of communication technology, and specifically relates to a sending method, user equipment, network side equipment and readable storage media.
  • RSRP Reference Signal Receiving Power
  • UE User Equipment
  • network-side equipment can use Artificial Intelligence (AI) models to predict the Reference Signal Receiving Power (RSRP) of beam pairs.
  • RSRP Reference Signal Receiving Power
  • some beams can be used The RSRP of the pair is taken as input, so that the RSRP results of all beam pairs can be output through the AI model to achieve prediction of the RSRP of the beam pair, where the beam pair includes a transmit beam and a receive beam.
  • the above-mentioned AI model is obtained through training.
  • the AI model may be obtained through UE training, or the AI model obtained by network side device training, and the trained AI
  • the inference location of the model may be on the network side or on the UE side, depending on the usage method and deployment location of the AI model. Therefore, there may be an AI model that needs to be transmitted from one side to the device on the other side, and there may be the number of inputs and outputs of the AI model.
  • the deployment side device and/or the training side device of the AI model since after the AI model training is completed, the number of input data and output data of the AI model, as well as the type of the AI model are determined, if there is no interaction of additional information, it may As a result, the model deployment side cannot obtain a sufficient number of model input parameters for model inference, or cannot obtain a sufficient number of model input and output parameters for model training, etc., resulting in AI model performance degradation or unusability.
  • the AI model deployment side has When many models are used in different scenarios/configurations, if the wrong model is selected on the model deployment side, the performance of the AI model will drop sharply or even become unusable.
  • Embodiments of this application provide a sending method, user equipment, network side equipment and readable storage media, which can improve the performance and applicability of AI models.
  • a first aspect provides a sending method, which method includes: the UE sends a target request related to the AI model; the target request includes at least one of the following: a first request, a second request, and a third request; wherein, the first request It is used to request the network side device to send the first resource; the first resource is the reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain the number information of the transmission beam of the network side device; the third request is used to request Replace the beam information associated with the first resource or request to send multiple first resources; the UE receives the response information corresponding to the target request; the response information is used to perform the target operation, and the target operation includes any of the following: Select the AI model corresponding to the response message , processing AI models.
  • a sending device which includes: a sending module and a receiving module; the sending module is used to send target requests related to the AI model.
  • the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send information about the first resource; the first resource is used by the network side device for beam scanning. Reference signal resources; the second request is used to request to obtain the number information of the transmission beams of the network side device; the third request is used to request to change the beam information associated with the first resource or to request to send multiple first resources; the receiving module is used to receive The response information corresponding to the target request; the response information is used to perform the target operation.
  • the target operation includes any of the following: selecting the AI model corresponding to the response information and processing the AI model.
  • a sending method includes: the network side device receives a target request related to the AI model; the target request includes at least one of the following: a first request, a second request, and a third request; wherein, the One request is used to request the network side device to send the first resource; the first resource is the reference signal resource used by the network side device for beam scanning; the second request is used to request the number information of the transmission beam of the network side device; the third request is used to Request the network side device to change the beam information associated with the first resource or request the network side device to send multiple first resources; the network side device sends response information corresponding to the target request; the response information is used to perform the target operation, and the target operation includes any of the following Items: Select the AI model corresponding to the response information and process the AI model.
  • a sending device which includes: a receiving module and a sending module; a receiving module configured to receive a target request related to the AI model.
  • the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send a first resource; the first resource is a reference signal used by the network side device for beam scanning.
  • the second request is used to request the number information of the transmission beams of the network side device;
  • the third request is used to request the network side device to change the beam information associated with the first resource or to request the network side device to send multiple first resources;
  • the sending module Used to send response information corresponding to the target request received by the receiving module;
  • the response information is used to perform target operations, and the target operations include any of the following: selecting the AI model corresponding to the response information, and processing the AI model.
  • a UE in a fifth aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • the program or instructions When the program or instructions are executed by the processor, the following implementations are implemented: The steps of the method described in one aspect.
  • a UE including a processor and a communication interface, wherein the processor is configured to send a target request related to an AI model; and receive response information corresponding to the target request.
  • a network side device in a seventh aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • the program or instructions are executed by the processor.
  • a network-side device including a processor and a communication interface, wherein the processor is configured to receive a target request related to the AI model; and send response information corresponding to the target request.
  • a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented; or the steps of the method are implemented as described in the first aspect. The steps of the method described in the third aspect.
  • a chip in a tenth aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. The steps; or the steps of implementing the method described in the third aspect.
  • a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the first aspect The steps of the method; or the steps of implementing the method described in the third aspect.
  • the UE sends a target request related to the AI model; the target request includes at least one of the following: a first request, a second request, and a third request; where the first request is used to request the network side device to send a third request.
  • One resource the first resource is a reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain the number information of the transmission beam of the network side device; the third request is used to request to change the beam information associated with the first resource Or request to send multiple first resources; the UE receives response information corresponding to the target request; the response information is used to perform target operations, and the target operations include any of the following: selecting an AI model corresponding to the response message, and processing the AI model.
  • the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • a target request related to the AI model for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • the deployment end of the AI model can obtain a sufficient number of input parameters to process the AI model, thereby improving the performance of the AI model, and because the first device can also send a third request, that is, requesting to replace the first device.
  • Resource-associated beam information or requests are sent to multiple first resources, thus ensuring the normal use of the AI model in different scenarios/configurations. In this way, the performance of the AI model is improved and the applicability of the AI model is ensured.
  • Figure 1 is a schematic architectural diagram of a communication system provided by an embodiment of the present application.
  • Figure 2 is a schematic diagram of the composition of an AI neural network provided by an embodiment of the present application.
  • Figure 3 is a schematic diagram of the structure of a neuron provided by an embodiment of the present application.
  • Figure 4 is a schematic structural diagram of a feedback report provided by an embodiment of the present application.
  • Figure 5 is a schematic structural diagram of a feedback report of a group-based beam report provided by an embodiment of the present application.
  • Figure 6 is a schematic diagram of beam prediction using the AI method provided by the embodiment of the present application.
  • Figure 7 is a schematic diagram of using an AI method to enhance beam prediction performance provided by an embodiment of the present application.
  • Figure 8 is a schematic diagram of using an AI method to improve enhanced beam prediction performance provided by an embodiment of the present application.
  • Figure 9 is one of the flow charts of a sending method provided by an embodiment of the present application.
  • Figure 10 is one of the interaction diagrams of a sending method provided by the embodiment of the present application.
  • Figure 11 is the second interaction diagram of a sending method provided by the embodiment of the present application.
  • Figure 12 is the second flow chart of a sending method provided by the embodiment of the present application.
  • Figure 13 is one of the structural schematic diagrams of a sending device provided by an embodiment of the present application.
  • Figure 14 is the second structural schematic diagram of a sending device provided by an embodiment of the present application.
  • Figure 15 is a schematic diagram of the hardware structure of a communication device provided by an embodiment of the present application.
  • Figure 16 is a schematic diagram of the hardware structure of a UE provided by an embodiment of the present application.
  • Figure 17 is a schematic diagram of the hardware structure of a network-side device provided by an embodiment of the present application.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • NR New Radio
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • Wireless communication system includes terminal 11 and network side equipment 12.
  • the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • PDA Personal Digital Assistant
  • UMPC ultra-mobile personal computer
  • UMPC mobile Internet device
  • Mobile Internet Device MID
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • VUE vehicle-mounted equipment
  • PUE pedestrian terminal
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • PC personal computers
  • teller machines or self-service Terminal devices such as mobile phones
  • wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), Smart wristbands, smart clothing, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device 12 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or Wireless access network unit.
  • the access network device 12 may include a base station, a WLAN access point or a WiFi node, etc.
  • the base station may be called a Node B, an evolved Node B (eNB), an access point, a Base Transceiver Station (BTS), a radio Base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home B-Node, Home Evolved B-Node, Transmitting Receiving Point (TRP) or belonging to Some other appropriate terminology in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the embodiment of this application, only the base station in the NR system is used as an example for introduction. Define the specific type of base station.
  • AI has been widely used in many fields, and AI networks have many implementation methods, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc.
  • FIG. 2 shows a schematic diagram of the composition of an AI neural network. As shown in Figure 2, the AI neural network is composed of neurons.
  • Figure 3 shows a schematic diagram of the composition of a neuron.
  • a 1 , a 2 ,...a K is the input
  • w is the weight (or multiplicative coefficient)
  • b is the bias (or is called the additive coefficient)
  • ⁇ (.) is the activation function.
  • Common activation functions include Sigmoid, tanh, Rectified Linear Unit (ReLU), etc.
  • the parameters of neural networks can be optimized through optimization algorithms.
  • An optimization algorithm is a type of algorithm that can assist developers or users in minimizing or maximizing an objective function (also known as: loss function).
  • the objective function is often a mathematical combination of model parameters and data. For example: given data X and its corresponding label Y, developers can build a neural network model f(.), and through this neural network model f(.), the predicted output can be obtained based on the input x f(x), and the difference between the predicted value and the true value (f(x)-Y) can be calculated, which is the loss function.
  • the developer's purpose is to find appropriate W and b so that the value of the above loss function can be minimized. The smaller the loss value, the closer the model is to the real situation.
  • the optimization algorithm in the embodiment of this application may be based on the error back propagation (Error Back Propagation, BP) algorithm.
  • BP error back propagation
  • the basic idea of BP algorithm is that the learning process consists of two processes: forward propagation of signals and back propagation of errors.
  • the input sample is passed in from the input layer, processed layer by layer by each hidden layer, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, it will enter the error backpropagation stage.
  • Error backpropagation is to backpropagate the output error in some form to the input layer layer by layer through the hidden layer, and allocate the error to all units in each layer, thereby obtaining the error signal of each layer unit. This error signal is used as a correction for each unit.
  • This process of adjusting the weights of each layer in forward signal propagation and error back propagation is carried out over and over again.
  • the process of continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until a preset number of learning times.
  • Optimization algorithms can also include gradient descent (Gradient Descent, GD), stochastic gradient descent (Stochastic Gradient Descent, SGD), mini-batch gradient descent (Mini-Batch Gradient Descent), momentum method (Momentum), and stochastic gradient descent with momentum ( Nesterov), adaptive gradient descent (ADAptive GRADient descent, Adagrad), Adadelta, root mean square error reduction (Root Mean Square Prop, RMSprop), adaptive momentum estimation (Adaptive Moment Estimation, Adam), etc.
  • the network side equipment can perform beam indication on the downlink and uplink channels or reference signals, which is used to establish beam links between the network side equipment and the UE to achieve channel or reference signal transmission.
  • RRC Radio Resource Control
  • K Transmission Configuration Indication
  • MAC CE Media Access Control Layer
  • the reference signal (Reference Signal) in this TCI state (such as periodic CSI-RS resource, semi-persistent CSI-RS resource, SS block, etc.) and the UE-specific PDCCH demodulation reference signal (Demodulation Reference Signal, DMRS) port are spatial QCL of.
  • the UE can learn which receiving beam is used to receive the PDCCH according to the TCI status.
  • the network side device configures M TCI states through RRC signaling, then uses the MAC CE command to activate 2 N TCI states, and then notifies the TCI state through the N-bit TCI field of DCI.
  • the TCI state in the TCI state The Reference Signal and the DMRS port of the PDSCH to be scheduled are QCL.
  • the UE can learn which receiving beam is used to receive the PDSCH according to the TCI status.
  • the network side device configures QCL information for the CSI-RS resource through RRC signaling.
  • the network side device uses the MAC CE command to indicate its QCL information when activating a CSI-RS resource from the CSI-RS resource set configured in RRC.
  • the network side device configures QCL for the CSI-RS resource through RRC signaling and uses DCI to trigger CSI-RS.
  • the network side device uses RRC signaling to configure the spatial relationship information (Spatial Relation Information) for each PUCCH resource through the PUCCH-SpatialRelationInfo parameter.
  • the spatial relationship information Spatial Relation Information
  • the configured Spatial Relation Information contains multiple spatial relation information
  • use MAC-CE to indicate or activate one of the spatial relation information.
  • the spatial relation information configured for the PUCCH resource only contains 1, no additional MAC CE command is required.
  • the spatial relation information of PUSCH is the uplink scheduling request indication information in DCI when the downlink control channel information (DCI) carried by PDCCH schedules the physical uplink shared channel (Physical Uplink Shared Channel, PUSCH).
  • DCI downlink control channel information
  • PUSCH Physical Uplink Shared Channel
  • Each SRI code point (codepoint) of the Schduling Request Indication field (SRI field) indicates an SRI, which is used to indicate the Spatial Relation Information of PUSCH.
  • the network configures Spatial Relation Information for the SRS resource through RRC signaling.
  • the SRS type is semi-persistent SRS
  • the network uses the MAC CE command to activate one from a set of Spatial Relation Information configured by RRC.
  • the SRS type is aperiodic SRS
  • the network configures Spatial Relation Information for the SRS resource through RRC signaling.
  • the unified transmission configuration indication status (unified TCI indication) is proposed The concept of indicating subsequent reference signals and beam information of multiple channels through the TCI domain in a DCI.
  • Beam information can usually be represented by TCI state information and QCL information.
  • Uplink beam information can usually be represented using Spatial Relation information.
  • the shaping weight of the analog beam is calculated by This is achieved by adjusting the parameters of equipment such as RF front-end phase shifters.
  • the polling method can be used for training of analog beamforming vectors, that is, the array elements of each polarization direction of each antenna panel sequentially send training signals (i.e., candidate shaping vectors) at an agreed time in a time-division multiplexing manner.
  • the terminal After measurement, the beam report is fed back, so that the network side can use the training signal to implement simulated beam transmission when transmitting services next time.
  • the content of the beam report usually includes several optimal transmit beam identifiers and the measured received power of each transmit beam.
  • the network side device When doing beam measurements, the network side device will configure a reference signal resource set (RS resource set), which includes at least one reference signal resource, such as SSB resource or CSI-RS resource.
  • RS resource set which includes at least one reference signal resource, such as SSB resource or CSI-RS resource.
  • the UE measures the L1-RSRP/L1-SINR of each RS resource and reports at least one of the best measurement results to the network side device.
  • the reported content includes SSBRI or CRI, and L1-RSRP/L1-SINR.
  • the report content reflects at least one optimal beam and its quality for the network side device to determine the beam used to send channels or signals to the UE.
  • the 7-bit quantization method is used, the quantization step is 1dB, and the quantization range is -140dBm to -44dBm.
  • the strongest RSRP quantization uses 7-bit quantization, and the remaining RSRP quantization uses the 4-bit differential quantization method. The steps are 2dB.
  • Figure 4 shows a schematic structural diagram of a feedback report.
  • Figure 5 shows a schematic diagram of the feedback report structure of group-based beam reporting.
  • the number of feedback reports is determined by the parameters configured by the network side device to the UE, and through the RRC configuration parameters, and the number of RS and RSRP that should be included in the feedback report of the UE.
  • the value of the quantity configuration is 1, 2 ,3,4, the default value is 1.
  • the number limit is based on the UE capability, and the UE will first report the maximum number it can support.
  • Figure 6 shows a schematic diagram of beam prediction using the AI method.
  • the RSRP of some beam pairs can be used as input, and the output of the AI model is the RSRP result of all beam pairs.
  • the beam pairs are composed of transmitting beams and receiving beams, and the input number of the AI model is the number of selected partial beam pairs, and the output number is the number of all beam pairs.
