WO2023066288A1 - 模型请求方法、模型请求处理方法及相关设备 - Google Patents

模型请求方法、模型请求处理方法及相关设备 Download PDF

Info

Publication number
WO2023066288A1
WO2023066288A1 PCT/CN2022/126134 CN2022126134W WO2023066288A1 WO 2023066288 A1 WO2023066288 A1 WO 2023066288A1 CN 2022126134 W CN2022126134 W CN 2022126134W WO 2023066288 A1 WO2023066288 A1 WO 2023066288A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
model
terminal
conforms
preset
Prior art date
Application number
PCT/CN2022/126134
Other languages
English (en)
French (fr)
Inventor
潘翔
杨晓东
金巴·迪·阿达姆布巴卡
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 WO2023066288A1 publication Critical patent/WO2023066288A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • 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

Definitions

  • the present application belongs to the technical field of communication, and in particular relates to a model request method, a model request processing method and related equipment.
  • AI artificial intelligence
  • the embodiment of the present application provides a model request method, a model request processing method, and related equipment, which can feed back the AI model of the response based on the request of the terminal device, so as to control the size of the AI model while reducing the problems caused by the insufficient generalization ability of the AI model. Impact.
  • a model request method including:
  • the terminal sends a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model;
  • the terminal receives the information of the AI model sent by the network side device;
  • the terminal obtains the AI model according to the information of the AI model.
  • a method for processing a model request including:
  • the network side device receives the target request sent by the terminal
  • the network side device obtains the information of the AI model according to the target request information
  • the network side device sends the information of the AI model to the terminal.
  • a device for requesting a model including:
  • the first sending module is configured to send a target request to the network side device, and the target request is used to request information of an artificial intelligence AI model;
  • the first receiving module is configured to receive the information of the AI model sent by the network side device;
  • a processing module configured to obtain the AI model according to the information of the AI model.
  • a model request processing device including:
  • the second receiving module is configured to receive the target request sent by the terminal
  • a generating module configured to obtain information of the AI model according to the target request information
  • the second sending module is configured to send the information of the AI model to the terminal.
  • a terminal includes a processor, a memory, and a program or instruction stored in the memory and operable on the processor.
  • the program or instruction is executed by the processor The steps of the method described in the first aspect are realized.
  • a terminal including a processor and a communication interface, wherein,
  • the communication interface is used to send a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model; receive the information of the AI model sent by the network side device;
  • the processor is configured to obtain the AI model according to the information of the AI model.
  • a network-side device includes a processor, a memory, and a program or instruction stored in the memory and operable on the processor, and the program or instruction is executed by the The processor implements the steps of the method described in the second aspect when executed.
  • a network side device including a processor and a communication interface, wherein,
  • the communication interface is used to receive a target request sent by a terminal
  • the processor is used to obtain the information of the artificial intelligence AI model according to the target request information
  • the communication interface is further configured to send the information of the AI model to the terminal.
  • a readable storage medium is provided, and programs or instructions are stored on the readable storage medium, and when the programs or instructions are executed by a processor, the steps of the method described in the first aspect are realized, or the steps of the method described in the first aspect are realized, or The steps of the method described in the second aspect.
  • the embodiment of the present application provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions, so as to implement the first aspect The steps of the method, or the steps of the method for realizing the second aspect.
  • a computer program product is provided, the computer program product is stored in a non-transitory storage medium, and the computer program product is executed by at least one processor to implement the method as described in the first aspect, Or implement the method as described in the second aspect.
  • a communication device configured to execute the method described in the first aspect, or execute the method described in the second aspect.
  • the terminal sends a target request to the network-side device, and the target request is used to request the information of the artificial intelligence AI model; the terminal receives the information of the AI model sent by the network-side device; the terminal according to the The information of the AI model is obtained from the AI model.
  • different AI models can be set based on different communication scenarios, and the information of the AI model can be dynamically obtained by the terminal through a request, so that while controlling the size of the AI model, the impact caused by the insufficient generalization ability of the AI model can be reduced. Improve transmission reliability.
  • FIG. 1 is a structural diagram of a network system applicable to an embodiment of the present application
  • FIG. 2 is a structural diagram of neurons applicable to embodiments of the present application.
  • FIG. 3 is a flow chart of a model request method provided by an embodiment of the present application.
  • Fig. 4 is a flowchart of a method for processing a model request provided by an embodiment of the present application
  • Fig. 5 is a structural diagram of a model requesting device provided by an embodiment of the present application.
  • Fig. 6 is a structural diagram of a model request processing device provided by an embodiment of the present application.
  • FIG. 7 is a structural diagram of a communication device provided by an embodiment of the present application.
  • FIG. 8 is a structural diagram of a terminal provided in an embodiment of the present application.
  • FIG. 9 is a structural diagram of a network side device provided by an embodiment of the present application.
  • first, second and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or described herein and that "first" and “second” distinguish objects. It is usually one category, and the number of objects is not limited. For example, there may be one or more first objects.
  • “and/or” in the description and claims means at least one of the connected objects, and the character “/” generally means that the related objects are an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced LTE-Advanced
  • LTE-A Long Term Evolution-Advanced
  • 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
  • system and “network” in the embodiments of the present application are often used interchangeably, and the described technology can be used for the above-mentioned system and radio technology, and can also be used for other systems and radio technologies.
  • the following description describes the New Radio (New Radio, NR) system for example purposes, and uses NR terms in most of the following descriptions. These technologies can also be applied to applications other than NR system applications, such as the 6th generation (6th Generation, 6G) communication system.
  • 6G 6th Generation
  • Fig. 1 shows a block diagram of a wireless communication system to which the embodiment of the present application is applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12 .
  • the terminal 11 can 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, a super mobile personal computer (ultra-mobile personal computer, UMPC), mobile Internet device (Mobile Internet Device, MID), augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) equipment, robot, wearable device (Wearable Device) , vehicle equipment (Vehicle User Equipment, VUE), pedestrian terminals (Pedestrian User Equipment, PUE), smart home (home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.) and other terminal-side equipment, wearable Devices include: smart watches, smart bracelets, smart headphones, smart glasses,
  • the network side device 12 may be a base station or a core network device, where a base station may be called a node B, an evolved node B, an access point, a base transceiver station (Base Transceiver Station, BTS), a radio base station, a radio transceiver, a basic Basic Service Set (BSS), Extended Service Set (ESS), Node B, Evolved Node B (eNB), Home Node B, Home Evolved Node B, Wireless Local Area Network, WLAN) access point, WiFi node, Transmitting Receiving Point (Transmitting Receiving Point, TRP) or some other suitable term in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms, what needs to be explained Yes, in the embodiment of the present application, only the base station in the NR system is taken as an example, but the specific type of the base station is not limited.
  • AI modules such as neural networks, decision trees, support vector machines, and Bayesian classifiers.
  • This application uses a neural network as an example for illustration, but does not limit the specific type of AI module.
  • the neural network is composed of neurons, and the structure of the neurons is shown in Figure 2, where a 1 , a 2 , ... a K is the input, w is the weight, that is, the multiplicative coefficient, and b is the bias, namely Additive coefficient, ⁇ (.) is the activation function.
  • Common activation functions include Sigmoid, tanh, linear rectification function (Rectified Linear Unit, ReLU), etc.
  • z a 1 w 1 + ⁇ +a k w k + ⁇ +a K w K +b.
  • the parameters of the neural network are optimized by an optimization algorithm.
  • An optimization algorithm is a type of algorithm that can help us minimize or maximize an objective function, which can also be called a loss function.
  • the objective function is often a mathematical combination of model parameters and data. For example, given the data X and its corresponding label Y, we construct a neural network model f(.), with the model, the predicted output f(x) can be obtained according to the input x, and the predicted value and the real value can be calculated The gap between (f(x)-Y), this is the loss function. Our purpose is to find the appropriate W and b to minimize the value of the above loss function. The smaller the loss value, the closer our model is to the real situation.
  • the current common optimization algorithms are basically based on the error back propagation (error Back Propagation, BP) algorithm.
  • BP error Back Propagation
  • the basic idea of the BP algorithm is that the learning process consists of two processes: the forward propagation of the signal and the back propagation of the error.
  • the input samples are passed in from the input layer, processed layer by layer by each hidden layer, and passed 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 transmit the output error layer by layer through the hidden layer to the input layer in some form, and distribute the error to all the units of each layer, so as to obtain the error signal of each layer unit, and this error signal is used as the correction unit Basis for weight.
  • This weight adjustment process of each layer of signal forward propagation and error back propagation is carried out repeatedly.
  • the process of continuously adjusting the weights is also the learning and training process of the network. This process has been carried out until the error of the network output is reduced to an acceptable level, or until the preset number of learning times.
  • optimization algorithms are based on the error/loss obtained by the loss function when the error is backpropagated, and the derivative/partial derivative of the current neuron is calculated, and the learning rate, the previous gradient/derivative/partial derivative, etc. are added to obtain the gradient. Pass the gradient to the previous layer.
  • the selected AI algorithm and the model used are also different.
  • the main method of using AI to improve 5G network performance is to enhance or replace existing algorithms or processing modules through neural network-based algorithms and models.
  • algorithms and models based on neural networks can achieve better performance than deterministic algorithms.
  • the more commonly used neural networks include deep neural networks, convolutional neural networks, and recurrent neural networks. With the help of existing AI tools, the construction, training and verification of neural networks can be realized.
  • the core of neural network design is to combine the problems and data to be solved to complete the construction of neural network.
  • the key steps in building a neural network include the selection of the stacking method of the multi-layer network, the selection of the number and depth of neurons in each layer of the network, and the selection of the activation function.
  • the depth of the network and the number of neurons increase, the number of hyperparameters that need to be trained also increases rapidly, and the computing resources consumed by neural network training and the difficulty of network training also increase significantly.
  • the problems of gradient disappearance and gradient explosion will become more prominent, which need to be fully considered when designing the activation function.
  • Algorithm and model performance are inseparable from the data set. If there is a large difference between the training data set and the test data set, it is difficult for the model obtained from the training set to achieve good performance on the test data set.
  • An ideal model is that the training data set, test data set and actual processing data have very good consistency, and the model trained under this condition can have better results. In the actual data set construction process, due to various practical conditions, the training and testing data sets and the data processed in the actual scene cannot guarantee a good match.
  • Model training and updating is another concern. Model training can be performed on network devices, cloud or edge devices. During model training, the data needs to be transmitted to the network element for training or to the cloud. The trained model is updated to the corresponding module.
  • the collection and transmission of data not only involves data privacy issues, but also brings certain demands on network transmission. If the model use and model training are not in the same network element, model transmission is also required, and frequent model transmission also consumes more network resources.
  • the computing power consumed by model updating, the need for network resources, and the timeliness requirements are also important factors to be considered when designing algorithms and models.
  • Figure 3 is a flow chart of a model request method provided in the embodiment of the present application, as shown in Figure 3, including the following steps:
  • Step 301 the terminal sends a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model;
  • the above-mentioned target request may be used to obtain information about an AI model that is not configured on the terminal side, or may be used to update the AI model on the terminal side.
  • the terminal when it does not have the corresponding AI model, it can send a target request to the network side device to obtain the information of the AI model, so as to configure the corresponding AI model.
  • the terminal When the terminal is pre-configured with the AI model, it can send the target request to the network side device Request to update the AI model on the terminal side to match the current communication scenario.
  • Step 302 the terminal receives the information of the AI model sent by the network side device
  • Step 303 the terminal obtains the AI model according to the information of the AI model.
  • the network side device can send the information of the AI model through a radio resource control (Radio Resource Control, RRC) configuration message, or send the information of the AI model through an RRC reconfiguration (RRCReconfiguration) message, or send the information of the AI model through a system information
  • RRC Radio Resource Control
  • RRCReconfiguration RRC reconfiguration
  • SIB System Information Block
  • the terminal After receiving the information of the AI model, the terminal can configure or update the AI model on the terminal side based on the information of the AI model. In this way, it can ensure that the terminal and the network side device have the same understanding of the AI model, thereby improving the reliability of transmission.
  • the terminal sends a target request to the network-side device, and the target request is used to request the information of the artificial intelligence AI model; the terminal receives the information of the AI model sent by the network-side device; the terminal according to the The information of the AI model is obtained from the AI model.
  • different AI models can be set based on different communication scenarios, and the information of the AI model can be dynamically obtained by the terminal through a request, so that while controlling the size of the AI model, the impact caused by the insufficient generalization ability of the AI model can be reduced. Improve transmission reliability.
  • the terminal sending the target request to the network side device includes:
  • the terminal sends the target request to the network side device; wherein,
  • the first condition includes at least one of the following:
  • the location of the terminal satisfies a first preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured first measurement result range
  • the first transmission parameter of the terminal satisfies a second preset condition
  • the terminal has a first requirement
  • the terminal does not have the AI model.
  • the foregoing first condition may be stipulated in a protocol, or may be configured by a network side device.
  • the first condition when the first condition is met, it means that the terminal does not have a corresponding AI model, that is, the network side device has not configured the corresponding AI model information in advance.
  • the information about configuring the AI model for the terminal is requested from the network side through the above target request.
  • the terminal After receiving the information of the AI model, the terminal may configure an AI model or generate an AI model based on the information of the AI model, so as to execute a corresponding terminal behavior through the AI model.
  • the AI model is used for terminal behaviors such as mobility management and positioning.
  • the first preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the first cell identity range corresponding to the pre-configured cell identity
  • the tracking area (Tracking Area, TA) where the terminal is located conforms to the first tracking area identification range corresponding to the pre-configured TA identification;
  • the access network notification area (RAN-based Notification Area, RNA) where the terminal is located conforms to the range of the first area corresponding to the pre-configured RNA;
  • PLMN Public Land Mobile Network
  • the geographic location where the terminal is located conforms to the preconfigured range of the first geographic location.
  • the above-mentioned first cell identity range can be understood as a pre-configured cell identity list, and when the cell identity of the serving cell where the terminal is located belongs to the cell identity list, it can be determined that the serving cell where the terminal is located conforms to the pre-configured cell Identify the corresponding first cell identification range.
  • the above-mentioned first tracking area identification range can be understood as a pre-configured tracking area list. When the TA identification corresponding to the TA where the terminal is located belongs to the first tracking area list, it can be determined that the TA where the terminal is located conforms to the pre-configured TA identification corresponding to the first tracking area list.
  • a tracking area identifies a range.
  • the above-mentioned first area range can be understood as an RNA list, and when the RNA where the terminal is located belongs to the RNA list, it can be determined that the RNA where the terminal is located conforms to the first area range corresponding to the pre-configured RNA.
  • the second preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured first parameter range
  • phase information between different transmitting antennas or between different transmitting ports conforms to preconfigured first phase information, or, the preset characterization of the phase information between different transmitting antennas or between different transmitting ports conforms to the first phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter conforms to a preset configured second parameter range
  • the information corresponding to different receiving beams conforms to the preconfigured first beam information, or the specific representation of the information corresponding to different receiving beams conforms to the first beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured second beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the second beam information;
  • the measurement parameter obtained on at least one frequency conforms to a pre-configured third parameter range
  • the data demodulation soft information conforms to preconfigured first preset information, or, the specific representation of the data demodulation soft information conforms to the first preset information;
  • the data packet transmission bit error rate conforms to preconfigured second preset information, or, the specific characterization of the data packet transmission bit error rate conforms to the second preset information;
  • the data packet transmission block error rate conforms to pre-configured third preset information, or, the specific characterization of the data packet transmission block error rate conforms to the third preset information;
  • the interference measurement information conforms to preconfigured fourth preset information, or the specific representation of the interference measurement information conforms to the fourth preset information.
  • the above data packets may include at least one of the following, physical layer data, media access control (Medium Access Control, MAC) layer data packets, radio link control (Radio Link Control, RLC) layer data packets, packet data aggregation Protocol (Packet Data Convergence Protocol, PDCP) layer data packets and Internet Protocol (Internet Protocol, IP) layer data packets.
  • physical layer data media access control (Medium Access Control, MAC) layer data packets
  • radio link control Radio Link Control, RLC
  • packet data aggregation Protocol Packet Data Convergence Protocol, PDCP
  • Internet Protocol Internet Protocol
  • the terminal sending the target request to the network side device includes:
  • the terminal sends the target request to the network side device; wherein the second condition is an update condition of the AI model;
  • Obtaining the AI model by the terminal according to the information of the AI model includes:
  • the terminal updates the original AI model according to the information of the AI model to obtain a target AI model.
