WO2023011655A1 - 一种通信方法及装置 - Google Patents

一种通信方法及装置 Download PDF

Info

Publication number
WO2023011655A1
WO2023011655A1 PCT/CN2022/110695 CN2022110695W WO2023011655A1 WO 2023011655 A1 WO2023011655 A1 WO 2023011655A1 CN 2022110695 W CN2022110695 W CN 2022110695W WO 2023011655 A1 WO2023011655 A1 WO 2023011655A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell
information
terminal device
future
target
Prior art date
Application number
PCT/CN2022/110695
Other languages
English (en)
French (fr)
Inventor
耿婷婷
曾宇
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP22852366.8A priority Critical patent/EP4369787A1/en
Publication of WO2023011655A1 publication Critical patent/WO2023011655A1/zh
Priority to US18/432,420 priority patent/US20240179603A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/24—Reselection being triggered by specific parameters
    • H04W36/32—Reselection being triggered by specific parameters by location or mobility data, e.g. speed data
    • H04W36/324—Reselection being triggered by specific parameters by location or mobility data, e.g. speed data by mobility data, e.g. speed data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/0005—Control or signalling for completing the hand-off
    • H04W36/0055—Transmission or use of information for re-establishing the radio link
    • H04W36/0061—Transmission or use of information for re-establishing the radio link of neighbour cell information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/0005—Control or signalling for completing the hand-off
    • H04W36/0055—Transmission or use of information for re-establishing the radio link
    • H04W36/0072—Transmission or use of information for re-establishing the radio link of resource information of target access point
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/16—Performing reselection for specific purposes
    • H04W36/18—Performing reselection for specific purposes for allowing seamless reselection, e.g. soft reselection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/02—Processing of mobility data, e.g. registration information at HLR [Home Location Register] or VLR [Visitor Location Register]; Transfer of mobility data, e.g. between HLR, VLR or external networks
    • H04W8/08—Mobility data transfer
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/16—Performing reselection for specific purposes
    • H04W36/22—Performing reselection for specific purposes for handling the traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements
    • H04W36/24—Reselection being triggered by specific parameters
    • H04W36/30—Reselection being triggered by specific parameters by measured or perceived connection quality data
    • H04W36/304—Reselection being triggered by specific parameters by measured or perceived connection quality data due to measured or perceived resources with higher communication quality