  • Figure 7 shows a schematic diagram of using AI methods to enhance beam prediction performance.
  • associated information can be added to the input side.
  • the associated information is generally angle-related information corresponding to the beam pairs selected for input, beam ID information, etc. Therefore, the number of inputs to this model is also related to the number of selected partial beam pairs, and the output number is still equal to the number of all beam pairs.
  • Figure 8 shows a schematic diagram of using AI methods to improve enhanced beam prediction performance. As shown in Figure 8, this method mainly affects the output of the AI model by changing the expected information.
  • the input type of the AI model includes at least one of the following:
  • the A-side sends beam-related association information
  • End B expects end A to send beam-related associated information
  • Desired forecast time related information is:
  • the associated information related to the beam refers to the beam information corresponding to the beam.
  • the beam information includes but is not limited to at least one of the following:
  • the beam ID information is used to characterize the identity of the beam, including but not limited to at least one of the following:
  • the beam angle information is used to characterize the angle-related information corresponding to the beam, including but not limited to at least one of the following:
  • angle information is relevant information used to characterize angles, such as angle, radian, index code value, code value processed by additional AI network, etc.
  • the training location and inference location of the AI model are not yet certain. Therefore, the training location and the inference location may be at the same location, such as at the UE, base station or central node, or the model training location and inference location are In two locations, for example, the training location is at the base station and the inference location is at the UE.
  • the performance of the AI model may be degraded; and since there are many feasible solutions for the implementation of the AI model, it may lead to AI model mismatch.
  • the implementation of the AI model solution requires some auxiliary information interaction to ensure the normal use of the AI model. Therefore, a method is urgently needed to ensure the normal use of the AI model and improve the performance of the AI model.
  • FIG. 9 shows a flow chart of a sending method provided by an embodiment of the present application.
  • the sending method provided by the embodiment of the present application may include the following steps 201 and 202.
  • Step 201 The user equipment UE sends a target request related to the AI model.
  • the target request includes at least one of the following: a first request, a second request, and a third request;
  • the first request is used to request the network side device to send the first resource;
  • the first resource is a reference signal resource used by the network side device for beam scanning;
  • the second request is used to request information on the number of transmission beams of the network side device
  • the third request is used to request the network side device to change the beam information associated with the first resource or to request to send multiple first resources.
  • Step 202 The UE receives response information corresponding to the target request.
  • the response information is used to perform a target operation.
  • the target operation includes any of the following: selecting an AI model corresponding to the response information and processing the AI model.
  • the embodiment of the present application provides a sending method.
  • the UE sends a target request related to the AI model; the target request includes at least one of the following: a first request, a second request, and a third request; wherein the first request is used to request the network side
  • the device sends a first resource; the first resource is a reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain information on the number of transmission beams of the network side device; the third request is used to request to change the first resource association
  • the beam information or request sends multiple first resources; the UE receives the response information corresponding to the target request; the response information is used to perform the target operation, and the target operation includes any of the following: selecting the AI model corresponding to the response message and processing the AI model.
  • the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • a target request related to the AI model for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • step 202 can be specifically implemented through the following step 202a.
  • Step 202a The UE receives the second resource.
  • the second resource is a reference signal resource used for beam scanning determined by the network side device according to the target request.
  • the sending method provided by the embodiment of this application also includes the following steps 202b and 202c.
  • Step 202b The UE obtains beam quality information of the beam corresponding to the second resource according to the second resource.
  • Step 202c The UE processes the AI model according to the beam quality information.
  • the UE after sending the first request, can receive the second resource determined by the network side device according to the first request.
  • the second resource is the reference signal resource for beam scanning; thus, the UE can receive the second resource according to the first request.
  • the second resource processes the AI model, such as training the AI model, inferring the AI model, etc. Therefore, not only can the interaction of information be completed before the AI model is processed, but also, the second resource obtained can be used to ensure the processing
  • the number of input data and output data of the AI model allows the UE to use the AI normally and perform subsequent processing.
  • the "processing AI model" in the above-mentioned step 202c can be specifically implemented through any one of the following steps 202c1 to 202c5.
  • Step 202c1 UE trains the AI model.
  • Step 202c2 The UE adjusts the parameters of the AI model.
  • Step 202c3 The UE monitors the performance of the AI model.
  • Step 202c4 UE infers the AI model.
  • Step 202c5 The UE collects the required data of the AI model.
  • the UE can perform various processes on the AI model, such as training, adjusting parameters, reasoning, or collecting the data it requires, thereby ensuring the applicability of the AI model or making model selection errors.
  • the first request is associated with at least one of the following:
  • Purpose information is used to indicate the purpose for which the UE requests the first resource, and the first quantity information is used to indicate at least one of the following:
  • the number of symbols occupied by the first resource is the number of symbols occupied by the first resource.
  • the UE while sending the first request, may carry the quantity information of the first resource requested in the first request, so that the receiving end can, after receiving the first request of the UE, carry out the first request according to the information carried in the first request.
  • the quantity information determines the number of resources required by the UE, or the number of beams corresponding to the resources, thereby ensuring that the UE can receive the required number of resources, or the number of beams corresponding to the resources, etc., and ensuring the accuracy of the AI model. Normal use.
  • the usage information is used to indicate at least one of the following:
  • the UE while sending the first request, may carry the purpose information for requesting the first resource in the first request, so that the receiving end can, after receiving the first request of the UE, carry out the first request according to the information carried in the first request.
  • the usage information determines the usage required by the UE, thereby ensuring that the UE can receive the required number of resources, or the number of beams corresponding to the resources, and ensuring the normal use of the AI model.
  • the usage information indicates: training the target artificial intelligence AI model
  • the usage information indicates: inferring the AI model, it means that the UE requires a smaller number of beam resources, and the smaller number of beam resources need to be used for the input and output of the AI model, so that the network side device can according to the prior Beam resources are configured with a determined, and/or negotiated, and/or agreed upon smaller number of characteristics.
  • the first quantity information is determined by the network side device according to the requested use of the first resource.
  • the quantity indicated by the first quantity information is determined by the first capability information reported by the UE, and the first capability information is the capability information of the AI model or the capability information of the UE.
  • the capability information of the AI model includes at least one of the following:
  • Quantitative information indicating the input of the AI model
  • Quantitative information indicating the output of the AI model
  • Quantity information indicating the first target information contained in the input of the AI model
  • Quantity information indicating the second target information contained in the output of the AI model.
  • the first target information and/or the second target information may include at least one of the following:
  • Beam information for the receive beam is
  • the beam information is information corresponding to the beam
  • the information corresponding to the beam includes at least one of the following:
  • Beam identification ID information of the beam
  • Beamwidth information for the beam
  • the first request is associated with or contains the first resource sending quantity and/or the minimum sending quantity, that is, the first request is associated with or contains the quantity indicated by the first quantity information.
  • the input quantity information and/or the output quantity information of the AI model are related to the quantity indicated by the first quantity information.
  • the UE can inform the receiving end, such as the network side device, of its AI model capability information by sending the first request, so that the network side device can send its AI model capability information to the UE based on the UE's AI model capability information. needs, or the value or range of information that the AI model can bear, so as to ensure the normal use of the AI model.
  • the quantity indicated by the first quantity information represents the quantity of the first resource, or represents the number of beams corresponding to the first resource.
  • the number indicated by the first quantity information may be the total number of beams sent. In this case, the number of beams sent is greater than that of the first resource. quantity.
  • the number of inputs to the AI model indicated by the AI model's capability information is less than or equal to the number indicated by the first quantity information
  • the quantity of output of the AI model indicated by the capability information of the AI model is less than or equal to the quantity indicated by the first quantity information.
  • the first target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the second target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the first target quantity is determined by the input quantity information of the AI model and the first target information; or, it is determined by the quantity information of the first target information;
  • the second target quantity is determined by the quantity information output by the AI model and the second target information; or, it is determined by the quantity information of the second target information.
  • the capability information of the UE includes second quantity information, and the second quantity information is used to indicate the number of beams required when processing the AI model.
  • the repeated configuration status of the first resource is closed.
  • the first resource is a resource configured with repetition off on the network side.
  • the quantity indicated by the first quantity information is the number of repetitions of the UE requesting to send the first resource. times; or the number of times the UE requests to send the beam corresponding to the first resource.
  • the repeated configuration state of the first resource is enabled.
  • the target request includes: a second request; the above-mentioned step 202 may specifically include the following step 202d.
  • Step 202d The UE receives the third quantity information.
  • the third quantity information is used to indicate the number of quantity information of transmitting beams.
  • the sending method provided by the embodiment of this application also includes the following step 202e.
  • Step 202e The UE selects the AI model corresponding to the third quantity information according to the third quantity information.
  • the network side device can send to the UE the information required by the UE or the applicable third party according to the quantity information. Quantity information, so that the UE can select an appropriate AI model based on the third quantity information, thereby avoiding AI model selection errors.
  • the target request includes a third request.
  • the sending method provided by the embodiment of the present application further includes the following step 301, and the above step 201 can be specifically performed by the following: Implemented in step 201a; or implemented through the following steps 201b and 201c.
  • Step 301 The UE measures the target resource and obtains the target measurement result.
  • the target resource is a resource used for beam measurement.
  • Step 201a If the first measurement result is less than and/or equal to the first threshold, the UE sends a third request.
  • the first measurement result is a measurement result that satisfies the first condition among the target measurement results.
  • the first condition is that there is a first preset quantity or a first preset ratio, and it is less than or equal to the first threshold value. .
  • Step 201b The UE determines the target feedback information according to the target measurement result.
  • Step 201c The UE sends target feedback information.
  • the target feedback information includes the target measurement result, and the target feedback information is used to implicitly indicate the third request.
  • the second measurement result is used to determine the third request
  • the second measurement result is a measurement result that satisfies the second condition among the target measurement results.
  • the second condition is that there is a second preset number or a second preset ratio and is less than or equal to the second threshold value.
  • the first measurement result or the second measurement result includes at least one of the following:
  • the first threshold value and/or the second threshold value are determined by at least one of the following:
  • the threshold value reported by the UE is the threshold value reported by the UE.
  • the minimum threshold value for quantification of measurement results agreed upon in the above protocol can be -140dBm;
  • the above threshold value related to the measurement results obtained through the protocol agreed upon can be a gate determined by calculating the mean value of the measured L1-RSPR. limit, or a value determined based on the noise coefficient, subcarrier spacing size, etc.;
  • the above threshold value reported by the UE can be a threshold value calculated and reported by the UE based on the noise coefficient, etc.
  • the target request information includes: a third request; the above step 202 can be specifically implemented through the following step 202f.
  • Step 202f The UE receives the third information.
  • the third information includes information of a third resource
  • the third resource is beam information associated with the first resource after replacement by the network side device or multiple first resources sent by the network side device.
  • the UE since the UE can send a third request to the network side device at any time, for example, when it is found that the number of the first resource or the beam information associated with the first resource is not applicable, the UE requests the deteriorated network side device to replace the first resource.
  • the embodiment of the present application provides a sending method, which may include the following steps 11 to 14.
  • Step 11 The UE sends the first request.
  • the first request includes first beam scanning request information
  • the first beam scanning request information is information used to request the network side device to send the first resource
  • the first resource is a reference for the network side device to use for beam scanning. Signal resources.
  • the first request is associated with at least one of the following:
  • Purpose information first quantity information
  • the purpose information is used to indicate the purpose for which the UE requests the first resource
  • the first quantity information is used to indicate at least one of the following:
  • the number of symbols occupied by the first resource is the number of symbols occupied by the first resource.
  • the usage information is used to indicate at least one of the following:
  • the network side device may determine the sending quantity and/or the minimum sending quantity of the first resource according to the usage information associated with the first request.
  • the quantity indicated by the first quantity information is determined by the first capability information reported by the UE, and the first capability information is the capability information of the AI model or the capability information of the UE.
  • the sending quantity and/or the minimum sending quantity of the first resource obtained based on the usage information can be determined by AI model capability reporting, that is, by the UE reporting the first capability information, and the first capability information is Capability information of the AI model or capability information of the UE.
  • the capability information of the AI model includes at least one of the following:
  • Quantitative information indicating the input of the AI model
  • Quantitative information indicating the output of the AI model
  • Quantity information indicating the first target information contained in the input of the AI model
  • Quantity information indicating the second target information contained in the output of the AI model.
  • the first target information and/or the second target information respectively include at least one of the following:
  • Beam information for the receive beam is
  • the number of inputs to the AI model indicated by the AI model's capability information is less than or equal to the number indicated by the first quantity information
  • the quantity of output of the AI model indicated by the capability information of the AI model is less than or equal to the quantity indicated by the first quantity information.
  • the first target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the second target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the first target quantity is determined by the input quantity information of the AI model and the first target information; or, it is determined by the quantity information of the first target information;
  • the second target quantity is determined by the quantity information output by the AI model and the second target information; or, it is determined by the quantity information of the second target information.
  • the capability information of the UE includes second quantity information, and the second quantity information is used to indicate the number of beams required when processing the AI model.
  • Step 12 The UE receives the second resource.
  • the second resource is a reference signal resource used for beam scanning determined by the network side device according to the target request.
  • Step 13 The UE obtains beam quality information of the beam corresponding to the second resource according to the second resource.
  • Step 14 The UE performs the target operation corresponding to the beam quality information according to the beam quality information.
  • Target actions include any of the following:
  • the repeated configuration status of the first resource is closed.
  • Figure 10 shows an interaction diagram of a sending method provided by the embodiment of the present application.
  • the AI model performs inference and training on the UE side.
  • the sending method may include the following steps a to steps d.
  • Step a The UE sends a first request to the network side device (such as a base station).
  • the network side device such as a base station.
  • Step b The network side device sends the first resource configured as repetition off according to the first request.
  • Step c The UE measures and obtains RSRP.
  • Step d The UE uses the RSRP as the input content of the AI model, thereby causing the AI model to output content.
  • the number of inputs of the AI model corresponds to the number of beams: 8; the number of outputs corresponding to the number of beams is: 32;
  • the UE when the UE performs AI model training on the AI model, the UE sends a first request, requesting the base station to send 32 beams;
  • the UE When the UE performs AI model inference on the AI model, the UE sends a first request, requesting the base station to send 8 beams;
  • the UE When the UE performs AI model inference on the AI model, the UE sends a first request, requesting the base station to send 8 beams; and when the base station sends 16 beams, it obtains 16 beam information based on UE measurements, and selects 8 of them as Input to the AI model.
  • This embodiment of the present application provides a sending method, which may include the following steps 15 to 17.
  • Step 15 The UE sends the second request.
  • the second request is used to request to obtain information on the number of transmission beams of the network side device.
  • Step 16 The UE receives the third quantity information.
  • the third quantity information is used to indicate the number of quantity information of transmitting beams.
  • Step 17 The UE selects an AI model based on the third quantity information.
  • Figure 11 shows an interaction diagram of a sending method provided by the embodiment of the present application.
  • the AI model performs inference and training on the UE side.
  • the sending method may include the following steps e to steps d.
  • Step e The UE sends a second request to the network side device (eg, base station).
  • the network side device eg, base station
  • Step f The base station sends the base station's transmission beam number information according to the second request.
  • Step g The UE selects an AI model based on the transmission beam number information.
  • the UE can send a second request, and the base station can send the base station's transmission according to the second request.
  • the number of beams is 32, then the UE selects AI model 2 based on the number of transmit beams.