  • the terminal is configured with a corresponding AI model, but the AI model needs to be updated.
  • the AI model can be updated through a target request.
  • the network side device can reconfigure the information of the AI model for the terminal through the RRC reconfiguration message, and the terminal updates the original AI model based on the information of the AI model.
  • the terminal can update the information of the AI model based on the information of the original AI model in a differential manner, and then update the AI model based on the information of the updated AI model, for example, after configuring or generating a new AI model, Then replace the original AI model.
  • the above information of the AI model may only include information that the AI model needs to be updated, or include all information after the AI model is updated, and no further limitation is made here.
  • the second condition includes at least one of the following:
  • the location of the terminal satisfies the third preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured second measurement result range
  • the first transmission parameter of the terminal satisfies a fourth preset condition
  • the terminal has a first requirement
  • the confidence degree of the reasoning result of the AI model reaches a threshold
  • the terminal moves outside the effective area of the AI model
  • the terminal has no valid AI model.
  • the fact that the terminal has no valid AI model may be understood as that the required AI model is in an invalid state or in a deactivated state.
  • the AI model can be determined to be in an invalid state when at least one of the following conditions is met:
  • the terminal moves outside the valid area of the AI model.
  • the foregoing effective area is determined based on at least one of the following: a cell list, a TA list, an RNA list, a PLMN list, and a geographic location.
  • the above-mentioned third preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the second cell identity range corresponding to the pre-configured cell identity
  • the TA where the terminal is located conforms to the second tracking area identification range corresponding to the pre-configured TA identification
  • the RNA where the terminal is located conforms to the range of the second region corresponding to the pre-configured RNA
  • the PLMN where the terminal is located conforms to the range of the second communication network corresponding to the pre-configured PLMN;
  • the geographical location of the terminal conforms to the preconfigured range of the second geographical location.
  • the difference between the above-mentioned third preset condition and the first preset condition is that the values corresponding to each range are different, wherein the values corresponding to each range in the first preset condition and the third preset condition can be It is set according to actual needs, and no further limitation is made here.
  • the fourth preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured fourth parameter range
  • phase information between different transmitting antennas or between different transmitting ports conforms to preconfigured second phase information, or, the preset characterization of the phase information between different transmitting antennas or between different transmitting ports conforms to the second phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter complies with a fifth parameter range of the preset configuration
  • the information corresponding to different receiving beams conforms to the preconfigured third beam information, or, the specific representation of the information corresponding to different receiving beams conforms to the third beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured fourth beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the fourth beam information;
  • the measured parameter obtained on at least one frequency conforms to a pre-configured sixth parameter range
  • the data demodulation soft information conforms to pre-configured fifth preset information, or, the specific representation of the data demodulation soft information conforms to the fifth preset information;
  • the packet transmission bit error rate complies with pre-configured sixth preset information, or, the specific characterization of the data packet transmission bit error rate complies with the sixth preset information;
  • the block error rate conforms to pre-configured seventh preset information, or, the specific characterization of the block error rate conforms to the seventh preset information;
  • the interference measurement information conforms to preconfigured eighth preset information, or the specific representation of the interference measurement information conforms to the eighth preset information.
  • the difference between the above-mentioned fourth preset condition and the second preset condition is that the values corresponding to each range are different, wherein the values corresponding to each range in the second preset condition and the fourth preset condition can be It is set according to actual needs, and no further limitation is made here.
  • the first measurement quantity includes at least one of the following: Signal Noise Ratio (Signal Noise Ratio, SNR), Reference Signal Received Power (Reference Signal Received Power, RSRP), signal and interference plus Noise ratio (Signal-to-noise and Interference Ratio, SINR), reference signal received quality (Reference Signal Received Quality, RSRQ), packet delay (packet delay), round-trip delay (Round-Trip Time, RTT), observation arrival Time difference (Observed time difference of arrival, OTDOA), channel state information (Channel State Information, CSI) corresponding measurement results and user quality of experience (Quality of Experience, QoE) corresponding measurement results.
  • Signal Noise Ratio Signal Noise Ratio
  • RSRP Reference Signal Received Power
  • SINR Signal and interference plus Noise ratio
  • SINR reference signal received quality
  • RSRQ Reference Signal Received Quality
  • packet delay packet delay
  • RTT Round-trip delay
  • observation arrival Time difference Observed time difference of arrival
  • OTDOA channel state information
  • the foregoing first measurement quantity includes a measurement quantity corresponding to a serving cell and a measurement quantity corresponding to a neighboring cell.
  • the measurement resource corresponding to the first measurement quantity includes at least one of the following:
  • Physical downlink control channel Physical downlink control channel (Physical downlink control channel, PDCCH) demodulation reference signal (Demodulation Reference Signal, DMRS);
  • PDCCH Physical downlink control channel
  • DMRS Demodulation Reference Signal
  • PDSCH Physical downlink shared channel
  • CSI-RS Channel State Information Reference Signal
  • Synchronization Signal and PBCH block (SSB);
  • PRS Positioning Reference Signal
  • the measurement resource corresponding to the first measurement quantity may be pre-configured or pre-agreed, and no further limitation is made here.
  • the first transmission parameter includes at least one of the following:
  • the foregoing beam pair includes a sending beam and a receiving beam.
  • the above data may include at least one of the following: physical layer data packets, MAC layer data packets, RLC layer data packets, PDCP layer data packets and IP layer data packets.
  • the above-mentioned first requirement includes at least one of the following: reporting a radio resource management (Radio resource management, RRM) measurement report; handover decision; redirection decision; generating a positioning result; reporting channel state information ( Channel State Information, CSI); user trajectory prediction; user business demand prediction; user slice demand prediction.
  • RRM Radio resource management
  • CSI Channel State Information
  • the fact that the terminal has the first requirement can be understood as that the terminal needs to perform the operation corresponding to the first requirement, for example, the terminal needs to report the RRM measurement report, or the terminal needs to make a handover decision, or the terminal needs to make a redirection decision, Or in a situation where the terminal needs to produce a positioning result, etc., it may be determined that the terminal has a corresponding first demand, so as to trigger the terminal to send a target request. It should be noted that before triggering the terminal to send the target request, the terminal first determines whether there is an available AI model, if there is an available AI model, then does not send the target request, and if there is no available AI model, then sends the target request. The absence of an available AI model can be understood as that there is no corresponding AI model, or there is a corresponding AI model, but the AI model is in an invalid state.
  • the information of the AI model includes at least one of the following: the identification of the AI model, the output parameters of the AI model, the structural information of the AI model, the model parameter information of the AI model, and the information of the AI model data. Processing information.
  • the above AI model identifier is used to uniquely identify a single AI model in the terminal.
  • the above structure information of the AI model may include a specific type and a specific structure of the AI model.
  • the specific type can refer to Gaussian process, support vector machine and various neural network methods, etc.
  • the specific structure can refer to the number of layers of the neural network, the number of neurons in each layer, and activation functions.
  • model parameter information of the AI model can be understood as the hyperparameter configuration of the AI model.
  • the processing method information can be understood as the preprocessing of the data before it is input to the AI model, including but not limited to: normalization, upsampling, downsampling, etc.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, AI model Default values corresponding to the input parameters of the model.
  • the update condition of the above AI model is used to guide the terminal to send the target request to update the AI model.
  • the update condition may be the above-mentioned second condition.
  • the terminal may update the second condition associated with the locally stored AI model.
  • the validity period of the aforementioned AI model can be understood as the effective time of the AI model.
  • the AI model is in the valid state or activated state, and the AI model can be used to perform related operations, such as making predictions or generating reported information.
  • a timer may be set, and the initial value of the timer is the effective duration of the AI model.
  • the effective start time of the AI model can be set according to actual needs. For example, after the terminal receives the above AI model information, it can start the timer, or it can start the timer after a preset period of time after the time unit where the AI model information is received, or the network side device can Start the timer at the indicated start time.
  • the manner in which the network-side device indicates the validity period of the AI model can be set according to actual needs.
  • the network-side device can indicate at least one of the following items to indicate the valid period of the AI model: the initial value and timing of the timer The start time of the device.
  • the aforementioned effective area of the AI model can be understood as the effective area of the AI model.
  • the AI model When the terminal is in the valid area, the AI model is in the valid state or the activated state, and the AI model can be used to perform related operations.
  • the AI model When the terminal is outside the effective area, the AI model is invalid.
  • each input parameter of the AI model is associated with a defaultable flag, and if the defaultable flag indicates that the input parameter can be defaulted, then if the input parameter cannot be obtained, the parameter may not be input.
  • Indicating the input parameters of the AI model can be understood as indicating the association relationship between the input parameters of the AI model and the default identifiers.
  • the input parameters indicated in the information of the AI model are input parameters that cannot be defaulted or output parameters that can be defaulted, or the information of the AI model directly indicates the association relationship between each input parameter of the AI model and the identifier that can be defaulted.
  • the default value may be used as the input.
  • the target request carries at least one of the following: terminal status information, wireless signal measurement results, AI capability information, and preference information for AI models.
  • the terminal status information includes at least one of the following:
  • the above location information may be a Global Positioning System (Global Positioning System, GPS) measurement result;
  • the above movement information may include movement direction and movement speed, and the above service information may include ongoing low-latency services.
  • GPS Global Positioning System
  • the wireless signal measurement results include at least one of the following:
  • the RSRQ of the reference signal of the network side device is the RSRQ of the reference signal of the network side device.
  • the RSRP of the reference signal of the network side device may include at least one of the RSRP of the serving cell reference signal and the RSRP of the neighbor cell reference signal.
  • the RSRQ of the reference signal of the network side device may include at least one of the RSRQ of the serving cell reference signal and the RSRQ of the neighbor cell reference signal.
  • the AI capability information includes at least one of the following:
  • Supported AI model input parameters such as whether a certain input parameter is supported.
  • the above preference information may be understood as terminal expectation information, or suggestion information.
  • the preference information includes at least one of a preference for the AI model structure and a preference for the AI model data processing method.
  • FIG. 4 is a flow chart of a model request processing method provided in the embodiment of the present application. As shown in FIG. 4, it includes the following steps:
  • Step 401 the network side device receives the target request sent by the terminal
  • Step 402 the network side device obtains the information of the artificial intelligence AI model according to the target request information
  • Step 403 the network side device sends the information of the AI model to the terminal.
  • the target request carries at least one of the following: terminal status information, wireless signal measurement results, AI capability information, and preference information for AI models.
  • the information of the AI model includes at least one of the following: AI model identification, AI model output parameters, AI model structure information, AI model model parameter information, and AI model data processing mode information.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, corresponding Defaults.
  • the terminal status information includes at least one of the following:
  • the wireless signal measurement results include at least one of the following:
  • the RSRQ of the reference signal of the network side device is the RSRQ of the reference signal of the network side device.
  • the AI capability information includes at least one of the following:
  • the preference information includes at least one of a preference for AI model structure and a preference for AI model data processing manner.
  • the information of the AI model includes at least one of the following: AI model identification, AI model output parameters, AI model structure information, AI model model parameter information, and AI model data processing mode information.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, corresponding Defaults.
  • this embodiment is an implementation of the network-side device corresponding to the embodiment shown in FIG. Avoid repeating descriptions, and will not repeat them here.
  • the execution subject may be a model request device, or a control module in the model request device for executing the model request method.
  • the model requesting device provided in the embodiment of the present application is described by taking the model requesting device executing the model requesting method as an example.
  • FIG. 5 is a structural diagram of a model requesting device provided by an embodiment of the present application. As shown in FIG. 5, the model requesting device 500 includes:
  • the first sending module 501 is configured to send a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model;
  • the first receiving module 502 is configured to receive the information of the AI model sent by the network side device;
  • the processing module 503 is configured to obtain the AI model according to the information of the AI model.
  • the first sending module 501 is specifically configured to: when the first condition is satisfied, the terminal sends the target request to the network side device; wherein,
  • the first condition includes at least one of the following:
  • the location of the terminal satisfies a first preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured first measurement result range
  • the first transmission parameter of the terminal satisfies a second preset condition
  • the terminal has a first requirement
  • the terminal does not have the AI model.
  • the first preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the first cell identity range corresponding to the pre-configured cell identity
  • the tracking area TA where the terminal is located conforms to the first tracking area identification range corresponding to the pre-configured TA identification;
  • the access network notification area where the terminal is located RNA conforms to the range of the first area corresponding to the pre-configured RNA;
  • the public land mobile communication network PLMN where the terminal is located conforms to the range of the first communication network corresponding to the pre-configured PLMN;
  • the geographic location where the terminal is located conforms to the preconfigured range of the first geographic location.
  • the second preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured first parameter range
  • phase information between different transmitting antennas or between different transmitting ports conforms to preconfigured first phase information, or, the preset characterization of the phase information between different transmitting antennas or between different transmitting ports conforms to the first phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter conforms to a preset configured second parameter range
  • the information corresponding to different receiving beams conforms to the preconfigured first beam information, or, the specific representation of the information corresponding to different receiving beams conforms to the first beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured second beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the second beam information;
  • the measurement parameter obtained on at least one frequency conforms to a pre-configured third parameter range
  • the data demodulation soft information conforms to preconfigured first preset information, or, the specific representation of the data demodulation soft information conforms to the first preset information;
  • the data packet transmission bit error rate conforms to preconfigured second preset information, or, the specific characterization of the data packet transmission bit error rate conforms to the second preset information;
  • the block error rate conforms to preconfigured third preset information, or the specific characterization of the block error rate conforms to the third preset information;
  • the interference measurement information conforms to preconfigured fourth preset information, or the specific representation of the interference measurement information conforms to the fourth preset information.
  • the first sending module 501 is specifically configured to: when the second condition is met, the terminal sends the target request to the network side device; wherein the second condition is the AI model's Update conditions;
  • the processing module 503 is specifically configured to: the terminal updates the original AI model according to the information of the AI model to obtain a target AI model.
  • the second condition includes at least one of the following:
  • the location of the terminal satisfies the third preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured second measurement result range
  • the first transmission parameter of the terminal satisfies a fourth preset condition
  • the terminal has a first requirement
  • the confidence degree of the reasoning result of the AI model reaches a threshold
  • the terminal moves outside the effective area of the AI model
  • the terminal has no valid AI model.
  • the third preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the second cell identity range corresponding to the pre-configured cell identity
  • the TA where the terminal is located conforms to the second tracking area identification range corresponding to the pre-configured TA identification
  • the RNA where the terminal is located conforms to the range of the second region corresponding to the pre-configured RNA
  • the PLMN where the terminal is located conforms to the range of the second communication network corresponding to the pre-configured PLMN;
  • the geographical location of the terminal conforms to the preconfigured range of the second geographical location.
  • the fourth preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured fourth parameter range
  • phase information between different transmitting antennas or between different transmitting ports conforms to preconfigured second phase information, or, the preset characterization of the phase information between different transmitting antennas or between different transmitting ports conforms to the second phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter complies with a fifth parameter range of the preset configuration
  • the information corresponding to different receiving beams conforms to the preconfigured third beam information, or, the specific representation of the information corresponding to different receiving beams conforms to the third beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured fourth beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the fourth beam information;
  • the measured parameter obtained on at least one frequency conforms to a pre-configured sixth parameter range
  • the data demodulation soft information conforms to pre-configured fifth preset information, or, the specific representation of the data demodulation soft information conforms to the fifth preset information;
  • the packet transmission bit error rate complies with pre-configured sixth preset information, or, the specific characterization of the data packet transmission bit error rate complies with the sixth preset information;
  • the block error rate conforms to pre-configured seventh preset information, or, the specific characterization of the block error rate conforms to the seventh preset information;
  • the interference measurement information conforms to preconfigured eighth preset information, or the specific representation of the interference measurement information conforms to the eighth preset information.
  • the first measurement quantity includes at least one of the following: signal-to-noise ratio, reference signal received power RSRP, signal-to-interference-plus-noise ratio SINR, reference signal received quality RSRQ, packet delay, round-trip delay RTT, observation Time difference of arrival OTDOA, measurement results corresponding to channel state information, and measurement results corresponding to user quality of experience.