Definitions

  • the embodiments of the present application relate to the field of communication technologies, and in particular, to a communication method and device.
  • the terminal can perform cell access.
  • the cell access includes the switching of the corresponding cell under the primary station, and the mobility of the secondary station.
  • the mobility of the secondary station includes adding, deleting or changing the corresponding cell of the secondary station. wait. For example, if the signal quality of the current serving cell of the terminal is poor, but the signal quality of the neighboring cell is good, the terminal can access the neighboring cell.
  • Cell access refers to the transfer of a terminal's wireless link connection from a source cell to a target cell under the control of network equipment, and is a basic technical means to maintain seamless mobile communication services. How the terminal performs cell handover is a problem worth studying.
  • Embodiments of the present application provide a communication method and device, so as to implement cell handover of terminal equipment.
  • a communication method is provided, the method is executed by the second network device, and may also be a component (processor, chip, circuit or others) configured in the second network device, or may be a software module, etc. , including: sending a first message to a first network device corresponding to a first target cell, where the first target cell is a predicted serving cell that the terminal device can access, and the first message is used to indicate a first reasoning result
  • the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device, future business information of the terminal device, or future movement track information of the terminal device.
  • the second network device sends the AI reasoning result for determining the first target cell, that is, the first reasoning result, to the first network device.
  • the first network device may reuse the AI reasoning result or perform other operations to improve the utilization rate of the AI reasoning result.
  • the first target cell is determined according to the first reasoning result.
  • the future movement information of the terminal device includes predicted at least one of the following: information about the future cell of the terminal device, information about the residence time of the terminal device in the future cell, the The manner in which the terminal device accesses the future cell, whether the terminal device leaves the connected state in the future cell, or the prediction accuracy of the future movement information of the terminal device.
  • the future service information of the terminal device includes predicted at least one of the following: the future service type of the terminal device, the service quality QoS requirement of the future service, and the traffic volume of the future service , or the time information of the future business.
  • the method further includes: receiving feedback information from the first network device, where the feedback information includes indication information of at least one of the following: the actual Residence time information, whether the terminal device actually leaves the connected state in the first target cell, a second reasoning result, or a second target cell.
  • the feedback information is used to optimize or update parameters of a model for determining the first reasoning result.
  • the first network device can send feedback information to the network device based on the actual operating parameters of the terminal device. Based on the feedback information, the first network device may optimize or update parameters related to the AI model, for example, the AI model itself or input parameters of the AI model, so as to improve the reasoning accuracy of the AI model.
  • the first reasoning result includes a reasoning result related to the mobility of the primary cell of the terminal device, and/or a reasoning result related to the mobility of the secondary station of the terminal device.
  • the first message when the first inference result includes an inference result related to the mobility of the terminal device's primary cell, the first message includes the terminal device's source secondary station, source secondary cell Indication information whether at least one of the group, the source primary secondary cell, or the source secondary cell needs to be changed.
  • a communication method is provided.
  • the execution body of the method is the first network device, and may also be a component (processor, chip, circuit or others) configured in the first network device, or may be a software module, etc. , including: receiving a first message from a second network device, where the first message is used to indicate a first inference result, and the first inference result includes at least one of the following predicted items: future movement information of the terminal device , future service information of the terminal device, or future movement track information of the terminal device.
  • the first network device can use the first reasoning result to perform corresponding operations, thereby improving the utilization rate of the first reasoning result.
  • the first network device can directly perform AI reasoning on the basis of the first reasoning result without having to start reasoning from scratch, saving the second Computing resources and storage resources of a network device.
  • the first message is used to request the first network device to allocate resources corresponding to a first target cell for the terminal device, and the first target cell is a predicted cell that the terminal device can access. Serve the community.
  • the method further includes: allocating resources of the first target cell to the terminal equipment in response to the first message; sending the second network equipment allocated for the terminal equipment to the second network equipment. Resource indication information of a target cell.
  • the future movement information of the terminal device includes predicted at least one of the following:
  • Information about the future cell of the terminal device residence time information of the terminal device in the future cell, the way the terminal device accesses the future cell, whether the terminal device leaves the connection in the future cell state, or the prediction accuracy of the future mobile information of the terminal device.
  • the future service information of the terminal device includes predicted at least one of the following:
  • the future service type of the terminal device the service quality QoS requirement of the future service, the service volume of the future service, or the time information of the future service.
  • the method further includes: sending feedback information to the second network device, where the feedback information includes indication information of at least one of the following: the actual camping position of the terminal device in the first target cell time information, whether the terminal device actually leaves the connected state in the first target cell, the second reasoning result, or the second target cell.
  • the feedback information is used to optimize or update parameters of a model for determining the first reasoning result.
  • the first reasoning result includes a reasoning result related to the mobility of the primary cell of the terminal device, and/or a reasoning result related to the mobility of the secondary station of the terminal device.
  • the first message when the first inference result includes an inference result related to the mobility of the terminal device's primary cell, the first message includes the terminal device's source secondary station, source secondary cell Indication information whether at least one of the group, the source primary secondary cell, or the source secondary cell needs to be changed.
  • the device may be a network device, or a device in a configured network device, or a device that can be matched with the network device.
  • the device includes a one-to-one unit for performing the method/operation/step/action described in the first aspect, and the unit may be a hardware circuit, or software, or a combination of hardware circuit and software.
  • the device may include a processing unit and a communication unit, and the processing unit and the communication unit may perform corresponding functions in any design example of the first aspect above, specifically: the processing unit is configured to generate the first message; the communication A unit, configured to send a first message to a first network device corresponding to a first target cell, where the first target cell is a predicted serving cell that the terminal device can access, and the first message is used to indicate the first reasoning
  • the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device, future service information of the terminal device, or future movement track information of the terminal device.
  • the device includes a memory for implementing the method described in the first aspect above.
  • the apparatus may also include memory for storing instructions and/or data.
  • the memory is coupled to the processor, and when the processor executes the program instructions stored in the memory, the method described in the first aspect above can be implemented.
  • the device may also include a communication interface for the device to communicate with other devices.
  • the communication interface may be a transceiver, circuit, bus, module, pin or other types of communication interface.
  • the device includes:
  • a processor configured to generate a first message
  • a communication interface sending a first message to a first network device corresponding to a first target cell, where the first target cell is a predicted serving cell that the terminal device can access, and the first message is used to indicate a first reasoning result
  • the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device, future business information of the terminal device, or future movement track information of the terminal device.
  • the device may be a network device, or a device configured in the network device, or a device that can be used in conjunction with the network device.
  • the device includes a one-to-one unit for performing the methods/operations/steps/actions described in the second aspect.
  • the unit may be a hardware circuit, or software, or a combination of hardware circuit and software.
  • the apparatus may include a processing unit and a communication unit, and the processing unit and the communication unit may perform the corresponding functions in any design example of the second aspect above, specifically: the communication unit is configured to receive information from the second network device A first message, where the first message is used to indicate a first reasoning result, where the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device, future business information of the terminal device, Or the future movement track information of the terminal device.
  • a processing unit configured to process the first message.
  • the device includes a memory for implementing the method described in the second aspect above.
  • the apparatus may also include memory for storing instructions and/or data.
  • the memory is coupled to the processor, and when the processor executes the program instructions stored in the memory, the method described in the second aspect above can be implemented.
  • the device may also include a communication interface for the device to communicate with other devices.
  • the communication interface may be a transceiver, circuit, bus, module, pin or other types of communication interface.
  • the device includes:
  • a communication interface receiving a first message from a second network device, where the first message is used to indicate a first reasoning result, and the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device , future service information of the terminal device, or future movement track information of the terminal device.
  • the processor is configured to process the first message.
  • the embodiment of the present application further provides a computer-readable storage medium, including instructions, which, when run on a computer, cause the computer to execute the method of any one of the first aspect or the second aspect.
  • the embodiment of the present application further provides a system-on-a-chip, where the system-on-a-chip includes a processor and may further include a memory, configured to implement the method in any one of the first aspect or the second aspect.
  • the system-on-a-chip may consist of chips, or may include chips and other discrete devices.
  • the embodiments of the present application further provide a computer program product, including instructions, which, when run on a computer, cause the computer to execute the method of any one of the first aspect or the second aspect.
  • the embodiment of the present application further provides a system, the system includes the device of the third aspect or the fourth aspect, and the device of the fifth aspect or the sixth aspect.
  • FIG. 1 is a schematic diagram of a communication architecture provided by an embodiment of the present application.
  • FIG. 1a to Figure 2d are schematic diagrams of the AI model provided by the embodiment of the present application.
  • FIG. 7 and Figure 8 are schematic diagrams of the device provided by the embodiment of the present application.
  • Figure 9a is a schematic diagram of the structure of a neuron
  • Fig. 9b is a schematic diagram of the layer relationship of the neural network.
  • FIG. 1 is a schematic structural diagram of a communication system 1000 applied in an embodiment of the present application.
  • the communication system includes a radio access network 100 and a core network 200 , and optionally, the communication system 1000 may also include the Internet 300 .
  • the radio access network 100 may include at least one radio access network device (such as 110a and 110b in FIG. 1 ), and may also include at least one terminal (such as 120a-120j in FIG. 1 ).
  • the terminal is connected to the wireless access network device in a wireless manner, and the wireless access network device is connected to the core network in a wireless or wired manner.
  • the core network equipment and the wireless access network equipment can be independent and different physical equipment, or the functions of the core network equipment and the logical functions of the wireless access network equipment can be integrated on the same physical equipment, or it can be a physical equipment It integrates some functions of core network equipment and some functions of radio access network equipment. Terminals and wireless access network devices may be connected to each other in a wired or wireless manner.
  • FIG. 1 is only a schematic diagram.
  • the communication system may also include other network devices, such as wireless relay devices and wireless backhaul devices, which are not shown in FIG. 1 .
  • the radio access network equipment can be a base station (base station), an evolved base station (evolved NodeB, eNodeB), a transmission reception point (transmission reception point, TRP), and the next generation in the fifth generation (5th generation, 5G) mobile communication system
  • Base station (next generation NodeB, gNB), the next generation base station in the sixth generation (6th generation, 6G) mobile communication system, the base station in the future mobile communication system or the access node in the wireless fidelity (wireless fidelity, WiFi) system etc.; it can also be a module or unit that completes some functions of the base station, for example, it can be a centralized unit (central unit, CU) or a distributed unit (distributed unit, DU).
  • the CU here completes the functions of the base station's radio resource control (radio resource control, RRC) protocol and packet data convergence protocol (PDCP), and can also complete the service data adaptation protocol (service data adaptation protocol, SDAP)
  • RRC radio resource control
  • PDCP packet data convergence protocol
  • SDAP service data adaptation protocol
  • the function; the DU completes the functions of the radio link control (radio link control, RLC) layer and medium access control (medium access control, MAC) layer of the base station, and can also complete part of the physical (PHY) layer or all physical layers.
  • RLC radio link control
  • MAC medium access control
  • PHY physical
  • the radio access network device may be a macro base station (such as 110a in Figure 1), a micro base station or an indoor station (such as 110b in Figure 1), or a relay node or a donor node.
  • the embodiment of the present application does not limit the specific technology and specific equipment form adopted by the radio access network equipment.
  • a base station is used as an example of a radio access network device for description below.
  • a terminal may also be called terminal equipment, user equipment (user equipment, UE), mobile station, mobile terminal, and so on.
  • Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things ( internet of things, IOT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grid, smart furniture, smart office, smart wearables, smart transportation, smart city, etc.
  • Terminals can be mobile phones, tablet computers, computers with wireless transceiver functions, wearable devices, vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, etc.
  • the embodiment of the present application does not limit the specific technology and specific device form adopted by the terminal.
  • UE is used as an example of a terminal for description below.
  • Base stations and terminals can be fixed or mobile. Base stations and terminals can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; they can also be deployed on aircraft, balloons and artificial satellites in the air. The embodiments of the present application do not limit the application scenarios of the base station and the terminal.
  • the helicopter or UAV 120i in FIG. base station for base station 110a, 120i is a terminal, that is, communication between 110a and 120i is performed through a wireless air interface protocol.
  • communication between 110a and 120i may also be performed through an interface protocol between base stations.
  • 120i compared to 110a, 120i is also a base station. Therefore, both the base station and the terminal can be collectively referred to as a communication device, 110a and 110b in FIG. 1 can be referred to as a communication device with a base station function, and 120a-120j in FIG. 1 can be referred to as a communication device with a terminal function.
  • the communication between the base station and the terminal, between the base station and the base station, and between the terminal and the terminal can be carried out through the licensed spectrum, the communication can also be carried out through the unlicensed spectrum, and the communication can also be carried out through the licensed spectrum and the unlicensed spectrum at the same time; Communications may be performed on frequency spectrums below megahertz (gigahertz, GHz), or communications may be performed on frequency spectrums above 6 GHz, or communications may be performed using both frequency spectrums below 6 GHz and frequency spectrums above 6 GHz.
  • the embodiments of the present application do not limit the frequency spectrum resources used for wireless communication.
  • the functions of the base station may also be performed by modules (such as chips) in the base station, or may be performed by a control subsystem including base station functions.
  • the control subsystem including base station functions here may be the control center in the above application scenarios such as smart grid, industrial control, intelligent transportation, and smart city.
  • the functions of the terminal may also be performed by a module (such as a chip or a modem) in the terminal, or may be performed by a device including the terminal function.
  • the base station sends a downlink signal or downlink information to the terminal, and the downlink information is carried on the downlink channel;
  • the terminal sends an uplink signal or uplink information to the base station, and the uplink information is carried on the uplink channel.
  • the terminal needs to establish a wireless connection with the cell controlled by the base station.
  • a cell with which a terminal has established a wireless connection is called a serving cell of the terminal.
  • the terminal communicates with the serving cell, it will also be interfered by signals from neighboring cells.
  • a UE may switch serving cells.
  • the base station to which the UE's current serving cell belongs may be referred to as a source base station
  • the base station to which the UE's serving cell to be switched belongs to may be referred to as a target base station.
  • the source base station or the AI device can perform artificial intelligence (AI) reasoning to determine the AI target cell, and send a handover request to the base station corresponding to the AI target cell, called the target base station. Afterwards, if the target base station agrees to the handover request of the source base station, the UE can be handed over to the AI target cell.
  • AI artificial intelligence
  • the above AI reasoning results can be sent to the target base station corresponding to the AI target cell in the handover request. Subsequently, the base station corresponding to the AI target cell can use the AI reasoning result to perform a series of operations, thereby improving the utilization rate of the AI reasoning result.
  • the embodiment of the present application involves the process of predicting the serving cell to which the UE can handover by using the AI technology
  • the AI technology is firstly introduced for ease of understanding. It should be understood that this introduction is not intended to limit the embodiments of the present application.
  • AI is a technology that performs complex calculations by simulating the human brain. With the improvement of data storage and capabilities, AI has been applied more and more.
  • the 3rd generation partnership project ( 3rd generation partnership project, 3GPP) version 17 (release17, R17) has passed the research project (study item, SI), and proposes to apply AI to new radio (new radio, NR).
  • Figure 2a is an example diagram of the first application framework of AI in NR:
  • Data source is used to store training data and inference data.
  • the model training host (model training host) obtains the AI model by analyzing or training the training data provided by the data source, and deploys the AI model in the model inference host (model inference host).
  • the model inference node uses the AI model to perform inference based on the inference data provided by the data source, and obtains the inference result.
  • the reasoning results are used to give reasonable predictions based on AI for network operation, or guide the network to make policy configurations or policy adjustments. Relevant policy configuration or policy adjustment is uniformly planned by the execution (actor) entity, and sent to multiple execution objects (for example, network entities) for execution. At the same time, after the relevant strategies are applied, the specific performance of the network can be input to the data source again for storage.
  • Figure 2b, Figure 2c or Figure 2d is an example diagram of the second application framework of AI in NR:
  • a first AI module independent of the base station receives training data.
  • the first AI module obtains an AI model by analyzing or training the training data.
  • the first AI module may use the corresponding AI model and reasoning data to perform inference to obtain the parameter, as shown in Figure 2b; or the first AI module may send the information of the AI model to the base station
  • the second AI module (or described as located in the RAN) uses the corresponding AI model and inference data to perform inference to obtain the parameter, as shown in FIG. 2c.
  • the AI model used by the second AI module for reasoning may also be obtained by the second AI module receiving training data and performing training on the training data, as shown in FIG. 2d.
  • the AI model can be simply referred to as a model, which can be regarded as a mapping from input measurement quantities (measurement information) to output parameters.
  • the input measurement can be one or more measurements
  • the output parameter can be one or more parameters.
  • the training data may include known input measurements, or known input measurements and corresponding output parameters, for training the AI model.
  • the training data may be data from the base station, CU, CU-CP, CU-UP, DU, radio frequency module, UE and/or other entities, and/or data inferred by AI technology, without limitation.
  • Inference data includes input measurements that are used to infer parameters using the model.
  • Inference data may be data from a base station, CU, CU-CP, CU-UP, DU, radio module, UE and/or other entities.
  • the inferred parameters can be regarded as policy information and sent to the execution object.
  • the inferred parameters can be sent to the base station, CU, CU-CP, CU-UP, DU, radio frequency module, or UE, etc. for policy configuration or policy adjustment.
  • the AI models used for inferring different parameters can be the same or different without limitation.
  • the UE and/or the base station may perform some or all of the steps in the embodiment of the present application, these steps or operations are only examples, and the embodiment of the present application may also perform other operations or various operations deformation.
  • each step may be performed in a different order presented in the embodiment of the present application, and it may not be necessary to perform all operations in the embodiment of the present application.
  • this embodiment of the present application provides a flow of a communication method, which at least includes:
  • Step 300 the source base station determines a first inference result, and the first inference is also called an AI inference result, a first AI inference result or other names, etc., which are not limited.
  • an AI model is deployed in the source base station, and the AI model can refer to the introduction in FIG. 2a, FIG. 2c or FIG. 2d.
  • the source base station may perform AI reasoning based on the AI model to obtain a first reasoning result.
  • the source base station may use at least one of the following information as the input of the AI model, for example, historical trajectory information of the UE, historical residence information of the UE, current moving direction of the UE, speed of the UE, network information subscribed by the UE (for example, telecommunications, China Unicom or China Mobile, etc.), or the service requirements of the UE, etc., are input into the AI model, and the output of the AI module is the first reasoning result.
  • the AI device is deployed separately, and the AI device may be called a remote intelligent communication, a wireless intelligent controller, an AI node, or others, without limitation.
  • An AI model is deployed in the AI device, and the AI model may refer to the introduction in FIG. 2a or FIG. 2b.
  • the AI device may perform AI reasoning based on the AI model, determine a first reasoning result, and send indication information of the first reasoning result to the source base station.
  • the AI model in this application can be composed of various deep neural networks.
  • a neural network is a specific implementation of machine learning. Neural networks can be used to perform classification tasks, prediction tasks, and can also be used to establish conditional probability distributions among variables.
  • DNN deep neural network
  • GNN generative neural network
  • FNN feedforward neural network
  • CNN convolutional neural network
  • RNN recurrent neural network
  • GNN includes Generative Adversarial Network (GAN) and Variational Autoencoder (VAE).
  • GAN Generative Adversarial Network
  • VAE Variational Autoencoder
  • each neuron performs a weighted summation operation on its input values, and the weighted summation result is passed through an activation function to generate an output.
  • Fig. 9a is a schematic diagram of neuron structure.
  • the bias of the weighted sum Set to b the form of the activation function can be diversified.
  • DNN generally has a multi-layer structure. Each layer of DNN can contain multiple neurons. The input layer passes the received value to the middle hidden layer after being processed by neurons. Similarly, the hidden layer then passes the calculation result to the final output layer to generate the final output of the DNN. As shown in Fig. 9b, Fig. 9b is a schematic diagram of the layer relationship of the neural network. DNN generally has one or more hidden layers, and hidden layers often directly affect the ability to extract information and fit functions.
  • the parameters of each neuron include weights, biases and activation functions, and the set of parameters of all neurons in DNN is called DNN parameters (or called neural network parameters).
  • DNN parameters or called neural network parameters.
  • the weights and biases of neurons can be optimized through the training process, so that DNN has the ability to extract data features and express mapping relationships.
  • the parameters of the neural network include information related to the neural network, for example, may include one or more of the following:
  • the type of neural network such as a deep neural network, or a generative neural network
  • Information related to the neural network structure such as the type of the neural network, the number of layers of the neural network, the number of neurons, etc.;
  • the parameters of each neuron in the neural network such as weights, biases, and activation functions.
  • the first reasoning result includes at least one of the following: future mobility information of the UE, future service information of the UE, or prediction information of a moving trajectory of the UE.
  • the UE's movement track prediction information may refer to predicted geographic location information of the UE in the future.
  • the UE's movement track prediction information may be predicted location information A of the UE at a first time in the future, location information B of the UE at a second time in the future, and so on.
  • the predicted future mobility information and/or future business service information of the UE is obtained.
  • the UE's future mobility information may include predicted at least one of the following information:
  • the information of the future cell of the UE may be the information of the cell that the UE may access in the future time.
  • the UE's future cell information may include cell 1 to cell 10 and so on.
  • Information about each cell may include cell global identifier (CGI), physical cell identifier (PCI) and frequency point, cell identifier (cell ID), non-public network identifier ( At least one of non-public network identifier, NPN ID), non-terrestrial network identifier (non-terrestrial network identifier, NTN ID) or other cell identifiers.
  • the CGI may include a public land mobile network (public land mobile network, PLMN ID) and a cell ID.
  • the information of the cell may also include a tracking area code (tracking area code, TAC) and/or identification information of a network device to which the cell belongs, such as a global network device identification.
  • TAC tracking area code
  • the dwell time information may refer to the time when the UE receives service in a certain cell, or the time when a certain cell is used as a serving cell, etc.
  • the residence time is specifically the start time and end time of receiving services in the cell.
  • the start time may be called the start time stamp time
  • the end time may be called the end time stamp time, etc., or may be The length of time for the UE to receive services in the cell, which may be referred to as a time range or the like.
  • the information of the future cells may be sorted according to the dwell time of the UE in different future cells.
  • the manner in which the UE camps on the cell may be: the UE handovers to the cell, or the UE selects and accesses the cell, or the UE reselects to the cell, or the UE reestablishes to the cell, etc.
  • the way the UE accesses the future cell may include normal handover (legacy handover or ordinary handover), dual active protocol stack handover (dual active protocol stack handover, DAPS HO), conditional handover (conditional handover, CHO), no random Access handover (RACH-less HO) or other access methods, etc.
  • normal handover legacy handover or ordinary handover
  • dual active protocol stack handover dual active protocol stack handover
  • conditional handover conditional handover, CHO
  • RACH-less HO no random Access handover
  • the information may specifically be whether the predicted UE will leave the connected state in the future cell. For example, if it is predicted that the UE will leave the connected state in the future cell, it can be represented by a first value (such as 1); if it is predicted that the UE will not leave the connected state in the future cell, it can be represented by a second value (such as 0).
  • a first value such as 1
  • a second value such as 0
  • the first value (such as 00) can be used to identify it; if it is predicted that the UE will leave the connected state and enter the idle state in the future cell, it can be Use the second value (for example, 01) to indicate; if it is predicted that the UE will not leave the connected state in the future cell, it can be indicated by the third value (for example, 11).
  • the information about the future cell of the UE, information about the residence time of the UE in the future cell, the manner in which the UE accesses the future cell, and whether the UE leaves the connected state in the future cell can all be referred to as future mobility information of the UE.
  • future mobility information of the UE For each item of information in the above UE's future mobility information, an accuracy can be predicted.
  • an accuracy at the cell level may be comprehensively predicted.
  • all future cells include cell 1 to cell 10 .
  • the prediction accuracy of the cell can be obtained comprehensively. For example, the comprehensive prediction accuracy of cell 1 is 95%, the comprehensive prediction accuracy of cell 2 is 98%, and so on.
  • the future service information of the UE includes predicted at least one of the following: the future service type of the UE, the service quality (quality of service, QoS) requirement of the UE's future service, the service volume of the future service, or the future service time information, etc.
  • Historical input information X his (0, N)
  • trajectory information including one or more of the following: trajectory information, resident information, moving direction, speed, contracted network information (for example, China Telecom, China Unicom or China Mobile, etc.), or UE The service requirements, etc., and the actual output information Y his corresponding to the historical input of the UE, such as the information of the cell actually accessed or camped on, the way of accessing the cell, etc.
  • the historical input information X his (0,x+1) of [T 0 ,...,T x ] can be selected as the input of the DNN model, and the inference information Y inf of [T x+1 ] can be obtained (x+1).
  • the loss function L(x+1) is obtained by comparing Y inf (x+1) with Y his (x+1).
  • the calculation method of the loss function can be, for example, commonly used mean square error loss, KL divergence (Kullback–Leibler divergence) loss, etc., which is not limited in this solution.
  • p means The number of parameters in , that is, the number of items of actual output information corresponding to historical input, Indicates the value of parameter i at (x+1) time, Indicates the inferred value of parameter i at time (x+1).
  • the accuracy of the current inference can be judged, wherein the setting of the specific preset threshold can be based on system requirements. For example, when the loss function value corresponding to the inference result at a certain moment is greater than the preset threshold 5, it is considered that the parameters of the model need to be adjusted to reduce the loss function value.
  • the model is adjusted so that the loss function at all moments is lower than the target loss function value, that is, the aforementioned preset threshold, it can be considered based on [T 0 ...T x ,T x+1 ...T N ]
  • the historical input, the actual output corresponding to the historical input, and the inference results of the historical input, the model has been trained and converged, and it is a usable model, that is, it can be applied to prediction.
  • Step 301 The source base station determines a first target cell according to a first reasoning result, and the first target cell may also be called an AI target cell. Alternatively, it may be described as that the first target cell is determined according to the first reasoning result.
  • the first target cell is a predicted serving cell that the UE can access.
  • the source base station may select a cell from the future cell information in the first reasoning result as the first target cell.
  • the future cell information in the first reasoning result includes cell 1 to cell 10 .
  • the source base station may select cell 1 as the first target cell.
  • the source base station can consider the mobility trajectory information of the UE. In the future, the UE will appear in the service range of cell 1, or the source base station can determine that the UE is in cell 1 based on the information about the residence time of the UE in the future cell.
  • the residence time of the cell is the longest or relatively long, select cell 1 as the first target cell, etc.
  • Step 302 The source base station sends a first message to a target base station corresponding to the first target cell, where the first message is used to indicate a first inference result.
  • the first message may be a handover request message or other messages, which is not limited.
  • the source base station may indicate all or part of the information of the first reasoning result in the foregoing first message. That is, the source base station may notify the target base station of all or part of the information of the first inference result.
  • the future cell information in the first reasoning result includes cell 1 to cell 10, and the source base station selects cell 1 as the first target cell.
  • the source base station may notify the target base station of the information about cells 2 to 10 in the first reasoning result.
  • Step 303 The source base station receives a second message from the target base station.
  • the second message is used to indicate whether the target base station agrees to the handover request of the source base station.
  • the second message may be called a handover response message or other messages.
  • the above second message may be an affirmative response message, for example, a handover request acknowledge (handover request acknowledge) message.
  • the first target cell does not agree to the handover request of the source base station, that is, does not agree to the handover of the UE to the first target cell
  • the above-mentioned second message may be a negative response message, such as a handover preparation failure (handover preparation failure) message, or Handover failure (handover failure) message, etc.
  • the target base station may allocate resources of the first target cell for the UE in response to the first message, and send information about resources of the first target cell allocated to the UE to the source base station Instructions.
  • the resource indication information of the first target cell may be carried in the second message.
  • the source base station may indicate to the UE the resource of the first target cell allocated to the UE.
  • the UE can access the first target cell. For example, after the UE accesses the first target cell, the target base station corresponding to the first target cell may use actual information after the UE accesses the target base station.
  • feedback information may be sent to the source base station, or feedback information may be sent to the source base station based on other conditions, without limitation.
  • specific trigger conditions for sending feedback information please refer to the description in step 304, so that the AI model determined by the source base station or the AI device for the first inference result is optimized or updated to make the inference of the AI model more accurate.
  • Step 304 The target base station sends indication information of the feedback information to the source base station or the AI device.
  • the target base station sends feedback information indication information to the source base station, and the source base station updates the parameters of the AI model based on the feedback information.
  • the target base station may send indication information of the feedback information to the source base station, and the source base station sends all or part of the feedback information to the AI device.
