WO2024120447A1 - 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统 - Google Patents

模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统 Download PDF

Info

Publication number
WO2024120447A1
WO2024120447A1 PCT/CN2023/136813 CN2023136813W WO2024120447A1 WO 2024120447 A1 WO2024120447 A1 WO 2024120447A1 CN 2023136813 W CN2023136813 W CN 2023136813W WO 2024120447 A1 WO2024120447 A1 WO 2024120447A1
Authority
WO
WIPO (PCT)
Prior art keywords
supervision
model
target
indication information
triggering
Prior art date
Application number
PCT/CN2023/136813
Other languages
English (en)
French (fr)
Inventor
贾承璐
邬华明
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 WO2024120447A1 publication Critical patent/WO2024120447A1/zh

Links

Definitions

  • the present application belongs to the field of communication technology, and specifically relates to a model supervision triggering method, device, UE, network side equipment, readable storage medium and communication system.
  • AI Artificial Intelligence
  • integrating artificial intelligence models into wireless communication networks can significantly improve technical indicators such as throughput, latency, and user capacity of wireless communication networks.
  • artificial intelligence is an important task for future wireless communication networks.
  • the model performance of artificial intelligence models can be improved by supervising the artificial intelligence models.
  • a model supervision triggering method which is applied to UE, and the method includes: the user equipment UE receives first indication information sent by a network side device, and the first indication information is used to indicate a triggering method of model supervision; the UE supervises the target model according to the first indication information; wherein the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • a model supervision triggering device which may include: a receiving module and a processing module.
  • the receiving module is used to receive first indication information sent by a network side device, wherein the first indication information is used to indicate a triggering mode of model supervision; the processing module is used to supervise the target model according to the first indication information.
  • the triggering mode includes at least one of the following: a periodic supervision mode, a semi-continuous supervision mode, a non-periodic supervision mode, and a supervision mode triggered by an event.
  • a model supervision triggering method is provided, which is applied to a network side device.
  • the method includes: the network side device sends a first indication information to a UE, and the first indication information is used to indicate a triggering method for model supervision; the first indication information is used by the UE to supervise the target model; wherein the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • a model supervision triggering device comprising: a sending module.
  • the sending module is used to send first indication information to a UE, wherein the first indication information is used to indicate a triggering mode of model supervision; the first indication information is used for the UE to supervise a target model; wherein the triggering mode comprises at least one of the following: a periodic supervision mode, a semi-continuous supervision mode, a non-periodic supervision mode, and a supervision mode triggered by an event.
  • a UE which includes a processor and a memory, wherein the memory stores a program or instruction that can be executed on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.
  • a UE comprising a processor and a communication interface, wherein the communication interface is used to connect Receive first indication information sent by a network side device, wherein the first indication information is used to indicate a triggering method for model supervision; the processor is used to supervise the target model according to the first indication information; wherein the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • a network side device which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the third aspect are implemented.
  • a network side device including a processor and a communication interface, wherein the processor is used to obtain first indication information, the communication interface is used to send the first indication information to a UE, and the first indication information is used to indicate a triggering mode of model supervision; the first indication information is used by the UE to supervise the target model;
  • the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • a communication system comprising: a UE and a network side device, wherein the UE can be used to execute the steps of the model supervision triggering method as described in the first aspect, and the network side device can be used to execute the steps of the model supervision triggering method as described in the third aspect.
  • it comprises: a model supervision triggering device as described in the second aspect and a model supervision triggering device as described in the fourth aspect, wherein the model supervision triggering device as described in the second aspect is used to execute the steps of the model supervision triggering method as described in the third aspect, and the model supervision triggering device as described in the fourth aspect is used to execute the steps of the model supervision triggering method as described in the third aspect.
  • a readable storage medium on which a program or instruction is stored.
  • the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the third aspect are implemented.
  • a computer program/program product is provided, wherein the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the steps of the model supervision triggering method as described in the first aspect.
  • the UE can receive the first indication message sent by the network side device to indicate a triggering mode for indicating model supervision, and the triggering mode includes at least one of the following: a periodic supervision mode, a semi-continuous supervision mode, a non-periodic supervision mode, and a supervision mode triggered by an event; that is, the network side device instructs the UE to trigger the triggering mode of model supervision, the UE can directly trigger the model supervision of the target model according to the first indication message.
  • FIG1 is a schematic diagram of the architecture of a wireless communication system provided in an embodiment of the present application.
  • FIG2 is a schematic diagram of the structure of a neural network provided by the related art
  • FIG3 is a schematic diagram of a neuron structure provided by the related art
  • FIG4 is a flow chart of a model supervision triggering method provided in an embodiment of the present application.
  • FIG5 is one of the structural schematic diagrams of a model supervision triggering device provided in an embodiment of the present application.
  • FIG6 is a second structural diagram of a model supervision triggering device provided in an embodiment of the present application.
  • FIG7 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application.
  • FIG8 is a hardware structure of a UE according to an embodiment of the present application.
  • FIG9 is a schematic diagram of a hardware structure of a network side device provided in an embodiment of the present application.
  • first, second, etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by “first” and “second” are generally of the same type, and the number of objects is not limited.
  • the first object can be one or more.
  • “and/or” in the specification and claims represents at least one of the connected objects, and the character “/" generally represents that the objects associated with each other are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • NR new radio
  • FIG1 shows a block diagram of a wireless communication system applicable to an embodiment of the present application.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer) or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a handheld computer, a netbook, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) device, a robot, a wearable device (Wearable Device), a vehicle-mounted device (VUE), a pedestrian terminal (PUE), a smart home (home appliances with wireless communication functions, such as refrigerators, televisions, washing machines or furniture, etc.), a game console, a personal computer (personal computer, PC), a teller machine or a self-service machine and other terminal side devices, and
  • the network side device 12 may include an access network device or a core network device, wherein the access network device 12 may also be referred to as a radio access network device, a radio access network (RAN), a radio access network function or a radio access network unit.
  • the access network device 12 may include a base station, a WLAN access point or a WiFi node, etc.
  • the base station may be referred to as a node B, an evolved node B (eNB), an access point, a base transceiver station (BTS), a radio base station, a radio transceiver, a basic service set (BSS), an extended service set (ESS), a home node B, a home evolved node B, a transmitting and receiving point (TRP) or other appropriate terms in the field, as long as the same technical effect is achieved, the base station is not limited to a specific technical vocabulary, it should be noted that in the embodiment of the present application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.
  • the core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (Mobility Management Entity, MME), access mobility management function (Access and Mobility Management Function, AMF), Session Management Function (Session Management Function, SMF), User Plane Function (User Plane Function, UPF), Policy Control Function (Policy Control Function, PCF), Policy and Charging Rules Function (Policy and Charging Rules Function, PCRF), Edge Application Server Discovery Function (Edge Application Server Discovery Function, EASDF), Unified Data Management (Unified Data Management, UDM), Unified Data Repository (Unified Data Repository, UDR), Home Subscriber Server (Home Subscriber Server, HSS), Centralized Network Configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF, or L-NEF), Binding Support Function (BSF), Application Function (AF), etc.
  • MME mobility management entity
  • AMF Access and Mobility Management Function
  • AI is the integration of artificial intelligence into wireless communication networks. Significantly improving technical indicators such as throughput, latency, and user capacity is an important task for future wireless communication networks.
  • AI modules such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. The embodiments of this application are described using neural networks as an example, but do not limit the specific types of AI modules.
  • the neural network includes an input layer, a hidden layer and an output layer, X1, X2, ..., Xn are inputs, and Y is output.
  • the neural network is composed of neurons, as shown in Figure 3, which is a schematic diagram of neurons.
  • a1, a2, ... aK are inputs
  • w is the weight (multiplicative coefficient)
  • b is the bias (additive coefficient)
  • z a1w1 + ... + akwk + ... + aKwK + b
  • ⁇ (z) is the activation function.
  • the activation function includes Sigmoid, tanh, ReLU (Rectified Linear Unit, linear rectification function, rectified linear unit), etc.
  • the parameters of the neural network are optimized using a gradient optimization algorithm.
  • the gradient optimization algorithm is a type of algorithm that minimizes or maximizes an objective function (also called a loss function), which is a mathematical combination of model parameters and data.
  • an objective function also called a loss function
  • a neural network model f(.) can be constructed. With the model, the predicted output f(x) can be obtained based on the input x, and the difference between the predicted value and the true value (f(x)-Y) can be calculated. This is the loss function. Find the appropriate W,b to minimize the value of the above loss function. The smaller the loss value, the closer the model is to the actual situation.
  • the most common optimization algorithm currently used is based on the BP (error Back Propagation) algorithm.
  • 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 sample is transmitted from the input layer, processed by each hidden layer layer by layer, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, the error back propagation stage is entered.
  • Error back propagation is to propagate the output error layer by layer through the hidden layer to the input layer in some form, and distribute the error to all units in each layer, so as to obtain the error signal of each layer unit, and this error signal is used as the basis for correcting the weights of each unit.
  • This process of adjusting the weights of each layer of the signal forward propagation and error back propagation is repeated over and over again.
  • the process of continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until the pre-set number of learning times is reached.
  • FIG. 4 shows a flow chart of a model supervision triggering method provided by an embodiment of the present application.
  • the model supervision triggering method provided by an embodiment of the present application may include the following steps 401 to 403.
  • Step 401 The network side device sends first indication information to the UE.
  • Step 402 The UE receives first indication information.
  • the first indication information may be used to indicate a triggering method for model supervision.
  • the triggering method indicated by the above-mentioned first indication information may include at least one of the following: method 1, periodic supervision method, method 2, semi-continuous supervision method, method 3, non-periodic supervision method, method 4, supervision method triggered by an event.
  • the network side device can indicate the triggering method of the UE model supervision through the first indication information, so that the UE can trigger the model supervision according to the triggering method indicated by the network side device.
  • Step 403 The UE supervises the target model according to the first indication information.
  • the UE in method 1, may periodically supervise the target model; in method 2 and method 3, the UE may supervise the target model at least once after receiving the first indication information; in method 4, the UE may supervise the target model when a specific event occurs.
  • the UE can obtain the triggering condition for triggering model supervision in advance. Specifically, in Mode 1, the UE can obtain the period value of triggering model supervision in advance, that is, know the supervision period in advance; in Mode 4, the UE needs to obtain the event that triggers model supervision in advance.
  • the periodic value and event that triggers model supervision may be agreed upon by a protocol, preset, or configured by a network-side device.
  • the UE's supervision of the target model can be understood as: during the movement of the target model, the UE supervises the operation of the target model to obtain the operation and performance of the target model.
  • the performance of the target model can be optimized according to the supervision result of the target model.
  • the supervision result may be reported to the network side device.
  • the UE may package the supervision results obtained from multiple supervisions and report them to the network side device.
  • the target model may be a positioning model, such as an AI positioning model. It is understandable that the AI positioning model is used for positioning.
  • model supervision triggering method provided in the embodiment of the present application is described in detail below using methods 1 to 4 as examples.
  • the first indication information may further include a periodic value of the model supervision.
  • the unit of the periodic value of the model supervision may be: OFDM symbol, time slot, subframe, frame, millisecond, second, minute, hour, day, etc.
  • the period of model supervision may be: at least one OFDM symbol, at least one time slot, at least one subframe, at least one frame, at least one millisecond, at least one second, at least one minute, at least one hour, at least one day, etc.
  • the period value is S time slots, where S is a positive integer; thus, after receiving the first indication information, the UE may perform model supervision on the target model once every S time slots.
  • the first indication information may also include at least one of the following: a cycle value of model supervision, a duration of model supervision, and a number of model supervision.
  • the cycle value, number of times and duration of model supervision can be any possible value, and the embodiments of the present application are not limited thereto.
  • the first indication information may specifically include one of the following:
  • the UE supervises the target model once every S time slots within the duration of T1 after receiving the first indication information.
  • the difference between the semi-continuous supervision mode and the periodic supervision mode is that during the semi-continuous supervision process, the UE can continue to receive the supervision indication from the network side device and perform additional supervision according to the supervision indication.
  • the semi-continuous supervision mode is equivalent to a combination of the periodic supervision mode and the aperiodic supervision mode.
  • the UE can supervise the target model within a period of time according to the first indication information, thereby being able to both supervise the operation of the target model and save UE power consumption.
  • the first indication information may further include at least one event that triggers model supervision.
  • the UE may supervise the target model at least once.
  • the UE may perform event monitoring.
  • the UE may perform at least one supervision on the target model.
  • the network side device can indicate the triggering method of the UE model supervision and the event that triggers the model supervision through the first indication information, the signaling overhead of the network side device can be saved.
  • the at least one event mentioned above includes any possible event that may affect the operating performance of the target model.
  • the at least one event may include at least one of the following:
  • the serving cell of the UE is switched
  • the area where the UE is located changes;
  • the reference signal configuration associated with the target model on the network side device changes
  • the service area corresponding to the UE changes;
  • the target bandwidth part (Bandwidth Part, BWP) is updated
  • the target measurement quantity satisfies the first condition.
  • the target measurement amount is related to at least one of the UE, the network side device and the target model
  • the target BWP may include at least one of a downlink BWP and an uplink BWP of the UE.
  • the switching of the serving cell of the UE may be: the serving cell of the UE is switched from cell C1 to cell C2.
  • the area where the UE is located refers to a physical area where the UE is located.
  • the area identifier of the area where the UE is located switches from A1 to A2, it indicates that the area where the UE is located changes.
  • the service area corresponding to the UE includes at least one of the following: a service area of a network operator corresponding to the UE, and a service area of a network equipment provider corresponding to the UE.
  • the change in the service area corresponding to the UE may include at least one of the following:
  • the UE enters the service area of a new network operator, such as the UE enters the service area of network operator 2 from the service area of network operator 1;
  • the UE enters the service area of a new network equipment provider, for example, the UE enters the service area of network equipment provider 2 from the service area of network equipment provider 1.
  • the operation of the target model may be different depending on the service area corresponding to the UE.
  • the UE can monitor the target model when the service area corresponding to the UE changes, the operation status of the target model in different service areas can be known.
  • the target measurement quantity satisfies the first condition and may include at least one of the following:
  • the movement speed of the UE is greater than or equal to the first threshold
  • the network synchronization error is greater than or equal to a second threshold
  • the channel state information (CSI) satisfies the second condition
  • the first reference signal related information satisfies a third condition
  • the output change value of the target model within M consecutive time units is greater than or equal to a third threshold, where M is a positive integer;
  • the timing error of the UE is greater than or equal to a fourth threshold.
  • the network synchronization error can be understood as follows: the network synchronization error is a deviation in the time of the network side device.
  • the UE timing error is an error in the timing of the UE side.
  • the network synchronization error may be an error in the timing of the sending and receiving nodes TRP of the network side device.
  • the traditional positioning method is to calculate the distance based on the time it takes for the signal to travel from the sending end to the receiving end.
  • the ranging accuracy will decrease, thereby affecting the positioning accuracy.
  • first threshold the second threshold, the third threshold and the fourth threshold are determined according to actual usage requirements and are not limited in the embodiments of the present application.
