WO2024027683A1 - Procédé et appareil de mise en correspondance de modèles, dispositif de communication, et support de stockage lisible - Google Patents

Procédé et appareil de mise en correspondance de modèles, dispositif de communication, et support de stockage lisible Download PDF

Info

Publication number
WO2024027683A1
WO2024027683A1 PCT/CN2023/110481 CN2023110481W WO2024027683A1 WO 2024027683 A1 WO2024027683 A1 WO 2024027683A1 CN 2023110481 W CN2023110481 W CN 2023110481W WO 2024027683 A1 WO2024027683 A1 WO 2024027683A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
data set
channel
information
channel information
Prior art date
Application number
PCT/CN2023/110481
Other languages
English (en)
Chinese (zh)
Inventor
任千尧
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2024027683A1 publication Critical patent/WO2024027683A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G06N3/0455—Auto-encoder networks; Encoder-decoder networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/0985—Hyperparameter optimisation; Meta-learning; Learning-to-learn
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel

Definitions

  • This application belongs to the field of communication technology, and specifically relates to a model matching method, device, communication equipment and readable storage medium.
  • model-based channel information feedback includes a coding part on the terminal side and a decoding part on the base station side.
  • the coding part and the decoding part are jointly trained and match each other.
  • the base station and the terminal need to independently train their respective required models. In this case, the coding model on the terminal side and the decoding model on the base station side will not match, and the output of the coding model on the terminal side cannot be restored through the decoding model on the base station side, affecting channel information feedback.
  • Embodiments of the present application provide a model matching method, device, communication equipment and readable storage medium, which can solve the problem in related technologies that the models on the terminal side and the network side cannot be matched.
  • the first aspect provides a model matching method, including:
  • the first device acquires first channel characteristic information, which is obtained by the second device using the first model to process the first channel information
  • the first device matches the second model of the first device with the first model according to the first channel characteristic information and the second channel information corresponding to the first channel characteristic information;
  • the first channel information is the same as the second channel information; the first model and the second model are models trained in different devices; when the first device is a network side device, the When the second device is a terminal, the second model is used to process the channel characteristic information obtained by the first device; or when the first device is a terminal and the second device is a network side device , the second model is used to process channel information.
  • a model matching device including:
  • An acquisition module configured to acquire first channel characteristic information, which is obtained by processing the first channel information by the second device using the first model
  • a matching module configured to match the second model of the first device with the first model according to the first channel characteristic information and the second channel information corresponding to the first channel characteristic information
  • the first channel information and the second channel information are the same; the first model and the second model are Models trained in different devices; when the first device is a network-side device and the second device is a terminal, the second model is used to process the channel characteristic information obtained by the first device; Or, when the first device is a terminal and the second device is a network-side device, the second model is used to process channel information.
  • a communication device in a third aspect, includes a processor and a memory.
  • the memory stores a program or instructions that can be run on the processor.
  • the program or instructions are implemented when executed by the processor. The steps of the method as described in the first aspect.
  • a communication device is provided.
  • the communication device is a first device and includes a processor and a communication interface.
  • the processor is used to obtain first channel characteristic information.
  • the first channel characteristic information is used by the second device.
  • the first model is obtained by processing the first channel information; according to the first channel characteristic information and the second channel information corresponding to the first channel characteristic information, the second model of the first device and the third A model is used for matching; the first channel information is the same as the second channel information; the first model and the second model are models trained in different devices; when the first device is a network side device, when the second device is a terminal, the second model is used to process the channel characteristic information obtained by the first device; or, when the first device is a terminal, the second device is a network When a side device is used, the second model is used to process channel information.
  • a communication system including: a first device and a second device.
  • the first device can be used to perform the steps of the model matching method as described in the first aspect.
  • the second device can utilize the second device.
  • a model processes the first channel information, obtains first channel characteristic information, and sends the first channel characteristic information to the first device.
  • a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented.
  • a chip in a seventh aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. A step of.
  • a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the method described in the first aspect Method steps.
  • the first channel characteristic information is obtained by the second device using the first model to process the first channel information
  • the first channel characteristic information is obtained according to the first channel characteristic information and the first channel characteristic information.
  • the second channel information corresponding to the channel characteristic information matches the second model of the first device with the first model, so that the first model and the second model that are independently trained can be matched, so that the channel processed by the terminal side model Characteristic information can be recovered through the network side model to obtain corresponding channel information, thereby ensuring channel information feedback.