  • the measurement resource corresponding to the first measurement quantity includes at least one of the following:
  • Synchronization signal block SSB Synchronization signal block
  • the first transmission parameter includes at least one of the following:
  • the first requirement includes at least one of the following: reporting a radio resource management RRM measurement report; handover decision; redirection decision; generating positioning results; reporting channel state information CSI; user trajectory prediction; user service demand prediction; user Slice Demand Forecast.
  • the effective area is determined based on at least one of the following: a cell list, a TA list, an RNA list, a PLMN list, and a geographic location.
  • the information of the AI model includes at least one of the following: AI model identification, AI model output parameters, AI model structure information, AI model model parameter information, and AI model data processing mode information.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, corresponding Defaults.
  • the target request carries at least one of the following: terminal status information, wireless signal measurement results, AI capability information, and preference information for AI models.
  • the terminal status information includes at least one of the following:
  • the wireless signal measurement results include at least one of the following:
  • the RSRQ of the reference signal of the network side device is the RSRQ of the reference signal of the network side device.
  • the AI capability information includes at least one of the following:
  • the preference information includes at least one of a preference for AI model structure and a preference for AI model data processing manner.
  • the model requesting device provided in the embodiment of the present application can implement each process in the method embodiment in FIG. 3 , and details are not repeated here to avoid repetition.
  • FIG. 6 is a structural diagram of a model request processing device provided in the embodiment of the present application. As shown in FIG. 6, the model request processing device 600 includes:
  • the second receiving module 601 is configured to receive the target request sent by the terminal
  • Generating module 602 is used for obtaining the information of artificial intelligence AI model according to target request information
  • the second sending module 603 is configured to send the information of the AI model to the terminal.
  • the target request carries at least one of the following: terminal status information, wireless signal measurement results, AI capability information, and preference information for AI models.
  • the information of the AI model includes at least one of the following: AI model identification, AI model output parameters, AI model structure information, AI model model parameter information, and AI model data processing mode information.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, corresponding Defaults.
  • the terminal status information includes at least one of the following:
  • the wireless signal measurement results include at least one of the following:
  • the RSRQ of the reference signal of the network side device is the RSRQ of the reference signal of the network side device.
  • the AI capability information includes at least one of the following:
  • the preference information includes at least one of a preference for AI model structure and a preference for AI model data processing manner.
  • the model request processing device provided in the embodiment of the present application can implement each process in the method embodiment in FIG. 3 , and details are not repeated here to avoid repetition.
  • the model requesting device and the model request processing device in the embodiment of the present application may be a device, a device with an operating system or an electronic device, or a component, an integrated circuit, or a chip in a terminal.
  • the device may be a mobile terminal or a non-mobile terminal.
  • the mobile terminal may include but not limited to the types of terminals 11 listed above, and the non-mobile terminal may be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television ( television, TV), teller machines or self-service machines, etc., are not specifically limited in this embodiment of the present application.
  • the embodiment of the present application further provides a communication device 700, including a processor 701, a memory 702, and programs or instructions stored in the memory 702 and operable on the processor 701,
  • a communication device 700 including a processor 701, a memory 702, and programs or instructions stored in the memory 702 and operable on the processor 701
  • the communication device 700 is a terminal
  • the program or instruction is executed by the processor 701
  • each process of the above-mentioned model request method embodiment can be realized, and the same technical effect can be achieved.
  • the communication device 700 is a network-side device
  • the program or instruction is executed by the processor 701
  • each process of the above-mentioned model request processing method embodiment can be achieved, and the same technical effect can be achieved. To avoid repetition, details are not repeated here.
  • the embodiment of the present application also provides a terminal, including a processor and a communication interface, the communication interface is used to send a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model; receive the network side device The information of the AI model sent; the processor, configured to obtain the AI model according to the information of the AI model.
  • This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment, and each implementation process and implementation mode of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
  • FIG. 8 is a schematic diagram of a hardware structure of a terminal implementing various embodiments of the present application.
  • the terminal 800 includes but not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809, and a processor 810, etc. At least some parts.
  • the terminal 800 may also include a power supply (such as a battery) for supplying power to various components, and the power supply may be logically connected to the processor 810 through the power management system, so as to manage charging, discharging, and power consumption through the power management system. Management and other functions.
  • the terminal structure shown in FIG. 8 does not constitute a limitation on the terminal, and the terminal may include more or fewer components than shown in the figure, or combine some components, or arrange different components, which will not be repeated here.
  • the input unit 804 may include a graphics processor (Graphics Processing Unit, GPU) and a microphone, and the graphics processor is controlled by an image capture device (such as a camera) in a video capture mode or an image capture mode.
  • the obtained image data of still picture or video is processed.
  • the display unit 806 may include a display panel, and the display panel may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 807 includes a touch panel and other input devices. Touch panel, also known as touch screen.
  • the touch panel can include two parts: a touch detection device and a touch controller.
  • Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be repeated here.
  • the radio frequency unit 801 receives the downlink data from the network side device, and then processes it to the processor 810; in addition, sends the uplink data to the network side device.
  • the radio frequency unit 801 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like.
  • the memory 809 can be used to store software programs or instructions as well as various data.
  • the memory 809 may mainly include a program or instruction storage area and a data storage area, wherein the program or instruction storage area may store an operating system, an application program or instructions required by at least one function (such as a sound playback function, an image playback function, etc.) and the like.
  • the memory 809 may include a high-speed random access memory, and may also include a non-transitory memory, wherein the non-transitory memory may be a read-only memory (Read-Only Memory, ROM), a 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
  • PROM erasable programmable read-only memory
  • Erasable PROM Erasable PROM
  • EPROM electrically erasable programmable read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • flash memory for example at least one disk storage device, flash memory device, or other non-transitory solid state storage device.
  • the processor 810 may include one or more processing units; optionally, the processor 810 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, application programs or instructions, etc., Modem processors mainly handle wireless communications, such as baseband processors. It can be understood that the foregoing modem processor may not be integrated into the processor 810 .
  • the radio frequency unit 801 is configured to send a target request to the network side device, and the target request is used to request the information of the artificial intelligence AI model; receive the information of the AI model sent by the network side device;
  • the processor 810 is configured to obtain the AI model according to the information of the AI model.
  • the radio frequency unit 801 is specifically configured to: send the target request to the network side device when the first condition is satisfied; wherein,
  • the first condition includes at least one of the following:
  • the location of the terminal satisfies a first preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured first measurement result range
  • the first transmission parameter of the terminal satisfies a second preset condition
  • the terminal has a first requirement
  • the terminal does not have the AI model.
  • the first preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the first cell identity range corresponding to the pre-configured cell identity
  • the tracking area TA where the terminal is located conforms to the first tracking area identification range corresponding to the pre-configured TA identification;
  • the access network notification area where the terminal is located RNA conforms to the range of the first area corresponding to the pre-configured RNA;
  • the public land mobile communication network PLMN where the terminal is located conforms to the range of the first communication network corresponding to the pre-configured PLMN;
  • the geographic location where the terminal is located conforms to the preconfigured range of the first geographic location.
  • the second preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured first parameter range
  • phase information between different transmitting antennas or between different transmitting ports conforms to preconfigured first phase information, or, the preset characterization of the phase information between different transmitting antennas or between different transmitting ports conforms to the first phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter conforms to a preset configured second parameter range
  • the information corresponding to different receiving beams conforms to the preconfigured first beam information, or the specific representation of the information corresponding to different receiving beams conforms to the first beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured second beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the second beam information;
  • the measurement parameter obtained on at least one frequency conforms to a pre-configured third parameter range
  • the data demodulation soft information conforms to preconfigured first preset information, or, the specific representation of the data demodulation soft information conforms to the first preset information;
  • the data packet transmission bit error rate conforms to preconfigured second preset information, or, the specific characterization of the data packet transmission bit error rate conforms to the second preset information;
  • the block error rate conforms to preconfigured third preset information, or the specific characterization of the block error rate conforms to the third preset information;
  • the interference measurement information conforms to preconfigured fourth preset information, or the specific representation of the interference measurement information conforms to the fourth preset information.
  • the radio frequency unit 801 is specifically configured to: send the target request to the network side device when the second condition is satisfied; wherein the second condition is an update condition of the AI model;
  • the processor 810 is specifically configured to: update the original AI model according to the information of the AI model to obtain a target AI model.
  • the second condition includes at least one of the following:
  • the location of the terminal satisfies the third preset condition
  • the first measurement quantity of the terminal conforms to a pre-configured second measurement result range
  • the first transmission parameter of the terminal satisfies a fourth preset condition
  • the terminal has a first requirement
  • the confidence degree of the reasoning result of the AI model reaches a threshold
  • the terminal moves outside the effective area of the AI model
  • the terminal has no valid AI model.
  • the third preset condition includes at least one of the following:
  • the serving cell where the terminal is located conforms to the second cell identity range corresponding to the pre-configured cell identity
  • the TA where the terminal is located conforms to the second tracking area identification range corresponding to the pre-configured TA identification
  • the RNA where the terminal is located conforms to the range of the second region corresponding to the pre-configured RNA
  • the PLMN where the terminal is located conforms to the range of the second communication network corresponding to the pre-configured PLMN;
  • the geographical location of the terminal conforms to the preconfigured range of the second geographical location.
  • the fourth preset condition includes at least one of the following:
  • the channel characteristic parameter or the preset characterization of the channel parameter conforms to the pre-configured fourth parameter range
  • phase information between different transmit antennas or between different transmit ports conforms to preconfigured second phase information, or, the preset characterization of the phase information between different transmit antennas or between different transmit ports conforms to the second phase information ;
  • the beam measurement parameter or the specific characterization of the beam measurement parameter complies with a fifth parameter range of the preset configuration
  • the information corresponding to different receiving beams conforms to the preconfigured third beam information, or, the specific representation of the information corresponding to different receiving beams conforms to the third beam information;
  • the information corresponding to different beam pairs conforms to the preconfigured fourth beam information, or, the specific representation of the information corresponding to different beam pairs conforms to the fourth beam information;
  • the measured parameter obtained on at least one frequency conforms to a pre-configured sixth parameter range
  • the data demodulation soft information conforms to pre-configured fifth preset information, or, the specific representation of the data demodulation soft information conforms to the fifth preset information;
  • the data packet transmission bit error rate complies with pre-configured sixth preset information, or, the specific characterization of the data packet transmission bit error rate complies with the sixth preset information;
  • the block error rate conforms to pre-configured seventh preset information, or, the specific characterization of the block error rate conforms to the seventh preset information;
  • the interference measurement information conforms to preconfigured eighth preset information, or the specific representation of the interference measurement information conforms to the eighth preset information.
  • the first measurement quantity includes at least one of the following: signal-to-noise ratio, reference signal received power RSRP, signal-to-interference-plus-noise ratio SINR, reference signal received quality RSRQ, packet delay, round-trip delay RTT, observation Time difference of arrival OTDOA, measurement results corresponding to channel state information, and measurement results corresponding to user quality of experience.
  • the measurement resource corresponding to the first measurement quantity includes at least one of the following:
  • Synchronization signal block SSB Synchronization signal block
  • the first transmission parameter includes at least one of the following:
  • the first requirement includes at least one of the following: reporting a radio resource management RRM measurement report; handover decision; redirection decision; generating positioning results; reporting channel state information CSI; user trajectory prediction; user service demand prediction; user Slice Demand Forecast.
  • the effective area is determined based on at least one of the following: a cell list, a TA list, an RNA list, a PLMN list, and a geographic location.
  • the information of the AI model includes at least one of the following: AI model identification, AI model output parameters, AI model structure information, AI model model parameter information, and AI model data processing mode information.
  • the information of the AI model further includes at least one of the following: update conditions of the AI model, validity period of the AI model, valid area of the AI model, input parameters of the AI model, corresponding Defaults.
  • the target request carries at least one of the following: terminal status information, wireless signal measurement results, AI capability information, and preference information for AI models.
  • the terminal state information includes at least one of the following: location information of the terminal; mobile information of the terminal; service information of the terminal.
  • the wireless signal measurement results include at least one of the following:
  • the RSRQ of the reference signal of the network side device is the RSRQ of the reference signal of the network side device.
  • the AI capability information includes at least one of the following:
  • the preference information includes at least one of a preference for AI model structure and a preference for AI model data processing manner.
  • the embodiment of the present application also provides a network side device, including a processor and a communication interface, the communication interface is used to receive the target request sent by the terminal; the processor is used to obtain the artificial intelligence AI model according to the target request information Information; the communication interface is further configured to send the information of the AI model to the terminal.
  • the network-side device embodiment corresponds to the above-mentioned network-side device method embodiment, and each implementation process and implementation mode of the above-mentioned method embodiment can be applied to this network-side device embodiment, and can achieve the same technical effect.
  • the embodiment of the present application also provides a network side device.
  • the network side device 900 includes: an antenna 901 , a radio frequency device 902 , and a baseband device 903 .
  • the antenna 901 is connected to the radio frequency device 902 .
  • the radio frequency device 902 receives information through the antenna 901, and sends the received information to the baseband device 903 for processing.
  • the baseband device 903 processes the information to be sent and sends it to the radio frequency device 902
  • the radio frequency device 902 processes the received information and sends it out through the antenna 901 .
  • the foregoing frequency band processing device may be located in the baseband device 903 , and the method performed by the network side device in the above embodiments may be implemented in the baseband device 903 , and the baseband device 903 includes a processor 904 and a memory 905 .
  • the baseband device 903 may include at least one baseband board, for example, a plurality of chips are arranged on the baseband board, as shown in FIG. The operation of the network side device shown in the above method embodiments.
  • the baseband device 903 may also include a network interface 906, configured to exchange information with the radio frequency device 902, such as a common public radio interface (common public radio interface, CPRI).
  • a network interface 906 configured to exchange information with the radio frequency device 902, such as a common public radio interface (common public radio interface, CPRI).
  • CPRI common public radio interface
  • the network-side device in the embodiment of the present application further includes: instructions or programs stored in the memory 905 and executable on the processor 904, and the processor 904 calls the instructions or programs in the memory 905 to execute the various operations shown in FIG.
  • the method of module execution achieves the same technical effect, so in order to avoid repetition, it is not repeated here.
  • the embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by the processor, each process of the above-mentioned embodiment of the model request method or the model request processing method is realized , and can achieve the same technical effect, in order to avoid repetition, it will not be repeated here.
  • the processor is the processor in the electronic device described in the above embodiments.
  • the readable storage medium includes computer readable storage medium, such as computer read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
  • the 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, and the processor is used to run programs or instructions to implement the above-mentioned model request method or model request.
  • chips mentioned in the embodiments of the present application may also be called system-on-chip, system-on-chip, system-on-a-chip, or system-on-a-chip.
  • the embodiment of the present application further provides a program product, the program product is stored in a non-transitory storage medium, and the program product is executed by at least one processor to implement the above-mentioned model request method or model request processing method embodiment
  • Each process can achieve the same technical effect, so in order to avoid repetition, it will not be repeated here.
  • the term “comprising”, “comprising” or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase “comprising a " does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element.
  • the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions are performed, for example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
  • the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also 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 computer software products, which are stored in a storage medium (such as ROM/RAM, magnetic disk, etc.) , optical disc), including several instructions to enable a terminal (which may be a mobile phone, computer, server, air conditioner, or base station, etc.) to execute the methods described in various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Pure & Applied Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种模型请求方法、模型请求处理方法及相关设备，属于通信技术领域。本申请实施例的模型请求方法包括：终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；所述终端接收网络侧设备发送的所述AI模型的信息；所述终端根据所述AI模型的信息得到所述AI模型。

