ABSTRACT
Abstract
A neural network training method and a related apparatus are provided. The method includes: a first device receives first channel sample information from a second device. The first device determines a first neural network. The first neural network is obtained through training based on the first channel sample information, and is used to perform inference based on the first channel sample information to obtain second channel sample information. According to this method, air interface signaling overheads can be effectively reduced, adaptability to a channel environment is achieved, and communication performance is improved.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is a continuation of International Application No. PCT/CN2020/142103, filed on Dec. 31, 2020, the disclosure of which is hereby incorporated by reference in its entirety.
TECHNICAL FIELD
This application relates to the field of communication technologies, and in particular, to a neural network training method and a related apparatus.
BACKGROUND
A wireless communication system may include three parts: a transmitter, a channel, and a receiver. The channel is used for transmission of signals exchanged between the transmitter and the receiver. For example, the transmitter may be an access network device, for example, a base station (BS), and the receiver may be a terminal device. For another example, the transmitter may be a terminal device, and the receiver may be an access network device.
To optimize performance of the communication system, the transmitter and the receiver may be optimized. The transmitter and the receiver each may have an independent mathematical model. Therefore, the transmitter and the receiver are independently optimized based on respective mathematical models. For example, a mathematical channel model may be used to generate a channel sample to optimize the transmitter and the receiver.
However, because the mathematical channel model is non-ideal and non-linear, the channel sample generated by the mathematical channel model defined in a protocol can hardly reflect an actual channel environment. Transmission of a large quantity of actual channel samples between the transmitter and the receiver occupies excessive air interface resources, affecting data transmission efficiency.
SUMMARY
According to a first aspect, an embodiment of this application provides a neural network training method, including the following operations.
A first device receives first channel sample information from a second device. The first device determines a first neural network. The first neural network is obtained through training based on the first channel sample information, and is used to perform inference to obtain second channel sample information.
Optionally, an example in which the first device is an access network device and the second device is a terminal device is used for description. It may be understood that the first device may be the access network device, a chip used in the access network device, a circuit used in the access network device, or the like; and the second device may be the terminal device, a chip used in the terminal device, a circuit used in the terminal device, or the like.
In a possible design, the method includes: the first device obtains the first neural network through training based on the first channel sample information. The first neural network is used to generate new channel sample information, for example, the second channel sample information.
In a possible design, the method includes: the first device receives information about the first neural network from a third device, and determines the first neural network based on the information about the first neural network. The first neural network is obtained by the third device through training based on the first channel sample information.
According to this method, the first neural network may be obtained through training based on the first channel sample information. The first neural network is used to perform inference to obtain the second channel sample information. According to this method, air interface signaling overheads during channel sample information transmission can be effectively reduced.
In a possible design, the second channel sample information is used to train a second neural network and/or a third neural network, and the second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device trains the second neural network and/or the third neural network based on the second channel sample information or based on the second channel sample information and the first channel sample information. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device receives information about the second neural network and/or information about the third neural network from the third device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device. The second neural network and/or the third neural network are/is obtained by the third device through training based on the second channel sample information or based on the second channel sample information and the first channel sample information.
According to this method, air interface signaling overheads can be effectively reduced, and a channel environment in which a trained neural network is located can be adapted. The second neural network and the third neural network obtained through training are closer to an actual channel environment, and communication performance is improved. A speed of training the second neural network and the third neural network is also greatly improved.
In a possible design, the method further includes: the first device sends a first reference signal to the second device. Optionally, the first reference signal includes a demodulation reference signal DMRS or a channel state information reference signal CSI-RS. Optionally, a sequence type of the first reference signal includes a ZC sequence or a gold sequence.
In a possible design, the first channel sample information includes but is not limited to a second reference signal and/or channel state information (CSI). The second reference signal is the first reference signal propagated through a channel. Alternatively, it is described as that the second reference signal is the first reference signal received by the second device from the first device.
In a possible design, the information about the first neural network includes a model variation of the first neural network relative to a reference neural network.
In a possible design, the information about the first neural network includes one or more of the following: a weight of a neural network, an activation function of a neuron, a quantity of neurons at each layer of the neural network, an inter-layer cascading relationship of the neural network, and a network type of each layer of the neural network.
In a possible design, the first neural network is a generative neural network. Optionally, the first neural network is a generative adversarial network (GAN) or a variational autoencoder (VAE).
According to this method, the second channel sample information that has a same distribution as or has a similar distribution to the first channel sample information may be obtained, so that the second channel sample information is closer to an actual channel environment.
In a possible design, the method further includes: the first device receives capability information of the second device from the second device. The capability information indicates one or more of the following information:
(1) whether the second device supports using the neural network to replace or implement a function of a communication module, where the communication module includes but is not limited to an OFDM modulation module, an OFDM demodulation module, a constellation mapping module, a constellation demapping module, a channel encoding module, a channel decoding module, a precoding module, an equalization module, an interleaving module, and/or a de-interleaving module; (2) whether the second device supports a network type of the third neural network; (3) whether the second device supports receiving the information about the third neural network using signaling; (4) the reference neural network stored by the second device; (5) memory space that may be used by the second device to store the third neural network; and (6) computing power information that may be used by the second device to run the neural network.
According to this method, the first device may receive the capability information sent by the second device. The capability information notifies the first device of related information about the second device. The first device may perform an operation related to the third neural network based on the capability information, to ensure that the second device can normally use the third neural network.
In a possible design, the method further includes: the first device sends the information about the third neural network to the second device.
In this embodiment, after completing the training of the third neural network, the first device sends the information about the third neural network to the second device. The information about the third neural network includes but is not limited to: a weight of a neural network, an activation function of a neuron, a quantity of neurons at each layer of the neural network, an inter-layer cascading relationship of the neural network, and/or a network type of each layer of the neural network. For example, the information about the third neural network may indicate different activation functions for different neurons.
In a possible design, when the third neural network is preconfigured (or predefined) in the second device, the information about the third neural network may alternatively be a model variation of the third neural network. The model variation includes but is not limited to: a weight of a changed neural network, a changed activation function, a quantity of neurons at one or more layers of the changed neural network, an inter-layer cascading relationship of the changed neural network, and/or a network type of one or more layers of the changed neural network. For example, the pre-configuration may be performed by the access network device using signaling, and the pre-definition may be performed in a protocol. For example, the third neural network in the terminal device is predefined as a neural network A in the protocol.
In embodiments of this application, there may be a plurality of implementation solutions for the information about the third neural network, so that implementation flexibility of the solution is improved.
According to a second aspect, an embodiment of this application provides a neural network training method, including: a second device performs channel estimation based on a first reference signal received from a first device, to determine first channel sample information. The second device sends the first channel sample information to the first device. The second device receives information about a third neural network from the first device. The third neural network is used for transmission of target information between the first device and the second device.
In a possible design, the method further includes: the second device sends capability information of the second device to the first device.
