WO2025035369A1 - 用于无线通信的方法及设备 - Google Patents

用于无线通信的方法及设备 Download PDF

Info

Publication number
WO2025035369A1
WO2025035369A1 PCT/CN2023/112991 CN2023112991W WO2025035369A1 WO 2025035369 A1 WO2025035369 A1 WO 2025035369A1 CN 2023112991 W CN2023112991 W CN 2023112991W WO 2025035369 A1 WO2025035369 A1 WO 2025035369A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
terminal device
information
network device
timer
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
PCT/CN2023/112991
Other languages
English (en)
French (fr)
Inventor
沈嘉
田文强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Oppo Mobile Telecommunications Corp Ltd
Original Assignee
Guangdong Oppo Mobile Telecommunications Corp Ltd
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 Guangdong Oppo Mobile Telecommunications Corp Ltd filed Critical Guangdong Oppo Mobile Telecommunications Corp Ltd
Priority to PCT/CN2023/112991 priority Critical patent/WO2025035369A1/zh
Priority to CN202380101304.4A priority patent/CN121773649A/zh
Publication of WO2025035369A1 publication Critical patent/WO2025035369A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W16/00Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22Traffic simulation tools or models

Definitions

  • the present application relates to the field of communication technology, and more specifically, to a method and device for wireless communication.
  • AI artificial intelligence
  • the present application provides a method and device for wireless communication. Various aspects involved in the present application are introduced below.
  • a method for wireless communication comprising: if a first condition is met, a terminal device communicates with a network device based on a first model, wherein the first condition includes that the terminal device is not configured or is not indicated with model information, and/or that a first timer expires, and the first timer is associated with the time at which the terminal device communicates with the network device based on a second model.
  • a method for wireless communication comprising: if a first condition is met, a network device communicates with a terminal device based on a third model, the first condition comprising that the terminal device is not configured or is not indicated with model information, and/or that a first timer expires; wherein the first model is associated with the third model, the first model is the model used by the terminal device when the first condition is met, and the first timer is associated with the time at which the terminal device communicates with the network device based on the second model.
  • a terminal device comprising: a communication unit, configured to communicate with a network device based on a first model if a first condition is met, wherein the first condition includes that the terminal device is not configured or is not indicated with model information, and/or that a first timer expires, and the first timer is associated with the time at which the terminal device communicates with the network device based on a second model.
  • a network device comprising: a communication unit, used to communicate with a terminal device based on a third model if a first condition is met, the first condition including that the terminal device is not configured or is not indicated with model information, and/or that a first timer expires; wherein the first model is associated with the third model, the first model is the model used by the terminal device when the first condition is met, and the first timer is associated with the time when the terminal device communicates with the network device based on the second model.
  • a terminal device comprising a memory and a processor, wherein the memory is used to store programs, and the processor is used to call the programs in the memory so that the device executes part or all of the steps in the method of the first aspect.
  • a network device comprising a memory and a processor, wherein the memory is used to store programs, and the processor is used to call the programs in the memory so that the device executes part or all of the steps in the method of the second aspect.
  • a terminal device comprising a processor for calling a program from a memory so that the device executes part or all of the steps in the method of the first aspect.
  • a network device comprising a processor for calling a program from a memory so that the device executes part or all of the steps in the method of the second aspect.
  • a chip comprising a processor for calling a program from a memory so that the chip executes part or all of the steps in the method of the first aspect or the second aspect.
  • a computer-readable storage medium on which a program is stored, wherein the program enables a wireless communication device to execute part or all of the steps in the method of the first aspect or the second aspect.
  • a computer program product comprising a computer program product storing a computer program product.
  • the computer program product may be a non-transitory computer-readable storage medium of a computer program, wherein the computer program is operable to cause a communication device to perform some or all of the steps in the above-mentioned various aspects of the method.
  • the computer program product may be a software installation package.
  • a computer program is provided, wherein the computer program enables a wireless communication device to execute part or all of the steps in the method of the first aspect or the second aspect.
  • the embodiment of the present application helps to improve system performance by using a model (such as the first model) with good generalization capability that can work in multiple deployment scenarios to communicate when the terminal device cannot determine the model information to be used or cannot obtain the model information applicable to the current scenario in a timely manner, such as when the first condition is met.
  • a model such as the first model
  • FIG1 is a wireless communication system applied in an embodiment of the present application.
  • FIG. 2 is a schematic diagram of a neural network applicable to an embodiment of the present application.
  • FIG3 is a schematic diagram of a convolutional neural network applicable to an embodiment of the present application.
  • FIG. 4 is a diagram showing an example of the use of an AI/ML model.
  • FIG5 is a flow chart of a method for wireless communication provided in an embodiment of the present application.
  • FIG6 is a diagram showing an example of the use of an initial model (single-sided) provided in an embodiment of the present application.
  • FIG. 7 is a diagram showing an example of the use of an initial model (double-sided) provided in an embodiment of the present application.
  • FIG8 is a diagram showing an example of the use of a default model (single-sided) provided in an embodiment of the present application.
  • FIG. 9 is a diagram showing an example of the use of a default model (bilateral) provided in an embodiment of the present application.
  • FIG. 10 is a diagram showing an example of the use of another default model (single-sided) provided in an embodiment of the present application.
  • FIG. 11 is a diagram showing an example of the use of another default model (bilateral) provided in an embodiment of the present application.
  • FIG. 12 is a diagram showing an example of the use of another default model (single-sided) provided in an embodiment of the present application.
  • FIG13 is a diagram showing an example of the use of another default model (bilateral) provided in an embodiment of the present application.
  • FIG. 14 is a diagram showing an example of the use of another default model (single-sided) provided in an embodiment of the present application.
  • FIG. 15 is a diagram showing an example of the use of another default model (bilateral) provided in an embodiment of the present application.
  • FIG. 16 is a schematic diagram of the structure of a terminal device according to an embodiment of the present application.
  • FIG. 17 is a schematic diagram of the structure of a network device according to an embodiment of the present application.
  • FIG. 18 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • FIG1 is a wireless communication system 100 used in an embodiment of the present application.
  • the wireless communication system 100 may include a network device 110 and a terminal device 120.
  • the network device 110 may be a device that communicates with the terminal device 120.
  • the network device 110 may provide communication coverage for a specific geographical area, and may communicate with the terminal device 120 located in the coverage area.
  • FIG1 exemplarily shows a network device and two terminals.
  • the wireless communication system 100 may include multiple network devices and each network device may include other number of terminal devices within its coverage area, which is not limited in the embodiments of the present application.
  • the wireless communication system 100 may also include other network entities such as a network controller and a mobility management entity, which is not limited in the embodiments of the present application.
  • network entities such as a network controller and a mobility management entity, which is not limited in the embodiments of the present application.
  • the technical solutions of the embodiments of the present application can be applied to various communication systems, such as: the fifth generation (5th generation, 5G) system or new radio (new radio, NR), long term evolution (long term evolution, LTE) system, LTE frequency division duplex (frequency division duplex, FDD) system, LTE time division duplex (time division duplex, TDD), etc.
  • 5G fifth generation
  • NR new radio
  • long term evolution long term evolution
  • LTE long term evolution
  • LTE frequency division duplex frequency division duplex
  • FDD frequency division duplex
  • TDD time division duplex
  • future communication systems such as the sixth generation mobile communication system, satellite communication system, etc.
  • the terminal device in the embodiment of the present application may also be referred to as user equipment (UE), access terminal, User unit, user station, mobile station, mobile station (MS), mobile terminal (MT), remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device.
  • UE user equipment
  • MS mobile station
  • MT mobile terminal
  • remote station remote terminal
  • mobile device user terminal
  • terminal wireless communication device
  • user agent or user device The terminal device in the embodiment of the present application
  • the terminal device in the embodiment of the present application can be a device that provides voice and/or data connectivity to the user, and can be used to connect people, objects and machines, such as handheld devices with wireless connection functions, vehicle-mounted devices, etc.
  • the terminal device in the embodiment of the present application can be a mobile phone, a tablet computer (Pad), a laptop computer, a handheld computer, a mobile Internet device (MID), a wearable device, a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical surgery, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, etc.
  • the UE can be used to act as a base station.
  • the UE can act as a scheduling entity that provides sidelink signals between UEs in V2X or D2D, etc.
  • a cell phone and a car communicate with each other using sidelink signals.
  • a cell phone and a smart home device communicate with each other without relaying the communication signal through a base station.
  • the network device in the embodiment of the present application may be a device for communicating with a terminal device, and the network device may also be referred to as an access network device or a wireless access network device, such as a base station.
  • the network device in the embodiment of the present application may refer to a radio access network (RAN) node (or device) that connects the terminal device to a wireless network, a core network device, a model monitoring and management device, or an operation administration and maintenance (OAM) device, etc.
  • RAN radio access network
  • OAM operation administration and maintenance
  • the base station may broadly cover the following various names, or be replaced with the following names, such as: Node B (NodeB), evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmission point (TRP), transmitting point (TP), master station MeNB, auxiliary station SeNB, multi-standard wireless (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, base Base band unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), centralized unit-control plane (CU-CP), centralized unit-user plane (CU-UP) positioning node, etc.
  • NodeB Node B
  • eNB evolved NodeB
  • gNB next generation NodeB
  • relay station access point
  • TRP transmission point
  • TRP transmission point
  • TP transmitting point
  • TP master station MeNB
  • auxiliary station SeNB multi-
  • Core network equipment can broadly cover the following various names, or be replaced with the following names, such as: location management function (LMF) network element, network slice selection function (NSSF), authentication server function (AUSF), unified data management (UDM), access and mobility management function (AMF), session management function (SMF), policy control function (PCF), user plane function (UPF), sensing control function (SF), network data analysis (NWDAF) network element.
  • LMF location management function
  • NSSF network slice selection function
  • AUSF authentication server function
  • UDM unified data management
  • AMF access and mobility management function
  • SMF session management function
  • PCF policy control function
  • UPF user plane function
  • sensing control function SF
  • NWDAF network data analysis
  • the base station may be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof.
  • the base station may also refer to a communication module, a modem or a chip for being arranged in the aforementioned device or apparatus.
  • the base station may also be a mobile switching center and a device to device D2D, vehicle-to-everything (V2X), a device that performs the base station function in machine-to-machine (M2M) communication, a network side device in a 6G network, and a device that performs the base station function in a future communication system.
  • the base station may support networks with the same or different access technologies.
  • the embodiments of the present application do not limit the specific technology and specific device form adopted by the network device.
  • the base station may be fixed or mobile.
  • a helicopter or a drone may be configured to act as a mobile base station, and one or more cells may move according to the location of the mobile base station.
  • a helicopter or a drone may be configured to be used as a device for communicating with another base station.
  • Base stations can be fixed or mobile.
  • a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move based on the location of the mobile base station.
  • a helicopter or drone can be configured to act as a mobile base station.
  • the network device in the embodiments of the present application may refer to a CU or a DU, or the network device includes a CU and a DU.
  • the gNB may also include an AAU.
  • the network equipment and terminal equipment can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on the water surface; they can also be deployed on aircraft, balloons and satellites in the air.
  • the embodiments of the present application do not limit the scenarios in which the network equipment and terminal equipment are located.
  • FIG2 is a schematic diagram of a neural network applicable to an embodiment of the present application.
  • the neural network shown in FIG2 can be divided into three categories according to the positions of different layers: an input layer 210, a hidden layer 220, and an output layer 230.
  • the first layer is the input layer 210
  • the last layer is the output layer 230
  • the intermediate layers between the first layer and the last layer are all hidden layers 220.
  • Samples can be input from the input layer 210, processed by the hidden layer 220, and the final result is generated in the output layer 230.
  • each node represents a processing unit, which can be considered to simulate a neuron. Multiple neurons form a layer of a neural network, and multiple layers of information transmission and processing construct an overall neural network.
  • neural network deep learning algorithms have been proposed in recent years, and more hidden layers have been introduced.
  • the learning and processing capabilities of neural networks have been greatly improved, and they have been widely used in pattern recognition, signal processing, optimization combination, anomaly detection, etc.
  • FIG. 3 is a schematic diagram of a convolutional neural network to which an embodiment of the present application is applicable.
  • the basic structure of a convolutional neural network may include: an input layer 310, multiple convolutional layers 320, multiple pooling layers 330, a fully connected layer 340, and an output layer 350.
  • the introduction of convolutional layers 320 and pooling layers 330 effectively controls the sharp increase in network parameters, limits the number of parameters, and taps into the characteristics of local structures, thereby improving the robustness of the algorithm.
  • 5G/6G air interface systems may choose AI/ML models for certain functions and features in appropriate application scenarios to replace traditional non-AI/ML technical solutions.
  • the terminal device can report its own capabilities.
  • the terminal device can report capability information associated with model operation, such as storage resources, computing resources, and power consumption.
  • the 5G ⁇ 6G system can be configured based on the terminal's capability report, such as configuring a set of AI/ML models that can be activated (such as AI/ML model A and AI/ML model B shown in Figure 4).
  • the network device may select and activate an AI/ML model from the AI/ML model set, and communicate based on the AI/ML model.
  • FIG4 exemplarily shows a schematic diagram of selecting AI/ML model A from AI/ML model A and AI/ML model B included in the AI/ML model set.
  • AI/ML models are specially optimized for application scenarios. Different AI/ML models may be suitable for different deployment scenarios.
  • the network device detects that a certain AI/ML model is not working well, the network device can switch to another AI/ML model through indication information. This process is called model switching. For example, as the terminal moves from one cell to another, the deployment scenario may change significantly, and the original AI/ML model is not suitable for the new deployment scenario. After the network device detects performance deterioration, it needs to switch to another model that is adapted to the current deployment scenario. Based on this, the above-mentioned AI/ML model usage process can also include model monitoring and model switching.
  • network devices are required to select models.
  • network devices can indicate the selection of models through connection state indication information, such as radio resource control (RRC) information and downlink control information (DCI).
  • RRC radio resource control
  • DCI downlink control information
  • the terminal device may not be able to obtain a model applicable to the current scenario in a timely manner, thereby affecting the performance of the wireless communication system.
  • the selection of the best AI/ML model requires accurate monitoring of the working status of the AI/ML model by network devices.
  • network devices may not be able to detect in time that the model has entered a poor performance state, and cannot promptly instruct the terminal device to switch to an AI/ML model that is more suitable for the current scenario, causing the system performance to deteriorate.
  • the model library of the terminal device may not contain an AI/ML model that is adapted to the new scenario. It is necessary to download a new model from the network or perform online training on the existing model to adapt it to the new scenario. However, this takes a certain amount of time to achieve, and before the new model is obtained, the performance of the original model may become very poor.
  • the terminal devices may not be able to continue using the original AI/ML model, but they are unable to promptly notify the network devices to switch to another AI/ML model, causing the network side to have different understandings of the AI/ML model used by the terminal devices, resulting in the network side and the terminal devices using incompatible AI/ML models, causing data transmission errors.
  • the terminal device cannot obtain the model information used or cannot obtain the model information applicable to the current scenario in a timely manner, it may not bring performance gains and may even affect system performance.
  • an embodiment of the present application provides a method for wireless communication, which helps to improve system performance by using a model (such as the first model) with good generalization capability that can work in multiple deployment scenarios to communicate when the terminal device cannot determine the model information used or cannot obtain the model information applicable to the current scenario in a timely manner, such as when the first condition is met.
  • a model such as the first model
  • Fig. 5 is a schematic flow chart of a method for wireless communication provided in an embodiment of the present application.
  • the method shown in Fig. 5 may include step S510.
  • step S510 if the first condition is met, the terminal device communicates with the network device based on the first model.
  • the first condition may include that the terminal device is not configured or indicated with model information.
  • the information that the terminal device is not configured or indicated with the model may mean that the network device is unable to configure or indicate model information to the terminal device, such as during the initial access process mentioned above, when the terminal device and the network device have not established an RRC connection, the network device is unable to configure or indicate model information to the network device.
  • the terminal device cannot obtain the model information applicable to the current scenario in a timely manner. Therefore, the above-mentioned first condition may include the expiration of the first timer (also known as overtime) to avoid the deterioration of system performance caused by the terminal device using an inappropriate model.
  • the first timer also known as overtime
  • the timing duration of the first timer can be set according to the needs of the system. For example, the timing duration of the first timer can be determined based on the time that the system can tolerate for possible performance degradation. If the time that the system can tolerate for possible performance degradation is short, the timing duration of the first timer can be shorter. If the time that the system can tolerate for possible performance degradation is longer, the timing duration of the first timer can be longer. For another example, the timing duration of the first timer can be determined based on the changes in the scene in which the terminal device is located. If the scene in which the terminal device is located changes rapidly, the terminal device may need to switch models multiple times to adapt to the changes in the scene.
  • the terminal device may not need to switch models, or the number of model switching times may be small. Based on this, as an example, if the scene in which the terminal device is located changes rapidly, the timing duration of the first timer can be shorter; if the scene in which the terminal device is located changes slowly, or does not change, the timing duration of the first timer can be longer.
  • the scene change of the terminal device may refer to, for example, a location change of the terminal device, a communication quality change, etc.
  • the reasonable design of the timing duration of the first timer can take into account both the switching frequency of the model and the improvement of the system performance.
  • the above-mentioned first timer is associated with the time when the terminal device communicates with the network device based on the second model.
  • the action of the first timer such as start, pause, etc.
  • the start time of the first timer is associated with the time when the terminal device starts to use the second model, or the start time of the first timer is associated with the time when the terminal device receives the indication information (which can be called the first indication information) for instructing the terminal device to use the second model.
  • the pause time of the first timer can be associated with the time when the terminal device receives the first indication information.
  • the first condition may include that the expiration of the first timer may include that before the expiration of the first timer, the terminal device has not received the indication information about the model information sent by the network device, or the first condition may include that the time when the terminal device communicates with the network device based on the second model exceeds the timing duration of the first timer.