Description

模型请求方法、模型请求处理方法及相关设备
相关申请的交叉引用
本申请主张在2021年10月21日在中国提交的中国专利申请No.202111229121.X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，尤其涉及一种模型请求方法、模型请求处理方法及相关设备。
背景技术
随着通信技术的发展，在通信系统中应用了人工智能(Artificial Intelligence，AI)模块对信息进行处理和分析。为了适用于不同的通信环境，兼容不同的通信场景，AI模型训练过程需要大量的训练数据，从而使得训练后的网络模型较大；若AI模型训练过程仅采用特定场景下的数据进行训练，训练后的模型较小，但这样将容易导致网络模型泛化能力不足，出现过拟合的问题。
发明内容
本申请实施例提供一种模型请求方法、模型请求处理方法及相关设备，能够基于终端设备的请求，反馈响应的AI模型，从而在控制AI模型大小的同时，减少AI模型泛化能力不足带来的影响。
第一方面，提供了一种模型请求方法，包括：
终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
所述终端接收网络侧设备发送的所述AI模型的信息；
所述终端根据所述AI模型的信息得到所述AI模型。
第二方面，提供了一种模型请求处理方法，包括：
网络侧设备接收终端发送的目标请求；
所述网络侧设备根据目标请求信息获得AI模型的信息；
所述网络侧设备向所述终端发送所述AI模型的信息。
第三方面，提供了一种模型请求装置，包括：
第一发送模块，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
第一接收模块，用于接收网络侧设备发送的所述AI模型的信息；
处理模块，用于根据所述AI模型的信息得到所述AI模型。
第四方面，提供了一种模型请求处理装置，包括：
第二接收模块，用于接收终端发送的目标请求；
生成模块，用于根据目标请求信息获得AI模型的信息；
第二发送模块，用于向所述终端发送所述AI模型的信息。
第五方面，提供了一种终端，该终端包括处理器、存储器及存储在所述存储器上并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，提供了一种终端，包括处理器及通信接口，其中，
所述通信接口，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；接收网络侧设备发送的所述AI模型的信息；
所述处理器，用于根据所述AI模型的信息得到所述AI模型。
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、存储器及存储在所述存储器上并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二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，
所述通信接口，用于接收终端发送的目标请求；
所述处理器，用于根据目标请求信息获得人工智能AI模型的信息；
所述通信接口，还用于向所述终端发送所述AI模型的信息。
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，或者实现如第二方面所述的方法的步骤。
第十方面，本申请实施例提供了一种芯片，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，或者实现如第二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产品，所述计算机程序产品存储在非瞬态的存储介质中，所述计算机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，或实现如第二方面所述的方法。
第十二方面，提供了一种通信设备，被配置为执行如第一方面所述的方法，或执行如第二方面所述的方法。
本申请实施例通过终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；所述终端接收网络侧设备发送的所述AI模型的信息；所述终端根据所述AI模型的信息得到所述AI模型。这样可以基于不同的通信场景设置不同的AI模型，由终端通过请求的方式动态获取AI模型的信息，从而可以在控制AI模型大小的同时，减少AI模型泛化能力不足带来的影响，进而可以提升传输的可靠性。
附图说明
图1是本申请实施例可应用的一种网络系统的结构图；
图2是本申请实施例可应用的神经元的结构图；
图3是本申请实施例提供的一种模型请求方法的流程图；
图4是本申请实施例提供的一种模型请求处理方法的流程图；
图5是本申请实施例提供的一种模型请求装置的结构图；
图6是本申请实施例提供的一种模型请求处理装置的结构图；
图7是本申请实施例提供的一种通信设备的结构图；
图8是本申请实施例提供的一种终端的结构图；
图9是本申请实施例提供的一种网络侧设备的结构图。
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，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(Vehicle User Equipment，VUE)、行人终端(Pedestrian User Equipment，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)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、游戏机等。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可以是基站或核心网设备，其中，基站可被称为节点B、演进节点B、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)、B节点、演进型B节点(eNB)、 家用B节点、家用演进型B节点、无线局域网(Wireless Local Area Network，WLAN)接入点、WiFi节点、发送接收点(Transmitting Receiving Point，TRP)或所述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，但是并不限定基站的具体类型。
为了方便理解，以下对本申请实施例涉及的一些内容进行说明：
一、人工智能
人工智能目前在各个领域获得了广泛的应用。AI模块有多种实现方式，例如神经网络、决策树、支持向量机和贝叶斯分类器等。本申请以神经网络为例进行说明，但是并不限定AI模块的具体类型。
可选地，神经网络由神经元组成，神经元的结构如图2所示，其中a 1，a 2，…a K为输入，w为权值，即乘性系数，b为偏置，即加性系数，σ(.)为激活函数。常见的激活函数包括Sigmoid、tanh、线性整流函数(Rectified Linear Unit，ReLU)等。z＝a 1w 1+···+a kw k+···+a Kw K+b。
神经网络的参数通过优化算法进行优化。优化算法就是一种能够帮我们最小化或者最大化目标函数的一类算法，目标函数也可以称之为损失函数。而目标函数往往是模型参数和数据的数学组合。例如给定数据X和其对应的标签Y，我们构建一个神经网络模型f(.)，有了模型后，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)，这个就是损失函数。我们的目的是找到合适的W和b使上述的损失函数的值达到最小，损失值越小，则说明我们的模型越接近于真实情况。
目前常见的优化算法，基本都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。
常见的优化算法包括：梯度下降(Gradient Descent)、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)和自适应动量估计(Adaptive Moment Estimation，Adam)等。
这些优化算法在误差反向传播时，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，得到梯度，将梯度传给上一层。
二、AI模型复杂度、泛化能力及过拟合问题
一般而言，根据解决类型不同，选取的AI算法和采用的模型也有所差别。根据目前发表文章及公开研究成果，借助AI提升5G网络性能的主要方法是通过基于神经网络的算法和模型增强或者替代目前已有的算法或处理模块。在特定场景下，基于神经网络的算法和模型可以取得比基于确定性算法更好的性能。比较常用的神经网络包括深度神经网络、卷积神经网络和循环神经网络等。借助已有AI工具，可以实现神经网络的搭建、训练与验证工作。
神经网络的设计核心是结合需要解决的问题和数据，完成神经网络的搭建。神经网络搭建比较关键的步骤包括多层网络的堆叠方式选择、每层网络的神经元个数与深度选择、激活函数选择等。为了取得更好的性能，往往需要增加网络的深度和神经元个数。随着网络深度和神经元个数增加，需要训练的超参个数也快速提升，神经网络训练消耗计算资源和网络训练难度也大幅提升。同时，随着网络深度增加，梯度消失和梯度爆炸问题也将更加突出，需要在设计激活函数时进行充分考虑。
大的神经网络不仅有训练难度大的问题，还容易产生过拟合的问题。过拟合的表现主要是在训练用数据集表现好，而在非训练用数据集表现不好。训练数据集表现好，而非训练数据集表现不好的模型往往被认为模型的泛化能力不好。泛化能力是5G引入基于AI的算法和模型的核心考虑因素。为提升模型的泛化能力，防止过拟合现象的发生，可以利用已有的一些模型工具，或者牺牲一定的性能采用较小的网络。
算法及模型性能与数据集的关系密不可分。如果训练数据集与测试数据集存在较大差异，那么依据训练集得到的模型很难在测试数据集上取得很好的性能。一种比较理想的模式是训练数据集、测试数据集和实际处理数据具有非常好的一致性，在此条件下训练出的模型可以有比较好的效果。而在实际的数据集构建过程中，受各种实际条件限制，训练与测试用数据集与实际场景中处理的数据并不能保证很好的匹配。这时不仅需要考虑不断丰富数据，还要结合有限数据条件下算法与模型特点，结合实际情况来更加合理的构建数据集，形成数据集构建和算法与模型探索的良性互动，提升算法与模型性能同时构建经典的数据集。
模型的训练和更新是另一个需要关注的问题。模型的训练可以在网络设 备、云端或者边缘设备中进行。在进行模型训练时，数据需要传送到进行训练的网元或者云端。训练完的模型再更新到对应的模块中。数据的收集和传输不仅涉及到数据的隐私性问题，还会对网络传输带来一定的需求。如果模型使用和模型训练不在同一网元，还需要进行模型传输，频繁的模型传输也需要消耗比较多的网络资源。模型更新消耗的算力、需要网络资源和时效性要求也是在进行算法与模型设计时需要考虑的重要因素。
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模型请求方法进行详细地说明。
请参见图3，图3是本申请实施例提供的一种模型请求方法的流程图，如图3所示，包括以下步骤：
步骤301，终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
本申请实施例中，上述目标请求可以用于获取终端侧未被配置的AI模型的信息，也可以用于更新终端侧的AI模型。例如，当终端不存在相应的AI模型时，可以向网络侧设备发送目标请求以获取AI模型的信息，从而配置相应的AI模型，当终端预先配置了AI模型时，可以向网络侧设备发送目标请求以更新终端侧的AI模型，从而匹配当前的通信场景。
步骤302，所述终端接收网络侧设备发送的所述AI模型的信息；
步骤303，所述终端根据所述AI模型的信息得到所述AI模型。
本申请实施例中，网络侧设备可以通过无线资源控制(Radio Resource Control，RRC)配置消息发送AI模型的信息，也可以通过RRC重配置(RRCReconfiguration)消息发送AI模型的信息，还可以通过系统信息块(System Information Block，SIB)广播AI模型的信息。
终端接收到该AI模型的信息后，可以基于该AI模型的信息可以配置或者更新终端侧的AI模型，这样，可以保证终端和网络侧设备对AI模型的理解一致，从而提高传输的可靠性。
本申请实施例通过终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；所述终端接收网络侧设备发送的所述AI模型的信息；所述终端根据所述AI模型的信息得到所述AI模型。这样可以基于不同的通信场景设置不同的AI模型，由终端通过请求的方式动态获取AI模型的信息，从而可以在控制AI模型大小的同时，减少AI模型泛化能力不足带来的影响，进而可以提升传输的可靠性。
可选地，在一些实施例中，所述终端向网络侧设备发送目标请求包括：
在第一条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，
所述第一条件包括以下至少一项：
所述终端所在位置满足第一预设条件；
所述终端的第一测量量符合预先配置的第一测量结果范围；
所述终端的第一传输参数满足第二预设条件；
所述终端存在第一需求；
所述终端无所述AI模型。
本申请实施例中，上述第一条件可以由协议约定，也可以由网络侧设备配置。应理解，本申请实施例中，在满足第一条件的请求下，表示终端不存在相应的AI模型，即网络侧设备未提前配置相应的AI模型的信息。通过上述目标请求向网络侧请求为终端配置AI模型的信息。终端接收到该AI模型的信息后，可以基于所述AI模型的信息配置AI模型或者生成AI模型，以通过该AI模型执行相应的终端行为。例如利用该AI模型用于移动性管理和定位等终端行为。
可选地，所述第一预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第一小区标识范围；
终端所在的跟踪区(Tracking Area，TA)符合预先配置的TA标识对应的第一跟踪区标识范围；
终端所在的接入网通知区域(RAN-based Notification Area，RNA)符合预先配置的RNA对应的第一区域范围；
终端所在的公共陆地移动通信网(Public Land Mobile Network，PLMN)符合预先配置的PLMN对应的第一通信网范围；
终端所在的地理位置符合预先配置的第一地理位置范围。
本申请实施例中，上述第一小区标识范围可以理解为预先配置的小区标识列表，当终端所在的服务小区的小区标识属于该小区标识列表时，可以确定终端所在的服务小区符合预先配置的小区标识对应的第一小区标识范围。上述第一跟踪区标识范围可以理解为预先配置的跟踪区列表，当终端所在的TA对应的TA标识属于该第一跟踪区列表时，可以确定终端所在的TA符合预先配置的TA标识对应的第一跟踪区标识范围。上述第一区域范围可以理解为RNA列表，当终端所在的RNA属于该RNA列表，则可以确定终端所在的RNA符合预先配置的RNA对应的第一区域范围。
可选地，在一些实施中，所述第二预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第一参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第一相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第一相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第二参数范围；
不同接收波束对应的信息符合预先配置的第一波束信息，或者，不同接收波束对应的信息的特定表征符合所述第一波束信息；
不同波束对对应的信息符合预先配置的第二波束信息，或者，不同波束对对应的信息的特定表征符合所述第二波束信息；
至少一个频率上获得的测量参数符合预先配置的第三参数范围；
数据解调软信息符合预先配置的第一预设信息，或者，数据解调软信息的特定表征符合所述第一预设信息；
数据包传输误比特率符合预先配置的第二预设信息，或者，数据包传输误比特率的特定表征符合所述第二预设信息；
数据包传输误块率符合预先配置的第三预设信息，或者，数据包传输误块率的特定表征符合所述第三预设信息；