  • the CSI satisfies the second condition including at least one of the following: a rate of change of the CSI is greater than or equal to a fifth threshold; a correlation of the CSI within the first time interval is less than or equal to a sixth threshold.
  • the fifth threshold and the sixth threshold may be determined based on the impact of the CSI on the performance of the target model, such as based on the historical impact of the CSI on the performance of the target model.
  • the correlation of CSI at two moments may describe the similarity of CSI at two moments. It is generally believed that within a smaller time interval, CSI should be relatively similar.
  • the first time period may be any possible time period, such as the first time period is a time interval, such as L time slots.
  • the first reference signal related information satisfying the third condition may include at least one of the following:
  • the first reference signal receiving power (Reference Signal Receiving Power, RSRP) is less than or equal to a seventh threshold;
  • the first reference signal receiving quality (RSRQ) is less than or equal to an eighth threshold
  • a signal to interference plus noise ratio (SINR) of the first reference signal is less than or equal to a ninth threshold
  • the signal-to-noise ratio (Signal Noise Ratio, SNR) of the first reference signal is less than or equal to the tenth threshold.
  • the first reference signal may include downlink reference signals of different TRPs from the network side device.
  • the first reference signal can be any possible reference signal such as a positioning reference signal, a channel state information reference signal (CSI-RS), a synchronization signal and a broadcast channel block (Synchronization Signal and PBCH (Physical Broadcast Channel) block, SSB), etc.
  • CSI-RS channel state information reference signal
  • PBCH Physical Broadcast Channel
  • the average RSRP, minimum RSRP or maximum RSRP of these TRPs is less than or equal to the seventh threshold.
  • the RSRP of the TRP associated with the target model is less than or equal to the seventh threshold.
  • SINR or RSRQ refer to the relevant description of the first reference signal related information being RSRP. In order to avoid repetition, it will not be repeated here.
  • the target measurement amount may be detected and measured by the UE itself, or may be acquired by the UE from a network-side device, and may be specifically determined according to the target measurement amount.
  • the target measurement quantity may include: an instantaneous measurement quantity or a statistical measurement quantity.
  • the target measurement quantity when the target measurement quantity is a statistical measurement quantity, the target measurement quantity may specifically include:
  • a measurement that is statistically measured over time such as a measurement that is statistically obtained over a period of time
  • Spatial statistical measurements such as measurements taken by UEs at different geographical locations
  • the target measurement quantity when the target measurement quantity is a statistical measurement quantity, the target measurement quantity may specifically be an average of multiple measurement quantities.
  • the target measurement quantity may include at least one of the following:
  • N TRPs are TRPs associated with the target model.
  • the identifiers of the above N TRPs are configured by network-side equipment or agreed upon by protocol.
  • the above-mentioned steps 401 and 402 can be specifically implemented by the following steps 401a and 402a.
  • Step 401a The network side device sends a first signaling to the UE.
  • Step 402a The UE receives a first signaling.
  • the first signaling includes first indication information.
  • the UE may parse the first indication information in the first signaling, and supervise the target model according to the indication of the first indication information.
  • the UE may supervise the target model at least once immediately after receiving the first signaling, or may supervise the target model at least once after a preset time period after receiving the first signaling.
  • the first signaling may include any one of the following: downlink control information (Downlink Control Information, DCI), radio resource control (Radio Resource Control, RRC), long-term evolution positioning protocol (LTE (Long Term Evolution) Positioning Protocol, LPP), medium access control-control element (Medium Access Control)-Control Element, MAC-CE).
  • DCI Downlink Control Information
  • RRC Radio Resource Control
  • LTE Long Term Evolution
  • LPP Long Term Evolution Positioning Protocol
  • MAC-CE medium access control-control element
  • the first indication information is used to indicate a triggering method for indicating model supervision, and the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event; that is, the network side device instructs the UE to trigger the triggering method of model supervision, so the UE can directly trigger the model supervision of the target model according to the first indication information.
  • step 403 can be specifically implemented by the following step 403a.
  • the model supervision triggering method provided in the embodiment of the present application can also include the following steps 404 and 405.
  • Step 403a When a target event in at least one event occurs, the UE monitors the target model.
  • Step 404 The UE reports target information to the network side device.
  • Step 405 The network side device receives the target information.
  • the target information may include at least one of the following: a supervision result of a target model; or relevant information of a target event.
  • the supervision result of the target model may indicate the validity of the target model, that is, the supervision result may indicate whether the target model is valid or the validity period.
  • the network side device may judge the validity of the target model based on the supervision result of the target model.
  • the network side device may modify the model parameters of the target model based on the target information, that is, update the target model to improve the performance of the target model; or, the network side device may update and switch the target model to another model based on the target information; or, the network side device may modify the at least one event based on the target information.
  • the first threshold value, the second threshold value, etc. may be updated.
  • the relevant information of the target event may include any of the following:
  • the area identifier of the UE after the change
  • the network equipment vendor identifier of the changed service area of the UE The network equipment vendor identifier of the changed service area of the UE
  • the network operator identifier of the changed service area of the UE
  • the UE reports the serving cell identifier of the UE after the switching.
  • the terminal reports the changed area identifier of the UE, such as identifier A2;
  • the UE reports its measured movement speed
  • the UE reports the measured SINR; if the target event is: the SNR of the UE's measurement channel reference signal is less than or equal to the tenth threshold, the UE reports the measured SNR.
  • the UE reports the measured network synchronization error
  • the UE reports the measured CSI change rate or the CSI at P moments, where P is a positive integer. It can be understood that the CSI change rate is determined by the CSI at P moments.
  • the UE reports the correlation of the CSI in the first time period
  • the UE reports the measurement The first RSRP
  • the UE reports the measured first RSRQ
  • the UE reports the output change value of the target model in M consecutive time units
  • the target event is that the UE detects that the timing error of the UE is greater than or equal to the fourth threshold, the UE reports the measured UE timing error
  • the UE reports the changed configuration of the reference signal associated with the target model on the network side; for example, the changed configuration may be measured by the UE;
  • the UE reports the changed network operator's identifier
  • the UE reports the changed network equipment provider's identity
  • the UE reports the identifier of the changed target BWP. For example, if the uplink BWP of the UE changes, the UE can report the changed uplink BWP; if the downlink BWP of the UE changes, the UE can report the changed downlink BWP.
  • the network side device since the UE can report the supervision result and at least one of the relevant information of the target event triggering this supervision to the network side device, it is convenient for the network side device to update the model parameters of the target model and the event that triggers the model supervision, thereby improving the accuracy of subsequent triggered model supervision.
  • the target model as an AI positioning model as an example to exemplarily illustrate the model supervision triggering method provided in the embodiment of the present application.
  • an embodiment of the present application provides a positioning model supervision triggering method.
  • the network side device can indicate the triggering mode of the UE model supervision, including at least one of a periodic supervision mode, an aperiodic supervision mode, a semi-continuous supervision mode, and a supervision mode triggered by an event.
  • the positioning model supervision triggering method provided in this embodiment may include the following steps:
  • Step 50 The network side device sends first indication information for triggering model supervision to the UE;
  • Step 51 UE receives first indication information.
  • Step 52 The UE supervises the AI positioning model according to the first indication information.
  • the first indication information is used to indicate a triggering method of model supervision.
  • the first indication information is used to indicate at least one of: (1) a periodic supervision method, (2) a non-periodic supervision method, (3) a semi-continuous supervision method, and (4) a supervision method triggered by an event.
  • the first indication information when the first indication information indicates a periodic supervision method, includes: indicating the period of model supervision, such as performing model supervision once every N time slots; such as the first indication information instructs the UE to perform model supervision once every N time slots.
  • the first indication information indicates the semi-continuous supervision mode
  • the first indication information is included in the first signaling sent by the network side device, such as DCI, RRC, LPP, MAC CE signaling, etc.
  • the first indication information may include any of the following items: the number of model supervisions, the duration of model supervisions, the number of model supervisions and the period value, the duration and the period value of model supervisions.
  • the first indication information indicates that the UE performs model supervision once every N time slots, for a total of K supervisions or for a duration of L, where K is a positive integer and L is greater than 0.
  • the first indication information indicates non-periodic supervision
  • the first indication information is included in the first signaling sent by the network side device, such as DCI, RRC, LPP, MAC CE signaling, etc.
  • the UE may immediately perform a model supervision on the AI positioning model, or may perform a model supervision on the AI positioning model after a preset time period after receiving the first indication information.
  • the network side device can send the first indication information to the UE when the UE is required to perform model supervision on the AI positioning model, so that the UE can supervise the AI positioning model.
  • the first indication information may include at least one event that triggers model supervision, and the at least one event includes at least one of the following:
  • Event 1 UE switches cells, such as the serving cell switches from C1 to C2;
  • Event 2 The area where the UE is located changes, such as the area ID switches from A1 to A2;
  • Event 3 The UE moving speed exceeds the threshold T1, i.e., the first threshold;
  • Event 4 The SINR of the channel measured by the UE is less than or equal to the threshold T2;
  • Event 5 The SNR of the channel measured by the UE is less than or equal to the threshold T10;
  • the threshold value T2 and the threshold value T10 may be the same or different.
  • Event 5 The UE detects that the network synchronization error is greater than or equal to the threshold T3;
  • Event 6 The change rate of the CSI obtained by the UE is greater than or equal to the threshold T4;
  • Event 7 The correlation of the CSI acquired by the UE within a time interval is less than or equal to the threshold T5;
  • Event 8 RSRP measured by the UE is less than or equal to threshold T6;
  • Event 9 The RSRQ measured by the UE is less than or equal to the threshold T7;
  • Event 10 The output change value of the AI positioning model obtained by the UE in M consecutive time units is greater than or equal to the threshold T8;
  • Event 11 The UE detects that the timing error of the UE is greater than or equal to the threshold T9;
  • Event 12 The UE detects that the reference signal configuration associated with the AI positioning model on the network side device has changed;
  • Event 13 The UE enters the service area of a new network operator, that is, the UE enters the service area of one network operator to the service area of another network operator;
  • Event 14 The UE enters the service area of a new network equipment provider, that is, the UE enters the service area of one network equipment provider to the service area of another network equipment provider;
  • Event 15 The uplink and/or downlink BWP of the terminal changes.
  • the UE may determine that when any of the at least one event mentioned above occurs, the AI positioning model is supervised once.
  • the UE may perform model supervision on the AI model. Then, the UE may report relevant information of the target event to the network side device. For example:
  • the UE reports the serving cell identifier of the UE after the switching.
  • the terminal reports the changed area identifier of the UE, such as identifier A2;
  • the UE reports its measured movement speed
  • the UE reports the measured SINR; if the target event is: the SNR of the UE's measurement channel reference signal is less than or equal to the tenth threshold, the UE reports the measured SNR.
  • the UE reports the measured network synchronization error
  • the UE reports the measured CSI change rate or the CSI at P moments, where P is a positive integer. It can be understood that the CSI change rate is determined by the CSI at P moments.
  • the UE reports the correlation of the CSI in the first time period
  • the UE reports the measurement The first RSRQ
  • the UE reports the measured first RSRQ
  • the UE reports the output change value of the target model in M consecutive time units
  • the target event is that the UE detects that the timing error of the UE is greater than or equal to the fourth threshold, the UE reports the measured UE timing error
  • the UE reports the changed configuration of the reference signal associated with the target model on the network side; for example, the changed configuration may be measured by the UE;
  • the UE reports the changed network operator's identifier
  • the UE reports the changed network equipment provider's identity
  • the UE reports the identifier of the changed target BWP. For example, if the uplink BWP of the UE changes, the UE can report the changed uplink BWP; if the downlink BWP of the UE changes, the UE can report the changed downlink BWP.
  • the UE can report the supervision results of the AI model and information related to the target event to the network side device.
  • the measurement quantities in the above events 3 to 11 may include instantaneous values and statistical values, such as the average of multiple measurement values.
  • the multiple measurement values may be measurement values obtained by measuring at different times, different UEs or different areas.
  • the measurement quantities in the above events 3 to 11 may include an average of N TRP measurement quantities, a minimum measurement quantity among the N TRPs, and a maximum measurement quantity among the N TRPs.
  • the N TRPs are TRPs associated with the AI positioning model.
  • the TRP identifier associated with the AI positioning model can be configured in advance by the network side device.
  • the AI positioning model triggering method provided in the embodiment of the present application can provide a triggering method for triggering supervision, so that the performance of the supervision model can be supervised based on the indicated supervision method to ensure the positioning accuracy of the AI positioning model.
  • the model supervision triggering method provided in the embodiment of the present application can be executed by a model supervision triggering device.
  • the model supervision triggering device executing the model supervision triggering method is taken as an example to illustrate the model supervision triggering device provided in the embodiment of the present application.
  • FIG. 5 shows a structural schematic diagram of the model supervision trigger device provided by the embodiment of the present application.
  • the model supervision trigger device 500 provided by the embodiment of the present application may include: a receiving module 501 and a processing module 502.
  • the receiving module 501 is used to receive first indication information sent by a network side device, where the first indication information is used to indicate a triggering mode of model supervision;
  • the processing module 502 is used to supervise the target model according to the first indication information received by the receiving module 501;
  • the triggering method indicated by the first indication information includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • the first indication information when the triggering method indicated by the first indication information includes a periodic supervision method, the first indication information also includes a periodic value of the model supervision.
  • the first indication information when the triggering method indicated by the above-mentioned first indication information includes a semi-continuous supervision method, the first indication information also includes at least one of the following: a periodic value of model supervision, a duration of model supervision, and a number of model supervision times.
  • the first indication information when the triggering mode indicated by the first indication information includes an event-triggered supervision mode, the first indication information also includes at least one event that triggers model supervision.
  • the at least one event includes at least one of the following:
  • the serving cell of the UE is switched
  • the area where the UE is located changes;
  • the reference signal configuration associated with the target model on the network side device changes
  • the service area corresponding to the UE changes;
  • the target partial bandwidth BWP is updated
  • the target measurement quantity satisfies the first condition
  • the target measurement amount is related to at least one of the UE, the network side device and the target model
  • the target BWP includes at least one of a downlink BWP and an uplink BWP of the UE.
  • the target measurement quantity satisfies the first condition including at least one of the following:
  • the movement speed of the UE is greater than or equal to the first threshold
  • the network synchronization error is greater than or equal to a second threshold
  • the first reference signal related information satisfies a third condition
  • the output change value of the target model within M consecutive time units is greater than or equal to a third threshold, where M is a positive integer;
  • the timing error of the UE is greater than or equal to a fourth threshold.
  • the CSI satisfies the second condition including at least one of the following:
  • the change rate of the CSI is greater than or equal to a fifth threshold
  • the correlation of the CSI in the first time interval is less than or equal to a sixth threshold.
  • the first reference signal related information satisfies the third condition including at least one of the following:
  • the first reference signal received power RSRP is less than or equal to a seventh threshold
  • the first reference signal reception quality RSRQ is less than or equal to an eighth threshold
  • a signal to interference plus noise ratio SINR of the first reference signal is less than or equal to a ninth threshold
  • a signal-to-noise ratio SNR of the first reference signal is less than or equal to a tenth threshold.
  • the service area corresponding to the UE includes at least one of the following:
  • the target measurement quantity includes: an instantaneous measurement quantity or a statistical measurement quantity.
  • the target measurement quantity includes at least one of the following:
  • N TRPs are TRPs associated with the target model.
  • the identifiers of the above-mentioned N TRPs are configured by network-side equipment or agreed upon by protocol.
  • the processing module 502 is specifically configured to supervise the target model when a target event in at least one event occurs;
  • the model supervision triggering device may further include a sending module, which is used to report target information to the network side device after supervising the target model.
  • the target information includes at least one of the following: the supervision result of the target model; and relevant information of the target event.
