WO2023015499A1 - Procédé et dispositif de communication sans fil - Google Patents

Procédé et dispositif de communication sans fil Download PDF

Info

Publication number
WO2023015499A1
WO2023015499A1 PCT/CN2021/112118 CN2021112118W WO2023015499A1 WO 2023015499 A1 WO2023015499 A1 WO 2023015499A1 CN 2021112118 W CN2021112118 W CN 2021112118W WO 2023015499 A1 WO2023015499 A1 WO 2023015499A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
channel
terminal device
network device
model
Prior art date
Application number
PCT/CN2021/112118
Other languages
English (en)
Chinese (zh)
Inventor
刘文东
田文强
肖寒
Original Assignee
Oppo广东移动通信有限公司
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 Oppo广东移动通信有限公司 filed Critical Oppo广东移动通信有限公司
Priority to CN202180097891.5A priority Critical patent/CN117322044A/zh
Priority to PCT/CN2021/112118 priority patent/WO2023015499A1/fr
Publication of WO2023015499A1 publication Critical patent/WO2023015499A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W36/00—Hand-off or reselection arrangements

Definitions

  • the embodiments of the present application relate to the communication field, and in particular to a method and device for wireless communication.
  • terminal devices can use encoders to compress channel information
  • network devices can use decoders to reconstruct channel information.
  • the channels of different cells also have different potential characteristics. In this case, how to perform channel information feedback is an urgent problem to be solved.
  • the present application provides a wireless communication method and device.
  • migration training is performed on the encoding network on the terminal device and the decoding network on the network device based on the data set corresponding to the changed channel scene.
  • the adaptive update of the encoding network and the decoding network for channel scene changes is realized, the adaptation and generalization capabilities of the encoding network and the decoding network are improved, and it is beneficial to improve the compression feedback accuracy of channel information feedback when the channel environment characteristics change.
  • a wireless communication method including: when the channel scene is changed from the first channel scene to the second channel scene, the network device performs the first encoding network deployment on the terminal device according to the target training data set Perform migration training with the first decoding network deployed on the network device to obtain a second encoding network and a second decoding network, wherein the target training data set includes channel data in the second channel scenario, and the second an encoding network and the first decoding network are adapted to the first channel scenario;
  • the network device decodes the target bit stream through the second decoding network to obtain target channel data, where the target bit stream is the target bit stream obtained by the terminal device under the second channel scenario through the second encoding network obtained by encoding the channel data.
  • a wireless communication method including: a network device sends second information to a terminal device, and the second information is used for the terminal device to update the terminal device when the channel scene changes.
  • Migration training is performed on the first encoding network deployed on the network device and the first decoding network deployed on the network device to obtain the second encoding network and the second decoding network, wherein the first encoding network and the first decoding network are adapted to The channel scene before the change is adapted, and the second encoding network and the second decoding network are adapted to the channel scene after the change.
  • a wireless communication method including: when the channel scenario is changed from the first channel scenario to the second channel scenario, the terminal device performs the first channel deployment on the terminal device according to the target training data set.
  • the encoding network and the first decoding network deployed on the network device perform migration training to obtain a second encoding network and a second decoding network, wherein the target training data set includes channel data in the second channel scenario, and the second an encoding network and the first decoding network are adapted to the first channel scenario;
  • the terminal device encodes the channel data in the second channel scenario through the second encoding network to obtain a target bit stream.
  • a wireless communication method including: a terminal device sends first information to a network device, and the first information is used for the network device to deploy information on the terminal device when the channel scene changes. Migration training is performed on the first encoding network and the first decoding network deployed on the network device to obtain the second encoding network and the second decoding network, wherein the first encoding network and the first decoding network are adaptively changed In the previous channel scenario, the second encoding network and the second decoding network adapt to the changed channel scenario.
  • a terminal device is provided, configured to execute the method in any one of the above-mentioned third aspect to the fourth aspect or in each implementation manner thereof.
  • the terminal device includes a functional module configured to execute any one of the third aspect to the fourth aspect or the method in each implementation manner thereof.
  • a network device configured to execute the method in any one of the first aspect to the second aspect or each implementation manner thereof.
  • the network device includes a functional module configured to execute any one of the above first aspect to the second aspect or the method in each implementation manner thereof.
  • a terminal device including a processor and a memory.
  • the memory is used to store a computer program
  • the processor is used to call and run the computer program stored in the memory, and execute the functional modules of any one of the above third to fourth aspects or the method in each implementation manner.
  • a network device including a processor and a memory.
  • the memory is used to store a computer program
  • the processor is used to invoke and run the computer program stored in the memory to execute any one of the above first to second aspects or the method in each implementation manner.
  • a chip configured to implement any one of the foregoing first to fourth aspects or the method in each implementation manner thereof.
  • the chip includes: a processor, configured to call and run a computer program from the memory, so that the device installed with the device executes any one of the above-mentioned first to fourth aspects or any of the implementations thereof. method.
  • a computer-readable storage medium for storing a computer program, and the computer program causes a computer to execute any one of the above-mentioned first to fourth aspects or the method in each implementation manner thereof.
  • a computer program product including computer program instructions, the computer program instructions causing a computer to execute any one of the above first to fourth aspects or the method in each implementation manner thereof.
  • a twelfth aspect provides a computer program that, when running on a computer, causes the computer to execute any one of the above first to fourth aspects or the method in each implementation manner.
  • the network device or terminal device can migrate the encoding network deployed on the terminal device and the decoding network deployed on the network device according to the target training data set corresponding to the changed channel scene Training, so as to realize the adaptive update of the encoding network and the decoding network when the channel scene changes, improve the adaptation and generalization ability of the encoding network and the decoding network, and help improve the compression feedback of the channel information feedback when the channel environment characteristics change precision.
  • FIG. 1 is a schematic diagram of a communication system architecture applied in an embodiment of the present application.
  • Fig. 2 is a schematic diagram of a neural network provided by the present application.
  • Fig. 3 is a schematic diagram of a convolutional neural network provided in the present application.
  • Fig. 4 is a schematic diagram of an LSTM unit provided in the present application.
  • Fig. 5 is a schematic diagram of channel information feedback provided by the present application.
  • Fig. 6 is a schematic diagram of another channel information feedback provided by the present application.
  • Fig. 7 is a schematic interaction diagram of a wireless communication method provided according to an embodiment of the present application.
  • FIG. 8 is a schematic interaction diagram of an example of network equipment performing channel scene migration training provided by an embodiment of the present application.
  • FIG. 9 is a schematic interaction diagram of another example of channel scene migration training performed by a network device provided in an embodiment of the present application.
  • FIG. 10 is another schematic interaction diagram of a network device performing channel scene migration training provided by an embodiment of the present application.
  • FIG. 11 is a schematic interaction diagram of yet another example of network equipment performing channel scene migration training provided by an embodiment of the present application.
  • Fig. 12 is a schematic interaction diagram of another wireless communication method provided according to an embodiment of the present application.
  • FIG. 13 is a schematic interaction diagram of an example of channel scene migration training performed by a terminal device provided in an embodiment of the present application.
  • FIG. 14 is another schematic interaction diagram of a terminal device performing channel scene migration training provided by an embodiment of the present application.
  • Fig. 15 is a schematic block diagram of a network device provided according to an embodiment of the present application.
  • Fig. 16 is a schematic block diagram of a terminal device provided according to an embodiment of the present application.
  • Fig. 17 is a schematic block diagram of another terminal device provided according to an embodiment of the present application.
  • Fig. 18 is a schematic block diagram of another network device provided according to an embodiment of the present application.
  • Fig. 19 is a schematic block diagram of a communication device provided according to an embodiment of the present application.
  • Fig. 20 is a schematic block diagram of a device provided according to an embodiment of the present application.
  • Fig. 21 is a schematic block diagram of a communication system provided according to an embodiment of the present application.
  • GSM Global System of Mobile communication
  • CDMA Code Division Multiple Access
  • WCDMA Wideband Code Division Multiple Access
  • GPRS General Packet Radio Service
  • LTE Long Term Evolution
  • LTE-A Advanced long term evolution
  • NR New Radio
  • NTN Non-Terrestrial Networks
  • UMTS Universal Mobile Telecommunications System
  • WLAN Wireless Local Area Networks
  • Wireless Fidelity Wireless Fidelity
  • D2D Device to Device
  • M2M Machine to Machine
  • MTC Machine Type Communication
  • V2V Vehicle to Vehicle
  • V2X Vehicle to everything
  • the communication system in the embodiment of the present application can be applied to a carrier aggregation (Carrier Aggregation, CA) scenario, can also be applied to a dual connectivity (Dual Connectivity, DC) scenario, and can also be applied to an independent (Standalone, SA ) meshing scene.
  • Carrier Aggregation, CA Carrier Aggregation
  • DC Dual Connectivity
  • SA independent meshing scene
  • the communication system in the embodiment of the present application can be applied to an unlicensed spectrum, where the unlicensed spectrum can also be considered as a shared spectrum; or, the communication system in the embodiment of the present application can also be applied to a licensed spectrum, Wherein, the licensed spectrum can also be regarded as a non-shared spectrum.
  • the embodiments of the present application describe various embodiments in conjunction with network equipment and terminal equipment, wherein the terminal equipment may also be referred to as user equipment (User Equipment, UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.
  • user equipment User Equipment, UE
  • access terminal user unit
  • user station mobile station
  • mobile station mobile station
  • remote station remote terminal
  • mobile device user terminal
  • terminal wireless communication device
  • wireless communication device user agent or user device
  • the terminal device can be a station (STATION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (Session Initiation Protocol, SIP) phone, a wireless local loop (Wireless Local Loop, WLL) station, a personal digital assistant (Personal Digital Assistant, PDA) devices, handheld devices with wireless communication functions, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices, wearable devices, next-generation communication systems such as terminal devices in NR networks, or future Terminal equipment in the evolved public land mobile network (Public Land Mobile Network, PLMN) network, etc.
  • PLMN Public Land Mobile Network
  • the terminal device can be deployed on land, including indoor or outdoor, handheld, wearable or vehicle-mounted; it can also be deployed on water (such as ships, etc.); it can also be deployed in the air (such as aircraft, balloons and satellites) superior).