  • This embodiment of the present application provides a sending method, which may include the following steps 18 to 22.
  • Step 18 The UE sends the third request.
  • the third request is used to request to change the beam information associated with the first resource or to request to send multiple first resources.
  • the sending method provided by the embodiment of the present application also includes the following step 19, and the embodiment of the present application provides two ways of sending the third request, namely an explicit way and an implicit way, where , the display mode can be realized through step 19 and step 20, and the implicit mode can be realized through step 19, step 21 and step 22.
  • Step 19 The UE measures the target resource and obtains the target measurement result.
  • the target resource is a resource used for beam measurement.
  • Step 20 If the first measurement result is less than and/or equal to the first threshold value, the UE directly sends the third request.
  • the first measurement result is a measurement result that satisfies the first condition among the target measurement results.
  • the first condition is that there is a first preset quantity or a first preset ratio, and it is less than or equal to the first threshold value. .
  • Step 21 The UE determines the target feedback information based on the target measurement results
  • Step 22 The UE sends target feedback information.
  • the target feedback information includes the target measurement result, and the target feedback information is used to implicitly indicate the third request.
  • the second measurement result is used to determine the third request;
  • the second measurement result is a measurement result that satisfies the second condition among the target measurement results, and the second condition is that there is a second preset number or a third
  • the two preset ratios are less than or equal to the second threshold value.
  • the first threshold value and/or the second threshold value are determined by at least one of the following:
  • the threshold value reported by the UE is the threshold value reported by the UE.
  • This embodiment of the present application provides a sending method, which may include the following step 23.
  • Step 23 The UE sends fifth request information.
  • the fifth request information includes second beam scanning request information, and the second beam scanning request information is information used to request the network side device to send the first resource; the first resource is used by the network side device for beam scanning. Reference signal resources.
  • the repeated configuration state of the first resource is on, that is, repetition on.
  • the fifth request information is associated with at least one of the following:
  • Purpose information first quantity information
  • the purpose information is used to indicate the purpose for which the UE requests the first resource
  • the first quantity information is used to indicate at least one of the following:
  • the number of symbols occupied by the first resource is the number of symbols occupied by the first resource.
  • the usage information is used to indicate at least one of the following:
  • the network side device may determine the number of repetitions and/or the minimum number of repetitions of the first resource based on the usage information associated with the first request.
  • the fifth request information is associated with or includes the number of repetitions or the minimum number of repetitions of the first resource.
  • the number of inputs and/or the number of outputs of the AI model is related to the number of repetitions or the minimum number of repetitions of the first resource.
  • the number of inputs and/or the number of outputs of the AI model is less than or equal to the number of repetitions or the minimum number of repetitions of the first resource.
  • the number of repetitions or the minimum number of repetitions of the first resource represents the number of the first resources, or the number of beams corresponding to the first resources.
  • the quantity indicated by the first quantity information is the number of repetitions of the UE's request to send the first resource; or the number of repetitions of the UE's request to send the beam corresponding to the first resource.
  • Embodiment 5 For the specific implementation method of Embodiment 5, reference can be made to Embodiment 1, which will not be described again here.
  • FIG. 12 shows a flow chart of a sending method provided by an embodiment of the present application.
  • the sending method provided by the embodiment of the present application may include the following steps 401 and 402.
  • Step 401 The network side device receives a target request related to the AI model.
  • the target request includes at least one of the following: a first request, a second request, and a third request;
  • the first request is used to request the network side device to send the first resource;
  • the first resource is a reference signal resource used by the network side device for beam scanning;
  • the second request is used to request information on the number of transmission beams of the network side device
  • the third request is used to request the network side device to change the beam information associated with the first resource or to request the network side device to send multiple first resources;
  • Step 402 The network side device sends response information corresponding to the target request.
  • the response information is used to perform target operations
  • the target operations include any of the following: selecting an AI model corresponding to the response information and processing the AI model.
  • the above target information includes: a first request.
  • the above step 402 may be performed through the following step 402a. accomplish.
  • Step 402a The network side device sends the second resource.
  • the second resource is a reference signal resource used for beam scanning; the second resource is used for processing the AI model.
  • the target request includes: a second request; the above step 402 can be specifically implemented through the following step 402b.
  • Step 402b The network side device sends the third quantity information.
  • the third quantity information is used to indicate the number of quantity information of transmitting beams; the third quantity information is used to select an AI model.
  • the target request includes: a third request; the above-mentioned step 401 can be implemented specifically through the following step 401c; or through the following steps 401d and 401e.
  • Step 401c The network side device receives the third request.
  • Step 401d The network side device receives the target feedback information.
  • Step 401e The network side device obtains the third request according to the target feedback information.
  • the target feedback information is the result of the UE measuring the target resource; the target resource is a resource used for beam measurement.
  • the target request includes: a third request; the above step 402 can be specifically implemented through the following step 402c.
  • Step 402c The network side device sends third information.
  • the third information includes information about the third resource, and the third resource is the beam information associated with the first resource after the network side device has been replaced or multiple first resources sent by the network side device; the third information is used to Processing AI models.
  • Embodiments of the present application provide a sending method.
  • a network side device receives a target request related to an AI model; the target request includes at least one of the following: a first request, a second request, and a third request; wherein the first request is used to request
  • the network side device sends a first resource; the first resource is a reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain information on the number of transmission beams of the network side device; the third request is used to request to replace the first Resource-associated beam information or requests to send multiple first resources; the network side device sends response information corresponding to the target request; the response information is used to perform target operations, and the target operations include any of the following: selecting an AI model corresponding to the response message; Processing AI models.
  • the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • a target request related to the AI model for example, requesting the network side device to send reference signal resources for beam scanning, requesting the number of beams to be sent, so that the UE can send the number of beams based on the reference signal resources and/or the information.
  • the execution subject may be a sending device.
  • the sending device executing the sending method is taken as an example to describe the sending device provided by the embodiment of the present application.
  • Figure 13 shows a possible structural diagram of the sending device involved in the embodiment of the present application.
  • the sending device 40 may include: a sending module 41 and a receiving module 42 .
  • the sending module 41 is used to send a target request related to the AI model; the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send a third request.
  • Information about one resource is the first resource is the reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain the network
  • the third request is used to request to change the beam information associated with the first resource or to request to send multiple first resources.
  • the receiving module 42 is used to receive response information corresponding to the target request; the response information is used to perform a target operation, and the target operation includes any of the following: selecting an AI model corresponding to the response information and processing the AI model.
  • the embodiment of the present application provides a sending device. Since the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning, requesting to send the number information of the beams, so that the UE can send the target request based on the reference signal.
  • a target request related to the AI model for example, requesting the network side device to send reference signal resources for beam scanning, requesting to send the number information of the beams, so that the UE can send the target request based on the reference signal.
  • Resource and/or information on the number of transmission beams select the AI model, or process the AI model, such as training the AI model or inferring the AI model, etc., because before the processing of the AI model is completed, information can be exchanged, such as the interaction of the first request , the second request or the third request, therefore, the amount of input data and output data of the AI model used for processing or selection can be guaranteed, or the type of AI model used for processing or selection, so that the deployment end of the AI model can obtain
  • a sufficient number of input parameters are used to process the AI model, thereby improving the performance of the AI model, and because the first device can also send a third request, that is, requesting to change the beam information associated with the first resource or requesting to send multiple Therefore, it can ensure the normal use of the AI model under different scenarios/configurations. In this way, the performance of the AI model is improved and the applicability of the AI model is ensured.
  • the above target information includes a first request; the above receiving module 42 is specifically configured to receive a second resource from a network side device; the second resource is a reference signal resource used for beam scanning.
  • the sending method provided by the embodiment of the present application also includes: an acquisition module and a processing module; an acquisition module, configured to acquire the beam quality of the beam corresponding to the second resource according to the second resource received by the receiving module information.
  • the processing module is used to process the AI model based on the beam quality information obtained by the acquisition module.
  • the above processing module is specifically used to
  • the above first request is associated with at least one of the following:
  • Purpose information is used to indicate the purpose for which the UE requests the first resource, and the first quantity information is used to indicate at least one of the following:
  • the number of symbols occupied by the first resource is the number of symbols occupied by the first resource.
  • the above usage information is used to indicate at least one of the following:
  • the above first quantity information is determined by the network side device according to the requested use of the first resource.
  • the quantity indicated by the first quantity information is determined by the first capability information reported by the UE, and the first capability information is the capability information of the AI model or the capability information of the UE.
  • the capability information of the AI model includes at least one of the following:
  • Quantitative information indicating the input of the AI model
  • Quantitative information indicating the output of the AI model
  • Quantity information indicating the first target information contained in the input of the AI model
  • Quantity information indicating the second target information contained in the output of the AI model.
  • the above-mentioned first target information and/or second target information includes at least one of the following:
  • Beam information for the receive beam is
  • the information corresponding to the above-mentioned beam includes at least one of the following:
  • Beam identification ID information of the beam
  • Beamwidth information for the beam
  • the number of inputs of the AI model indicated by the capability information of the AI model is less than or equal to the number indicated by the first quantity information
  • the quantity of output of the AI model indicated by the capability information of the AI model is less than or equal to the quantity indicated by the first quantity information.
  • the first target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the second target quantity indicated by the AI model's capability information is less than or equal to the quantity indicated by the first quantity information
  • the first target quantity is determined by the input quantity information of the AI model and the first target information; or, it is determined by the quantity information of the first target information;
  • the second target quantity is determined by the quantity information output by the AI model and the second target information; or, it is determined by the quantity information of the second target information.
  • the capability information of the UE includes second quantity information, and the second quantity information is used to indicate the number of beams required when processing the AI model.
  • the repeated configuration state of the first resource is closed.
  • the quantity indicated by the first quantity information is the number of repetitions of the UE's request to send the first resource; or the number of repetitions of the UE's request to send the beam corresponding to the first resource.
  • the repeated configuration state of the above-mentioned first resource association is enabled.
  • the above target request includes: a second request; and the above receiving module 42 is specifically configured to receive third quantity information, where the third quantity information is used to indicate the number of quantity information of transmitting beams.
  • the sending device provided by the embodiment of the present application further includes: a selection module.
  • the selection module is configured to select the AI model corresponding to the third quantity information according to the third quantity information received by the receiving module 42 .
  • the above target request includes a third request
  • the sending device provided by the embodiment of the present application further includes: a measurement module; a measurement module configured to, before the sending module 41 sends the target request related to the AI model, The target resource is measured to obtain the target measurement result; the target resource is a resource used for beam measurement.
  • the above-mentioned sending module 41 is specifically configured to send a third request if the first measurement result is less than and/or equal to the first threshold value, and the first measurement result is the target measurement result that satisfies the first threshold value.
  • the measurement result of a condition, the first condition is that there is a first preset quantity or a first preset proportion, and it is less than or equal to the first threshold value.
  • the above-mentioned sending module 41 is specifically used to determine the target based on the target measurement results. target feedback information; and send target feedback information; wherein the target feedback information includes target measurement results, and the target feedback information is used to implicitly indicate the third request.
  • the second measurement result is used to determine the third request;
  • the second measurement result is a measurement result that satisfies a second condition among the target measurement results, and the second condition is that there is a second preset number or a third
  • the two preset ratios are less than or equal to the second threshold value.
  • the above-mentioned first threshold value and/or the second threshold value is determined by at least one of the following:
  • the threshold value reported by the UE is the threshold value reported by the UE.
  • the above target request includes: a third request; the sending device provided by the embodiment of the present application further includes: a processing module.
  • the above-mentioned receiving module is also used to receive third information; the third information includes information of a third resource, and the third resource is the beam information associated with the first resource after the network side device has replaced it or a plurality of first resources sent by the network side device. resource.
  • the above-mentioned processing module is also used to process the AI model based on the third information.
  • the sending device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • Figure 14 shows a possible structural diagram of the sending device involved in the embodiment of the present application.
  • the sending device 50 may include: a receiving module 51 and a sending module 52 .
  • the receiving module 51 is used to receive target requests related to the AI model.
  • the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send a first resource; the first resource is a reference signal used by the network side device for beam scanning.
  • the second request is used to request the number information of the transmission beams of the network side device;
  • the third request is used to request the network side device to change the beam information associated with the first resource or to request the network side device to send multiple first resources;
  • sending module 52 used to send response information corresponding to the target request;
  • the response information is used to perform target operations, and the target operations include any of the following: selecting the AI model corresponding to the response information, and processing the AI model.
  • the embodiment of the present application provides a sending device. Since the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning, requesting to send the number information of the beams, so that the UE can send the target request based on the reference signal.
  • a target request related to the AI model for example, requesting the network side device to send reference signal resources for beam scanning, requesting to send the number information of the beams, so that the UE can send the target request based on the reference signal.
  • Resource and/or information on the number of transmission beams select the AI model, or process the AI model, such as training the AI model or inferring the AI model, etc., because before the processing of the AI model is completed, information can be exchanged, such as the interaction of the first request , the second request or the third request, therefore, the amount of input data and output data of the AI model used for processing or selection can be guaranteed, or the type of AI model used for processing or selection, so that the deployment end of the AI model can obtain
  • a sufficient number of input parameters are used to process the AI model, thereby improving the performance of the AI model, and because the first device can also send a third request, that is, requesting to change the beam information associated with the first resource or requesting to send multiple Therefore, it can ensure the normal use of the AI model under different scenarios/configurations. In this way, the performance of the AI model is improved and the applicability of the AI model is ensured.
  • the target information includes: a first request, sending module 52, specifically used to send a second resource; the second resource is a reference signal resource used for beam scanning; the second resource is used to process the AI model .
  • the target request includes: a second request; a sending module 52, specifically configured to send third quantity information, where the third quantity information is used to indicate the number of quantity information to send beams; the third quantity information is To choose an AI model.
  • the target request includes: a third request; a receiving module 51, specifically configured for the network side device to receive the third request; or the receiving module 51, specifically configured to receive target feedback information; and based on the target feedback information to obtain the third request; the target feedback information is the result obtained by the UE measuring the target resource; the target resource is a resource used for beam measurement.
  • the sending module 52 is specifically used by the network side device to send third information; the third information includes information about the third resource, and the third resource is the beam associated with the first resource after the network side device replaces it. Information or multiple first resources sent by network side devices; the third information is used to process the AI model.
  • the sending device provided by the embodiments of the present application can implement each process implemented by the above method embodiments and achieve the same technical effect. To avoid duplication, details will not be described here.
  • this embodiment of the present application also provides a communication device 700, which includes a processor 701 and a memory 702.
  • the memory 702 stores programs or instructions that can be run on the processor 701, for example.
  • the communication device 700 is a terminal, when the program or instruction is executed by the processor 701, each step of the above sending method embodiment is implemented, and the same technical effect can be achieved.
  • the communication device 700 is a network-side device, when the program or instruction is executed by the processor 701, each step of the above-mentioned sending method embodiment is implemented, and the same technical effect can be achieved. To avoid duplication, the details are not repeated here.
  • An embodiment of the present application also provides a UE (terminal), which includes a processor and a communication interface, and the processor is configured to send a target request.
  • This terminal embodiment corresponds to the above-mentioned UE-side method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this UE embodiment, and can achieve the same technical effect.
  • FIG. 16 is a schematic diagram of the hardware structure of a UE that implements an embodiment of the present application.