  • the target base station may directly send the feedback information to the AI device through an interface between the target base station and the AI device.
  • the first message in step 302 above may carry information about the AI device, for example, the address information of the AI device, or the address information of the AI device.
  • the AI device Based on the feedback information, the AI device optimizes or adjusts the parameters of the AI model.
  • the feedback information is used to optimize or update the parameters of the model for determining the first reasoning result.
  • the input parameters of the AI model may be updated or prioritized according to the above feedback information, and/or the AI model itself may be optimized or updated, etc., without limitation.
  • the target base station can send feedback information to the source base station or the AI device when at least one of the following trigger conditions is met:
  • the target base station determines to handover the UE to the second target cell. For example, due to factors such as UE movement, the first target base station considers that the UE needs to be handed over from the first target cell to the second target cell. Regarding the manner in which the target base station determines the second target cell, it may be: the first target cell performs AI inference based on the AI model, and determines the second inference result. Alternatively, the AI device performs AI reasoning based on the AI model, determines a second reasoning result, and sends indication information of the second reasoning result to the first target cell. The first target cell determines the second target cell, etc. based on the second reasoning result. For example, the first target cell is cell 1, and the second target cell is cell 2. When the base station corresponding to cell 1 determines that the UE needs to be handed over to cell 2, the base station corresponding to cell 1 considers that the trigger condition is satisfied, and the base station corresponding to cell 1 can send feedback information to the source base station.
  • the service information of the UE in the first target cell changes.
  • the target base station may compare the service information of the UE in the first target cell with the service information predicted in the first reasoning result. When the difference between the two exceeds a certain range, it can be considered that the trigger condition is met, and feedback information can be sent to the source base station.
  • the second target cell determined by the target base station is different from the predicted cell in the received first reasoning result.
  • the second target upper area does not belong to the cell in the future cell information of the first inference result.
  • the future cell information in the first pushing result includes cell 1 to cell 10
  • the target base station determines that the cell to be handed over to by the UE next time is cell 20, it can be considered that the trigger condition is met, and feedback information can be sent to the source base station.
  • the access mode of the UE predicted by the target base station in the second target cell is different from the access mode predicted in the first reasoning result.
  • the target base station may determine the second reasoning result.
  • the second target cell and the access manner in the second target cell may be determined according to the second reasoning result.
  • the predicted access modes in the first reasoning result may include A, B, and C, and so on.
  • the UE access mode predicted by the target base station is F, it can be considered that the trigger condition is met, and feedback information can be sent to the source base station.
  • the UE's actual trajectory deviates from the UE's movement trajectory predicted in the first inference result.
  • the UE movement track predicted in the first reasoning result is that at the first time, the UE is predicted to be at position A; at the second time, the UE is predicted to be at position B; but at the first time, the UE is actually at position C; when the position When the distance between A and position C is greater than the preset condition, it can be considered that the trigger condition is met, and feedback information can be sent.
  • the feedback information sent by the target base station to the source base station or the AI device may include at least one of the following:
  • the information may be information about the actual residence time of the UE in cell 1.
  • the actual dwell time information may be the actual start time and the actual end time of the UE's stay in the cell 1 .
  • the actual resident time information may be the actual resident duration information of the UE in the cell 1 and the like.
  • the target base station may determine the second reasoning result according to the first reasoning result sent by the source base station.
  • the target base station may input the first inference result into the AI model as an input of the AI model, and the output of the AI model is the second inference result.
  • the target base station may send the first inference result to the AI device, and the AI device performs AI inference based on the first inference result, determines the second inference result, and sends the second inference result to the target base station and so on.
  • the type of information included in the second reasoning result refer to the description of the first reasoning result above.
  • the target base station determines the second target cell according to the second reasoning result.
  • the future cell information included in the second reasoning result is cell 2 to cell 10
  • the second target cell determined by the target base station may be cell 2 and so on.
  • the service type of the UE in the second target cell or the manner of accessing the second cell may be predicted by the target base station. For example, it may be predicted or inferred by the target base station based on the first inference result.
  • step 300, step 301, step 303 or step 304 in the process shown in FIG. 3 are all optional.
  • the source base station indicates the first reasoning result to the target base station.
  • the first target device can directly use the first reasoning result to perform AI reasoning to determine the second reasoning result; and determine the second target cell based on the second reasoning result. Since the source base station or the AI device consumes a lot of computing resources or storage resources when inferring the first inference result, doing so can improve the utilization rate of the first inference result. Furthermore, the first target device directly uses the first inference result to perform AI inference without inferring from scratch, which can also reduce the consumption of computing resources or storage resources of the first target device.
  • the first target cell may determine a trigger condition for feedback information according to the first reasoning result.
  • feedback information may be sent to the source base station or the AI device. Based on the feedback information, the parameters of the AI model are optimized or updated to improve the accuracy of subsequent AI reasoning and improve system efficiency.
  • the source base station sends the first reasoning result to the target base station in step 302 above.
  • How the target base station or other devices use the first reasoning result is not limited in this embodiment of the present application.
  • the above process of using the first reasoning result is only a schematic illustration.
  • the dual connectivity technology of the UE is firstly introduced.
  • the UE maintains connections with two base stations at the same time and receives services, which is called a dual connectivity architecture.
  • the dual connection architecture supported in the NR system also known as multi-radio dual connectivity (MR-DC) includes: dual connections composed of LTE base stations and NR base stations, or composed of NR base stations and NR base stations dual connectivity, or dual connectivity composed of an LTE base station and an LTE base station, etc.
  • the LTE base station includes an LTE base station connected to 4G core network equipment, or an LTE base station connected to 5G core network equipment.
  • NR base stations include NR base stations connected to 4G core network equipment, or NR base stations connected to 5G core network equipment.
  • the UE can maintain connections with two base stations, which are called a master node (MN) and a secondary node (SN) respectively.
  • MN master node
  • SN secondary node
  • the primary cell group includes at least one cell.
  • the primary cell group may include a primary cell (primary cell, PCell), and may also include at least one secondary cell (secondary cell, SCell) when carrier aggregation (carrier aggregation, CA) is configured.
  • the cell group in which the secondary station provides air interface resources for the UE is called a secondary cell group (SCG).
  • the secondary cell group includes at least one cell.
  • the secondary cell group may include a primary secondary cell (PSCell), and may also include at least one secondary cell when CA is configured.
  • a flow of a communication method is provided.
  • the flow may be a specific application of the flow shown in FIG. 3 in a dual connection architecture, and at least includes the following steps:
  • Step 400 The source master station determines the first reasoning result.
  • the first reasoning result may include a reasoning result related to primary cell mobility and/or a reasoning result related to SN mobility.
  • the inference result related to the mobility of the primary cell and the inference result related to the SN mobility may be inferred by using the same AI model, or may be inferred by using different AI models, etc., without limitation.
  • the reasoning results related to the mobility of the primary cell and the reasoning results related to the SN mobility may be derived by the source master station based on the AI model, or by the AI device based on the AI model. Or, any one of the above two inference results is inferred by the source master station, and the other one is inferred by the AI device, etc., without limitation.
  • the inference result related to the mobility of the primary cell may include at least one of the following: mobility information of the UE's future primary cell/primary station/primary cell group, future service information of the UE in the future primary cell/primary station/primary cell group, Or UE's movement trajectory prediction information, etc.
  • the reasoning results related to SN mobility may include at least one of the following: mobility information of the UE's future primary secondary cell/secondary station/secondary cell group, future service information of the UE in the future primary secondary cell/secondary station/secondary cell group, or Mobile trajectory prediction information of the UE, etc.
  • Step 401 The source primary station determines a first target primary cell based on a first reasoning result, and the first target primary cell is a predicted primary cell that the UE can access.
  • the source primary station may determine the first target primary cell according to the inference result related to the mobility of the primary cell in the first inference result.
  • the future primary cells in the inference result related to the mobility of the primary cell include primary cell 1 to primary cell 10 .
  • the source master station finds through judgment that the primary cell 1 can be used as the primary cell of the UE, and then the primary cell 1 can be regarded as the above-mentioned first target primary cell.
  • Step 402 The source primary station sends a first message to the base station corresponding to the first target primary cell, and the base station corresponding to the first target primary cell may be called a target primary station.
  • the first message is used to request handover of the primary cell of the UE to the first target primary cell, and the first message may include indication information of the above-mentioned first reasoning result.
  • the first message may include at least one of the mobility inference result of the primary cell and the mobility inference result of the SN. If the target base station agrees to the request of the first message, the target base station may configure information related to the primary cell for the UE when receiving the mobility inference result of the primary cell, for example, configure a primary cell group for the UE. After the UE accesses the first target primary cell, the first target primary cell can add, change or delete secondary stations for the UE according to the mobility reasoning result of the SN.
  • the target master station after the target master station changes or adds a secondary station for the UE, it can also add or update the relevant information of the secondary station for the UE configuration according to the reasoning results related to the SN mobility, for example, for the added or updated secondary station configuration Relevant information of the secondary cell group, etc. or,
  • the target master station can determine the triggering condition of the feedback information of the reasoning result related to the mobility of the primary cell according to the reasoning result related to the mobility of the primary cell. For example, when the difference between the predicted information in the reasoning results related to the main cell and the actual information after the UE accesses the first main station exceeds the preset value, the target main station can send feedback information to the source base station or AI equipment, etc., To optimize or update the parameters related to the AI model. Similarly, the target master station may also determine the triggering conditions of the feedback information of the SN mobility-related reasoning results according to the SN mobility-related reasoning results.
  • the first message may include the source secondary station, source secondary cell group, and source primary secondary cell of the terminal device. Or indication information indicating whether at least one item in the source secondary cell needs to be changed.
  • Step 403 The source master station receives a second message from the target master station, and the second message may be a response message to the above-mentioned first message.
  • the second message may be an acknowledgment message, indicating that the target primary station agrees to the request of the source primary station, and the primary cell of the UE may be handed over to the first target primary cell.
  • the primary cell of the UE is handed over to the first target primary cell, and it can be considered that the primary station of the UE is handed over from the source primary station to the target primary station.
  • the target primary station agrees to the UE's request, it can determine the relevant configuration of the primary station according to the reasoning result of the mobility of the primary cell. For example, after switching the target master station to the master station, the configuration of the master cell group, etc.
  • the relevant configuration of the master station may be carried in the second message in step 403 above.
  • the target primary station can configure the relevant information of the secondary station for the UE according to the reasoning result related to the mobility of the SN.
  • the second message may be a negative response message, indicating that the target primary station does not agree to the request of the source primary station, and the primary cell of the UE cannot be handed over to the first target primary cell.
  • Step 404 The target master station sends feedback information to the source master station or the AI device, where the feedback information is used to update or optimize the parameters of the AI model used to determine the first inference result.
  • the feedback information may include feedback information on inference results related to the mobility of the primary cell, and/or feedback information on inference results related to SN mobility.
  • the related description in step 304 which will not be repeated here.
  • the difference from the above is that if the reasoning results related to the mobility of the main cell and the reasoning results related to the SN mobility are inferred from different AI models, the above reasoning results related to the mobility of the main cell are used for the corresponding AI model
  • the parameters are optimized or updated.
  • the inference results related to SN mobility are used to optimize or update the parameters of the corresponding AI model.
  • the relevant description in step 304 which will not be repeated here.
  • the target master station sends the feedback information of the above-mentioned first reasoning result to the source master station or AI device, and the source master station or AI device can optimize or update the relevant parameters of the AI model of the first reasoning result based on the above-mentioned feedback information , so that the deduced configuration of the primary cell to be handed over by the UE or the mobility of the SN is more accurate. Further, the target master station or the source master station can configure more reasonable multi-connection configuration for the UE according to whether SN mobility is configured for the UE according to the first reasoning result, and upgrade system information.
  • a flow of a communication method is provided, which is mainly used for adding or changing UE secondary stations triggered by the primary station, and at least includes the following steps:
  • Step 500 the master station determines a first reasoning result.
  • the first reasoning result may include a reasoning result related to SN mobility.
  • the reasoning results related to SN mobility reference may be made to step 400, which will not be repeated here.
  • Step 501 the primary station determines the first target secondary station according to the first reasoning result.
  • the primary station may determine whether to add or change the secondary station of the UE according to the first reasoning result. For example, based on the future trajectory information of the UE in the inference results related to SN mobility, the primary station determines that the current secondary station cannot provide services for the UE in the future, or the service quality of the current secondary station cannot meet the requirements in the future, etc.
  • the primary station can determine the first target secondary station according to the method. For example, the primary station may determine the first target secondary station according to information such as the future primary-secondary cell/secondary station/primary-secondary cell group in the inference results related to SN mobility.
  • the master station discovers that the future cells 1 to 3 in the reasoning results related to SN mobility can be used as the secondary cell group of the UE in the future, and the base stations corresponding to the above cells 1 to 3 can be called the first target secondary cells. stand.
  • Step 502 the primary station sends a first message to the first target secondary station, the first message may be a request message requesting to add or change the first target secondary station to a secondary station of the UE, and the first message may include the first reasoning An indication of the result.
  • the first target secondary station when the first target secondary station agrees to be changed or added as a UE secondary station, the first target secondary station can determine the information related to the secondary station configured for the SN according to the first reasoning result, for example, the secondary station Cell group information, primary and secondary cell information, or secondary cell information, etc. Or, after the UE accesses the first target secondary station, the first target secondary station may perform AI reasoning according to the first reasoning result to determine future target secondary stations that the UE may add or change.
  • Step 503 the primary station receives the second message from the first target secondary station.
  • the second message may be an acknowledgment message, indicating that the first target secondary station agrees to be added or changed as a secondary station of the UE.
  • the second message may be a negative response message, indicating that the first target secondary station does not agree to be added or changed as a secondary station of the UE.
  • the second message may include secondary station related information configured by the first target secondary station for the UE, for example, secondary cell group information, primary and secondary Cell information or secondary cell information, etc.
  • the first target secondary station may notify the master station of the secondary station-related information configured for the UE through the second message, and the master station forwards it to the UE.
  • the first target secondary station may directly notify the UE of the information related to the secondary station configured above, without limitation.
  • Step 504 the first target secondary station sends feedback information to the primary station or the AI device.
  • the first target secondary station can optimize or update the parameters related to the AI model for inferring SN mobility according to the above feedback information.
  • this is based on the reasoning of the SN mobility obtained by the main station according to the AI model. under the premise of the result.
  • the first target secondary station can directly send the above feedback information to the AI device, or can send it to the master station, and the master station forwards it to the AI device wait.
  • steps 500, 501, 503 or 504 may be optional.
  • the source secondary station obtains the first target secondary station according to the first reasoning result. Afterwards, the source secondary station notifies the master station of the first reasoning result and the first target secondary station, and the master station then triggers the process of adding or changing the first target secondary station, which at least includes the following steps:
  • Step 600 the source secondary station determines the first reasoning result.
  • the first reasoning result may include a reasoning result related to SN mobility.
  • the reasoning results related to SN mobility reference may be made to step 400, which will not be repeated here.
  • the first reasoning result may be obtained by the source secondary station according to the AI model reasoning, or the AI device reasoning according to the AI model, and then notifying the source secondary station.
  • Step 601 The source secondary station determines the first target secondary station based on the first reasoning result.
  • Step 602 the source secondary station sends the first inference result and the indication information of the first target secondary station to the master station.
  • the source secondary station may only send the indication information of the first reasoning result to the primary station.
  • the primary station determines the first target secondary station according to the first reasoning result.
  • the indication information of the first target secondary station may be identification information of the first target secondary station, such as a global node identifier.
  • Step 603 the primary station sends a first message to the first target secondary station, the first message includes indication information of the first inference result, and the first message is used to request to add or change the first target secondary station as a secondary station of the UE stand.
  • Step 604 the primary station receives a second message from the first target secondary station, and the second message may be a response message to the first message.
  • Step 605 the first target secondary station sends indication information of the feedback message to the primary station.
  • Step 606 the primary station sends the indication information of the feedback message to the source secondary station.
  • the first target secondary station may send the indication information of the feedback information to the primary station, and the primary station forwards it to the source secondary station.
  • the source secondary station optimizes or updates the AI model that infers the inference result related to the mobility of the SN.
  • the first target secondary station may directly send the indication information of the feedback information to the source secondary station.
  • the first message in step 603 may need to carry address information or identification information of the source secondary station.
  • the first target secondary station can directly send the indication information of the feedback information to the AI device, or forward it to the AI device via the primary station and the source secondary station.
  • Equipment, etc. are not limited.
  • step 601 if the source base station determines to release the source base station according to the first reasoning result. Then the source secondary station may also send release instruction information to the master station, and the master station forwards the release instruction information to the UE source secondary station. Optionally, after releasing the source secondary station, the master station may send indication information of a feedback message to the source secondary station.
  • steps in the various processes described in FIG. 3 to FIG. 6 are not all the steps that need to be executed, and some steps can be added or deleted based on the actual needs of each process. For example, steps 300, 301, 303 and 304 in the above-mentioned process of FIG. 3 can be selectively executed.
  • FIG. 3 to FIG. 6 a hardware device as a whole is used as an example to describe, and actions of various modules inside the hardware device are not described.
  • the operations between internal modules of the hardware device and the operations of each module are also within the protection scope of the embodiments of the present application.
  • functions of an access network device may be implemented by multiple modules of common standards.
  • the functions of a base station may be implemented by a CU module or a DU module.
  • the actions of the source base station may be described as a whole: the source base station determines the first reasoning result, determines the first target cell according to the first reasoning result, and sends the first target cell to the target base station.
  • the entire processing process shown in Figure 3 may include: the CU determines the first inference result, determines the first target cell according to the first inference result; sends the first inference result to the DU ; The DU sends a first message to the target base station, where the first message includes indication information of the first reasoning result.
  • the description of "carrying certain indication information in a message" is adopted.
  • the indication information of the first reasoning result and the like are carried in the first message.
  • the message may directly indicate the corresponding information, for example, the information is directly carried in the message.
  • the message may indirectly indicate corresponding information.
  • the message A includes indication information of the information X
  • the data A may directly indicate the information X, for example, the data A carries the information X.
  • this data A may indicate information X indirectly.
  • the data A may carry other information of the information X and the like.
  • the apparatus 700 may include: a communication unit 701 configured to support communication between the apparatus 700 and other devices.
  • the communication unit 701 is also referred to as a transceiver unit, and may include a receiving unit and/or a sending unit, configured to perform receiving and sending operations respectively.
  • the processing unit 702 is configured to support the device to perform processing.
  • the device 700 may further include a storage unit 703 for storing program codes and/or data of the device 700 .
  • the foregoing apparatus 700 may be a network device or a module, chip or circuit in the network device.
  • the communication unit 701 is configured to execute the transceiving operation of the source base station in the flow shown in FIG. 3 above;
  • the processing unit 702 is configured to execute the processing operation of the source base station in the flow shown in FIG. 3 above.
  • the processing unit 702 is configured to generate the first message; the communication unit 701 is configured to send the first message to the first network device corresponding to the first target cell, the first target cell is predicted and the terminal device can access , the first message is used to indicate a first reasoning result, and the first reasoning result includes at least one of the following predicted items: future movement information of the terminal device, future service information of the terminal device, Or the future movement track information of the terminal device.
  • the first target cell is determined according to the first reasoning result.
  • the future movement information of the terminal device includes predicted at least one of the following: information about the future cell of the terminal device, information about the residence time of the terminal device in the future cell, the The manner in which the terminal device accesses the future cell, whether the terminal device leaves the connected state in the future cell, or the prediction accuracy of the future movement information of the terminal device.
  • the future service information of the terminal device includes predicted at least one of the following: the future service type of the terminal device, the service quality QoS requirement of the future service, and the traffic volume of the future service , or the time information of the future business.
  • the communication unit 701 is further configured to: receive feedback information from the first network device, where the feedback information includes at least one of the following indication information; Information about the actual residence time of a target cell, whether the terminal device actually leaves the connected state in the first target cell, a second reasoning result, or a second target cell.
  • the feedback information is used to optimize or update parameters of a model for determining the first reasoning result.
  • the first reasoning result includes a reasoning result related to the mobility of the primary cell of the terminal device, and/or a reasoning result related to the mobility of the secondary station of the terminal device.
  • the first message when the first inference result includes an inference result related to the mobility of the terminal device's primary cell, the first message includes the terminal device's source secondary station, source secondary cell Indication information whether at least one of the group, the source primary secondary cell, or the source secondary cell needs to be changed.
  • the foregoing apparatus 700 may be a network device or a module, chip or circuit in the network device.
  • the communication unit 701 is configured to execute the transceiving operation of the target base station in the flow shown in FIG. 3 above;
  • the processing unit 702 is configured to execute the processing operation of the target base station in the flow shown in FIG. 3 above.
  • the communication unit 701 is configured to receive a first message from a second network device, where the first message is used to indicate a first inference result, and the first inference result includes at least one of the following predictions: the terminal The future movement information of the equipment, the future service information of the terminal equipment, or the future movement track information of the terminal equipment.
  • the processing unit 702 is configured to process the first message.
  • the first message is used to request the first network device to allocate resources corresponding to a first target cell for the terminal device, and the first target cell is a predicted cell that the terminal device can access. Serve the community.
  • the processing unit 702 is further configured to allocate resources of the first target cell for the terminal device in response to the first message; the communication unit 701 is further configured to send the resource of the second The network device sends indication information of resources of the first target cell allocated for the terminal device.
  • the future movement information of the terminal device includes predicted at least one of the following: information about the future cell of the terminal device, information about the residence time of the terminal device in the future cell, the The manner in which the terminal device accesses the future cell, whether the terminal device leaves the connected state in the future cell, or the prediction accuracy of the future movement information of the terminal device.
  • the future service information of the terminal device includes predicted at least one of the following: the future service type of the terminal device, the service quality QoS requirement of the future service, and the traffic volume of the future service , or the time information of the future business.
  • the communication unit 701 is further configured to: send feedback information to the second network device, where the feedback information includes indication information of at least one of the following: Information about the actual residence time of the target cell, whether the terminal device actually leaves the connected state in the first target cell, the second reasoning result, or the second target cell.
  • the feedback information is used to optimize or update parameters of a model for determining the first reasoning result.
  • the first reasoning result includes a reasoning result related to the mobility of the primary cell of the terminal device, and/or a reasoning result related to the mobility of the secondary station of the terminal device.
  • the first message when the first inference result includes an inference result related to the mobility of the terminal device's primary cell, the first message includes the terminal device's source secondary station, source secondary cell Indication information whether at least one of the group, the source primary secondary cell, or the source secondary cell needs to be changed.
  • each unit in the device can be implemented in the form of software called by the processing element; they can also be implemented in the form of hardware; some units can also be implemented in the form of software called by the processing element, and some units can be implemented in the form of hardware.
  • each unit can be a separate processing element, or it can be integrated in a certain chip of the device.
  • it can also be stored in the memory in the form of a program, which is called and executed by a certain processing element of the device. Function.
  • all or part of these units can be integrated together, or implemented independently.
  • the processing element mentioned here may also be a processor, which may be an integrated circuit with signal processing capabilities.
  • each operation of the above method or each unit above may be realized by an integrated logic circuit of hardware in the processor element, or implemented in the form of software called by the processing element.
  • the units in any of the above devices may be one or more integrated circuits configured to implement the above method, for example: one or more specific integrated circuits (application specific integrated circuit, ASIC), or, one or Multiple microprocessors (digital signal processor, DSP), or, one or more field programmable gate arrays (field programmable gate array, FPGA), or a combination of at least two of these integrated circuit forms.
  • ASIC application specific integrated circuit
  • DSP digital signal processor
  • FPGA field programmable gate array
  • the units in the device can be implemented in the form of a processing element scheduler
  • the processing element can be a processor, such as a general-purpose central processing unit (central processing unit, CPU), or other processors that can call programs.
  • CPU central processing unit
  • these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).
  • the above unit for receiving is an interface circuit of the device for receiving signals from other devices.
  • the receiving unit is an interface circuit for the chip to receive signals from other chips or devices.
  • the above sending unit is an interface circuit of the device, and is used to send signals to other devices.
  • the sending unit is an interface circuit used by the chip to send signals to other chips or devices.
  • the network device may be an access network device (such as a source base station or a target base station, etc.).
  • the access network device 800 may include one or more DUs 801 and one or more CUs 802.
  • the DU 801 may include at least one antenna 8011, at least one radio frequency unit 8012, at least one processor 8013 and at least one memory 8014.
  • the DU801 is mainly used for transmitting and receiving radio frequency signals, converting radio frequency signals and baseband signals, and processing part of the baseband.
  • the CU 802 may include at least one processor 8022 and at least one memory 8021 .
  • the CU802 part is mainly used for baseband processing, controlling access network equipment, and the like.
  • the DU801 and the CU802 may be physically set together, or physically separated, that is, a distributed base station.
  • the CU802 is the control center of the access network equipment, and can also be called a processing unit, which is mainly used to complete the baseband processing function.
  • the CU 802 may be used to control the access network device to execute the operation procedures related to the access network device in the foregoing method embodiments.
  • the access network device 800 may include one or more radio frequency units, one or more DUs, and one or more CUs.
  • the DU may include at least one processor 8013 and at least one memory 8014
  • the radio frequency unit may include at least one antenna 8011 and at least one radio frequency unit 8012
  • the CU may include at least one processor 8022 and at least one memory 8021.
  • the CU802 can be composed of one or more single boards, and multiple single boards can jointly support a wireless access network (such as a 5G network) with a single access indication, or can separately support wireless access networks of different access standards.
  • Access network (such as LTE network, 5G network or other networks).
  • the memory 8021 and the processor 8022 may serve one or more boards. That is to say, memory and processors can be set independently on each single board. It may also be that multiple single boards share the same memory and processor. In addition, necessary circuits can also be set on each single board.
  • the DU801 can be composed of one or more single boards, and multiple single boards can jointly support a wireless access network (such as a 5G network) with a single access indication, or can respectively support wireless access networks of different access standards (such as a 5G network). LTE network, 5G network or other networks).
  • the memory 8014 and the processor 8013 may serve one or more boards. That is to say, memory and processors can be set independently on each single board. It may also be that multiple single boards share the same memory and processor. In addition, necessary circuits can also be set on each single board.
  • the access network equipment shown in FIG. 8 can implement various processes involving the source base station and the target base station in the foregoing method embodiments.
  • the operations and/or functions of the various modules in the access network device shown in FIG. 8 are respectively for realizing the corresponding processes in FIGS. 3 to 6 in the above method embodiment.
  • system and “network” in the embodiments of the present application may be used interchangeably.
  • “At least one” means one or more, and “plurality” means two or more.
  • “And/or” describes the association relationship of associated objects, indicating that there can be three types of relationships, for example, A and/or B, which can mean: A exists alone, A and B exist simultaneously, and B exists alone, where A, B can be singular or plural.
  • the character “/” generally indicates that the contextual objects are an “or” relationship.
  • “At least one of the following” or similar expressions refer to any combination of these items, including any combination of single or plural items. For example "at least one of A, B or C” includes A, B, C, AB, AC, BC or ABC. And, unless otherwise specified, ordinal numerals such as “first” and “second” mentioned in the embodiments of this application are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects degree etc.
  • the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions
  • the device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