  • the terminal device may be a mobile phone (Mobile Phone), a tablet computer (Pad), a computer with a wireless transceiver function, a virtual reality (Virtual Reality, VR) terminal device, an augmented reality (Augmented Reality, AR) terminal Equipment, wireless terminal equipment in industrial control, wireless terminal equipment in self driving, wireless terminal equipment in remote medical, wireless terminal equipment in smart grid , wireless terminal equipment in transportation safety, wireless terminal equipment in smart city, or wireless terminal equipment in smart home.
  • a virtual reality (Virtual Reality, VR) terminal device an augmented reality (Augmented Reality, AR) terminal Equipment
  • wireless terminal equipment in industrial control wireless terminal equipment in self driving
  • wireless terminal equipment in remote medical wireless terminal equipment in smart grid
  • wireless terminal equipment in transportation safety wireless terminal equipment in smart city, or wireless terminal equipment in smart home.
  • the terminal device may also be a wearable device.
  • Wearable devices can also be called wearable smart devices, which is a general term for the application of wearable technology to intelligently design daily wear and develop wearable devices, such as glasses, gloves, watches, clothing and shoes.
  • a wearable device is a portable device that is worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not only a hardware device, but also achieve powerful functions through software support, data interaction, and cloud interaction.
  • Generalized wearable smart devices include full-featured, large-sized, complete or partial functions without relying on smart phones, such as smart watches or smart glasses, etc., and only focus on a certain type of application functions, and need to cooperate with other devices such as smart phones Use, such as various smart bracelets and smart jewelry for physical sign monitoring.
  • the network device may be a device for communicating with the mobile device, and the network device may be an access point (Access Point, AP) in WLAN, a base station (Base Transceiver Station, BTS) in GSM or CDMA , or a base station (NodeB, NB) in WCDMA, or an evolved base station (Evolutional Node B, eNB or eNodeB) in LTE, or a relay station or access point, or a vehicle-mounted device, a wearable device, and an NR network A network device or a base station (gNB) in a network device or a network device in a future evolved PLMN network or a network device in an NTN network.
  • AP Access Point
  • BTS Base Transceiver Station
  • NodeB, NB base station
  • Evolutional Node B, eNB or eNodeB evolved base station
  • LTE Long Term Evolution
  • eNB evolved base station
  • gNB base station
  • the network device may have a mobile feature, for example, the network device may be a mobile device.
  • the network equipment may be a satellite, balloon station.
  • the satellite can be a low earth orbit (low earth orbit, LEO) satellite, a medium earth orbit (medium earth orbit, MEO) satellite, a geosynchronous earth orbit (geosynchronous earth orbit, GEO) satellite, a high elliptical orbit (High Elliptical Orbit, HEO) satellite. ) Satellite etc.
  • the network device may also be a base station installed on land, in water, or other locations.
  • the network device may provide services for a cell, and the terminal device communicates with the network device through the transmission resources (for example, frequency domain resources, or spectrum resources) used by the cell, and the cell may be a network device ( For example, a cell corresponding to a base station), the cell may belong to a macro base station, or may belong to a base station corresponding to a small cell (Small cell), and the small cell here may include: a metro cell (Metro cell), a micro cell (Micro cell), a pico cell ( Pico cell), Femto cell, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.
  • the transmission resources for example, frequency domain resources, or spectrum resources
  • the cell may be a network device (
  • the cell may belong to a macro base station, or may belong to a base station corresponding to a small cell (Small cell)
  • the small cell here may include: a metro cell (Metro cell), a micro cell (Micro
  • the communication system 100 may include a network device 110, and the network device 110 may be a device for communicating with a terminal device 120 (or called a communication terminal, terminal).
  • the network device 110 can provide communication coverage for a specific geographical area, and can communicate with terminal devices located in the coverage area.
  • FIG. 1 exemplarily shows one network device and two terminal devices.
  • the communication system 100 may include multiple network devices and each network device may include other numbers of terminal devices within the coverage area. This embodiment of the present application does not limit it.
  • the communication system 100 may further include other network entities such as a network controller and a mobility management entity, which is not limited in this embodiment of the present application.
  • a device with a communication function in the network/system in the embodiment of the present application may be referred to as a communication device.
  • the communication equipment may include a network equipment 110 and a terminal equipment 120 with communication functions.
  • the network equipment 110 and the terminal equipment 120 may be the specific equipment described above, and will not be repeated here.
  • the communication device may also include other devices in the communication system 100, such as network controllers, mobility management entities and other network entities, which are not limited in this embodiment of the present application.
  • the "indication" mentioned in the embodiments of the present application may be a direct indication, may also be an indirect indication, and may also mean that there is an association relationship.
  • a indicates B which can mean that A directly indicates B, for example, B can be obtained through A; it can also indicate that A indirectly indicates B, for example, A indicates C, and B can be obtained through C; it can also indicate that there is an association between A and B relation.
  • the term "corresponding" may indicate that there is a direct or indirect correspondence between the two, or that there is an association between the two, or that it indicates and is indicated, configuration and is configuration etc.
  • predefined or “preconfigured” can be realized by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in devices (for example, including terminal devices and network devices).
  • the application does not limit its specific implementation.
  • pre-defined may refer to defined in the protocol.
  • the "protocol” may refer to a standard protocol in the communication field, for example, may include the LTE protocol, the NR protocol, and related protocols applied to future communication systems, which is not limited in the present application.
  • the eigenvector feedback based on the codebook is mainly used to enable the base station to obtain the downlink CSI.
  • the base station sends a downlink channel state information reference signal (Channel State Information Reference Signal, CSI-RS) to the terminal device, and the terminal device uses the CSI-RS to estimate the CSI of the downlink channel, and performs eigenvalue decomposition on the estimated downlink channel, Obtain the eigenvector corresponding to this downlink channel.
  • CSI-RS Downlink channel state information reference signal
  • a neural network is an operational model composed of multiple neuron nodes connected to each other, in which the connection between nodes represents the weighted value from the input signal to the output signal, called weight; each node performs weighted summation of different input signals (summation, SUM), and output through a specific activation function (f).
  • a simple neural network is shown in Figure 2, which includes an input layer, a hidden layer, and an output layer. Through different connection methods, weights, and activation functions of multiple neurons, different outputs can be generated, and then fitted from input to output. Mapping relations.
  • Deep learning uses a deep neural network with multiple hidden layers, which greatly improves the ability of the network to learn features, and can fit complex nonlinear mappings from input to output, so it is widely used in the fields of speech and image processing.
  • deep learning also includes common basic structures such as convolutional neural network (CNN), recurrent neural network (Recurrent Neural Network, RNN).
  • CNN convolutional neural network
  • RNN Recurrent Neural Network
  • the basic structure of a convolutional neural network includes: an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer, as shown in Figure 3.
  • Each neuron of the convolution kernel in the convolution layer is locally connected to its input, and the local maximum or average feature of a certain layer is extracted by introducing a pooling layer, which effectively reduces the parameters of the network and mines local features. It enables the convolutional neural network to converge quickly and obtain excellent performance.
  • RNN is a neural network that models sequence data. It has achieved remarkable results in the field of natural language processing, such as machine translation and speech recognition. The specific performance is that the network memorizes the information of the past moment and uses it in the calculation of the current output, that is, the nodes between the hidden layers are no longer connected but connected, and the input of the hidden layer includes not only the input layer but also the Includes the output of the hidden layer at the previous moment.
  • Commonly used RNNs include structures such as Long Short-Term Memory (LSTM) and gated recurrent unit (GRU).
  • Figure 4 shows a basic LSTM cell structure, which can contain a tanh activation function. Unlike RNN, which only considers the nearest state, the cell state of LSTM will determine which states should be kept and which states should be forgotten, solving the traditional Shortcomings of RNN in long-term memory.
  • AI artificial intelligence
  • deep learning has achieved great success in computer vision, natural language processing, etc.
  • the field of communication has begun to try to use deep learning to solve technical problems that are difficult to solve by traditional communication methods, such as deep learning.
  • the neural network architecture commonly used in deep learning is nonlinear and data-driven. It can extract features from the actual channel matrix data and restore the channel matrix information compressed and fed back by the terminal side as much as possible on the base station side. It is possible to reduce the CSI feedback overhead on the terminal side.
  • the CSI feedback based on deep learning regards the channel information as the image to be compressed, uses the deep learning self-encoder to compress the channel information, and reconstructs the compressed channel image at the sending end, which can preserve the channel information to a greater extent , as shown in Figure 5.
  • a typical channel information feedback system is shown in FIG. 6 .
  • the entire feedback system is divided into encoder and decoder parts, which are deployed at the sending end and receiving end respectively.
  • the transmitting end obtains the channel information through channel estimation
  • the channel information matrix is compressed and encoded through the neural network of the encoder, and the compressed bit stream is fed back to the receiving end through the air interface feedback link, and the receiving end passes the decoder according to the feedback bit stream
  • the channel information is restored to obtain complete feedback channel information.
  • the encoder shown in Figure 6 uses the superposition of multiple fully connected layers, and the design of the convolutional layer and residual structure is used in the decoder. Under the condition that the encoding and decoding framework remains unchanged, the network model structure inside the encoder and decoder can be flexibly designed.
  • the channel information feedback in the NR system is a codebook-based feedback scheme. Since this scheme performs quantitative feedback on the channel information according to the preset codebook vector, the codebook itself cannot be adjusted according to the real-time changes of the channel environment, so the codebook feedback obtained by the base station has a large error compared with the original CSI vector, and then Resulting in limited accuracy of CSI recovery.
  • AI-based channel information feedback considers that the encoder of the AI autoencoder is used to compress the channel information at the sending end, and the decoder of the AI autoencoder is used to reconstruct the channel information at the receiving end.
  • the AI-based solution uses the nonlinear fitting ability of the neural network to compress and feed back the channel information, which can greatly improve the compression efficiency and feedback accuracy.
  • the channels of different cells also have different potential characteristics.
  • the inherent disadvantage of the generalization problem of the neural network itself in practical applications makes the trained network only suitable for the channel test set with the same characteristics as the channel data of the training set, that is, the training set is often difficult to cover all situations, when the scene characteristics change When , it is difficult for the trained model to continue to maintain good generalization performance. Therefore, how to realize the online update of the CSI autoencoder model is an urgent problem to be solved.
  • this application proposes a technical solution.
  • the network device or terminal device performs online scene migration training for the encoding network at the sending end and the decoding network at the receiving end, and realizes the encoding network and decoding network.
  • the adaptive update when the channel scene changes improves the adaptation and generalization capabilities of the encoding network and the decoding network, thereby improving the compression feedback accuracy of the channel information feedback when the channel environment characteristics change.