For descriptions of the first channel sample information, the information about the third neural network, the capability information of the second device, and the like, refer to the first aspect. Details are not described herein again.
According to a third aspect, an embodiment of this application provides a neural network training method, including the following operations.
A first device sends a first reference signal to a second device. The first device receives information about a first neural network from the second device. The first neural network is used to perform inference to obtain second channel sample information.
In the method, the second device (for example, a terminal device) obtains the first neural network through training based on the first channel sample information. The first neural network is used to perform inference to obtain the second channel sample information. According to this method, air interface signaling overheads during channel sample information transmission can be effectively reduced.
In a possible design, the second channel sample information is used to train a second neural network and/or a third neural network, and the second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device trains the second neural network and/or the third neural network based on the second channel sample information or based on the second channel sample information and the first channel sample information. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device receives information about the second neural network and/or information about the third neural network from a third device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device. The second neural network and/or the third neural network are/is obtained by the third device through training based on the second channel sample information or based on the second channel sample information and the first channel sample information.
According to this method, air interface signaling overheads can be effectively reduced, and a channel environment in which a trained neural network is located can be adapted. The second neural network and the third neural network obtained through training are closer to an actual channel environment, and communication performance is improved. A speed of training the second neural network and the third neural network is also greatly improved.
Specifically, for descriptions of the first reference signal, the first neural network, the information about the first neural network, the second neural network, and/or the third neural network, refer to the first aspect. Details are not described again.
In a possible design, the method further includes: the first device sends the inf
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is a continuation of International Application No. PCT/CN2020/142103, filed on Dec. 31, 2020, the disclosure of which is hereby incorporated by reference in its entirety.
TECHNICAL FIELD
This application relates to the field of communication technologies, and in particular, to a neural network training method and a related apparatus.
BACKGROUND
A wireless communication system may include three parts: a transmitter, a channel, and a receiver. The channel is used for transmission of signals exchanged between the transmitter and the receiver. For example, the transmitter may be an access network device, for example, a base station (BS), and the receiver may be a terminal device. For another example, the transmitter may be a terminal device, and the receiver may be an access network device.
To optimize performance of the communication system, the transmitter and the receiver may be optimized. The transmitter and the receiver each may have an independent mathematical model. Therefore, the transmitter and the receiver are independently optimized based on respective mathematical models. For example, a mathematical channel model may be used to generate a channel sample to optimize the transmitter and the receiver.
However, because the mathematical channel model is non-ideal and non-linear, the channel sample generated by the mathematical channel model defined in a protocol can hardly reflect an actual channel environment. Transmission of a large quantity of actual channel samples between the transmitter and the receiver occupies excessive air interface resources, affecting data transmission efficiency.
SUMMARY
According to a first aspect, an embodiment of this application provides a neural network training method, including the following operations.
A first device receives first channel sample information from a second device. The first device determines a first neural network. The first neural network is obtained through training based on the first channel sample information, and is used to perform inference to obtain second channel sample information.
Optionally, an example in which the first device is an access network device and the second device is a terminal device is used for description. It may be understood that the first device may be the access network device, a chip used in the access network device, a circuit used in the access network device, or the like; and the second device may be the terminal device, a chip used in the terminal device, a circuit used in the terminal device, or the like.
In a possible design, the method includes: the first device obtains the first neural network through training based on the first channel sample information. The first neural network is used to generate new channel sample information, for example, the second channel sample information.
In a possible design, the method includes: the first device receives information about the first neural network from a third device, and determines the first neural network based on the information about the first neural network. The first neural network is obtained by the third device through training based on the first channel sample information.
According to this method, the first neural network may be obtained through training based on the first channel sample information. The first neural network is used to perform inference to obtain the second channel sample information. According to this method, air interface signaling overheads during channel sample information transmission can be effectively reduced.
In a possible design, the second channel sample information is used to train a second neural network and/or a third neural network, and the second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device trains the second neural network and/or the third neural network based on the second channel sample information or based on the second channel sample information and the first channel sample information. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device receives information about the second neural network and/or information about the third neural network from the third device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device. The second neural network and/or the third neural network are/is obtained by the third device through training based on the second channel sample information or based on the second channel sample information and the first channel sample information.
According to this method, air interface signaling overheads can be effectively reduced, and a channel environment in which a trained neural network is located can be adapted. The second neural network and the third neural network obtained through training are closer to an actual channel environment, and communication performance is improved. A speed of training the second neural network and the third neural network is also greatly improved.
In a possible design, the method further includes: the first device sends a first reference signal to the second device. Optionally, the first reference signal includes a demodulation reference signal DMRS or a channel state information reference signal CSI-RS. Optionally, a sequence type of the first reference signal includes a ZC sequence or a gold sequence.
In a possible design, the first channel sample information includes but is not limited to a second reference signal and/or channel state information (CSI). The second reference signal is the first reference signal propagated through a channel. Alternatively, it is described as that the second reference signal is the first reference signal received by the second device from the first device.
In a possible design, the information about the first neural network includes a model variation of the first neural network relative to a reference neural network.
In a possible design, the information about the first neural network includes one or more of the following: a weight of a neural network, an activation function of a neuron, a quantity of neurons at each layer of the neural network, an inter-layer cascading relationship of the neural network, and a network type of each layer of the neural network.
In a possible design, the first neural network is a generative neural network. Optionally, the first neural network is a generative adversarial network (GAN) or a variational autoencoder (VAE).
According to this method, the second channel sample information that has a same distribution as or has a similar distribution to the first channel sample information may be obtained, so that the second channel sample information is closer to an actual channel environment.
In a possible design, the method further includes: the first device receives capability information of the second device from the second device. The capability information indicates one or more of the following information:
(1) whether the second device supports using the neural network to replace or implement a function of a communication module, where the communication module includes but is not limited to an OFDM modulation module, an OFDM demodulation module, a constellation mapping module, a constellation demapping module, a channel encoding module, a channel decoding module, a precoding module, an equalization module, an interleaving module, and/or a de-interleaving module; (2) whether the second device supports a network type of the third neural network; (3) whether the second device supports receiving the information about the third neural network using signaling; (4) the reference neural network stored by the second device; (5) memory space that may be used by the second device to store the third neural network; and (6) computing power information that may be used by the second device to run the neural network.
According to this method, the first device may receive the capability information sent by the second device. The capability information notifies the first device of related information about the second device. The first device may perform an operation related to the third neural network based on the capability information, to ensure that the second device can normally use the third neural network.
In a possible design, the method further includes: the first device sends the information about the third neural network to the second device.
In this embodiment, after completing the training of the third neural network, the first device sends the information about the third neural network to the second device. The information about the third neural network includes but is not limited to: a weight of a neural network, an activation function of a neuron, a quantity of neurons at each layer of the neural network, an inter-layer cascading relationship of the neural network, and/or a network type of each layer of the neural network. For example, the information about the third neural network may indicate different activation functions for different neurons.