  • the UE 100 includes but is not limited to: at least one of the radio frequency unit 101, the network module 102, the audio output unit 103, the input unit 104, the sensor 105, the display unit 106, the user input unit 107, the interface unit 108, the memory 109, the processor 110, etc. Some parts.
  • the terminal 100 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 110 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
  • the terminal structure shown in FIG. 16 does not constitute a limitation on the terminal.
  • the terminal may include more or fewer components than shown in the figure, or some components may be combined or arranged differently, which will not be described again here.
  • the input unit 104 may include a graphics processing unit (Graphics Processing Unit, GPU) 1041 and a microphone 1042.
  • the graphics processor 1041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 106 may include a display panel 1061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 107 includes a touch panel 1071 and at least one of other input devices 1072 .
  • Touch panel 1071 is also called a touch screen.
  • the touch panel 1071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 1072 may include, but are not limited to, physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 101 after receiving downlink data from the network side device, the radio frequency unit 101 can transmit it to the processor 110 for processing; in addition, the radio frequency unit 101 can send uplink data to the network side device.
  • the radio frequency unit 101 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
  • Memory 109 may be used to store software programs or instructions as well as various data.
  • the memory 109 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc.
  • memory 109 may include volatile memory or nonvolatile memory, or memory 109 may include both volatile and nonvolatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • ROM Read-Only Memory
  • PROM programmable read-only memory
  • Erasable PROM Erasable PROM
  • EPROM electrically erasable programmable read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • the processor 110 may include one or more processing units; optionally, the processor 110 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 110 .
  • the radio frequency unit 101 is used to send a target request related to the AI model; the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send a third request.
  • Information about a resource is a reference signal resource used by the network side device for beam scanning; the second request is used to request to obtain information on the number of transmit beams of the network side device; the third request is used to request to change the number associated with the first resource.
  • Beam information or requests are sent to multiple first resources, and response information corresponding to the target request is received; the response information is used to perform target operations, and the target operations include any of the following: selecting an AI model corresponding to the response information, and processing the AI model.
  • the embodiment of the present application provides a UE. Since the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning and requesting to send the number information of the beams, the UE can request the target request based on the reference signal resources.
  • the number information of the beams select the AI model, or process the AI model, such as training the AI model or inferring the AI model, etc., because before the processing of the AI model is completed, information can be exchanged, such as the interaction of the first request,
  • the second request or the third request therefore, can guarantee the amount of input data and output data of the AI model used for processing or selection, or the type of AI model used for processing or selection, so that the deployment end of the AI model can obtain enough
  • the number of input parameters to process the AI model therefore, improves the performance of the AI model, and because the first device can also send a third request, that is, a request to change the beam information associated with the first resource or a request to send multiple
  • the first resource therefore, can ensure the normal use of the AI model under different scenarios/configurations. In this way, the performance of the AI model is improved and the applicability of the AI model is ensured.
  • the above target information includes the first request, and the above radio frequency unit 101 is specifically configured to receive a second resource from a network side device; the second resource is a reference signal resource used for beam scanning.
  • the processor 110 is further configured to obtain, according to the second resource, beam quality information of the beam corresponding to the second resource. And process the AI model based on the acquired beam quality information.
  • the processor 110 is specifically used to perform
  • the above target request includes: a second request; and the radio frequency unit 101, specifically configured to receive third quantity information, where the third quantity information is used to indicate the number of quantity information of transmission beams.
  • the processor 110 is also configured to select an AI model corresponding to the third quantity information based on the third quantity information.
  • the above target request includes a third request.
  • the processor 110 is also configured to measure the target resource before sending the target request related to the AI model to obtain the target measurement result; the target resource is used for beam measurement. H.
  • the radio frequency unit 101 is specifically configured to send a third request if the first measurement result is less than and/or equal to the first threshold value, and the first measurement result is a measurement result that satisfies the first condition among the target measurement results,
  • the first condition is that there is a first preset quantity or a first preset proportion, and it is less than or equal to the first threshold value.
  • the radio frequency unit 101 is specifically configured to determine the target feedback information according to the target measurement result; and send the target feedback information; the target feedback information includes the target measurement result, and the target feedback information is used to implicitly indicate the third request.
  • the radio frequency unit 101 is specifically configured to receive third information; the third information includes information of a third resource, and the third resource is the beam information associated with the first resource after the network side device has been replaced or the network side device sends of multiple first resources.
  • the processor 110 is also used to process the AI model according to the third information.
  • Embodiments of the present application also provide a readable storage medium on which a program or instructions are stored.
  • a program or instructions are stored.
  • each process of the above sending method embodiment is implemented and the same can be achieved. To avoid repetition, the technical effects will not be repeated here.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • Embodiments of the present application also provide a network-side device, including a processor and a communication interface.
  • the processor is configured to receive a target request related to the AI model and send a response message corresponding to the target request.
  • This network-side device embodiment corresponds to the above-mentioned method embodiment when the first device is a network-side device.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this network-side device embodiment, and can achieve the same technical effects.
  • the embodiment of the present application also provides a network side device.
  • the network side device 700 includes: an antenna 71 , a radio frequency device 72 , a baseband device 73 , a processor 74 and a memory 75 .
  • the antenna 71 is connected to the radio frequency device 72 .
  • the radio frequency device 72 receives information through the antenna 71 and sends the received information to the baseband device 73 for processing.
  • the baseband device 73 processes the information to be sent and sends it to the radio frequency device 72.
  • the radio frequency device 72 processes the received information and then sends it out through the antenna 71.
  • the method performed by the network side device in the above embodiment can be implemented in the baseband device 73, which includes a baseband processor.
  • the baseband device 73 may include, for example, at least one baseband board on which multiple chips are disposed, as shown in FIG. Program to perform the network device operations shown in the above method embodiments.
  • the network side device may also include a network interface 76, which is, for example, a common public radio interface (CPRI).
  • a network interface 76 which is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 700 in this embodiment of the present invention also includes: instructions or programs stored in the memory 75 and executable on the processor 74.
  • the processor 74 calls the instructions or programs in the memory 75 to execute each of the steps shown in Figure 17. The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.
  • the radio frequency device 72 is used to receive target requests related to the AI model.
  • the target request includes at least one of the following: a first request, a second request and a third request; wherein the first request is used to request the network side device to send a first resource; the first resource is a reference signal used by the network side device for beam scanning.