  • the first model may be a model based on AI technology, such as an AI/ML model.
  • AI technology such as an AI/ML model.
  • the first model is a model that can work in various deployment scenarios, or the first model is a model that can work well in multiple deployment scenarios, such as a model with high reliability, best generalization ability, and can work in various deployment scenarios.
  • the terminal device can use the first model to communicate with the network device, which helps to obtain better system performance compared to not using the model.
  • the terminal device can use the first model to communicate with the network device, that is, after using a certain model for a long time, the terminal device can automatically switch to (or can be called automatically falling back to) the first model.
  • the above-mentioned automatic fallback to the first model helps to avoid the mismatch between the model and the deployment scenario not being monitored in time, or the misdetection of the model switching instruction, etc., which causes the degradation of system performance.
  • the first model may be different.
  • the first model may include multiple models, and for different first conditions, the terminal device may communicate with the network device based on different models among the multiple models.
  • the first model may include a first initial model. If the first condition is that the terminal device is not configured or the model information is not indicated, the terminal device communicates with the network device based on the first initial model. In this way, compared with not using a model, when the deployment scenario of the terminal device cannot be detected and the network device cannot indicate which model to use for the terminal device, the terminal device communicates with the network device based on the first initial model to obtain better performance.
  • the first model may include a first default model. If the first condition is that the first timer expires, the terminal device communicates with the network device based on the first default model.
  • the terminal device if the terminal device receives the first indication information sent by the network device, the terminal device starts a first timer, and the first indication information can be used to instruct the terminal device to communicate with the network device based on the second model.
  • the second model mentioned here can be a model selected by the network device based on the deployment scenario.
  • the first indication information can be used to instruct the terminal device to communicate with the network device based on the second model, and can also mean: the first indication information is used to instruct the terminal device to use the second model to process data during communication with the network device.
  • the first indication information can be carried in RRC information, DCI indication information or media access control element (MAC CE) control information.
  • RRC information RRC information
  • DCI indication information DCI indication information
  • MAC CE media access control element
  • the deployment scenario may have changed, or the second model may not match the deployment scenario. If the network device does not detect this situation and the terminal device continues to use the second model, system performance will be degraded. If the network device detects this situation, but in some cases, such as when the channel quality is poor, the signaling from the network device instructing the terminal device to switch models may be misdetected, then the network side and the terminal side have different understandings of the selected or activated model, resulting in system performance degradation. Therefore, when the first timer expires, the terminal device can use the first default model to communicate with the network device, that is, the terminal device can switch from the second model to the default model. The first default model helps to avoid degradation of system performance.
  • the terminal device can adjust the model usage policy based on the indication information, or determine the usage timing of the first default model based on the indication information.
  • the terminal device may perform a first operation, where the first operation is associated with the use of the first default model.
  • the first operation may include restarting the first timer, or the first operation may include resetting the first timer.
  • the terminal device receives the first indication information, that is, the second model matches the deployment scenario, or the second model is applicable to the scenario in which the current terminal device communicates with the network device. Therefore, restarting the first timer at this time can extend the use time of the non-first default model, such as the second model, thereby reducing unnecessary model switching, which in turn helps to obtain better adaptability of the model to the deployment scenario and better model reasoning performance.
  • the terminal device may process data using the second model (e.g., based on the instruction information of the network device, the terminal device processes data using the second model).
  • the terminal device communicates with the network device based on the first default model, or the terminal device switches from the second model to the first default model, which may interrupt the data processing process based on the second model, thereby causing performance loss.
  • the first operation may include the terminal device pausing (also known as terminating) the first timer and continuing to process data based on the second model; if the terminal device completes the data processing based on the second model, the terminal device resumes running the first timer.
  • the first operation may include maintaining the operation of the first timer, and when the first timer expires, determining the model used by the terminal device according to whether the data processing based on the second model is completed. For example, when the first timer expires, if the data processing based on the second model of the terminal device has been completed, the terminal device communicates with the network device based on the first default model; if the data processing based on the second model of the terminal device has not been completed, the terminal device communicates with the network device based on the first default model after the data processing based on the second model is completed.
  • the first operation may include increasing the timing duration of the first timer by a first duration. That is, the conflict between the expiration of the first timer and the incomplete model-based data processing is avoided by extending the timing duration of the first timer.
  • the first duration may be a preset duration, or the first duration may be indicated by the network device (or the first duration may be determined based on the indication information sent by the network device). For example, the first duration may be determined based on the RRC configuration information, DCI indication information or MAC CE control information sent by the network device.
  • the first operation may include increasing the timing duration of the first timer by the first duration, and maintaining the operation of the first timer.
  • the model used by the terminal device is determined according to whether the data processing based on the second model is completed, so as to further improve the system performance.
  • the first timer after increasing the timing duration of the first timer by the first duration, when the first timer expires, if the data processing based on the second model of the terminal device has been completed, it communicates with the network device based on the first default model; if the data processing based on the second model of the terminal device has not been completed, it communicates with the network device based on the first default model after the data processing based on the second model is completed.
  • the first model can be determined based on a first rule, and the first rule can be associated with one or more of the following information: the frequency band accessed by the terminal device; the operator identifier of the operator providing services to the terminal device; information on the geographical location of the terminal device; and historical usage information of the first model.
  • the deployment scenario of the model can be predicted to determine a suitable first initial model and/or first default model.
  • the first model may be determined based on historical usage information of the first model.
  • the historical usage information of the first model may include the model used by the terminal device when the first condition is met before the current moment, such as the model used by the terminal device when the first condition is met last time.
  • the first model may be the model that is used the most times among the multiple models.
  • the first model may be the model used by the terminal device when the first condition was met last time. Since the deployment scenario when the first condition was met last time may be closest to the current model deployment scenario, taking the model used by the terminal device when the first condition was met last time as the first model helps to improve the performance of the model while simplifying the model selection process.
  • the first model may include a first initial model and a first default model.
  • the first initial model may be determined based on historical usage information of the first initial model
  • the first default model may be determined based on historical usage information of the first initial model and/or usage information of the first default model.
  • the first default model may be the model used by the terminal device when the first timer expired last time, or the first default model may be the model used by the terminal device when the terminal device was not configured or indicated with model information last time.
  • the information of the first initial model may be carried in system information.
  • the system information may be system information in a physical broadcast channel (PBCH) (such as a master information block (MIB)) or a system information block (SIB).
  • PBCH physical broadcast channel
  • MIB master information block
  • SIB system information block
  • information of the first default model may be carried in RRC configuration information.
  • the information of the first initial model and/or the first default model may also be carried in information sent by other terminal devices. That is, the terminal device may also obtain the information of the first initial model and/or the first default model from other terminal devices, such as through direct communication between terminal devices. For example, when the terminal device switches from communication with other terminal devices to, or is about to switch to communication with a network device, the terminal device may obtain the information of the first initial model from other terminal devices.
  • the first model may be determined based on one or more of the above methods.
  • the method of determining the first model based on the first rule has the lowest priority among the above methods. That is, if the first model can be determined by other methods among the above methods except the method of determining the first model based on the first rule (which may be referred to as the above other methods), the first model is not determined based on the first rule, or, if the first model cannot be determined by any of the above other methods, the first model is determined based on the first rule.
  • the capabilities of different terminal devices may be different, and therefore, the models that different terminal devices can support may also be different.
  • the candidate first models that different terminal devices can support may also be different. Therefore, in some embodiments, the terminal device can report the candidate first models that it can support to the network device in order to determine the first model that matches the capabilities of the terminal device, or to determine the first model used by the terminal device from the candidate first models supported by the terminal device.
  • the candidate first model mentioned here can be a model that can be used by the terminal device when the first condition is met, or a model supported by the capabilities of the terminal device.
  • the information of the candidate first model may be included in the capability information reported by the terminal device.
  • the terminal device may send the first information to the network device, or the network device may receive the first information sent by the terminal device.
  • the first information includes information of models supported by the terminal device, wherein the models supported by the terminal device include the first model.
  • the first information may be a list of models supported by the terminal device.
  • the first model may be a model with a specific identifier among the models supported by the terminal device.
  • the first model may be a model with a specific identifier in the first information.
  • the model identified as A is a candidate first model supported by the terminal device.
  • the model identified as X can be used as the first initial model of the terminal device; the model identified as Y can be used as the first default initial model of the terminal device.
  • the model identified as M can be used as both the first initial model of the terminal device and the first default model of the terminal device.
  • logo X can be used alone or in combination.
  • logo Y can be used in combination.
  • logo M can be used alone.
  • the terminal device may send the second information to the network device, or the network device may receive the second information sent by the terminal device, wherein the second information may be used to indicate the first model.
  • the model used by the network device needs to be associated with the model used by the terminal device (i.e., the case of using a two-sided model).
  • the network device can use the channel state information decompression model associated with the channel state information compression model. Correct communication can be achieved.
  • the first model is associated with the third model, wherein the third model is the model used by the network device under the first condition. That is, when the terminal device uses the first model, the network device can use the third model.
  • the first model includes the first initial model and/or the first default model
  • the third model may include the second initial model and/or the second default model, wherein the first initial model is associated with the second initial model, and the first default model is associated with the second default model. That is, when the terminal device uses the first initial model, the network device may communicate using the second initial model associated with the first initial model; and when the terminal device uses the first default model, the network device may communicate using the second default model associated with the first default model.
  • the association between the first model and the third model may include an association relationship between the first model and the third model, and the association relationship may be configured by the network device.
  • the association relationship between the first initial model and the second initial model, and/or the association relationship between the first default model and the second default model may be configured by the network device.
  • the above-mentioned association relationship can be configured through a model identifier, such as an association relationship of a model ID or a model identifier group.
  • the network device can configure the identifier of the second initial model associated with the identifier of the first initial model, or the network device can configure the identifier of the second default model associated with the identifier of the first default model.
  • the first initial model can use the same model identifier as the second initial model, such as the first initial model is model 1 on the terminal device side, and the second initial model is model 1 on the network device side, to indicate their association relationship.
  • the network device can configure a combined identifier for the first initial model and the second initial model, or the network device can configure a combined identifier for the first default model and the second default model.
  • the terminal device when using the two-side model, if the first condition is met, the terminal device communicates with the network device based on the first model, and correspondingly, the network device communicates with the terminal device based on the third model.
  • the following will introduce the above method from the perspective of the network device. It should be understood that the description on the network device side can correspond to the description on the terminal device side. Therefore, for the parts not described in detail, refer to the method described on the terminal device side.
  • both the terminal device and the network device may be provided with a first timer, and the first timer mentioned in this embodiment may refer to the first timer set on the network device side.
  • the third model may include the second initial model, and if the first condition is that the terminal device is not configured or the model information is not indicated, the network device communicates with the terminal device based on the second initial model. That is, the first initial model is associated with the second initial model, and when the terminal device uses the first initial model, the network device can use the second initial model.
  • the third model may include the second default model, and if the first condition is that the first timer expires, the network device communicates with the terminal device based on the second default model.
  • the second default model is associated with the first default model, and when the terminal device uses the first default model, the network device can use the second default model.
  • the network device if the network device sends first indication information to the terminal device, the network device starts a first timer, and the first indication information is used to instruct the terminal device to communicate with the network device based on the second model.
  • the network device if the network device sends first indication information to the terminal device during the operation of the first timer, the network device performs a first operation, and the first operation is associated with the use of the second default model.
  • the first operation includes restarting a first timer.
  • the first operation includes: the network device suspending the first timer; if the terminal device completes data processing based on the second model, the network device resumes running the first timer.
  • the first operation includes: maintaining the operation of the first timer, and when the first timer expires, if the data processing based on the second model of the terminal device has been completed, communicating with the terminal device based on the second default model; if the data processing based on the second model of the terminal device has not been completed, communicating with the terminal device based on the second default model after the data processing based on the second model is completed.
  • the network device can, for example, determine the activation timing of the second default model based on the completion time of data processing of the fourth model.
  • the fourth model is associated with the second model.
  • the network device can determine the activation timing of the second default model based on the completion time of data processing of the fourth model.
  • the completion time of the data processing based on the second model is determined based on the data scheduling situation on the network device side.
  • the first operation includes increasing the timing duration of the first timer by a first duration.
  • the first duration is a preset duration, or the first duration is indicated by the network device.
  • start and pause of the first timer on the network device side is associated with the sending time of the first indication information
  • start and pause of the first timer on the terminal device side is associated with the receiving time of the first indication information.
  • the first time delay may cause the first timer on the terminal device side to be out of sync with the first timer on the network device side, which may lead to inconsistency in the time when the terminal device and the network device switch models. That is to say, within the first time delay range, the model used by the terminal device may not be associated with the model used by the network device, which may lead to a decrease in communication quality.
  • the model switching timing of the terminal device and/or the network device may be determined based on the expiration of the first timer and the time correction amount (or time interval).
  • the terminal device can switch to the first default model when the first timer expires, and the network device can switch to the second default model at a time interval of the first time correction amount after the first timer expires. Since the time when the network device sends the first indication information is earlier than the time when the terminal device receives the first indication information, the start time and the expiration time of the first timer on the network device side are respectively earlier than the start time and the expiration time of the first timer on the terminal device side. Therefore, the network device switches to the second default model at a time interval of the first time correction amount after the first timer expires, which helps to eliminate the influence of the above-mentioned first time delay on the timing of model switching.
  • the terminal device may switch to the first default model at a second time correction amount after the first timer expires, and the network device may switch to the second default model at a third time correction amount after the first timer expires, wherein the third time correction amount is greater than the second time correction amount.
  • the terminal device may switch to the first default model at a second time correction amount after the first timer expires, and the network device may switch to the second default model at a third time correction amount after the first timer expires, wherein the third time correction amount is greater than the second time correction amount.
  • the terminal device can pause the first timer, and continue to run the first timer after the terminal device completes the data processing based on the second model. Since the terminal device can obtain the time when it completes the data processing based on the second model, but the network device cannot obtain or cannot accurately obtain the time when the terminal device completes the data processing based on the second model, the network device cannot determine the time to resume the operation of the first timer, or there is an error between the time when the network device resumes the operation of the first timer and the time when the terminal device resumes the operation of the first timer. Therefore, there may be an error between the time when the first timer on the network device side expires and the time when the first timer on the terminal device side expires.
  • the terminal device can send a second indication message to the network device when completing data processing based on the second model.
  • the network device can resume running the first timer when receiving the second indication message sent by the terminal device.
  • the second time delay is greater than the first time delay, such as the second time delay is twice the first time delay. Therefore, the terminal device can switch to the first default model when the first timer expires, and the network device can switch to the second default model at a time interval of the fourth time correction amount after the first timer expires, which helps to improve the accuracy of the model switching timing.
  • the fourth time correction amount can be greater than the first time correction amount, such as the fourth time correction amount is twice the first time correction amount.
  • the network device may determine the time to resume running the first timer based on the data scheduling situation of the network device, for ease of implementation.
  • the network device may switch to the second default model when the first timer expires, and the terminal device may switch to the first default model at a time interval of the first time correction amount before the first timer expires.
  • the specific implementation method is similar to the method of correcting the timing of switching models of the network device mentioned above, and for the sake of brevity, it will not be repeated here.