In a possible design, when the third neural network is preconfigured (or predefined) in the second device, the information about the third neural network may alternatively be a model variation of the third neural network. The model variation includes but is not limited to: a weight of a changed neural network, a changed activation function, a quantity of neurons at one or more layers of the changed neural network, an inter-layer cascading relationship of the changed neural network, and/or a network type of one or more layers of the changed neural network. For example, the pre-configuration may be performed by the access network device using signaling, and the pre-definition may be performed in a protocol. For example, the third neural network in the terminal device is predefined as a neural network A in the protocol.
In embodiments of this application, there may be a plurality of implementation solutions for the information about the third neural network, so that implementation flexibility of the solution is improved.
According to a second aspect, an embodiment of this application provides a neural network training method, including: a second device performs channel estimation based on a first reference signal received from a first device, to determine first channel sample information. The second device sends the first channel sample information to the first device. The second device receives information about a third neural network from the first device. The third neural network is used for transmission of target information between the first device and the second device.
In a possible design, the method further includes: the second device sends capability information of the second device to the first device.
For descriptions of the first channel sample information, the information about the third neural network, the capability information of the second device, and the like, refer to the first aspect. Details are not described herein again.
According to a third aspect, an embodiment of this application provides a neural network training method, including the following operations.
A first device sends a first reference signal to a second device. The first device receives information about a first neural network from the second device. The first neural network is used to perform inference to obtain second channel sample information.
In the method, the second device (for example, a terminal device) obtains the first neural network through training based on the first channel sample information. The first neural network is used to perform inference to obtain the second channel sample information. According to this method, air interface signaling overheads during channel sample information transmission can be effectively reduced.
In a possible design, the second channel sample information is used to train a second neural network and/or a third neural network, and the second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device trains the second neural network and/or the third neural network based on the second channel sample information or based on the second channel sample information and the first channel sample information. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device.
In a possible design, the method includes: the first device receives information about the second neural network and/or information about the third neural network from a third device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and the second device. The second neural network and/or the third neural network are/is obtained by the third device through training based on the second channel sample information or based on the second channel sample information and the first channel sample information.
According to this method, air interface signaling overheads can be effectively reduced, and a channel environment in which a trained neural network is located can be adapted. The second neural network and the third neural network obtained through training are closer to an actual channel environment, and communication performance is improved. A speed of training the second neural network and the third neural network is also greatly improved.
Specifically, for descriptions of the first reference signal, the first neural network, the information about the first neural network, the second neural network, and/or the third neural network, refer to the first aspect. Details are not described again.
In a possible design, the method further includes: the first device sends the information about the third neural network to the second device.
Specifically, for descriptions of the information about the third neural network, refer to the first aspect. Details are not described again.
In a possible design, the method further includes the following operations.
The first device receives capability information from the second device. The capability information indicates one or more of the following information about the second device:
(1) whether to support using the neural network to replace or implement a function of a communication module; (2) whether to support a network type of the first neural network; (3) whether to support a network type of the third neural network; (4) whether to support receiving information about a reference neural network using signaling, where the reference neural network is used to train the first neural network; (5) whether to support receiving the information about the third neural network using signaling; (6) stored reference neural network; (7) memory space for storing the first neural network and/or the third neural network; (8) computing power information that may be used to run the neural network; and (9) location information of the second device.
According to a fourth aspect, an embodiment of this application provides a neural network training method, including the following operations.
A second device performs channel estimation based on a first reference signal received from a first device, to determine first channel sample information. The second device determines a first neural network. The first neural network is obtained through training based on first channel sample information. The second device sends information about the first neural network to the first device.
Specifically, for descriptions of the first reference signal, the first neural network, the first channel sample information, the information about the first neural network, and the like, refer to the third aspect. Details are not described again.
In a possible design, the method further includes the following operations.
The second device receives information about a third neural network from the first device. For the information about the third neural network, refer to the third aspect. Details are not described herein again.
In a possible design, the method further includes: the second device sends capability information to the first device. For the capability information, refer to the third aspect. Details are not described herein again.
According to a fifth aspect, an embodiment of this application provides a neural network training method, including: a third device receives first channel sample information from a first device. The third device obtains a first neural network through training based on the first channel sample information. The first neural network is used to perform inference to obtain second channel sample information.
In a possible design, the method further includes: training a second neural network and/or a third neural network based on the second channel sample information, and sending information about the second neural network and/or information about the third neural network to the first device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and a second device.
According to a sixth aspect, an apparatus is provided. The apparatus may be an access network device, an apparatus in the access network device, or an apparatus that can be used with the access network device.
In a possible design, the apparatus may include modules for performing the method/operations/steps/actions described in the first aspect. The modules may be hardware circuits, may be software, or may be implemented by using a combination of a hardware circuit and software. In a design, the apparatus may include a processing module and a transceiver module.
For example, the transceiver module is configured to receive first channel sample information from a second device.
The processing module is configured to determine a first neural network. The first neural network is obtained through training based on the first channel sample information, and is used to perform inference to obtain second channel sample information.
For descriptions of the first neural network, the first channel sample information, the second channel sample information, and other operations, refer to the first aspect. Details are not described herein again.
In a possible design, the apparatus may include modules for performing the method/operations/steps/actions described in the third aspect. The modules may be hardware circuits, may be software, or may be implemented by using a combination of a hardware circuit and software. In a design, the apparatus may include a processing module and a transceiver module.
For example, the transceiver module is configured to send a first reference signal to the second device, and receive information about a first neural network from the second device. The first neural network is used to perform inference to obtain the second channel sample information.
For descriptions of the first neural network, the first channel sample information, the second channel sample information, and other operations, refer to the third aspect. Details are not described herein again.
According to a seventh aspect, an apparatus is provided. The apparatus may be a terminal device, an apparatus in the terminal device, or an apparatus that can be used with the terminal device.
In a possible design, the apparatus may include modules for performing the method/operations/steps/actions described in the second aspect. The modules may be hardware circuits, may be software, or may be implemented by using a combination of a hardware circuit and software. In a design, the apparatus may include a processing module and a transceiver module.
For example, the processing module is configured to perform channel estimation based on a first reference signal received from a first device, to determine first channel sample information.
The transceiver module is configured to send the first channel sample information to the first device.
The transceiver module is further configured to receive information about a third neural network from the first device. The third neural network is used for transmission of target information between the first device and a second device.
For descriptions of the first reference signal, the first channel sample information, the third neural network, and other operations, refer to the second aspect. Details are not described herein again.
In a possible design, the apparatus may include modules for performing the method/operations/steps/actions described in the fourth aspect. The modules may be hardware circuits, may be software, or may be implemented by using a combination of a hardware circuit and software. In a design, the apparatus may include a processing module and a transceiver module.
For example, the processing module is configured to perform channel estimation based on the first reference signal received from the first device, to determine the first channel sample information.