Landscapes

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

Abstract

一种通信方法及装置，该方法包括：第二网络设备向第一目标小区对应的第一网络设备发送第一消息，第一目标小区为预测的、终端设备能够接入的服务小区，第一消息用于指示第一推理结果，第一推理结果中包括预测的以下至少一项：终端设备的未来移动信息、终端设备的未来业务信息、或终端设备的未来移动轨迹信息。第一网络设备可以根据接收的第一推理结果，再次进行AI推理或其它操作等，从而提高第一推理结果的利用率。

Description

一种通信方法及装置
相关申请的交叉引用
本申请要求在2021年08月06日提交中国专利局、申请号为202110900384.2、申请名称为“一种通信方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及通信技术领域，尤其涉及一种通信方法及装置。
背景技术
在通信系统中，终端可以进行小区接入，该小区接入包括主站下属对应小区的切换，和辅站的移动性，所述辅站的移动性包括辅站对应小区的添加、删除或变更等。比如，终端的当前服务小区的信号质量较差，而邻区的信号质量较好，则终端可以接入到邻区。小区接入可以指终端在网络设备的控制下完成从源小区到目标小区的无线链路连接的迁移，是保持无缝的移动通信服务的基本技术手段。终端如何进行小区切换，是一个值得研究的问题。
发明内容
本申请实施例提供一种通信方法及装置，以实现终端设备的小区切换。
第一方面，提供一种通信方法，该方法的执行主体为第二网络设备，还可以为配置于第二网络设备中的部件(处理器、芯片、电路或其它)，或者可以为软件模块等，包括：向第一目标小区对应的第一网络设备发送第一消息，所述第一目标小区为预测的、终端设备能够接入的服务小区，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
通过上述设计，第二网络设备将用于确定第一目标小区的AI推理结果，即第一推理结果发送给第一网络设备。第一网络设备可以再次利用所述AI推理结果或进行其它操作，提高AI推理结果的利用率。
在一种可能的设计中，所述第一目标小区是根据所述第一推理结果确定的。
在一种可能的设计中，所述终端设备的未来移动信息包括预测的以下至少一项：所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
在一种可能的设计中，所述终端设备的未来业务信息包括预测的以下至少一项：所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
在一种可能的设计中，还包括：接收来自所述第一网络设备的反馈信息，所述反馈信息中包括以下至少一项的指示信息：所述终端设备在所述第一目标小区的实际驻留时间信 息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
在一种可能的设计中，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
在该设计中，当终端设备实际接入第一网络设备后，第一网络设备可以基于终端设备的实际运行参数，向网络设备发送反馈信息。第一网络设备基于该反馈信息可以对AI模型相关的参数，例如，AI模型本身或AI模型的输入参数，进行优化或更新，提高AI模型的推理准确度。
在一种可能的设计中，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
在一种可能的设计中，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
第二方面，提供一种通信方法，该方法的执行主体为第一网络设备，还可以为配置于第一网络设备中的部件(处理器、芯片、电路或其它)，或者可以为软件模块等，包括：接收来自第二网络设备的第一消息，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
通过上述设计，第一网络设备可以利用第一推理结果进行相应的操作，从而提高第一推理结果的利用率。另一方面，以第一网络设备利用第一推理结果，再次进行AI推理为例，第一网络设备可以在第一推理结果的基础上，直接进行AI推理，而不必从头开始推理，节省了第一网络设备的计算资源和存储资源。
在一种可能的设计中，所述第一消息用于请求第一网络设备为所述终端设备分配第一目标小区对应的资源，所述第一目标小区为预测的、终端设备能够接入的服务小区。
在一种可能的设计中，还包括：响应于所述第一消息，为所述终端设备分配所述第一目标小区的资源；向所述第二网络设备发送为所述终端设备分配的第一目标小区的资源的指示信息。
在一种可能的设计中，所述终端设备的未来移动信息包括预测的以下至少一项：
所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
在一种可能的设计中，所述终端设备的未来业务信息包括预测的以下至少一项：
所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
在一种可能的设计中，还包括：向所述第二网络设备发送反馈信息，所述反馈信息中包括以下至少一项的指示信息：所述终端设备在所述第一目标小区的实际驻留时间信息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
在一种可能的设计中，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
在一种可能的设计中，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
在一种可能的设计中，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
第三方面，提供一种装置，有益效果可参见第一方面的记载，该装置可以是网络设备，或者是配置中网络设备中的装置，或者能够和网络设备匹配使用的装置。一种设计中，该装置包括执行第一方面中所描述的方法/操作/步骤/动作一一对应的单元，该单元可以是硬件电路，也可是软件，也可以是硬件电路结合软件实现。示例性地，该装置可以包括处理单元和通信单元，且处理单元和通信单元可以执行上述第一方面任一种设计示例中的相应功能，具体的：处理单元，用于生成第一消息；通信单元，用于向第一目标小区对应的第一网络设备发送第一消息，所述第一目标小区为预测的、终端设备能够接入的服务小区，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
关于处理单元和通信单元的具体执行过程可参见上述第一方面的记载。
第四方面，提供一种装置，有益效果可参见第一方面的记载，所述装置包括存储器，用于实现上述第一方面描述的方法。所述装置还可以包括存储器，用于存储指令和/或数据。所述存储器与所述处理器耦合，所述处理器执行所述存储器中存储的程序指令时，可以实现上述第一方面描述的方法。所述装置还可以包括通信接口，所述通信接口用于该装置和其它设备进行通信。示例性地，通信接口可以是收发器、电路、总线、模块、管脚或其它类型的通信接口。在一种可能的设计中，该装置包括：
存储器，用于存储程序指令；
处理器，用于生成第一消息；
通信接口，向第一目标小区对应的第一网络设备发送第一消息，所述第一目标小区为预测的、终端设备能够接入的服务小区，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
关于处理器和通信接口的具体执行过程可参见上述第一方面的记载。
第五方面，提供一种装置，有益效果可参见第二方面的记载，该装置可以是网络设备，或者是配置于网络设备中的装置，或者能够和网络设备匹配使用的装置。一种设计中，该装置包括执行第二方面中所描述的方法/操作/步骤/动作一一对应的单元，该单元可以是硬件电路，也可是软件，也可以是硬件电路结合软件实现。示例性地，该装置可以包括处理单元和通信单元，且处理单元和通信单元可以执行上述第二方面任一种设计示例中的相应功能，具体的：通信单元用于接收来自第二网络设备的第一消息，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。处理单元，用于对第一消息进行处理。
关于处理单元和通信单元的具体执行过程可参见上述第二方面的记载。
第六方面，提供一种装置，有益效果可参见第二方面的记载，所述装置包括存储器， 用于实现上述第二方面描述的方法。所述装置还可以包括存储器，用于存储指令和/或数据。所述存储器与所述处理器耦合，所述处理器执行所述存储器中存储的程序指令时，可以实现上述第二方面描述的方法。所述装置还可以包括通信接口，所述通信接口用于该装置和其它设备进行通信。示例性地，通信接口可以是收发器、电路、总线、模块、管脚或其它类型的通信接口。在一种可能的设计中，该装置包括：
存储器，用于存储程序指令；
通信接口，接收来自第二网络设备的第一消息，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
处理器，对第一消息进行处理。
关于处理器和通信接口的具体执行过程可参见上述第二方面的记载。
第七方面，本申请实施例还提供一种计算机可读存储介质，包括指令，当其在计算机上运行时，使得计算机执行第一方面或第二方面任一方面的方法。
第八方面，本申请实施例还提供一种芯片系统，该芯片系统包括处理器，还可以包括存储器，用于实现第一方面或第二方面任一方面的方法。该芯片系统可以由芯片构成，也可以包含芯片和其他分立器件。
第九方面，本申请实施例中还提供一种计算机程序产品，包括指令，当其在计算机上运行时，使得计算机执行第一方面或第二方面任一方面的方法。
第十方面，本申请实施例中还提供一种系统，该系统中包括第三方面或第四方面的装置、和第五方面或第六方面的装置。
附图说明
图1为本申请实施例提供的通信架构的示意图；
图2a至图2d为本申请实施例提供的AI模型的示意图；
图3至图6为本申请实施例提供的通信方法的流程图；
图7和图8为本申请实施例提供的装置的示意图；
图9a为神经元的结构示意图；
图9b为神经网络的层关系示意图。
具体实施方式
图1是本申请的实施例应用的通信系统1000的架构示意图。如图1所示，该通信系统包括无线接入网100和核心网200，可选的，通信系统1000还可以包括互联网300。其中，无线接入网100可以包括至少一个无线接入网设备(如图1中的110a和110b)，还可以包括至少一个终端(如图1中的120a-120j)。终端通过无线的方式与无线接入网设备相连，无线接入网设备通过无线或有线方式与核心网连接。核心网设备与无线接入网设备可以是独立的不同的物理设备，也可以是将核心网设备的功能与无线接入网设备的逻辑功能集成在同一个物理设备上，还可以是一个物理设备上集成了部分核心网设备的功能和部分的无线接入网设备的功能。终端和终端之间以及无线接入网设备和无线接入网设备之间可以通过有线或无线的方式相互连接。图1只是示意图，该通信系统中还可以包括其它网络设备， 如还可以包括无线中继设备和无线回传设备，在图1中未画出。
无线接入网设备可以是基站(base station)、演进型基站(evolved NodeB，eNodeB)、发送接收点(transmission reception point，TRP)、第五代(5th generation，5G)移动通信系统中的下一代基站(next generation NodeB，gNB)、第六代(6th generation，6G)移动通信系统中的下一代基站、未来移动通信系统中的基站或无线保真(wireless fidelity，WiFi)系统中的接入节点等；也可以是完成基站部分功能的模块或单元，例如，可以是集中式单元(central unit，CU)，也可以是分布式单元(distributed unit，DU)。这里的CU完成基站的无线资源控制(radio resource control，RRC)协议和分组数据汇聚层协议(packet data convergence protocol，PDCP)的功能，还可以完成业务数据适配协议(service data adaptation protocol，SDAP)的功能；DU完成基站的无线链路控制(radio link control，RLC)层和介质访问控制(medium access control，MAC)层的功能，还可以完成部分物理(physical，PHY)层或全部物理层的功能，有关上述各个协议层的具体描述，可以参考第三代合作伙伴计划(3rd generation partnership project，3GPP)的相关技术规范。无线接入网设备可以是宏基站(如图1中的110a)，也可以是微基站或室内站(如图1中的110b)，还可以是中继节点或施主节点等。本申请的实施例对无线接入网设备所采用的具体技术和具体设备形态不做限定。为了便于描述，下文以基站作为无线接入网设备的例子进行描述。
终端也可以称为终端设备、用户设备(user equipment，UE)、移动台、移动终端等。终端可以广泛应用于各种场景，例如，设备到设备(device-to-device，D2D)、车物(vehicle to everything，V2X)通信、机器类通信(machine-type communication，MTC)、物联网(internet of things，IOT)、虚拟现实、增强现实、工业控制、自动驾驶、远程医疗、智能电网、智能家具、智能办公、智能穿戴、智能交通、智慧城市等。终端可以是手机、平板电脑、带无线收发功能的电脑、可穿戴设备、车辆、无人机、直升机、飞机、轮船、机器人、机械臂、智能家居设备等。本申请的实施例对终端所采用的具体技术和具体设备形态不做限定。为了便于描述，下文以UE作为终端的例子进行描述。
基站和终端可以是固定位置的，也可以是可移动的。基站和终端可以部署在陆地上，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、气球和人造卫星上。本申请的实施例对基站和终端的应用场景不做限定。
基站和终端的角色可以是相对的，例如，图1中的直升机或无人机120i可以被配置成移动基站，对于那些通过120i接入到无线接入网100的终端120j来说，终端120i是基站；但对于基站110a来说，120i是终端，即110a与120i之间是通过无线空口协议进行通信的。当然，110a与120i之间也可以是通过基站与基站之间的接口协议进行通信的，此时，相对于110a来说，120i也是基站。因此，基站和终端都可以统一称为通信装置，图1中的110a和110b可以称为具有基站功能的通信装置，图1中的120a-120j可以称为具有终端功能的通信装置。
基站和终端之间、基站和基站之间、终端和终端之间可以通过授权频谱进行通信，也可以通过免授权频谱进行通信，也可以同时通过授权频谱和免授权频谱进行通信；可以通过6千兆赫(gigahertz，GHz)以下的频谱进行通信，也可以通过6GHz以上的频谱进行通信，还可以同时使用6GHz以下的频谱和6GHz以上的频谱进行通信。本申请的实施例对无线通信所使用的频谱资源不做限定。
在本申请的实施例中，基站的功能也可以由基站中的模块(如芯片)来执行，也可以 由包含有基站功能的控制子系统来执行。这里的包含有基站功能的控制子系统可以是智能电网、工业控制、智能交通、智慧城市等上述应用场景中的控制中心。终端的功能也可以由终端中的模块(如芯片或调制解调器)来执行，也可以由包含有终端功能的装置来执行。
在本申请中，基站向终端发送下行信号或下行信息，下行信息承载在下行信道上；终端向基站发送上行信号或上行信息，上行信息承载在上行信道上。终端为了与基站进行通信，需要与基站控制的小区建立无线连接。与终端建立了无线连接的小区称为该终端的服务小区。当终端与该服务小区进行通信的时候，还会受到来自邻区的信号的干扰。
在无线通信系统中，UE可以切换服务小区。在以下描述中，可以将UE当前服务小区所属的基站称为源基站，将UE待切换服务小区所属的基站称为目标基站。在一种设计中，源基站或AI设备可以进行人工智能(artificial intelligence，AI)推理，确定AI目标小区，且向AI目标小区对应的基站称为目标基站，发送切换请求。之后，若目标基站同意源基站的切换请求，UE可以切换到AI目标小区。在本申请实施例中考虑到源基站或AI设备的AI推理结果需要消耗源基站的大量计算资源和内存资源，可将上述AI推理结果，在切换请求中发送给AI目标小区对应的目标基站。后续，AI目标小区对应的基站可以利用该AI推理结果进行一系列操作，从而提高AI推理结果的利用率。
由于本申请实施例涉及利用AI技术，预测UE可以切换到的服务小区的过程，为了便于理解，首先对AI技术进行介绍。可以理解的是，该介绍并不作为对本申请实施例的限定。
AI，是一种通过模拟人脑进行复杂计算的技术。随着数据存储和能力的能升，AI得到了越来越多的应用。第三代合作伙伴计划(3 rd generation partnership project，3GPP)的版本17(release17，R17)通过了研究项目(study item，SI)，提出了将AI运用到新无线(new radio，NR)中。如图2a所示为AI在NR中的第一种应用框架的示例图：
数据源(data source)用于存储训练数据和推理数据。模型训练节点(model trainning host)通过对数据源提供的训练数据(training data)进行分析或训练，得到AI模型，且将AI模型部署在模型推理节点(model inference host)中。模型推理节点使用AI模型，基于数据源提供的推理数据进行推理，得到推理结果。该推理结果用于对网络运行给出基于AI的合理预测，或者指导网络做出策略配置或策略调整。相关的策略配置或者策略调整，由执行(actor)实体统一规划，并发送给多个执行对象(例如，网络实体)去执行。同时，应用了相关策略后，网络的具体表现，可以被再次输入到数据源存储起来。
如图2b、图2c或图2d所示为AI在NR中的第二种应用框架的示例图：
独立于基站的第一AI模块接收训练数据。第一AI模块通过对训练数据进行分析或训练，得到AI模型。针对某个参数，可以是第一AI模块利用相应的AI模型和推理数据进行推理，得到该参数，可参见图2b；或者可以是由第一AI模块将该AI模型的信息发送给位于基站中(或描述为位于RAN中)的第二AI模块，由第二AI模块利用相应的AI模型和推理数据进行推理，得到该参数，可参见图2c。或者，第二AI模块用于推理的AI模型也可以是第二AI模块接收训练数据，并通过对该训练数据进行训练得到的，可参见图2d。
需要说明的是，在上述图2a至图2d中的框架中，AI模型可以简称为模型，其可以看做是从输入的测量量(测量信息)到输出的参数之间的映射。输入的测量量可以 是一个或多个测量量，输出的参数可以是一个或多个参数。训练数据可以包括已知的输入测量量，或包括已知的输入测量量和对应的输出参数，用于训练AI模型。训练数据可以是来自基站、CU、CU-CP、CU-UP、DU、射频模块、UE和/或其它实体的数据，和/或是通过AI技术推理出的数据，不予限制。推理数据包括输入测量量，用于利用模型推理出参数。推理数据可以是来自基站、CU、CU-CP、CU-UP、DU、射频模块、UE和/或其它实体的数据。推理出的参数可以看做策略信息，发送给执行对象。推理出的参数可以被发送给基站、CU、CU-CP、CU-UP、DU、射频模块、或UE等，用于进行策略配置或策略调整。用于推理不同参数的AI模型可以是相同的，也可以是不同的，不予限制。
可以理解的是，本申请实施例中，UE和/或基站可以执行本申请实施例中的部分或全部步骤，这些步骤或操作仅是示例，本申请实施例还可以执行其它操作或者各种操作的变形。此外，各个步骤可以按照本申请实施例呈现的不同的顺序来执行，并且有可能并非要执行本申请实施例中的全部操作。
在本申请的各个实施例中，如果没有特殊说明以及逻辑冲突，不同的实施例之间的术语和/或描述具有一致性、且可以相互引用，不同的实施例中的技术特征根据其内在的逻辑关系可以组合形成新的实施例。
如图3所示，本申请实施例提供一种通信方法的流程，至少包括：
步骤300：源基站确定第一推理结果，该第一推理还称为AI推理结果、第一AI推理结果或其它名称等，不作限定。
在一种设计中，源基站中布署有AI模型，该AI模型可以参见图2a、图2c或图2d中的介绍。源基站可以基于该AI模型，进行AI推理，得到第一推理结果。例如，源基站可以将以下至少一项信息作为AI模型的输入，例如，UE的历史轨迹信息、UE历史驻留信息、UE当前移动方向、UE的速度，UE签约的网络信息(例如，电信、联通或移动等)，或UE的业务需求等，输入到AI模型中，该AI模块的输出为第一推理结果。