  • One of the first time correction value, the second time correction value, the third time correction value and the fourth time correction value can be determined based on the transmission delay between the terminal device and the network device in the current deployment scenario.
  • the terminal device can determine the above transmission delay based on the sending time carried in the first indication information and the time when the terminal device receives the first indication information.
  • One or more of the first time correction value, the second time correction value, the third time correction value and the fourth time correction value may be pre-configured or dynamically indicated before the model is switched.
  • the fourth time correction value may be carried in the second indication information.
  • the above-mentioned network device performs certain actions based on the first timer, which refers to the actions performed by the network device based on the first timer on the network device side; the above-mentioned terminal device performs certain actions based on the first timer, which refers to the actions performed by the terminal device based on the first timer on the terminal device side.
  • first timer can be an increasing timer or a decreasing timer, and this application does not limit this.
  • the use of a certain model or switching to a certain model mentioned in the embodiments of the present application may refer to the first device communicating with the second device based on the model, wherein the first device may be a terminal device or a network device, and the second device may be a terminal device or a network device.
  • Example 1 a case where a terminal device and a network device have not established an RRC connection (i.e., the first condition is that the terminal device has not been configured or has not been indicated with model information) is taken as an example, and the method provided in the embodiment of the present application is introduced in combination with Figures 6 and 7.
  • the terminal device has not yet established an RRC connection, and cannot determine which AI/ML model to use based on the RRC configuration, DCI, MAC CE, etc. on the network device side, and cannot detect the current deployment scenario, so it is impossible to determine which model is best adapted to the current deployment scenario. Therefore, an AI/ML model (i.e., the first initial model) can be used that can provide reliable performance, has the best generalization capability, and can definitely work in any deployment scenario.
  • the ID of the first initial model can be determined according to the first rule or the system information on the network device side.
  • the system can detect the deployment environment of the terminal device, so as to select the best AI/ML model (such as the second model) that is adapted to the current deployment environment, or switch to another AI/ML model (such as the second model) according to the indication information of the network device (RRC configuration information, DCI indication information, MAC CE control information, etc.). As the deployment environment changes, it can also switch to other AI/ML models according to the indication information of the network device.
  • the initial model on the network device side and the initial model on the terminal device side must also be used accordingly.
  • the terminal device has not yet established an RRC connection
  • the terminal device side uses the first initial model
  • the network device side also uses the corresponding second initial model.
  • the network device side instructs the terminal device to switch to the second model, and the network device also switches to the corresponding fourth model at the same time.
  • an AI/ML model can also be used to obtain better performance than a non-AI/ML model.
  • Embodiment 2 takes the first condition that the first timer expires as an example, and introduces the method provided in the embodiment of the present application in combination with FIG. 8 and FIG. 9 .
  • the terminal device in the RRC connection state, can select the best AI/ML model (such as the second model) adapted to the deployment scenario according to the instructions of the network device side.
  • a timer i.e., the first timer mentioned above
  • the default model on the network device side and the AI/ML model on the terminal device side must also be used accordingly.
  • the network device side when the terminal device side uses the second model, the network device side also uses the corresponding fourth model.
  • the terminal device automatically switches to the first default model, and the network device also automatically switches to the second default model.
  • the system after the system has been using an AI/ML model that is adapted to a specific deployment scenario for a long time, it can automatically fall back to a default AI/ML model that is adapted to various deployment scenarios and has good generalization performance.
  • This can avoid the mismatch between the AI/ML model and the deployment scenario not being detected by the system, or can avoid the network device side and the terminal device side having different understandings of the activated AI/ML due to the misdetection of the signaling indicating the model switching of the network device, thereby causing the system performance to degrade.
  • both the network device side and the terminal device side will consider that they should fall back to the default AI/ML model, thereby ensuring the reliability of the system performance.
  • Example 3 the first condition that the first timer expires is taken as an example, and the method provided in the embodiment of the present application is introduced in combination with Figures 10 and 11.
  • the timer should expire after time T and the model should be rolled back to the default model.
  • the default model on the network device side and the default model on the terminal device side should also be used accordingly.
  • the network device side after the second model on the terminal device side that is better adapted to the deployment scenario is activated, the network device side also uses the corresponding fourth model.
  • the timer should expire after time T, and when the timer expires, the models on the terminal device side and the network device side are rolled back to the default model.
  • the system when the system confirms that a non-default model is well adapted to the deployment scenario, it can reset the AI/ML model fallback timer to extend the usage time of the non-default AI/ML model, thereby reducing unnecessary frequent fallbacks and achieving better adaptability to the deployment scenario and better AI/ML model reasoning performance.
  • Embodiment 4 takes the first condition that the first timer expires as an example, and introduces the method provided in the embodiment of the present application in combination with Figures 12 to 15.
  • Selecting and activating an AI/ML model does not mean that the terminal device and network device will always use this model to process data.
  • the timer expires, if neither the terminal device nor the network device is using a non-default AI/ML model to process the data, you can immediately fall back to the default AI/ML model.
  • the terminal device and/or the network device is using a non-default AI/ML model to process the data, immediately falling back to the default AI/ML model will cause the data processing process to be interrupted and result in performance loss. Therefore, when the terminal device and/or the network device is using a non-default AI/ML model to process the data, the AI/ML model fallback timer can be terminated and the fallback time of the timer can be delayed.
  • the timer should expire after time T and the model will be rolled back to the default model.
  • the terminal device starts to use the second model to process data.
  • the terminal device automatically switches to the first default model.
  • the default model on the network device side and the default model on the terminal device side should also be used accordingly.
  • the network device side after the second model on the terminal device side that is better adapted to the deployment scenario is activated, the network device side also uses the corresponding fourth model.
  • the timer should expire after time T, and the models on the terminal device side and the network device side are both rolled back to the default model.
  • the network device and/or the terminal device begins to use the fourth model and/or the second model to process data.
  • Another method to avoid model fallback during model-based data processing is to delay fallback after the timer expires.
  • the terminal device starts to process data using the second model, but the timer is not stopped and operates normally until the timer expires. If the terminal device has not yet completed the data processing using the second model, it will not fall back to the first default model temporarily, but wait until the data is processed using the second model, and then the terminal device will automatically switch to the first default model. If the terminal device has completed the data processing using the second model when the timer expires, it will fall back to the first default model.
  • the default model on the network device side and the AI/ML model on the terminal device side must also be used accordingly.
  • the terminal device and/or the network device side begins to use the second model and/or the fourth model to process data, but the timer is not terminated and operates normally until it expires. If the terminal device and/or the network device has not completed the data processing using the second model and the fourth model at this time, it will not fall back to the default AI/ML model for the time being. Instead, after the data is processed using the second model and the fourth model, the terminal device and the network device side automatically switch to the first default model and the second default model.
  • FIG16 is a schematic diagram of the structure of a terminal device according to an embodiment of the present application.
  • the terminal device 1600 may include a communication unit 1610 .
  • the communication unit 1610 is used to communicate with the network device based on a first model if a first condition is met, wherein the first condition includes that the terminal device is not configured or not indicated with model information, and/or that a first timer expires, and the first timer is associated with the time at which the terminal device communicates with the network device based on a second model.
  • the first model includes a first initial model, and if the first condition is that the terminal device is not configured or is not indicated with model information, the terminal device communicates with the network device based on the first initial model.
  • the first model includes a first default model, and if the first condition is that the first timer expires, the terminal device communicates with the network device based on the first default model.
  • the device further includes: a starting unit, configured to start the first timer if the terminal device receives first indication information sent by the network device, wherein the first indication information is used to instruct the terminal device to communicate with the network device based on the second model.
  • the device further includes: an execution unit, configured to cause the terminal device to execute a first operation if the terminal device receives the first indication information sent by the network device during the operation of the first timer, and the first operation is associated with the use of the first default model.
  • the first operation includes restarting the first timer.
  • the first operation includes: the terminal device suspending the first timer and continuing Perform data processing based on the second model; if the terminal device completes the data processing based on the second model, the terminal device resumes running the first timer.
  • the first operation includes: maintaining the operation of the first timer, and when the first timer expires, if the data processing based on the second model of the terminal device has been completed, communicating with the network device based on the first default model; if the data processing based on the second model of the terminal device has not been completed, communicating with the network device based on the first default model after the data processing based on the second model is completed.
  • the first operation includes increasing the timing duration of the first timer by a first duration.
  • the first duration is a preset duration, or the first duration is indicated by the network device.
  • the first model is determined based on a first rule, and the first rule is associated with one or more of the following information: a frequency band accessed by the terminal device; an operator identifier of an operator providing services to the terminal device; information on the geographical location of the terminal device; and historical usage information of the first model.
  • information of the first initial model is carried in system information or information sent by other terminal devices.
  • information of the first default model is carried in RRC configuration information, or information sent by other terminal devices.
  • the device further includes: a first sending unit, configured to send first information to the network device, wherein the first information includes information of models supported by the terminal device, wherein the models supported by the terminal device include the first model.
  • a first sending unit configured to send first information to the network device, wherein the first information includes information of models supported by the terminal device, wherein the models supported by the terminal device include the first model.
  • the first model is a model with a specific identifier among the models supported by the terminal device.
  • the device further includes: a second sending unit, configured to send second information to the network device, wherein the second information is used to indicate the first model.
  • the first model is associated with a third model, wherein the third model is a model used by the network device under the first condition.
  • the associating of the first model with the third model includes an association relationship between the first model and the third model, and the association relationship is configured by the network device.
  • FIG17 is a schematic diagram of the structure of a network device according to an embodiment of the present application.
  • the network device 1700 may include a communication unit 1710 .
  • Communication unit 1710 is used to communicate with the terminal device based on a third model if a first condition is met, the first condition including that the terminal device is not configured or is not indicated with model information, and/or that a first timer expires; wherein the first model is associated with the third model, the first model is the model used by the terminal device when the first condition is met, and the first timer is associated with the time when the terminal device communicates with the network device based on the second model.
  • the third model includes a second initial model, and if the first condition is that the terminal device is not configured or is not indicated with model information, the network device communicates with the terminal device based on the second initial model.
  • the third model includes a second default model, and if the first condition is that the first timer expires, the network device communicates with the terminal device based on the second default model.
  • the device further includes: a starting unit, configured to start the first timer if the network device sends first indication information to the terminal device, wherein the first indication information is used to instruct the terminal device to communicate with the network device based on the second model.
  • the device further comprises: an execution unit, configured to execute a first operation if the network device sends the first indication information to the terminal device during the operation of the first timer, wherein the first operation is associated with the use of the second default model.
  • the first operation includes restarting the first timer.
  • the first operation includes: the network device pausing the first timer; if the terminal device completes data processing based on the second model, the network device resumes running the first timer.
  • the first operation includes: maintaining the operation of the first timer, and when the first timer expires, if the data processing based on the second model of the terminal device has been completed, communicating with the terminal device based on the second default model; if the data processing based on the second model of the terminal device has not been completed, communicating with the terminal device based on the second default model after the data processing based on the second model is completed.
  • the first operation includes increasing the timing duration of the first timer by a first duration.
  • the first duration is a preset duration, or the first duration is indicated by the network device.
  • the first model includes a first initial model, and the first initial model is used by the terminal device when the first condition is that the terminal device is not configured or is not indicated with model information.
  • the first model comprises a first default model, and the first default model is used by the terminal device when the first condition is the expiration of a first timer.
  • the first model is determined based on a first rule, and the first rule is associated with one or more of the following information: a frequency band accessed by the terminal device; an operator identifier of an operator providing services to the terminal device; information on the geographical location of the terminal device; and historical usage information of the first model.
  • information of the first initial model is carried in system information or information sent by other terminal devices.
  • information of the first default model is carried in RRC configuration information, or information sent by other terminal devices.
  • the device also includes: a first receiving unit, used to receive first information sent by the terminal device, the first information including information of models supported by the terminal device, wherein the models supported by the terminal device include a first model, and the first model is the model used by the terminal device under the first condition.
  • a first receiving unit used to receive first information sent by the terminal device, the first information including information of models supported by the terminal device, wherein the models supported by the terminal device include a first model, and the first model is the model used by the terminal device under the first condition.
  • the first model is a model with a specific identifier among the models supported by the terminal device.
  • the device further includes: a second receiving unit, configured to receive second information sent by the terminal device, wherein the second information is used to indicate the first model.
  • the associating of the first model with the third model includes an association relationship between the first model and the third model, and the association relationship is configured by the network device.
  • the sending unit, receiving unit, and communication unit mentioned above may be a transceiver 1830.
  • the terminal device 1600 and the network device 1700 may further include a processor 1810 and a memory 1820, as specifically shown in FIG. 18 .
  • FIG18 is a schematic structural diagram of a communication device according to an embodiment of the present application.
  • the dotted lines in FIG18 indicate that the unit or module is optional.
  • the device 1800 may be used to implement the method described in the above method embodiment.
  • the device 1800 may be a chip or a terminal device or a network device.
  • the device 1800 may include one or more processors 1810.
  • the processor 1810 may support the device 1800 to implement the method described in the above method embodiment.
  • the processor 1810 may be a general-purpose processor or a special-purpose processor.
  • the processor may be a central processing unit (CPU).
  • the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • DSP digital signal processor
  • ASIC application specific integrated circuits
  • FPGA field programmable gate arrays
  • a general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
  • the apparatus 1800 may further include one or more memories 1820.
  • the memory 1820 stores a program, which can be executed by the processor 1810, so that the processor 1810 executes the method described in the above method embodiment.
  • the memory 1820 may be independent of the processor 1810 or integrated in the processor 1810.
  • the apparatus 1800 may further include a transceiver 1830.
  • the processor 1810 may communicate with other devices or For example, the processor 1810 can transmit and receive data with other devices or chips through the transceiver 1830.
  • the present application also provides a computer-readable storage medium for storing a program.
  • the computer-readable storage medium can be applied to a terminal or network device provided in the present application, and the program enables a computer to execute the method performed by the terminal device or network device in each embodiment of the present application.
  • the embodiment of the present application also provides a computer program product.
  • the computer program product includes a program.
  • the computer program product can be applied to the terminal or network device provided in the embodiment of the present application, and the program enables the computer to execute the method performed by the terminal or network device in each embodiment of the present application.
  • the embodiment of the present application also provides a computer program.
  • the computer program can be applied to the terminal device or network device provided in the embodiment of the present application, and the computer program enables a computer to execute the method executed by the terminal device or network device in each embodiment of the present application.
  • the "indication" mentioned can be a direct indication, an indirect indication, or an indication of an association relationship.
  • a indicates B which can mean that A directly indicates B, for example, B can be obtained through A; it can also mean that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also mean that there is an association relationship between A and B.
  • the term “include” may refer to direct inclusion or indirect inclusion.
  • the term “include” in the embodiments of the present application may be replaced with “indicates” or “is used to determine”.
  • “A includes B” may be replaced with “A indicates B” or "A is used to determine B”.
  • B corresponding to A means that B is associated with A, and B can be determined according to A.
  • determining B according to A does not mean determining B only according to A, and B can also be determined according to A and/or other information.
  • the term "corresponding" may indicate that there is a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship of indication and being indicated, configuration and being configured, etc.
  • pre-definition or “pre-configuration” can be implemented by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in a device (for example, including a terminal device and a network device), and the present application does not limit the specific implementation method.
  • pre-definition can refer to what is defined in the protocol.
  • the “protocol” may refer to a standard protocol in the communication field, for example, it may include an LTE protocol, an NR protocol, and related protocols used in future communication systems, and the present application does not limit this.
  • the term "and/or" is only a description of the association relationship of the associated objects, indicating that there can be three relationships.
  • a and/or B can represent: A exists alone, A and B exist at the same time, and B exists alone.
  • the character "/" in this article generally indicates that the associated objects before and after are in an "or" relationship.
  • the size of the serial numbers of the above-mentioned processes does not mean the order of execution.
  • the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
  • the disclosed systems, devices and methods can be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.
  • Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separate.
  • the components may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.
  • the computer-readable storage medium may be any available medium that can be read by a computer or a data storage device such as a server or data center that includes one or more available media integrated.
  • the available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.
  • a magnetic medium e.g., a floppy disk, a hard disk, a magnetic tape
  • an optical medium e.g., a digital video disc (DVD)
  • DVD digital video disc
  • SSD solid state disk