The processing module is further configured to determine a first neural network. The first neural network is obtained through training based on the first channel sample information.
The transceiver module is configured to send information about the first neural network to the first device.
For descriptions of the first reference signal, the first channel sample information, the first neural network, and other operations, refer to the fourth aspect. Details are not described herein again.
According to an eighth aspect, an apparatus is provided. The apparatus may be an AI node, an apparatus in the AI node, or an apparatus that can be used with the AI node.
In a possible design, the apparatus may include modules for performing the method/operations/steps/actions described in the fifth aspect. The modules may be hardware circuits, may be software, or may be implemented by using a combination of a hardware circuit and software. In a design, the apparatus may include a processing module and a transceiver module.
For example, the transceiver module is configured to receive first channel sample information from a first device.
The processing module is configured to obtain a first neural network through training based on the first channel sample information. The first neural network is used to perform inference to obtain second channel sample information.
In a possible design, the processing module is further configured to train a second neural network and/or a third neural network based on the second channel sample information. The transceiver module is further configured to send information about the second neural network and/or information about the third neural network to the first device. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and a second device.
According to a ninth aspect, an embodiment of this application provides an apparatus.
In a possible design, the apparatus includes a processor, configured to implement the method described in the first aspect. The apparatus may further include a memory, configured to store instructions and data. The memory is coupled to the processor, and the processor may implement the method described in the first aspect when executing the instructions stored in the memory. The apparatus may further include a communication interface. The communication interface is used by the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.
In a possible design, the apparatus includes:
a memory, configured to store program instructions; and a processor, configured to receive first channel sample information from a second device through a communication interface.
The processor is further configured to determine a first neural network. The first neural network is obtained through training based on the first channel sample information, and is used to perform inference to obtain second channel sample information.
For descriptions of the first neural network, the first channel sample information, the second channel sample information, and other operations, refer to the first aspect. Details are not described herein again.
In a possible design, the apparatus includes a processor, configured to implement the method described in the third aspect. The apparatus may further include a memory, configured to store instructions and data. The memory is coupled to the processor, and the processor may implement the method described in the third aspect when executing the instructions stored in the memory. The apparatus may further include a communication interface. The communication interface is used by the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.
In a possible design, the apparatus includes:
a memory, configured to store program instructions; and a processor, configured to send a first reference signal to the second device through a communication interface, and receive information about the first neural network from the second device. The first neural network is used to perform inference to obtain the second channel sample information.
For descriptions of the first neural network, the first channel sample information, the second channel sample information, and other operations, refer to the third aspect. Details are not described herein again.
According to a tenth aspect, an embodiment of this application provides an apparatus.
In a possible design, the apparatus includes a processor, configured to implement the method described in the second aspect. The apparatus may further include a memory, configured to store instructions and data. The memory is coupled to the processor, and the processor may implement the method described in the second aspect when executing the instructions stored in the memory. The apparatus may further include a communication interface. The communication interface is used by the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.
In a possible design, the apparatus includes:
a memory, configured to store program instructions; and a processor, configured to perform channel estimation based on a first reference signal received from a first device through a communication interface, determine first channel sample information, send the first channel sample information to the first device, and receive information about a third neural network from the first device. The third neural network is used for transmission of target information between the first device and a second device.
For descriptions of the first reference signal, the first channel sample information, the third neural network, and other operations, refer to the second aspect. Details are not described herein again.
In a possible design, the apparatus includes a processor, configured to implement the method described in the fourth aspect. The apparatus may further include a memory, configured to store instructions and data. The memory is coupled to the processor, and the processor may implement the method described in the fourth aspect when executing the instructions stored in the memory. The apparatus may further include a communication interface. The communication interface is used by the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.
In a possible design, the apparatus includes:
a memory, configured to store program instructions; and a processor, configured to perform channel estimation based on the first reference signal received from the first device through a communication interface, determine the first channel sample information, determine a first neural network, and send information about the first neural network to the first device. The first neural network is obtained through training based on the first channel sample information.
For descriptions of the first reference signal, the first channel sample information, the first neural network, and other operations, refer to the fourth aspect. Details are not described herein again.
According to an eleventh aspect, an embodiment of this application provides an apparatus. The apparatus includes a processor, configured to implement the method described in the fifth aspect. The apparatus may further include a memory, configured to store instructions and data. The memory is coupled to the processor, and the processor may implement the method described in the fifth aspect when executing the instructions stored in the memory. The apparatus may further include a communication interface. The communication interface is used by the apparatus to communicate with another device. For example, the communication interface may be a transceiver, a circuit, a bus, a module, or another type of communication interface.
In a possible design, the apparatus includes:
a memory, configured to store program instructions; and a processor, configured to receive first channel sample information from a first device through a communication interface, and obtain a first neural network through training based on the first channel sample information. The first neural network is used to perform inference to obtain second channel sample information.
In a possible design, the processor is further configured to train a second neural network and/or a third neural network based on the second channel sample information. The processor further sends information about the second neural network and/or information about the third neural network to the first device through the communication interface. The second neural network and/or the third neural network are/is used for transmission of target information between the first device and a second device.
According to a twelfth aspect, an embodiment of this application further provides a computer-readable storage medium, including instructions. When the instructions are run on a computer, the computer is enabled to perform the method in any one of the first aspect to the fifth aspect.
According to a thirteenth aspect, an embodiment of this application further provides a computer program product, including instructions. When the instructions are run on a computer, the computer is enabled to perform the method in any one of the first aspect to the fifth aspect.
According to a fourteenth aspect, an embodiment of this application provides a chip system. The chip system includes a processor, and may further include a memory, configured to implement the method in any one of the first aspect to the fifth aspect. The chip system may include a chip, or may include a chip and another discrete component.
According to a fifteenth aspect, an embodiment of this application further provides a communication system, and the communication system includes:
the apparatus according to the sixth aspect and the apparatus according to the seventh aspect; the apparatus according to the sixth aspect, the apparatus according to the seventh aspect, and the apparatus according to the eighth aspect; the apparatus according to the ninth aspect and the apparatus according to the tenth aspect; or the apparatus according to the ninth aspect, the apparatus according to the tenth aspect, and the apparatus according to the eleventh aspect.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a schematic architectural diagram of a network according to an embodiment of this application;
FIG. 2 is a schematic diagram of a hardware structure of a communication apparatus according to an embodiment of this application;
FIG. 3 is a schematic structural diagram of a neuron according to an embodiment of this application;
FIG. 4 is a schematic diagram of a layer relationship in a neural network according to an embodiment of this application;
FIG. 5 is a schematic diagram of a convolutional neural network (CNN) according to an embodiment of this application;
FIG. 6 is a schematic diagram of a recurrent neural network (RNN) according to an embodiment of this application;
FIG. 7 is a schematic diagram of a generative adversarial network (GAN) according to an embodiment of this application;
FIG. 8 is a schematic diagram of a variational autoencoder (VAE) according to an embodiment of this application;
FIG. 9 is a schematic architectural diagram of a neural network of constellation modulation/demodulation optimized through joint sending and receiving according to an embodiment of this application;
FIG. 10 to FIG. 12 are schematic flowcharts of a neural network training method according to an embodiment of this application;
FIG. 13 is a schematic structural diagram of a generator network of a first neural network according to an embodiment of this application;
FIG. 14 is a schematic structural diagram of a discriminator network of a first neural network according to an embodiment of this application;
FIG. 15 a is a schematic structural diagram of a generator network of a first neural network according to an embodiment of this application;
FIG. 15 b is a schematic structural diagram of a discriminator network of a first neural network according to an embodiment of this application;
FIG. 16 a and FIG. 16 b each are a schematic structural diagram of a network according to an embodiment of this application; and
FIG. 17 is a schematic diagram of a communication apparatus according to an embodiment of this application.