  • Figure 1 is a block diagram of a wireless communication system applicable to the embodiment of the present application.
  • Figure 2 is a schematic diagram of a neural network in an embodiment of the present application.
  • FIG. 3 is a schematic diagram of neurons in the embodiment of the present application.
  • Figure 4 is a flow chart of a model matching method provided by an embodiment of the present application.
  • Figure 5 is a schematic structural diagram of a model matching device provided by an embodiment of the present application.
  • Figure 6 is a schematic structural diagram of a communication device provided by an embodiment of the present application.
  • Figure 7 is a schematic structural diagram of a terminal provided by an embodiment of the present application.
  • Figure 8 is a schematic structural diagram of a network side device provided by an embodiment of the present application.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • NR New Radio
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • PDA Personal Digital Assistant
  • UMPC ultra-mobile personal computer
  • UMPC mobile Internet device
  • MID mobile Internet Device
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • WUE Vehicle User Equipment
  • PUE Pedestrian User Equipment
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (personal computer, PC), teller machine or self-service machine and other terminal-side equipment.
  • Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart Necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a wireless access network unit.
  • Access network equipment may include a base station, a Wireless Local Area Network (WLAN) access point or a WiFi node, etc.
  • the base station may be called a Node B, an Evolved Node B (eNB), an access point, or a base station.
  • BTS Base Transceiver Station
  • BSS Basic Service Set
  • ESS Extended Service Set
  • TRP Transmitting Receiving Point
  • the base station is not limited to specific technical terms. It should be noted that in the embodiment of this application, only The base station in the NR system is taken as an example for introduction, and the specific type of base station is not limited.
  • Core network equipment may include but is not limited to at least one of the following: core network nodes, core network functions, mobility management entities (Mobility Management Entity, MME), access mobility management functions (Access and Mobility Management Function, AMF), session management functions (Session Management Function, SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Service Discovery function (Edge Application Server Discovery Function, EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), centralized network configuration ( Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF, or L-NEF), Binding Support Function (Binding Support Function, BSF), application function (Application Function, AF), etc.
  • MME mobility management entities
  • AMF Access and Mobility Management Function
  • SMF Session Management Function
  • UPF User Plane Function
  • PCF Policy Control Function
  • the model in the embodiment of this application may be an artificial intelligence (Artificial Intelligence, AI) model.
  • AI models have a variety of algorithm implementations, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. This application takes neural network as an example for explanation, but does not limit the specific type of AI module.
  • the schematic diagram of a neural network can be shown in Figure 2, in which X 1 , The results will continue to be passed to the next layer.
  • the input layer, hidden layer and output layer composed of these many neurons is a neural network.
  • the number of hidden layers and the number of neurons in each layer is the "network structure" of the neural network.
  • a neural network is composed of neurons, and the schematic diagram of the neurons can be shown in Figure 3, where a 1 , a k ...a K (i.e., X1, X2... shown in Figure 2) are inputs, and w is the weight. (can also be called: multiplicative coefficient), b is the bias (can also be called: additive coefficient), ⁇ () is the activation function, z is the output value, and the corresponding operation process can be expressed as: Common activation functions include but are not limited to Sigmoid function, hyperbolic tanh function, rectified linear unit (Rectified Linear Unit, ReLU), etc.
  • the parameter information of each neuron and the algorithm used are combined to form the "parameter information" of the entire neural network, which is also an important part of the AI model file.
  • the parameters of neural networks can be optimized through optimization algorithms.
  • An optimization algorithm is a type of algorithm that can minimize or maximize an objective function (sometimes also called a loss function).
  • the objective function is often a mathematical combination of model parameters and data. For example, given data x and its corresponding label Y, and constructing a neural network model f(.), with the model, the predicted output f(x) can be obtained based on the input data x, and the predicted value and the true value can be calculated The difference between the values (f(x)-Y), this is the loss function.
  • the purpose at this time is to find appropriate model parameters (such as weights/biases) to minimize the value of the above loss function. The smaller the loss value, the closer the model is to the real situation.
  • the optimization algorithm includes error back propagation (error Back Propagation, BP) algorithm.
  • error Back Propagation BP
  • the basic idea of BP algorithm is that the learning process consists of two processes: forward propagation of signals and back propagation of errors.
  • the back propagation of error is to back propagate the output error in some form to the input layer layer by layer through the hidden layer, and allocate the error to all units in each layer, thereby obtaining the error signal of each layer unit, and this error signal is used as a correction The basis for the weight of each unit.
  • This process of adjusting the weights of each layer in forward signal propagation and error back propagation is carried out over and over again.
  • the process of continuous adjustment of weights is the learning and training process of the neural network. This process continues until the error of the network output is reduced to an acceptable level, or until a preset number of learning times.
  • optimization algorithms may include but are not limited to: gradient descent (Gradient Descent), stochastic gradient descent (Stochastic Gradient Descent, SGD), mini-batch gradient descent (mini-batch gradient descent), momentum method (Momentum), driven momentum Stochastic gradient descent (such as Nesterov), adaptive gradient descent (ADAptive GRADient descent, Adagrad), Adadelta algorithm, root mean square error reduction (root mean square prop, RMSprop), adaptive momentum estimation (Adaptive Moment Estimation, Adam) wait.
  • gradient descent Gradient Descent