  • the second request is used to request the number information of the transmission beams of the network side device;
  • the third request is used to request the network side device to change the beam information associated with the first resource or to request the network side device to send multiple first resources;
  • processing 74 Used to send response information corresponding to the target request;
  • the response information is used to perform target operations, and the target operations include any of the following: selecting the AI model corresponding to the response information, and processing the AI model.
  • the embodiment of the present application provides a network side device. Since the UE can send a target request related to the AI model, for example, requesting the network side device to send reference signal resources for beam scanning and requesting to send beam quantity information, from The AI model can be selected based on the reference signal resources and/or the number of transmitted beams, or the AI model can be processed, such as training the AI model or inferring the AI model, etc.
  • the number of input data and output data of the AI model for processing or selection can be guaranteed, or the type of AI model for processing or selection, so that the AI model
  • the deployment end can obtain a sufficient number of input parameters to process the AI model, thus improving the performance of the AI model, and because the first device can also send a third request, that is, requesting to change the beam associated with the first resource Information or requests are sent to multiple first resources, thus ensuring the normal use of the AI model under different scenarios/configurations. In this way, the performance of the AI model is improved and the applicability of the AI model is ensured.
  • the target information includes: a first request, the processor 74, specifically used to send a second resource; the second resource is a reference signal resource used for beam scanning; and the second resource is used to process the AI model.
  • the target request includes: a second request; the processor 74, specifically configured to send third quantity information, the third quantity information is used to indicate the number of quantity information to send beams; the third quantity information is used to select an AI model.
  • the target request includes: a third request; the radio frequency device 72 is specifically configured to receive the third request; or, receive target feedback information; and obtain the third request according to the target feedback information; the target feedback information is the UE's response to the target resource.
  • the result of the measurement; the target resource is the resource used for beam measurement.
  • the processor 74 is specifically configured to send third information; the third information includes information of a third resource, and the third resource is beam information associated with the first resource after replacement by the network side device or multiple beams sent by the network side device.
  • the first resource; the third information is used to process the AI model.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement each of the above sending method embodiments. The process can achieve the same technical effect. To avoid repetition, it will not be described again here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement the above sending method embodiment.
  • Each process can achieve the same technical effect. To avoid duplication, it will not be described again here.
  • Embodiments of the present application also provide a sending system, including: a UE and a network side device.
  • the UE can be used to perform the steps of the sending method as described above.
  • the network side device can be used to perform the steps of the sending method as described above. step.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种发送方法、用户设备、网络侧设备及可读存储介质，属于通信领域，本申请实施例的发送方法包括：用户设备UE发送与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；UE接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应消息对应的AI模型、处理所述AI模型。

Description

发送方法、用户设备、网络侧设备及可读存储介质
相关申请的交叉引用
本申请主张在2022年08月30日在中国提交的中国专利申请号202211050043.1的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于通信技术领域，具体涉及一种发送方法、用户设备、网络侧设备及可读存储介质。
背景技术
目前，用户设备(User Equipment，UE)和网络侧设备均可以使用人工智能(Artificial Intelligence，AI)模型进行波束对的参考信号接收功率(Reference Signal Receiving Power，RSRP)预测，例如，可以使用部分波束对的RSRP作为输入，从而可以通过AI模型输出所有波束对的RSRP结果，以实现对波束对的RSRP的预测，其中，波束对包括发送波束和接收波束。
然而，上述AI模型是通过训练得到的，但由于AI模型的训练位置和推理位置的不确定性，可能通过UE训练得到的AI模型，或网络侧设备训练得到的AI模型，且训练得到的AI模型的推理位置可能在网络侧，也可能在UE侧，取决于AI模型的使用方法和部署位置，因此，存在AI模型需要从一侧传输到另一侧设备，可能存在AI模型的输入输出数量与AI模型的部署侧设备和/或训练侧设备有关，并且由于在AI模型训练完成后，AI模型输入数据和输出数据的数量，以及AI模型的类型均确定，若没有额外信息的交互，可能导致模型部署端无法获得足够的模型输入参数数量进行模型推理，或无法获得足够的模型输入和输出参数数量进行模型训练等，从而导致AI模型性能下降或无法使用，同时，若AI模型部署侧拥有较多模型用于适用不同场景/配置的情况下，模型部署侧选错了模型，从而会导致AI模型性能急剧下降，甚至无法使用。
发明内容
本申请实施例提供一种发送方法、用户设备、网络侧设备及可读存储介质，能够提升AI模型的性能和适用性。
第一方面，提供了一种发送方法，该方法包括：UE发送与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；UE接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应消息对应的AI模型、处理AI模型。
第二方面，提供了一种发送装置，该装置包括：发送模块和接收模块；发送模块，用于发送与AI模型相关目标请求。目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源的信息；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；接收模块，用于接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
第三方面，提供了一种发送方法，该方法包括：网络侧设备接收与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求网络侧设备的发送波束的数量信息；第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求网络侧设备发送多个第一资源；网络侧设备发送与目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
第四方面，提供了一种发送装置，该装置包括：接收模块和发送模块；接收模块，用于接收与AI模型相关的目标请求。目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求网络侧设备的发送波束的数量信息；第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求网络侧设备发送多个第一资源；发送模块，用于发送与接收模块接收的目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
第五方面，提供了一种UE，该UE包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面，提供了一种UE，包括处理器及通信接口，其中，所述处理器用于发送与AI模型相关的目标请求；并接收目标请求对应的响应信息。
第七方面，提供了一种网络侧设备，该网络侧设备包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第三方面所述的方法的步骤。
第八方面，提供了一种网络侧设备，包括处理器及通信接口，其中，所述处理器用于接收与AI模型相关的目标请求；并发送目标请求对应的响应信息。
第九方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤；或者实现如第三方面所述的方法的步骤。
第十方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第一方面所述的方法的步骤；或者实现如第三方面所述的方法的步骤。
第十一方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面所述的方法的步骤；或者实现如第三方面所述的方法的步骤。
在本申请实施例中，UE发送与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；UE接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应消息对应的AI模型、处理AI模型。由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型， 从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
附图说明
图1是本申请实施例提供的一种通信系统的架构示意图；
图2是本申请实施例提供的一种AI神经网络的构成示意图；
图3是本申请实施例提供的一种神经元的构成示意图；
图4是本申请实施例提供的一种反馈报告结构示意图；
图5是本申请实施例提供的一种基于组的波束报告的反馈报告结构示意图；
图6是本申请实施例提供的一种使用AI方法进行波束预测的示意图；
图7是本申请实施例提供的一种使用AI方法增强波束预测性能的示意图；
图8是本申请实施例提供的一种使用AI方法改进增强波束预测性能的示意图；
图9是本申请实施例提供的一种发送方法的流程图之一；
图10是本申请实施例提供的一种发送方法的交互图之一；
图11是本申请实施例提供的一种发送方法的交互图之二；
图12是本申请实施例提供的一种发送方法的流程图之二；
图13是本申请实施例提供的一种发送装置的结构示意图之一；
图14是本申请实施例提供的一种发送装置的结构示意图之二；
图15是本申请实施例提供的一种通信设备的硬件结构示意图；
图16是本申请实施例提供的一种UE的硬件结构示意图；
图17是本申请实施例提供的一种网络侧设备的硬件结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio，NR)系统，并且在以下大部分描述中使用NR术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6th Generation，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端 11和网络侧设备12。其中，终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备，如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer，PC)、柜员机或者自助机等终端侧设备，可穿戴式设备包括：智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是，在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备，其中，接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等，基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station，BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set，BSS)、扩展服务集(Extended Service Set，ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point，TRP)或所属领域中其他某个合适的术语，只要达到相同的技术效果，所述基站不限于特定技术词汇，需要说明的是，在本申请实施例中仅以NR系统中的基站为例进行介绍，并不限定基站的具体类型。
AI在众多领域均获得了广泛的应用，且AI网络有多种实现方式，例如：神经网络、决策树、支持向量机、贝叶斯分类器等。
图2示出了一种AI神经网络的构成示意图，如图2所示，AI神经网络由神经元组成。
图3示出了一种神经元的构成示意图，如图3所示，a1，a2，…aK为输入，w为权值(或称为乘性系数)，b为偏置(或称为加性系数)，σ(.)为激活函数。常见的激活函数包括Sigmoid、tanh、修正线性单元(Rectified Linear Unit，ReLU)等。
神经网络的参数可以通过优化算法进行优化。优化算法是一种能够协助开发人员或用户将目标函数(也称为：损失函数)最小化或者最大化的一类算法。而目标函数往往是模型参数和数据的数学组合。例如：在给定数据X和其对应的标签Y的情况下，开发人员可以构建一个神经网络模型f(.)，并且可以通过该神经网络模型f(.)，根据输入x就可以得到预测输出f(x)，并且可以计算出预测值和真实值之间的差距(f(x)-Y)，也就是损失函数。其中，开发人员的目的是找到合适的W，b，使得上述的损失函数的值可以达到最小，而损失值越小，则说明模型越接近于真实情况。
本申请实施例中优化算法可以是基于误差反向传播(Error Back Propagation，BP)算法。BP算法的基本思想是，学习过程由信号的正向传播与误差的反向传播两个过程组成。正向传播时，输入样本从输入层传入，经各隐层逐层处理后，传向输出层。若输出层的实际输出与期望的输出不符，则转入误差的反向传播阶段。误差反传是将输出误差以某种形式通过隐层向输入层逐层反传，并将误差分摊给各层的所有单元，从而获得各层单元的误差信号，此误差信号即作为修正各单元权值的依据。这种信号正向传播与误差反向传播的各层权值调整过程，是周而复始地进行的。权值不断调整的过程，也就是网络的学习训练过程。此过程一直进行到网络输出的误差减少到可接受的程度，或进行到预先设定的学习次数为止。
优化算法还可以包括梯度下降(Gradient Descent，GD)、随机梯度下降(Stochastic Gradient Descent，SGD)、小批量梯度下降(Mini-Batch Gradient Descent)、动量法(Momentum)、带动量的随机梯度下降(Nesterov)、自适应梯度下降(ADAptive GRADient  descent，Adagrad)、Adadelta、均方根误差降速(Root Mean Square Prop，RMSprop)、自适应动量估计(Adaptive Moment Estimation，Adam)等。
这些优化算法在误差反向传播时，其都是根据损失函数得到的误差/损失，对当前神经元求导数/偏导，加上学习速率、之前的梯度/导数/偏导等影响，从而可以得到梯度，并将梯度传给上一层。
下面对本申请实施例提供的确定方法涉及的一些概念和/或术语做一下解释说明。
波束指示(Beam Indication)机制
在经过波束测量和波束报告后，网络侧设备可以对下行链路与上行链路的信道或参考信号做波束指示，用于网络侧设备与UE之间建立波束链路，实现信道或参考信号的传输。
对于物理下行控制信道(Physical Downlink Control Channel，PDCCH)的波束指示，网络使用无线资源控制(Radio Resource Control，RRC)信令为每个核心集(CORESET)配置K个传输配置指示(Transmission Configuration Indication，TCI)state；当K>1时，由媒体接入层控制单元(Media Access Control Layer，MAC CE)指示或激活1个TCI state，当K＝1时，不需要额外的MAC CE命令。UE在监听PDCCH时，对CORESET内全部search space使用相同准共址(Quasi-colocation，QCL)，即相同的TCI state来监听PDCCH。该TCI状态中的参考信号(Reference Signal)(例如周期CSI-RS resource、半持续CSI-RS resource、SS block等)与UE-specific PDCCH解调参考信号(Demodulation Reference Signal，DMRS)端口是空间QCL的。UE根据该TCI状态即可获知使用哪个接收波束来接收PDCCH。