  • the relevant information of the target event includes any of the following:
  • the area identifier of the UE after the change
  • the network equipment vendor identifier of the changed service area of the UE The network equipment vendor identifier of the changed service area of the UE
  • the network operator identifier of the changed service area of the UE
  • the triggering method includes a semi-continuous supervision method or a non-periodic supervision method
  • the receiving module 501 is specifically configured to receive a first signaling sent by a network side device, where the first signaling includes first indication information.
  • the first signaling includes any one of the following: DCI, RRC, LPP, and MAC-CE.
  • the target model is a positioning model.
  • the model supervision triggering device can receive the first indication information sent by the network side device for indicating the triggering method for indicating model supervision, and the triggering method includes at least one of the following: periodic supervision method, semi-continuous supervision method, non-periodic supervision method, and supervision method triggered by an event; that is, the network side device instructs the model supervision triggering device to trigger the triggering method of model supervision, the model supervision triggering device can directly trigger the model supervision of the target model according to the first indication information.
  • the embodiment of the present application provides a model supervision triggering device
  • FIG6 shows a schematic diagram of the structure of the model supervision triggering device provided by the embodiment of the present application.
  • the model supervision triggering device 600 provided by the embodiment of the present application may include: a sending module 601.
  • the sending module 601 is used to send first indication information to the UE, and the first indication information is used to indicate the triggering mode of the model supervision; the first indication information is used for the UE to supervise the target model;
  • the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • the first indication information when the triggering method indicated by the first indication information includes a periodic supervision method, the first indication information also includes a periodic value of model supervision.
  • the first indication information when the triggering method indicated by the first indication information includes a semi-continuous supervision method, the first indication information also includes at least one of the following: a period value of model supervision, a duration of model supervision, and a number of model supervision times.
  • the first indication information when the triggering mode indicated by the first indication information includes an event-triggered supervision mode, the first indication information also includes at least one event that triggers model supervision.
  • the at least one event includes at least one of the following:
  • the serving cell of the UE is switched
  • the area where the UE is located changes;
  • the reference signal configuration associated with the target model on the network side device changes
  • the service area corresponding to the UE changes;
  • the target BWP is updated
  • the target measurement quantity satisfies the first condition
  • the target measurement amount is related to at least one of the UE, the network side device and the target model
  • the target BWP includes at least one of a downlink BWP and an uplink BWP of the UE.
  • the target measurement quantity satisfies the first condition including at least one of the following:
  • the movement speed of the UE is greater than or equal to the first threshold
  • the network synchronization error is greater than or equal to a second threshold
  • the first reference signal related information satisfies a third condition
  • the output change value of the target model within M consecutive time units is greater than or equal to a third threshold, where M is a positive integer;
  • the timing error of the UE is greater than or equal to a fourth threshold.
  • the CSI satisfies the second condition including at least one of the following:
  • the change rate of the CSI is greater than or equal to a fifth threshold
  • the correlation of the CSI in the first time interval is less than or equal to a sixth threshold.
  • the first reference signal related information satisfies the third condition, including at least one of the following:
  • the first RSRP is less than or equal to a seventh threshold
  • the first RSRQ is less than or equal to an eighth threshold
  • the SINR of the first reference signal is less than or equal to a ninth threshold
  • the SNR of the first reference signal is less than or equal to a tenth threshold.
  • the service area corresponding to the UE includes at least one of the following:
  • the target measurement quantity includes: an instantaneous measurement quantity or a statistical measurement quantity.
  • the target measurement quantity includes at least one of the following:
  • N TRPs are TRPs associated with the target model.
  • the model supervision triggering device 600 may further include a receiving module
  • the receiving module is used to receive target information reported by the UE after the sending module 601 sends the first indication information to the UE; the target information includes at least one of the following: the supervision result of the target model; and the relevant information of the target event.
  • the triggering method includes a semi-continuous supervision method or a non-periodic supervision method
  • the network side device sends first indication information to the UE, including:
  • the network side device sends a first signaling to the UE, where the first signaling includes first indication information.
  • the first signaling includes any one of the following: DCI, RRC, LPP, and MAC-CE.
  • the target model is a positioning model.
  • the model supervision triggering device 600 since the model supervision triggering device can send a first indication information to the UE for indicating a triggering method for indicating model supervision, and the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event; that is, the model supervision triggering device instructs the UE to trigger the triggering method of model supervision, the UE can directly trigger the model supervision of the target model according to the first indication information.
  • the model supervision trigger device in the embodiment of the present application can be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip.
  • the electronic device can be a terminal, or it can be other devices other than a terminal.
  • the terminal can include but is not limited to the type of terminal 11 listed in FIG. 1 above, and other devices can be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiment of the present application.
  • the model supervision triggering device provided in the embodiment of the present application can implement each process implemented by the UE in the method embodiment of Figure 4 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • the embodiment of the present application further provides a communication device 700, including a processor 701 and a memory 702, the memory 702 storing programs or instructions that can be run on the processor 701.
  • the communication device 700 is a terminal
  • the program or instruction is executed by the processor 701 to implement the various steps of the above-mentioned model supervision triggering method embodiment, and can achieve the same technical effect.
  • the communication device 700 is a network side device
  • the program or instruction is executed by the processor 701 to implement the various steps executed by the network side device in the above-mentioned model supervision triggering method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • the communication device When 700 is a UE, the program or instruction is executed by the processor 701 to implement the various steps executed by the UE in the above-mentioned model supervision triggering method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • the embodiment of the present application also provides a UE, including a processor and a communication interface, wherein the communication interface is used to receive a first indication message sent by a network side device, and the first indication message is used to indicate a triggering mode of model supervision; the processor is used to supervise the target model according to the first indication message; wherein the triggering mode includes at least one of the following: a periodic supervision mode, a semi-continuous supervision mode, a non-periodic supervision mode, and a supervision mode triggered by an event.
  • This UE embodiment corresponds to the above-mentioned UE method embodiment, and each implementation process and implementation mode of the above-mentioned method embodiment can be applied to the terminal embodiment, and can achieve the same technical effect.
  • Figure 8 is a schematic diagram of the hardware structure of a UE that implements an embodiment of the present application.
  • the UE 800 includes but is 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 at least some of the components in the processor 810.
  • UE800 can also include a power supply (such as a battery) for supplying power to each component, and the power supply can be logically connected to the processor 810 through a power management system, so as to implement functions such as charging, discharging, and power consumption management through the power management system.
  • a power supply such as a battery
  • the UE structure shown in FIG8 does not constitute a limitation on the terminal, and the UE may include more or fewer components than shown in the figure, or combine certain components, or arrange components differently, which will not be described in detail here.
  • the radio frequency unit 801 is used to receive first indication information sent by the network side device, where the first indication information is used to indicate a triggering mode of model supervision;
  • the processor 810 is configured to supervise the target model according to the first indication information received by the radio frequency unit 801;
  • the triggering method includes at least one of the following: a periodic supervision method, a semi-continuous supervision method, a non-periodic supervision method, and a supervision method triggered by an event.
  • the first indication information when the triggering method indicated by the first indication information includes a periodic supervision method, the first indication information also includes a periodic value of the model supervision.
  • the first indication information when the triggering method indicated by the above-mentioned first indication information includes a semi-continuous supervision method, the first indication information also includes at least one of the following: a periodic value of model supervision, a duration of model supervision, and a number of model supervision times.
  • the first indication information when the triggering mode indicated by the first indication information includes an event-triggered supervision mode, the first indication information further includes at least one event that triggers model supervision;
  • the at least one event includes at least one of the following:
  • the serving cell of the UE is switched
  • the area where the UE is located changes;
  • the reference signal configuration associated with the target model on the network side device changes
  • the service area corresponding to the UE changes;
  • the target partial bandwidth BWP is updated
  • the target measurement quantity satisfies the first condition
  • the target measurement amount is related to at least one of the UE, the network side device and the target model
  • the target BWP includes at least one of a downlink BWP and an uplink BWP of the UE.
  • the target measurement quantity satisfies the first condition including at least one of the following:
  • the movement speed of the UE is greater than or equal to the first threshold
  • the network synchronization error is greater than or equal to a second threshold
  • the first reference signal related information satisfies a third condition
  • the output change value of the target model within M consecutive time units is greater than or equal to a third threshold, where M is a positive integer;
  • the timing error of the UE is greater than or equal to a fourth threshold.
  • the CSI satisfies the second condition including at least one of the following:
  • the change rate of the CSI is greater than or equal to a fifth threshold
  • the correlation of the CSI in the first time interval is less than or equal to a sixth threshold.
  • the first reference signal related information satisfies the third condition including at least one of the following:
  • the first reference signal received power RSRP is less than or equal to a seventh threshold
  • the first reference signal reception quality RSRQ is less than or equal to an eighth threshold
  • a signal to interference plus noise ratio SINR of the first reference signal is less than or equal to a ninth threshold
  • a signal-to-noise ratio SNR of the first reference signal is less than or equal to a tenth threshold.
  • the service area corresponding to the UE includes at least one of the following:
  • the target measurement quantity includes: an instantaneous measurement quantity or a statistical measurement quantity.
  • the target measurement quantity includes at least one of the following:
  • N TRPs are TRPs associated with the target model.
  • the identifiers of the above-mentioned N TRPs are configured by network-side equipment or agreed upon by protocol.
  • the processor 810 is specifically configured to supervise the target model when a target event in at least one event occurs;
  • the radio frequency unit 801 is further used to report target information to the network side device after monitoring the target model.
  • the target information includes at least one of the following: the monitoring result of the target model; and the relevant information of the target event.
  • the relevant information of the target event includes any of the following:
  • the area identifier of the UE after the change
  • the network equipment vendor identifier of the changed service area of the UE The network equipment vendor identifier of the changed service area of the UE
  • the network operator identifier of the changed service area of the UE
  • the triggering method includes a semi-continuous supervision method or a non-periodic supervision method
  • the radio frequency unit 801 is specifically configured to receive a first signaling sent by a network side device, where the first signaling includes first indication information.
  • the first signaling includes any one of the following: DCI, RRC, LPP, and MAC-CE.
  • the target model is a positioning model.
  • the UE can receive the first indication information sent by the network side device to indicate the triggering method for indicating model supervision, and the triggering method includes at least one of the following: periodic supervision method, semi-continuous supervision method, non-periodic supervision method, and supervision method triggered by an event; that is, the network side device instructs the UE to trigger the triggering method of model supervision, the UE can directly trigger the model supervision of the target model according to the first indication information.
  • the input unit 804 may include a graphics processing unit (GPU) 8041 and a microphone 8042, and the graphics processor 8041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode.
  • the display unit 806 may include a display panel 8061, which may be in the form of a liquid crystal display, an organic light emitting diode, etc. The display panel 8061 is configured in a certain manner.
  • the user input unit 807 includes at least one of a touch panel 8071 and other input devices 8072.
  • the touch panel 8071 is also called a touch screen.
  • the touch panel 8071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 8072 may include but are not limited to a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, and a joystick, which will not be described in detail here.
  • the radio frequency unit 801 after receiving downlink data from the network side device, can transmit the data to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device.
  • the radio frequency unit 801 includes but is not limited to an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
  • the memory 809 can be used to store software programs or instructions and various data.
  • the memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
  • the memory 809 may include a volatile memory or a non-volatile memory, or the memory 809 may include both volatile and non-volatile memories.
  • the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
  • the volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM).
  • the memory 809 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.
  • the processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 810.
  • the embodiment of the present application also provides a network side device, including a processor and a communication interface, wherein the processor is used to obtain first indication information, and the communication interface is used to send the first indication information to the UE, and the first indication information is used to indicate the triggering method of model supervision; the first indication information is used for the UE to supervise the target model; wherein the triggering method includes at least one of the following: periodic supervision method, semi-continuous supervision method, non-periodic supervision method, and supervision method triggered by an event.
  • This network side device embodiment corresponds to the above-mentioned network side device method embodiment, and each implementation process and implementation method 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 91, a radio frequency device 92, a baseband device 93, a processor 94 and a memory 95.
  • the antenna 91 is connected to the radio frequency device 92.
  • the radio frequency device 92 receives information through the antenna 91 and sends the received information to the baseband device 93 for processing.
  • the baseband device 93 processes the information to be sent and sends it to the radio frequency device 92.
  • the radio frequency device 92 processes the received information and sends it out through the antenna 91.
  • the method executed by the network-side device in the above embodiment may be implemented in the baseband device 93, which includes a baseband processor.
  • the baseband device 93 may include, for example, at least one baseband board, on which a plurality of chips are arranged, as shown in FIG. 9 , wherein one of the chips is, for example, a baseband processor, which is connected to the memory 95 through a bus interface to call a program in the memory 95 and execute the network device operations shown in the above method embodiment.
  • the network side device may also include a network interface 96, which is, for example, a common public wireless interface (common public radio interface, CPRI).
  • a network interface 96 which is, for example, a common public wireless interface (common public radio interface, CPRI).
  • the network side device 900 of the embodiment of the present invention also includes: instructions or programs stored in the memory 95 and executable on the processor 94.
  • the processor 94 calls the instructions or programs in the memory 95 to execute the methods executed by the modules shown in Figure 5 and achieve the same technical effect. To avoid repetition, it will not be described here.
  • the embodiment of the present application further provides a network side device.
  • the network side device 1000 includes: a processor 1001, a network interface 1002, and a memory 1003.
  • the network interface 1002 is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 1000 of the embodiment of the present invention also includes: instructions or programs stored in the memory 1003 and executable on the processor 1001.
  • the processor 1001 calls the instructions or programs in the memory 1003 to execute the method executed by each module shown in Figure 6 and achieves the same technical effect. To avoid repetition, it will not be repeated here.
  • An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored.
  • a program or instruction is stored.
  • the various processes of the above-mentioned model supervision trigger method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.
  • An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned model supervision triggering method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
  • the embodiment of the present application further provides a computer program/program product, which is stored in a storage medium, and is executed by at least one processor to implement the various processes of the above-mentioned model supervision triggering method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • An embodiment of the present application also provides a communication system, including: a UE and a network side device, wherein the UE can be used to execute the steps of the UE method embodiment described above, and the network side device can be used to execute the steps of the network side method embodiment described above.
  • the technical solution of the present application can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM/RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.
  • a storage medium such as ROM/RAM, a magnetic disk, or an optical disk
  • a terminal which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.