  • stochastic gradient descent stochastic gradient descent
  • mini-batch gradient descent mini-batch gradient descent
  • momentum method Motionum
  • driven momentum Stochastic gradient descent such as Nesterov
  • adaptive gradient descent ADAptive GRADient descent, Adagrad
  • Adadelta algorithm root mean square error reduction (root mean square prop, RMSprop), adaptive momentum estimation (Adaptive Moment Estimation, Adam) wait.
  • the function of the encoding network (or coding model) is to compress channel information into channel characteristic information
  • the function of the decoding network (or called decoding model) is to restore the channel characteristic information into corresponding channel information. Therefore
  • the compression method of the encoding network and the recovery method of the decoding network need to match each other.
  • the base station and the terminal since models with good performance often have relatively large model sizes and high transmission overhead, and the base station and the terminal may not want the other party to know the model they use and the optimization processing, the base station and the terminal can independently train the required model.
  • Figure 4 is a flow chart of a model matching method provided by an embodiment of the present application.
  • the method is applied to a first device.
  • the first device can be a terminal or a network side device.
  • the network side device For example, it is a base station or core network equipment.
  • the method includes the following steps:
  • Step 41 The first device obtains the first channel characteristic information.
  • the first channel characteristic information is used by the second device.
  • the first model is obtained by processing the first channel information.
  • Step 42 The first device matches the second model of the first device with the first model according to the first channel characteristic information and the second channel information corresponding to the first channel characteristic information.
  • the above-mentioned first model and second model are models trained on different devices.
  • the second model is used to process the channel characteristic information obtained by the first device (i.e., network-side device, such as a base station); or, when the first device is a terminal , when the second device is a network-side device, the second model is used to process channel information.
  • the above-mentioned first channel information and the second channel information are the same, that is to say, the first channel information and the second channel information are the same channel information located in the second device and the first device respectively, and one is used for the first device.
  • One model processes the first channel information, and the other is used for model matching.
  • the above-mentioned first channel characteristic information may be obtained by the second device using the first model to encode/compress the first channel information.
  • the first model can be understood as a coding model trained in the second device.
  • the above processing of the channel characteristic information obtained by the first device can be understood as decoding/decompressing the channel characteristic information.
  • the above processing of channel information can be understood as encoding/compressing the channel information.
  • the above-mentioned first channel information/second channel information may be, but is not limited to, a channel matrix, a precoding matrix, etc.
  • the first model is a coding model trained in the terminal for encoding channel information
  • the second model is a coding model trained in the network-side device.
  • a decoding model used to decode the acquired channel characteristic information.
  • the encoding result of the channel information in the first model of the terminal is used as the input of the network-side decoding model, and this channel information is used as the output of the decoding model, and the decoding model can be trained.
  • the encoding model and the decoding model trained separately are matched, so that the channel characteristic information encoded by the terminal-side encoding model can be decoded by the network-side encoding model to obtain the corresponding channel information, thereby ensuring channel information feedback.
  • the first model is a coding model trained in the network-side device for encoding channel information
  • the second model is a coding model trained in the terminal. Coding model used to encode channel information. Since the encoding model and decoding model trained by the network side device match the channel information, and the encoding model and decoding model trained by the terminal match, therefore, by matching the second model with the first model, the network side device can The trained coding model matches the coding model trained by the terminal, so that the channel characteristic information encoded by the terminal-side coding model can be decoded by the network-side coding model to obtain the corresponding channel information, thereby ensuring channel information feedback.
  • the model matching method in the embodiment of the present application obtains the first channel characteristic information, which is obtained by the second device using the first model to process the first channel information, and based on the first channel characteristic information and the The second channel information corresponding to the first channel characteristic information matches the second model of the first device with the first model, so that the first model and the second model that are independently trained can be matched, so that the terminal side model processes the
  • the channel characteristic information can be recovered through the network side model to obtain the corresponding channel information, thereby ensuring channel information feedback.
  • model transmission can be avoided, thereby saving overhead and meeting the requirements of Network-side devices and terminals do not want the other party to know the model they are using and the need for corresponding optimization processing.
  • the first device may receive an identifier of the second channel information, and then determine the characteristics of the first channel based on the identifier of the second channel information.
  • the second channel information corresponding to the information is used to perform model matching based on the first channel characteristic information and the corresponding second channel information.
  • the base station sends indication information to instruct the terminal on the time, period, etc. to report matching information; then, the terminal determines the corresponding time based on the indication information or calculates the corresponding time based on the indicated period.
  • the matching information is reported on the configured time-frequency resources, including the first channel characteristic information and the identification (Identity, ID) of the corresponding channel information in the first data set.