对于PDSCH的波束指示，网络侧设备通过RRC信令配置M个TCI state，再使用MAC CE命令激活2N个TCI state，然后通过DCI的N-bit TCI field来通知TCI状态，该TCI状态中的Reference Signal与要调度的PDSCH的DMRS端口是QCL的。UE根据该TCI状态即可获知使用哪个接收波束来接收PDSCH。
对于CSI-RS的波束指示，当CSI-RS类型为周期CSI-RS时，网络侧设备通过RRC信令为CSI-RS资源(resource)配置QCL信息。当CSI-RS类型为半持续CSI-RS时，网络侧设备通过MAC CE命令来从RRC配置的CSI-RS resource set中激活一个CSI-RS resource时指示其QCL信息。当CSI-RS类型为非周期CSI-RS时，网络侧设备通过RRC信令为CSI-RS resource配置QCL，并使用DCI来触发CSI-RS。
对于物理上行链路控制信道(Physical Uplink Control Channel，PUCCH)的波束指示，网络侧设备使用RRC信令通过PUCCH-SpatialRelationInfo参数为每个PUCCH resource配置空间关系信息(Spatial Relation Information)，当为PUCCH resource配置的Spatial Relation Information包含多个时，使用MAC-CE指示或激活其中一个spatial relation information。当为PUCCH resource配置的spatial relation information只包含1个时，不需要额外的MAC CE命令。
对于PUSCH的波束指示，PUSCH的spatial relation信息是当PDCCH承载的下行控制信道信息(Downlink Control Information，DCI)调度物理上行共享信道(Physical Uplink Shared Channel，PUSCH)时，DCI中的上行调度请求指示信息域(Schduling Request Indication field，SRI field)的每个SRI代码点(codepoint)指示一个SRI，该SRI用于指示PUSCH的Spatial Relation Information。
对于SRS的波束指示，当SRS类型为周期SRS时，网络通过RRC信令为SRS resource配置Spatial Relation Information。当SRS类型为半持续SRS时，网络通过MAC CE命令来从RRC配置的一组Spatial Relation Information中激活一个。当SRS类型为非周期SRS时，网络通过RRC信令为SRS resource配置Spatial Relation Information。
对于进一步的波束指示改进，提出了统一传输配置指示状态(unified TCI indication) 的概念，即通过一个DCI中的TCI域，指示后续的各参考信号以及多个信道的波束信息。
需要说明的是，上述波束信息、Spatial Relation信息、空间域传输滤波器Spatial Domain Transmission Filter信息、空间滤波Spatial Filter信息、TCI State信息、QCL信息、QCL参数、Spatial Relation信息，波束关联关系等，其所表达的意思相同或相近。其中，下行波束信息通常可使用TCI state信息、QCL信息表示。上行波束信息通常可使用Spatial Relation信息表示。
解调灵敏度计算方法
接收灵敏度，其可以通过解调公式来实现，其中解调公式为：S(dBm)＝热噪声(dBm)+10log(BW)+NF(dB)+解调门限，热噪声为-174dbm/Hz。
忽略解调门限，以30GHz，120kH SCS为例，
一个子载波上的底噪＝-174+10*log10(120*10^3)+10＝-174+50.8+10＝-113.2dBm。
因此对于高频大子载波间隔来说，其底噪的能量相对较大。
关于波束测量和报告(Beam Measurement And Beam Reporting)
由于模拟波束赋形是全带宽发射的，并且每个高频天线阵列的面板上每个极化方向阵元仅能以时分复用的方式发送模拟波束，因此模拟波束的赋形权值是通过调整射频前端移相器等设备的参数来实现。
可使用轮询的方式进行模拟波束赋形向量的训练，即每个天线面板每个极化方向的阵元以时分复用方式依次在约定时间发送训练信号(即候选的赋形向量)，终端经过测量后反馈波束报告，供网络侧在下一次传输业务时采用该训练信号来实现模拟波束发射。波束报告的内容通常包括最优的若干个发射波束标识以及测量出的每个发射波束的接收功率。
在做波束测量时，网络侧设备会配置参考信号资源集合(RS resource set)，其中包括至少一个参考信号资源，例如SSB resource或CSI-RS resource。UE测量每个RS resource的L1-RSRP/L1-SINR，并将最优的至少一个测量结果上报给网络侧设备，上报内容包括SSBRI或CRI、及L1-RSRP/L1-SINR。该报告内容反映了至少一个最优的波束及其质量，供网络侧设备确定用来向UE发送信道或信号的波束。
当UE反馈报告中仅包含一个L1-RSRP时，使用7bit的量化方法，量化步进为1dB,量化范围是-140dBm到-44dBm。当UE被指示的反馈报告中包含多个L1-RSRP，或使能了基于组的波束报告Group Based Beam Report时，最强的RSRP量化使用7bit量化，其余RSRP量化使用4bit的差分量化方法，量化步进为2dB.
图4示出了一种反馈报告结构示意图。
图5示出了一种基于组的波束报告的反馈报告结构示意图。
其中，反馈报告数量是通过网络侧设备配置给UE的参数进行确定的，并通过RRC配置参数，以及配置UE的反馈报告中应该包含的RS以及RSRP的数量，数量配置的取值是1,2,3,4，默认值为1，此外，该数量限制是基于UE能力的，UE会先上报能支持的最大数量。
使用AI方法进行波束预测：
图6示出了一种使用AI方法进行波束预测的示意图。如图6所示，可以使用部分波束对的RSRP作为输入，AI模型的输出则是所有波束对的RSRP结果。其中波束对是由发送波束和接收波束组成的，并且该AI模型的输入数量是挑选出来的部分波束对的数量，输出数量则是所有波束对的数量。
图7示出了一种使用AI方法增强波束预测性能的示意图。如图7所示，可以在输入侧增加了关联信息，关联信息一般是挑选出来用于输入的波束对对应的角度相关信息，波束ID信息等。因此这种模型的输入数量还与挑选出来的部分波束对的数量相关，输出数量还是等于所有波束对的数量。
图8示出了一种使用AI方法改进增强波束预测性能的示意图。如图8所示，该方法主要是通过AI模型改变期望信息，来影响AI模型的输出。
其中AI模型的输入类型包括以下至少之一：
波束质量相关信息；
波束相关的关联信息；
A端发送波束相关的关联信息；
B端接收波束相关的关联信息；
B端期望的波束相关的关联信息；
B端期望的B端接收波束相关的关联信息；
B端期望的A端发送波束相关的关联信息；
与波束质量相关信息的时间相关信息；
期望的预测时间相关信息。
波束相关的关联信息是指所述波束对应的波束信息，波束信息包含但不限于以下至少之一：
波束ID信息；
波束角度信息；
波束增益信息；
波束宽度信息等。
其中，波束ID信息用于表征所述波束的身份识别的信息，包含但不限于以下至少之一：
发送波束ID；
接收波束ID；
波束ID；
所述波束对应的参考信号set ID；
所述波束对应的参考信号resource ID；
唯一标识的随机ID；
额外AI网络处理后的编码值；
波束角度信息等。
其中，波束角度信息用于表征所述波束对应的角度相关信息，包含但不限于以下至少之一：
角度信息；
发送角度信息；
接收角度信息。
其中，角度信息是用于表征角度的相关信息，例如，角度，弧度，索引编码值，额外AI网络处理后的编码值等
然而，对于AI模型的训练位置，推理位置，都还不确定，因此，训练位置和推理位置可能都在一个位置，例如都在UE，基站或中心节点等，或者，模型训练位置和推理位置是在两个位置，例如，训练位置在基站，推理位置在UE。
因此，由于AI模型的训练位置和推理位置的不确定性，可能会导致AI模型的性能下降；并且，由于AI模型的实现的可行方案也较多，因此可能会导致AI模型不匹配的情况出现，并且，AI模型方案的实现是需要一些辅助的信息交互，从而才能保证AI模型的正常使用，因此，亟需一种方法保证AI模型的正常使用，并且提升AI模型的性能。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的确定方法进行详细地说明。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的发送进行详细地说明。
实施例一
本申请实施例提供一种发送方法，图9示出了本申请实施例提供的一种发送方法的流程图。如图9所示，本申请实施例提供的发送方法可以包括下述的步骤201和步骤202。
步骤201、用户设备UE发送与AI模型相关目标请求。
本申请实施例中，目标请求包括以下至少之一：第一请求、第二请求和第三请求；
其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；
第二请求用于请求网络侧设备的发送波束的数量信息；
第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求发送多个第一资源。
步骤202、UE接收目标请求对应的响应信息。
本申请实施例中，响应信息用于执行目标操作，该目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
本申请实施例提供一种发送方法，UE发送与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；UE接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应消息对应的AI模型、处理AI模型。由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
可选地，上述步骤202具体可以通过下述的步骤202a实现。
步骤202a、UE接收第二资源。
本申请实施例中，第二资源为网络侧设备根据目标请求确定的，用于波束扫描的参考信号资源。
可选地，本申请实施例提供的发送方法还包括下述的步骤202b和步骤202c。
步骤202b、UE根据第二资源，获取第二资源对应的波束的波束质量信息。
步骤202c、UE根据波束质量信息，处理AI模型。
本申请实施例中，UE可以在发送了第一请求之后，接收由网络侧设备根据第一请求确定的第二资源，该第二资源为用于波束扫描的参考信号资源；从而UE可以根据该第二资源，对AI模型进行处理，例如训练AI模型，推理AI模型等，因此，不仅可以在对AI模型处理之前，完成信息的交互，并且，由于获得的第二资源可以用于保证处理的AI模型输入数据和输出数据的数量，从而可以使得UE可以正常对AI进行使用，以及后续的处理。
可选地，本申请实施例中，上述步骤202c中的“处理AI模型”具体可以通过下述的步骤202c1至步骤202c5中的任一项实现。
步骤202c1、UE训练AI模型。
步骤202c2、UE调整AI模型的参数。
步骤202c3、UE监测AI模型的性能。
步骤202c4、UE推理AI模型。
步骤202c5、UE收集AI模型的所需数据。
本申请实施例中，UE可以对AI模型执行多种处理，例如训练、调整参数、推理或收集其所需要的数据，从而可以保证AI模型的适用性，或模型选择错误。
可选地，本申请实施例中，第一请求关联以下至少之一：
用途信息、第一数量信息；用途信息用于指示UE请求第一资源的用途，第一数量信息用于指示以下至少之一：
第一资源的数量；
第一资源对应的波束的数量；
第一资源占用的符号的数量。
本申请实施例中，UE可以在发送第一请求的同时，在第一请求中携带请求第一资源的数量信息，从而接收端可以在接收到UE的第一请求之后，根据第一请求中携带的数量信息，确定UE所需要的资源的数量，或是资源对应的波束的数量，从而保证UE可以接收到所需的资源的数量，或是资源对应的波束的数量等，并且保证AI模型的正常使用。
可选地，本申请实施例中，用途信息用于指示以下至少之一：
训练AI模型；
推理AI模型；
收集AI模型的所需数据；
调整AI模型的参数；
监测AI模型的性能。
本申请实施例中，UE可以在发送第一请求的同时，在第一请求中携带请求第一资源的用途信息，从而接收端可以在接收到UE的第一请求之后，根据第一请求中携带的用途信息，确定UE所需要的用途，从而保证UE可以接收到所需的资源的数量，或是资源对应的波束的数量等，并且保证AI模型的正常使用。
示例性地，若用途信息指示：训练目标人工智能AI模型，则表示UE需求数量较多的波束资源，该数量较多的波束资源需要用于AI模型的输入和输出，从而网络侧设备可以根据事先确定的，和/或协商的，和/或约定的特性较多数量，进行波束资源的配置。
示例性地，若用途信息指示：推理所述AI模型，则表示UE需求数量较少的波束资源，该数量较少的波束资源需要用于AI模型的输入和输出，从而网络侧设备可以根据事先确定的，和/或协商的，和/或约定的特性较少数量，进行波束资源的配置。
可选地，本申请实施例中，第一数量信息是网络侧设备根据第一资源的请求用途确定的。
可选地，本申请实施例中，第一数量信息指示的数量是通过UE上报的第一能力信息确定的，该第一能力信息为AI模型的能力信息或UE的能力信息。
可选地，本申请实施例中，AI模型的能力信息包括以下至少之一：
指示AI模型的输入的数量信息；
指示AI模型的输出的数量信息；
指示AI模型的输入包含的第一目标信息；
指示AI模型的输出包含的第二目标信息；
指示AI模型的输入包含的第一目标信息的数量信息；
指示AI模型的输出包含的第二目标信息的数量信息。
可选地，本申请实施例中，第一目标信息和/或第二目标信息可以包括以下至少之一：
参考信号接收功率RSRP信息；
波束的波束信息；
发送波束的波束信息；
接收波束的波束信息。
可选地，本申请实施例中，波束信息为波束对应的信息；
其中，波束对应的信息包含以下至少之一：
波束的波束身份识别标识ID信息；
波束对应的波束角度信息；
波束的波束增益信息；
波束的波束宽度信息。
可选的，本申请实施例中，第一请求关联或包含第一资源发送数量和/或最小发送数量，即第一请求关联或包含第一数量信息指示的数量。
可选的，本申请实施例中，AI模型的输入的数量信息和/或输出的数量信息与第一数量信息指示的数量有关。
本申请实施例中，UE可以通过发送的第一请求，告知接收端，例如网络侧设备，其AI模型的能力信息，从而网络侧设备可以根据UE的AI模型的能力信息，向UE发送其所需要，或是AI模型所能承受的信息的数值或数值范围，从而可以保证AI模型的正常使用。
可选的，本申请实施例中，第一数量信息指示的数量表征第一资源的数量，或表征的第一资源对应的波束的数量。
示例性地，若在第一资源中的N个资源中，配置了repetition on，则第一数量信息指示的数可以为发送的波束的总数量，此时，发送波束的数量大于第一资源的数量。
可选地，本申请实施例中，
AI模型的能力信息指示的AI模型的输入的数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的AI模型的输出的数量小于或等于第一数量信息指示的数量。
和/或，
AI模型的能力信息指示的第一目标数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的第二目标数量小于或等于第一数量信息指示的数量；
其中，第一目标数量为通过AI模型的输入的数量信息，和第一目标信息确定的；或者，通过第一目标信息的数量信息确定的；
第二目标数量为通过AI模型的输出的数量信息和第二目标信息确定的；或者，通过第二目标信息的数量信息确定的。
可选地，本申请实施例中，UE的能力信息包括第二数量信息，第二数量信息用于指示处理AI模型时所需的波束的数量。
可选地，本申请实施例中，第一资源的重复配置状态为关闭。
可选地，本申请实施例中，第一资源为网络侧配置了repetition off的资源。
可选地，本申请实施例中，第一数量信息指示的数量为UE请求发送第一资源的重复 次数；或UE请求发送第一资源对应的波束的重复次数。
可选地，本申请实施例中，第一资源的重复配置状态为开启。
可选地，本申请实施例中，目标请求包括：第二请求；上述步骤202具体可以通过下述的步骤202d。
步骤202d、UE接收第三数量信息。
本申请实施例中，第三数量信息用于指示发送波束的数量信息的数量。
可选地，本申请实施例提供的发送方法还包括下述的步骤202e。
步骤202e、UE根据第三数量信息，选择与第三数量信息对应的AI模型。
本申请实施例中，由于UE可以向网络侧设备发送其所需要的发送波束的数量信息，从而使得网络侧设备可以根据该数量信息，向UE发送UE所需要的，或是所适用的第三数量信息，从而UE可以根据该第三数量信息，选择合适的AI模型，从而可以避免AI模型的选择错误。
可选地，本申请实施例中，目标请求包括第三请求，在上述步骤201之前，本申请实施例提供的发送方法还包括下述的步骤301、且上述的步骤201具体可以通过下述的步骤201a实现；或者，通过下述的步骤201b和步骤201c实现。
步骤301、UE对目标资源进行测量，得到目标测量结果。
目标资源为用于进行波束测量的资源。
步骤201a、若第一测量结果小于和/或等于第一门限值，则UE发送第三请求。
本申请实施例中，第一测量结果为目标测量结果中满足第一条件的测量结果，第一条件为存在第一预设数量或第一预设比例、且小于或小于等于第一门限值。
步骤201b、UE根据目标测量结果，确定目标反馈信息。
步骤201c、UE发送目标反馈信息。
其中，目标反馈信息中包括目标测量结果，目标反馈信息用于隐式指示第三请求。
可选地，本申请实施例中，第二测量结果用于确定第三请求；
第二测量结果为目标测量结果中满足第二条件的测量结果，第二条件为存在第二预设数量或第二预设比例、且小于或小于等于第二门限值。
可选地，本申请实施例中，第一测量结果或所述第二测量结果包括以下至少之一：
层一的信干噪比；
层一的参考信号接收功率；
层一的参考信号接收质量；
层三的信干噪比；
层三的参考信号接收功率；
层三的参考信号接收质量
可选地，本申请实施例中，第一门限值，和/或第二门限值由以下至少之一确定：
协议约定的测量结果量化的最小门限值；
通过协议约定方式获得的与测量结果有关的门限值；
网络侧设备配置的门限值；
UE上报的门限值。
示例性地，上述协议约定的测量结果量化的最小门限值可以为-140dBm；上述通过协议约定方式获得的与测量结果有关的门限值可以通过计算测量的L1-RSPR的均值确定的一个门限值，或根据噪声系数，子载波间隔大小等确定的数值；上述UE上报的门限值可以为UE根据噪声系数等计算出的，且进行上报的一个门限值。
可选地，本申请实施例中，目标请求信息包括：第三请求；上述步骤202具体可以通过下述的步骤202f实现。
步骤202f、UE接收第三信息。
本申请实施例中，第三信息包括第三资源的信息，该第三资源为所述网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源。
本申请实施例中，由于UE可以随时向网络侧设备发送第三请求，例如在发现第一资源的数量或第一资源关联的波束信息并不适用的情况下，请求网络侧恶化设备更换第一资源的波束信息，或请求网络侧设备发送更多的第一资源，从而使得在不同场景/配置的情况下，AI模型的正常使用。因此，提升了AI模型的性能，保证了AI模型的适用性。
实施例二
本申请实施例提供一种发送方法，该发送方法可以包括下述的步骤11至步骤14。
步骤11、UE发送第一请求。
其中，第一请求包含第一波束扫描请求信息，该第一波束扫描请求信息是用于请求网络侧设备发送第一资源的信息；该第一资源为所述网络侧设备用于波束扫描的参考信号资源。
可选的，本申请实施例中，第一请求关联以下至少之一：
用途信息、第一数量信息；
其中，用途信息用于指示UE请求第一资源的用途，第一数量信息用于指示以下至少之一：
第一资源的数量；
第一资源对应的波束的数量；
第一资源占用的符号的数量。
可选地，本申请实施例中，用途信息用于指示以下至少之一：
训练目标人工智能AI模型；
推理AI模型；
收集AI模型的所需数据；
调整AI模型的参数；
监测AI模型的性能。
可选地，本申请实施例中，网络侧设备在接收到第一请求之后，可以根据第一请求关联的用途信息，确定第一资源的发送数量和/或最小发送数量。
可选地，本申请实施例中，第一数量信息指示的数量是通过UE上报的第一能力信息确定的，该第一能力信息为AI模型的能力信息或UE的能力信息。
可选地，本申请实施例中，基于用途信息获取的第一资源的发送数量和/或最小发送数量可以通过AI模型能力上报确定，即通过UE上报第一能力信息确定，第一能力信息为AI模型的能力信息或所述UE的能力信息。
可选地，本申请实施例中，AI模型的能力信息包括以下至少之一：
指示AI模型的输入的数量信息；
指示AI模型的输出的数量信息；
指示AI模型的输入包含的第一目标信息；
指示AI模型的输出包含的第二目标信息；
指示AI模型的输入包含的第一目标信息的数量信息；
指示AI模型的输出包含的第二目标信息的数量信息。。
可选地，本申请实施例中，第一目标信息和/或第二目标信息，分别包括以下至少之一：
RSRP信息；
波束的波束信息；
发送波束的波束信息；
接收波束的波束信息。
可选地，本申请实施例中，
AI模型的能力信息指示的AI模型的输入的数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的AI模型的输出的数量小于或等于第一数量信息指示的数量。
和/或，
AI模型的能力信息指示的第一目标数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的第二目标数量小于或等于第一数量信息指示的数量；
其中，第一目标数量为通过AI模型的输入的数量信息，和第一目标信息确定的；或者，通过第一目标信息的数量信息确定的；
第二目标数量为通过AI模型的输出的数量信息和第二目标信息确定的；或者，通过第二目标信息的数量信息确定的。
可选地，本申请实施例中，UE的能力信息包括第二数量信息，第二数量信息用于指示处理AI模型时所需的波束的数量。
步骤12、UE接收第二资源。
本申请实施例中，第二资源为网络侧设备根据目标请求确定的，用于波束扫描的参考信号资源。
步骤13、UE根据第二资源，获取第二资源对应的波束的波束质量信息。
步骤14、UE根据波束质量信息，执行与波束质量信息对应的目标操作。
目标操作包括以下任一项：
训练AI模型；
调整AI模型的参数；
监测AI模型的性能；
推理AI模型；