  • FIG. 7 is a schematic interaction diagram of a wireless communication method 200 provided according to an embodiment of the present application. As shown in FIG. 7, the method 200 includes the following content:
  • the network device performs migration training on the first encoding network and the first decoding network according to the target training data set to obtain a second encoding network and a second decoding network.
  • the encoding network in the embodiment of the present application may also be called an encoder, and the decoding network may also be called a decoder.
  • the encoder and decoder can constitute a CSI autoencoder, or a channel information feedback system. That is, the first encoding network may correspond to an encoder in the CSI autoencoder, and the first decoding network may correspond to a decoder in the CSI autoencoder.
  • the encoding network may be deployed on terminal devices, and the decoding network may be deployed on network devices.
  • the terminal device can encode the channel data through the encoding network to obtain the target bit stream, and further send the target bit stream to the network device, and the network device decodes the target bit stream through the decoding network to obtain the target channel data.
  • the first encoding network and the first decoding network correspond to a first channel scenario, or in other words, a first channel environment.
  • the first encoding network and the first decoding network are adapted to the first channel scenario.
  • adapting the first encoding network and the first decoding network to the first channel scenario may refer to:
  • the terminal device encodes the channel data based on the first encoding network to obtain the target bit stream, and the network device decodes the target bit stream based on the first encoding network to obtain the target channel data, which can achieve relatively Excellent CSI feedback and recovery performance.
  • the autoencoder model based on the first encoding network and the first decoding network can achieve better CSI feedback and recovery performance.
  • the network device when the channel scene changes, can perform migration training on the currently used model based on the channel data set corresponding to the changed channel scene, so that the model after migration training can adapt to the changed channel scenarios to improve CSI feedback and recovery performance.
  • the network device can perform migration training on the currently used first encoding network and first decoding network according to the target training data set to obtain the second encoding network and a second decoding network, wherein the target training data set includes channel data corresponding to a second channel scenario.
  • the method 200 further includes:
  • the terminal device acquires the model parameters of the second coding network from the network device.
  • the network device may send the trained model parameters of the second encoding network to the terminal device.
  • the method 200 further includes:
  • the terminal device may encode the channel data in the second channel scenario through the second encoding network to obtain the target bit stream;
  • the terminal device sends the target bit stream to the network device.
  • the network device may decode the target bit stream through the second decoding network to obtain target channel data.
  • the network device can update the currently used model based on the changed channel data set, so that the network device and the terminal device can use the model adapted to the changed channel scene. Compression and feedback of channel data in different channel scenarios is beneficial to improve CSI feedback and restoration performance.
  • the change of the channel scene may be determined by the network device, or may also be determined by the terminal device.
  • the network device can monitor the channel quality indicator to determine whether the channel scene changes. For example, the network device may continuously or periodically detect the channel quality indicator, and further determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the channel quality indicator may include at least one of the following: CSI, Reference Signal Receiving Power (RSRP), Reference Signal Receiving Quality (Reference Signal Receiving Quality, RSRQ), received signal Received Signal Strength Indication (RSSI), Signal to Interference plus Noise Ratio (SINR).
  • RSRP Reference Signal Receiving Power
  • RSRQ Reference Signal Receiving Quality
  • RSSI received signal Received Signal Strength Indication
  • SINR Signal to Interference plus Noise Ratio
  • the network device determines that the channel scene is changed when the change amount of the channel quality index is greater than the first threshold.
  • the network device may determine that the channel scene is changed when the change amount of the RSRP is greater than the first RSRP threshold.
  • the first threshold may be determined according to the recovery performance of CSI feedback, for example, when the variation of the channel quality index is greater than a certain value, continue to use the current encoder model and decoder model, resulting in the recovery of CSI feedback The performance becomes worse, or the requirement of feedback accuracy is not met. In this case, this value may be set as the first threshold.
  • different channel scenarios may correspond to corresponding channel quality indicator ranges, and the network device may determine the changed channel scenario according to the changed channel quality indicator and the channel quality indicator range.
  • channel scene change in the embodiment of the present application may include any situation that causes model incompatibility, for example, the change amount of the channel quality index exceeds a certain threshold, or the terminal device Moving from one cell to another etc., the application is not limited thereto.
  • the method 200 further includes:
  • the network device sends first indication information to the terminal device, where the first indication information is used to indicate that a model is updated and/or a channel scene is changed, where the model update includes a model update of the encoding network and/or Model updates for the decoding network.
  • the first indication information may be sent through any downlink message or signaling, for example, the first indication information is sent through downlink control information (Downlink Control Information, DCI).
  • DCI Downlink Control Information
  • the terminal device may monitor the channel quality indicator to determine whether the channel scene changes. For example, the terminal device may continuously or periodically detect the channel quality indicator, and further determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the manner in which the terminal device monitors the channel quality indicator may refer to the relevant implementation of the network device, and for the sake of brevity, details are not repeated here.
  • the method 200 further includes:
  • the network device receives second instruction information sent by the terminal device, where the second instruction information is used to indicate model update and/or channel scene change, where the model update includes model update of the coding network and/or Or model updates for decoding networks.
  • the second indication information may be sent through any uplink message or signaling, for example, the second indication information is sent through uplink control information (Uplink Control Information, UCI).
  • uplink control information Uplink Control Information, UCI.
  • the method 200 further includes:
  • the network device acquires the model parameters of the first coding network from the terminal device.
  • the terminal device when the terminal device receives the first indication information sent by the network device, it reports the currently used encoder model to the network device, so that the network device performs migration training according to the encoder model and the decoder model currently used by the network device .
  • the terminal device when the terminal device determines that the channel scene changes by monitoring the channel quality index, it reports the currently used encoder model to the network device, so that the network device performs migration training based on the encoder model and the decoder model currently used by the network device.
  • the data set on which the network device performs migration training can be obtained from the terminal device, or it can also be pre-stored on the network device, and this application does not limit the method of obtaining the data set .
  • the method 200 further includes:
  • the network device receives the target training data set reported by the terminal device.
  • the terminal device when the terminal device receives the first indication information sent by the network device, it reports the target training data set to the network device, so that the network device can perform an evaluation of the encoder model and The decoder model currently used by the network device performs migration training to obtain an encoder model and a decoder model adapted to the changed channel scenario.
  • the terminal device when the terminal device determines that the channel scene changes by monitoring the channel quality index, it reports the target training data set to the network device, so that the network device can use the target training data set to update the encoder model and network currently used by the terminal device.
  • the decoder model currently used by the device performs migration training to obtain an encoder model and a decoder model adapted to the changed channel scenario.
  • model parameters of the first encoding network and the target training data set may be sent through the same message, or may also be sent through different messages, and this application does not make any limited.
  • the second indication information and the model parameters of the first encoding network may be sent through the same message, or may also be sent through different messages, which is not limited in this application .
  • the terminal device may send the second indication information to the network device, and at the same time report the currently used encoder model to the network device.
  • the target training data set includes a CSI vector on at least one frequency domain unit within a first time period after the first channel scene is changed to the second channel scene.
  • the unit of the first duration may be a subframe, a time slot, or a symbol, or may also be other time units, which are not limited in the present application.
  • the first duration includes N time slots, where N is a positive integer.
  • the N is predefined or configured by the network device, for example, the network device can configure the parameter N through DCI.
  • the frequency domain unit may be a subband, a physical resource block (Physical Resource Block, PRB), or other frequency domain units, which are not limited in this application.
  • PRB Physical Resource Block
  • the at least one frequency domain unit includes K subbands, where K is a positive integer.
  • the K is predefined, or configured by the network device, for example, the network device can configure the parameter K through DCI.
  • the at least one frequency domain unit includes M PRBs, where M is a positive integer.
  • the M is predefined or configured by the network device, for example, the network device can configure the parameter M through DCI.
  • the target training data set is used for model migration training, in order to meet the requirements of online update of the model, higher CSI feedback accuracy is required, and the CSI vector in the target training data set can be obtained based on the first codebook quantization, wherein, the first codebook has higher precision, for example, the precision of the first codebook is higher than the precision of the Type 2 codebook.
  • the first codebook may be a dedicated codebook for the data set for model migration training, that is, only use this codebook for feedback on the data set for model migration training.
  • the terminal device may use higher-precision quantization bits to quantize the fed-back amplitude and phase to obtain the target training data set.
  • the configuration of the first codebook is sent by the network device through DCI.
  • the target training data set is determined according to at least one data set in multiple data sets prestored on the network device.
  • the network device can pre-store data sets corresponding to various channel scenarios, such as Line-of-Sight (LoS) scenarios, Non-Line-of-Sight (NLoS) scenarios, indoor scenes, outdoor scenes, low-speed moving scenes, high-speed moving scenes and other scenes corresponding to the data sets.
  • LoS Line-of-Sight
  • NoS Non-Line-of-Sight
  • the network device may determine the target training data set from the multiple pre-stored data sets according to the changed channel scene.
  • the network device may determine samples in the data set corresponding to the changed channel scene as the target training data set.
  • the target training data set includes samples in multiple data sets, where each data set corresponds to a corresponding weight, and the weight corresponding to each data set is based on the applicable channel scenario of the data set and the Correlation determination of the above-mentioned changed channel scene. For example, a data set corresponding to a channel scene with a large correlation with the changed channel scene can be given a larger weight, and a data set corresponding to a channel scene with a small correlation with the changed channel scene can be given a smaller weight . That is, the samples with high matching degree of channel scene are given higher weight, and the samples with low matching degree of channel scene are given lower weight.