Landscapes

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

Abstract

提供了一种用于无线通信的方法及设备,该方法包括:若满足第一条件,终端设备基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。本申请实施例通过在终端设备无法确定所使用的模型信息或者无法及时获取适用于当前场景的模型信息时,如满足第一条件时,使用一个泛化能力较好,可以在多种部署场景工作的模型(如第一模型)进行通信,有助于改善系统性能。

Description

用于无线通信的方法及设备 技术领域
本申请涉及通信技术领域,并且更为具体地,涉及用于无线通信的方法及设备。
背景技术
相比于传统通信技术,基于人工智能(artificial intelligence,AI)模型的通信技术可能会带来一定的性能增益。但是在终端设备无法获取所使用的模型信息或者无法及时获取适用于当前场景的模型信息时,可能无法带来性能增益,甚至影响系统性能。
发明内容
本申请提供一种用于无线通信的方法及设备,下面对本申请涉及的各个方面进行介绍。
第一方面,提供了一种用于无线通信的方法,包括:若满足第一条件,终端设备基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
第二方面,提供一种用于无线通信的方法,包括:若满足第一条件,网络设备基于第三模型与终端设备进行通信,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期;其中,第一模型与所述第三模型关联,所述第一模型为满足所述第一条件时所述终端设备所使用的模型,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
第三方面,提供一种终端设备,包括:通信单元,用于若满足第一条件,基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
第四方面,提供一种网络设备,包括:通信单元,用于若满足第一条件,基于第三模型与终端设备进行通信,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期;其中,第一模型与所述第三模型关联,所述第一模型为满足所述第一条件时所述终端设备所使用的模型,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
第五方面,提供一种终端设备,包括存储器和处理器,所述存储器用于存储程序,所述处理器用于调用所述存储器中的程序,以使所述设备执行如第一方面的方法中的部分或全部步骤。
第六方面,提供一种网络设备,包括存储器和处理器,所述存储器用于存储程序,所述处理器用于调用所述存储器中的程序,以使所述设备执行如第二方面的方法中的部分或全部步骤。
第七方面,提供一种终端设备,包括处理器,用于从存储器中调用程序,以使所述设备执行如第一方面的方法中的部分或全部步骤。
第八方面,提供一种网络设备,包括处理器,用于从存储器中调用程序,以使所述设备执行如第二方面的方法中的部分或全部步骤。
第九方面,提供了一种芯片,包括处理器,用于从存储器中调用程序,以使所述芯片执行如第一方面或第二方面的方法中的部分或全部步骤。
第十方面,提供了一种计算机可读存储介质,其上存储有程序,所述程序使得无线通信设备执行如第一方面或第二方面的方法中的部分或全部步骤。
第十一方面,提供了一种计算机程序产品,其中,所述计算机程序产品包括存储了计 算机程序的非瞬时性计算机可读存储介质,所述计算机程序可操作来使通信设备执行上述各个方面的方法中的部分或全部步骤。在一些实现方式中,该计算机程序产品可以为一个软件安装包。
第十二方面,提供一种计算机程序,所述计算机程序使得无线通信设备执行如第一方面或第二方面的方法中的部分或全部步骤。
本申请实施例通过在终端设备无法确定所使用的模型信息或者无法及时获取适用于当前场景的模型信息时,如满足第一条件时,使用一个泛化能力较好,可以在多种部署场景工作的模型(如第一模型)进行通信,有助于改善系统性能。
附图说明
图1是本申请实施例应用的无线通信系统。
图2是本申请实施例适用的神经网络的示意图。
图3是本申请实施例适用的卷积神经网络的示意图。
图4是一种AI/ML模型的使用示例图。
图5是本申请实施例提供的用于无线通信的方法的流程示意图。
图6是本申请实施例提供的一种初始模型(单侧)的使用示例图。
图7是本申请实施例提供的一种初始模型(双侧)的使用示例图。
图8是本申请实施例提供的一种缺省模型(单侧)的使用示例图。
图9是本申请实施例提供的一种缺省模型(双侧)的使用示例图。
图10是本申请实施例提供的另一种缺省模型(单侧)的使用示例图。
图11是本申请实施例提供的另一种缺省模型(双侧)的使用示例图。
图12是本申请实施例提供的又一种缺省模型(单侧)的使用示例图。
图13是本申请实施例提供的又一种缺省模型(双侧)的使用示例图。
图14是本申请实施例提供的再一种缺省模型(单侧)的使用示例图。
图15是本申请实施例提供的再一种缺省模型(双侧)的使用示例图。
图16是本申请实施例的一种终端设备的结构示意图。
图17是本申请实施例的一种网络设备的结构示意图。
图18是本申请实施例的通信装置的示意性结构图。
具体实施方式
下面将结合附图,对本申请中的技术方案进行描述。
图1是本申请实施例应用的无线通信系统100。该无线通信系统100可以包括网络设备110和终端设备120。网络设备110可以是与终端设备120通信的设备。网络设备110可以为特定的地理区域提供通信覆盖,并且可以与位于该覆盖区域内的终端设备120进行通信。
图1示例性地示出了一个网络设备和两个终端,可选地,该无线通信系统100可以包括多个网络设备并且每个网络设备的覆盖范围内可以包括其它数量的终端设备,本申请实施例对此不做限定。
可选地,该无线通信系统100还可以包括网络控制器、移动管理实体等其他网络实体,本申请实施例对此不作限定。
应理解,本申请实施例的技术方案可以应用于各种通信系统,例如:第五代(5th generation,5G)系统或新无线(new radio,NR)、长期演进(long term evolution,LTE)系统、LTE频分双工(frequency division duplex,FDD)系统、LTE时分双工(time division duplex,TDD)等。本申请提供的技术方案还可以应用于未来的通信系统,如第六代移动通信系统,又如卫星通信系统,等等。
本申请实施例中的终端设备也可以称为用户设备(user equipment,UE)、接入终端、 用户单元、用户站、移动站、移动台(mobile station,MS)、移动终端(mobile terminal,MT)、远方站、远程终端、移动设备、用户终端、终端、无线通信设备、用户代理或用户装置。本申请实施例中的终端设备可以是指向用户提供语音和/或数据连通性的设备,可以用于连接人、物和机,例如具有无线连接功能的手持式设备、车载设备等。本申请的实施例中的终端设备可以是手机(mobile phone)、平板电脑(Pad)、笔记本电脑、掌上电脑、移动互联网设备(mobile internet device,MID)、可穿戴设备,虚拟现实(virtual reality,VR)设备、增强现实(augmented reality,AR)设备、工业控制(industrial control)中的无线终端、无人驾驶(self driving)中的无线终端、远程手术(remote medical surgery)中的无线终端、智能电网(smart grid)中的无线终端、运输安全(transportation safety)中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端等。可选地,UE可以用于充当基站。例如,UE可以充当调度实体,其在V2X或D2D等中的UE之间提供侧行链路信号。比如,蜂窝电话和汽车利用侧行链路信号彼此通信。蜂窝电话和智能家居设备之间通信,而无需通过基站中继通信信号。
本申请实施例中的网络设备可以是用于与终端设备通信的设备,该网络设备也可以称为接入网设备或无线接入网设备,如网络设备可以是基站。本申请实施例中的网络设备可以是指将终端设备接入到无线网络的无线接入网(radio access network,RAN)节点(或设备)、核心网设备、模型监测管理设备或者操作管理维护(operation administration and maintenance,OAM)设备等。基站可以广义的覆盖如下中的各种名称,或与如下名称进行替换,比如:节点B(NodeB)、演进型基站(evolved NodeB,eNB)、下一代基站(next generation NodeB,gNB)、中继站、接入点、传输点(transmitting and receiving point,TRP)、发射点(transmitting point,TP)、主站MeNB、辅站SeNB、多制式无线(MSR)节点、家庭基站、网络控制器、接入节点、无线节点、接入点(access point,AP)、传输节点、收发节点、基带单元(base band unit,BBU)、射频拉远单元(remote radio unit,RRU)、有源天线单元(active antenna unit,AAU)、射频头(remote radio head,RRH)、中心单元(central unit,CU)、分布式单元(distributed unit,DU)、集中单元-控制面(centralized unit-control plane,CU-CP)、集中单元-用户面(centralized unit-user plane,CU-UP)定位节点等。核心网设备可以广义的覆盖如下中的各种名称,或与如下名称进行替换,比如:位置管理功能(location management function,LMF)网元、网络切片选择功能(network slice selection function,NSSF)、身份验证服务器功能(authentication server function,AUSF)、统一数据管理(unified data management,UDM)、接入和移动性管理功能(access and mobility management function,AMF)、会话管理功能(session management function,SMF)、策略控制功能(policy control function,PCF)、用户面功能(user plane function,UPF)、感知控制功能(sensing function,SF)、网络数据分析(network data analytics function,NWDAF)网元。
基站可以是宏基站、微基站、中继节点、施主节点或类似物,或其组合。基站还可以指用于设置于前述设备或装置内的通信模块、调制解调器或芯片。基站还可以是移动交换中心以及设备到设备D2D、车辆外联(vehicle-to-everything,V2X)、机器到机器(machine-to-machine,M2M)通信中承担基站功能的设备、6G网络中的网络侧设备、未来的通信系统中承担基站功能的设备等。基站可以支持相同或不同接入技术的网络。本申请的实施例对网络设备所采用的具体技术和具体设备形态不做限定。基站可以是固定的,也可以是移动的。例如,直升机或无人机可以被配置成充当移动基站,一个或多个小区可以根据该移动基站的位置移动。在其他示例中,直升机或无人机可以被配置成用作与另一基站通信的设备。
基站可以是固定的,也可以是移动的。例如,直升机或无人机可以被配置成充当移动基站,一个或多个小区可以根据该移动基站的位置移动。在其他示例中,直升机或无人机 可以被配置成用作与另一基站通信的设备。
在一些部署中,本申请实施例中的网络设备可以是指CU或者DU,或者,网络设备包括CU和DU。gNB还可以包括AAU。
网络设备和终端设备可以部署在陆地上,包括室内或室外、手持或车载;也可以部署在水面上;还可以部署在空中的飞机、气球和卫星上。本申请实施例中对网络设备和终端设备所处的场景不做限定。
应理解,本申请中的通信设备的全部或部分功能也可以通过在硬件上运行的软件功能来实现,或者通过平台(例如云平台)上实例化的虚拟化功能来实现。
AI
近年来,以神经网络为代表的人工智能研究在很多领域都取得了非常大的成果,其也将在未来很长一段时间内在人们的生产生活中起到重要的作用。
图2是本申请实施例适用的神经网络的示意图。图2所示的神经网络按照不同层的位置划分可以分为三类:输入层210,隐藏层220和输出层230。一般来说,第一层是输入层210、最后一层是输出层230,第一层和最后一层之间的中间层都是隐藏层220。样本可以从输入层210输入,经过隐藏层220的处理,最后的结果在输出层230产生。在这其中,各个节点代表一个处理单元,可以认为是模拟了一个神经元,多个神经元组成一层神经网络,多层的信息传递与处理构造出一个整体的神经网络。
随着神经网络研究的不断发展,近年来又提出了神经网络深度学习算法,较多的隐层被引入。通过多隐层的神经网络逐层训练进行特征学习,极大地提升了神经网络的学习和处理能力,并在模式识别、信号处理、优化组合、异常探测等方面广泛被应用。
同样,随着深度学习的发展,卷积神经网络也被进一步研究。图3是本申请实施例适用的卷积神经网络的示意图。卷积神经网络的基本结构可以包括:输入层310、多个卷积层320、多个池化层330、全连接层340及输出层350。卷积层320和池化层330的引入,有效地控制了网络参数的剧增,限制了参数的个数并挖掘了局部结构的特点,提高了算法的鲁棒性。
当前,业界已经开始对用于5G、6G空中接口的AI/机器学习(machine learning,ML)模型开展研究,认为针对5G/6G空中接口的某些功能(functionality)、特性(feature),基于AI/ML模型的技术方案可能比传统基于非AI/ML的技术方案获得一定的性能增益。因此在未来的5G/6G空中接口系统可能在适合的应用场景下,针对某个功能、特性,选择AI/ML模型,替代传统非AI/ML的技术方案。
下面结合图4对无线通信系统中AI/ML模型的使用流程进行简单的介绍。
首先,终端设备可以上报自身的能力。例如,终端设备可以上报与模型运行关联的能力信息,如存储资源、算力资源以及电量等。
其次,如果针对一个功能、特性存在一个AI/ML模型,5G\6G系统可以根据终端的能力上报进行配置,如配置一个可供被激活的AI/ML模型集合(如图4中所示的AI/ML模型A和AI/ML模型B)。
然后,可以由网络设备从AI/ML模型集合中选择、激活一个AI/ML模型,并基于该AI/ML模型进行通信。图4示例性的给出了从AI/ML模型集合中包括的AI/ML模型A和AI/ML模型B中选择了AI/ML模型A的示意图。
针对应用场景专门优化的AI/ML模型,不同的AI/ML模型可能适用于不同的部署场景,当网络设备监测到某个AI/ML模型工作状态不佳的时候,网络设备可以通过指示信息切换到另一个AI/ML模型,这个过程称为模型切换(model switching)。例如,随着终端的位置移动,从一个小区切换到另一个小区,部署场景可能发生明显变化,原来使用的AI/ML模型不适应新的部署场景。网络设备检测到性能恶化后,需要切换到另一种适配当前部署场景的模型。基于此,在上述AI/ML模型使用流程中,还可以包括模型监测以及模型切换等。
可以看出,在模型,如AI/ML模型的使用过程中,需要网络设备来选择模型,如网络设备可以通过连接态的指示信息,如无线资源控制(radio resource control,RRC)信息、下行控制信息(downlink control information,DCI)来指示模型的选择。如此一来,在某些状态下(如初始接入阶段),网络设备还没有进入RRC连接状态,可能无法通过RRC配置或DCI来指示模型。或者说,在终端设备未被配置或未被指示模型信息时,无法确定终端设备的使用模型。
另外,在一些情况下,终端设备可能无法及时获取适用当前场景的模型,从而影响无线通信系统的性能。
例如,最佳的AI/ML模型的选择,需要基于网络设备对AI/ML模型工作状态的精确监测。但是受限于模型性能监测的复杂度、不准确性、非实时性,网络设备可能无法及时发现模型进入性能较差的工作状态,不能及时指示终端设备切换到更适合当前场景的AI/ML模型,使系统的性能恶化。
又如,对于终端设备从未遇到过的新的部署场景,即使及时发现了原AI/ML模型不再适用,但终端设备的模型库中也不一定存有与新场景适配的AI/ML模型,需要从网络下载新的模型,或对现有的模型进行在线训练(online training),使其适应新的场景。但这都需要一定时间才能实现,在获得新模型之前,原有模型的性能可能变得很差。
又如,终端设备由于自身能力的变化(如算力资源变少、存储空间下降、电能不足),可能无法继续使用原来的AI/ML模型,但又无法及时通知网络设备切换到另一个AI/ML模型,造成网络侧对终端设备使用的AI/ML模型产生不同的理解,导致网络侧和终端设备使用不适配的AI/ML模型,造成数据传输错误。
因此,在终端设备无法获取所使用的模型信息或者无法及时获取适用于当前场景的模型信息时,可能无法带来性能增益,甚至影响系统性能。
为了解决上述问题,本申请实施例提供了一种用于无线通信的方法,通过在终端设备无法确定所使用的模型信息或者无法及时获取适用于当前场景的模型信息时,如满足第一条件时,使用一个泛化能力较好,可以在多种部署场景工作的模型(如第一模型)进行通信,有助于改善系统性能。
图5是本申请实施例提供的用于无线通信的方法的流程示意图。图5所示的方法可以包括步骤S510。
在步骤S510中,若满足第一条件,终端设备基于第一模型与网络设备进行通信。
上述第一条件可以包括终端设备未被配置或未被指示模型信息。终端设备未被配置或未被指示模型的信息,可以指网络设备无法向终端设备配置或指示模型信息,如前文提到的初始接入过程中,终端设备与网络设备尚未建立RRC连接时,网络设备无法向网络设备配置或指示模型信息。
前文提到,在一些情况下,终端设备无法及时获取适用于当前场景的模型信息,因此,上述第一条件可以包括第一计时器到期(也可以称为超期),以避免终端设备使用不合适的模型带来的系统性能恶化。
在一些实施例中,第一计时器的计时时长可以根据系统的需求设定。例如,第一计时器的计时时长可以根据系统能够容忍的可能出现性能恶化的时间确定。如果系统能够容忍的可能出现性能恶化的时间较短,第一计时器的计时时长可以较小。如果系统能够容忍的可能出现性能恶化的时间较长,第一计时器的计时时长可以较大。又如,第一计时器的计时时长可以根据终端设备所处的场景的变化情况确定。如果终端设备所处的场景的变化较快,那么终端设备可能需要多次切换模型以适应场景的变化。如果终端设备的场景变化较慢,或者未发生变化,那么终端设备可能不需要切换模型,或者模型的切换次数较少。基于此,作为一个示例,如果终端设备所处的场景变化较快,第一计时器的计时时长可以较小,如果终端设备所处的场景变化较慢,或者未发生变化,第一计时器的计时时长可以较 大。上述终端设备所处的场景变化例如可以指终端设备的位置变化、通信质量变化等。
在本申请实施例中,第一计时器的计时时长的合理设计,可以兼顾模型的切换频次与系统性能的改善。
上述第一计时器与终端设备基于第二模型与网络设备进行通信的时间关联。例如,第一计时器的动作,如开启、暂停等与终端设备基于第二模型与网络设备进行通信的时间关联。作为一个示例,第一计时器的开启时间与终端设备开始使用第二模型的时间关联,或者第一计时器的开启时间与终端设备接收到用于指示终端设备使用第二模型的指示信息(可以称为第一指示信息)的时间关联。作为另一个示例,第一计时器的暂停时间可以与终端设备接收到第一指示信息的时间关联。