收集AI模型的所需数据。
可选地，本申请实施例中，第一资源的重复配置状态为关闭。
示例性地，图10示出了本申请实施例提供的一种发送方法交互图，如图10所示，AI模型在UE侧进行推理和训练，该发送方法可以包括下述的步骤a至步骤d。
步骤a、UE向网络侧设备(例如基站)发送第一请求。
步骤b、网络侧设备根据第一请求发送配置为repetition off的第一资源。
步骤c、UE测量获得RSRP。
步骤d、UE将RSRP作为AI模型的输入内容，从而使得AI模型输出内容。
例如：若AI模型的输入数量对应的波束数量为：8；输出数量对应的波束数量为：32；
则，UE在对AI模型进行AI模型训练时，UE发送第一请求，请求基站发送32个波束；
或者，
UE在对AI模型进行AI模型推理时，UE发送第一请求，请求基站发送8个波束；
或者，
UE在对AI模型进行AI模型推理时，UE发送第一请求，请求基站发送8个波束；并在基站发送16个波束的情况下，根据UE测量获得16个波束信息，挑选其中8个作为 AI模型的输入。
实施例三
本申请实施例提供一种发送方法，该发送方法可以包括下述的步骤15至步骤17。
步骤15、UE发送第二请求。
本申请实施例中，第二请求用于请求获取网络侧设备的发送波束的数量信息。
步骤16、UE接收第三数量信息。
本申请实施例中，第三数量信息用于指示发送波束的数量信息的数量。
步骤17、UE根据第三数量信息，选择AI模型。
示例性地，图11示出了本申请实施例提供的一种发送方法交互图，如图11所示，AI模型在UE侧进行推理和训练，该发送方法可以包括下述的步骤e至步骤d。
步骤e、UE向网络侧设备(例如基站)发送第二请求。
步骤f、基站根据第二请求，发送基站的发送波束数量信息。
步骤g、UE根据发送波束数量信息，选择AI模型。
例如：若AI模型1的输出数量对应的波束数量为：16、AI模型2的输出数量对应的波束数量为：32；UE可以发送第二请求，基站可以根据该第二请求，发送基站的发送波束数量为32，则UE根据发送波束数量选择AI模型2。
实施例四
本申请实施例提供一种发送方法，该发送方法可以包括下述的步骤18至步骤22。
步骤18、UE发送第三请求。
本申请实施例中，第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源。
可选地，在上述步骤18之前，本申请实施例提供的发送方法还包括下述的步骤19，并且本申请实施例提供两种发送第三请求的方式，即显示方式和隐式方式，其中，显示方式可以通过步骤19和步骤20实现，隐式方式可以通过步骤19、步骤21和步骤22实现。
步骤19、UE对目标资源进行测量，得到目标测量结果。
本申请实施例中，目标资源为用于进行波束测量的资源。
步骤20、若第一测量结果小于和/或等于第一门限值，则UE直接发送第三请求。
本申请实施例中，第一测量结果为目标测量结果中满足第一条件的测量结果，第一条件为存在第一预设数量或第一预设比例、且小于或小于等于第一门限值。
步骤21、UE根据目标测量结果，确定目标反馈信息；
步骤22、UE发送目标反馈信息。
其中，目标反馈信息中包括目标测量结果，目标反馈信息用于隐式指示第三请求。
可选地，本申请实施例中，第二测量结果用于确定第三请求；第二测量结果为目标测量结果中满足第二条件的测量结果，第二条件为存在第二预设数量或第二预设比例、且小于或小于等于第二门限值。
可选地，本申请实施例中，第一门限值，和/或第二门限值由以下至少之一确定：
协议约定的测量结果量化的最小门限值；
通过协议约定方式获得的与测量结果有关的门限值；
网络侧设备配置的门限值；
UE上报的门限值。
实施例五
本申请实施例提供一种发送方法，该发送方法可以包括下述的步骤23。
步骤23、UE发送第五请求信息。
本申请实施例中，
其中，第五请求信息包含第二波束扫描请求信息，该第二波束扫描请求信息是用于请求网络侧设备发送第一资源的信息；该第一资源为所述网络侧设备用于波束扫描的参考信号资源。
可选的，本申请实施例中，第一资源的重复配置状态为开启，即repetition on。
可选的，本申请实施例中，第五请求信息关联以下至少之一：
用途信息、第一数量信息；
其中，用途信息用于指示UE请求第一资源的用途，第一数量信息用于指示以下至少之一：
第一资源的数量；
第一资源对应的波束的数量；
第一资源占用的符号的数量。
可选地，本申请实施例中，用途信息用于指示以下至少之一：
训练目标人工智能AI模型；
推理AI模型；
收集AI模型的所需数据；
调整AI模型的参数；
监测AI模型的性能。
可选地，本申请实施例中，网络侧设备在接收到第五请求信息之后，可以根据第一请求关联的用途信息，确定第一资源的重复数量和/或最小重复数量。
可选的，本申请实施例中，第五请求信息关联或包含第一资源的重复数量或最小重复数量。
可选地，本申请实施例中，AI模型的输入数量和/或输出数量与第一资源的重复数量或最小重复数量有关。
可选地，本申请实施例中，AI模型的输入数量和/或输出数量小于或等于第一资源的重复数量或最小重复数量。
可选地，本申请实施例中，第一资源的重复数量或最小重复数量表征第一资源的数量，或表征的第一资源对应的波束的数量。
可选地，本申请实施例中，第一数量信息指示的数量为UE请求发送第一资源的重复次数；或UE请求发送第一资源对应的波束的重复次数。
实施例五的具体实现方法可以参考实施例一，此处不再赘述。
实施例六
本申请实施例提供一种发送方法，图12示出了本申请实施例提供的一种发送方法的流程图。如图12所示，本申请实施例提供的发送方法可以包括下述的步骤401和步骤402。
步骤401、网络侧设备接收与AI模型相关的目标请求。
本申请实施例中，目标请求包括以下至少之一：第一请求、第二请求和第三请求；
其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；
第二请求用于请求网络侧设备的发送波束的数量信息；
第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求网络侧设备发送多个第一资源；
步骤402、网络侧设备发送与目标请求对应的响应信息。
本申请实施例中，响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
可选地，上述目标信息包括：第一请求，上述步骤402具体可以通过下述的步骤402a 实现。
步骤402a、网络侧设备发送第二资源。
本申请实施例中，第二资源为用于波束扫描的参考信号资源；第二资源用于处理AI模型。
可选地，目标请求包括：第二请求；上述步骤402具体可以通过下述的步骤402b实现。
步骤402b、网络侧设备发送第三数量信息。
本申请实施例中，第三数量信息用于指示发送波束的数量信息的数量；第三数量信息用于选择AI模型。
可选地，目标请求包括：第三请求；上述步骤401具体可以通过下述的步骤401c；或者通过下述的步骤401d和步骤401e实现。
步骤401c、网络侧设备接收第三请求。
步骤401d、网络侧设备接收目标反馈信息。
步骤401e、网络侧设备根据目标反馈信息，获取第三请求。
本申请实施例中，目标反馈信息为UE对目标资源进行测量得到的结果；目标资源为用于进行波束测量的资源。
可选地，目标请求包括：第三请求；上述步骤402具体可以通过下述的步骤402c实现。
步骤402c、网络侧设备发送第三信息。
本申请实施例中，第三信息包括第三资源的信息，第三资源为网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源；第三信息用于处理AI模型。
本申请实施例提供一种发送方法，网络侧设备接收与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源；网络侧设备发送目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应消息对应的AI模型、处理AI模型。由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
本申请实施例提供的发送方法，执行主体可以为发送装置。本申请实施例中以发送装置执行发送方法为例，说明本申请实施例提供的发送装置。
图13示出了本申请实施例中涉及的发送装置的一种可能的结构示意图。如图13所示，该发送装置40可以包括：发送模块41和接收模块42。
其中，发送模块41，用于发送与AI模型相关目标请求；该目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源的信息；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网 络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源。接收模块42，用于接收目标请求对应的响应信息；该响应信息用于执行目标操作，该目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
本申请实施例提供一种发送装置，由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
在一种可能实现的方式中，上述目标信息包括第一请求；上述接收模块42，具体用于从网络侧设备接收第二资源；该第二资源为用于波束扫描的参考信号资源。
在一种可能实现的方式中，本申请实施例提供的发送方法还包括：获取模块和处理模块；获取模块，用于根据接收模块接收的第二资源，获取第二资源对应的波束的波束质量信息。处理模块，用于根据获取模块获取的波束质量信息，处理AI模型。
在一种可能实现的方式中，上述处理模块，具体用于
训练AI模型；
调整AI模型的参数；
监测AI模型的性能；
推理AI模型；
收集AI模型的所需数据。
在一种可能实现的方式中，上述第一请求关联以下至少之一：
用途信息、第一数量信息；该用途信息用于指示UE请求第一资源的用途，该第一数量信息用于指示以下至少之一：
第一资源的数量；
第一资源对应的波束的数量；
第一资源占用的符号的数量。
在一种可能实现的方式中，上述用途信息用于指示以下至少之一：
训练目标人工智能AI模型；
推理AI模型；
收集AI模型的所需数据；
调整AI模型的参数；
监测AI模型的性能。
在一种可能实现的方式中，上述第一数量信息是网络侧设备根据第一资源的请求用途确定的。
在一种可能实现的方式中，第一数量信息指示的数量是通过UE上报的第一能力信息确定的，该第一能力信息为AI模型的能力信息或所述UE的能力信息。
在一种可能实现的方式中，AI模型的能力信息包括以下至少之一：
指示AI模型的输入的数量信息；
指示AI模型的输出的数量信息；
指示AI模型的输入包含的第一目标信息；
指示AI模型的输出包含的第二目标信息；
指示AI模型的输入包含的第一目标信息的数量信息；
指示AI模型的输出包含的第二目标信息的数量信息。。
在一种可能实现的方式中，上述第一目标信息和/或第二目标信息包括以下至少之一：
参考信号接收功率RSRP信息；
波束的波束信息；
发送波束的波束信息；
接收波束的波束信息。
在一种可能实现的方式中，上述波束对应的信息包含以下至少之一：
波束的波束身份识别标识ID信息；
波束对应的波束角度信息；
波束的波束增益信息；
波束的波束宽度信息。
在一种可能实现的方式中，AI模型的能力信息指示的AI模型的输入的数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的AI模型的输出的数量小于或等于第一数量信息指示的数量。
和/或，
AI模型的能力信息指示的第一目标数量小于或等于第一数量信息指示的数量；
和/或，
AI模型的能力信息指示的第二目标数量小于或等于第一数量信息指示的数量；
其中，第一目标数量为通过AI模型的输入的数量信息，和第一目标信息确定的；或者，通过第一目标信息的数量信息确定的；
第二目标数量为通过AI模型的输出的数量信息和第二目标信息确定的；或者，通过第二目标信息的数量信息确定的。
在一种可能实现的方式中，UE的能力信息包括第二数量信息，第二数量信息用于指示处理AI模型时所需的波束的数量。
在一种可能实现的方式中，上述第一资源的重复配置状态为关闭。
在一种可能实现的方式中，上述第一数量信息指示的数量为UE请求发送第一资源的重复次数；或UE请求发送第一资源对应的波束的重复次数。
在一种可能实现的方式中，上述第一资源关联的重复配置状态为开启。
在一种可能实现的方式中，上述目标请求包括：第二请求；上述接收模块42，具体用于接收第三数量信息，该第三数量信息用于指示发送波束的数量信息的数量。
在一种可能实现的方式中，本申请实施例提供的发送装置还包括：选择模块。选择模块，用于根据接收模块42接收的第三数量信息，选择与第三数量信息对应的AI模型。
在一种可能实现的方式中，上述目标请求包括第三请求，本申请实施例提供的发送装置还包括：测量模块；测量模块，用于在发送模块41发送与AI模型相关的目标请求之前，对目标资源进行测量，得到目标测量结果；该目标资源为用于进行波束测量的资源。
在一种可能实现的方式中，上述发送模块41，具体用于若第一测量结果小于和/或等于第一门限值，则发送第三请求，第一测量结果为目标测量结果中满足第一条件的测量结果，第一条件为存在第一预设数量或第一预设比例、且小于或小于等于第一门限值。
在一种可能实现的方式中，上述发送模块41，具体用于根据目标测量结果，确定目 标反馈信息；并发送目标反馈信息；其中，目标反馈信息中包括目标测量结果，目标反馈信息用于隐式指示第三请求。
在一种可能实现的方式中，第二测量结果用于确定第三请求；该第二测量结果为目标测量结果中满足第二条件的测量结果，第二条件为存在第二预设数量或第二预设比例、且小于或小于等于第二门限值。
在一种可能实现的方式中，上述第一门限值，和/或第二门限值由以下至少之一确定：
协议约定的测量结果量化的最小门限值；
通过协议约定方式获得的与测量结果有关的门限值；
网络侧设备配置的门限值；
UE上报的门限值。
在一种可能实现的方式中，上述目标请求包括：第三请求；本申请实施例提供的发送装置还包括：处理模块。上述接收模块，还用于接收第三信息；该第三信息包括第三资源的信息，第三资源为网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源。
在一种可能实现的方式中，上述处理模块，还用于根据第三信息，处理AI模型。
本申请实施例中的发送装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
图14示出了本申请实施例中涉及的发送装置的一种可能的结构示意图。如图14所示，该发送装置50可以包括：接收模块51和发送模块52。
其中，接收模块51，用于接收与AI模型相关的目标请求。目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求网络侧设备的发送波束的数量信息；第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求网络侧设备发送多个第一资源；发送模块52，用于发送与目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
本申请实施例提供一种发送装置，由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
在一种可能实现的方式中，目标信息包括：第一请求，发送模块52，具体用于发送第二资源；第二资源为用于波束扫描的参考信号资源；第二资源用于处理AI模型。
在一种可能实现的方式中，目标请求包括：第二请求；发送模块52，具体用于发送第三数量信息，第三数量信息用于指示发送波束的数量信息的数量；第三数量信息用于选择AI模型。
在一种可能实现的方式中，目标请求包括：第三请求；接收模块51，具体用于网络侧设备接收第三请求；或者，接收模块51，具体用于接收目标反馈信息；并根据目标反馈信息，获取第三请求；目标反馈信息为UE对目标资源进行测量得到的结果；目标资源为用于进行波束测量的资源。
在一种可能实现的方式中，发送模块52，具体用于网络侧设备发送第三信息；第三信息包括第三资源的信息，第三资源为网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源；第三信息用于处理AI模型。
本申请实施例提供的发送装置能够实现上述方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图15所示，本申请实施例还提供一种通信设备700，包括处理器701和存储器702，存储器702上存储有可在所述处理器701上运行的程序或指令，例如，该通信设备700为终端时，该程序或指令被处理器701执行时实现上述发送方法实施例的各个步骤，且能达到相同的技术效果。该通信设备700为网络侧设备时，该程序或指令被处理器701执行时实现上述发送方法实施例的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种UE(终端)，包括处理器和通信接口，处理器用于用于发送目标请求。该终端实施例与上述UE侧方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该UE实施例中，且能达到相同的技术效果。具体地，图16为实现本申请实施例的一种UE的硬件结构示意图。
该UE100包括但不限于：射频单元101、网络模块102、音频输出单元103、输入单元104、传感器105、显示单元106、用户输入单元107、接口单元108、存储器109以及处理器110等中的至少部分部件。
本领域技术人员可以理解，终端100还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器110逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图16中示出的终端结构并不构成对终端的限定，终端可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
应理解的是，本申请实施例中，输入单元104可以包括图形处理单元(Graphics Processing Unit，GPU)1041和麦克风1042，图形处理器1041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元106可包括显示面板1061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板1061。用户输入单元107包括触控面板1071以及其他输入设备1072中的至少一种。触控面板1071，也称为触摸屏。触控面板1071可包括触摸检测装置和触摸控制器两个部分。其他输入设备1072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此不再赘述。
本申请实施例中，射频单元101接收来自网络侧设备的下行数据后，可以传输给处理器110进行处理；另外，射频单元101可以向网络侧设备发送上行数据。通常，射频单元101包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器109可用于存储软件程序或指令以及各种数据。存储器109可主要包括存储程序或指令的第一存储区和存储数据的第二存储区，其中，第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器109可以包括易失性存储器或非易失性存储器，或者，存储器109可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器 (Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。本申请实施例中的存储器109包括但不限于这些和任意其它适合类型的存储器。
处理器110可包括一个或多个处理单元；可选的，处理器110集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器110中。
其中，射频单元101，用于发送与AI模型相关的目标请求；目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源的信息；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求获取网络侧设备的发送波束的数量信息；第三请求用于请求更换第一资源关联的波束信息或请求发送多个第一资源，并接收目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
本申请实施例提供一种UE，由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
可选地，上述目标信息包括所述第一请求，上述射频单元101，具体用于从网络侧设备接收第二资源；第二资源为用于波束扫描的参考信号资源。
可选地，处理器110，还用于根据第二资源，获取第二资源对应的波束的波束质量信息。并根据获取的波束质量信息，处理AI模型。
可选地，处理器110，具体用于
训练AI模型；
调整AI模型的参数；
监测AI模型的性能；
推理AI模型；
收集AI模型的所需数据。
可选地，上述目标请求包括：第二请求；射频单元101，具体用于接收第三数量信息，该第三数量信息用于指示发送波束的数量信息的数量。
可选地，处理器110，还用于根据第三数量信息，选择与第三数量信息对应的AI模型。
可选地，上述目标请求包括第三请求，处理器110，还用于在发送与AI模型相关的目标请求之前，对目标资源进行测量，得到目标测量结果；目标资源为用于进行波束测量 的资源。
可选地，射频单元101，具体用于若第一测量结果小于和/或等于第一门限值，则发送第三请求，第一测量结果为目标测量结果中满足第一条件的测量结果，第一条件为存在第一预设数量或第一预设比例、且小于或小于等于第一门限值。
可选地，射频单元101，具体用于根据目标测量结果，确定目标反馈信息；并发送目标反馈信息；该目标反馈信息中包括目标测量结果，目标反馈信息用于隐式指示第三请求。
可选地，射频单元101，具体用于接收第三信息；该第三信息包括第三资源的信息，该第三资源为网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源。
可选地，处理器110，还用于根据第三信息，处理AI模型。
本申请实施例还提供一种可读存储介质，所述可读存储介质上存储有程序或指令，该程序或指令被处理器执行时实现上述发送方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例还提供一种网络侧设备，包括处理器和通信接口，处理器用于接收与AI模型相关的目标请求，并发送与目标请求对应的响应消息。该网络侧设备实施例与上述第一设备为网络侧设备时的方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中，且能达到相同的技术效果。