干扰测量信息符合预先配置的第四预设信息，或者，干扰测量信息的特定表征符合所述第四预设信息。
应理解，上述数据包可以包括以下至少一项，物理层数据，媒体接入控制(Medium Access Control，MAC)层数据包、无线链路控制(Radio Link Control，RLC)层数据包、分组数据汇聚协议(Packet Data Convergence Protocol，PDCP)层数据包和互联网协议(Internet Protocol，IP)层数据包。
可选地，所述终端向网络侧设备发送目标请求包括：
在第二条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，所述第二条件为所述AI模型的更新条件；
所述终端根据所述AI模型的信息得到所述AI模型包括：
所述终端根据所述AI模型的信息对原AI模型进行更新，得到目标AI模型。
本申请实施例中，终端配置了相应的AI模型，但该AI模型需要进行更新，例如在该AI模型处于失效状态，或者去激活状态时，可以通过目标请求对该AI模型进行更新。网络侧设备可以通过RRC重配置消息重新为终端配置该AI模型的信息，终端基于该AI模型的信息对原有的AI模型进行更新。
需要说明的是，终端可以按照差分方式在原有的AI模型的信息的基础上更新AI模型的信息，然后基于更新后的AI模型的信息进行AI模型更新，例如配置或者生成新的AI模型后，再替换原有的AI模型。应理解，上述AI模型的信息可以仅包括AI模型需要更新的信息，或者包括AI模型更新后的所有信息，在此不做进一步的限定。
可选地，在一些实施例中，所述第二条件包括以下至少一项：
所述终端所在位置满足第三预设条件；
所述终端的第一测量量符合预先配置的第二测量结果范围；
所述终端的第一传输参数满足第四预设条件；
所述终端存在第一需求；
所述AI模型的推理结果置信度达到阈值；
所述AI模型有效期限超时；
所述终端移动至所述AI模型的有效区域外；
所述终端无有效AI模型。
本申请实施例中，所述终端无有效AI模型可以理解为，所需要的AI模型处于无效状态，或者去激活状态。其中，在满足以下至少一项条件的情况下，可以确定AI模型处于无效状态：
AI模型的有效期限超时；
终端移动至AI模型的有效区域外。
应理解，在本申请实施例中，上述有效区域基于以下至少一项确定：小区列表、TA列表、RNA列表、PLMN列表和地理位置。
可选地，上述第三预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第二小区标识范围；
终端所在的TA符合预先配置的TA标识对应的第二跟踪区标识范围；
终端所在的RNA符合预先配置的RNA对应的第二区域范围；
终端所在的PLMN符合预先配置的PLMN对应的第二通信网范围；
终端所在的地理位置符合预先配置的第二地理位置范围。
本申请实施例中，上述第三预设条件与第一预设条件的区别在于各范围对应的取值不同，其中第一预设条件和第三预设条件中各范围对应的取值大小可以根据实际需要进行设置，在此不做进一步的限定。
可选地，在一些实施例中，所述第四预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第四参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第二相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第二相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第五参数范围；
不同接收波束对应的信息符合预先配置的第三波束信息，或者，不同接收波束对应的信息的特定表征符合所述第三波束信息；
不同波束对对应的信息符合预先配置的第四波束信息，或者，不同波束对对应的信息的特定表征符合所述第四波束信息；
至少一个频率上获得的测量参数符合预先配置的第六参数范围；
数据解调软信息符合预先配置的第五预设信息，或者，数据解调软信息 的特定表征符合所述第五预设信息；
数据包传输误比特率符合预先配置的第六预设信息，或者，数据包传输误比特率的特定表征符合所述第六预设信息；
误块率符合预先配置的第七预设信息，或者，误块率的特定表征符合所述第七预设信息；
干扰测量信息符合预先配置的第八预设信息，或者，干扰测量信息的特定表征符合所述第八预设信息。
本申请实施例中，上述第四预设条件与第二预设条件的区别在于各范围对应的取值不同，其中第二预设条件和第四预设条件中各范围对应的取值大小可以根据实际需要进行设置，在此不做进一步的限定。
可选地，在一些实施例中，所述第一测量量包括以下至少一项：信噪比(Signal Noise Ratio，SNR)、参考信号接收功率(Reference Signal Received Power，RSRP)、信号与干扰加噪声比(Signal-to-noise and Interference Ratio，SINR)、参考信号接收质量(Reference Signal Received Quality，RSRQ)、包时延(packet delay)、往返时延(Round-Trip Time，RTT)、观测到达时间差(Observed time difference of arrival，OTDOA)、信道状态信息(Channel State Information，CSI)对应的测量结果和用户体验质量(Quality of Experience，QoE)对应的测量结果。
需要说明的是，本申请实施例中，上述第一测量量包括服务小区对应的测量量和邻小区对应的测量量。
可选地，在一些实施例中，所述第一测量量对应的测量资源包括以下至少一项：
物理下行控制信道(Physical downlink control channel，PDCCH)解调参考信号(Demodulation Reference Signal，DMRS)；
物理下行共享信道(Physical downlink shared channel，PDSCH)DMRS；
信道状态信息参考信号(Channel State Information Reference Signal，CSI-RS)；
同步信号块(Synchronization Signal and PBCH block，SSB)；
定位参考信号(P ositioning Reference Signal，PRS)。
本申请实施例中，上述第一测量量对应的测量资源可以是预先配置的或者预先约定的，在此不做进一步的限定。
可选地，在一些实施例中，所述第一传输参数包括以下至少一项：
信道特征参数；
不同发送天线之间或不同发送端口之间的相位信息；
至少一个波束上获得的测量参数；
不同波束对对应的信息；
至少一个频率上获得的测量参数；
数据解调软信息；
数据包传输误比特率或误块率；
干扰测量信息。
应理解，上述波束对包括发送波束和接收波束。上述数据包括可以包括以下至少一项：物理层数据包，MAC层数据包，RLC层数据包，PDCP层数据包和IP层数据包。
可选地，在一些实施例中，上述第一需求包括以下至少一项：上报无线资源管理(Radio resource management，RRM)测量报告；切换判决；重定向判决；生成定位结果；上报信道状态信息(Channel State Information，CSI)；用户轨迹预测；用户业务需求预测；用户切片需求预测。
本申请实施例中，终端存在第一需求可以理解为，终端需要执行第一需求对应的操作，例如，在终端需要上报RRM测量报告，或者终端需要进行切换判决，或者终端需要进行重定向判决，或者在终端需要生产定位结果等情况下，可以确定终端存在对应的第一需求，以触发终端发送目标请求。需要说明的是，在触发终端发送目标请求前，终端首先确定是否存在可用的AI模型，若存在可用的AI模型，则不发送目标请求，若不存在可用的AI模型，则发送目标请求。不存在可用的AI模型可以理解为，无相应的AI模型，或者存在相应的AI模型，但AI模型处于失效状态。
可选地，在一些实施例中，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
本申请实施例中，上述AI模型的标识用于在终端中唯一标识单个AI模型。
可选地，上述AI模型的结构信息可以包括AI模型的具体类型以及具体结构。其中具体类型可以指高斯过程、支持向量机和各种神经网络方法等，具体结构可以指神经网络的层数、各层神经元个数和激活函数等。
可选地，AI模型的模型参数信息可以理解为AI模型的超参数配置。
可选地，处理方式信息可以理解为数据在输入到AI模型之前对数据的预处理，包括但不限于：归一化，上采样，降采样等。
可选地，在一些实施例中，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
可选地，上述AI模型的更新条件，用于指导终端发送所述目标请求，以 更新AI模型。例如，该更新条件可以为上述第二条件。具体地，终端在接收到更新条件后可以对本地存储的AI模型关联的第二条件进行更新。
可选地，上述AI模型的有效期限可以理解为AI模型的生效时间。在AI模型的有效期限内，该AI模型为有效状态或者激活状态，可以使用该AI模型执行相关的操作，例如，进行预测或者上报信息的生成等。具体的，可以设置计时器，该计时器的初始值为AI模型的有效时长。应理解，AI模型的生效起始时间可以根据实际需要进行设置。例如，在终端接收到上述AI模型的信息后，可以启动该计时器，也可以在接收到该AI模型的信息所在的时间单元之后的预设时长后启动该计时器，还可以在网络侧设备指示的起始时刻启动该计时器。其中，网络侧设备指示AI模型的有效期限的方式可以根据实际需要进行设置，在一些实施例中，网络侧设备可以指示以下至少一项来指示AI模型的有效时长：计时器的初始值和计时器的开始启动时刻。
可选地，上述AI模型的有效区域可以理解为AI模型的生效区域。在终端处于有效区域内，该AI模型为有效状态或者激活状态，可以使用该AI模型执行相关的操作。当终端处于有效区域外，该AI模型无效。
可选地，AI模型的各输入参数关联可缺省标识，若可缺省标识指示该输入参数可缺省，则当该输入参数无法获取时，可不输入该参数。指示AI模型的输入参数可以理解为指示该AI模型的输入参数与可缺省标识的关联关系。例如AI模型的信息中指示的输入参数为不可缺省的输入参数或可缺省的输出参数，或者，AI模型的信息直接指示AI模型的各输入参数与可缺省标识的关联关系。
可选地，AI模型的信息中指示了AI模型的输入参数对应的默认值的情况下，则在终端无法获取到相应的输入值时，可以使用该默认值作为输入。
可选地，在一些实施例中，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
可选地，所述终端状态信息包括以下至少一项：
所述终端的位置信息；
所述终端的移动信息；
所述终端的业务信息。
本申请实施例中，上述位置信息可以为全球定位系统(Global Positioning System，GPS)测量结果；上述移动信息可以包括移动方向和移动速度，上述业务信息可以包括正在进行低时延业务。
可选地，所述无线信号测量结果包括以下至少一项：
所述网络侧设备参考信号的RSRP；
所述网络侧设备参考信号的RSRQ。
本申请实施例中，网络侧设备参考信号的RSRP可以包括服务小区参考信号的RSRP和邻小区参考信号的RSRP中的至少一项。网络侧设备参考信号的RSRQ可以包括服务小区参考信号的RSRQ和邻小区参考信号的RSRQ中的至少一项。
可选地，所述AI能力信息包括以下至少一项：
支持的AI模型结构；
支持的AI模型数据处理方式；
支持的AI模型输入参数，例如是否支持某个输入参数。
可选地，上述偏好信息可以理解为终端的期望信息，或建议信息。具体的，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
请参见图4，图4是本申请实施例提供的一种模型请求处理方法的流程图，如图4所示，包括以下步骤：
步骤401，网络侧设备接收终端发送的目标请求；
步骤402，所述网络侧设备根据目标请求信息获得人工智能AI模型的信息；
步骤403，所述网络侧设备向所述终端发送所述AI模型的信息。
可选地，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
可选地，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
可选地，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
可选地，所述终端状态信息包括以下至少一项：
所述终端的位置信息；
所述终端的移动信息；
所述终端的业务信息。
可选地，所述无线信号测量结果包括以下至少一项：
所述网络侧设备参考信号的RSRP；
所述网络侧设备参考信号的RSRQ。
可选地，所述AI能力信息包括以下至少一项：
支持的AI模型结构；
支持的AI模型数据处理方式；
支持的AI模型输入参数。
可选地，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
可选地，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
可选地，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
需要说明的是，本实施例作为图2所示的实施例对应的网络侧设备的实施方式，其具体的实施方式可以参见图2所示的实施例相关说明，以及达到相同的有益效果，为了避免重复说明，此处不再赘述。
需要说明的是，本申请实施例提供的模型请求方法，执行主体可以为模型请求装置，或者，该模型请求装置中的用于执行模型请求方法的控制模块。本申请实施例中以模型请求装置执行模型请求方法为例，说明本申请实施例提供的模型请求装置。
请参见图5，图5是本申请实施例提供的一种模型请求装置的结构图，如图5所示，模型请求装置500包括：
第一发送模块501，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
第一接收模块502，用于接收网络侧设备发送的所述AI模型的信息；
处理模块503，用于根据所述AI模型的信息得到所述AI模型。
可选地，所述第一发送模块501具体用于：在第一条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，
所述第一条件包括以下至少一项：
所述终端所在位置满足第一预设条件；
所述终端的第一测量量符合预先配置的第一测量结果范围；
所述终端的第一传输参数满足第二预设条件；
所述终端存在第一需求；
所述终端无所述AI模型。
可选地，所述第一预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第一小区标识范围；
终端所在的跟踪区TA符合预先配置的TA标识对应的第一跟踪区标识范围；
终端所在的接入网通知区域RNA符合预先配置的RNA对应的第一区域 范围；
终端所在的公共陆地移动通信网PLMN符合预先配置的PLMN对应的第一通信网范围；
终端所在的地理位置符合预先配置的第一地理位置范围。
可选地，所述第二预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第一参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第一相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第一相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第二参数范围；
不同接收波束对应的信息符合预先配置的第一波束信息，或者，不同接收波束对应的信息的特定表征符合所述第一波束信息；
不同波束对对应的信息符合预先配置的第二波束信息，或者，不同波束对对应的信息的特定表征符合所述第二波束信息；
至少一个频率上获得的测量参数符合预先配置的第三参数范围；
数据解调软信息符合预先配置的第一预设信息，或者，数据解调软信息的特定表征符合所述第一预设信息；
数据包传输误比特率符合预先配置的第二预设信息，或者，数据包传输误比特率的特定表征符合所述第二预设信息；
误块率符合预先配置的第三预设信息，或者，误块率的特定表征符合所述第三预设信息；