或者，在另一种设计中，单独布署AI设备，该AI设备可称为远程智能通信、无线智能控制器、AI节点或其它等，不作限定。该AI设备中部署有AI模型，该AI模型可参见图2a或图2b中的介绍。该AI设备可以基于AI模型进行AI推理，确定第一推理结果，且将第一推理结果的指示信息发送给源基站。应当指出，本申请中的AI模型可以由各种深度神经网络构成。神经网络是机器学习的一种具体实现形式。神经网络可以用来执行分类任务、预测任务，也可以用来建立变量间的条件概率分布。常见的神经网络包括深度神经网络(deep neural network,DNN)、生成型神经网络(generative neural network，GNN)等。根据网络的构建方式，DNN可包括前馈神经网络(feedforward neural network,FNN)、卷积神经网络(convolutional neural networks,CNN)和递归神经网络(recurrent neural network,RNN)等。GNN包括生成对抗网络(Generative Adversarial Network，GAN)和变分自编码器(Variational Autoencoder,VAE)。神经网络是以神经元为基础而构造的，下面以DNN为例介绍神经网络的计算和优化机制，可以理解的是，本发明实施例中对神经网络的具体实现方式不做限制。DNN网络中，每个神经元都对其输入值做加权求和运算，将加权求和结果通过一个激 活函数产生输出。如图9a所示，图9a为神经元结构示意图。假设神经元的输入为x＝[x 0,x 1,...x n],与输入对应的权值为w＝[w 0,w 1,...w n]，加权求和的偏置为b，激活函数的形式可以多样化，作为示例，激活函数为：y＝f(x)＝max{0,x}，则一个神经元的执行的输出为：
Figure PCTCN2022110695-appb-000001
其中，w ix i表示w i与x i的乘积。DNN一般具有多层结构，DNN的每一层都可包含多个神经元，输入层将接收到的数值经过神经元处理后，传递给中间的隐藏层。类似的，隐藏层再将计算结果传递给最后的输出层，产生DNN的最后输出。如图9b所示，图9b为神经网络的层关系示意图。DNN一般具有一个或多个隐藏层，隐藏层往往直接影响提取信息和拟合函数的能力。增加DNN的隐藏层数或扩大每一层的神经元的个数都可以提高DNN的函数拟合能力。每个神经元的参数包括权值、偏置和激活函数，DNN中所有神经元的参数构成的集合称为DNN参数(或称为神经网络参数)。神经元的权值和偏置可以通过训练过程得到优化，从而使得DNN具备提取数据特征、表达映射关系的能力。
得益于神经网络在建模和提取信息特征的优势，可以设计基于神经网络的通信方案。为了支持不同的应用场景并获得良好的结果，需要对神经网络的参数进行设置与优化。所述神经网络的参数包括神经网络相关的信息，示例性的，可以包括以下内容的一项或多项：
神经网络的类型，例如深度神经网络，或，生成型神经网络；
神经网络结构相关的信息，例如所述神经网络的类型，神经网络的层数，神经元的数量等；
神经网络中每个神经元的参数，例如权值、偏置和激活函数等。
示例的，所述第一推理结果中包括以下至少一项：UE的未来移动性信息、UE的未来业务信息、或UE的移动轨迹预测信息等。所述UE的移动轨迹预测信息可以指预测的UE在未来时间的地理位置信息。比如，该UE的移动轨迹预测信息可以为预测的在未来的第一时间UE的位置信息A，在未来的第二时间UE的位置信息B等。
在一种设计中，以DNN网络为例，神经元的输入为x＝[x 0,x 1,...x n]，n为整数，其中，x 0,x 1,...,x n可以分别对应以下中的一项或多项：UE的历史轨迹信息、UE历史驻留信息、UE当前移动方向、UE的速度，UE签约的网络信息(例如，电信、联通或移动等)，或UE的业务需求等。经过图9a和图9b中所示的计算之后，得到预测的UE的未来移动性信息和/或未来业务服务信息。
举例来说，UE的未来移动性信息可以包括预测的以下至少一项信息：
-UE未来小区的信息，该UE未来小区的信息可以为UE在未来时间内，可能接入小区的信息。比如，UE未来小区的信息可以包括小区1至小区10等。关于每个小区的信息可以包括小区的小区全局标识(cell global identifier，CGI)、物理小区标识(physical cell identifier，PCI)和频点、小区标识(cell identifier，cell ID)、非公网标识(non-public network identifier,NPN ID)、非陆地网络标识(non-terrestrial network identifier,NTN ID)或者其它小区标识中的至少一种。CGI可以包括公共陆地移动网络(public land mobile network，PLMN ID)和小区cell ID。可选的，小区的信息还可以包括跟踪区码(tracking area code，TAC)和/或小区所属的网络设备的标识信息，比如全局网络设备标识。
-UE在未来小区的驻留时间信息。驻留时间信息可以指UE在某小区接受服务的时间，或者称为将某小区作为服务小区的时间等。针对某个未来小区，该驻留时间具体为在该小区接受服务的开始时间和结束时间，该开始时间可称为开始时间戳时间，结束时间可称为结束时间戳时间等，或者，可以为UE在该小区接受服务的时长，该时长可称为时间范围等。可选的，可以按照UE在不同未来小区的驻留时间，对未来小区的信息进行排序。UE驻留到小区的方式可以为：UE切换到该小区，或者UE小区选择接入到该小区，或者UE重选到该小区，或者UE重建到该小区等。
-UE接入未来小区的方式。可选的，UE接入未来小区的方式可包括正常切换(legacy handover或者ordinary handover)、双激活协议栈切换(dual active protocol stack handover，DAPS HO)、条件切换(conditional handover，CHO)、无随机接入切换(RACH-less HO)或其它接入方式等。
-UE在所述未来小区中是否离开连接态。该信息可具体为预测的UE在未来小区是否会离开连接态。例如，若预测UE在未来小区中会离开连接态，可用第一值(例如1)表示；若预测UE在未来小区中不会离开连接态，可用第二值(例如0)表示等。又例如，若预测UE在未来小区中会离开连接态，且进入去活动态，可以用第一值(例如00)标识；若预测UE在未来小区中会离开连接态，且进入空闲态，可以用第二值(例如01)标识；若预测UE在未来小区中不会离开连接态，可用第三值(例如11)表示等。
-UE的未来移动性的预测准确度。上述UE未来小区的信息、UE在未来小区的驻留时间信息、UE接入未来小区的方式和UE在所述未来小区中是否离开连接态等信息，都可以称为UE的未来移动性信息。针对上述UE的未来移动性信息中的每一项信息，可以均预测一个准确度。或者，可以对上述UE的未来移动性信息中的所有信息，综合预测小区级的一个准确度等。例如，所有未来小区中包括小区1至小区10。针对每个小区的未来移动性信息，可以综合得到该小区的预测准确度。例如，小区1的综合预测准确度为95％，小区2的综合预测准确度为98％等。
举例来说，UE的未来业务信息包括预测的以下至少一项：UE的未来业务类型，UE未来业务的服务质量(quality of service，QoS)需求、所述未来业务的业务量，或未来业务的时间信息等。
在一种AI模型的训练方法中，以对DNN模型的训练为例，假设已有[T 0,...,T x,T x+1,...,T N]时刻实际的UE的历史输入信息X his(0,N)，包括以下中的一项或多项：轨迹信息、驻留信息、移动方向、速度，签约的网络信息(例如，电信、联通或移动等)，或UE的业务需求等，并同时有UE历史输入所对应的实际输出信息Y his，例如实际接入或驻留的小区信息，接入小区的方式等。在训练过程中，可以选择[T 0,...,T x]的历史输入信息X his(0,x+1)作为DNN模型的输入，并得到[T x+1]的推理信息Y inf(x+1)。通过将Y inf(x+1)与Y his(x+1)进行对比，得到损失函数L(x+1)。其中损失函数的计算方法，可以是例如常用的均方差损失、KL散度(Kullback–Leibler divergence)损失等，本方案对此不做限制。以均方差损失为例，其
Figure PCTCN2022110695-appb-000002
Figure PCTCN2022110695-appb-000003
其中p表示
Figure PCTCN2022110695-appb-000004
中参数的数目，即历史输入所对应的实际输出信息的项目的数目，
Figure PCTCN2022110695-appb-000005
表示(x+1)时刻的参数i的值，
Figure PCTCN2022110695-appb-000006
表示 (x+1)时刻的参数i的推理值。通过损失函数的计算结果与预设门限的对比，可以判断当前推理精确度的高低，其中，具体预设门限的设置可以基于系统需求。例如当某时刻的推理结果所对应的损失函数值大于预设门限5时，认为还需要对模型的参数进行调整，以降低损失函数值。例如L(x+1)＝6，则调整模型参数，例如上文所提到的w＝[w 0,w 1,...w n]，加权求和的偏置b，使(x+1)时刻的损失函数降到5以下。当模型调整到使所有时刻的损失函数都低于目标损失函数值，即，前述预设门限，时，则可认为基于[T 0...T x,T x+1...T N]的历史输入、历史输入所对应的实际输出，以及历史输入的推理结果，模型已经训练收敛，是可以使用的模型，即可以应用于预测。
步骤301：源基站根据第一推理结果，确定第一目标小区，该第一目标小区还可以称为AI目标小区。或者描述为，第一目标小区是根据第一推理结果所确定的。该第一目标小区为预测的UE可以接入的服务小区。
在一种可能的实现方式中，源基站可以在第一推理结果中的未来小区信息中，选择一个小区，作为第一目标小区。比如，第一推理结果中的未来小区信息包括小区1至小区10。源基站可以选择小区1作为第一目标小区。关于源基站具体基于何种条件，选择出小区1，不作限定。比如，源基站可以考虑UE的移动性轨迹信息，在未来的时间中，UE将出现在小区1的服务范围内，或者，源基站根据UE在未来小区的驻留时间信息，确定UE在小区1的驻留时间最长或较长等，选择小区1作为第一目标小区等。
步骤302：源基站向第一目标小区对应的目标基站发送第一消息，该第一消息用于指示第一推理结果。该第一消息可以为切换请求消息，或者其它消息，不作限定。
在一种设计中，源基站可以在上述第一消息中指示第一推理结果的全部信息或者部分信息。即源基站可以将第一推理结果的全部信息或部分信息通知目标基站。比如，第一推理结果中的未来小区信息包括小区1至小区10，源基站选择小区1作为第一目标小区。源基站可以把第一推理结果中关于小区2至小区10的信息通知目标基站。
步骤303：源基站接收来自目标基站的第二消息，该第二消息用于指示目标基站是否同意源基站的切换请求，该第二消息可以称为切换响应消息或其它消息。
示例的，若目标基站同意源基站的切换请求，即同意UE切换到第一目标小区，则上述第二消息可以为肯定性应答消息，例如，切换请求确认(handover request acknowledge)消息。或者，若第一目标小区不同意源基站的切换请求，即不同意UE切换到第一目标小区，则上述第二消息可以为否定性应答消息，比如切换准备失败(handover preparation failure)消息，或者切换失败(handover failure)消息等。
在一种设计中，若目标基站同意源基站的切换请求，目标基站可以响应于第一消息，为UE分配第一目标小区的资源，向源基站发送为UE分配的第一目标小区的资源的指示信息。可选的，上述第一目标小区的资源的指示信息可以携带于第二消息中。源基站可以将上述为UE分配的第一目标小区的资源指示给UE。UE可以接入到第一目标小区。示例的，在UE接入到第一目标小区后，第一目标小区对应的目标基站可以采用UE接入目标基站后的实际信息。当该实际信息与第一推理结果中的预测信息两者的差别满足预设条件时，可以向源基站发送反馈信息，或者可基于其它条件向源基站发送反馈信息等，不作限定。具体关于反馈信息发送的触发条件可以参见步骤304中的说明，以使得源基站或AI设备等确定第一推理结果的AI模型进行优化或更新等，使得该AI模型的推理更加准确。
步骤304：目标基站向源基站或AI设备发送反馈信息的指示信息。
在一种设计中，若AI模型布署在源基站中，则目标基站向源基站发送反馈信息的指示信息，源基站基于该反馈信息，更新该AI模型的参数。或者，若单独布署AI设备，目标基站可以向源基站发送反馈信息的指示信息，源基站将反馈信息的全部或部分信息，发送给AI设备。或者，目标基站可以通过目标基站与AI设备的接口，直接将反馈信息发送给AI设备。应当指出，在第一目标设备直接向AI设备发送反馈信息的方案中，在上述步骤302中的第一消息中可以携带有AI设备的相关信息，例如，AI设备的地址信息，或者AI设备的标识信息等。AI设备基于该反馈信息，对AI模型的参数进行优化或调整。所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。例如，可以根据上述反馈信息，更新或优先所述AI模型的输入参数，和/或，对AI模型本身进行优化或更新等，不作限定。
示例的，目标基站可以在满足以下触发条件中的至少一项时，目标基站向源基站或AI设备发送反馈信息：
-目标基站确定切换UE到第二目标小区。例如，由于UE移动等因素，第一目基站认为UE由需要由第一目标小区切换到第二目标小区。关于目标基站确定第二目标小区的方式，可以为：第一目标小区基于AI模型进行AI推理，确定第二推理结果。或者，AI设备基于AI模型进行AI推理，确定第二推理结果，向第一目标小区发送第二推理结果的指示信息。第一目标小区基于第二推理结果，确定第二目标小区等。例如，第一目标小区为小区1，第二目标小区为小区2。当小区1对应的基站确定UE需要切换到小区2时，小区1对应的基站认为满足该项触发条件，小区1对应的基站可以向源基站发送反馈信息。
-UE在第一目标小区的业务信息发生改变。比如，目标基站可以将UE在第一目标小区的业务信息，与第一推理结果中预测的业务信息相比较。当两者的差值超过一定范围时，可以认为满足该项触发条件，可以向源基站发送反馈信息。
-UE在第一目标小区的实际驻留时间与第一推理结果中预测的在第一目标小区的驻留时间之间的差值超过预设条件等。
-目标基站确定的第二目标小区不同于接收的第一推理结果中的预测的小区。或者称为第二目标上区不属于第一推理结果的未来小区信息中的小区。比如，第一推结果中的未来小区信息包括小区1至小区10，目标基站确定的UE下次待切换的小区为小区20，则可以认为满足该项触发条件，可以向源基站发送反馈信息。
-目标基站预测的UE在第二目标小区的接入方式不同于第一推理结果中预测的接入方式。比如，目标基站可以确定第二推理结果，关于确定第二推理结果的方式可参见前述。根据第二推理结果可以确定第二目标小区和在第二目标小区的接入方式。举个简单的示例，第一推理结果中预测接入方式可包括A、B和C等。但目标基站预测的UE接入第二目标小区的方式为F，则可以认为满足该项触发条件，可以向源基站发送反馈信息等。
-UE的实际轨迹偏移第一推理结果中预测的UE移动轨迹。比如，第一推理结 果中预测的UE移动轨迹为在第一时间，预测UE位于位置A；在第二时间，预测UE位于位置B；但在第一时间，UE的实际位于位置C；当位置A与位置C间的距离大于预设条件时，可以认为满足该项触发条件，可以发送反馈信息等。
示例的，目标基站向源基站或AI设备发送的反馈信息可以包括以下至少一项：
-UE在所述第一目标小区的实际驻留时间信息。例如，以第一目标小区为小区1为例，该信息可以为UE在小区1的实际驻留时间信息。同样，该实际驻留时间信息可以为UE在该小区1实际的开始驻留时间和实际的结束驻留时间。或者，该实际驻留时间信息可以为UE在该小区1的实际驻留时长信息等。
-UE在所述第一目标小区中实际是否离开连接态。
-目标基站确定的推理结果，本实施例中也可以称为第二推理结果。可选的，目标基站可以根据源基站发送的第一推理结果，确定第二推理结果。例如，目标基站可以将第一推理结果作为AI模型的输入，输入到AI模型中，该AI模型的输出为第二推理结果。或者，目标基站可以将第一推理结果发送给AI设备，AI设备基于该第一推理结果进行AI推理，确定第二推理结果，且将第二推理结果发送给目标基站等。关于第二推理结果中包括信息的类型，可参见上述第一推理结果的描述。
-第二目标小区信息。可选的，目标基站根据第二推理结果，确定第二目标小区。例如，第二推理结果中包括的未来小区信息为小区2至小区10，目标基站所确定的第二目标小区可以为小区2等。
-UE在第二目标小区的业务类型。
-UE接入第二目标小区的方式。关于UE在第二目标小区的业务类型或接入第二小区的方式可以是目标基站预测的。例如，可以是目标基站基于第一推理结果预测或推理的等。
-至少部分第一推理结果的准确度等。
应当指出，上述图3所示流程中的步骤300、步骤301、步骤303或步骤304都是可选的。在上图3的描述中，通过步骤302，源基站将第一推理结果指示给目标基站。关于上述第一推理结果的作用主要作了两方面的介绍：
第一方面：第一目标设备可以直接利用该第一推理结果，作AI推理，确定第二推理结果；基于第二推理结果，确定第二目标小区。由于源基站或AI设备在推理第一推理结果时消耗了大量了计算资源或存储资源，这样做可以提高第一推理结果的利用率。再者，第一目标设备直接利用第一推理结果进行AI推理，而不用从头开始推理，这样也可以减少第一目标设备的计算资源或存储资源的消耗。
第二方面：第一目标小区可以根据该第一推理结果，确定反馈信息的触发条件。当UE在第一目标小区的实际性能参数，与上述第一推理结果中预测的各种参数的性能差别不同或超过阈值时，可以向源基站或AI设备发送反馈信息。基于该反馈信息，对AI模型的参数进行优化或更新，以提高后续AI推理的准度性，提升系统效率。
应当指出，在本申请实施例中，重点保护上述步骤302中源基站将第一推理结果发送给目标基站。关于目标基站或其它设备如何利用该第一推理结果，本申请实施例中并不作限定。上述对第一推理结果的利用过程，仅为示意性说明。
为了便于理解，首先对UE的双连接技术进行介绍。UE同时与两个基站保持连接，并接收服务，称之为双连接架构。NR系统中支持的双连接架构，又称之为多空口的双连接(multi-radio dual connectivity，MR-DC)，包括：LTE基站和NR基站组成的双连接，或者，NR基站与NR基站组成的双连接，或者，LTE基站与LTE基站组成的双连接等。可以理解的是，LTE基站包括连接4G核心网设备的LTE基站，或者连接5G核心网设备的LTE基站。NR基站包括连接4G核心网设备的NR基站，或者连接5G核心网设备的NR基站。
在双连接架构中，UE可以与两个基站保持连接，分别称为主站(master node，MN)和辅站(secondary node，SN)。其中，主站为UE提供空口资源的小区组，称为主小区组(master cell group，MCG)。该主小区组中包括至少一个小区。例如，主小区组中可以包含主小区(primary cell,PCell)，在配置了载波聚合(carrier aggregation，CA)的情况下，还可以包含至少一个辅小区(secondary cell,SCell)。辅站为UE提供空口资源的小区组，称为辅小区组(secondary cell group，SCG)。该辅小区组中包括至少一个小区。例如，辅小区组中，可以包含主辅小区(primary secondary cell,PSCell)，在配置了CA的情况下，还可以包含至少一个辅小区等。
如图4所示，提供一种通信方法的流程，该流程可以为在双连接架构中，对图3所示流程的一种具体应用，至少包括以下步骤：
步骤400：源主站确定第一推理结果。
其中，第一推理结果可以包括主小区移动性相关的推理结果和/或SN移动性相关的推理结果。关于主小区移动性相关的推理结果与SN移动性相关的推理结果可以利用相同的AI模型推理出，或者可以利用不同的AI模型推理得出等，不作限定。应当指出，上述主小区移动性相关的推理结果和SN移动性相关的推理结果可以是源主站基于AI模型推理得出的，或者是AI设备基于AI模型推理得出的。或者，上述两个推理结果中的任一个是源主站推理得出的，剩余的另一个是由AI设备推理得到的等，不作限定。