具体地，本申请实施例还提供了一种网络侧设备。如图17所示，该网络侧设备700包括：天线71、射频装置72、基带装置73、处理器74和存储器75。天线71与射频装置72连接。在上行方向上，射频装置72通过天线71接收信息，将接收的信息发送给基带装置73进行处理。在下行方向上，基带装置73对要发送的信息进行处理，并发送给射频装置72，射频装置72对收到的信息进行处理后经过天线71发送出去。
以上实施例中网络侧设备执行的方法可以在基带装置73中实现，该基带装置73包括基带处理器。
基带装置73例如可以包括至少一个基带板，该基带板上设置有多个芯片，如图17所示，其中一个芯片例如为基带处理器，通过总线接口与存储器75连接，以调用存储器75中的程序，执行以上方法实施例中所示的网络设备操作。
该网络侧设备还可以包括网络接口76，该接口例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本发明实施例的网络侧设备700还包括：存储在存储器75上并可在处理器74上运行的指令或程序，处理器74调用存储器75中的指令或程序执行图17所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
其中，射频装置72，用于接收与AI模型相关的目标请求。目标请求包括以下至少之一：第一请求、第二请求和第三请求；其中，第一请求用于请求网络侧设备发送第一资源；第一资源为网络侧设备用于波束扫描的参考信号资源；第二请求用于请求网络侧设备的发送波束的数量信息；第三请求用于请求网络侧设备更换第一资源关联的波束信息或请求网络侧设备发送多个第一资源；处理74，用于发送与目标请求对应的响应信息；响应信息用于执行目标操作，目标操作包括以下任一项：选择与响应信息对应的AI模型、处理AI模型。
本申请实施例提供一种网络侧设备，由于UE可以发送与AI模型相关的目标请求，例如，请求网络侧设备发送用于波束扫描的参考信号资源，请求发送波束的数量信息，从 而可以根据参考信号资源和/或发送波束的数量信息，选择AI模型，或是对AI模型进行处理，例如训练AI模型或推理AI模型等等，由于在处理AI模型完成之前，可以交互信息，例如交互第一请求、第二请求或第三请求，因此，可以保证用于处理或选择的AI模型输入数据和输出数据的数量，或用于处理或选择的AI模型的类型，从而使得AI模型的部署端可以获取足够的数量的输入参数，以对AI模型进行处理，因此，提升了AI模型的性能，并且，由于第一设备还可以发送第三请求，即请求更换第一资源关联的波束信息或请求发送多个第一资源，因此，可以保证在不同场景/配置的情况下，AI模型的正常使用。如此，提升了AI模型的性能，保证了AI模型的适用性。
可选地，目标信息包括：第一请求，处理器74，具体用于发送第二资源；第二资源为用于波束扫描的参考信号资源；第二资源用于处理AI模型。
可选地，目标请求包括：第二请求；处理器74，具体用于发送第三数量信息，第三数量信息用于指示发送波束的数量信息的数量；第三数量信息用于选择AI模型。
可选地，目标请求包括：第三请求；射频装置72，具体用于接收第三请求；或者，接收目标反馈信息；并根据目标反馈信息，获取第三请求；目标反馈信息为UE对目标资源进行测量得到的结果；目标资源为用于进行波束测量的资源。
可选地，处理器74，具体用于发送第三信息；第三信息包括第三资源的信息，第三资源为网络侧设备更换后的第一资源关联的波束信息或网络侧设备发送的多个第一资源；第三信息用于处理AI模型。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现上述发送方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现上述发送方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供了一种发送系统，包括：UE及网络侧设备，所述UE可用于执行如上所述的发送方法的步骤，所述网络侧设备可用于执行如上所述的发送方法的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机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(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施 方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (35)

  1. 一种发送方法，所述方法包括：
    用户设备UE发送与AI模型相关的目标请求；
    所述目标请求包括以下至少之一：第一请求、第二请求和第三请求；
    其中，所述第一请求用于请求网络侧设备发送第一资源；所述第一资源为所述网络侧设备用于波束扫描的参考信号资源；
    所述第二请求用于请求所述网络侧设备的发送波束的数量信息；
    所述第三请求用于请求所述网络侧设备更换所述第一资源关联的波束信息或请求所述网络侧设备发送多个所述第一资源；
    所述UE接收所述目标请求对应的响应信息；所述响应信息用于执行目标操作，所述目标操作包括以下任一项：选择与所述响应信息对应的AI模型、处理所述AI模型。
  2. 根据权利要求1所述的方法，其中，所述目标信息包括所述第一请求，
    所述UE接收所述目标请求对应的响应信息，包括：
    所述UE从所述网络侧设备接收第二资源；
    所述第二资源为用于波束扫描的参考信号资源。
  3. 根据权利要求2所述的方法，其中，所述方法还包括：
    所述UE根据所述第二资源，获取所述第二资源对应的波束的波束质量信息；
    所述UE根据所述波束质量信息，处理所述AI模型。
  4. 根据权利要求3所述的方法，其中，所述处理所述AI模型包括以下任一项：
    训练所述AI模型；
    调整所述AI模型的参数；
    监测所述AI模型的性能；
    推理所述AI模型；
    收集所述AI模型的所需数据。
  5. 根据权利要求1所述的方法，其中，所述第一请求关联以下至少之一：
    用途信息；所述用途信息用于指示所述UE请求所述第一资源的用途；
    第一数量信息，所述第一数量信息用于指示以下至少之一：
    所述第一资源的数量；
    所述第一资源对应的波束的数量；
    所述第一资源占用的符号的数量。
  6. 根据权利要求5所述的方法，其中，所述用途信息还用于指示以下至少之一：
    训练所述AI模型；
    推理所述AI模型；
    收集所述AI模型的所需数据；
    调整所述AI模型的参数；
    监测所述AI模型的性能。
  7. 根据权利要求5所述的方法，其中，所述第一数量信息是根据所述用途信息确定的。
  8. 根据权利要求7所述的方法，其中，
    所述第一数量信息指示的数量是根据UE上报的第一能力信息确定的，所述第一能力信息包括AI模型的能力信息或所述UE的能力信息。
  9. 根据权利要求8所述的方法，其中，所述AI模型的能力信息包括以下至少之一：
    指示所述AI模型的输入的数量信息；
    指示所述AI模型的输出的数量信息；
    指示所述AI模型的输入包含的第一目标信息；
    指示所述AI模型的输出包含的第二目标信息；
    指示所述AI模型的输入包含的第一目标信息的数量信息；
    指示所述AI模型的输出包含的第二目标信息的数量信息。
  10. 根据权利要求9所述的方法，其中，所述第一目标信息和/或第二目标信息包括以下至少之一：
    参考信号接收功率RSRP信息；
    波束的波束信息；
    发送波束的波束信息；
    接收波束的波束信息。
  11. 根据权利要求10所述的方法，其中，
    所述波束信息包含以下至少之一：
    所述波束的波束身份识别标识ID信息；
    所述波束对应的波束角度信息；
    所述波束的波束增益信息；
    所述波束的波束宽度信息。
  12. 根据权利要求9所述的方法，其中，
    所述AI模型的能力信息指示的AI模型的输入的数量小于或等于所述第一数量信息指示的数量；
    和/或，
    所述AI模型的能力信息指示的AI模型的输出的数量小于或等于所述第一数量信息指示的数量；
    和/或，
    AI模型的能力信息指示的第一目标数量小于或等于所述第一数量信息指示的数量；
    和/或，
    AI模型的能力信息指示的第二目标数量小于或等于所述第一数量信息指示的数量；
    其中，所述第一目标数量为通过所述AI模型的输入的数量信息，和所述第一目标信息确定的；或者，通过所述第一目标信息的数量信息确定的；
    所述第二目标数量为通过所述AI模型的输出的数量信息和所述第二目标信息确定的；或者，通过所述第二目标信息的数量信息确定的。
  13. 根据权利要求8所述的方法，其中，所述UE的能力信息包括第二数量信息，所述第二数量信息用于指示所述处理所述AI模型时所需的波束的数量。
  14. 根据权利要求1所述的方法，其中，
    所述第一资源的重复配置状态为关闭。
  15. 根据权利要求5所述的方法，其中，所述第一数量信息指示的数量为所述UE请求发送所述第一资源的重复次数；或所述UE请求发送所述第一资源对应的波束的重复次数。
  16. 根据权利要求15所述的方法，其中，
    所述第一资源的重复配置状态为开启。
  17. 根据权利要求1所述的方法，其中，所述目标请求包括：第二请求；
    所述UE接收所述目标请求对应的响应信息，包括：
    所述UE接收第三数量信息，所述第三数量信息用于指示所述发送波束的数量信 息的数量。
  18. 根据权利要求17所述的方法，其中，所述方法还包括：
    所述UE根据所述第三数量信息，选择与所述第三数量信息对应的AI模型。
  19. 根据权利要求1所述的方法，其中，所述目标请求包括所述第三请求，所述UE发送与AI模型相关的目标请求之前，所述方法还包括：
    所述UE对目标资源进行测量，得到目标测量结果；
    所述目标资源为用于进行波束测量的资源。
  20. 根据权利要求19所述的方法，其中，所述UE发送与AI模型相关的目标请求，包括：
    若第一测量结果小于和/或等于第一门限值，则所述UE发送所述第三请求，所述第一测量结果为所述目标测量结果中满足第一条件的测量结果，所述第一条件为存在第一预设数量或第一预设比例、且小于或小于等于所述第一门限值。
  21. 根据权利要求20所述的方法，其中，所述UE发送与AI模型相关的目标请求，包括：
    所述UE根据所述目标测量结果，确定目标反馈信息；
    所述UE发送所述目标反馈信息；
    其中，所述目标反馈信息中包括所述目标测量结果，所述目标反馈信息用于隐式指示所述第三请求。
  22. 根据权利要求21所述的方法，其中，
    第二测量结果用于确定所述第三请求；
    所述第二测量结果为所述目标测量结果中满足第二条件的测量结果，所述第二条件为存在第二预设数量或第二预设比例、且小于或小于等于第二门限值。
  23. 根据权利要求20或22所述的方法，其中，
    所述第一门限值，和/或所述第二门限值由以下至少之一确定：
    协议约定的测量结果量化的最小门限值；
    通过协议约定方式获得的与测量结果有关的门限值；
    所述网络侧设备配置的门限值；
    所述UE上报的门限值。
  24. 根据权利要求1或18所述的方法，其中，所述UE接收所述目标请求对应的响应信息，包括：
    所述UE接收所述第三信息；所述第三信息包括第三资源的信息，所述第三资源为所述网络侧设备更换后的所述第一资源关联的波束信息或所述网络侧设备发送的多个所述第一资源。
  25. 根据权利要求24所述的方法，其中，所述方法还包括：
    所述UE根据所述第三信息，处理所述AI模型。
  26. 一种发送方法，所述方法包括：
    网络侧设备接收与AI模型相关的目标请求；
    所述目标请求包括以下至少之一：第一请求、第二请求和第三请求；
    其中，所述第一请求用于请求网络侧设备发送第一资源；所述第一资源为所述网络侧设备用于波束扫描的参考信号资源；
    所述第二请求用于请求所述网络侧设备的发送波束的数量信息；
    所述第三请求用于请求所述网络侧设备更换所述第一资源关联的波束信息或请求所述网络侧设备发送多个所述第一资源；
    所述网络侧设备发送与所述目标请求对应的响应信息；所述响应信息用于执行目标操作，所述目标操作包括以下任一项：选择与所述响应信息对应的AI模型、处理所 述AI模型。
  27. 根据权利要求26所述的方法，其中，所述目标信息包括：所述第一请求，
    所述网络侧设备发送与所述目标请求对应的响应信息，包括：
    所述网络侧设备发送第二资源；
    所述第二资源为用于波束扫描的参考信号资源；所述第二资源用于处理所述AI模型。
  28. 根据权利要求26所述的方法，其中，所述目标请求包括：所述第二请求；
    所述网络侧设备发送与所述目标请求对应的响应信息，包括：
    所述网络侧设备发送第三数量信息，所述第三数量信息用于指示所述发送波束的数量信息的数量；所述第三数量信息用于选择AI模型。
  29. 根据权利要求26所述的方法，其中，所述目标请求包括：所述第三请求；
    网络侧设备接收与AI模型相关的目标请求，包括：
    所述网络侧设备接收所述第三请求；
    或者，
    所述网络侧设备接收目标反馈信息；
    所述网络侧设备根据所述目标反馈信息，获取所述第三请求；
    所述目标反馈信息为UE对目标资源进行测量得到的结果；所述目标资源为用于进行波束测量的资源。
  30. 根据权利要求26或29所述的方法，其中，所述网络侧设备发送与所述目标请求对应的响应信息，包括：
    所述网络侧设备发送第三信息；所述第三信息包括第三资源的信息，所述第三资源为所述网络侧设备更换后的所述第一资源关联的波束信息或所述网络侧设备发送的多个所述第一资源；所述第三信息用于处理所述AI模型。
  31. 一种发送装置，所述装置包括：发送模块和接收模块；
    所述发送模块，用于发送与AI模型相关目标请求；
    所述目标请求包括以下至少之一：第一请求、第二请求和第三请求；
    其中，所述第一请求用于请求网络侧设备发送第一资源的信息；所述第一资源为所述网络侧设备用于波束扫描的参考信号资源；
    所述第二请求用于请求获取所述网络侧设备的发送波束的数量信息；
    所述第三请求用于请求所述网络侧设备更换所述第一资源关联的波束信息或请求所述网络侧设备发送多个所述第一资源；
    所述接收模块，用于接收所述目标请求对应的响应信息；所述响应信息用于执行目标操作，所述目标操作包括以下任一项：选择与所述响应信息对应的AI模型、处理所述AI模型。
  32. 一种发送装置，所述装置包括：接收模块和发送模块；
    所述接收模块，用于接收与AI模型相关的目标请求；
    所述目标请求包括以下至少之一：第一请求、第二请求和第三请求；
    其中，所述第一请求用于请求网络侧设备发送第一资源；所述第一资源为所述网络侧设备用于波束扫描的参考信号资源；
    所述第二请求用于请求所述网络侧设备的发送波束的数量信息；
    所述第三请求用于请求所述网络侧设备更换所述第一资源关联的波束信息或请求所述网络侧设备发送多个所述第一资源；
    所述发送模块，用于发送与所述接收模块接收的所述目标请求对应的响应信息；所述响应信息用于执行目标操作，所述目标操作包括以下任一项：选择与所述响应信息对应的AI模型、处理所述AI模型。
  33. 一种用户设备UE，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至25中任一项所述的发送方法的步骤。
  34. 一种网络侧设备，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求26至30中任一项所述的发送方法的步骤。
  35. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至25中任一项所述的发送方法的步骤；或者实现如权利要求26至30中任一项所述的发送方法的步骤。
PCT/CN2023/114677 2022-08-30 2023-08-24 发送方法、用户设备、网络侧设备及可读存储介质 WO2024046202A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211050043.1 2022-08-30
CN202211050043.1A CN117676544A (zh) 2022-08-30 2022-08-30 发送方法、用户设备、网络侧设备及可读存储介质

Publications (1)

Publication Number Publication Date
WO2024046202A1 true WO2024046202A1 (zh) 2024-03-07

Family

ID=90084943

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/114677 WO2024046202A1 (zh) 2022-08-30 2023-08-24 发送方法、用户设备、网络侧设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN117676544A (zh)
WO (1) WO2024046202A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021112592A1 (ko) * 2019-12-03 2021-06-10 엘지전자 주식회사 무선 통신 시스템에서 인공 지능 기반의 빔 관리 방법 및 이에 대한 장치
CN114390580A (zh) * 2020-10-20 2022-04-22 维沃移动通信有限公司 波束上报方法、波束信息确定方法及相关设备
CN114423019A (zh) * 2020-10-09 2022-04-29 诺基亚通信公司 监测波束
CN114531698A (zh) * 2020-11-23 2022-05-24 维沃移动通信有限公司 一种波束的处理方法及装置、通信设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021112592A1 (ko) * 2019-12-03 2021-06-10 엘지전자 주식회사 무선 통신 시스템에서 인공 지능 기반의 빔 관리 방법 및 이에 대한 장치
CN114423019A (zh) * 2020-10-09 2022-04-29 诺基亚通信公司 监测波束
CN114390580A (zh) * 2020-10-20 2022-04-22 维沃移动通信有限公司 波束上报方法、波束信息确定方法及相关设备
CN114531698A (zh) * 2020-11-23 2022-05-24 维沃移动通信有限公司 一种波束的处理方法及装置、通信设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ZTE CORPORATION: "Evaluation assumptions on AI/ML for beam management", 3GPP DRAFT; R1-2203250, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. RAN WG1, no. e-Meeting; 20220509 - 20220520, 29 April 2022 (2022-04-29), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052152892 *

Also Published As

Publication number Publication date
CN117676544A (zh) 2024-03-08

Similar Documents

Publication Publication Date Title
CN114390580B (zh) 波束上报方法、波束信息确定方法及相关设备
CN115843054A (zh) 参数选择方法、参数配置方法、终端及网络侧设备
WO2024046202A1 (zh) 发送方法、用户设备、网络侧设备及可读存储介质
WO2024046206A1 (zh) 接收方法、设备及可读存储介质
WO2024041420A1 (zh) 测量反馈处理方法、装置、终端及网络侧设备
WO2024041419A1 (zh) 信息传输方法、装置、终端及网络侧设备
WO2024160085A1 (zh) 通信处理方法、装置、设备及可读存储介质
WO2024041421A1 (zh) 测量反馈处理方法、装置、终端及网络侧设备
WO2024093713A1 (zh) 资源配置方法、装置、通信设备及可读存储介质
WO2024104070A1 (zh) 波束报告的发送方法、接收方法、装置及通信设备
WO2024125391A2 (zh) Ai模型监测方法、ai模型性能的测量方法、装置及设备
WO2024160086A1 (zh) 通信处理方法、装置、设备及可读存储介质
WO2024164962A1 (zh) 通信处理方法、装置、设备及可读存储介质
WO2024078405A1 (zh) 传输方法、装置、通信设备及可读存储介质
WO2023134628A1 (zh) 传输方法、装置和设备
CN117674915A (zh) 确定方法、设备及可读存储介质
WO2023134761A1 (zh) 波束信息交互方法、装置、设备及存储介质
WO2024067281A1 (zh) Ai模型的处理方法、装置及通信设备
WO2024099091A1 (zh) 波束预测方法、装置、终端、网络侧设备及存储介质
WO2024032606A1 (zh) 信息传输方法、装置、设备、系统及存储介质
WO2024149156A1 (zh) 信息传输方法、装置、终端及网络侧设备
WO2024032694A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2024099094A1 (zh) 波束测量方法、装置、终端、网络侧设备及存储介质
CN117676668A (zh) 信息传输方法、装置、终端及网络侧设备
WO2023103911A1 (zh) 测量方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23859235

Country of ref document: EP

Kind code of ref document: A1