  • This ID is used as the second channel information. identification; then, after receiving the ID, the base station finds the corresponding second channel information in the first data set.
  • the first device may determine the second model corresponding to the first channel characteristic information based on the obtained time-frequency domain position of the first channel characteristic information.
  • Channel information to perform model matching based on the first channel characteristic information and the corresponding second channel information For example, the terminal can send channel characteristic information at a designated time-frequency domain position, and the base station determines channel information corresponding to the channel characteristic information based on the time-frequency domain position.
  • the above time-frequency domain position can be configured by the network side.
  • the base station configures the time-frequency domain location for the terminal to send channel characteristic information.
  • the above-mentioned first channel information and second channel information may be channel information in a first data set, and the channel information in the first data set is used for model matching. That is, the terminal and the network side device can perform model processing based on the same first data set containing channel information.
  • the first data set may be referred to as a matching data set.
  • the terminal generates the corresponding coding result, that is, channel characteristic information, based on the channel information in the matching data set.
  • the terminal sends the channel characteristic information to the base station.
  • the base station codes the terminal's coding model based on the channel information in the matching data set and the corresponding channel characteristic information. Make a match.
  • the terminal can use data set A to train a channel state information (CSI) compressed AI model, including a coding model and a decoding model
  • the base station can use data set B to train a CSI compressed AI model, including a coding model. and decoding models.
  • data set A and data set B can be the same or different.
  • the matching data set used by the further model matching process can be different from both data set A and data set B.
  • the order of the first channel information and the second channel information in the channel information contained in the first data set may be agreed by the protocol or configured by the network side.
  • the order of the first channel information and the second channel information in the channel information contained in the first data set is fixed.
  • the first device may receive the first data set from the second device.
  • the first data set can be sent by the base station to the terminal in real time.
  • the terminal encodes certain channel information in the received first data set to obtain the channel characteristic information and feeds it back to the base station.
  • the base station then codes based on the channel characteristic information and the corresponding channel. information for model matching.
  • the first data set may be measured by the terminal in real time.
  • the terminal sends the estimated channel information and the encoded channel characteristic information to the base station, and the base station performs model matching based on the channel characteristic information and corresponding channel information.
  • the above-mentioned first data set can satisfy at least one of the following:
  • the third device may be understood as an independent device, an independent node, etc.
  • the terminal may receive configuration information of the network side device, where the configuration information is used to configure the third data set.
  • An identifier of a data set select the first data set from the second data set according to the identifier of the first data set.
  • the range of the second data set is agreed upon by the protocol and divided into multiple subsets, and the base station selects and instructs the terminal to use the subset (ie, the first data set).
  • the second data set may be agreed upon in a protocol, or may be collected or updated by a third device different from the first device and the second device.
  • the collection and/or updating of the second data set may be offline or long-term.
  • the network side equipment such as the base station can indicate in the second data set that part of it is the first data set.
  • the indication information can be downlink control information (Downlink Control Information, DCI), medium access control control element (Medium Access Control Control Element, MAC). CE) or Radio Resource Control (Radio Resource Control, RRC) signaling, etc.
  • the first device may perform any of the following: communicate with the second device and/or Or the third device interacts with the updated version of the first data set, interacts with the second device and/or the third device with the updated first data set; and uses the updated first data set to perform model matching. And/or, after the first data set is updated, the updated first data set and/or the updated version of the first data set may be interacted between the third device and the second device. That is to say, after the third device updates the first data set, the third device can notify the first device and/or the second device of the updated first data set through interaction with the first device and/or the second device. , and the first device and the second device can also interact with the updated first data set.
  • the base station and the terminal may exchange the updated version of the first data set through signaling to use the same version of the first data set for model matching.
  • the third device can periodically update the first data set and send it to the first device and/or the second device.
  • the first data set can be sent via the data plane.
  • the first device may perform any of the following: communicate with the second device and/or or the third device interacts with the updated version of the second data set, and interacts with the second device and/or the third device with the updated second data set. And/or, after the second data set is updated, the updated second data set and/or a version of the updated second data set may be interacted between the third device and the second device.
  • the third device can communicate with the first device
  • the first device and/or the second device may interact with each other to inform the first device and/or the second device of the updated second data set, and the first device and the second device may also interact with the updated second data set.
  • the above-mentioned matching of the second model of the first device with the first model may include at least one of the following:
  • the first device uses the first channel characteristic information as the input of the second model, uses the second channel information as the output of the second model, and retrains the second model;
  • the first device adjusts the parameters of the second model according to the first channel characteristic information and the second channel information.
  • the output of the first model can be used as the input of the second model (such as decoding model/decoder).
  • Train a second model eg, decoding model/decoder
  • parameters of the second model eg, decoding model/decoder
  • the first model and the second model that are independently trained can be matched, so that the channel characteristic information processed by the terminal side model can be passed through the network side.
  • the model is restored to obtain corresponding channel information, thereby ensuring channel information feedback.
  • the above-mentioned matching of the second model of the first device with the first model may include at least one of the following:
  • the first device uses the first channel characteristic information as the output of the second model, uses the second channel information as the input of the second model, and retrains the second model;
  • the first device adjusts the parameters of the second model according to the first channel characteristic information and the second channel information.
  • the inputs of the first model (such as encoding model/encoder) and the second model (such as encoding model/encoder) are the same channel information, and the second model can be trained by model (such as encoding model/encoder) or adjust the second model (such as encoding model/encoder) so that the output result of this second model (such as encoding model/encoder) is consistent with the first model (such as encoding model/encoder) )'s output results match, that is, they are the same.
  • the first model and the second model that are independently trained can be matched, so that the channel characteristic information processed by the terminal side model can be passed through the network side.
  • the model is restored to obtain corresponding channel information, thereby ensuring channel information feedback.
  • the execution subject may be a model matching device.
  • the model matching device executing the model matching method is used as an example to illustrate the model matching device provided by the embodiment of the present application.
  • Figure 5 is a schematic structural diagram of a model matching device provided by an embodiment of the present application.
  • the device is applied to a first device.
  • the first device can be a terminal or a network side device.
  • the network side device is, for example, Base station or core network equipment.
  • the model matching device 50 includes:
  • the acquisition module 51 is used to obtain the first channel characteristic information, which is obtained by the second device using the first model to process the first channel information;
  • Matching module 52 configured to match the first channel characteristic information and the first channel characteristic information corresponding to the first channel characteristic information.
  • second channel information matching the second model of the first device with the first model;
  • the first channel information is the same as the second channel information; the first model and the second model are models trained in different devices; when the first device is a network side device, the When the second device is a terminal, the second model is used to process the channel characteristic information obtained by the first device; or when the first device is a terminal and the second device is a network side device , the second model is used to process channel information.
  • model matching device 50 also includes:
  • a first receiving module configured to receive the identification of the second channel information
  • a first determination module configured to determine the second channel information corresponding to the first channel characteristic information according to the identification of the second channel information.
  • model matching device 50 also includes:
  • the second determination module is configured to determine the second channel information corresponding to the first channel characteristic information according to the time-frequency domain position of the first channel characteristic information.
  • the time-frequency domain position is configured by the network side.
  • the first channel information and the second channel information are channel information in a first data set, and the channel information in the first data set is used for model matching.
  • the order of the first channel information and the second channel information in the channel information contained in the first data set is agreed upon by the protocol or configured by the network side.
  • model matching device 50 also includes:
  • a second receiving module configured to receive the first data set from the second device.
  • the first data set satisfies at least one of the following:
  • a third device different from the first device and the second device.
  • the model matching device 50 when the first data set is a subset of the second data set indicated by the network side device, and if the first device is a terminal, the model matching device 50 further includes:
  • a third receiving module configured to receive configuration information of the network side device, where the configuration information is used to configure the identification of the first data set;
  • a selection module configured to select the first data set from the second data set according to the identification of the first data set.
  • the model matching device 50 when the first data set is provided by a third device different from the first device and the second device, the model matching device 50 further includes:
  • a first execution model configured to perform any of the following after the first data set is updated: interact with the updated version of the first data set with the second device and/or the third device, interact with the The second device and/or the third device communicate The first data set after mutual updates;
  • the matching module 52 is also configured to perform model matching using the updated first data set.
  • the updated first data set and/or the updated version of the first data set may be interacted between the third device and the second device.
  • the model matching device 50 when the second data set is provided by a third device different from the first device and the second device, the model matching device 50 further includes:
  • a second execution model configured to perform any of the following after the second data set is updated: interact with the updated version of the second data set with the second device and/or the third device; The second device and/or the third device interact with the updated second data set.