  • the changed channel scenario may be learned by the network device by monitoring the channel quality index, or may also be determined according to the third indication information sent by the terminal device, where the third indication information is used to indicate the changed channel scenario, wherein the terminal device can determine the changed channel scenario according to the changed channel quality indicator.
  • the third indication information and the encoder model currently used by the terminal device may be sent through the same message, or may also be sent through different messages, which is not limited in this application.
  • the terminal device may send to the network device first information for assisting the network device in performing model migration training, and the first information may include at least one of the following:
  • the second indication information is used to indicate that the model is updated and/or the channel scene is changed
  • the third indication information is used to indicate the changed channel scene
  • the encoder model currently used by the terminal device is the encoder model currently used by the terminal device.
  • the network device can learn the target training data set used for model migration training, the encoder model currently used by the terminal device, and the decoder model currently used by the network device. Further, the network device can perform migration training on the currently used encoder model and decoder model based on the target training data set. For example, fine-tuning the model parameters, as an example, fixing the parameters of some layers in the network structure and only adjusting the parameters of other layers, etc. This application does not limit the specific adjustment methods. Finally, the updated encoder model and decoder model converge on the target training data set.
  • the training rounds on the target training data set during migration training and the training loss function used for migration training, and the migration training parameters such as the training optimizer can be configured by the network device, or can be is predefined.
  • the network device After the network device completes the online model update, the network device sends the updated encoder model parameters to the terminal device, that is, the network device can send the model parameters of the second encoding network obtained through migration training to the terminal device.
  • the terminal device may download the updated encoder model from the network device for subsequent CSI compression feedback.
  • the embodiment of the present application does not limit the network implementation of the encoder and decoder, and may include at least one network structure such as DNN, CNN, and recurrent neural network (eg, LSTM).
  • model updates can be performed according to the technical solutions of the embodiments of the present application.
  • the encoder and decoder are implemented by CNN, and the terminal device can encode the channel data as an image to be compressed through CNN to obtain the target bit stream, and further send the target bit stream to the network device.
  • the network device uses the CNN to use the target bit stream as information obtained by encoding an image to decode the target bit stream to obtain target channel data.
  • the encoder and decoder are implemented using a cyclic neural network
  • the terminal device can encode each CSI vector in the channel data as an element of the sequence through the cyclic neural network to obtain the target bit stream, and further the target bit stream sent to network devices.
  • the network device decodes the target bit stream by using the target bit stream as the information obtained by encoding the sequence through the cyclic neural network to obtain the target channel data.
  • FIG. 8 is a schematic interaction diagram of an example of migration training performed by a network device provided by an embodiment of the present application. As shown in Figure 8, the following steps may be included:
  • the terminal device monitors the channel quality index
  • the terminal device continuously or periodically monitors the channel quality indicator, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the terminal device continuously or periodically monitors the channel quality indicator, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the terminal device sends second indication information to the network device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device reports the target training data set and the currently used encoder model to the network device.
  • the target training data set includes channel data corresponding to the changed channel scene, and for the specific implementation of the target training data set, refer to the relevant description of the foregoing embodiments.
  • the terminal device may directly report the current encoder model to the network device without sending the second indication information to the network device, and the network device may receive the encoder model reported by the terminal device.
  • the network device may receive the encoder model reported by the terminal device.
  • it may be determined that a model update is required.
  • the network device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and decoder model.
  • the terminal device downloads the updated encoder model from the network device, or the network device sends the updated encoder model to the terminal device.
  • the updated model can be adapted to the changed channel scene, and the network device and the terminal device can perform CSI compression feedback based on the updated model when the channel scene does not change.
  • the terminal device continues to monitor the channel quality index to determine whether the channel scene is changed. If the channel is changed, the above steps may be performed to implement online update of the model.
  • FIG. 9 is another schematic interaction diagram of a network device performing migration training provided by an embodiment of the present application. As shown in Figure 9, the following steps may be included:
  • the network device monitors the channel quality index
  • the network device monitors the channel quality indicator continuously or periodically, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the network device monitors the channel quality indicator continuously or periodically, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the network device sends first indication information to the terminal device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device reports the target training data set and the currently used encoder model to the network device.
  • the terminal device when receiving the first indication information, reports the target training data set and the currently used encoder model to the network device.
  • the target training data set includes channel data corresponding to the changed channel scene, and for specific implementation, refer to relevant descriptions of the foregoing embodiments.
  • the network device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and decoder model.
  • the terminal device downloads the updated encoder model from the network device, or the network device sends the updated encoder model to the terminal device.
  • the updated model can be adapted to the changed channel scene, and the network device and the terminal device can perform CSI compression feedback based on the updated model when the channel scene does not change.
  • the network device continues to monitor the channel quality index to determine whether the channel scene is changed. If the channel is changed, the above steps may be performed to implement online update of the model.
  • FIG. 10 is another schematic interaction diagram of a network device performing migration training provided by an embodiment of the present application. As shown in Figure 10, the following steps may be included:
  • the terminal device performs scene monitoring.
  • a terminal device may perform scene monitoring by monitoring channel quality indicators.
  • the terminal device may monitor the channel quality indicator continuously or periodically, and determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the terminal device may monitor the channel quality indicator continuously or periodically, and determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the terminal device sends second indication information to the network device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device sends the third indication information and the currently used encoder model to the network device.
  • the third indication information is used to indicate the changed channel scenario.
  • the target training data set used by the network device for migration training may be determined according to the pre-stored data set of the network device.
  • the specific determination method refer to the relevant description in the previous embodiment, and for the sake of brevity, details are not repeated here.
  • the terminal device may not send the second indication information to the network device, but directly send the third indication information and the current encoder model to the network device, and the network device may 3. Indicate the information and/or the encoder model reported by the terminal device, and determine that a model update is required.
  • the network device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and a decoder model.
  • the terminal device downloads the updated encoder model from the network device, or the network device sends the updated encoder model to the terminal device.
  • the updated model can be adapted to the changed channel scene, and the network device and the terminal device can perform CSI compression feedback based on the updated model when the channel scene does not change.
  • the terminal device continues to monitor the scene to determine whether the channel scene changes, and if the channel changes, the above steps can be performed to implement online update of the model.
  • FIG. 11 is a schematic interaction diagram of yet another example of migration training performed by a network device provided in an embodiment of the present application. As shown in Figure 11, the following steps may be included:
  • the network device performs scene monitoring.
  • a network device can perform scenario monitoring by monitoring channel quality indicators.
  • the network device may continuously or periodically monitor the channel quality indicator, and determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the network device sends first indication information to the terminal device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device sends the currently used encoder model to the network device.
  • the target training data set used by the network device for migration training may be determined according to the pre-stored data set of the network device.
  • the specific determination method refer to the relevant description in the previous embodiment, and for the sake of brevity, details are not repeated here.
  • the network device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and a decoder model.
  • the terminal device downloads the updated encoder model from the network device, or the network device sends the updated encoder model to the terminal device.
  • the updated model can be adapted to the changed channel scene, and the network device and the terminal device can perform CSI compression feedback based on the updated model when the channel scene does not change.
  • the network device continues to monitor the scene to determine whether the channel scene changes, and if the channel changes, the above steps can be performed to implement online update of the model.
  • the network device can update the encoder model and decoder model online according to the data set corresponding to the changed channel scene according to the change of the wireless channel scene, and support the target training data set, encoding
  • the air interface transmission of the decoder model and the decoder model enables the updated model to adapt to the changed channel environment, thereby achieving better CSI feedback and restoration performance.
  • FIG. 12 is a schematic interaction diagram of another wireless communication method 1000 provided according to an embodiment of the present application. As shown in FIG. 12, the method 1000 includes the following content:
  • the terminal device performs transfer training on the first encoding network and the first decoding network according to the target training data set, to obtain the second encoding network and the second decoding network.
  • the encoding network in the embodiment of the present application may also be called an encoder, and the decoding network may also be called a decoder.
  • the encoder and decoder can constitute a CSI autoencoder, or in other words, a channel information feedback system. That is, the first encoding network may correspond to an encoder in the CSI autoencoder, and the first decoding network may correspond to a decoder in the CSI autoencoder.
  • the encoding network may be deployed on terminal devices, and the decoding network may be deployed on network devices.
  • the terminal device can encode the channel data through the encoding network to obtain the target bit stream, and further send the target bit stream to the network device, and the network device decodes the target bit stream through the decoding network to obtain the target channel data.
  • the first encoding network and the first decoding network correspond to a first channel scenario, or in other words, a first channel environment. In other words, the first encoding network and the first decoding network are adapted to the first channel scenario.
  • adapting the first encoding network and the first decoding network to the first channel scenario may refer to:
  • the terminal device encodes the channel data based on the first encoding network to obtain the target bit stream, and the network device decodes the target bit stream based on the first encoding network to obtain the target channel data, which can achieve relatively Excellent CSI feedback and recovery performance.