DESCRIPTION OF EMBODIMENTS
A wireless communication system includes communication devices, and the communication devices may perform wireless communication by using a radio resource. The communication devices may include an access network device and a terminal device, and the access network device may also be referred to as an access side device. The radio resource may include a link resource and/or an air interface resource. The air interface resource may include at least one of a time domain resource, a frequency domain resource, a code resource, and a space resource. In embodiments of this application, âat least one (type)â may alternatively be described as âone (type) or more (types)â, and âa plurality of (types)â may be two (types), three (types), four (types), or more (types). This is not limited in embodiments of this application.
In embodiments of this application, â/â may represent an âorâ relationship between associated objects. For example, A/B may represent A or B. âAnd/orâ may be used to indicate that three relationships exist between associated objects. For example, A and/or B may represent the following three cases: Only A exists, both A and B exist, and only B exists. A and B may be singular or plural. To facilitate descriptions of the technical solutions in embodiments of this application, terms such as âfirstâ and âsecondâ may be used to distinguish between technical features with same or similar functions. The terms such as âfirstâ and âsecondâ do not limit a quantity and an execution sequence, and the terms such as âfirstâ and âsecondâ do not indicate a definite difference. In embodiments of this application, the terms such as âexampleâ or âfor exampleâ are used to represent an example, an illustration, or a description. Any embodiment or design scheme described with âexampleâ or âfor exampleâ should not be explained as being more preferred or having more advantages than another embodiment or design scheme. Use of the terms such as âexampleâ or âfor exampleâ is intended to present a related concept in a specific manner for ease of understanding.
FIG. 1 is a schematic architectural diagram of a network to which an embodiment of this application is applicable. A communication system in embodiments of this application may be a system including the access network device (for example, a base station shown in FIG. 1 ) and the terminal device, or may be a system including two or more terminal devices. In the communication system, the access network device may send configuration information to the terminal device, and the terminal device performs corresponding configuration based on the configuration information. The access network device may send downlink data to the terminal device, and/or the terminal device may send uplink data to the access network device. In the communication system (for example, internet of vehicles) including two or more terminal devices, a terminal device 1 may send configuration information to a terminal device 2, and the terminal device 2 performs corresponding configuration based on the configuration information. The terminal device 1 may send data to the terminal device 2, and the terminal device 2 may also send data to the terminal device 1. Optionally, in the communication system shown in FIG. 1 , the access network device may implement one or more of the following artificial intelligence (AI) functions: model training and inference. Optionally, in the communication system shown in FIG. 1 , a network side may include a node independent of the access network device, configured to implement one or more of the following AI functions: model training and inference. The node may be referred to as an AI node, a model training node, an inference node, a wireless intelligent controller, or another name. This is not limited. For example, the access network device may implement the model training function and the inference function. Alternatively, the AI node may implement the model training function and the inference function. Alternatively, the AI node may implement the model training function, and send information about the model to the access network device, and the access network device implements the inference function. Optionally, if the AI node implements the inference function, the AI node may send an inference result to the access network device for use by the access network device, and/or the AI node may send the inference result to the terminal device via the access network device for use by the terminal device. If the access network device implements the inference function, the access network device may use the inference result, or send the inference result to the terminal device for use by the terminal device. If the AI node is used to implement the model training function and the inference function, the AI node may be divided into two nodes. One node thereof implements the model training function, and the other node thereof implements the inference function.
A specific quantity of network elements in the communication system is not limited in embodiments of this application.
The terminal device in embodiments of this application may also be referred to as a terminal or an access terminal, and may be a device having a wireless transceiver function. The terminal device may communicate with one or more core networks (CNs) via the access network device. The terminal device may be a subscriber unit, a subscriber station, a mobile station, a mobile console, a remote station, a remote terminal, a mobile device, a user terminal, user equipment (UE), a user agent, a user apparatus, or the like. The terminal device may be deployed on land, including indoor or outdoor, in a handheld manner or vehicle-mounted manner, may be deployed on water (for example, on a ship), or may be deployed in air (for example, on a plane, a balloon, or a satellite). The terminal device may be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a smartphone, a mobile phone, a wireless local loop (WLL) station, or a personal digital assistant (PDA). Alternatively, the terminal device may be a handheld device, a computing device, or another device having a wireless communication function, a vehicle-mounted device, a wearable device, an unmanned aerial vehicle device, a terminal in internet of things or internet of vehicles, a terminal in a fifth generation (5G) mobile communication network, relay user equipment, or a terminal in a future evolved mobile communication network, or the like. The relay user equipment may be, for example, a 5G residential gateway (RG). For another example, the terminal device may be a virtual reality (virtual reality, VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, or the like. This is not limited in embodiments of this application. In embodiments of this application, an apparatus configured to implement a function of the terminal may be a terminal, or may be an apparatus that can support the terminal in implementing the function, for example, a chip system. The apparatus may be installed in the terminal or used with the terminal. In embodiments of this application, the chip system may include a chip, or may include a chip and another discrete component.
The access network device may be considered as a sub-network of a carrier network, and is an implementation system between a service node in the carrier network and the terminal device. To access the carrier network, the terminal device may first pass through the access network device, and then may be connected to the service node in the carrier network via the access network device. The access network device in embodiments of this application is a device that is located in a (radio) access network ((R)AN) and that can provide the wireless communication function for the terminal device. The access network device includes a base station, and includes but is not limited to, for example, a next generation NodeB (gNB) in a 5G system and an evolved NodeB (eNB) in a long term evolution (LTE) system, a radio network controller (RNC), a NodeB (NB), a base station controller (BSC), a base transceiver station (BTS), a home base station (for example, home evolved NodeB, or home NodeB, HNB), a base band unit (BBU), a transmitting and receiving point (TRP), a transmitting point (TP), a pico base station device, a mobile switching center, and an access network device in a future network. In systems using different radio access technologies, devices having a function of the access network device may have different names. In embodiments of this application, an apparatus configured to implement the function of the access network device may be an access network device, or may be an apparatus that can support the access network device in implementing the function, for example, a chip system. The apparatus may be installed in the access network device or used with the access network device.