干扰测量信息符合预先配置的第四预设信息，或者，干扰测量信息的特定表征符合所述第四预设信息。
可选地，所述第一发送模块501具体用于：在第二条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，所述第二条件为所述AI模型的更新条件；
所述处理模块503具体用于：所述终端根据所述AI模型的信息对原AI模型进行更新，得到目标AI模型。
可选地，所述第二条件包括以下至少一项：
所述终端所在位置满足第三预设条件；
所述终端的第一测量量符合预先配置的第二测量结果范围；
所述终端的第一传输参数满足第四预设条件；
所述终端存在第一需求；
所述AI模型的推理结果置信度达到阈值；
所述AI模型有效期限超时；
所述终端移动至所述AI模型的有效区域外；
所述终端无有效AI模型。
可选地，所述第三预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第二小区标识范围；
终端所在的TA符合预先配置的TA标识对应的第二跟踪区标识范围；
终端所在的RNA符合预先配置的RNA对应的第二区域范围；
终端所在的PLMN符合预先配置的PLMN对应的第二通信网范围；
终端所在的地理位置符合预先配置的第二地理位置范围。
可选地，所述第四预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第四参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第二相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第二相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第五参数范围；
不同接收波束对应的信息符合预先配置的第三波束信息，或者，不同接收波束对应的信息的特定表征符合所述第三波束信息；
不同波束对对应的信息符合预先配置的第四波束信息，或者，不同波束对对应的信息的特定表征符合所述第四波束信息；
至少一个频率上获得的测量参数符合预先配置的第六参数范围；
数据解调软信息符合预先配置的第五预设信息，或者，数据解调软信息的特定表征符合所述第五预设信息；
数据包传输误比特率符合预先配置的第六预设信息，或者，数据包传输误比特率的特定表征符合所述第六预设信息；
误块率符合预先配置的第七预设信息，或者，误块率的特定表征符合所述第七预设信息；
干扰测量信息符合预先配置的第八预设信息，或者，干扰测量信息的特定表征符合所述第八预设信息。
可选地，所述第一测量量包括以下至少一项：信噪比、参考信号接收功率RSRP、信号与干扰加噪声比SINR、参考信号接收质量RSRQ、包时延、往返时延RTT、观测到达时间差OTDOA、信道状态信息对应的测量结果和用户体验质量对应的测量结果。
可选地，所述第一测量量对应的测量资源包括以下至少一项：
物理下行控制信道PDCCH解调参考信号DMRS；
物理下行共享信道PDSCH DMRS；
信道状态信息参考信号CSI-RS；
同步信号块SSB；
定位参考信号PRS。
可选地，所述第一传输参数包括以下至少一项：
信道特征参数；
不同发送天线之间或不同发送端口之间的相位信息；
至少一个波束上获得的测量参数；
不同波束对对应的信息；
至少一个频率上获得的测量参数；
数据解调软信息；
数据包传输误比特率或误块率；
干扰测量信息。
可选地，所述第一需求包括以下至少一项：上报无线资源管理RRM测量报告；切换判决；重定向判决；生成定位结果；上报信道状态信息CSI；用户轨迹预测；用户业务需求预测；用户切片需求预测。
可选地，所述有效区域基于以下至少一项确定：小区列表、TA列表、RNA列表、PLMN列表和地理位置。
可选地，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
可选地，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
可选地，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
可选地，所述终端状态信息包括以下至少一项：
所述终端的位置信息；
所述终端的移动信息；
所述终端的业务信息。
可选地，所述无线信号测量结果包括以下至少一项：
所述网络侧设备参考信号的RSRP；
所述网络侧设备参考信号的RSRQ。
可选地，所述AI能力信息包括以下至少一项：
支持的AI模型结构；
支持的AI模型数据处理方式；
支持的AI模型输入参数。
可选地，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
本申请实施例提供的模型请求装置能够实现图3的方法实施例中各个过程，为避免重复，这里不再赘述。
请参见图6，图6是本申请实施例提供的一种模型请求处理装置的结构图，如图6所示，模型请求处理装置600包括：
第二接收模块601，用于接收终端发送的目标请求；
生成模块602，用于根据目标请求信息获得人工智能AI模型的信息；
第二发送模块603，用于向所述终端发送所述AI模型的信息。
可选地，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
可选地，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
可选地，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
可选地，所述终端状态信息包括以下至少一项：
所述终端的位置信息；
所述终端的移动信息；
所述终端的业务信息。
可选地，所述无线信号测量结果包括以下至少一项：
所述网络侧设备参考信号的RSRP；
所述网络侧设备参考信号的RSRQ。
可选地，所述AI能力信息包括以下至少一项：
支持的AI模型结构；
支持的AI模型数据处理方式；
支持的AI模型输入参数。
可选地，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
本申请实施例提供的模型请求处理装置能够实现图3的方法实施例中各个过程，为避免重复，这里不再赘述。
本申请实施例中的模型请求装置和模型请求处理装置可以是装置，具有操作系统的装置或电子设备，也可以是终端中的部件、集成电路、或芯片。该装置可以是移动终端，也可以为非移动终端。示例性的，移动终端可以包 括但不限于上述所列举的终端11的类型，非移动终端可以为服务器、网络附属存储器(Network Attached Storage，NAS)、个人计算机(personal computer，PC)、电视机(television，TV)、柜员机或者自助机等，本申请实施例不作具体限定。
可选的，如图7所示，本申请实施例还提供一种通信设备700，包括处理器701，存储器702，存储在存储器702上并可在所述处理器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执行时实现上述模型请求方法实施例的各个过程，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执行时实现上述模型请求处理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一种终端，包括处理器和通信接口，所述通信接口，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；接收网络侧设备发送的所述AI模型的信息；所述处理器，用于根据所述AI模型的信息得到所述AI模型。该终端实施例是与上述终端侧方法实施例对应的，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终端实施例中，且能达到相同的技术效果。具体地，图8为实现本申请各个实施例的一种终端的硬件结构示意图。
该终端800包括但不限于：射频单元801、网络模块802、音频输出单元803、输入单元804、传感器805、显示单元806、用户输入单元807、接口单元808、存储器809以及处理器810等中的至少部分部件。
本领域技术人员可以理解，终端800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810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8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804可以包括图形处理器(Graphics Processing Unit，GPU)和麦克风，图形处理器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806可包括显示面板，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。用户输入单元807包括触控面板以及其他输入设备。触控面板，也称为触摸屏。触控面板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801将来自网络侧设备的下行数据接收后， 给处理器810处理；另外，将上行的数据发送给网络侧设备。通常，射频单元801包括但不限于天线、至少一个放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器809可用于存储软件程序或指令以及各种数据。存储器809可主要包括存储程序或指令区和存储数据区，其中，存储程序或指令区可存储操作系统、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809可以包括高速随机存取存储器，还可以包括非瞬态性存储器，其中，非瞬态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。例如至少一个磁盘存储器件、闪存器件、或其他非瞬态性固态存储器件。
处理器810可包括一个或多个处理单元；可选的，处理器810可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操作系统、用户界面和应用程序或指令等，调制解调处理器主要处理无线通信，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810中。
其中，射频单元801，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；接收网络侧设备发送的所述AI模型的信息；
处理器810，用于根据所述AI模型的信息得到所述AI模型。
可选地，所述射频单元801具体用于：在第一条件满足的情况下，向网络侧设备发送所述目标请求；其中，
所述第一条件包括以下至少一项：
所述终端所在位置满足第一预设条件；
所述终端的第一测量量符合预先配置的第一测量结果范围；
所述终端的第一传输参数满足第二预设条件；
所述终端存在第一需求；
所述终端无所述AI模型。
可选地，所述第一预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第一小区标识范围；
终端所在的跟踪区TA符合预先配置的TA标识对应的第一跟踪区标识范围；
终端所在的接入网通知区域RNA符合预先配置的RNA对应的第一区域范围；
终端所在的公共陆地移动通信网PLMN符合预先配置的PLMN对应的第 一通信网范围；
终端所在的地理位置符合预先配置的第一地理位置范围。
可选地，所述第二预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第一参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第一相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第一相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第二参数范围；
不同接收波束对应的信息符合预先配置的第一波束信息，或者，不同接收波束对应的信息的特定表征符合所述第一波束信息；
不同波束对对应的信息符合预先配置的第二波束信息，或者，不同波束对对应的信息的特定表征符合所述第二波束信息；
至少一个频率上获得的测量参数符合预先配置的第三参数范围；
数据解调软信息符合预先配置的第一预设信息，或者，数据解调软信息的特定表征符合所述第一预设信息；
数据包传输误比特率符合预先配置的第二预设信息，或者，数据包传输误比特率的特定表征符合所述第二预设信息；
误块率符合预先配置的第三预设信息，或者，误块率的特定表征符合所述第三预设信息；
干扰测量信息符合预先配置的第四预设信息，或者，干扰测量信息的特定表征符合所述第四预设信息。
可选地，所述射频单元801具体用于：在第二条件满足的情况下，向网络侧设备发送所述目标请求；其中，所述第二条件为所述AI模型的更新条件；
所述处理器810具体用于：根据所述AI模型的信息对原AI模型进行更新，得到目标AI模型。
可选地，所述第二条件包括以下至少一项：
所述终端所在位置满足第三预设条件；
所述终端的第一测量量符合预先配置的第二测量结果范围；
所述终端的第一传输参数满足第四预设条件；
所述终端存在第一需求；
所述AI模型的推理结果置信度达到阈值；
所述AI模型有效期限超时；
所述终端移动至所述AI模型的有效区域外；
所述终端无有效AI模型。
可选地，所述第三预设条件包括以下至少一项：
终端所在的服务小区符合预先配置的小区标识对应的第二小区标识范围；
终端所在的TA符合预先配置的TA标识对应的第二跟踪区标识范围；
终端所在的RNA符合预先配置的RNA对应的第二区域范围；
终端所在的PLMN符合预先配置的PLMN对应的第二通信网范围；
终端所在的地理位置符合预先配置的第二地理位置范围。
可选地，所述第四预设条件包括以下至少一项：
信道特征参数或信道参数的预设表征符合预先配置的第四参数范围；
不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第二相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第二相位信息；
波束测量参数或波束测量参数的特定表征符合预设配置的第五参数范围；
不同接收波束对应的信息符合预先配置的第三波束信息，或者，不同接收波束对应的信息的特定表征符合所述第三波束信息；
不同波束对对应的信息符合预先配置的第四波束信息，或者，不同波束对对应的信息的特定表征符合所述第四波束信息；
至少一个频率上获得的测量参数符合预先配置的第六参数范围；
数据解调软信息符合预先配置的第五预设信息，或者，数据解调软信息的特定表征符合所述第五预设信息；
数据包传输误比特率符合预先配置的第六预设信息，或者，数据包传输误比特率的特定表征符合所述第六预设信息；
误块率符合预先配置的第七预设信息，或者，误块率的特定表征符合所述第七预设信息；
干扰测量信息符合预先配置的第八预设信息，或者，干扰测量信息的特定表征符合所述第八预设信息。
可选地，所述第一测量量包括以下至少一项：信噪比、参考信号接收功率RSRP、信号与干扰加噪声比SINR、参考信号接收质量RSRQ、包时延、往返时延RTT、观测到达时间差OTDOA、信道状态信息对应的测量结果和用户体验质量对应的测量结果。
可选地，所述第一测量量对应的测量资源包括以下至少一项：
物理下行控制信道PDCCH解调参考信号DMRS；
物理下行共享信道PDSCH DMRS；
信道状态信息参考信号CSI-RS；
同步信号块SSB；
定位参考信号PRS。
可选地，所述第一传输参数包括以下至少一项：
信道特征参数；
不同发送天线之间或不同发送端口之间的相位信息；
至少一个波束上获得的测量参数；
不同波束对对应的信息；
至少一个频率上获得的测量参数；
数据解调软信息；
数据包传输误比特率或误块率；
干扰测量信息。
可选地，所述第一需求包括以下至少一项：上报无线资源管理RRM测量报告；切换判决；重定向判决；生成定位结果；上报信道状态信息CSI；用户轨迹预测；用户业务需求预测；用户切片需求预测。
可选地，所述有效区域基于以下至少一项确定：小区列表、TA列表、RNA列表、PLMN列表和地理位置。
可选地，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