  • the autoencoder model based on the first encoding network and the first decoding network can achieve better CSI feedback and recovery performance.
  • the terminal device can perform migration training on the currently used model based on the channel data set corresponding to the changed channel scene, so that the model after migration training can adapt to the changed channel scenarios to improve CSI feedback and recovery performance.
  • the terminal device can perform migration training on the first encoding network and the first decoding network according to the target training data set to obtain the second encoding network and the second A decoding network, wherein the target training data set includes channel data corresponding to a second channel scenario.
  • the method 1000 also includes:
  • the terminal device sends the model parameters of the second decoding network to the network device.
  • the terminal device may send the trained model parameters of the second decoding network to the network device.
  • the method 1000 also includes:
  • the terminal device may encode the channel data in the second channel scenario through the second encoding network to obtain the target bit stream;
  • the terminal device sends the target bit stream to the network device.
  • the network device may decode the target bit stream through the second decoding network to obtain target channel data.
  • the terminal device can update the currently used model based on the channel data set corresponding to the changed channel scene, so that the network device and the terminal device can use the model adapted to the changed channel scene.
  • the model compresses and feeds back the channel data in the changed channel scenario, which is beneficial to improve the CSI feedback and recovery performance.
  • the network device may send second information to the terminal device, where the second information is used to assist the terminal device in performing model migration training.
  • the change of the channel scene may be determined by the network device, or may also be determined by the terminal device.
  • the network device can monitor the channel quality indicator to determine whether the channel scene changes. For example, the network device may continuously or periodically detect the channel quality indicator, and further determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the channel quality indicator may include at least one of the following: CSI, RSRP, RSRQ, RSSI, and SINR.
  • the network device determines that the channel scene is changed when the change amount of the channel quality index is greater than the first threshold.
  • the network device may determine that the channel scene is changed when the change amount of the RSRP is greater than the first RSRP threshold.
  • different channel scenarios may correspond to corresponding channel quality indicator ranges, and the network device may determine the changed channel scenario according to the changed channel quality indicator and the channel quality indicator range.
  • channel scene change in the embodiment of the present application may include any scene that causes model incompatibility, for example, the change amount of the channel quality index exceeds a certain threshold, or the terminal device Moving from one cell to another etc., the application is not limited thereto.
  • the method 200 further includes:
  • the network device sends first indication information to the terminal device, where the first indication information is used to indicate that a model is updated and/or a channel scene is changed, where the model update includes a model update of the encoding network and/or Model updates for the decoding network.
  • the second information may include the first indication information.
  • the first indication information may be sent through any downlink message or signaling, for example, the first indication information is sent through DCI.
  • the terminal device may monitor the channel quality indicator to determine whether the channel scene changes. For example, the terminal device may continuously or periodically detect the channel quality indicator, and further determine whether the channel scene is changed according to the change of the channel quality indicator.
  • the manner in which the terminal device monitors the channel quality indicator may refer to the relevant implementation of the network device, and for the sake of brevity, details are not repeated here.
  • the method 200 further includes:
  • the terminal device sends second indication information to the network device, where the second indication information is used to indicate that a model is updated and/or a channel scene is changed, where the model update includes a model update of the coding network and/or Model updates for the decoding network.
  • the second indication information may be sent through any uplink message or signaling, for example, the second indication information is sent through UCI.
  • the method 200 further includes:
  • the terminal device acquires the model parameters of the first decoding network from the network device.
  • the terminal device acquires the decoder model currently used by the network device from the network device when receiving the first indication information sent by the network device.
  • the terminal device determines that the channel scene changes by monitoring the channel quality index, it acquires the decoder model currently used by the network device from the network device.
  • the network device may send the currently used decoder model to the terminal device when it detects that the channel scene changes, or when receiving the second indication information sent by the terminal device.
  • the second information includes model parameters of the first decoding network.
  • the first instruction information may not be sent to the terminal device, but the currently used encoder model may be directly sent to the terminal device, and the terminal device may use the coder model sent by the network device to The model determines that a model update is required.
  • the data set on which the terminal device performs migration training can be obtained in real time from the terminal device, or it can also be pre-stored on the terminal device.
  • the method of obtaining the data set in this application Not limited.
  • the terminal device collects the target training data set when receiving the first indication information sent by the network device. For another example, the terminal device collects the target training data set when it determines that the channel scene changes by monitoring the channel quality index.
  • the target training data set includes a CSI vector on at least one frequency domain unit within a first time period after the first channel scene is changed to the second channel scene.
  • the unit of the first duration may be a subframe, a time slot, a symbol, or other time units, which is not limited in the present application.
  • the first duration includes N time slots, where N is a positive integer.
  • the N is predefined or configured by the network device, for example, the network device can configure the parameter N through DCI.
  • the frequency domain unit may be a subband, a physical resource block (Physical Resource Block, PRB), or other frequency domain units, which are not limited in this application.
  • PRB Physical Resource Block
  • the at least one frequency domain unit includes K subbands, where K is a positive integer.
  • the K is predefined, or configured by the network device, for example, the network device can configure the parameter K through DCI.
  • the at least one frequency domain unit includes M PRBs, where M is a positive integer.
  • the M is predefined or configured by the network device, for example, the network device can configure the parameter M through DCI.
  • the target training data set is used for model migration training, in order to meet the requirements of online update of the model, higher CSI feedback accuracy is required, and the CSI vector in the target training data set can be obtained based on the first codebook quantization, wherein, the first codebook has higher precision, for example, the precision of the first codebook is higher than the precision of the Type 2 codebook.
  • the first codebook may be a dedicated codebook for the data set for model migration training, that is, only use this codebook for feedback on the data set for model migration training.
  • the terminal device may use higher-precision quantization bits to quantize the fed-back amplitude and phase to obtain the target training data set.
  • the configuration of the first codebook is sent by the network device through DCI.
  • the target training data set is determined according to at least one data set in multiple data sets prestored on the terminal device.
  • the terminal device may pre-store data sets corresponding to multiple channel scenarios, such as LoS scenarios, NLoS scenarios, indoor scenarios, outdoor scenarios, low-speed moving scenarios, high-speed moving scenarios and other scenarios.
  • the terminal device may determine a target training data set from the multiple pre-stored data sets according to the changed channel scenario.
  • the terminal device may determine samples in the data set corresponding to the changed channel scene as the target training data set.
  • the target training data set includes samples in multiple data sets, where each data set corresponds to a corresponding weight, and the weight corresponding to each data set is based on the applicable channel scenario of the data set and the Correlation determination of the above-mentioned changed channel scene. For example, a data set corresponding to a channel scene with a large correlation with the changed channel scene can be given a larger weight, and a data set corresponding to a channel scene with a small correlation with the changed channel scene can be given a smaller weight . That is, the samples with high matching degree of channel scene are given higher weight, and the samples with low matching degree of channel scene are given lower weight.
  • the changed channel scenario may be known by the terminal device by monitoring the channel quality index, or it may be determined according to the fourth indication information sent by the network device, where the fourth indication information is used to indicate the changed channel scenario, wherein the network device can determine the changed channel scenario according to the changed channel quality index.
  • the second information sent by the network device to the terminal device for assisting the terminal device in performing model migration training may include at least one of the foregoing first indication information, the first decoding network, and the fourth indication information.
  • the terminal device can know the encoder model currently used by the terminal device and the decoder model currently used by the network device, as well as the target training data set used for model migration training. Further, the terminal device can perform transfer training on the currently used encoder model and decoder model based on the target training data set. For example, by fine-tuning the model parameters, as an example, the parameters of some layers in the network structure are fixed, and only the parameters of other layers are adjusted. This application does not limit the specific adjustment method. Finally, the updated encoder model and decoder model converge on the target training data set.
  • the training rounds on the target training data set during migration training and the training loss function used for migration training, and the migration training parameters such as the training optimizer can be configured by the network device, or can be is predefined.
  • the terminal device After the terminal device completes the online update of the model, the terminal device sends the updated model parameters of the decoder to the network device, that is, the terminal device can send the model parameters of the second decoding network obtained through migration training to the network device.
  • the network device may download the updated decoder model from the terminal device side.
  • the embodiment of the present application does not limit the network implementation of the encoder and decoder, and may include at least one network structure such as DNN, CNN, and recurrent neural network (eg, LSTM).
  • the model can be updated according to the method described in this application.
  • the encoder and decoder are implemented using CNN, and the terminal device can encode the channel data as an image to be compressed through CNN to obtain the target bit stream.
  • the network device uses CNN to encode the target bit stream as The information obtained by encoding the image is used to decode the target bit stream to obtain target channel data.
  • the encoder and decoder are implemented using a cyclic neural network
  • the terminal device can encode each CSI vector in the channel data as an element of the sequence through the cyclic neural network to obtain a target bit stream.
  • the network device decodes the target bit stream by using the target bit stream as the information obtained by encoding the sequence through the cyclic neural network to obtain the target channel data.
  • FIG. 13 is a schematic interaction diagram of an example of migration training performed by a terminal device according to an embodiment of the present application. As shown in Figure 13, the following steps may be included:
  • the terminal device monitors a channel quality index.
  • the terminal device monitors the channel quality indicator continuously or periodically, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the terminal device sends second indication information to the network device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device acquires the decoder model currently used by the network device from the network device.