The technical solutions provided in embodiments of this application may be applied to various communication systems, for example, an LTE system, a 5G system, a wireless-fidelity (Wi-Fi) system, a future sixth generation mobile communication system, or a system integrating a plurality of communication systems. This is not limited in embodiments of this application. 5G may also be referred to as new radio (NR).
The technical solutions provided in embodiments of this application may be applied to various communication scenarios, for example, may be applied to one or more of the following communication scenarios: enhanced mobile broadband (eMBB) communication, ultra-reliable low-latency communication (URLLC), machine type communication (MTC), massive machine type
CLAIMS
Claims ( 24 )
What is claimed is:
1 . A neural network training method, comprising:
sending, by a first device, a first reference signal to a second device; receiving, by the first device, first channel sample information from the second device; and determining, by the first device, a first neural network, wherein the first neural network is obtained through training based on the first channel sample information, and is used to perform inference to obtain second channel sample information, wherein the second channel sample information is used to train a second neural network, and the second neural network is used for transmission of target information between the first device and the second device.
2 . The method according to claim 1 , wherein the first channel sample information is further used to train the second neural network .
3 . The method according to claim 1 , wherein the method further comprises:
sending, by the first device, information about a third neural network to the second device.
4 . The method according to claim 1 , wherein the first channel sample information comprises channel state information (CSI) or a second reference signal, and the second reference signal is the first reference signal propagated through a channel.
5 . The method according to claim 1 , wherein the first neural network is a generative adversarial network or a variational autoencoder.
6 . The method according to claim 1 , wherein the first reference signal comprises a demodulation reference signal (DMRS) or a channel state information reference signal (CSI-RS).
7 . The method according to claim 1 , wherein a sequence type of the first reference signal comprises a Zadoff-Chu (ZC) sequence or a gold sequence.
8 . A neural network training method, comprising:
performing, by a second device, channel estimation based on a first reference signal received from a first device, to determine first channel sample information; sending, by the second device, the first channel sample information to the first device; and receiving, by the second device, information about a third neural network from the first device, wherein the third neural network is used for transmission of target information between the first device and the second device.
9 . The method according to claim 8 , wherein the first channel sample information comprises channel state information (CSI) or a second reference signal, and the second reference signal is the first reference signal received by the second device through a channel.
10 . The method according to claim 8 , wherein a first neural network is a generative adversarial network or a variational autoencoder.
11 . The method according to claim 8 , wherein the first reference signal comprises a demodulation reference signal (DMRS) or a channel state information reference signal (CSI-RS).
12 . The method according to claim 8 , wherein a sequence type of the first reference signal comprises a Zadoff-Chu (ZC) sequence or a gold sequence.
13 - 20 . (canceled)
21 . A neural network training method, comprising:
performing, by a second device, channel estimation based on a first reference signal received from a first device, to determine first channel sample information; determining, by the second device, a first neural network, wherein the first neural network is obtained through training based on the first channel sample information; and sending, by the second device, information about the first neural network to the first device.
22 . The method according to claim 21 , wherein the information about the first neural network comprises a model variation of the first neural network relative to a reference neural network, and the reference neural network is used to train the first neural network.
23 . The method according to claim 21 , wherein the information about the first neural network comprises one or more of the following: a weight of a neural network, an activation function of a neuron, a quantity of neurons at each layer of the neural network, an inter-layer cascading relationship of the neural network, and/or a network type of each layer of the neural network.
24 . The method according to claim 21 , wherein the method further comprises:
receiving, by the second device, information about a third neural network from the first device.
25 . The method according to claim 21 , wherein the method further comprises:
sending, by the second device, capability information to the first device, wherein the capability information indicates one or more of the following information about the second device:
whether to support using the neural network to replace or implement a function of a communication module;
whether to support a network type of the first neural network;
whether to support a network type of the third neural network;
whether to support receiving information about the reference neural network using signaling, wherein the reference neural network is used to train the first neural network;
whether to support receiving the information about the third neural network using signaling;
the stored reference neural network; memory space for storing the first neural network and/or the third neural network; computing power information that may be used to run the neural network; or location information of the second device.
26 . The method according to claim 21 , wherein the first neural network is a generative adversarial network or a variational autoencoder.
27 - 28 . (canceled)
31 . A communication apparatus, comprising a processor and a memory, wherein the memory is coupled to the processor, and the processor is configured to perform the method according to claim 1 .
32 . (canceled)
33 . A communication apparatus, comprising a processor and a memory, wherein the memory is coupled to the processor, and the processor is configured to perform the method according to claim 8 .
34 - 42 . (canceled)
US18/345,904
2020-12-31
2023-06-30
Neural network training method and related apparatus
Pending
US20230342593A1
( en )
Applications Claiming Priority (1)
Application Number
Priority Date
Filing Date
Title
PCT/CN2020/142103
WO2022141397A1
( en )
2020-12-31
2020-12-31
Neural network training method and related apparatus
Related Parent Applications (1)
Application Number
Title
Priority Date
Filing Date
PCT/CN2020/142103
Continuation
WO2022141397A1
( en )
2020-12-31
2020-12-31
Neural network training method and related apparatus
Publications (1)
Publication Number
Publication Date
US20230342593A1
true
US20230342593A1 ( en )
2023-10-26
Family
ID=82258815
Family Applications (1)
Application Number
Title
Priority Date
Filing Date
US18/345,904
Pending
US20230342593A1
( en )
2020-12-31
2023-06-30
Neural network training method and related apparatus
Country Status (4)
Country
Link
US
( 1 )
US20230342593A1
( en )
EP
( 1 )
EP4262121A4
( en )
CN
( 2 )
CN116458103B
( en )
WO
( 1 )
WO2022141397A1
( en )
Cited By (11)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
US20230031702A1
( en )
*
2021-07-14
2023-02-02
Google Llc
Neural Networks based Multimodal Transformer for Multi-Task User Interface Modeling
US20240097843A1
( en )
*
2022-08-19
2024-03-21
Nvidia Corporation
Reference signal configuration information transmission
US20240154722A1
( en )
*
2022-10-27
2024-05-09
Samsung Electronics Co., Ltd.
Method and system for rate-matching auto-encoders
US20240187285A1
( en )
*
2021-04-15
2024-06-06
Nokia Solutions And Networks Oy
Machine learning for channel estimation against diverse doppler effects
US20240323059A1
( en )
*
2021-12-03
2024-09-26
Guangdong Oppo Mobile Telecommunications Corp., Ltd.