可选地，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
可选地，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
可选地，所述终端状态信息包括以下至少一项：所述终端的位置信息；所述终端的移动信息；所述终端的业务信息。
可选地，所述无线信号测量结果包括以下至少一项：
所述网络侧设备参考信号的RSRP；
所述网络侧设备参考信号的RSRQ。
可选地，所述AI能力信息包括以下至少一项：
支持的AI模型结构；
支持的AI模型数据处理方式；
支持的AI模型输入参数。
可选地，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
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，所述通信接口，用于接收终端发送的目标请求；所述处理器，用于根据目标请求信息获得人工智能AI模型的信息；所述通信接口，还用于向所述终端发送所述 AI模型的信息。该网络侧设备实施例是与上述网络侧设备方法实施例对应的，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9所示，该网络侧设备900包括：天线901、射频装置902、基带装置903。天线901与射频装置902连接。在上行方向上，射频装置902通过天线901接收信息，将接收的信息发送给基带装置903进行处理。在下行方向上，基带装置903对要发送的信息进行处理，并发送给射频装置902，射频装置902对收到的信息进行处理后经过天线901发送出去。
上述频带处理装置可以位于基带装置903中，以上实施例中网络侧设备执行的方法可以在基带装置903中实现，该基带装置903包括处理器904和存储器905。
基带装置903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9所示，其中一个芯片例如为处理器904，与存储器905连接，以调用存储器905中的程序，执行以上方法实施例中所示的网络侧设备操作。
该基带装置903还可以包括网络接口906，用于与射频装置902交互信息，该接口例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本申请实施例的网络侧设备还包括：存储在存储器905上并可在处理器904上运行的指令或程序，处理器904调用存储器905中的指令或程序执行图6所示的各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
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模型请求方法或模型请求处理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电子设备中的处理器。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(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等。
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，实现上述模型请求方法或模型请求处理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程序产品，所述程序产品存储在非瞬态的存储介质中，所述程序产品被至少一个处理器执行以实现上述模型请求方法或模型请求处理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。
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，或者基站等)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. 一种模型请求方法，包括：
    终端向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
    所述终端接收网络侧设备发送的所述AI模型的信息；
    所述终端根据所述AI模型的信息得到所述AI模型。
  2. 根据权利要求1所述的方法，其中，所述终端向网络侧设备发送目标请求包括：
    在第一条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，
    所述第一条件包括以下至少一项：
    所述终端所在位置满足第一预设条件；
    所述终端的第一测量量符合预先配置的第一测量结果范围；
    所述终端的第一传输参数满足第二预设条件；
    所述终端存在第一需求；
    所述终端无所述AI模型。
  3. 根据权利要求2所述的方法，其中，所述第一预设条件包括以下至少一项：
    终端所在的服务小区符合预先配置的小区标识对应的第一小区标识范围；
    终端所在的跟踪区TA符合预先配置的TA标识对应的第一跟踪区标识范围；
    终端所在的接入网通知区域RNA符合预先配置的RNA对应的第一区域范围；
    终端所在的公共陆地移动通信网PLMN符合预先配置的PLMN对应的第一通信网范围；
    终端所在的地理位置符合预先配置的第一地理位置范围。
  4. 根据权利要求2所述的方法，其中，所述第二预设条件包括以下至少一项：
    信道特征参数或信道参数的预设表征符合预先配置的第一参数范围；
    不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第一相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第一相位信息；
    波束测量参数或波束测量参数的特定表征符合预设配置的第二参数范围；
    不同接收波束对应的信息符合预先配置的第一波束信息，或者，不同接 收波束对应的信息的特定表征符合所述第一波束信息；
    不同波束对对应的信息符合预先配置的第二波束信息，或者，不同波束对对应的信息的特定表征符合所述第二波束信息；
    至少一个频率上获得的测量参数符合预先配置的第三参数范围；
    数据解调软信息符合预先配置的第一预设信息，或者，数据解调软信息的特定表征符合所述第一预设信息；
    数据包传输误比特率符合预先配置的第二预设信息，或者，数据包传输误比特率的特定表征符合所述第二预设信息；
    误块率符合预先配置的第三预设信息，或者，误块率的特定表征符合所述第三预设信息；
    干扰测量信息符合预先配置的第四预设信息，或者，干扰测量信息的特定表征符合所述第四预设信息。
  5. 根据权利要求1所述的方法，其中，所述终端向网络侧设备发送目标请求包括：
    在第二条件满足的情况下，所述终端向网络侧设备发送所述目标请求；其中，所述第二条件为所述AI模型的更新条件；
    所述终端根据所述AI模型的信息得到所述AI模型包括：
    所述终端根据所述AI模型的信息对原AI模型进行更新，得到目标AI模型。
  6. 根据权利要求5所述的方法，其中，所述第二条件包括以下至少一项：
    所述终端所在位置满足第三预设条件；
    所述终端的第一测量量符合预先配置的第二测量结果范围；
    所述终端的第一传输参数满足第四预设条件；
    所述终端存在第一需求；
    所述AI模型的推理结果置信度达到阈值；
    所述AI模型有效期限超时；
    所述终端移动至所述AI模型的有效区域外；
    所述终端无有效AI模型。
  7. 根据权利要求6所述的方法，其中，所述第三预设条件包括以下至少一项：
    终端所在的服务小区符合预先配置的小区标识对应的第二小区标识范围；
    终端所在的TA符合预先配置的TA标识对应的第二跟踪区标识范围；
    终端所在的RNA符合预先配置的RNA对应的第二区域范围；
    终端所在的PLMN符合预先配置的PLMN对应的第二通信网范围；
    终端所在的地理位置符合预先配置的第二地理位置范围。
  8. 根据权利要求6所述的方法，其中，所述第四预设条件包括以下至少一项：
    信道特征参数或信道参数的预设表征符合预先配置的第四参数范围；
    不同发送天线之间或不同发送端口之间的相位信息符合预先配置的第二相位信息，或者，不同发送天线之间或不同发送端口之间的所述相位信息的预设表征符合所述第二相位信息；
    波束测量参数或波束测量参数的特定表征符合预设配置的第五参数范围；
    不同接收波束对应的信息符合预先配置的第三波束信息，或者，不同接收波束对应的信息的特定表征符合所述第三波束信息；
    不同波束对对应的信息符合预先配置的第四波束信息，或者，不同波束对对应的信息的特定表征符合所述第四波束信息；
    至少一个频率上获得的测量参数符合预先配置的第六参数范围；
    数据解调软信息符合预先配置的第五预设信息，或者，数据解调软信息的特定表征符合所述第五预设信息；
    数据包传输误比特率符合预先配置的第六预设信息，或者，数据包传输误比特率的特定表征符合所述第六预设信息；
    误块率符合预先配置的第七预设信息，或者，误块率的特定表征符合所述第七预设信息；
    干扰测量信息符合预先配置的第八预设信息，或者，干扰测量信息的特定表征符合所述第八预设信息。
  9. 根据权利要求2或6所述的方法，其中，所述第一测量量包括以下至少一项：信噪比、参考信号接收功率RSRP、信号与干扰加噪声比SINR、参考信号接收质量RSRQ、包时延、往返时延RTT、观测到达时间差OTDOA、信道状态信息对应的测量结果和用户体验质量对应的测量结果。
  10. 根据权利要求2或6所述的方法，其中，所述第一测量量对应的测量资源包括以下至少一项：
    物理下行控制信道PDCCH解调参考信号DMRS；
    物理下行共享信道PDSCH DMRS；
    信道状态信息参考信号CSI-RS；
    同步信号块SSB；
    定位参考信号PRS。
  11. 根据权利要求2或6所述的方法，其中，所述第一传输参数包括以下至少一项：
    信道特征参数；
    不同发送天线之间或不同发送端口之间的相位信息；
    至少一个波束上获得的测量参数；
    不同波束对对应的信息；
    至少一个频率上获得的测量参数；
    数据解调软信息；
    数据包传输误比特率或误块率；
    干扰测量信息。
  12. 根据权利要求2或6所述的方法，其中，所述第一需求包括以下至少一项：上报无线资源管理RRM测量报告；切换判决；重定向判决；生成定位结果；上报信道状态信息CSI；用户轨迹预测；用户业务需求预测；用户切片需求预测。
  13. 根据权利要求6所述的方法，其中，所述有效区域基于以下至少一项确定：小区列表、TA列表、RNA列表、PLMN列表和地理位置。
  14. 根据权利要求1至8中任一项所述的方法，其中，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
  15. 根据权利要求14所述的方法，其中，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
  16. 根据权利要求1至8中任一项所述的方法，其中，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
  17. 根据权利要求16所述的方法，其中，所述终端状态信息包括以下至少一项：
    所述终端的位置信息；
    所述终端的移动信息；
    所述终端的业务信息。
  18. 根据权利要求16所述的方法，其中，所述无线信号测量结果包括以下至少一项：
    所述网络侧设备参考信号的RSRP；
    所述网络侧设备参考信号的RSRQ。
  19. 根据权利要求16所述的方法，其中，所述AI能力信息包括以下至少一项：
    支持的AI模型结构；
    支持的AI模型数据处理方式；
    支持的AI模型输入参数。
  20. 根据权利要求16所述的方法，其中，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
  21. 一种模型请求处理方法，包括：
    网络侧设备接收终端发送的目标请求；
    所述网络侧设备根据目标请求获得人工智能AI模型的信息；
    所述网络侧设备向所述终端发送所述AI模型的信息。
  22. 根据权利要求21所述的方法，其中，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
  23. 根据权利要求22所述的方法，其中，所述AI模型的信息还包括以下至少一项：所述AI模型的更新条件、AI模型的有效期限、AI模型的有效区域、AI模型的输入参数、AI模型的输入参数对应的默认值。
  24. 根据权利要求21至23中任一项所述的方法，其中，所述目标请求携带有以下至少一项：终端状态信息、无线信号测量结果、AI能力信息和针对AI模型的偏好信息。
  25. 根据权利要求24所述的方法，其中，所述终端状态信息包括以下至少一项：
    所述终端的位置信息；
    所述终端的移动信息；
    所述终端的业务信息。
  26. 根据权利要求24所述的方法，其中，所述无线信号测量结果包括以下至少一项：
    所述网络侧设备参考信号的RSRP；
    所述网络侧设备参考信号的RSRQ。
  27. 根据权利要求24所述的方法，其中，所述AI能力信息包括以下至少一项：
    支持的AI模型结构；
    支持的AI模型数据处理方式；
    支持的AI模型输入参数。
  28. 根据权利要求24所述的方法，其中，所述偏好信息包括对AI模型结构的偏好和对AI模型数据处理方式的偏好中的至少一项。
  29. 一种模型请求装置，包括：
    第一发送模块，用于向网络侧设备发送目标请求，所述目标请求用于请求人工智能AI模型的信息；
    第一接收模块，用于接收网络侧设备发送的所述AI模型的信息；
    处理模块，用于根据所述AI模型的信息得到所述AI模型。
  30. 根据权利要求29所述的装置，其中，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
  31. 一种模型请求处理装置，包括：
    第二接收模块，用于接收终端发送的目标请求；
    生成模块，用于根据目标请求信息获得人工智能AI模型的信息；
    第二发送模块，用于向所述终端发送所述AI模型的信息。
  32. 根据权利要求31所述的装置，其中，所述AI模型的信息包括以下至少一项：AI模型的标识、AI模型的输出参数、AI模型的结构信息、AI模型的模型参数信息和AI模型数据的处理方式信息。
  33. 一种终端，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的程序，其中，所述程序被所述处理器执行时实现如权利要求1至20中任一项所述的模型请求方法中的步骤。
  34. 一种网络侧设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的程序或指令，其中，所述程序或指令被所述处理器执行时实现如权利要求21至28中任一项所述的模型请求处理方法中的步骤。
  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，其中，所述程序或指被处理器执行时实现如权利要求1至20中任一项所述的模型请求方法的步骤，或者所述程序或指令被处理器执行时实现如权利要求21至28中任一项所述的模型请求处理方法的步骤。
PCT/CN2022/126134 2021-10-21 2022-10-19 模型请求方法、模型请求处理方法及相关设备 WO2023066288A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111229121.X 2021-10-21
CN202111229121.XA CN116017493A (zh) 2021-10-21 2021-10-21 模型请求方法、模型请求处理方法及相关设备