其中，主小区移动性相关的推理结果可以包括以下至少一项：UE的未来主小区/主站/主小区组移动性信息、UE在未来主小区/主站/主小区组的未来业务信息、或UE的移动轨迹预测信息等。SN移动性相关的推理结果可以包括以下至少一项：UE的未来主辅小区/辅站/辅小区组移动性信息、UE在未来主辅小区/辅站/辅小区组的未来业务信息、或UE的移动轨迹预测信息等。关于主小区移动性/SN移动性相关的推理结果中包括信息的详细描述可以参考步骤300中第一推理结果的相关描述，此处不再赘述。
步骤401：源主站基于第一推理结果，确定第一目标主小区，该第一目标主小区为预测的，UE能够接入的主小区。
例如，源主站可以根据第一推理结果中的主小区移动性相关的推理结果，确定第一目标主小区。例如，主小区移动性相关的推理结果中的未来主小区中包括主小区1至主小区10。源主站通过判断发现，主小区1可以作为UE的主小区，则主小区1可认为是上述第一目标主小区。
步骤402：源主站向第一目标小区对应的基站发送第一消息，第一目标主小区对应的基站可称为目标主站。该第一消息用于请求将UE的主小区切换到第一目标主小区，该第一消息中可以包括上述第一推理结果的指示信息。
与上述图3所示的流程不同的是，该第一消息中可以包括主小区的移动性推理结果和 SN的移动性推理结果中的至少一项。若目标基站同意第一消息的请求，目标主站在接收到上述主小区的移动性推理结果时，可以为UE配置主小区相关的信息，例如，为UE配置主小区组等。当UE接入第一目标主小区后，第一目标主小区可以根据SN的移动性推理结果，为UE添加、更改或删除辅站等。可选的，在目标主站为UE更改或添加辅站后，还可以根据SN移动性相关的推理结果，为UE配置添加或更新辅站的相关信息，例如，对添加或更新的辅站配置辅小区组的相关信息等。或者，
在UE接入到目标主站后，目标主站可以根据主小区移动性相关的推理结果，确定主小区移动性相关的推理结果的反馈信息的触发条件。比如，当主小区相关的推理结果中预测信息，与UE接入第一主站后的实际信息之间的差值超出预设值时，目标主站可以向源基站或AI设备发送反馈信息等，以对AI模型相关的参数进行优化或更新等。类似的，目标主站也可以根据SN移动性相关的推理结果，确定SN移动性相关推理结果的反馈信息的触发条件等。
可选的，当所述第一推理结果包括所述主小区移动性相关的推理结果时，所述第一消息中可以包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
步骤403：源主站接收来自目标主站的第二消息，该第二消息可以上述第一消息的响应消息。
其中，第二消息可以为肯定应答的消息，表示目标主站同意源主站的请求，UE的主小区可以切换到第一目标主小区。在一种说明中，UE的主小区切换到第一目标主小区，可以认为UE的主站由源主站切换到目标主站。目标主站在同意UE的请求时，可以根据主小区移动性的推理结果，确定主站的相关配置。比如，在将目标主站切换为主站后，主小区组的配置等。上述主站的相关配置可以携带于上述步骤403中的第二消息中。在UE切换到目标主站时，目标主站可以根据SN移动性相关的推理结果，为UE配置辅站的相关信息等。或者，第二消息可以为否定应答的消息，表示目标主站不同意源主站的请求，UE的主小区不可以切换到第一目标主小区。
步骤404：目标主站向源主站或AI设备发送反馈信息，该反馈信息用于对用于确定第一推理结果的AI模型的参数进行更新或优化。
其中，反馈信息可以包括对主小区移动性相关的推理结果的反馈信息，和/或，SN移动性相关的推理结果的反馈信息。反馈信息的内容可以参考步骤304中的相关描述，此处不再赘述。与上述不同的是，若主小区移动性相关的推理结果与SN移动性相关的推理结果由不同的AI模式推理出的，则上述主小区移动性相关的推理结果用于对其对应的AI模型的参数进行优化或更新。SN移动性相关的推理结果用于对其对应的AI模型的参数进行优化或更新等。关于目标主站发送反馈信息的触发条件可以参考步骤304中的相关描述，此处不再赘述。
在该实施例中，目标主站向源主站或AI设备发送上述第一推理结果的反馈信息，源主站或AI设备可以基于上述反馈信息优化或更新第一推理结果的AI模型的相关参数，从而使得推理出的UE待切换的主小区或者SN移动性的配置更加准确。进一步，目标主站或源主站可以根据第一推理结果是否为UE配置SN移动性，为UE配置更合理的多连接配置，提升系统消息。
如图5所示，提供一种通信方法的流程，该流程主要用于由主站触发的对UE辅站的添加或变更等，至少包括以下步骤：
步骤500：主站确定第一推理结果。
其中，第一推理结果可以包括SN移动性相关的推理结果。SN移动性相关的推理结果的具体描述可以参考步骤400，此处不再赘述。
步骤501：主站根据第一推理结果，确定第一目标辅站。
例如，主站可以根据第一推理结果，确定是否需要添加或变更UE的辅站。例如，主站根据SN移动性相关的推理结果中的UE未来轨迹信息，确定在未来的时间内，当前辅站不能为UE提供服务，或者当前辅站在未来时间的服务质量不能满足要求等，主站可以根据确定第一目标辅站。例如，主站可以根据SN移动性相关的推理结果中的未来主辅小区/辅站/主辅小区组等信息，确定第一目标辅站。例如，主站发现SN移动性相关的推理结果中的未来小区1至3，在未来时间内，可以作为UE的辅小区组，上述小区1至小区3对应的基站可以称为上述第一目标辅站。
步骤502：主站向第一目标辅站发送第一消息，第一消息可以是请求将第一目标辅站添加或变更为UE的辅站的请求消息，该第一消息中可以包括第一推理结果的指示信息。
在一种设计中，在第一目标辅站同意变更或添加为UE辅站的情况下，第一目标辅站可以根据第一推理结果，确定为SN配置的辅站相关的信息，例如，辅小区组信息、主辅小区信息或辅小区信息等。或者，在UE接入第一目标辅站之后，第一目标辅站可以根据第一推理结果，进行AI推理，确定UE可以添加或变更的未来目标辅站等。
步骤503：主站接收来自第一目标辅站的第二消息。
其中，第二消息可以为肯定应答消息，表示第一目标辅站同意将添加或变更为UE的辅站。或者，第二消息可以为否定应答消息，表示第一目标辅站不同意添加或变更为UE的辅站。可选的，若第一目标辅站同意变更或添加为UE的辅站，则第二消息中可以包括第一目标辅站为UE配置的辅站相关信息，例如，辅小区组信息、主辅小区信息或辅小区信息等。应当指出，第一目标辅站可以通过第二消息将为UE配置的辅站相关信息通知主站，由主站再转发给UE。或者，在UE与目标主站建立连接的情况下，第一目标辅站可以将上述配置的辅站相关的信息直接通知UE等，不作限定。
步骤504：第一目标辅站向主站或AI设备发送反馈信息。
应当指出，第一目标辅站可以根据上述反馈信息，对推理SN移动性的AI模型相关的参数进行优化或更新等，当然这是建立在由主站根据AI模型，推理得到SN移动性的推理结果的前提下。或者，若由AI设备根据AI模型，推理得到的SN移动性推理结果，则第一目标辅站可将上述反馈信息直接发送给AI设备，或者可以发送给主站，由主站转发给AI设备等。
应当指出，上述步骤500、501、503或504可以是可选的。
如图6所示，提供一种通信方法的流程，与上述图5所示流程的区别在于，由源辅站根据第一推理结果，得到第一目标辅站。之后，源辅站将第一推理结果和第一目标辅站通知主站，由主站再触发第一目标辅站的添加或更改流程，至少包括以下步骤：
步骤600：源辅站确定第一推理结果。
其中，第一推理结果可以包括SN移动性相关的推理结果。SN移动性相关的推理结果的具体描述可以参考步骤400，此处不再赘述。需要说明的是，第一推理结果可以是由源辅站根据AI模型推理得到的，或者，是AI设备根据AI模型推理得到的，然后通知源辅站的。
步骤601：源辅站基于第一推理结果，确定第一目标辅站。
步骤602：源辅站向主站发送第一推理结果和第一目标辅站的指示信息。或者，源辅站可以只向主站发送第一推理结果的指示信息。由主站根据第一推理结果，确定第一目标辅站。第一目标辅站的指示信息可以为第一目标辅站的标识信息，比如全局节点标识等。
步骤603：主站向第一目标辅站发送第一消息，该第一消息中包括第一推理结果的指示信息，该第一消息用于请求将第一目标辅站添加或变更为UE的辅站。
步骤604：主站接收来自第一目标辅站的第二消息，该第二消息可以第一消息的应答消息。
步骤605：第一目标辅站向主站发送反馈消息的指示信息。
步骤606：主站向源辅站发送反馈消息的指示信息。
应当指出，在上述步骤605和步骤606中，第一目标辅站可以将反馈信息的指示信息发送给主站，由主站转发给源辅站。源辅站根据该反馈信息，对推理得到SN移动性相关的推理结果的AI模型进行优化或更新。或者，第一目标辅站可以将反馈信息的指示信息直接发送给源辅站，此时可能需要在上述步骤603中的第一消息中携带源辅站的地址信息，或者标识信息等。或者，若由AI设备进行推理得到的上述SN移动性的推理结果，则第一目标辅站可以直接将反馈信息的指示信息发送给AI设备，或者可以经由主站和源辅站等转发给AI设备等，不作限定。
应当指出，在上述图6中，除步骤603外，其它步骤都是可选的。在步骤601中，若源基站根据第一推理结果，确定释放源基站。则源辅站还可以向主站发送释放指示信息，主站向转发UE源辅站释放指示信息。可选的，主站在释放源辅站后，可以向源辅站发送反馈消息的指示信息等。
在上述图5或图6的实施例中，引入SN添加或SN变更场景下的主辅小区相关，或者辅站相关的AI模型结果，以及对应的性能反馈信息的交互，使得对应的节点可以进一步优化AI模块，提升AI性能。
针对上述图3至图6的流程，需要说明的是：
1、上文侧重描述了图3至图6的区别之处，除区别之外的其它内容，可相互参见。
2、图3至图6所描述的各个流程中的步骤并非全部需要执行的步骤，可以根据实际需要各个流程的基础上增添或删除部分步骤。比如上述图3的流程中除步骤300、301、303和304都可以选择性执行。
3、在上述图3至图6的描述中，以某个硬件设备作为一个整体为例进行描述的，并没有描述该硬件设备内部各模块的动作。为了支持该硬件作为一个整体，实现上述实施例描述的相关功能，该硬件设备内部模块间的操作以及各模块的操作，也在本申请实施例的保护范围内。
例如，在一种设计中，随着开放无线接入网络(open，radio access network，RAN， O-RAN)的提出，接入网设备的功能可能被多个通用标准的模块实现。例如，基站的功能可能由CU模块或DU模块实现。举例来说，在上述图3的流程中，从整体上描述，源基站的动作可以为：源基站确定第一推理结果，根据第一推理结果，确定第一目标小区，向目标基站发送第一消息，该第一消息中包括第一推理结果的指示信息。
若源基站包括CU模块和DU模块，则整个图3所示流程的处理过程可包括：CU确定第一推理结果，根据第一推理结果，确定第一目标小区；将第一推理结果发送给DU；DU向目标基站发送第一消息，该第一消息中包括第一推理结果的指示信息。
5、在上述图3至图6的描述中，采用在“一个消息中携带某个指示信息”的描述。例如，在第一消息中携带第一推理结果的指示信息等。关于上述描述，作如下说明：该消息可以直接指示对应的信息，例如，在该消息中直接携带该信息。或者，该消息可以间接指示对应的信息。举例来说，消息A中包括信息X的指示信息，该数据A可以直接指示信息X，例如，该数据A中携带信息X。或者，该数据A可以间接指示信息X。例如，该数据A中可以携带信息X的其它信息等。
以上结合图3至图6详细说明了本申请实施例的方法，以下结合图7和图8详细说明本申请实施例提供的装置。应理解，装置实施例的描述与方法实施例的描述相互对应，因此，未详细描述的内容可参见上文方法实施例中的描述。
图7示出了本申请实施例所涉及的装置的可能的框图。如图7所示，装置700可以包括：通信单元701用于支持装置700与其他设备的通信。可选的，通信单元701也称为收发单元，可以包括接收单元和/或发送单元，分别用于执行接收和发送操作。处理单元702用于支持装置进行处理。可选的，装置700还可以包括存储单元703，用于存储装置700的程序代码和/或数据。
在第一个实施例中，上述装置700可以为网络设备或网络设备中的模块、芯片或电路。通信单元701用于执行上文图3所示流程中源基站的收发操作；处理单元702用于执行上文图3所示流程中源基站的处理操作。
例如，处理单元702，用于生成第一消息；通信单元701，用于向第一目标小区对应的第一网络设备发送第一消息，所述第一目标小区为预测的、终端设备能够接入的服务小区，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
在一种可能的设计中，所述第一目标小区是根据所述第一推理结果确定的。
在一种可能的设计中，所述终端设备的未来移动信息包括预测的以下至少一项：所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
在一种可能的设计中，所述终端设备的未来业务信息包括预测的以下至少一项：所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
在一种可能的设计中，通信单元701，还用于：接收来自所述第一网络设备的反馈信息，所述反馈信息中包括以下至少一项的指示信息；所述终端设备在所述第一目标小区的 实际驻留时间信息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
在一种可能的设计中，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
在一种可能的设计中，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
在一种可能的设计中，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
在第二个实施例中，上述装置700可以为网络设备或网络设备中的模块、芯片或电路。通信单元701用于执行上文图3所示流程中目标基站的收发操作；处理单元702用于执行上文图3所示流程中目标基站的处理操作。
例如，通信单元701，用于接收来自第二网络设备的第一消息，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。处理单元702，用于对第一消息进行处理。
在一种可能的设计中，所述第一消息用于请求第一网络设备为所述终端设备分配第一目标小区对应的资源，所述第一目标小区为预测的、终端设备能够接入的服务小区。
在一种可能的设计中，处理单元702，还用于响应于所述第一消息，为所述终端设备分配所述第一目标小区的资源；通信单元701，还用于向所述第二网络设备发送为所述终端设备分配的第一目标小区的资源的指示信息。
在一种可能的设计中，所述终端设备的未来移动信息包括预测的以下至少一项：所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
在一种可能的设计中，所述终端设备的未来业务信息包括预测的以下至少一项：所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
在一种可能的设计中，通信单元701，还用于：向所述第二网络设备发送反馈信息，所述反馈信息中包括以下至少一项的指示信息：所述终端设备在所述第一目标小区的实际驻留时间信息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
在一种可能的设计中，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
在一种可能的设计中，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
在一种可能的设计中，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
应理解以上装置中单元的划分仅仅是一种逻辑功能的划分，实际实现时可以全部或部 分集成到一个物理实体上，也可以物理上分开。且装置中的单元可以全部以软件通过处理元件调用的形式实现；也可以全部以硬件的形式实现；还可以部分单元以软件通过处理元件调用的形式实现，部分单元以硬件的形式实现。例如，各个单元可以为单独设立的处理元件，也可以集成在装置的某一个芯片中实现，此外，也可以以程序的形式存储于存储器中，由装置的某一个处理元件调用并执行该单元的功能。此外这些单元全部或部分可以集成在一起，也可以独立实现。这里所述的处理元件又可以成为处理器，可以是一种具有信号的处理能力的集成电路。在实现过程中，上述方法的各操作或以上各个单元可以通过处理器元件中的硬件的集成逻辑电路实现或者以软件通过处理元件调用的形式实现。
在一个例子中，以上任一装置中的单元可以是被配置成实施以上方法的一个或多个集成电路，例如：一个或多个特定集成电路(application specific integrated circuit，ASIC)，或，一个或多个微处理器(digital singnal processor，DSP)，或，一个或者多个现场可编程门阵列(field programmable gate array，FPGA)，或这些集成电路形式中至少两种的组合。再如，当装置中的单元可以通过处理元件调度程序的形式实现时，该处理元件可以是处理器，比如通用中央处理器(central processing unit，CPU)，或其它可以调用程序的处理器。再如，这些单元可以集成在一起，以片上系统(system-on-a-chip，SOC)的形式实现。
以上用于接收的单元是一种该装置的接口电路，用于从其它装置接收信号。例如，当该装置以芯片的方式实现时，该接收单元是该芯片用于从其它芯片或装置接收信号的接口电路。以上用于发送的单元是一种该装置的接口电路，用于向其它装置发送信号。例如，当该装置以芯片的方式实现时，该发送单元是该芯片用于向其它芯片或装置发送信号的接口电路。
参见图8，为本申请实施例提供的网络设备的结构示意图，该网络设备可以为接入网设备(如源基站或目标基站等)。接入网设备800可包括一个或多个DU801和一个或多个CU 802。所述DU 801可以包括至少一个天线8011，至少一个射频单元8012，至少一个处理器8013和至少一个存储器8014。所述DU801部分主要用于射频信号的收发以及射频信号与基带信号的转换，以及部分基带处理。CU802可以包括至少一个处理器8022和至少一个存储器8021。
所述CU802部分主要用于进行基带处理，对接入网设备进行控制等。所述DU801与CU802可以是物理上设置在一起，也可以物理上分离设置的，即分布式基站。所述CU802为接入网设备的控制中心，也可以称为处理单元，主要用于完成基带处理功能。例如所述CU802可以用于控制接入网设备执行上述方法实施例中关于接入网设备的操作流程。
此外，可选的，接入网设备800可以包括一个或多个射频单元，一个或多个DU和一个或多个CU。其中，DU可以包括至少一个处理器8013和至少一个存储器8014，射频单元可以包括至少一个天线8011和至少一个射频单元8012，CU可以包括至少一个处理器8022和至少一个存储器8021。
在一个实例中，所述CU802可以由一个或多个单板构成，多个单板可以共同支持单一接入指示的无线接入网(如5G网)，也可以分别支持不同接入制式的无线接入网(如LTE网，5G网或其他网)。所述存储器8021和处理器8022可以服务于一个或多个单板。也就是说，可以每个单板上单独设置存储器和处理器。也可以是多个单板共用相同的存储器和处理器。此外每个单板上还可以设置有必要的电路。所述DU801可以由一个或多个单板构成，多个单板可以共同支持单一接入指示的无线接入网(如5G网)，也可以分别支持不同 接入制式的无线接入网(如LTE网，5G网或其他网)。所述存储器8014和处理器8013可以服务于一个或多个单板。也就是说，可以每个单板上单独设置存储器和处理器。也可以是多个单板共用相同的存储器和处理器。此外每个单板上还可以设置有必要的电路。
图8所示的接入网设备能够实现上述方法实施例中涉及源基站和目标基站的各个过程。图8所示的接入网设备中的各个模块的操作和/或功能，分别为了实现上述方法实施例中图3至图6中的相应流程。具体可参见上述方法实施例中的描述，为避免重复，此处适当省略详述描述。
本申请实施例中的术语“系统”和“网络”可被互换使用。“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A、同时存在A和B、单独存在B的情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如“A，B或C中的至少一个”包括A，B，C，AB，AC，BC或ABC。以及，除非有特别说明，本申请实施例提及“第一”、“第二”等序数词是用于对多个对象进行区分，不用于限定多个对象的顺序、时序、优先级或者重要程度等。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (21)