Data acquisition method and apparatus, and device, medium, chip, product and program
US20240333601A1
( en )
*
2021-06-25
2024-10-03
Google Llc
Wireless network employing neural networks for channel state feedback
US20240396766A1
( en )
*
2021-09-14
2024-11-28
Telefonaktiebolaget Lm Ericsson (Publ)
Improved pilot assisted radio propagation channel estimation based on machine learning
US20240430062A1
( en )
*
2023-06-20
2024-12-26
Qualcomm Incorporated
Ml based dynamic bit loading and rate control
US20240429985A1
( en )
*
2021-10-20
2024-12-26
Lg Electronics Inc.
Method and device for transmitting and receiving channel state information in wireless communication system
EP4597935A3
( en )
*
2024-01-31
2025-11-12
Nokia Solutions and Networks Oy
Communication systems
US12580795B2
( en )
*
2021-04-13
2026-03-17
Samsung Electronics Co., Ltd.
Electronic device for processing wireless signal, and operating method therefor
Families Citing this family (17)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
US12323210B2
( en )
*
2021-01-29
2025-06-03
Qualcomm Incorporated
Techniques for determining channel state information using a neural network model
CN117692032A
( en )
*
2022-08-12
2024-03-12
ç»´æ²ç§»å¨éä¿¡æéå ¬å¸
Information transmission methods, devices, equipment, systems and storage media
CN117648966A
( en )
*
2022-08-16
2024-03-05
ä¸å½ç§»å¨éä¿¡æéå ¬å¸ç 究é¢
AI model training methods, devices, equipment and readable storage media
CN117715106A
( en )
*
2022-09-07
2024-03-15
ç»´æ²ç§»å¨éä¿¡æéå ¬å¸
Information transmission method, AI network model training method, device and communication equipment
US12506521B2
( en )
2023-01-30
2025-12-23
Nokia Technologies Oy
Method for data augmentation for channel state information feedback
WO2024207222A1
( en )
*
2023-04-04
2024-10-10
Oppo广ä¸ç§»å¨éä¿¡æéå ¬å¸
Channel information determination method and apparatus, and medium, chip, product and program
CN116404760B
( en )
*
2023-06-09
2023-10-20
西å®æ°è§ç©ºé´ä¿¡æ¯ç§ææéå ¬å¸
Distributed power grid transient steady-state operation method and device based on digital twin map
WO2025039193A1
( en )
*
2023-08-22
2025-02-27
å为ææ¯æéå ¬å¸
Neural network training method and communication apparatus
WO2025050325A1
( en )
*
2023-09-06
2025-03-13
Oppo广ä¸ç§»å¨éä¿¡æéå ¬å¸
Wireless communication method and communication device
CN119886383A
( en )
*
2023-10-25
2025-04-25
ä¸å½ç§»å¨éä¿¡æéå ¬å¸ç 究é¢
Model training method and device, communication equipment and computer readable storage medium
WO2025118186A1
( en )
*
2023-12-06
2025-06-12
Oppo广ä¸ç§»å¨éä¿¡æéå ¬å¸
Signal modulation method, signal demodulation method, and communication device
CN120151917A
( en )
*
2023-12-12
2025-06-13
æé½å为ææ¯æéå ¬å¸
A communication method and a communication device
WO2025166741A1
( en )
*
2024-02-08
2025-08-14
å¯å£«éæ ªå¼ä¼ç¤¾
Data receiving method and apparatus, data sending method and apparatus, and communication system
CN120785384A
( en )
*
2024-04-03
2025-10-14
å为ææ¯æéå ¬å¸
Communication method and related device
CN120956566A
( en )
*
2024-05-13
2025-11-14
å为ææ¯æéå ¬å¸
A communication method and apparatus
CN121645315A
( en )
*
2024-09-04
2026-03-10
å为ææ¯æéå ¬å¸
A sensing method, device, system, storage medium, and program product
WO2026081135A1
( en )
*
2024-10-16
2026-04-23
Oppo广ä¸ç§»å¨éä¿¡æéå ¬å¸
Communication method and communication device
Citations (6)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
US20210110261A1
( en )
*
2019-10-10
2021-04-15
Samsung Electronics Co., Ltd.
Method and apparatus for transceiving signal using artificial intelligence in wireless communication system
US20220078777A1
( en )
*
2020-09-10
2022-03-10
Qualcomm Incorporated
Assigning time domain resources to fragmented orthogonal frequency division multiplexing (ofdm) symbols
US20220123966A1
( en )
*
2020-10-19
2022-04-21
Qualcomm Incorporated
Data-driven probabilistic modeling of wireless channels using conditional variational auto-encoders
US20220376957A1
( en )
*
2019-09-19
2022-11-24
Nokia Solution and Networka Oy
Machine Learning-Based Channel Estimation
US20230237841A1
( en )
*
2020-10-16
2023-07-27
Shanghai Bilibili Technology Co., Ltd.
Occlusion Detection
US20230300654A1
( en )
*
2020-07-03
2023-09-21
Telefonaktiebolaget Lm Ericsson (Publ)
Methods, UE and Network Node for Failure Predictions
Family Cites Families (9)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
CN111434049B
( en )
*
2017-06-19
2021-06-11
å¼åå°¼äºç§æç¥è¯äº§ææéå ¬å¸
Encoding and decoding of information transmitted wirelessly using a multi-antenna transceiver
CN110198282B
( en )
*
2018-02-27
2020-06-23
ä¸æµ·è¯ºåºäºè´å°è¡ä»½æéå ¬å¸
Method, apparatus and computer readable medium for channel equalization
US10531415B2
( en )
*
2018-03-02
2020-01-07
DeepSig Inc.
Learning communication systems using channel approximation
US10637551B2
( en )
*
2018-08-09
2020-04-28
At&T Intellectual Property I, L.P.
Generic reciprocity based channel state information acquisition frameworks for advanced networks
CN111628946B
( en )
*
2019-02-28
2021-10-26
å为ææ¯æéå ¬å¸
Channel estimation method and receiving equipment
US11533115B2
( en )
*
2019-05-15
2022-12-20
Huawei Technologies Co., Ltd.
Systems and methods for wireless signal configuration by a neural network
CN111954309B
( en )
*
2019-05-17
2025-12-19
æ ªå¼ä¼ç¤¾Ntté½ç§æ©
Terminals and base stations
CN110661734B
( en )
*
2019-09-20
2020-07-28
西å®äº¤é大å¦
Channel estimation method, device and readable storage medium based on deep neural network
CN111342867A
( en )
*
2020-02-28
2020-06-26
西å®äº¤é大å¦
MIMO iterative detection method based on deep neural network
2020
2020-12-31
CN
CN202080107089.5A
patent/CN116458103B/en
active
Active
2020-12-31
WO
PCT/CN2020/142103
patent/WO2022141397A1/en
not_active
Ceased
2020-12-31
CN
CN202510125089.2A
patent/CN120090899A/en
active
Pending
2020-12-31
EP
EP20967736.8A
patent/EP4262121A4/en
active
Pending
2023
2023-06-30
US
US18/345,904
patent/US20230342593A1/en
active
Pending
Patent Citations (6)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
US20220376957A1
( en )
*
2019-09-19
2022-11-24
Nokia Solution and Networka Oy
Machine Learning-Based Channel Estimation
US20210110261A1
( en )
*
2019-10-10
2021-04-15
Samsung Electronics Co., Ltd.