Landscapes

  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统，属于通信技术领域，本申请实施例的模型监督触发方法包括：用户设备UE接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述UE根据所述第一指示信息，对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。

Description

模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统
相关申请的交叉引用
本申请主张在2023年12月09日在中国提交申请号为202211586245.8的中国专利的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于通信技术领域，具体涉及一种模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统。
背景技术
人工智能(Artificial Intelligence，AI)模型在各个领域获得了广泛的应用。
示例性地，将人工智能模型融入无线通信网络中，可以显著提升无线通信网络的吞吐量、时延以及用户容量等技术指标。显而易见，人工智能是未来的无线通信网络的重要任务。其中，可以通过对人工智能模型进行监督，以提升人工智能模型的模型性能。
如此，如何触发人工智能模型的监督是亟待解决的问题。
发明内容
本申请实施例提供一种模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统，能够解决如何触发人工智能模型监督的问题。
第一方面，提供了一种模型监督触发方法，应用于UE，该方法包括：用户设备UE接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述UE根据所述第一指示信息，对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第二方面，提供了一种模型监督触发装置，该模型监督触发装置可以包括：接收模块和处理模块。接收模块，用于接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；处理模块，用于根据所述第一指示信息，对目标模型进行监督。其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第三方面，提供了一种模型监督触发方法，应用于网络侧设备，该方法包括：网络侧设备向UE发送第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述第一指示信息用于所述UE对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第四方面，提供了一种模型监督触发装置，该装置包括：发送模块。发送模块，用于向UE发送第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述第一指示信息用于所述UE对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第五方面，提供了一种UE，该UE包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面，提供了一种UE，包括处理器及通信接口，其中，所述通信接口用于接 收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述处理器用于根据所述第一指示信息，对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第七方面，提供了一种网络侧设备，该网络侧设备包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第三方面所述的方法的步骤。
第八方面，提供了一种网络侧设备，包括处理器及通信接口，其中，所述处理器用于获取第一指示信息，所述通信接口用于向UE发送所述第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述第一指示信息用于所述UE对目标模型进行监督；
其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
第九方面，提供了一种通信系统，包括：UE及网络侧设备，所述UE可用于执行如第一方面所述的模型监督触发方法的步骤，所述网络侧设备可用于执行如第三方面所述的模型监督触发方法的步骤。或者包括：如第二方面所述的模型监督触发装置和如第四方面所述的模型监督触发装置，其中，如第二方面所述的模型监督触发装置用于执行如第三方面所述的模型监督触发方法的步骤，如第四方面所述的模型监督触发装置用于执行如第三方面所述的模型监督触发方法的步骤。
第十方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤，或者实现如第三方面所述的方法的步骤。
第十一方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第第一方面所述的方法，或实现如第三方面所述的方法。
第十二方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面所述的模型监督触发方法的步骤。
在本申请实施例中，UE接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述UE根据所述第一指示信息，对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。通过该方案，由于UE可以接收网络侧设备发送的用于指示用于指示模型监督的触发方式的第一指示信息，且该触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式；即网络侧设备指示UE触发模型监督的触发方式，因此UE可以直接根据第一指示信息触发目标模型的模型监督。
附图说明
图1是本申请实施例提供的一种无线通信系统的架构示意图；
图2是相关技术提供的一种神经网络的结构示意图；
图3是相关技术提供的一种神经元的结构示意图；
图4是本申请实施例提供的一种模型监督触发方法的流程示意图；
图5是本申请实施例提供的一种模型监督触发装置的结构示意图之一；
图6是本申请实施例提供的一种模型监督触发装置的结构示意图之二；
图7是本申请实施例提供的一种通信设备的结构示意图；
图8是本申请实施例的一种UE的硬件结构；
图9是本申请实施例提供的一种网络侧设备的硬件结构示意图之一；
图10是本申请实施例提供的一种网络侧设备的硬件结构示意图之二。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio，NR)系统，并且在以下大部分描述中使用NR术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6th Generation，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中，终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备，如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer，PC)、柜员机或者自助机等终端侧设备，可穿戴式设备包括：智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是，在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备，其中，接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等，基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station，BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set，BSS)、扩展服务集(Extended Service Set，ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point，TRP)或所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系统中的基站为例进行介绍，并不限定基站的具体类型。核心网设备可以包含但不限于如下至少一项：核心网节点、核心网功能、移动管理实体(Mobility Management Entity，MME)、接入移动管理功能(Access and Mobility  Management Function，AMF)、会话管理功能(Session Management Function，SMF)、用户平面功能(User Plane Function，UPF)、策略控制功能(Policy Control Function，PCF)、策略与计费规则功能单元(Policy and Charging Rules Function，PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function，EASDF)、统一数据管理(Unified Data Management，UDM)，统一数据仓储(Unified Data Repository，UDR)、归属用户服务器(Home Subscriber Server，HSS)、集中式网络配置(Centralized network configuration，CNC)、网络存储功能(Network Repository Function，NRF)，网络开放功能(Network Exposure Function，NEF)、本地NEF(Local NEF，或L-NEF)、绑定支持功能(Binding Support Function，BSF)、应用功能(Application Function，AF)等。需要说明的是，在本申请实施例中仅以NR系统中的核心网设备为例进行介绍，并不限定核心网设备的具体类型。
下面对本申请实施例提供的一种模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统中涉及的一些概念和/或术语做一下解释说明
一、人工智能(AI)
AI是将人工智能融入无线通信网络，显著提升吞吐量、时延以及用户容量等技术指标是未来的无线通信网络的重要任务。AI模块有多种实现方式，例如神经网络、决策树、支持向量机、贝叶斯分类器等。本申请实施例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。
示例性地，如图2所示，为一个神经网络的示意图。该神经网络包括输入层、隐层和输出层，X1,X2,…,Xn为输入，Y为输出。
其中，神经网络由神经元组成，如图3所示，为神经元的示意图。其中，a1,a2,…aK为输入，w为权值(乘性系数)，b为偏置(加性系数)，z＝a1w1+…+akwk+…+aKwK+b，σ(z)为激活函数。通常，激活函数包括Sigmoid、tanh、ReLU(Rectified Linear Unit、线性整流函数、修正线性单元)等。
神经网络的参数通过梯度优化算法进行优化。梯度优化算法是一类最小化或者最大化目标函数(也可以称为损失函数)的算法，而目标函数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，可以构建一个神经网络模型f(.)，有了模型后，可以根据输入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。找到合适的W,b使上述的损失函数的值达到最小，损失值越小，则说明模型越接近于真实情况。
目前常见的优化算法，为基于BP(error Back Propagation，误差反向传播)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。
常见的优化算法有梯度下降(Gradient Descent)、随机梯度下降(Stochastic Gradient Descent，SGD)、mini-batch gradient descent(小批量梯度下降)、动量法(Momentum)、Nesterov(发明者的名字，具体为带动量的随机梯度下降)、Adagrad(ADAptive GRADient descent，自适应梯度下降)、Adadelta、RMSprop(root mean square prop，均方根误差降速)、Adam(Adaptive Moment Estimation，自适应动量估计)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。
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一种模型监督触发方法、装置、UE、网络侧设备、可读存储介质及通信系统进行详细地说明。
本申请实施例提供一种模型监督触发方法，图4示出了本申请实施例提供的一种模型监督触发方法的流程示意图，如图4所示，本申请实施例提供模型监督触发方法可以包括下述的步骤401至403。
步骤401、网络侧设备向UE发送第一指示信息。
步骤402、UE接收第一指示信息。
其中，第一指示信息可以用于指示模型监督的触发方式。
可选地，上述第一指示信息指示的触发方式可以包括以下至少之一：方式1，周期性监督方式、方式2，半持续监督方式、方式3，非周期性监督方式、方式4，由事件触发的监督方式。
需要说明的是，对于模型监督在UE侧的情况，网络侧设备可以通过第一指示信息指示UE模型监督的触发方式，以便于UE可以根据网络侧设备指示的触发方式触发模型监督。
步骤403、UE根据第一指示信息，对目标模型进行监督。
本申请实施例中，在方式1中，UE可以周期性地对目标模型进行监督；在方式2和方式3中，UE可以在接收到第一指示信息之后，对目标模型进行至少一次监督；在方式4中，UE可以在发生特定事件的情况下，对目标模型进行监督。
可以理解，在方式1和方式4中，UE可以提前获取触发模型监督的触发条件。具体的，在方式1中，UE可以提前获取触发模型监督的周期值，即提前获知监督周期；在方式4中，UE需要提前获取触发模型监督的事件。
可选地，触发模型监督的周期值和事件可以是协议约定的、预设的或者网络侧设备配置的。
需要说明的是，UE对目标模型进行监督可以理解为：在目标模型运动的过程中，UE对目标模型的运行情况进行监督，以获知目标模型的运行情况和性能。可选地，可以根据对目标模型的监督结果对目标模型的性能进行优化。
可选地，UE对目标模型进行监督之后，可以将监督结果上报至网络侧设备。例如，UE每对目标模型进行一次监督，即将该次监督的监督结果上报至网络侧设备；或者，UE可以将多次监督得到的监督结果打包上报至网络侧设备。
可选地，目标模型可以为定位模型，如AI定位模型。可以理解，AI定位模型用于定位。
下面分别以方式1～4为例对本申请实施例提供的模型监督触发方法进行详细说明。
方式1
可选地，在第一指示信息指示的触发方式包括周期性监督方式的情况下，第一指示信息还可以包括模型监督的周期值。
可选地，模型监督的周期值的单位可以为：OFDM符号、时隙、子帧、帧、毫秒、秒、分钟、小时、天等。
例如，模型监督的周期可以为：至少一个OFDM符号、至少一个时隙、至少一个子帧、至少一帧、至少一毫秒、至少一秒、至少一分钟、至少一小时、至少一天等。
例如，周期值为S个时隙，S为正整数；如此，UE在接收到第一指示信息之后，可以每S个时隙对目标模型执行一次模型监督。
如此，由于可以在第一指示信息中指示执行模型监督的周期值，因此可以节省网 络侧设备的信令开销。
方式2
可选地，在第一指示信息指示的触发方式包括所述半持续监督方式的情况下，第一指示信息还可以包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
其中，模型监督的周期值、次数和时长可以为任意可能的值，本申请实施例不作限定。
可选地，第一指示信息中具体可以包括以下之一：
模型监督的周期值；
模型监督的时长；
模型监督的次数；
模型监督的周期值和时长；
模型监督的周期值和次数；
模型监督的周期值、次数和时长。
例如，假设第一指示信息中包括S个时隙、时长T1；那么UE在接收到第一指示信息之后的时长T1内，每S个时隙对目标模型进行一次监督。
又例如，假设第一指示信息中包括周期值：S个时隙、K1次；那么UE在接收到第一指示信息之后，可以每S个时隙对目标模型进行一次监督，共监督K1次。
需要说明的是，半持续监督方式与周期性监督方式的区别是：在半持续监督过程中，UE可以继续接收网络侧设备的监督指示，并根据该监督指示进行额外的监督。其中，当第一指示信息中仅包括监督周期时，半持续监督方式相当于周期性监督方式和非周期性监督方式的组合。
如此，由于第一指示信息中可以包括模型监督的周期值、次数和时长中的至少一项，因此UE可以根据第一指示信息在一段时间内对目标模型进行监督，从而既能够实现对目标模型的运行情况进行监督；又能够节省UE的功耗。
方式3
可选地，在第一指示信息指示的触发方式包括所述事件触发监督方式的情况下，第一指示信息中还可以包括触发模型监督的至少一事件。
本申请实施例中，当上述至少一个事件中的任一事件发生时，UE均可以对目标模型进行至少一次监督。
本申请实施例中，UE在接收到第一指示信息之后，可以进行事件监测，在监测到上述至少一个事件中的任一事件发生时，UE可以目标模型进行至少一次监督。
如此，由于网络侧设备可以通过第一指示信息指示UE模型监督的触发方式和触发模型监督的事件，因此一方面可以节省网络侧设备的信令开销。
本申请实施例中，上述至少一个事件包括可能影响目标模型的运行性能的任意可能的事件。
可选地，上述至少一个事件可以包括以下至少之一：
UE的服务小区发生切换；
UE所在区域发生变化；
网络侧设备与目标模型关联的参考信号配置发生变化；
UE对应的服务区域发生变化；
目标部分带宽(Bandwidth Part，BWP)发生更新；
目标测量量满足第一条件.
其中，目标测量量与UE、网络侧设备和目标模型中的至少之一相关；
目标BWP可以包括UE的下行BWP和上行BWP中的至少之一。
可选地，UE的服务小区发生切换可以为：UE的服务小区从小区C1切换为小区C2。
可选地，UE所在区域是指，UE所在的物理区域。可选地，当UE从区域1进入到区域2，或者UE所在区域的区域标识从A1切换为A2时，表示UE所在区域发生变化。
可选地，UE对应的服务区域包括以下至少之一：UE对应的网络运营商的服务区域、UE对应的网络设备商的服务区域。
可选地，UE对应的服务区域发生变化可以包括以下至少之一：