  1. 一种通信方法，其特征在于，包括：
    向第一目标小区对应的第一网络设备发送第一消息，所述第一目标小区为预测的、终端设备能够接入的服务小区，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：
    所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
  2. 如权利要求1所述的方法，其特征在于，所述第一目标小区是根据所述第一推理结果确定的。
  3. 如权利要求1或2所述的方法，其特征在于，所述终端设备的未来移动信息包括预测的以下至少一项：
    所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
  4. 如权利要求1至3中任一项所述的方法，其特征在于，所述终端设备的未来业务信息包括预测的以下至少一项：
    所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
  5. 如权利要求1至4中任一项所述的方法，其特征在于，还包括：
    接收来自所述第一网络设备的反馈信息，所述反馈信息中包括以下至少一项的指示信息：
    所述终端设备在所述第一目标小区的实际驻留时间信息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
  6. 如权利要求5所述的方法，其特征在于，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
  7. 如权利要求1至6中任一项所述的方法，其特征在于，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
  8. 如权利要求7所述的方法，其特征在于，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
  9. 一种通信方法，其特征在于，包括：
    接收来自第二网络设备的第一消息，所述第一消息用于指示第一推理结果，所述第一推理结果中包括预测的以下至少一项：
    所述终端设备的未来移动信息、所述终端设备的未来业务信息、或所述终端设备的未来移动轨迹信息。
  10. 如权利要求9所述的方法，其特征在于，所述第一消息用于请求第一网络设备为所述终端设备分配第一目标小区对应的资源，所述第一目标小区为预测的、终端设备能够接入的服务小区。
  11. 如权利要求10所述的方法，其特征在于，还包括：
    响应于所述第一消息，为所述终端设备分配所述第一目标小区的资源；
    向所述第二网络设备发送为所述终端设备分配的第一目标小区的资源的指示信息。
  12. 如权利要求9至11中任一项所述的方法，其特征在于，所述终端设备的未来移动信息包括预测的以下至少一项：
    所述终端设备未来小区的信息、所述终端设备在所述未来小区的驻留时间信息、所述终端设备接入所述未来小区的方式、所述终端设备在所述未来小区中是否离开连接态、或所述终端设备的未来移动信息的预测准确度。
  13. 如权利要求9至12中任一项所述的方法，其特征在于，所述终端设备的未来业务信息包括预测的以下至少一项：
    所述终端设备的未来业务类型、所述未来业务的服务质量QoS需求、所述未来业务的业务量，或所述未来业务的时间信息。
  14. 如权利要求9至13中任一项所述的方法，其特征在于，还包括：
    向所述第二网络设备发送反馈信息，所述反馈信息中包括以下至少一项的指示信息：
    所述终端设备在所述第一目标小区的实际驻留时间信息、所述终端设备在所述第一目标小区中实际是否离开连接态、第二推理结果、或第二目标小区。
  15. 如权利要求14所述的方法，其特征在于，所述反馈信息用于对确定所述第一推理结果的模型的参数进行优化或更新。
  16. 如权利要求9至15中任一项所述的方法，其特征在于，所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果，和/或，所述终端设备的辅站移动性相关的推理结果。
  17. 如权利要求16所述的方法，其特征在于，当所述第一推理结果包括所述终端设备的主小区移动性相关的推理结果时，所述第一消息中包括所述终端设备的源辅站、源辅小区组、源主辅小区或源辅小区中的至少一项是否需要变更的指示信息。
  18. 一种通信装置，其特征在于，包括用于实现权利要求1至17中任一项所述的方法的单元。
  19. 一种装置，其特征在于，包括处理器和存储器，所述处理器和存储器耦合，所述处理器用于实现权利要求1至17中任一项所述的方法。
  20. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有指令，当所述指令在计算机上运行时，使得计算机执行权利要求1至17中任一项所述的方法。
  21. 一种计算机程序产品，其特征在于，包括指令，当所述指令在计算机上运行时，使得计算机执行权利要求1至17中任一项所述的方法。
PCT/CN2022/110695 2021-08-06 2022-08-05 一种通信方法及装置 WO2023011655A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP22852366.8A EP4369787A1 (en) 2021-08-06 2022-08-05 Communication method and apparatus
US18/432,420 US20240179603A1 (en) 2021-08-06 2024-02-05 Communication method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110900384.2A CN115707047A (zh) 2021-08-06 2021-08-06 一种通信方法及装置
CN202110900384.2 2021-08-06