  • the updated second data set and/or the updated version of the second data set may be interacted between the third device and the second device.
  • the matching module 52 is used for at least one of the following:
  • the matching module 52 is used for at least one of the following:
  • the model matching device 50 provided by the embodiment of the present application can implement each process implemented by the method embodiment in Figure 4 and achieve the same technical effect. To avoid duplication, details will not be described here.
  • this embodiment of the present application also provides a communication device 60, which includes a processor 61 and a memory 62.
  • the memory 62 stores programs or instructions that can be run on the processor 61.
  • the communication device 60 may be a terminal or a network side device, such as a base station or a core network device.
  • An embodiment of the present application also provides a communication device.
  • the communication device is a first device and includes a processor and a communication interface.
  • the processor is configured to obtain first channel characteristic information.
  • the first channel characteristic information is obtained by the second device using the first
  • the model is obtained by processing the first channel information; according to the first channel characteristic information and the second channel information corresponding to the first channel characteristic information, the second model of the first device and the first model are Matching is performed; the first channel information is the same as the second channel information; the first model and the second model are models trained in different devices; when the first device is a network side device, When the second device is a terminal, the second model is used to process the channel characteristic information obtained by the first device; or when the first device is a terminal, the second device is a network When using a network side device, the second model is used to process channel information.
  • This embodiment corresponds to the above-mentioned method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this embodiment
  • FIG. 7 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
  • the terminal 700 includes but is not limited to: a radio frequency unit 701, a network module 702, an audio output unit 703, an input unit 704, a sensor 705, a display unit 706, a user input unit 707, an interface unit 708, a memory 709, a processor 710, etc. At least some parts.
  • the terminal 700 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 710 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
  • the terminal structure shown in FIG. 7 does not constitute a limitation on the terminal.
  • the terminal may include more or fewer components than shown in the figure, or some components may be combined or arranged differently, which will not be described again here.
  • the input unit 704 may include a graphics processing unit (Graphics Processing Unit, GPU) 7041 and a microphone 7042.
  • the graphics processor 7041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 707 includes a touch panel 7071 and at least one of other input devices 7072 .
  • Touch panel 7071 also called touch screen.
  • the touch panel 7071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 7072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 701 after receiving downlink data from the network side device, can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network side device.
  • the radio frequency unit 701 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
  • Memory 709 may be used to store software programs or instructions as well as various data.
  • the memory 709 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc.
  • memory 709 may include volatile memory or non-volatile memory, or memory 709 may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • the processor 710 may include one or more processing units; optionally, the processor 710 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above-mentioned modem processor may not be integrated into the processor 710.
  • the processor 710 is used to obtain the first channel characteristic information, which is obtained by processing the first channel information by the network side device using the first model; according to the first channel characteristic information and the first channel characteristic information,
  • the second channel information corresponding to the channel characteristic information matches the second model in the terminal 700 with the first model; the first channel information is the same as the second channel information; the first model is the same as the second channel information.
  • the second model is a model trained in different devices; the second model is used to process channel information.
  • the terminal 700 provided by the embodiment of the present application can implement each process implemented by the terminal in the method embodiment of Figure 4 and achieve the same technical effect. To avoid duplication, details will not be described here.
  • the embodiment of the present application also provides a network side device.
  • the network side device 80 includes: a processor 81 , a network interface 82 and a memory 83 .
  • the network interface 82 is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 80 in this embodiment of the present invention also includes: instructions or programs stored in the memory 83 and executable on the processor 81.
  • the processor 81 calls the instructions or programs in the memory 83 to execute what is shown in Figure 5 To avoid duplication, the methods for executing each module and achieving the same technical effect will not be described in detail here.
  • Embodiments of the present application also provide a readable storage medium.
  • Programs or instructions are stored on the readable storage medium.
  • the program or instructions are executed by a processor, each process of the above model matching method embodiment is implemented, and the same can be achieved. The technical effects will not be repeated here to avoid repetition.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the above model matching method embodiment. Each process can achieve the same technical effect. To avoid duplication, it will not be described again here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement the above model matching method embodiment.
  • Each process can achieve the same technical effect. To avoid repetition, we will not go into details here.
  • An embodiment of the present application also provides a communication system, including: a first device and a second device.
  • the first device can be used to perform the steps of the model matching method as described above.
  • the second device can utilize the first model. Process the first channel information to obtain first channel characteristic information, and send the first channel characteristic information to the first device.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