Publications (1)

Publication Number Publication Date
WO2023066288A1 true WO2023066288A1 (zh) 2023-04-27

Family

ID=86023555

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/126134 WO2023066288A1 (zh) 2021-10-21 2022-10-19 模型请求方法、模型请求处理方法及相关设备

Country Status (2)

Country Link
CN (1) CN116017493A (zh)
WO (1) WO2023066288A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117835262A (zh) * 2022-09-26 2024-04-05 维沃移动通信有限公司 Ai模型的处理方法、装置及通信设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200413316A1 (en) * 2018-03-08 2020-12-31 Telefonaktiebolaget Lm Ericsson (Publ) Managing communication in a wireless communications network
CN112512058A (zh) * 2020-05-24 2021-03-16 中兴通讯股份有限公司 网络优化方法、服务器、客户端设备、网络设备和介质
CN112698848A (zh) * 2020-12-31 2021-04-23 Oppo广东移动通信有限公司 机器学习模型的下载方法、装置、终端及存储介质
CN113191502A (zh) * 2021-04-21 2021-07-30 烽火通信科技股份有限公司 一种人工智能模型在线训练方法及系统
CN113420888A (zh) * 2021-06-03 2021-09-21 中国石油大学(华东) 一种基于泛化域自适应的无监督联邦学习方法
CN113435585A (zh) * 2021-07-15 2021-09-24 支付宝(杭州)信息技术有限公司 一种业务处理方法、装置及设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200413316A1 (en) * 2018-03-08 2020-12-31 Telefonaktiebolaget Lm Ericsson (Publ) Managing communication in a wireless communications network
CN112512058A (zh) * 2020-05-24 2021-03-16 中兴通讯股份有限公司 网络优化方法、服务器、客户端设备、网络设备和介质
CN112698848A (zh) * 2020-12-31 2021-04-23 Oppo广东移动通信有限公司 机器学习模型的下载方法、装置、终端及存储介质
CN113191502A (zh) * 2021-04-21 2021-07-30 烽火通信科技股份有限公司 一种人工智能模型在线训练方法及系统
CN113420888A (zh) * 2021-06-03 2021-09-21 中国石油大学(华东) 一种基于泛化域自适应的无监督联邦学习方法
CN113435585A (zh) * 2021-07-15 2021-09-24 支付宝(杭州)信息技术有限公司 一种业务处理方法、装置及设备

Also Published As

Publication number Publication date
CN116017493A (zh) 2023-04-25

Similar Documents

Publication Publication Date Title
US11399303B2 (en) Configuration of a neural network for a radio access network (RAN) node of a wireless network
CN114363921B (zh) Ai网络参数的配置方法和设备
US20220190883A1 (en) Beam prediction for wireless networks
WO2023109827A1 (zh) 客户端筛选方法及装置、客户端及中心设备
WO2023066288A1 (zh) 模型请求方法、模型请求处理方法及相关设备
WO2023134650A1 (zh) 信息交互方法、装置及通信设备
WO2023143572A1 (zh) 基于人工智能ai模型的定位方法及通信设备
US20240137783A1 (en) Signalling support for split ml-assistance between next generation random access networks and user equipment
WO2023040885A1 (zh) 参数选择方法、参数配置方法、终端及网络侧设备
WO2024067281A1 (zh) Ai模型的处理方法、装置及通信设备
WO2023125951A1 (zh) 通信模型配置方法、装置和通信设备
WO2024032694A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2023179540A1 (zh) 信道预测方法、装置及无线通信设备
WO2024067434A1 (zh) Rs配置方法、装置、终端及网络侧设备
WO2023109828A1 (zh) 数据收集方法及装置、第一设备、第二设备
WO2023088387A1 (zh) 信道预测方法、装置、ue及系统
WO2023174253A1 (zh) Ai模型的处理方法及设备
WO2023020383A1 (zh) 信息处理方法、装置及终端
WO2023174325A1 (zh) Ai模型的处理方法及设备
WO2022237822A1 (zh) 训练数据集获取方法、无线传输方法、装置及通信设备
WO2024067280A1 (zh) 更新ai模型参数的方法、装置及通信设备
WO2023040888A1 (zh) 数据传输方法及装置
WO2023072239A1 (zh) 信道预测方法、装置、网络侧设备及终端
WO2024017239A1 (zh) 数据采集方法及装置、通信设备
WO2023098661A1 (zh) 定位方法及通信设备

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: 22882893

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2022882893

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2022882893

Country of ref document: EP

Effective date: 20240521