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/432,420 Continuation US20240179603A1 (en) 2021-08-06 2024-02-05 Communication method and apparatus

Publications (1)

Publication Number Publication Date
WO2023011655A1 true WO2023011655A1 (zh) 2023-02-09

Family

ID=85154849

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/110695 WO2023011655A1 (zh) 2021-08-06 2022-08-05 一种通信方法及装置

Country Status (4)

Country Link
US (1) US20240179603A1 (zh)
EP (1) EP4369787A1 (zh)
CN (1) CN115707047A (zh)
WO (1) WO2023011655A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040203831A1 (en) * 2002-04-11 2004-10-14 Khan Moinul H. Reduction of QoS impairment during the hand-off process
CN101365242A (zh) * 2008-08-29 2009-02-11 同济大学 基于移动预测的群体切换方法及系统
CN104902529A (zh) * 2015-04-27 2015-09-09 努比亚技术有限公司 一种网络切换方法、装置及基站
CN108810965A (zh) * 2017-05-05 2018-11-13 华为技术有限公司 一种随机接入资源分配方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040203831A1 (en) * 2002-04-11 2004-10-14 Khan Moinul H. Reduction of QoS impairment during the hand-off process
CN101365242A (zh) * 2008-08-29 2009-02-11 同济大学 基于移动预测的群体切换方法及系统
CN104902529A (zh) * 2015-04-27 2015-09-09 努比亚技术有限公司 一种网络切换方法、装置及基站
CN108810965A (zh) * 2017-05-05 2018-11-13 华为技术有限公司 一种随机接入资源分配方法及装置

Also Published As

Publication number Publication date
EP4369787A1 (en) 2024-05-15
CN115707047A (zh) 2023-02-17
US20240179603A1 (en) 2024-05-30

Similar Documents

Publication Publication Date Title
WO2021243619A1 (zh) 信息传输方法、装置、通信设备及存储介质
WO2022022334A1 (zh) 基于人工智能的通信方法和通信装置
US20230209390A1 (en) Intelligent Radio Access Network
WO2020156127A1 (zh) 无线通信的方法和通信设备
US20230224982A1 (en) Computing data transmission method and apparatus
WO2023143267A1 (zh) 一种模型配置方法及装置
JP2023520274A (ja) 無線通信方法、端末機器及びネットワーク機器
WO2023116772A1 (zh) 一种卫星切换方法、装置、存储介质和芯片系统
US20230189054A1 (en) Relay communication method, and communication apparatus
Li et al. A reinforcement learning based user association algorithm for UAV networks
EP4319321A1 (en) Energy-saving method and apparatus
WO2022134854A1 (zh) 信息共享方法及通信装置
WO2023130897A1 (zh) 一种通信方法、装置及系统
WO2023011655A1 (zh) 一种通信方法及装置
Zhang et al. Three-sided matching game based joint bandwidth and caching resource allocation for UAVs
WO2024093739A1 (zh) 一种通信方法及装置
WO2024067248A1 (zh) 一种获取训练数据集的方法和装置
US20240056926A1 (en) Communication method and apparatus for obtaining load information
WO2023125599A1 (zh) 一种通信方法及通信装置
WO2022126347A1 (zh) 通信方法、装置及系统
WO2024067283A1 (zh) 一种测量报告发送方法、通信装置与通信系统
WO2023125598A1 (zh) 一种通信方法及通信装置
WO2024082274A1 (zh) Ai任务指示的方法、通信装置和系统
WO2023272670A1 (zh) 一种网络连接的方法、装置及系统
Wang et al. Airborne computing platform based joint offload and deployment optimization algorithm

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: 22852366

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2022852366

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2022852366

Country of ref document: EP

Effective date: 20240207

NENP Non-entry into the national phase

Ref country code: DE