UE进入新的网络运营商的服务区域，如UE从网络运营上1的服务区域进入到网络运营上2的服务区域；
UE进入新的网络设备商的服务区域，如，UE从网络设备商1的服务区域进入到网络设备商2的服务区域。
需要说明的是，UE对应的服务区域不同，目标模型的运行情况可能存在差异。
如此，由于UE可以在UE对应的服务区域发生变化时，对目标模型进行监督，从而可以可以获知目标模型在不同服务区域下的运行情况。
可选地，目标测量量满足第一条件可以包括以下至少之一：
UE的运动速度大于或等于第一阈值；
网络同步误差大于或等于第二阈值；
信道状态信息(Channel State Information，CSI)满足第二条件；
第一参考信号相关信息满足第三条件；
目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
UE的定时误差大于或等于第四阈值。
本申请实施例中，网络同步误差可以理解为：网络同步误差是网络侧设备的时间存在偏差。UE定时误差是UE侧的定时存在误差。
具体的，网络同步误差可以为网络侧设备的发送与接收节点TRP的定时存在误差。以目标模型为定位模型为例，传统定位方法是按信号从发送端到接收端的经过的时间去计算距离，当由多个TRP进行信号收发时，可以根据三角定位原理去计算位置。此时，各TRP接收的信号在空中经过的时间＝接收的时间t1-发送的时间t0，当任意一个时间存在误差的时候，都会导致测距精度下降，进而影响定位精度。
需要说明的是，第一阈值、第二阈值、第三阈值和第四阈值的具体数值根据实际使用需求确定，本申请实施例不作限定。
可选地，CSI满足第二条件包括以下至少之一：CSI的变化率大于或等于第五阈值；CSI在第一时间间隔内的相关性小于或等于第六阈值。
第五阈值和第六阈值可以根据CSI对目标模型的性能的影响情况确定，如可以根据CSI对目标模型的性能的历史影响情况确定。
可选地，两个时刻CSI的相关性可以描述两个时刻CSI的相似性，一般认为，在较小的时间间隔内，CSI应该比较相似。
可选地，第一时间段可以为任意可能的时间段，如第一时间段为一个时间间隔，如L个时隙。
可选地，第一参考信号相关信息满足第三条件可以包括以下至少之一：
第一参考信号接收功率(Reference Signal Receiving Power，RSRP)小于或等于第七阈值；
第一参考信号接收质量(Reference Signal Receiving Power，RSRQ)小于或等于第八阈值；
第一参考信号的信号与干扰加噪声比(Signal to Interference plus Noise Ratio，SINR)小于或等于第九阈值；
第一参考信号的信噪比(Signal Noise Ratio，SNR)小于或等于第十阈值。
可选地，第一参考信号可以包括来自网络侧设备的不同TRP的下行参考信号。
可选地，第一参考信号可以为定位参考信号、信道状态信息参考信号(Channel State Information-Reference Signal，CSI-RS)、同步信号和广播信道块(Synchronization Signal and PBCH(Physical Broadcast Channel)block，SSB)等任意可能的参考信号。
可选地，以第一参考信号相关信息为RSRP为例，对于网络侧设备的所有TRP共享同一目标模型的情况，则这些TRP的平均RSRP、最小RSRP或最大RSRP小于或等于第七阈值。对于网络侧设备的每个TRP分别对应一个模型的情况，则与目标模型关联的TRP的RSRP小于等于第七阈值。对于第一参考信号相关信息为SINR或RSRQ的描述，参见对第一参考信号相关信息为RSRP的相关描述，为了避免重复，此处不再赘述。
可选地，目标测量量可以为UE自身检测、测量到的，或者可以为UE从网络侧设备获取的，具体可以根据目标测量量确定。
可选地，目标测量量可以包括：瞬时测量量或统计测量量。
可选地，当目标测量量为统计测量量时，目标测量量具体可以包括：
在时间上统计的测量量，比如在一段时间内统计得到的测量量；
在空间上的统计的测量量，比如不同地理位置的UE测量的测量量；
在时间和空间上统计的测量量。
可选地，当目标测量量为统计测量量时，目标测量量具体可以为多次测量量的均值。
可选地，目标测量量可以包括以下至少之一：
通过N个发送与接收点TRP获取的测量量的均值；
通过N个TRP获取的测量量中的最小测量量；
通过N个TRP获取的测量量中最大的测量量；
其中，N个TRP为目标模型关联的TRP。
可选地，上述N个TRP的标识由网络侧设备配置或协议约定。
可选地，在上述方式2或方式3，即第一指示信息指示的触发方式包括半持续监督方式或非周期性监督方式的情况下，上述步骤401和步骤402具体可以通过下述的步骤401a和步骤402a实现。
步骤401a、网络侧设备向UE发送第一信令。
步骤402a、UE接收第一信令。
其中，第一信令中包括第一指示信息。
本申请实施例中，UE在接收到第一信令之后，可以解析第一信令中的第一指示信息，并根据第一指示信息的指示，对目标模型进行监督。
可以理解，UE可以在接收到第一信令后立刻对目标模型进行至少一次监督，或者可以在接收到第一信令后的预设时长之后对目标模型进行至少一次监督。
可选地，第一信令可以包括以下任一项：下行控制信息(Downlink Control Information，DCI)、无线资源控制(Radio Resource Control，RRC)、长期演进定位协议(LTE(Long Term Evolution)Positioning Protocol，LPP)、媒体接入控制-控制单元((Medium Access Control)-Control Element，MAC-CE)。如此，由于可以在DCI，RRC、LPP或者MAC-CE中的任一项中携带第一指示信息，因此可以提高网络侧设备发送第一指示信息的灵活性。
在本申请实施例提供的模型监督触发方法中，由于UE可以接收网络侧设备发送 的用于指示用于指示模型监督的触发方式的第一指示信息，且该触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式；即网络侧设备指示UE触发模型监督的触发方式，因此UE可以直接根据第一指示信息触发目标模型的模型监督。
可选地，在方式4中，上述步骤403具体可以通过下述的步骤403a实现，在步骤403a之后，本申请实施例提供的模型监督触发方法还可以包括下述的步骤404和步骤405。
步骤403a、在发生至少一个事件中的目标事件的情况下，UE对所述目标模型进行监督。
步骤404、UE向网络侧设备上报目标信息。
步骤405、网络侧设备接收目标信息。
其中，目标信息中可以包括以下至少之一：目标模型的监督结果；目标事件的相关信息。
可选地，目标模型的监督结果可以指示目标模型的有效性，即监督结果可以指示目标模型是否有效或有效期限。从而网络侧设备可以目标模型的监督结果判断目标模型的有效性。
可选地，在网络侧设备接收到目标信息之后，可以基于目标信息，对目标模型的模型参数进行修正，即更新目标模型，以提升目标模型的性能；或者，网络侧设备可以基于目标信息，更新将目标模型切换为另一个模型；或者，网络侧设备可以基于目标信息，对上述至少一个事件进行修正。例如，可以更新上述第一阈值、第二阈值等。
可选地，目标事件的相关信息可以包括以下任一项：
UE切换后的服务小区标识；
UE变化后的区域标识；
网络侧设备与目标模型关联的参考信号变化后的配置；
UE变化后的服务区域所属的网络设备商标识；
UE变化后的服务区域所属的网络运营商标识；
UE更新后的BWP标识；
目标测量量。
示例性地，以UE上报目标事件的相关信息为例：
i.若目标事件为UE发生服务小区切换，则UE上报UE的切换后的服务小区标识。
ii.若目标事件为UE所在区域发生变化，则终端上报UE变化后的区域标识，如标识A2；
iii.若目标事件为UE的移动速度超过阈值第一阈值，则UE上报其测量的运动速度；
iv.若目标事件为UE的测量信道参考信号的SINR小于或等于第九阈值，则UE上报测量的SINR；若目标事件为：UE的测量信道参考信号的SNR小于或等于第十阈值，则UE上报测量的SNR。
v.若目标事件为UE检测到网络同步误差大于或等于第二阈值，则UE上报测量的网络同步误差；
vi.若目标事件为UE获取的CSI的变化率大于或等于第五阈值，则UE上报测量的CSI变化率或者P个时刻的CSI，P为正整数。可以理解，CSI的变化率由P个时刻的CSI确定。
vii.若目标事件为UE获取的CSI在第一时间段的相关性小于或等于第六阈值，则UE上报CSI在第一时间段内的相关性；
viii.若目标事件为UE测量的第一RSRP小于或等于第七阈值，则UE上报测量 的第一RSRP；
ix.若目标事件为UE测量的第一RSRQ小于或等于第八阈值，则UE上报测量的第一RSRQ；
x.若目标事件为目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，则UE上报目标模型在连续M个时间单位的输出变化值；
xi.若目标事件为UE检测到UE的定时误差大于或等于第四阈值，则UE上报测量的UE定时误差；
xii.若目标事件为检测到网络侧设备与目标模型关联的参考信号配置发生变化，则UE上报网络侧设备与目标模型关联的参考信号变化后的配置；例如，该变化后的配置可以为UE测量的到的；
xiii.若目标事件为UE对应的的网络运营商的服务区域发生变化，则UE上报变化后的网络侧运营商的标识；
xiv.若目标事件为UE对应的网络设备商的服务区域发生变化，则UE上报变化后的网络设备商的标识；
xv.若目标事件为UE的目标BWP发生变化，则UE上报变化后的目标BWP的标识。例如，若UE的上行BWP发生变化，则UE可以上报变化后的上行BWP；若UE的下行BWP发生变化，则UE可以上报变化后的下行BWP。
如此，当第一指示信息指示模型触发方式为由事件触发的触发方式时，由于UE可以将监督结果和触发本次监督的目标事件的相关信息中的至少一项上报至网络侧设备，因此便于网络侧设备对目标模型的模型参数和触发模型监督的事件进行更新，从而可以提高后续触发模型监督的准确性。
下面以目标模型为AI定位模型为例，对本申请实施例提供的模型监督触发方法进行示例性地说明。
示例性地，本申请实施例提供一种定位模型监督触发方法，对于模型监督在UE侧执行的情况，网络侧设备可以指示UE模型监督的触发方式，包括周期性监督方式、非周期性监督方式、半持续监督方式以及由事件触发的监督方式中的至少一种。本实施例提供的定位模型监督触发方法可以包括下述的步骤：
步骤50、网络侧设备向UE发送触发模型监督的第一指示信息；
步骤51、UE接收第一指示信息。
步骤52，UE根据第一指示信息，对AI定位模型进行监督。
可选地，第一指示信息用于指示模型监督的触发方式，具体的，第一指示信息用于指示：(1)周期性监督方式、(2)非周期性监督方式、(3)半持续监督方式以及(4)由事件触发的监督方式中的至少一种。
其中，(1)当第一指示信息指示周期性监督方式时，第一指示信息中包括：指示模型监督的周期，如每N个时隙执行一次模型监督；如第一指示信息指示UE每N个时隙执行一次模型监督。
(2)当第一指示信息指示半持续监督方式时，第一指示信息包含在网络侧设备发送的第一信令中，如DCI、RRC、LPP、MAC CE信令等。
第一指示信息中可以包括以下任一项：模型监督的次数、模型监督的时长、模型监督的次数和周期值、模型监督的时长和周期值。如，第一指示信息指示UE每N个时隙执行一次模型监督，一共监督K次或监督L时长，K为正整数，L大于0。
(3)当第一指示信息指示非周期性监督时，第一指示信息包含在网络侧设备发送的第一信令中，如DCI、RRC、LPP、MAC CE信令等。且UE接收到第一指示信息之后，可以立即对AI定位模型进行一次模型监督，或者可以在接收到第一指示信息之后的一个预设时长后，对AI定位模型进行一次模型监督。
可以理解，在(2)和(3)中，网络侧设备可以在需求UE对AI定位模型进行模型监督的时候，向第UE发送第一指示信息，以便于UE对AI定位模型进行监督。
(4)当第一指示信息指示由事件触发的触发方式时，第一指示信息可以包括触发模型监督的至少一个事件，该至少一个事件包括以下至少一项：
事件1：UE发生小区切换，如服务小区由C1切换到C2；
事件2：UE所在区域发生变化，如区域ID由A1切换到A2；
事件3：UE移动速度超过阈值T1，即第一阈值；
事件4：UE测量信道的SINR小于或等于阈值T2；
事件5：UE测量信道的SNR小于或等于阈值T10；
其中，阈值T2和阈值T10可以相同，也可以不同。
事件5：UE检测到网络同步误差大于或等于阈值T3；
事件6：UE获取的CSI的变化率大于或等于阈值T4；
事件7：UE获取的CSI在一个时间间隔内的相关性小于或等于阈值T5；
事件8：UE测量的RSRP小于或等于阈值T6；
事件9：UE测量的RSRQ小于或等于阈值T7；
事件10：UE获取的、AI定位模型在连续M个时间单位的输出变化值大于或等于阈值T8；
事件11：UE检测到UE的定时误差大于或等于阈值T9；
事件12：UE检测到网络侧设备与AI定位模型关联的参考信号配置发生变化；
事件13：UE进入新的网络运营商的服务区域，即UE从一个网络运营商的服务区域进入到另一个网络运营商的服务区域；
事件14：UE进入新的网络设备商的服务区域，即即UE从一个网络设备商的服务区域进入到另一个网络设备商的服务区域；
事件15：终端的上行和/或下行BWP发生变化。
可选地，在(4)中，UE可以确定发生上述至少一个事件中的任一事件时，对AI定位模型进行一次监督。
可选地，在上述(4)中，UE接收到第一指示信息之后，若发生了上述至少一个事件中的目标事件，则UE可以对AI模型进行一次模型监督。然后，UE可以向网络侧设备上报目标事件的相关信息。例如：
i.若目标事件为UE发生服务小区切换，则UE上报UE的切换后的服务小区标识。
ii.若目标事件为UE所在区域发生变化，则终端上报UE变化后的区域标识，如标识A2；
iii.若目标事件为UE的移动速度超过阈值第一阈值，则UE上报其测量的运动速度；
iv.若目标事件为UE的测量信道参考信号的SINR小于或等于第九阈值，则UE上报测量的SINR；若目标事件为：UE的测量信道参考信号的SNR小于或等于第十阈值，则UE上报测量的SNR。
v.若目标事件为UE检测到网络同步误差大于或等于第二阈值，则UE上报测量的网络同步误差；
vi.若目标事件为UE获取的CSI的变化率大于或等于第五阈值，则UE上报测量的CSI变化率或者P个时刻的CSI，P为正整数。可以理解，CSI的变化率由P个时刻的CSI确定。
vii.若目标事件为UE获取的CSI在第一时间段的相关性小于或等于第六阈值，则UE上报CSI在第一时间段内的相关性；
viii.若目标事件为UE测量的第一RSRP小于或等于第七阈值，则UE上报测量 的第一RSRQ；
ix.若目标事件为UE测量的第一RSRP小于或等于第八阈值，则UE上报测量的第一RSRQ；
x.若目标事件为目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，则UE上报目标模型在连续M个时间单位的输出变化值；
xi.若目标事件为UE检测到UE的定时误差大于或等于第四阈值，则UE上报测量的UE定时误差；
xii.若目标事件为检测到网络侧设备与目标模型关联的参考信号配置发生变化，则UE上报网络侧设备与目标模型关联的参考信号变化后的配置；例如，该变化后的配置可以为UE测量的到的；
xiii.若目标事件为UE对应的的网络运营商的服务区域发生变化，则UE上报变化后的网络侧运营商的标识；
xiv.若目标事件为UE对应的网络设备商的服务区域发生变化，则UE上报变化后的网络设备商的标识；
xv.若目标事件为UE的目标BWP发生变化，则UE上报变化后的目标BWP的标识。例如，若UE的上行BWP发生变化，则UE可以上报变化后的上行BWP；若UE的下行BWP发生变化，则UE可以上报变化后的下行BWP。
可选地，UE可以将对AI模型的监督结果和目标事件相关信息一起上报至网络侧设备。
可选地，上述事件3～事件11中的测量量，可以包括瞬时值和统计值，如统计至可以为多次测量值的均值。
其中，该多次测量值可以为在不同时间、不同UE或不同区域测量得到的测量值。
可选地，上述事件3～事件11中的测量量，可以包括N个TRP测量量的均值、N个TRP中最小的测量量、N个TRP中最大的测量量。该N个TRP为AI定位模型关联的TRP。
可选地，AI定位模型关联的TRP标识可以由网络侧设备提前配置。
本申请实施例提供的AI定位模型触发方法可以提供触发监督的触发方式，从而可以基于指示的监督方式，监督模型的性能，保证AI定位模型的定位精度。
本申请实施例提供的模型监督触发方法，执行主体可以为模型监督触发装置。本申请实施例中以模型监督触发装置执行模型监督触发方法为例，说明本申请实施例提供的模型监督触发装置。
本申请实施例提供一种AI模型监督触发装置，图5示出了本申请实施例提供的模型监督触发装置的结构示意图，如图5所示，本申请实施例提供的模型监督触发装置500可以包括：接收模块501和处理模块502。
接收模块501，用于接收网络侧设备发送的第一指示信息，第一指示信息用于指示模型监督的触发方式；
处理模块502，用于根据接收模块501接收的第一指示信息，对目标模型进行监督；
其中，第一指示信息指示的触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
一种可能的实现方式，上述第一指示信息指示的触发方式包括周期性监督方式的情况下，第一指示信息还包括模型监督的周期值。
一种可能的实现方式，上述第一指示信息指示的触发方式包括半持续监督方式的情况下，第一指示信息还包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
一种可能的实现方式，上述第一指示信息指示的触发方式包括事件触发监督方式的情况下，第一指示信息还包括触发模型监督的至少一个事件。
一种可能的实现方式，上述至少一个事件包括以下至少之一：
UE的服务小区发生切换；
UE所在区域发生变化；
网络侧设备与目标模型关联的参考信号配置发生变化；
UE对应的服务区域发生变化；
目标部分带宽BWP发生更新；
目标测量量满足第一条件；
其中，目标测量量与UE、网络侧设备和目标模型中的至少之一相关；
目标BWP包括UE的下行BWP和上行BWP中的至少之一。
一种可能的实现方式，上述目标测量量满足第一条件包括以下至少之一：
UE的运动速度大于或等于第一阈值；
网络同步误差大于或等于第二阈值；
CSI满足第二条件；
第一参考信号相关信息满足第三条件；
目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
UE的定时误差大于或等于第四阈值。
一种可能的实现方式，上述CSI满足第二条件包括以下至少之一：
CSI的变化率大于或等于第五阈值；
CSI在第一时间间隔内的相关性小于或等于第六阈值。
一种可能的实现方式，上述第一参考信号相关信息满足第三条件包括以下至少之一：
第一参考信号接收功率RSRP小于或等于第七阈值；