La présente demande se rapporte au domaine technique des communications, et divulgue un procédé et un appareil de mise en correspondance de modèles, un dispositif de communication, et un support de stockage lisible. Le procédé de mise en correspondance de modèles de modes de réalisation de la présente demande comprend les étapes suivantes : un premier dispositif acquiert des premières informations de caractéristique de canal, les premières informations de caractéristique de canal étant obtenues par un second dispositif traitant des premières informations de canal au moyen d'un premier modèle ; et met en correspondance un second modèle du premier dispositif avec le premier modèle selon les premières informations de caractéristique de canal et des secondes informations de canal correspondant aux premières informations de caractéristique de canal, le premier modèle et le second modèle représentant des modèles obtenus par entraînement sur différents dispositifs. Lorsque le premier dispositif représente un dispositif côté réseau et que le second dispositif représente un terminal, le second modèle est utilisé pour traiter des informations de caractéristique de canal acquises par le premier dispositif ; ou lorsque le premier dispositif représente un terminal et que le second dispositif représente un dispositif côté réseau, le second modèle est utilisé pour traiter des informations de canal.
PCT/CN2023/110481 2022-08-04 2023-08-01 Procédé et appareil de mise en correspondance de modèles, dispositif de communication, et support de stockage lisible WO2024027683A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210934857.5 2022-08-04
CN202210934857.5A CN117579201A (zh) 2022-08-04 2022-08-04 模型匹配方法、装置、通信设备及可读存储介质