  • the network device may send the currently used decoder model to the terminal device.
  • the terminal device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and decoder model.
  • the terminal device sends the updated decoder model to the network device.
  • the terminal device can upload the updated decoder model to the network device.
  • the updated model can be adapted to the changed channel scene, and when the channel scene does not change, the network device and the terminal device can perform CSI compression feedback based on the updated model.
  • the terminal device continues to monitor the channel quality index to determine whether the channel scene is changed. If the channel is changed, the above steps may be performed to implement online update of the model.
  • FIG. 14 is a schematic interaction diagram of an example of migration training performed by a terminal device according to an embodiment of the present application. As shown in Figure 14, the following steps may be included:
  • the network device monitors the channel quality index.
  • the network device monitors the channel quality indicator continuously or periodically, and determines whether the channel scene is changed according to the change of the channel quality indicator.
  • the network device sends first indication information to the terminal device, which is used to indicate that the model is updated and/or the channel scene is changed.
  • the terminal device acquires the currently used decoder model from the network device.
  • the network device when the network device determines that the channel scene changes, it may directly send the currently used decoder model to the terminal device without sending the first indication information, and the terminal device , it can be determined that a model update is required.
  • the terminal device performs joint migration training on the encoder model currently used by the terminal device and the decoder model currently used by the network device according to the target training data set, to obtain an updated encoder model and decoder model.
  • the terminal device sends the updated decoder model to the network device.
  • the updated model can be adapted to the changed channel scene, and when the channel scene does not change, the network device and the terminal device can perform CSI compression feedback based on the updated model.
  • the network device continues to monitor the channel quality index to determine whether the channel scene is changed. If the channel is changed, the above steps may be performed to implement online update of the model.
  • the terminal device can update the encoder model and decoder model online according to the data set corresponding to the changed channel scene according to the change of the wireless channel scene, and support the target training data set, encoding
  • the air interface transmission of the decoder model and the decoder model enables the updated model to adapt to the changed channel environment, thereby achieving better CSI feedback and restoration performance.
  • Fig. 15 is a schematic block diagram of a network device according to an embodiment of the present application.
  • the network device 1100 of Figure 15 includes:
  • the processing unit 1110 is configured to, when the channel scene is changed from the first channel scene to the second channel scene, according to the target training data set, the first encoding network deployed on the terminal device and the first decoding network deployed on the network device.
  • the network performs migration training to obtain a second encoding network and a second decoding network, wherein the target training data set includes channel data in the second channel scenario, and the first encoding network and the first decoding network are adapted to Configure the first channel scenario.
  • processing unit 1110 is further configured to:
  • processing unit 1110 is further configured to:
  • the change amount of the channel quality index is greater than the first threshold, it is determined that the channel scene is changed.
  • the network device 1110 further includes:
  • a communication unit configured to send first indication information to the terminal device, where the first indication information is used to indicate that a model is updated and/or a channel scene is changed, wherein the model update includes an encoding network model update and/or Or model updates for decoding networks.
  • the first indication information is sent through downlink control information DCI.
  • the network device 1100 further includes:
  • a communication unit configured to receive second indication information sent by the terminal device, where the second indication information is used to indicate that a model is updated and/or a channel scene is changed, where the model update includes a model update of the coding network and /or model updates for the decoding network.
  • the second indication information is sent through uplink control information UCI.
  • the network device 1100 further includes:
  • a communication unit configured to acquire the model parameters of the first encoding network from the terminal device.
  • the network device 1100 further includes:
  • a communication unit configured to receive the target training data set reported by the terminal device.
  • the target training data set includes channel state information CSI vectors on at least one frequency domain unit within a first time period after the first channel scene is changed to the second channel scene.
  • the CSI vector is obtained through quantization based on the first codebook, and the precision of the first codebook is higher than the precision of the type 2 codebook.
  • the configuration of the first codebook is sent by the network device through downlink control information DCI.
  • the first duration includes N time slots
  • the at least one frequency domain unit includes K subbands or M PRBs, where N is a positive integer, and the K is a positive integer, and the M is a positive integer.
  • the N is predefined or configured by the network device
  • the K is predefined or configured by the network device
  • the M is predefined or configured by the network device.
  • processing unit 1110 is further configured to:
  • the target training data set is determined in multiple prestored data sets, where each data set in the multiple data sets corresponds to a corresponding channel scene.
  • the network device 1100 further includes:
  • a communication unit configured to receive third indication information sent by the terminal device, where the third indication information is used to indicate a changed channel scenario.
  • the target training data set only includes the data set corresponding to the second channel scene
  • the target training data set includes a plurality of data sets, wherein each data set corresponds to a corresponding weight, and the weight corresponding to each data set is based on the correlation between the channel scene applicable to the data set and the second channel scene gender certainty.
  • the network device 1100 further includes:
  • a communication unit configured to send the model parameters of the second encoding network obtained through migration training to the terminal device.
  • the network device 1100 further includes:
  • a communication unit configured to receive a target bit stream sent by the terminal device, where the target bit stream is obtained by the terminal device encoding channel data in the second channel scenario through the second encoding network.
  • processing unit 1110 is further configured to:
  • the target bit stream is decoded by the second decoding network to obtain target channel data.
  • the above-mentioned communication unit may be a communication interface or a transceiver, or an input-output interface of a communication chip or a system-on-chip.
  • the aforementioned processing unit may be one or more processors.
  • the network device 1100 may correspond to the network device in the method embodiment of the present application, and the above-mentioned and other operations and/or functions of each unit in the network device 1100 are to realize the For the sake of brevity, the corresponding flow of the network device in the shown method 200 is not repeated here.
  • Fig. 16 shows a schematic block diagram of a terminal device 1200 according to an embodiment of the present application.
  • the terminal device 1200 includes:
  • the processing unit 1210 is configured to, when the channel scene is changed from the first channel scene to the second channel scene, according to the target training data set, the first encoding network deployed on the terminal device and the first decoding network deployed on the network device.
  • the network performs migration training to obtain a second encoding network and a second decoding network, wherein the target training data set includes channel data in the second channel scenario, and the first encoding network and the first decoding network are adapted to Configure the first channel scenario.
  • processing unit 1210 is further configured to:
  • processing unit 1210 is further configured to:
  • the terminal device 1200 further includes:
  • a communication unit configured to send second indication information to the network device, where the second indication information is used to indicate that a model is updated and/or a channel scene is changed, where the model update includes an encoding network model update and/or Or model updates for decoding networks.
  • the second indication information is sent through uplink control information UCI.
  • the terminal device 1200 further includes:
  • a communication unit configured to receive the first instruction information sent by the network device, for instructing to update the model and/or change the channel scene, wherein the update of the model includes an update of the model of the encoding network and/or a model of the decoding network renew.
  • the first indication information is sent through downlink control information DCI.
  • the target training data set includes channel state information CSI vectors on at least one frequency domain unit within a first time period after the first channel scene is changed to the second channel scene.
  • the CSI vector is obtained through quantization based on the first codebook, and the precision of the first codebook is higher than the precision of the type 2 codebook.
  • the configuration of the first codebook is sent by the network device through downlink control information DCI.
  • the first duration includes N time slots
  • the at least one frequency domain unit includes K subbands or M PRBs, where N is a positive integer, and the K is a positive integer, and the M is a positive integer.
  • the N is predefined or configured by the network device
  • the K is predefined or configured by the network device
  • the M is predefined or configured by the network device.
  • the terminal device 1200 further includes:
  • a communication unit configured to acquire model parameters of the first decoding network from the network device.
  • the terminal device 1200 further includes:
  • a communication unit configured to send the model parameters of the second decoding network obtained through migration training to the network device.
  • processing unit 1200 is also used to:
  • the channel data in the second channel scenario is encoded by the second encoding network to obtain a target bit stream.
  • the terminal device 1200 further includes: a communication unit, configured to send the target bit stream to the network device.
  • the above-mentioned communication unit may be a communication interface or a transceiver, or an input-output interface of a communication chip or a system-on-chip.
  • the aforementioned processing unit may be one or more processors.
  • terminal device 1200 may correspond to the terminal device in the method embodiment of the present application, and the above-mentioned and other operations and/or functions of each unit in the terminal device 1200 are to realize the For the sake of brevity, the corresponding process of the terminal device in the shown method 200 will not be repeated here.
  • Fig. 17 shows a schematic block diagram of a terminal device 1300 according to an embodiment of the present application. As shown in Figure 17, the terminal device 1300 includes:
  • the communication unit 1310 is configured to send first information to the network device, where the first information is used for the network device to update the first encoding network deployed on the terminal device and the network device on the network device when the channel scene changes.
  • the deployed first decoding network performs migration training to obtain a second encoding network and a second decoding network, wherein the first encoding network and the first decoding network adapt to the channel scene before the change, and the second encoding network Adapting to the changed channel scenario with the second decoding network.
  • the terminal device 1310 further includes: a processing unit configured to monitor a channel quality indicator, and determine whether a channel scene is changed according to a change amount of the channel quality indicator.
  • the processing unit is also used for:
  • the first information includes second indication information
  • the second indication information is used to indicate that the model is updated and/or the channel scene is changed, wherein the model update includes the encoding network Model update and/or model update of the decoding network.
  • the second indication information is sent through uplink control information UCI.
  • the communication unit 1310 is also used to:
  • the network device receiving first instruction information sent by the network device, where the first instruction information is used to indicate model update and/or channel scene change, where the model update includes model update of the encoding network and/or model update of the decoding network Model updates.