在一些实施例中,第一条件包括第一计时器到期可以包括第一计时器到期之前,终端设备未接收到网络设备发送的关于模型信息的指示信息,或者,第一条件可以包括在终端设备基于第二模型与网络设备进行通信的时间超过第一计时器的计时时长。
在一些实施例中,第一模型可以为基于AI技术的模型,如AI/ML模型。例如,第一模型为在各种部署场景均可以工作的模型,或者第一模型为在多种部署场景均可以良好工作的模型,如,第一模型为可靠性较高、泛化能力最佳、在各种部署场景均可以工作的模型。
在本申请实施例中,针对终端设备未被配置或指示模型信息的情况,终端设备可以使用第一模型与网络设备进行通信,与不使用模型相比,有助于获得更好的系统性能。另外,在第一定时器到期时,终端设备可以使用第一模型与网络设备进行通信,也就是说,在长期使用某个模型后,终端设备可以自动切换到(或者可以称为自动回落到)第一模型。由于第一模型的泛化性能较好,可以适用于多种部署场景,因此上述自动回落到第一模型有助于避免模型与部署场景失配未被及时监测到,或者模型切换指令的误检等造成的系统性能下降。
针对不同的第一条件,第一模型可以不同。或者说,第一模型可以包括多个模型,针对不同的第一条件,终端设备可以基于多个模型中不同的模型与网络设备进行通信。
例如,第一模型可以包括第一初始模型。若第一条件为终端设备未被配置或未被指示模型信息,终端设备基于第一初始模型与网络设备进行通信。如此一来,与不使用模型相比,在无法探测终端设备的部署场景、网络设备无法指示终端设备使用哪个模型的情况下,终端设备基于第一初始模型与网络设备进行通信可以获得更好的性能。
又如,第一模型可以包括第一缺省模型,若第一条件为第一计时器到期,终端设备基于第一缺省模型与网络设备进行通信。
在一些实施例中,若终端设备接收到网络设备发送的第一指示信息,则终端设备启动第一计时器,第一指示信息可以用于指示终端设备基于第二模型与网络设备进行通信。这里提到的第二模型,可以为网络设备基于部署场景选择的模型。
需要说明的是,第一指示信息可以用于指示终端设备基于第二模型与网络设备进行通信,还可以指:第一指示信息用于指示终端设备使用第二模型处理与网络设备通信过程中的数据。
其中,第一指示信息可以承载在RRC信息、DCI指示信息或者媒体接入控制单元(media access control control element,MAC CE)控制信息中。
在第二模型使用一段时间之后,部署场景可能发生了变化,或者说第二模型与部署场景可能发生失配。如果网络设备未监测到该情况,终端设备继续使用第二模型,则会造成系统性能下降。如果网络设备监测到该情况,但是在一些情况下,如信道质量较差的情况下,网络设备指示终端设备进行模型切换的信令可能被误检,那么网络侧和终端侧对选择或者激活的模型产生不同的理解,从而造成系统性能下降。因此,当第一计时器到期时,终端设备可以使用第一缺省模型与网络设备进行通信,即终端设备可以从第二模型切换到 第一缺省模型,从而有助于避免系统性能的下降。
如果在第一计时器运行期间,终端设备接收到网络设备发送的与模型关联的指示信息,那么终端设备可以基于该指示信息调整模型的使用策略,或者基于该指示信息确定第一缺省模型的使用时机。
在一些实施例中,若在第一计时器运行期间,终端设备接收到网络设备发送的第一指示信息,则终端设备可以执行第一操作,第一操作与第一缺省模型的使用关联。
例如,第一操作可以包括重启第一计时器,或者说第一操作可以包括第一计时器复位。在第一计时器运行期间,终端设备接收到第一指示信息,也就是说,第二模型与部署场景匹配,或者第二模型适用于当前终端设备与网络设备通信的场景,因此,此时重启第一计时器,可以延长非第一缺省模型,如第二模型的使用时间,从而减少非必要的模型切换,进而有助于获得更好的模型与部署场景的适配性,以及更好的模型推理性能。
在第一计时器运行期间,终端设备可能会使用第二模型处理数据(如基于网络设备的指示信息使用第二模型处理数据)。在第一计时器到期时,终端设备基于第一缺省模型与网络设备进行通信,或者说终端设备从第二模型切换到第一缺省模型,可能会中断基于第二模型的数据处理过程,从而带来性能损失。
作为解决上述问题的一种方法,第一操作可以包括终端设备暂停(也可以称为中止)第一计时器,并继续基于第二模型进行数据处理;如果终端设备完成基于第二模型的数据处理,则终端设备恢复运行第一计时器。
作为解决上述问题的另一种方法,第一操作可以包括维持第一计时器的运行,当第一计时器到期时,根据基于第二模型的数据处理是否完成,确定终端设备所使用的模型。例如,当第一计时器到期时,如果终端设备基于第二模型的数据处理已完成,则基于第一缺省模型与网络设备进行通信;如果终端设备基于第二模型的数据处理未完成,则在基于第二模型的数据处理完成之后,基于第一缺省模型与网络设备进行通信。
作为解决上述问题的又一种方法,第一操作可以包括将第一计时器的计时时长增加第一时长。也就是说,通过延长第一计时器的计时时长来避免第一计时器到期和基于模型的数据处理未完成的冲突。在一些实施例中,第一时长可以为预设时长,或者第一时长可以是网络设备指示的(或者说,第一时长可以基于网络设备发送的指示信息确定)。例如,第一时长可以基于网络设备发送的RRC配置信息,DCI指示信息或者MAC CE控制信息确定。
需要说明的是,上述三种解决方法可以单独使用,也可以结合使用。例如,第一操作可以包括将第一计时器的计时时长增加第一时长,以及维持第一计时器的运行,当第一计时器到期时,根据基于第二模型的数据处理是否完成,确定终端设备所使用的模型,以进一步提高系统性能。也就是说,将第一计时器的计时时长增加第一时长之后,当第一计时器到期时,如果终端设备基于第二模型的数据处理已完成,则基于第一缺省模型与网络设备进行通信;如果终端设备基于第二模型的数据处理未完成,则在基于第二模型的数据处理完成之后,基于第一缺省模型与网络设备进行通信。
在一些实施例中,第一模型可以基于第一规则确定,第一规则可以与以下信息中的一种或多种关联:终端设备接入的频段;为终端设备提供服务的运营商的运营商标识;终端设备所处的地理位置的信息;以及第一模型的历史使用信息。
基于终端设备接入的频段,为终端设备提供服务的运营商的运营商标识以及终端设备所处的地理位置的信息中的一种或多种,可以对模型的部署场景进行预测,从而确定合适的第一初始模型和/或第一缺省模型。
在一些实施例中,第一模型可以基于第一模型的历史使用信息确定。其中,第一模型的历史使用信息可以包括当前时刻之前满足第一条件时,终端设备所使用的模型,如上一次满足第一条件时,终端设备所使用的模型。例如,如果第一模型的历史使用信息中包括 多个模型的使用信息,那么第一模型可以为多个模型中使用次数最多的模型。又如,第一模型可以为上一次满足第一条件时,终端设备所使用的模型。由于上一次满足第一条件时的部署场景可能与当前的模型部署场景最为接近,因此,将上一次满足第一条件时,终端设备所使用的模型作为第一模型,有助于在提高模型的性能的同时简化模型的选择过程。
前文提到,第一模型可以包括第一初始模型和第一缺省模型。在一些实施例中,第一初始模型可以基于第一初始模型的历史使用信息确定,第一缺省模型可以基于第一初始模型的历史使用信息和/或第一缺省模型的使用信息确定。例如,第一缺省模型可以为上一次第一计时器到期时,终端设备所使用的模型,或者,第一缺省模型可以为上一次终端设备未被配置或指示模型信息的情况下,终端设备所使用的模型。
在一些实施例中,第一初始模型的信息可以承载在系统信息中。例如,系统信息可以为物理广播信道(physical broadcast channel,PBCH)中的系统信息(如主信息块(master information block,MIB))或系统信息块(system information block,SIB)中的系统信息。
在一些实施例中,第一缺省模型的信息可以承载在RRC配置信息中。
在一些实施例中,第一初始模型和/或第一缺省模型的信息还可以承载在其他终端设备发送的信息中。也就是说,终端设备还可以从其他终端设备,如通过终端设备之间的直接通信获取第一初始模型和/或第一缺省模型的信息。例如,在终端设备从与其他终端设备的通信切换为,或者准备切换为与网络设备的通信时,终端设备可以从其他终端设备获取第一初始模型的信息。
第一模型可以基于上述一种或多种方法确定。在一些实施例中,上述方法中基于第一规则确定第一模型的方法优先级最低。也就是说,如果可以通过上述方法中,除基于第一规则确定第一模型的方法之外的其他方法确定第一模型(可以简称为上述其他方法),则不基于第一规则确定第一模型,或者说,如果通过上述其他方法均无法确定第一模型,则基于第一规则确定第一模型。
如前文所述,不同终端设备的能力可能不同,因此,不同的终端设备可以支持的模型可能也不相同。同理,不同的终端设备可以支持的候选第一模型可能也不相同。因此,在一些实施例中,终端设备可以将其可以支持的候选第一模型上报给网络设备,以便确定与终端设备能力相匹配的第一模型,或者说从终端设备支持的候选第一模型中确定终端设备所使用的第一模型。这里提到的候选第一模型可以为满足第一条件时,终端设备可以使用的模型,或者说终端设备的能力支持的模型。
例如,候选第一模型的信息可以包含在终端设备上报的能力信息中。作为一个示例,终端设备可以向网络设备发送第一信息,或者,网络设备接收终端设备发送的第一信息。第一信息包括终端设备支持的模型的信息,其中,终端设备支持的模型包括第一模型。在一些实现方式中,第一信息可以为终端设备支持的模型列表。
在一些实施例中,第一模型可以为终端设备支持的模型中具有特定标识的模型。或者说,第一模型可以为第一信息中具有特定标识的模型。例如,标识为A的模型为终端设备支持的候选第一模型。又如,标识为X的模型为可以作为终端设备的第一初始模型;标识为Y的模型为可以作为终端设备的第一缺省始模型。另如,标识为M的模型既可以作为终端设备的第一初始模型,也可以作为终端设备的第一缺省模型。
需要说明的是,上述标识可以单独使用,也可以结合使用。例如,标识X、标识Y、标识M可以结合使用。又如,标识A可以单独使用。
在一些实施例中,终端设备可以向网络设备发送第二信息,或者,网络设备接收终端设备发送的第二信息。其中,第二信息可以用于指示第一模型。
在一些情况下,同一时刻,网络设备所使用的模型需要与终端设备所使用的模型关联(即使用两侧模型(two-sided model)的情况)。例如,当终端设备使用信道状态信息压缩模型时,网络设备可以使用与该信道状态信息压缩模型关联的信道状态信息解压缩模型才 能实现正确通信。
在一些实施例中,第一模型与第三模型关联,其中,第三模型为第一条件下网络设备所使用的模型。也就是说,当终端设备使用第一模型时,网络设备可以使用第三模型。
例如,第一模型包括第一初始模型和/或第一缺省模型,第三模型可以包括第二初始模型和/或第二缺省模型,其中第一初始模型与第二初始模型关联,第一缺省模型与第二缺省模型关联。也就是说,当终端设备使用第一初始模型时,网络设备可以使用与第一初始模型关联的第二初始模型进行通信;当终端设备使用第一缺省模型时,网络设备可以使用与第一缺省模型关联的第二缺省模型进行通信。
上述第一模型与第三模型关联可以包括第一模型与第三模型之间具有关联关系,该关联关系可以由网络设备进行配置。例如,第一初始模型与第二初始模型的关联关系,和/或第一缺省模型与第二缺省模型的关联关系可以由网络设备进行配置。
在一些实现方式中,上述关联关系可以通过模型标识,如模型ID的关联关系或模型标识组进行配置。例如,网络设备可以配置与第一初始模型的标识关联的第二初始模型的标识,或者网络设备可以配置与第一缺省模型的标识关联的第二缺省模型的标识。作为一个示例,第一初始模型可以和第二初始模型使用相同的模型标识,如第一初始模型为终端设备侧model 1,第二初始模型为网络设备侧model 1,以指示其关联关系。又如,网络设备可以为第一初始模型和第二初始模型配置组合标识,或者网络设备可以为第一缺省模型和第二缺省模型配置组合标识。
前文提到,在使用两侧模型的情况下,若满足第一条件,终端设备基于第一模型与网络设备进行通信,相应的,网络设备基于第三模型与终端设备进行通信。下文将站在网络设备的角度对上述方法进行介绍,应理解,网络设备侧的描述可以与终端设备侧的描述相互对应,因此,未详细描述的部分可以参见前面终端设备侧描述的方法。
需要说明的是,终端设备与网络设备可以均设有第一计时器,本实施例中提到的第一计时器可以指网络设备侧设置的第一计时器。
在一些实施例中,第三模型可以包括第二初始模型,若第一条件为终端设备未被配置或未被指示模型信息,网络设备基于第二初始模型与终端设备进行通信。也就是说,第一初始模型与第二初始模型关联,当终端设备使用第一初始模型时,网络设备可以使用第二初始模型。
在一些实施例中,第三模型可以包括第二缺省模型,若第一条件为第一计时器到期,网络设备基于第二缺省模型与终端设备进行通信。也就是说,第二缺省模型与第一缺省模型关联,当终端设备使用第一缺省模型时,网络设备可以使用第二缺省模型。
在一些实施例中,若网络设备向终端设备发送第一指示信息,则网络设备启动第一计时器,第一指示信息用于指示终端设备基于第二模型与网络设备进行通信。
在一些实施例中,若在第一计时器运行期间,网络设备向终端设备发送第一指示信息,则网络设备执行第一操作,第一操作与第二缺省模型的使用关联。
在一些实施例中,第一操作包括重启第一计时器。
在一些实施例中,第一操作,包括:网络设备暂停第一计时器;如果终端设备完成基于第二模型的数据处理,则网络设备恢复运行第一计时器。
在一些实施例中,第一操作,包括:维持第一计时器的运行,当第一计时器到期时,如果终端设备基于第二模型的数据处理已完成,则基于第二缺省模型与终端设备进行通信;如果终端设备基于第二模型的数据处理未完成,则在基于第二模型的数据处理完成之后,基于第二缺省模型与终端设备进行通信。
由于网络设备无法获取,或者无法准确获取基于第二模型的数据处理的完成时间,因此,对于使用两侧模型的情况,网络设备例如可以基于第四模型的数据处理的完成时间,确定第二缺省模型的激活时机。其中,第四模型与第二模型关联。又如,网络设备可以基 于网络设备侧的数据调度情况确定基于第二模型的数据处理的完成时间。
在一些实施例中,第一操作包括将第一计时器的计时时长增加第一时长。例如,第一时长为预设时长,或者第一时长是网络设备指示的。
可以看出,网络设备侧的第一计时器的启动、暂停与第一指示信息的发送时间关联,而终端设备侧的第一计时器的启动、暂停与第一指示信息的接收时间关联。但是,终端设备接收到第一指示信息的时间与网络设备发送第一指示信息存在第一时间延迟。
上述第一时间延迟可能会导致终端设备侧的第一计时器的与网络设备侧的第一计时器的动作不同步,进而导致终端设备与网络设备切换模型的时间不一致。也就是说在上述第一时间延迟范围内,终端设备使用的模型可能与网络设备所使用的模型不关联,从而可能会导致通信质量下降。
为了解决上述问题,可以基于第一计时器到期以及时间修正量(或者称为时间间隔)确定终端设备和/或网络设备的模型切换时机。
在一些实施例中,终端设备可以在第一计时器到期时,切换为第一缺省模型,网络设备可以在第一计时器到期之后间隔第一时间修正量的时刻,切换为第二缺省模型。由于网络设备发送第一指示信息的时间比终端设备接收到第一指示信息的时间早,那么网络设备侧的第一计时器的启动时刻,到期时刻分别早于终端设备侧的第一计时器的启动时刻,到期时刻。因此,网络设备通过在第一计时器到期之后间隔第一时间修正量的时刻切换为第二缺省模型,有助于消除上述第一时间延迟对模型切换时机的影响。
在一些实施例中,终端设备可以在第一计时器到期之后间隔第二时间修正量的时刻,切换为第一缺省模型,网络设备可以在第一计时器到期之后间隔第三时间修正量的时刻,切换为第二缺省模型,其中,第三时间修正量大于第二间修正量。为了进一步消除上述第一时间延迟对模型切换时机的影响,在第一计时器到期之后,终端设备和网络设备均不使用模型。
前文提到,在第一计时器运行期间,如果终端设备使用第二模型处理数据,那么终端设备可以暂停第一计时器,待终端设备完成基于第二模型的数据处理之后,继续运行第一计时器。由于终端设备可以获取其完成基于第二模型的数据处理的时间,但网络设备无法获取或者说无法准确获取终端设备完成基于第二模型的数据处理的时间,那么网络设备无法确定恢复运行第一计时器的时间,或者说,网络设备恢复运行第一计时器的时间与终端设备恢复运行第一计时器的时间存在误差。因此,网络设备侧的第一计时器到期的时刻与终端设备侧的第一计时器到期的时刻可能存在误差。
为了解决上述问题,终端设备可以在完成基于第二模型的数据处理时,向网络设备发送第二指示信息。网络设备可以在接收到终端设备发送的第二指示信息时,恢复运行第一计时器。如此一来,网络设备侧的第一计时器到期的时刻与终端设备侧的第一计时器到期的时刻可能存在第二时间延迟。其中,第二时间延迟大于第一时间延迟,如第二时间延迟为第一时间延迟的两倍。因此,终端设备可以在第一计时器到期时,切换为第一缺省模型,网络设备可以在第一计时器到期之后间隔第四时间修正量的时刻,切换为第二缺省模型,有助于提高模型切换时机的准确度。其中第四时间修正量可以大于第一时间修正量,如第四时间修正量为第一时间修正量的两倍。
在一些实施例中,网络设备可以基于网络设备的数据调度情况确定恢复运行第一计时器的时间,便于实现。
在一些实施例中,网络设备可以在第一计时器到期时,切换为第二缺省模型,终端设备可以在第一计时器到期之前间隔第一时间修正量的时刻,切换为第一缺省模型。具体实现方式与前文提到的对网路设备切换模型的时机进行修正的方法相似,为了简洁,此处不再赘述。
上述第一时间修正量、第二时间修正量、第三时间修正量以及第四时间修正量中的一 种或多种可以基于当前部署场景下,终端设备与网络设备之间的传输时延确定。例如,终端设备可以基于第一指示信息中携带的发送时间以及终端设备接收到第一指示信息的时间确定上述传输时延。
上述第一时间修正量、第二时间修正量、第三时间修正量以及第四时间修正量中的一种或多种可以预配置,也可以在模型切换之前进行动态指示。例如,第四时间修正量可以承载在第二指示信息中。
需要说明的是,上述网络设备基于第一计时器执行某些动作,是指网络设备基于网络设备侧的第一计时器执行的动作;上述终端设备基于第一计时器执行某些动作,是指终端设备基于终端设备侧的第一计时器执行的动作。
需要说明的是,上述第一计时器可以为递增计时器,也可以为递减计时器,本申请对此不作限定。
需要说明的是,本申请实施例中提到的使用某个模型或者切换为某个模型,均可以指第一设备基于该模型与第二设备进行通信,其中第一设备可以为终端设备,也可以为网络设备,第二设备可以为终端设备,也可以为网络设备。
下文将以第一模型为AI/ML模型为例,结合实施例一至实施例四对本申请实施例提供的方法进行详细的介绍。
实施例一
实施例一以终端设备与网络设备未建立RRC连接(即第一条件为终端设备未被配置或未被指示模型信息的一种情况)为例,结合图6和图7对本申请实施例提供的方法进行介绍。
参见图6,在初始接入过程中,终端设备尚未建立RRC连接,无法基于网络设备侧的RRC配置、DCI、MAC CE等确定该使用哪个AI/ML模型,且也无法对当前的部署场景进行探测,从而无法判断哪个模型对当前部署场景适配的最好。因此可以采用一个能够提供可靠性能、泛化能力最佳、在任何部署场景都肯定能工作的AI/ML模型(即第一初始模型)。其中,第一初始模型的ID可以根据第一规则或网络设备侧的系统信息确定。
终端设备在RRC连接建立之后,系统可以对终端设备所处的部署环境进行探测,从而选择和当前部署环境适配的最好的AI/ML模型(如第二模型),也可以根据网络设备的指示信息(RRC配置信息、DCI指示信息、MAC CE控制信息等)切换到另一个AI/ML模型(如第二模型)。后续随着部署环境的变化,还可以根据网络设备的指示信息切换到其他AI/ML模型。
当网络设备侧和终端设备侧需要使用对应的AI/ML模型时(如使用两侧模型),网络设备侧的初始模型和终端设备侧的初始模型也要对应使用。如图7所示,在初始接入过程中,终端设备尚未建立RRC连接,终端设备侧使用第一初始模型,而网络设备侧也使用相对应的第二初始模型。终端设备在RRC连接建立之后,网络设备侧指示终端设备切换到第二模型,网络设备也同时切换到与之对应的第四模型。