第一参考信号接收质量RSRQ小于或等于第八阈值；
第一参考信号的信号与干扰加噪声比SINR小于或等于第九阈值；
第一参考信号的信噪比SNR小于或等于第十阈值。
一种可能的实现方式，上述UE对应的服务区域包括以下至少之一：
UE对应的网络运营商的服务区域、UE对应的网络设备商的服务区域。
一种可能的实现方式，上述目标测量量包括：瞬时测量量或统计测量量。
一种可能的实现方式，上述目标测量量包括以下至少之一：
通过N个发送与接收点TRP获取的测量量的均值；
通过N个TRP获取的测量量中的最小测量量；
通过N个TRP获取的测量量中最大的测量量；
其中，N个TRP为目标模型关联的TRP。
一种可能的实现方式，上述N个TRP的标识由网络侧设备配置或协议约定。
一种可能的实现方式，上述处理模块502，具体用于在发生至少一个事件中的目标事件的情况下，对目标模型进行监督；
模型监督触发装置还可以包括发送模块；发送模块，用于在对目标模型进行监督之后，向网络侧设备上报目标信息。目标信息中包括以下至少之一：目标模型的监督结果；目标事件的相关信息。
一种可能的实现方式，上述目标事件的相关信息包括以下任一项：
UE切换后的服务小区标识；
UE变化后的区域标识；
网络侧设备与目标模型关联的参考信号变化后的配置；
UE变化后的服务区域所属的网络设备商标识；
UE变化后的服务区域所属的网络运营商标识；
UE更新后的BWP标识；
目标测量量。
一种可能的实现方式，上述触发方式包括半持续监督方式或非周期性监督方式；
接收模块501，具体用于接收网络侧设备发送的第一信令，第一信令中包括第一指示信息。
一种可能的实现方式，上述第一信令包括以下任一项：DCI、RRC、LPP、MAC-CE。
一种可能的实现方式，上述目标模型为定位模型。
在本申请实施例提供的模型监督触发装置中，由于模型监督触发装置可以接收网络侧设备发送的用于指示用于指示模型监督的触发方式的第一指示信息，且该触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式；即网络侧设备指示模型监督触发装置触发模型监督的触发方式，因此模型监督触发装置可以直接根据第一指示信息触发目标模型的模型监督。
本申请实施例提供一种模型监督触发装置，图6示出了本申请实施例提供的模型监督触发装置的结构示意图，如图6所示，本申请实施例提供的模型监督触发装置600可以包括：发送模块601。发送模块601，用于向UE发送第一指示信息，第一指示信息用于指示模型监督的触发方式；第一指示信息用于UE对目标模型进行监督；
其中，触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
一种可能的实现方式中，上述第一指示信息指示的触发方式包括周期性监督方式的情况下，第一指示信息还包括模型监督的周期值。
一种可能的实现方式中，上述第一指示信息指示的触发方式包括半持续监督方式的情况下，第一指示信息还包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
一种可能的实现方式中，上述第一指示信息指示的触发方式包括事件触发监督方式的情况下，第一指示信息还包括触发模型监督的至少一个事件。
其中，上述至少一个事件包括以下至少之一：
UE的服务小区发生切换；
UE所在区域发生变化；
网络侧设备与目标模型关联的参考信号配置发生变化；
UE对应的服务区域发生变化；
目标BWP发生更新；
目标测量量满足第一条件；
其中，目标测量量与UE、网络侧设备和目标模型中的至少之一相关；
目标BWP包括UE的下行BWP和上行BWP中的至少之一。
一种可能的实现方式中，上述目标测量量满足第一条件包括以下至少之一：
UE的运动速度大于或等于第一阈值；
网络同步误差大于或等于第二阈值；
CSI满足第二条件；
第一参考信号相关信息满足第三条件；
目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
UE的定时误差大于或等于第四阈值。
一种可能的实现方式中，上述CSI满足第二条件包括以下至少之一：
CSI的变化率大于或等于第五阈值；
CSI在第一时间间隔内的相关性小于或等于第六阈值。
一种可能的实现方式中，上述第一参考信号相关信息满足第三条件包括以下至少之一：
第一RSRP小于或等于第七阈值；
第一RSRQ小于或等于第八阈值；
第一参考信号的SINR小于或等于第九阈值；
第一参考信号的SNR小于或等于第十阈值。
一种可能的实现方式中，上述UE对应的服务区域包括以下至少之一：
UE对应的网络运营商的服务区域、UE对应的网络设备商的服务区域。
一种可能的实现方式中，上述目标测量量包括：瞬时测量量或统计测量量。
一种可能的实现方式中，上述目标测量量包括以下至少之一：
通过N个TRP获取的测量量的均值；
通过N个TRP获取的测量量中的最小测量量；
通过N个TRP获取的测量量中最大的测量量；
其中，N个TRP为目标模型关联的TRP。
一种可能的实现方式中，上述模型监督触发装置600还可以包括接收模块；
接收模块，用于在发送模块601向UE发送第一指示信息之后，接收UE上报的目标信息；目标信息中包括以下至少之一：目标模型的监督结果；目标事件的相关信息。
一种可能的实现方式中，上述触发方式包括半持续监督方式或非周期性监督方式；
网络侧设备向UE发送第一指示信息，包括：
网络侧设备向UE发送第一信令，第一信令中包括第一指示信息。
一种可能的实现方式中，上述第一信令包括以下任一项：DCI、RRC、LPP、MAC-CE。
一种可能的实现方式中，上述目标模型为定位模型。
在本申请实施例提供的模型监督触发装置600中，由于模型监督触发装置可以向UE发送的用于指示用于指示模型监督的触发方式的第一指示信息，且该触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式；即模型监督触发装置指示UE触发模型监督的触发方式，因此UE可以直接根据第一指示信息触发目标模型的模型监督。
本申请实施例中的模型监督触发装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述图1所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的模型监督触发装置能够实现图4的方法实施例中UE实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图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和存储器702，存储器702上存储有可在所述处理器701上运行的程序或指令，例如，该通信设备700为终端时，该程序或指令被处理器701执行时实现上述模型监督触发方法实施例的各个步骤，且能达到相同的技术效果。该通信设备700为网络侧设备时，该程序或指令被处理器701执行时实现上述模型监督触发方法实施例中网络侧设备执行的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。该通信设备 700为UE时，该程序或指令被处理器701执行时实现上述模型监督触发方法实施例中UE执行的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种UE，包括处理器和通信接口，其中，所述通信接口用于接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述处理器用于根据所述第一指示信息，对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。该UE实施例与上述UE方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中，且能达到相同的技术效果。具体地，图8为实现本申请实施例的一种UE的硬件结构示意图。
该UE800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，UE800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UE结构并不构成对终端的限定，UE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。
其中，射频单元801，用于接收网络侧设备发送的第一指示信息，第一指示信息用于指示模型监督的触发方式；
处理器810，用于根据射频单元801接收的第一指示信息，对目标模型进行监督；
其中，触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
一种可能的实现方式，上述第一指示信息指示的触发方式包括周期性监督方式的情况下，第一指示信息还包括模型监督的周期值。
一种可能的实现方式，上述第一指示信息指示的触发方式包括半持续监督方式的情况下，第一指示信息还包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
一种可能的实现方式，上述第一指示信息指示的触发方式包括事件触发监督方式的情况下，第一指示信息还包括触发模型监督的至少一个事件；
其中，上述至少一个事件包括以下至少之一：
UE的服务小区发生切换；
UE所在区域发生变化；
网络侧设备与目标模型关联的参考信号配置发生变化；
UE对应的服务区域发生变化；
目标部分带宽BWP发生更新；
目标测量量满足第一条件；
其中，目标测量量与UE、网络侧设备和目标模型中的至少之一相关；
目标BWP包括UE的下行BWP和上行BWP中的至少之一。
一种可能的实现方式，上述目标测量量满足第一条件包括以下至少之一：
UE的运动速度大于或等于第一阈值；
网络同步误差大于或等于第二阈值；
CSI满足第二条件；
第一参考信号相关信息满足第三条件；
目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
UE的定时误差大于或等于第四阈值。
一种可能的实现方式，上述CSI满足第二条件包括以下至少之一：
CSI的变化率大于或等于第五阈值；
CSI在第一时间间隔内的相关性小于或等于第六阈值。
一种可能的实现方式，上述第一参考信号相关信息满足第三条件包括以下至少之一：
第一参考信号接收功率RSRP小于或等于第七阈值；
第一参考信号接收质量RSRQ小于或等于第八阈值；
第一参考信号的信号与干扰加噪声比SINR小于或等于第九阈值；
第一参考信号的信噪比SNR小于或等于第十阈值。
一种可能的实现方式，上述UE对应的服务区域包括以下至少之一：
UE对应的网络运营商的服务区域、UE对应的网络设备商的服务区域。
一种可能的实现方式，上述目标测量量包括：瞬时测量量或统计测量量。
一种可能的实现方式，上述目标测量量包括以下至少之一：
通过N个发送与接收点TRP获取的测量量的均值；
通过N个TRP获取的测量量中的最小测量量；
通过N个TRP获取的测量量中最大的测量量；
其中，N个TRP为目标模型关联的TRP。
一种可能的实现方式，上述N个TRP的标识由网络侧设备配置或协议约定。
一种可能的实现方式，上述处理器810，具体用于在发生至少一个事件中的目标事件的情况下，对目标模型进行监督；
射频单元801，还用于在对目标模型进行监督之后，向网络侧设备上报目标信息。目标信息中包括以下至少之一：目标模型的监督结果；目标事件的相关信息。
一种可能的实现方式，上述目标事件的相关信息包括以下任一项：
UE切换后的服务小区标识；
UE变化后的区域标识；
网络侧设备与目标模型关联的参考信号变化后的配置；
UE变化后的服务区域所属的网络设备商标识；
UE变化后的服务区域所属的网络运营商标识；
UE更新后的BWP标识；
目标测量量。
一种可能的实现方式，上述触发方式包括半持续监督方式或非周期性监督方式；
射频单元801，具体用于接收网络侧设备发送的第一信令，第一信令中包括第一指示信息。
一种可能的实现方式，上述第一信令包括以下任一项：DCI、RRC、LPP、MAC-CE。
一种可能的实现方式，上述目标模型为定位模型。
在本申请实施例提供的UE中，由于UE可以接收网络侧设备发送的用于指示用于指示模型监督的触发方式的第一指示信息，且该触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式；即网络侧设备指示UE触发模型监督的触发方式，因此UE可以直接根据第一指示信息触发目标模型的模型监督。
应理解的是，本申请实施例中，输入单元804可以包括图形处理单元(Graphics Processing Unit，GPU)8041和麦克风8042，图形处理器8041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8061，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8061。用户输入单元807包括触控面板8071以及其他输入设备8072中的至少一种。触控面板8071，也称为触摸屏。触控面板8071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8072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接收来自网络侧设备的下行数据后，可以传输给处理器810进行处理；另外，射频单元801可以向网络侧设备发送上行数据。通常，射频单元801包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器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的第一存储区和存储数据的第二存储区，其中，第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器809可以包括易失性存储器或非易失性存储器，或者，存储器809可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-OnlyMemory，ROM)、可编程只读存储器(ProgrammableROM，PROM)、可擦除可编程只读存储器(ErasablePROM，EPROM)、电可擦除可编程只读存储器(ElectricallyEPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(RandomAccessMemory，RAM)，静态随机存取存储器(StaticRAM，SRAM)、动态随机存取存储器(DynamicRAM，DRAM)、同步动态随机存取存储器(SynchronousDRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(DoubleDataRateSDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(EnhancedSDRAM，ESDRAM)、同步连接动态随机存取存储器(SynchlinkDRAM，SLDRAM)和直接内存总线随机存取存储器(DirectRambusRAM，DRRAM)。本申请实施例中的存储器809包括但不限于这些和任意其它适合类型的存储器。
处理器810可包括一个或多个处理单元；可选的，处理器810集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器810中。
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，其中，所述处理器用于获取第一指示信息，所述通信接口用于向UE发送所述第一指示信息，所述第一指示信息用于指示模型监督的触发方式；所述第一指示信息用于所述UE对目标模型进行监督；其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。该网络侧设备实施例与上述网络侧设备方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中，且能达到相同的技术效果。
具体地，本申请实施例还提供了一种网络侧设备。如图9所示，该网络侧设备900包括：天线91、射频装置92、基带装置93、处理器94和存储器95。天线91与射频装置92连接。在上行方向上，射频装置92通过天线91接收信息，将接收的信息发送给基带装置93进行处理。在下行方向上，基带装置93对要发送的信息进行处理，并发送给射频装置92，射频装置92对收到的信息进行处理后经过天线91发送出去。
以上实施例中网络侧设备执行的方法可以在基带装置93中实现，该基带装置93包括基带处理器。
基带装置93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基带处理器，通过总线接口与存储器95连接，以调用存储器95中的程序，执行以上方法实施例中所示的网络设备操作。
该网络侧设备还可以包括网络接口96，该接口例如为通用公共无线接口(common  public radio interface，CPRI)。
具体地，本发明实施例的网络侧设备900还包括：存储在存储器95上并可在处理器94上运行的指令或程序，处理器94调用存储器95中的指令或程序执行图5所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
具体地，本申请实施例还提供了一种网络侧设备。如图10所示，该网络侧设备1000包括：处理器1001、网络接口1002和存储器1003。其中，网络接口1002例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本发明实施例的网络侧设备1000还包括：存储在存储器1003上并可在处理器1001上运行的指令或程序，处理器1001调用存储器1003中的指令或程序执行图6所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
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监督触发方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，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，实现上述模型监督触发方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现上述模型监督触发方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供了一种通信系统，包括：UE及网络侧设备，所述UE可用于执行如上所述的UE方法实施例的步骤，所述网络侧设备可用于执行如上所述的网络侧方法实施例的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (28)