Publications (1)

Publication Number Publication Date
WO2024027683A1 true WO2024027683A1 (fr) 2024-02-08

Family

ID=89848560

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/110481 WO2024027683A1 (fr) 2022-08-04 2023-08-01 Procédé et appareil de mise en correspondance de modèles, dispositif de communication, et support de stockage lisible

Country Status (2)

Country Link
CN (1) CN117579201A (fr)
WO (1) WO2024027683A1 (fr)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110255613A1 (en) * 2010-04-15 2011-10-20 Futurewei Technologies, Inc. System and Method for Feeding Back Channel Information
US20210273707A1 (en) * 2020-02-28 2021-09-02 Qualcomm Incorporated Neural network based channel state information feedback
CN113810086A (zh) * 2020-06-12 2021-12-17 华为技术有限公司 信道信息反馈方法、通信装置及存储介质
CN113938907A (zh) * 2020-07-13 2022-01-14 华为技术有限公司 通信的方法及通信装置
CN114172615A (zh) * 2020-09-11 2022-03-11 维沃移动通信有限公司 传输方法、装置、设备及可读存储介质
CN114679355A (zh) * 2020-12-24 2022-06-28 华为技术有限公司 通信方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110255613A1 (en) * 2010-04-15 2011-10-20 Futurewei Technologies, Inc. System and Method for Feeding Back Channel Information
US20210273707A1 (en) * 2020-02-28 2021-09-02 Qualcomm Incorporated Neural network based channel state information feedback
CN113810086A (zh) * 2020-06-12 2021-12-17 华为技术有限公司 信道信息反馈方法、通信装置及存储介质
CN113938907A (zh) * 2020-07-13 2022-01-14 华为技术有限公司 通信的方法及通信装置
CN114172615A (zh) * 2020-09-11 2022-03-11 维沃移动通信有限公司 传输方法、装置、设备及可读存储介质
CN114679355A (zh) * 2020-12-24 2022-06-28 华为技术有限公司 通信方法和装置

Also Published As

Publication number Publication date
CN117579201A (zh) 2024-02-20

Similar Documents

Publication Publication Date Title
WO2024027683A1 (fr) Procédé et appareil de mise en correspondance de modèles, dispositif de communication, et support de stockage lisible
WO2024032694A1 (fr) Procédé et appareil de traitement prédiction de csi, dispositif de communication et support de stockage lisible
WO2024037380A1 (fr) Procédés et appareil de traitement d'informations de canal, dispositif de communication et support de stockage
WO2024104126A1 (fr) Procédé et appareil de mise à jour de modèle de réseau d'ia, et dispositif de communication
WO2024032606A1 (fr) Procédé et appareil de transmission d'informations, dispositif, système et support de stockage
WO2024217495A1 (fr) Procédé de traitement d'informations, appareil de traitement d'informations, terminal et dispositif côté réseau
WO2024067280A1 (fr) Procédé et appareil de mise à jour de paramètre de modèle d'ia, et dispositif de communication
WO2024041421A1 (fr) Procédé et appareil de traitement de rétroaction de mesure, terminal et dispositif côté réseau
WO2024067665A1 (fr) Procédé et appareil de traitement de prédiction de csi, dispositif de communication et support de stockage lisible
WO2024041420A1 (fr) Procédé et appareil de traitement de rétroaction de mesure, terminal et dispositif côté réseau
WO2023179570A1 (fr) Procédé et appareil de transmission d'informations de caractéristique de canal, terminal et dispositif côté réseau
WO2024164962A1 (fr) Procédés et appareil de traitement de communication, dispositif et support de stockage lisible
WO2024222573A1 (fr) Procédé de traitement d'information, appareil de traitement d'information, terminal et dispositif côté réseau
WO2023134628A1 (fr) Procédé et appareil de transmission, et dispositif
WO2024149156A1 (fr) Procédé et appareil de transmission d'informations, et terminal et dispositif côté réseau
WO2024067281A1 (fr) Procédé et appareil de traitement de modèle d'ia, et dispositif de communication
WO2024078405A1 (fr) Procédés de transmission, appareil, dispositif de communication et support de stockage lisible
WO2024120358A1 (fr) Procédé de transmission d'informations, appareil de transmission d'informations et dispositif de communication
WO2024125391A2 (fr) Procédé et appareil de surveillance de modèle d'ia, procédé et appareil de mesure de performance de modèle d'ia, et dispositif
WO2023186091A1 (fr) Procédé, appareil et dispositif de détermination d'échantillon
WO2024160085A1 (fr) Procédé et appareil de traitement de communication, dispositif et support de stockage lisible
WO2024051564A1 (fr) Procédé de transmission d'informations, procédé d'entraînement de modèle de réseau d'ia, appareil, et dispositif de communication
WO2024012303A1 (fr) Procédé et appareil d'interaction de modèle de réseau ai, et dispositif de communication
WO2023125934A1 (fr) Procédé et appareil de transmission d'informations de réseau d'ia, et dispositif de communication
WO2023098535A1 (fr) Procédé et appareil d'interaction d'informations, et dispositif de communication

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: 23849390

Country of ref document: EP

Kind code of ref document: A1