本申请实施例中,在系统在无法探测终端设备部署场景、网络设备也无法指示终端设备使用哪个模型的初始接入阶段,也可以采用AI/ML模型,获得比非AI/ML模型更好的性能。
实施例二
实施例二以第一条件为第一计时器到期为例,结合图8和图9对本申请实施例提供的方法进行介绍。
参见图8,在RRC连接状态中,终端设备可以根据网络设备侧的指示,选择与部署场景适配的最好的AI/ML模型(如第二模型)。但为防止对部署场景变化的探测及对AI/ML模型运行性能的监测的不准确性和不及时性,应在终端设备开始使用第二模型时启动一个计时器(timer)(即前文提到的第一计时器)。例如,计时器为时长为T的timer,计时器 从t=T开始时间递减。当计时器到期(expire,即t=0)后,自动切换到所述第一缺省模型。
当网络设备侧和终端设备侧需要使用对应的AI/ML模型时(如使用两侧模型),网络设备侧的缺省模型和终端设备侧的AI/ML模型也要对应使用。参见图9,终端设备侧使用第二模型时,网络设备侧也使用相对应的第四模型。当计时器到期后,终端设备自动切换到第一缺省模型,同时网络设备也自动切换到第二缺省模型。
在本申请实施例中,系统在长期使用某个与特定部署场景相适配的AI/ML模型后,可以自动回落到一个适配各种部署场景的泛化性能较好的缺省AI/ML模型,可以避免AI/ML模型与部署场景失配没有被系统监测到,或者可以避免网络设备指示模型切换的信令误检造成的网络设备侧和终端设备侧对激活AI/ML产生不同理解,从而造成系统性能下降。即使AI/ML模型与部署场景失配没有被系统监测到,或网络设备指示模型切换的信令被误检,在计时器到期后,网络设备侧和终端设备侧都会认为应回落到缺省AI/ML模型,从而可以保证了系统性能的可靠性。
实施例三
实施例三以第一条件为第一计时器到期为例,结合图10和图11对本申请实施例提供的方法进行介绍。
参见图10,与部署场景适配得更好的AI/ML模型(如第二模型)被激活以后,计时器应在时间T以后到期,并将模型回退到缺省模型。但如果在计时器运行期间,网络设备通过指示信息再次指示激活第二模型,说明网络设备根据对部署场景的探测和第二模型与部署场景适配性的监测,确认仍适合继续使用第二模型。因此可在网络设备通过指示信息再次指示激活第二模型时将计时器复位到t=T,使计时器重新开始计时,等到计时器到期后,终端设备自动切换到第一缺省模型。
当网络设备侧和终端设备侧需要使用对应的AI/ML模型时,网络设备侧的缺省模型和终端设备侧的缺省模型也要对应使用。参见图11,终端设备侧与部署场景适配得更好的第二模型被激活以后,网络设备侧也使用相对应的第四模型。计时器应在时间T以后到期,且在计时器到期时,将终端设备侧和网络设备侧的模型均回退到缺省模型。但如果在计时器运行期间,网络设备通过指示信息再次指示激活第二模型和第四模型,说明网络设备根据对部署场景的探测以及第二模型、第四模型与部署场景适配性的监测,确认仍适合继续使用第二模型和/或第四模型。因此可在网络设备通过指示信息再次指示激活第二模型和/或第四模型时,将计时器复位(或者也可以称为重启)到t=T,使计时器重新开始计时,等到计时器到期后,终端设备和网络设备侧自动切换到第一缺省模型和第二缺省模型。
在本申请实施例中,系统可以在确认某个非缺省模型与部署场景适配的很好的情况下,通过复位AI/ML模型回落计时器,延长非缺省AI/ML模型的使用时间,从而减少不必要的频繁回落,获得更好的与部署场景的适配性和更好的AI/ML模型推理性能。
实施例四
实施例四以第一条件为第一计时器到期为例,结合图12至图15对本申请实施例提供的方法进行介绍。
选择、激活一个AI/ML模型并不意味着终端设备、网络设备一直使用此模型对数据进行处理。当计时器到期,如果终端设备和网络设备都未使用非缺省AI/ML模型对数据进行处理,则可以马上回落到缺省AI/ML模型。但是如果终端设备和/或网络设备正在使用非缺省AI/ML模型对数据进行处理,马上回落到缺省AI/ML模型将造成数据处理过程的中断,带来性能损失。因此可在终端设备和/或网络设备正在使用非缺省AI/ML模型对数据进行处理时,中止AI/ML模型回落计时器,延后计时器的回落时间。
参见图12,在与部署场景适配得更好的第二模型被激活以后,计时器应在时间T以后到期,并将模型回退到缺省模型。但是当计时器时间减少到t=T1时,终端设备开始使用第二模型处理数据,为避免在数据处理过程中计时器时间减少到t=0出现模型回落,可以在 终端设备使用第二模型处理数据的过程中,中止计时器的运行,等数据处理完之后再从t=T1恢复计时器运行,等到计时器到期后,终端设备自动切换到第一缺省模型。
当网络设备侧和终端设备侧需要使用对应的AI/ML模型时,网络设备侧的缺省模型和终端设备侧的缺省模型也要对应使用。参见图13,终端设备侧与部署场景适配得更好的第二模型被激活以后,网络设备侧也使用相对应的第四模型,计时器应在时间T以后到期,并将终端设备侧和网络设备侧模型均回退到缺省模型。但是当计时器时间减少到t=T1时,网络设备和/或终端设备开始使用第四模型和/或第二模型处理数据,为避免在数据处理过程中计时器时间减少到t=0出现模型回落,可以在网络设备和终端设备使用第四模型和/或第二模型处理数据的过程中,中止计时器的运行,等数据处理完之后再从t=T1恢复计时器运行。计时器到期后,终端设备和网络设备自动切换到第一缺省模型和第二缺省模型。
另一种避免在基于模型的数据处理过程中发生模型回落的方法,是在计时器到期后延迟回落。参见图14,当计时器时间减少到t=T1时,终端设备开始使用第二模型处理数据,但计时器并不中止,正常运行直至计时器到期。如果此时终端设备尚未完成采用第二模型的数据处理,则暂不回落到第一缺省模型,而是等采用第二模型处理完数据之后,终端设备再自动切换到第一缺省模型。如果计时器到期时,终端设备已经完成采用第二模型的数据处理,则回落到第一缺省模型。
当网络设备侧和终端设备侧需要使用对应的AI/ML模型时,网络设备侧的缺省模型和终端设备侧的AI/ML模型也要对应使用。参见图15,当计时器时间减少到t=T1时,终端设备和/或网络设备侧开始使用第二模型和/或第四模型处理数据,但计时器并不中止,正常运行直至到期。如果此时终端设备和/或网络设备尚未完成采用第二模型、第四模型的数据处理,则暂不回落到缺省AI/ML模型,而是等采用第二模型、第四模型处理完数据之后,终端设备和网络设备侧自动切换到第一缺省模型和第二缺省模型。
在本申请实施例中,通过计时器的中止或延迟回落,可以避免模型回落中断采用AI/ML模型处理数据的过程,从而避免数据处理过程中断带来的性能损失。
上文结合图1至图15,详细描述了本申请的方法实施例,下面结合图16至图18,详细描述本申请的装置实施例。应理解,方法实施例的描述与装置实施例的描述相互对应,因此,未详细描述的部分可以参见前面方法实施例。
图16是本申请实施例的一种终端设备的结构示意图。如图16所示,终端设备1600可以包括通信单元1610。
通信单元1610,用于若满足第一条件,基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
在一些实施例中,所述第一模型包括第一初始模型,若所述第一条件为所述终端设备未被配置或未被指示模型信息,所述终端设备基于所述第一初始模型与所述网络设备进行通信。
在一些实施例中,所述第一模型包括第一缺省模型,若所述第一条件为所述第一计时器到期,所述终端设备基于所述第一缺省模型与所述网络设备进行通信。
在一些实施例中,所述设备还包括:启动单元,用于若所述终端设备接收到所述网络设备发送的第一指示信息,则启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
在一些实施例中,所述设备还包括:执行单元,用于若在所述第一计时器运行期间,所述终端设备接收到所述网络设备发送的所述第一指示信息,则所述终端设备执行第一操作,所述第一操作与所述第一缺省模型的使用关联。
在一些实施例中,所述第一操作包括重启所述第一计时器。
在一些实施例中,所述第一操作,包括:所述终端设备暂停所述第一计时器,并继续 基于所述第二模型进行数据处理;如果所述终端设备完成基于所述第二模型的数据处理,则所述终端设备恢复运行所述第一计时器。
在一些实施例中,所述第一操作,包括:维持所述第一计时器的运行,当所述第一计时器到期时,如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第一缺省模型与所述网络设备进行通信;如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第一缺省模型与所述网络设备进行通信。
在一些实施例中,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
在一些实施例中,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
在一些实施例中,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:所述终端设备接入的频段;为所述终端设备提供服务的运营商的运营商标识;所述终端设备所处的地理位置的信息;以及所述第一模型的历史使用信息。
在一些实施例中,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
在一些实施例中,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
在一些实施例中,所述设备还包括:第一发送单元,用于向所述网络设备发送第一信息,所述第一信息包括所述终端设备支持的模型的信息,其中,所述终端设备支持的模型包括所述第一模型。
在一些实施例中,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
在一些实施例中,所述设备还包括:第二发送单元,用于向所述网络设备发送第二信息,所述第二信息用于指示所述第一模型。
在一些实施例中,所述第一模型与第三模型关联,其中,所述第三模型为所述第一条件下所述网络设备所使用的模型。
在一些实施例中,所述第一模型与第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
图17是本申请实施例的一种网络设备的结构示意图。如图17所示,网络设备1700可以包括通信单元1710。
通信单元1710,用于若满足第一条件,基于第三模型与终端设备进行通信,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期;其中,第一模型与所述第三模型关联,所述第一模型为满足所述第一条件时所述终端设备所使用的模型,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
在一些实施例中,所述第三模型包括第二初始模型,若所述第一条件为所述终端设备未被配置或未被指示模型信息,所述网络设备基于所述第二初始模型与所述终端设备进行通信。
在一些实施例中,所述第三模型包括第二缺省模型,若所述第一条件为所述第一计时器到期,所述网络设备基于所述第二缺省模型与所述终端设备进行通信。
在一些实施例中,所述设备还包括:启动单元,用于若所述网络设备向所述终端设备发送第一指示信息,则启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
在一些实施例中,所述设备还包括:执行单元,用于若在所述第一计时器运行期间,所述网络设备向所述终端设备发送所述第一指示信息,则执行第一操作,所述第一操作与所述第二缺省模型的使用关联。
在一些实施例中,所述第一操作包括重启所述第一计时器。
在一些实施例中,所述第一操作,包括:所述网络设备暂停所述第一计时器;如果所述终端设备完成基于所述第二模型的数据处理,则所述网络设备恢复运行所述第一计时器。
在一些实施例中,所述第一操作,包括:维持所述第一计时器的运行,当所述第一计时器到期时,如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第二缺省模型与所述终端设备进行通信;如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第二缺省模型与所述终端设备进行通信。
在一些实施例中,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
在一些实施例中,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
在一些实施例中,所述第一模型包括第一初始模型,所述第一初始模型用于所述终端设备在所述第一条件为所述终端设备未被配置或未被指示模型信息的情况下使用。
在一些实施例中,所述第一模型包括第一缺省模型,所述第一缺省模型用于所述终端设备在所述第一条件为第一计时器到期的情况下使用。
在一些实施例中,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:所述终端设备接入的频段;为所述终端设备提供服务的运营商的运营商标识;所述终端设备所处的地理位置的信息;以及所述第一模型的历史使用信息。
在一些实施例中,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
在一些实施例中,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
在一些实施例中,所述设备还包括:第一接收单元,用于接收所述终端设备发送的第一信息,所述第一信息包括所述终端设备支持的模型的信息,其中,所述终端设备支持的模型包括第一模型,所述第一模型为所述第一条件下,所述终端设备所使用的模型。
在一些实施例中,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
在一些实施例中,所述设备还包括:第二接收单元,用于接收所述终端设备发送的第二信息,所述第二信息用于指示所述第一模型。
在一些实施例中,所述第一模型与所述第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
在可选地实施例中,上文所述的发送单元、接收单元、通信单元可以为收发器1830。终端设备1600、网络设备1700还可以包括处理器1810和存储器1820,具体如图18所示。
图18是本申请实施例的通信装置的示意性结构图。图18中的虚线表示该单元或模块为可选的。该装置1800可用于实现上述方法实施例中描述的方法。装置1800可以是芯片或终端设备或网络设备。
装置1800可以包括一个或多个处理器1810。该处理器1810可支持装置1800实现前文方法实施例所描述的方法。该处理器1810可以是通用处理器或者专用处理器。例如,该处理器可以为中央处理单元(central processing unit,CPU)。或者,该处理器还可以是其他通用处理器、数字信号处理器(digital signal processor,DSP)、专用集成电路(application specific integrated circuit,ASIC)、现场可编程门阵列(field programmable gate array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
装置1800还可以包括一个或多个存储器1820。存储器1820上存储有程序,该程序可以被处理器1810执行,使得处理器1810执行前文方法实施例所描述的方法。存储器1820可以独立于处理器1810也可以集成在处理器1810中。
装置1800还可以包括收发器1830。处理器1810可以通过收发器1830与其他设备或 芯片进行通信。例如,处理器1810可以通过收发器1830与其他设备或芯片进行数据收发。
本申请实施例还提供一种计算机可读存储介质,用于存储程序。该计算机可读存储介质可应用于本申请实施例提供的终端或网络设备中,并且该程序使得计算机执行本申请各个实施例中的由终端设备或网络设备执行的方法。
本申请实施例还提供一种计算机程序产品。该计算机程序产品包括程序。该计算机程序产品可应用于本申请实施例提供的终端或网络设备中,并且该程序使得计算机执行本申请各个实施例中的由终端或网络设备执行的方法。
本申请实施例还提供一种计算机程序。该计算机程序可应用于本申请实施例提供的终端设备或网络设备中,并且该计算机程序使得计算机执行本申请各个实施例中的由终端设备或网络设备执行的方法。
应理解,本申请中术语“系统”和“网络”可以被可互换使用。另外,本申请使用的术语仅用于对本申请的具体实施例进行解释,而非旨在限定本申请。本申请的说明书和权利要求书及所述附图中的术语“第一”、“第二”、“第三”和“第四”等是用于区别不同对象,而不是用于描述特定顺序。此外,术语“包括”和“具有”以及它们任何变形,意图在于覆盖不排他的包含。
在本申请的实施例中,提到的“指示”可以是直接指示,也可以是间接指示,还可以是表示具有关联关系。举例说明,A指示B,可以表示A直接指示B,例如B可以通过A获取;也可以表示A间接指示B,例如A指示C,B可以通过C获取;还可以表示A和B之间具有关联关系。
本申请的实施例,提到的“包括”可以指直接包括,也可以指间接包括。可选地,可以将本申请实施例中提到的“包括”替换为“指示”或“用于确定”。例如,A包括B,可以替换为A指示B,或A用于确定B。
在本申请实施例中,“与A相应的B”表示B与A相关联,根据A可以确定B。但还应理解,根据A确定B并不意味着仅仅根据A确定B,还可以根据A和/或其它信息确定B。
在本申请实施例中,术语“对应”可表示两者之间具有直接对应或间接对应的关系,也可以表示两者之间具有关联关系,也可以是指示与被指示、配置与被配置等关系。
本申请实施例中,“预定义”或“预配置”可以通过在设备(例如,包括终端设备和网络设备)中预先保存相应的代码、表格或其他可用于指示相关信息的方式来实现,本申请对于其具体的实现方式不做限定。比如预定义可以是指协议中定义的。
本申请实施例中,所述“协议”可以指通信领域的标准协议,例如可以包括LTE协议、NR协议以及应用于未来的通信系统中的相关协议,本申请对此不做限定。
本申请实施例中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表示前后关联对象是一种“或”的关系。
在本申请的各种实施例中,上述各过程的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的 部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
在上述实施例中,可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时,可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时,全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中,或者从一个计算机可读存储介质向另一个计算机可读存储介质传输,例如,所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line,DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够读取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质,(例如,软盘、硬盘、磁带)、光介质(例如,数字通用光盘(digital video disc,DVD))或者半导体介质(例如,固态硬盘(solid state disk,SSD))等。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (86)