  1. 一种模型监督触发方法，包括：
    用户设备UE接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；
    所述UE根据所述第一指示信息，对目标模型进行监督；
    其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
  2. 根据权利要求1所述的方法，其中，所述第一指示信息指示的触发方式包括所述周期性监督方式的情况下，所述第一指示信息还包括模型监督的周期值。
  3. 根据权利要求1所述的方法，其中，所述第一指示信息指示的触发方式包括所述半持续监督方式的情况下，所述第一指示信息还包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
  4. 根据权利要求1所述的方法，其中，所述第一指示信息指示的触发方式包括所述事件触发监督方式的情况下，所述第一指示信息还包括触发模型监督的至少一个事件；
    其中，所述至少一个事件包括以下至少之一：
    所述UE的服务小区发生切换；
    所述UE所在区域发生变化；
    所述网络侧设备与所述目标模型关联的参考信号配置发生变化；
    所述UE对应的服务区域发生变化；
    目标部分带宽BWP发生更新；
    目标测量量满足第一条件；
    其中，所述目标测量量与所述UE、所述网络侧设备和所述目标模型中的至少之一相关；
    所述目标BWP包括所述UE的下行BWP和上行BWP中的至少之一。
  5. 根据权利要求4所述的方法，其中，所述目标测量量满足第一条件包括以下至少之一：
    所述UE的运动速度大于或等于第一阈值；
    网络同步误差大于或等于第二阈值；
    信道状态信息CSI满足第二条件；
    第一参考信号相关信息满足第三条件；
    所述目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
    所述UE的定时误差大于或等于第四阈值；
    其中，所述CSI满足第二条件包括以下至少之一：
    CSI的变化率大于或等于第五阈值；
    CSI在第一时间间隔内的相关性小于或等于第六阈值；
    其中，所述第一参考信号相关信息满足第三条件包括以下至少之一：
    第一参考信号接收功率RSRP小于或等于第七阈值；
    第一参考信号接收质量RSRQ小于或等于第八阈值；
    第一参考信号的信号与干扰加噪声比SINR小于或等于第九阈值；
    第一参考信号的信噪比SNR小于或等于第十阈值。
  6. 根据权利要求4所述的方法，其中，所述UE对应的服务区域包括以下至少之一：
    所述UE对应的网络运营商的服务区域、所述UE对应的网络设备商的服务区域。
  7. 根据权利要求4所述的方法，其中，所述目标测量量包括：瞬时测量量或统计 测量量。
  8. 根据权利要求4所述的方法，其中，所述目标测量量包括以下至少之一：
    通过N个发送与接收点TRP获取的测量量的均值；
    通过所述N个TRP获取的测量量中的最小测量量；
    通过所述N个TRP获取的测量量中最大的测量量；
    其中，所述N个TRP为所述目标模型关联的TRP。
  9. 根据权利要求4所述的方法，其中，所述UE根据所述第一指示信息，对目标模型进行监督，包括：
    在发生所述至少一个事件中的目标事件的情况下，所述UE对所述目标模型进行监督；
    所述UE对所述目标模型进行监督之后，所述方法还包括：
    所述UE向所述网络侧设备上报目标信息；
    所述目标信息中包括以下至少之一：
    所述目标模型的监督结果；
    所述目标事件的相关信息；
    其中，所述目标事件的相关信息包括以下任一项：
    所述UE切换后的服务小区标识；
    所述UE变化后的区域标识；
    所述网络侧设备与所述目标模型关联的参考信号变化后的配置；
    所述UE变化后的服务区域所属的网络设备商标识；
    所述UE变化后的服务区域所属的网络运营商标识；
    所述UE更新后的BWP标识；
    目标测量量。
  10. 根据权利要求1所述的方法，其中，所述触发方式包括所述半持续监督方式或所述非周期性监督方式；
    所述UE接收网络侧设备发送第一指示信息，包括：
    所述UE接收网络侧设备发送的第一信令，所述第一信令中包括所述第一指示信息。
  11. 根据权利要求1所述的方法，其中，所述目标模型为定位模型。
  12. 一种模型监督触发方法，包括：
    网络侧设备向UE发送第一指示信息，所述第一指示信息用于指示模型监督的触发方式；
    所述第一指示信息用于所述UE对目标模型进行监督；
    其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
  13. 根据权利要求12所述的方法，其中，所述第一指示信息指示的触发方式包括所述周期性监督方式的情况下，所述第一指示信息还包括模型监督的周期值。
  14. 根据权利要求12所述的方法，其中，所述第一指示信息指示的触发方式包括所述半持续监督方式的情况下，所述第一指示信息还包括以下至少之一：模型监督的周期值、模型监督的时长、模型监督的次数。
  15. 根据权利要求12所述的方法，其中，所述第一指示信息指示的触发方式包括所述事件触发监督方式的情况下，所述第一指示信息还包括触发模型监督的至少一个事件；
    其中，所述至少一个事件包括以下至少之一：
    所述UE的服务小区发生切换；
    所述UE所在区域发生变化；
    所述网络侧设备与所述目标模型关联的参考信号配置发生变化；
    所述UE对应的服务区域发生变化；
    目标BWP发生更新；
    目标测量量满足第一条件；
    其中，所述目标测量量与所述UE.所述网络侧设备和所述目标模型中的至少之一相关；
    所述目标BWP包括所述UE的下行BWP和上行BWP中的至少之一。
  16. 根据权利要求15所述的方法，其中，所述目标测量量满足第一条件包括以下至少之一：
    所述UE的运动速度大于或等于第一阈值；
    网络同步误差大于或等于第二阈值；
    CSI满足第二条件；
    第一参考信号相关信息满足第三条件；
    所述目标模型在连续的M个时间单位内的输出变化值大于或等于第三阈值，M为正整数；
    所述UE的定时误差大于或等于第四阈值；
    其中，所述CSI满足第二条件包括以下至少之一：
    CSI的变化率大于或等于第五阈值；
    CSI在第一时间间隔内的相关性小于或等于第六阈值；
    其中，所述第一参考信号相关信息满足第三条件包括以下至少之一：
    第一RSRP小于或等于第七阈值；
    第一RSRQ小于或等于第八阈值；
    第一参考信号的SINR小于或等于第九阈值；
    第一参考信号的SNR小于或等于第十阈值。
  17. 根据权利要求15所述的方法，其中，所述UE对应的服务区域包括以下至少之一：
    所述UE对应的网络运营商的服务区域.所述UE对应的网络设备商的服务区域。
  18. 根据权利要求15所述的方法，其中，所述目标测量量包括：瞬时测量量或统计测量量。
  19. 根据权利要求15所述的方法，其中，所述目标测量量包括以下至少之一：
    通过N个TRP获取的测量量的均值；
    通过所述N个TRP获取的测量量中的最小测量量；
    通过所述N个TRP获取的测量量中最大的测量量；
    其中，所述N个TRP为所述目标模型关联的TRP。
  20. 根据权利要求15所述的方法，其中，所述网络侧设备向UE发送第一指示信息之后，所述方法还包括：
    所述网络侧设备接收所述UE上报的目标信息；
    所述目标信息中包括以下至少之一：
    所述目标模型的监督结果；
    所述目标事件的相关信息；
    其中，所述目标事件的相关信息包括以下任一项：
    所述UE切换后的服务小区标识；
    所述UE变化后的区域标识；
    所述网络侧设备与所述目标模型关联的参考信号变化后的配置；
    所述UE变化后的服务区域所属的网络设备商标识；
    所述UE变化后的服务区域所属的网络运营商标识；
    所述UE更新后的BWP标识；
    目标测量量。
  21. 根据权利要求12所述的方法，其中，所述第一指示信息指示的触发方式包括所述半持续监督方式或所述非周期性监督方式的情况下；
    所述网络侧设备向UE发送第一指示信息，包括：
    所述网络侧设备向UE发送第一信令，所述第一信令中包括所述第一指示信息。
  22. 根据权利要求12所述的方法，其中，所述目标模型为定位模型。
  23. 一种模型监督触发装置，所述装置包括：接收模块和处理模块；
    所述接收模块，用于接收网络侧设备发送的第一指示信息，所述第一指示信息用于指示模型监督的触发方式；
    所述处理模块，用于根据所述接收模块接收的所述第一指示信息，对目标模型进行监督；
    其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
  24. 一种模型监督触发装置，所述装置包括：发送模块；
    所述发送模块，用于向UE发送第一指示信息，所述第一指示信息用于指示模型监督的触发方式；
    所述第一指示信息用于所述UE对目标模型进行监督；
    其中，所述触发方式包括以下至少之一：周期性监督方式、半持续监督方式、非周期性监督方式、由事件触发的监督方式。
  25. 一种UE，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至11中任一项所述的模型监督触发方法的步骤。
  26. 一种网络侧设备，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求12至22中任一项所述的模型监督触发方法的步骤。
  27. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至11中任一项所述的模型监督触发方法，或者实现如权利要求12至22中任一项所述的模型监督触发方法的步骤。
  28. 一种通信系统，所述通信系统包括如权利要求23所述的模型监督触发装置以及如权利要求24所述的模型监督触发装置；或者，
    所述通信系统包括如权利要求25所述的UE和如权利要求26所述的网络侧设备。
PCT/CN2023/136813 2022-12-09 2023-12-06 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统 WO2024120447A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211586245.8 2022-12-09
CN202211586245.8A CN118175048A (zh) 2022-12-09 2022-12-09 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统