  • the first indication information is sent through downlink control information DCI.
  • the first information includes model parameters of the first encoding network.
  • the first information includes a target training data set
  • the target training data set includes channel data in a changed channel scenario
  • the target training data set includes channel state information CSI vectors on at least one frequency domain unit within the first time period after the channel scene is changed.
  • the CSI vector is obtained by quantization based on the first codebook, and the precision of the first codebook is higher than that of the type 2 codebook.
  • the configuration of the first codebook is sent by the network device through DCI.
  • the first duration includes N time slots
  • the at least one frequency domain unit includes K subbands or M PRBs, where N is a positive integer, and the K is a positive integer, and the M is a positive integer.
  • the N is predefined or configured by the network device
  • the K is predefined or configured by the network device
  • the M is predefined or configured by the network device.
  • the first information includes third indication information
  • the third indication information is used to indicate a changed channel scenario.
  • the communication unit 1310 is also used to:
  • the terminal device 1300 further includes: a processing unit, configured to use the second encoding network to encode the channel data in the second channel scenario to obtain a target bit stream.
  • the communication unit 1310 is also used to:
  • the above-mentioned communication unit may be a communication interface or a transceiver, or an input-output interface of a communication chip or a system-on-chip.
  • the aforementioned processing unit may be one or more processors.
  • terminal device 1300 may correspond to the terminal device in the method embodiment of the present application, and the above-mentioned and other operations and/or functions of each unit in the terminal device 1300 are to realize the For the sake of brevity, the corresponding process of the terminal device in the shown method 1000 is not repeated here.
  • Fig. 18 shows a schematic block diagram of a network device 1800 according to an embodiment of the present application.
  • the network device 1800 includes:
  • the communication unit 1810 is configured to send second information to the terminal device, where the second information is used for the terminal device to update the first encoding network deployed on the terminal device and the network when the channel scene changes.
  • the first decoding network deployed on the device performs migration training to obtain a second encoding network and a second decoding network, wherein the first encoding network and the first decoding network adapt to the channel scene before the change, and the second The encoding network and the second decoding network are adapted to the changed channel scenario.
  • the network device 1800 further includes: a processing unit configured to monitor a channel quality indicator, and determine whether a channel scene is changed according to a change amount of the channel quality indicator.
  • the second information includes first indication information
  • the first indication information is used to indicate that a model is updated and/or a channel scene is changed
  • the model update includes a model of the coding network Update and/or model updates for the decoding network.
  • the first indication information is sent through downlink control information DCI.
  • the second information includes model parameters of the first decoding network.
  • the communication unit 1810 is also used to:
  • the network device 1800 further includes: a processing unit, configured to decode the target bit stream through the second decoding network to obtain target channel data, wherein the target bit stream is the It is obtained by the terminal device by encoding the channel data in the second channel scenario through the second encoding network.
  • the above-mentioned communication unit may be a communication interface or a transceiver, or an input-output interface of a communication chip or a system-on-chip.
  • the aforementioned processing unit may be one or more processors.
  • the network device 1800 may correspond to the network device in the method embodiment of the present application, and the above-mentioned and other operations and/or functions of each unit in the network device 1800 are to realize the For the sake of brevity, the corresponding flow of the network device in the shown method 1000 is not repeated here.
  • FIG. 19 is a schematic structural diagram of a communication device 1400 provided by an embodiment of the present application.
  • the communication device 1400 shown in FIG. 19 includes a processor 1410, and the processor 1410 can call and run a computer program from a memory, so as to implement the method in the embodiment of the present application.
  • the communication device 1400 may further include a memory 1420 .
  • the processor 1410 can invoke and run a computer program from the memory 1420, so as to implement the method in the embodiment of the present application.
  • the memory 1420 may be an independent device independent of the processor 1410 , or may be integrated in the processor 1410 .
  • the communication device 1400 may further include a transceiver 1430, and the processor 1410 may control the transceiver 1430 to communicate with other devices, specifically, to send information or data to other devices, or to receive other Information or data sent by the device.
  • the transceiver 1430 may include a transmitter and a receiver.
  • the transceiver 1430 may further include antennas, and the number of antennas may be one or more.
  • the communication device 1400 may specifically be the network device of the embodiment of the present application, and the communication device 1400 may implement the corresponding processes implemented by the network device in each method of the embodiment of the present application. For the sake of brevity, details are not repeated here. .
  • the communication device 1400 may specifically be the mobile terminal/terminal device of the embodiment of the present application, and the communication device 1400 may implement the corresponding processes implemented by the mobile terminal/terminal device in each method of the embodiment of the present application, for the sake of brevity , which will not be repeated here.
  • FIG. 20 is a schematic structural diagram of a chip according to an embodiment of the present application.
  • the chip 1500 shown in FIG. 20 includes a processor 1510, and the processor 1510 can call and run a computer program from a memory, so as to implement the method in the embodiment of the present application.
  • the chip 1500 may further include a memory 1520 .
  • the processor 1510 can invoke and run a computer program from the memory 1520, so as to implement the method in the embodiment of the present application.
  • the memory 1520 may be an independent device independent of the processor 1510 , or may be integrated in the processor 1510 .
  • the chip 1500 may also include an input interface 1530 .
  • the processor 1510 can control the input interface 1530 to communicate with other devices or chips, specifically, can obtain information or data sent by other devices or chips.
  • the chip 1500 may also include an output interface 1540 .
  • the processor 1510 can control the output interface 1540 to communicate with other devices or chips, specifically, can output information or data to other devices or chips.
  • the chip can be applied to the network device in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the network device in the methods of the embodiment of the present application.
  • the chip can implement the corresponding processes implemented by the network device in the methods of the embodiment of the present application.
  • the chip can be applied to the mobile terminal/terminal device in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the mobile terminal/terminal device in the various methods of the embodiments of the present application.
  • the chip can implement the corresponding processes implemented by the mobile terminal/terminal device in the various methods of the embodiments of the present application.
  • the chip can implement the corresponding processes implemented by the mobile terminal/terminal device in the various methods of the embodiments of the present application.
  • the chip can be applied to the mobile terminal/terminal device in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the mobile terminal/terminal device in the various methods of the embodiments of the present application.
  • the chip mentioned in the embodiment of the present application may also be called a system-on-chip, a system-on-chip, a system-on-a-chip, or a system-on-a-chip.
  • Fig. 21 is a schematic block diagram of a communication system 900 provided by an embodiment of the present application. As shown in FIG. 21 , the communication system 900 includes a terminal device 910 and a network device 920 .
  • the terminal device 910 can be used to realize the corresponding functions realized by the terminal device in the above method
  • the network device 920 can be used to realize the corresponding functions realized by the network device in the above method.
  • the processor in the embodiment of the present application may be an integrated circuit chip, which has a signal processing capability.
  • each step of the above-mentioned method embodiments may be completed by an integrated logic circuit of hardware in a processor or instructions in the form of software.
  • the above-mentioned processor can be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf programmable gate array (Field Programmable Gate Array, FPGA) or other available Program logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field Programmable Gate Array
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register.
  • the storage medium is located in the memory, and the processor reads the information in the memory, and completes the steps of the above method in combination with its hardware.
  • the memory in the embodiments of the present application may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories.
  • 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), electronically programmable Erase Programmable Read-Only Memory (Electrically EPROM, EEPROM) or Flash.
  • the volatile memory can be Random Access Memory (RAM), which acts as external cache memory.
  • RAM Static Random Access Memory
  • SRAM Static Random Access Memory
  • DRAM Dynamic Random Access Memory
  • Synchronous Dynamic Random Access Memory Synchronous Dynamic Random Access Memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM, DDR SDRAM enhanced synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM synchronous connection dynamic random access memory
  • Synchlink DRAM, SLDRAM Direct Memory Bus Random Access Memory
  • Direct Rambus RAM Direct Rambus RAM
  • the memory in the embodiment of the present application may also be a static random access memory (static RAM, SRAM), a 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, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection Dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is, the memory in the embodiments of the present application is intended to include, but not be limited to, these and any other suitable types of memory.
  • the embodiment of the present application also provides a computer-readable storage medium for storing computer programs.
  • the computer-readable storage medium can be applied to the network device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the network device in the methods of the embodiments of the present application.
  • the computer program enables the computer to execute the corresponding processes implemented by the network device in the methods of the embodiments of the present application.
  • the computer-readable storage medium can be applied to the mobile terminal/terminal device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the mobile terminal/terminal device in the various methods of the embodiments of the present application , for the sake of brevity, it is not repeated here.
  • the embodiment of the present application also provides a computer program product, including computer program instructions.
  • the computer program product may be applied to the network device in the embodiment of the present application, and the computer program instructions cause the computer to execute the corresponding process implemented by the network device in each method of the embodiment of the present application.
  • the Let me repeat for the sake of brevity, the Let me repeat.
  • the computer program product can be applied to the mobile terminal/terminal device in the embodiments of the present application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the mobile terminal/terminal device in the methods of the embodiments of the present application, For the sake of brevity, details are not repeated here.
  • the embodiment of the present application also provides a computer program.
  • the computer program can be applied to the network device in the embodiment of the present application.
  • the computer program executes the corresponding process implemented by the network device in each method of the embodiment of the present application.
  • the computer program executes the corresponding process implemented by the network device in each method of the embodiment of the present application.
  • the computer program can be applied to the mobile terminal/terminal device in the embodiment of the present application.
  • the computer program executes each method in the embodiment of the present application to be implemented by the mobile terminal/terminal device
  • the corresponding process will not be repeated here.
  • the disclosed systems, devices and methods may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