  1. 一种用于无线通信的方法,其特征在于,包括:
    若满足第一条件,终端设备基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
  2. 根据权利要求1所述的方法,其特征在于,所述第一模型包括第一初始模型,若所述第一条件为所述终端设备未被配置或未被指示模型信息,所述终端设备基于所述第一初始模型与所述网络设备进行通信。
  3. 根据权利要求1或2所述的方法,其特征在于,所述第一模型包括第一缺省模型,若所述第一条件为所述第一计时器到期,所述终端设备基于所述第一缺省模型与所述网络设备进行通信。
  4. 根据权利要求3所述的方法,其特征在于,所述方法还包括:
    若所述终端设备接收到所述网络设备发送的第一指示信息,则所述终端设备启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
  5. 根据权利要求4所述的方法,其特征在于,所述方法还包括:
    若在所述第一计时器运行期间,所述终端设备接收到所述网络设备发送的所述第一指示信息,则所述终端设备执行第一操作,所述第一操作与所述第一缺省模型的使用关联。
  6. 根据权利要求5所述的方法,其特征在于,所述第一操作包括重启所述第一计时器。
  7. 根据权利要求5所述的方法,其特征在于,所述第一操作,包括:
    所述终端设备暂停所述第一计时器,并继续基于所述第二模型进行数据处理;
    如果所述终端设备完成基于所述第二模型的数据处理,则所述终端设备恢复运行所述第一计时器。
  8. 根据权利要求5所述的方法,其特征在于,所述第一操作,包括:
    维持所述第一计时器的运行,当所述第一计时器到期时,
    如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第一缺省模型与所述网络设备进行通信;
    如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第一缺省模型与所述网络设备进行通信。
  9. 根据权利要求5所述的方法,其特征在于,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
  10. 根据权利要求9所述的方法,其特征在于,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
  11. 根据权利要求1-10中任一项所述的方法,其特征在于,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:
    所述终端设备接入的频段;
    为所述终端设备提供服务的运营商的运营商标识;
    所述终端设备所处的地理位置的信息;以及
    所述第一模型的历史使用信息。
  12. 根据权利要求2所述的方法,其特征在于,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
  13. 根据权利要求3-10中任一项所述的方法,其特征在于,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
  14. 根据权利要求1-13中任一项所述的方法,其特征在于,所述方法还包括:
    所述终端设备向所述网络设备发送第一信息,所述第一信息包括所述终端设备支持的模型的信息,其中,所述终端设备支持的模型包括所述第一模型。
  15. 根据权利要求14所述的方法,其特征在于,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
  16. 根据权利要求14所述的方法,其特征在于,所述方法还包括:
    所述终端设备向所述网络设备发送第二信息,所述第二信息用于指示所述第一模型。
  17. 根据权利要求1-16中任一项所述的方法,其特征在于,所述第一模型与第三模型关联,其中,所述第三模型为所述第一条件下所述网络设备所使用的模型。
  18. 根据权利要求17所述的方法,其特征在于,所述第一模型与第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
  19. 一种用于无线通信的方法,其特征在于,包括:
    若满足第一条件,网络设备基于第三模型与终端设备进行通信,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期;
    其中,第一模型与所述第三模型关联,所述第一模型为满足所述第一条件时所述终端设备所使用的模型,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
  20. 根据权利要求19所述的方法,其特征在于,所述第三模型包括第二初始模型,若所述第一条件为所述终端设备未被配置或未被指示模型信息,所述网络设备基于所述第二初始模型与所述终端设备进行通信。
  21. 根据权利要求19或20所述的方法,其特征在于,所述第三模型包括第二缺省模型,若所述第一条件为所述第一计时器到期,所述网络设备基于所述第二缺省模型与所述终端设备进行通信。
  22. 根据权利要求21所述的方法,其特征在于,所述方法还包括:
    若所述网络设备向所述终端设备发送第一指示信息,则所述网络设备启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
  23. 根据权利要求22所述的方法,其特征在于,所述方法还包括:
    若在所述第一计时器运行期间,所述网络设备向所述终端设备发送所述第一指示信息,则所述网络设备执行第一操作,所述第一操作与所述第二缺省模型的使用关联。
  24. 根据权利要求23所述的方法,其特征在于,所述第一操作包括重启所述第一计时器。
  25. 根据权利要求23所述的方法,其特征在于,所述第一操作,包括:
    所述网络设备暂停所述第一计时器;
    如果所述终端设备完成基于所述第二模型的数据处理,则所述网络设备恢复运行所述第一计时器。
  26. 根据权利要求23所述的方法,其特征在于,所述第一操作,包括:
    维持所述第一计时器的运行,当所述第一计时器到期时,
    如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第二缺省模型与所述终端设备进行通信;
    如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第二缺省模型与所述终端设备进行通信。
  27. 根据权利要求23所述的方法,其特征在于,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
  28. 根据权利要求27所述的方法,其特征在于,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
  29. 根据权利要求19所述的方法,其特征在于,所述第一模型包括第一初始模型,所述第一初始模型用于所述终端设备在所述第一条件为所述终端设备未被配置或未被指示模型信息的情况下使用。
  30. 根据权利要求19所述的方法,其特征在于,所述第一模型包括第一缺省模型,所述第一缺省模型用于所述终端设备在所述第一条件为第一计时器到期的情况下使用。
  31. 根据权利要求29或30所述的方法,其特征在于,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:
    所述终端设备接入的频段;
    为所述终端设备提供服务的运营商的运营商标识;
    所述终端设备所处的地理位置的信息;以及
    所述第一模型的历史使用信息。
  32. 根据权利要求29所述的方法,其特征在于,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
  33. 根据权利要求30所述的方法,其特征在于,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
  34. 根据权利要求19-33中任一项所述的方法,其特征在于,所述方法还包括:
    所述网络设备接收所述终端设备发送的第一信息,所述第一信息包括所述终端设备支持的模型的信息,其中,所述终端设备支持的模型包括第一模型,所述第一模型为所述第一条件下,所述终端设备所使用的模型。
  35. 根据权利要求34所述的方法,其特征在于,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
  36. 根据权利要求34所述的方法,其特征在于,所述方法还包括:
    所述网络设备接收所述终端设备发送的第二信息,所述第二信息用于指示所述第一模型。
  37. 根据权利要求34-36中任一项所述的方法,其特征在于,所述第一模型与所述第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
  38. 一种终端设备,其特征在于,包括:
    通信单元,用于若满足第一条件,基于第一模型与网络设备进行通信,其中,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
  39. 根据权利要求38所述的设备,其特征在于,所述第一模型包括第一初始模型,若所述第一条件为所述终端设备未被配置或未被指示模型信息,所述终端设备基于所述第一初始模型与所述网络设备进行通信。
  40. 根据权利要求38或39所述的设备,其特征在于,所述第一模型包括第一缺省模型,若所述第一条件为所述第一计时器到期,所述终端设备基于所述第一缺省模型与所述网络设备进行通信。
  41. 根据权利要求40所述的设备,其特征在于,所述设备还包括:
    启动单元,用于若所述终端设备接收到所述网络设备发送的第一指示信息,则启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
  42. 根据权利要求41所述的设备,其特征在于,所述设备还包括:
    执行单元,用于若在所述第一计时器运行期间,所述终端设备接收到所述网络设备发送的所述第一指示信息,则所述终端设备执行第一操作,所述第一操作与所述第一缺省模型的使用关联。
  43. 根据权利要求42所述的设备,其特征在于,所述第一操作包括重启所述第一计时器。
  44. 根据权利要求42所述的设备,其特征在于,所述第一操作,包括:
    所述终端设备暂停所述第一计时器,并继续基于所述第二模型进行数据处理;
    如果所述终端设备完成基于所述第二模型的数据处理,则所述终端设备恢复运行所述第一计时器。
  45. 根据权利要求42所述的设备,其特征在于,所述第一操作,包括:
    维持所述第一计时器的运行,当所述第一计时器到期时,
    如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第一缺省模型与所述网络设备进行通信;
    如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第一缺省模型与所述网络设备进行通信。
  46. 根据权利要求42所述的设备,其特征在于,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
  47. 根据权利要求46所述的设备,其特征在于,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
  48. 根据权利要求38-47中任一项所述的设备,其特征在于,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:
    所述终端设备接入的频段;
    为所述终端设备提供服务的运营商的运营商标识;
    所述终端设备所处的地理位置的信息;以及
    所述第一模型的历史使用信息。
  49. 根据权利要求39所述的设备,其特征在于,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
  50. 根据权利要求40-47中任一项所述的设备,其特征在于,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
  51. 根据权利要求38-50中任一项所述的设备,其特征在于,所述设备还包括:
    第一发送单元,用于向所述网络设备发送第一信息,所述第一信息包括所述终端设备支持的模型的信息,其中,所述终端设备支持的模型包括所述第一模型。
  52. 根据权利要求51所述的设备,其特征在于,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
  53. 根据权利要求51所述的设备,其特征在于,所述设备还包括:
    第二发送单元,用于向所述网络设备发送第二信息,所述第二信息用于指示所述第一模型。
  54. 根据权利要求38-53中任一项所述的设备,其特征在于,所述第一模型与第三模型关联,其中,所述第三模型为所述第一条件下所述网络设备所使用的模型。
  55. 根据权利要求54所述的设备,其特征在于,所述第一模型与第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
  56. 一种网络设备,其特征在于,包括:
    通信单元,用于若满足第一条件,基于第三模型与终端设备进行通信,所述第一条件包括所述终端设备未被配置或未被指示模型信息,和/或,第一计时器到期;
    其中,第一模型与所述第三模型关联,所述第一模型为满足所述第一条件时所述终端设备所使用的模型,所述第一计时器与所述终端设备基于第二模型与所述网络设备进行通信的时间关联。
  57. 根据权利要求56所述的设备,其特征在于,所述第三模型包括第二初始模型,若 所述第一条件为所述终端设备未被配置或未被指示模型信息,所述网络设备基于所述第二初始模型与所述终端设备进行通信。
  58. 根据权利要求56或57所述的设备,其特征在于,所述第三模型包括第二缺省模型,若所述第一条件为所述第一计时器到期,所述网络设备基于所述第二缺省模型与所述终端设备进行通信。
  59. 根据权利要求58所述的设备,其特征在于,所述设备还包括:
    启动单元,用于若所述网络设备向所述终端设备发送第一指示信息,则启动所述第一计时器,所述第一指示信息用于指示所述终端设备基于所述第二模型与所述网络设备进行通信。
  60. 根据权利要求59所述的设备,其特征在于,所述设备还包括:
    执行单元,用于若在所述第一计时器运行期间,所述网络设备向所述终端设备发送所述第一指示信息,则执行第一操作,所述第一操作与所述第二缺省模型的使用关联。
  61. 根据权利要求60所述的设备,其特征在于,所述第一操作包括重启所述第一计时器。
  62. 根据权利要求60所述的设备,其特征在于,所述第一操作,包括:
    所述网络设备暂停所述第一计时器;
    如果所述终端设备完成基于所述第二模型的数据处理,则所述网络设备恢复运行所述第一计时器。
  63. 根据权利要求60所述的设备,其特征在于,所述第一操作,包括:
    维持所述第一计时器的运行,当所述第一计时器到期时,
    如果所述终端设备基于所述第二模型的数据处理已完成,则基于所述第二缺省模型与所述终端设备进行通信;
    如果所述终端设备基于所述第二模型的数据处理未完成,则在基于所述第二模型的数据处理完成之后,基于所述第二缺省模型与所述终端设备进行通信。
  64. 根据权利要求60所述的设备,其特征在于,所述第一操作包括将所述第一计时器的计时时长增加第一时长。
  65. 根据权利要求64所述的设备,其特征在于,所述第一时长为预设时长,或者所述第一时长是所述网络设备指示的。
  66. 根据权利要求56所述的设备,其特征在于,所述第一模型包括第一初始模型,所述第一初始模型用于所述终端设备在所述第一条件为所述终端设备未被配置或未被指示模型信息的情况下使用。
  67. 根据权利要求56所述的设备,其特征在于,所述第一模型包括第一缺省模型,所述第一缺省模型用于所述终端设备在所述第一条件为第一计时器到期的情况下使用。
  68. 根据权利要求66或67所述的设备,其特征在于,所述第一模型基于第一规则确定,所述第一规则与以下信息中的一种或多种关联:
    所述终端设备接入的频段;
    为所述终端设备提供服务的运营商的运营商标识;
    所述终端设备所处的地理位置的信息;以及
    所述第一模型的历史使用信息。
  69. 根据权利要求66所述的设备,其特征在于,所述第一初始模型的信息承载在系统信息,或者其他终端设备发送的信息中。
  70. 根据权利要求67所述的设备,其特征在于,所述第一缺省模型的信息承载在RRC配置信息,或者其他终端设备发送的信息中。
  71. 根据权利要求56-70中任一项所述的设备,其特征在于,所述设备还包括:
    第一接收单元,用于接收所述终端设备发送的第一信息,所述第一信息包括所述终端 设备支持的模型的信息,其中,所述终端设备支持的模型包括第一模型,所述第一模型为所述第一条件下,所述终端设备所使用的模型。
  72. 根据权利要求71所述的设备,其特征在于,所述第一模型为所述终端设备支持的模型中具有特定标识的模型。
  73. 根据权利要求71所述的设备,其特征在于,所述设备还包括:
    第二接收单元,用于接收所述终端设备发送的第二信息,所述第二信息用于指示所述第一模型。
  74. 根据权利要求71-73中任一项所述的设备,其特征在于,所述第一模型与所述第三模型关联包括所述第一模型与所述第三模型之间具有关联关系,所述关联关系是由所述网络设备配置的。
  75. 一种终端设备,其特征在于,包括存储器和处理器,所述存储器用于存储程序,所述处理器用于调用所述存储器中的程序,以使所述终端设备执行如权利要求1-18中任一项所述的方法。
  76. 一种网络设备,其特征在于,包括存储器和处理器,所述存储器用于存储程序,所述处理器用于调用所述存储器中的程序,以使所述网络设备执行如权利要求19-37中任一项所述的方法。
  77. 一种装置,其特征在于,包括处理器,用于从存储器中调用程序,以使所述装置执行如权利要求1-18中任一项所述的方法。
  78. 一种装置,其特征在于,包括处理器,用于从存储器中调用程序,以使所述装置执行如权利要求19-37中任一项所述的方法。
  79. 一种芯片,其特征在于,包括处理器,用于从存储器调用程序,使得安装有所述芯片的设备执行如权利要求1-18中任一项所述的方法。
  80. 一种芯片,其特征在于,包括处理器,用于从存储器调用程序,使得安装有所述芯片的设备执行如权利要求19-37中任一项所述的方法。
  81. 一种计算机可读存储介质,其特征在于,其上存储有程序,所述程序使得计算机执行如权利要求1-18中任一项所述的方法。
  82. 一种计算机可读存储介质,其特征在于,其上存储有程序,所述程序使得计算机执行如权利要求19-37中任一项所述的方法。
  83. 一种计算机程序产品,其特征在于,包括程序,所述程序使得计算机执行如权利要求1-18中任一项所述的方法。
  84. 一种计算机程序产品,其特征在于,包括程序,所述程序使得计算机执行如权利要求19-37中任一项所述的方法。
  85. 一种计算机程序,其特征在于,所述计算机程序使得计算机执行如权利要求1-18中任一项所述的方法。
  86. 一种计算机程序,其特征在于,所述计算机程序使得计算机执行如权利要求19-37中任一项所述的方法。
PCT/CN2023/112991 2023-08-14 2023-08-14 用于无线通信的方法及设备 Pending WO2025035369A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2023/112991 WO2025035369A1 (zh) 2023-08-14 2023-08-14 用于无线通信的方法及设备
CN202380101304.4A CN121773649A (zh) 2023-08-14 2023-08-14 用于无线通信的方法及设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2023/112991 WO2025035369A1 (zh) 2023-08-14 2023-08-14 用于无线通信的方法及设备