Method and apparatus for transceiving signal using artificial intelligence in wireless communication system
US20230300654A1
( en )
*
2020-07-03
2023-09-21
Telefonaktiebolaget Lm Ericsson (Publ)
Methods, UE and Network Node for Failure Predictions
US20220078777A1
( en )
*
2020-09-10
2022-03-10
Qualcomm Incorporated
Assigning time domain resources to fragmented orthogonal frequency division multiplexing (ofdm) symbols
US20230237841A1
( en )
*
2020-10-16
2023-07-27
Shanghai Bilibili Technology Co., Ltd.
Occlusion Detection
US20220123966A1
( en )
*
2020-10-19
2022-04-21
Qualcomm Incorporated
Data-driven probabilistic modeling of wireless channels using conditional variational auto-encoders
Cited By (14)
* Cited by examiner, â Cited by third party
Publication number
Priority date
Publication date
Assignee
Title
US12580795B2
( en )
*
2021-04-13
2026-03-17
Samsung Electronics Co., Ltd.
Electronic device for processing wireless signal, and operating method therefor
US20240187285A1
( en )
*
2021-04-15
2024-06-06
Nokia Solutions And Networks Oy
Machine learning for channel estimation against diverse doppler effects
US20240333601A1
( en )
*
2021-06-25
2024-10-03
Google Llc
Wireless network employing neural networks for channel state feedback
US12394191B2
( en )
*
2021-07-14
2025-08-19
Google Llc
Neural networks based multimodal transformer for multi-task user interface modeling
US20230031702A1
( en )
*
2021-07-14
2023-02-02
Google Llc
Neural Networks based Multimodal Transformer for Multi-Task User Interface Modeling
US20240396766A1
( en )
*
2021-09-14
2024-11-28
Telefonaktiebolaget Lm Ericsson (Publ)
Improved pilot assisted radio propagation channel estimation based on machine learning
US20240429985A1
( en )
*
2021-10-20
2024-12-26
Lg Electronics Inc.
Method and device for transmitting and receiving channel state information in wireless communication system
US20240323059A1
( en )
*
2021-12-03
2024-09-26
Guangdong Oppo Mobile Telecommunications Corp., Ltd.
Data acquisition method and apparatus, and device, medium, chip, product and program
US12476755B2
( en )
*
2022-08-19
2025-11-18
Nvidia Corporation
Reference signal configuration information transmission
US20240097843A1
( en )
*
2022-08-19
2024-03-21
Nvidia Corporation
Reference signal configuration information transmission
US20240154722A1
( en )
*
2022-10-27
2024-05-09
Samsung Electronics Co., Ltd.
Method and system for rate-matching auto-encoders
US20240430062A1
( en )
*
2023-06-20
2024-12-26
Qualcomm Incorporated
Ml based dynamic bit loading and rate control
US12621109B2
( en )
*
2023-06-20
2026-05-05
Qualcomm Incorporated
ML based dynamic bit loading and rate control
EP4597935A3
( en )
*
2024-01-31
2025-11-12
Nokia Solutions and Networks Oy
Communication systems
Also Published As
Publication number
Publication date
CN116458103A
( en )
2023-07-18
WO2022141397A1
( en )
2022-07-07
EP4262121A4
( en )
2024-03-13
CN120090899A
( en )
2025-06-03
EP4262121A1
( en )
2023-10-18
CN116458103B
( en )
2025-02-18
Similar Documents
Publication
Publication Date
Title
EP4262121A1
( en )
2023-10-18
Neural network training method and related apparatus
EP3852326B1
( en )
2022-06-29
Transmitter
US11929853B2
( en )
2024-03-12
Data-driven probabilistic modeling of wireless channels using conditional variational auto-encoders
JP7809818B2
( en )
2026-02-02
Communication method and apparatus
US20230259742A1
( en )
2023-08-17
Communication method, apparatus, and system
CN116318481A
( en )
2023-06-23
A communication method and device
CN115118566B
( en )
2024-06-04
Improved transmission of information in wireless communications
US11456834B2
( en )
2022-09-27
Adaptive demodulation reference signal (DMRS)
WO2024051789A1
( en )
2024-03-14
Beam management method
US10797930B1
( en )
2020-10-06
Apparatus and method for detection of received signals
EP4009597B1
( en )
2025-12-10
Demapping received data
WO2025054840A1
( en )
2025-03-20
Information transmission method and apparatus
WO2025118274A1
( en )
2025-06-12
Data transmission method and communication device
CN120238456A
( en )
2025-07-01
A communication method and related equipment
WO2025025002A9
( en )
2026-01-15
Wireless communication method and communication device
WO2025118271A1
( en )
2025-06-12
Wireless communication method and communication device
WO2025025016A1
( en )
2025-02-06
Wireless communication method and communication device
WO2025184808A1
( en )
2025-09-12
Information detection method, terminal device, and network device
CN121509266A
( en )
2026-02-10
A communication method and related apparatus
CN120263341A
( en )
2025-07-04
Communication method, device and storage medium
WO2024138692A1
( en )
2024-07-04
Spatial filter prediction methods, terminal device and network device
WO2025076724A1
( en )
2025-04-17
Method for transmitting reference signal, terminal device, and network device
WO2025224478A1
( en )
2025-10-30
Interference mitigation
CN121241353A
( en )
2025-12-30
Communication methods and communication devices
CN120018167A
( en )
2025-05-16
A communication method and related equipment
Legal Events
Date
Code
Title
Description
2023-07-25
STPP
Information on status: patent application and granting procedure in general
Free format text : DOCKETED NEW CASE - READY FOR EXAMINATION
Free format text : NON FINAL ACTION COUNTED, NOT YET MAILED
2025-09-03
AS
Assignment
Owner name : HUAWEI TECHNOLOGIES CO., LTD., CHINA
Free format text : ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:SUN, YAN;WU, YIQUN;REEL/FRAME:072144/0195
Effective date : 20250819
2026-02-04
STPP
Information on status: patent application and granting procedure in general
Free format text : NON FINAL ACTION COUNTED, NOT YET MAILED
2026-02-06
STPP
Information on status: patent application and granting procedure in general
Free format text : NON FINAL ACTION MAILED
2026-03-19
STPP
Information on status: patent application and granting procedure in general
Free format text : RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER
2026-04-14
STPP
Information on status: patent application and granting procedure in general
Free format text : NON FINAL ACTION COUNTED, NOT YET MAILED
2026-04-20
STPP
Information on status: patent application and granting procedure in general
Free format text : NON FINAL ACTION MAILED