Landscapes

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

Abstract

Procédé et dispositif de communication sans fil qui réalisent une mise à jour adaptative d'un réseau de codage et d'un réseau de décodage lorsqu'un scénario de canal change, améliorent les capacités de généralisation adaptative du réseau de codage et du réseau de décodage, et facilitent l'amélioration de la précision de rétroaction de compression de retour d'informations de canal lorsque des caractéristiques d'environnement de canal changent. Le procédé consiste : lorsque le scénario de canal change d'un premier scénario de canal à un second scénario de canal, à effectuer une formation par transfert, en fonction d'un ensemble de données de formation cible, sur un premier réseau de codage déployé sur un dispositif terminal et un premier réseau de décodage déployé sur un dispositif de réseau pour obtenir un second réseau de codage et un second réseau de décodage, l'ensemble de données de formation cible comprenant des données de canal dans le second scénario de canal, et le premier réseau de codage et le premier réseau de décodage s'adaptant au premier scénario de canal.
PCT/CN2021/112118 2021-08-11 2021-08-11 Procédé et dispositif de communication sans fil WO2023015499A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202180097891.5A CN117322044A (zh) 2021-08-11 2021-08-11 无线通信的方法和设备
PCT/CN2021/112118 WO2023015499A1 (fr) 2021-08-11 2021-08-11 Procédé et dispositif de communication sans fil

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/112118 WO2023015499A1 (fr) 2021-08-11 2021-08-11 Procédé et dispositif de communication sans fil

Publications (1)

Publication Number Publication Date
WO2023015499A1 true WO2023015499A1 (fr) 2023-02-16

Family

ID=85200480

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/112118 WO2023015499A1 (fr) 2021-08-11 2021-08-11 Procédé et dispositif de communication sans fil

Country Status (2)

Country Link
CN (1) CN117322044A (fr)
WO (1) WO2023015499A1 (fr)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014077758A1 (fr) * 2012-11-14 2014-05-22 Telefonaktiebolaget L M Ericsson (Publ) Systèmes et procédés pour améliorer des propriétés de transmission en liaison montante dans un réseau de communication
CN108989880A (zh) * 2018-06-21 2018-12-11 北京邮电大学 一种码率自适应切换方法及***
CN110891292A (zh) * 2019-11-20 2020-03-17 三星电子（中国）研发中心 一种终端自动切换网络方法和装置
CN111556453A (zh) * 2020-04-27 2020-08-18 南京邮电大学 一种基于信道状态信息和BiLSTM的多场景室内动作识别方法
CN112135325A (zh) * 2020-09-30 2020-12-25 Oppo广东移动通信有限公司 网络切换方法、装置、存储介质以及终端
CN112233160A (zh) * 2020-10-15 2021-01-15 杭州知路科技有限公司 一种基于双目摄像头的实时深度及置信度的预测方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014077758A1 (fr) * 2012-11-14 2014-05-22 Telefonaktiebolaget L M Ericsson (Publ) Systèmes et procédés pour améliorer des propriétés de transmission en liaison montante dans un réseau de communication
CN108989880A (zh) * 2018-06-21 2018-12-11 北京邮电大学 一种码率自适应切换方法及***
CN110891292A (zh) * 2019-11-20 2020-03-17 三星电子（中国）研发中心 一种终端自动切换网络方法和装置
CN111556453A (zh) * 2020-04-27 2020-08-18 南京邮电大学 一种基于信道状态信息和BiLSTM的多场景室内动作识别方法
CN112135325A (zh) * 2020-09-30 2020-12-25 Oppo广东移动通信有限公司 网络切换方法、装置、存储介质以及终端
CN112233160A (zh) * 2020-10-15 2021-01-15 杭州知路科技有限公司 一种基于双目摄像头的实时深度及置信度的预测方法

Also Published As

Publication number Publication date
CN117322044A (zh) 2023-12-29

Similar Documents

Publication Publication Date Title
US20230019669A1 (en) Systems and methods for enhanced feedback for cascaded federated machine learning
WO2021217519A1 (fr) Procédé et appareil de réglage de réseau neuronal
WO2022001822A1 (fr) Procédé et dispositif pour acquérir un réseau de neurones artificiels
CN116419257A (zh) 一种通信方法及装置
CN116569527A (zh) 信道信息反馈的方法、发送端设备和接收端设备
WO2022236785A1 (fr) Procédé de rétroaction d'informations de canal, dispositif d'extrémité de réception et dispositif d'extrémité de transmission
CN114492784A (zh) 神经网络的测试方法和装置
WO2023015499A1 (fr) Procédé et dispositif de communication sans fil
WO2023004638A1 (fr) Procédés de rétroaction d'informations de canal, dispositifs d'extrémité d'émission et dispositifs d'extrémité de réception
US20240259072A1 (en) Model processing method, electronic device, network device, and terminal device
CN117255996A (zh) 信号处理方法、通信设备及通信***
WO2023004563A1 (fr) Procédé d'obtention de signal de référence et dispositifs de communication
EP4398527A1 (fr) Procédé de traitement de modèle, dispositif électronique, dispositif de réseau et dispositif terminal
WO2024020793A1 (fr) Procédé de rétroaction d'informations d'état de canal (csi), dispositif terminal et dispositif réseau
WO2024046288A1 (fr) Procédé et appareil de communication
WO2023283785A1 (fr) Procédé de traitement de signal, et récepteur
WO2024138673A1 (fr) Procédés de communication sans fil et dispositifs de communication
WO2023115254A1 (fr) Procédé et dispositif de traitement de données
WO2024008004A1 (fr) Procédé et appareil de communication
WO2022151063A1 (fr) Procédé d'envoi d'informations, procédé de réception d'informations, appareil, dispositif et support
WO2023060503A1 (fr) Procédé et appareil de traitement d'informations, dispositif, support, puce, produit et programme
WO2024077621A1 (fr) Procédé de rétroaction d'informations de canal, dispositif d'extrémité de transmission, et dispositif d'extrémité de réception
US20230354096A1 (en) Binary variational (biv) csi coding
WO2024108356A1 (fr) Procédé de rétroaction de csi, dispositif transmetteur et dispositif récepteur
WO2024138689A1 (fr) Procédé de communication sans fil 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: 21953122

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202180097891.5

Country of ref document: CN

NENP Non-entry into the national phase

Ref country code: DE