Publications (1)

Publication Number Publication Date
WO2024120447A1 true WO2024120447A1 (zh) 2024-06-13

Family

ID=91349413

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/136813 WO2024120447A1 (zh) 2022-12-09 2023-12-06 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统

Country Status (2)

Country Link
CN (1) CN118175048A (zh)
WO (1) WO2024120447A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110533251A (zh) * 2019-09-03 2019-12-03 北京天泽智云科技有限公司 提升预测性维护模型适应能力的方法及装置
US20210256391A1 (en) * 2020-02-13 2021-08-19 International Business Machines Corporation Unsupervised, semi-supervised, and supervised learning using deep learning based probabilistic generative models
CN113692594A (zh) * 2019-04-08 2021-11-23 国际商业机器公司 通过强化学习的公平性改进
CN114363921A (zh) * 2020-10-13 2022-04-15 维沃移动通信有限公司 Ai网络参数的配置方法和设备
CN115201608A (zh) * 2022-07-26 2022-10-18 广东粤电靖海发电有限公司 一种基于神经网络的电厂设备运行参数监测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113692594A (zh) * 2019-04-08 2021-11-23 国际商业机器公司 通过强化学习的公平性改进
CN110533251A (zh) * 2019-09-03 2019-12-03 北京天泽智云科技有限公司 提升预测性维护模型适应能力的方法及装置
US20210256391A1 (en) * 2020-02-13 2021-08-19 International Business Machines Corporation Unsupervised, semi-supervised, and supervised learning using deep learning based probabilistic generative models
CN114363921A (zh) * 2020-10-13 2022-04-15 维沃移动通信有限公司 Ai网络参数的配置方法和设备
CN115201608A (zh) * 2022-07-26 2022-10-18 广东粤电靖海发电有限公司 一种基于神经网络的电厂设备运行参数监测方法

Also Published As

Publication number Publication date
CN118175048A (zh) 2024-06-11

Similar Documents

Publication Publication Date Title
US11451452B2 (en) Model update method and apparatus, and system
WO2024120447A1 (zh) 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统
WO2023066288A1 (zh) 模型请求方法、模型请求处理方法及相关设备
WO2023040887A1 (zh) 信息上报方法、装置、终端及可读存储介质
WO2024067434A1 (zh) Rs配置方法、装置、终端及网络侧设备
CN114125962A (zh) 一种自适应网络切换方法、系统及存储介质
WO2024067281A1 (zh) Ai模型的处理方法、装置及通信设备
WO2024125525A1 (zh) Ai算力上报方法、终端及网络侧设备
WO2023103911A1 (zh) 测量方法、装置、设备及存储介质
WO2024067665A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2024093771A1 (zh) 参考信号确定方法、终端及网络侧设备
WO2024099094A1 (zh) 波束测量方法、装置、终端、网络侧设备及存储介质
WO2024099091A1 (zh) 波束预测方法、装置、终端、网络侧设备及存储介质
WO2024093997A1 (zh) 确定模型适用性的方法、装置及通信设备
WO2023186099A1 (zh) 信息反馈方法、装置及设备
WO2023186091A1 (zh) 样本确定方法、装置及设备
WO2023040885A1 (zh) 参数选择方法、参数配置方法、终端及网络侧设备
WO2024125391A2 (zh) Ai模型监测方法、ai模型性能的测量方法、装置及设备
WO2024093713A1 (zh) 资源配置方法、装置、通信设备及可读存储介质
WO2024067280A1 (zh) 更新ai模型参数的方法、装置及通信设备
WO2023072239A1 (zh) 信道预测方法、装置、网络侧设备及终端
WO2023155839A1 (zh) Ai模型的在线学习方法、装置、通信设备及可读存储介质
WO2024078400A1 (zh) 模型请求方法、装置、通信设备及可读存储介质
WO2023040888A1 (zh) 数据传输方法及装置
WO2023134761A1 (zh) 波束信息交互方法、装置、设备及存储介质