Publications (1)

Publication Number Publication Date
WO2025035369A1 true WO2025035369A1 (zh) 2025-02-20

Family

ID=94631917

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/112991 Pending WO2025035369A1 (zh) 2023-08-14 2023-08-14 用于无线通信的方法及设备

Country Status (2)

Country Link
CN (1) CN121773649A (zh)
WO (1) WO2025035369A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115767570A (zh) * 2022-11-04 2023-03-07 中国信息通信研究院 一种无线通信系统ai模型注册方法和设备
CN115843054A (zh) * 2021-09-18 2023-03-24 维沃移动通信有限公司 参数选择方法、参数配置方法、终端及网络侧设备
CN115942298A (zh) * 2021-08-17 2023-04-07 华为技术有限公司 一种人工智能ai模型传输方法及装置
CN116074813A (zh) * 2021-10-29 2023-05-05 中国电信股份有限公司 无线通信方法及相关设备
CN116249166A (zh) * 2021-12-07 2023-06-09 华为技术有限公司 一种无线通信的方法和装置
WO2023115567A1 (en) * 2021-12-24 2023-06-29 Nec Corporation Methods, devices, and computer readable medium for communication

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115942298A (zh) * 2021-08-17 2023-04-07 华为技术有限公司 一种人工智能ai模型传输方法及装置
CN115843054A (zh) * 2021-09-18 2023-03-24 维沃移动通信有限公司 参数选择方法、参数配置方法、终端及网络侧设备
CN116074813A (zh) * 2021-10-29 2023-05-05 中国电信股份有限公司 无线通信方法及相关设备
CN116249166A (zh) * 2021-12-07 2023-06-09 华为技术有限公司 一种无线通信的方法和装置
WO2023115567A1 (en) * 2021-12-24 2023-06-29 Nec Corporation Methods, devices, and computer readable medium for communication
CN115767570A (zh) * 2022-11-04 2023-03-07 中国信息通信研究院 一种无线通信系统ai模型注册方法和设备

Also Published As

Publication number Publication date
CN121773649A (zh) 2026-03-31

Similar Documents

Publication Publication Date Title
WO2015109747A1 (zh) 小小区基站状态切换方法及装置、计算机存储介质
WO2025168095A1 (zh) 候选主小区配置信息的处理方法、设备、存储介质及产品
CN105359617B (zh) 用于改进(c+g)dsds设备中的调离操作的装置和方法
WO2025138269A1 (zh) 信息记录、接收方法和装置、终端、网络设备和存储介质
EP4633244A1 (en) Energy-saving control method and apparatus, and storage medium
WO2024169706A1 (zh) 一种用于通信的方法、装置、存储介质和程序产品
EP4068838B1 (en) Communication method and communication apparatus
WO2025030562A1 (zh) 用于无线通信的方法、终端设备和网络设备
WO2025030460A1 (zh) 无线通信的方法、终端设备以及网络设备
CN114501510B (zh) 一种通信方法和通信装置
WO2025168028A1 (zh) 小区切换方法、设备、存储介质及产品
WO2025035369A1 (zh) 用于无线通信的方法及设备
WO2025020132A1 (zh) 无线链路恢复方法及终端设备
US20240406808A1 (en) Confirming ue-triggered mobility using low-layer information
EP3864903B1 (en) Apparatus, method and computer program
WO2023108416A1 (zh) 无线通信的方法、终端设备及网络设备
US20260075649A1 (en) Cyclic prefix extension ramping for sensing in shared sidelink channel communications
US12593253B2 (en) Indicating UE-triggered mobility using low-layer information
WO2025156218A1 (zh) 无线通信方法、终端设备及网络设备
WO2025000294A1 (zh) 用于无线通信的方法及设备
WO2025020204A1 (zh) 用于激活scg的方法、终端设备以及网络设备
WO2025167353A1 (zh) 通信方法及通信装置
CN117279051A (zh) 中断影响的确定方法、网络侧设备和终端
WO2024169778A1 (zh) 信号传输方法、装置及存储介质
WO2026067534A1 (zh) 一种通信方法及通信装置

Legal Events

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

Ref document number: 23948804

Country of ref document: EP

Kind code of ref document: A1

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112026003240

Country of ref document: BR

WWE Wipo information: entry into national phase

Ref document number: 202617026577

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 2023948804

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

WWP Wipo information: published in national office

Ref document number: 202617026577

Country of ref document: IN