Information acquisition method and communication device

Abstract

This application provides an information acquisition method and a communication device. The method relates to the field of communication technology and includes: a terminal device or access network device determining a first neural network model, the first neural network model being associated with a first reference signal density, and the first neural network model being used to determine a first reference signal pattern; then, acquiring communication resources of the first reference signal in the first reference signal pattern; and further, receiving the first reference signal on a first channel based on the communication resources of the first reference signal. This method enables efficient use of reference signal pattern designs based on neural networks between the terminal device and the access network device, saving signaling overhead and thus helping to ensure the accuracy of reconstructed channel information.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to information acquisition methods and communication devices. Background Technology

[0002] Machine learning is an important branch of artificial intelligence and a technological means to achieve artificial intelligence. Among the many research directions in machine learning, neural network algorithms have become a highly promising technique due to their ability to infinitely approximate any continuous function, granted by the universal approximation theorem. They can accurately abstract and model complex high-dimensional problems. Deep neural networks and deep learning have already achieved significant results in applications such as image processing, speech processing, and natural language processing.

[0003] With technological advancements, deep neural networks can also be applied to the physical layer of wireless communication (such as multiple-input multiple-output (MIMO)). For instance, reference signal pattern design based on neural networks can effectively select communication resources that facilitate the reconstruction of complete channel information. Furthermore, channel reconstruction based on neural networks can ensure the accuracy of the reconstructed channel information. However, currently, reference signal pattern design or channel reconstruction based on neural networks lacks compatibility with air interfaces, and there is a lack of corresponding technical details and specifications supporting interface and process design. Therefore, how to use reference signal pattern design or channel reconstruction based on neural networks between terminal devices and access network devices to ensure the accuracy of reconstructed channel information is a pressing issue that needs to be addressed. Summary of the Invention

[0004] This application provides an information acquisition method and a communication device. Based on the method described in this application, reference signal pattern design or channel reconstruction based on neural networks can be used efficiently between terminal devices and access network devices, thereby ensuring the accuracy of the reconstructed channel information.

[0005] Firstly, embodiments of this application provide an information acquisition method, which can be applied to a receiving device. Exemplarily, the method can be applied to the terminal side, such as a terminal or a communication / processing module within the terminal, or circuits or chips in the terminal responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-chip (SoC) chip containing a modem core, or a system-in-package (SIP) chip), or circuits or chips in the terminal responsible for processing functions (such as a graphics processing unit (GPU), an artificial intelligence (AI) processor, or an application-specific integrated circuit (ASIC)). The method can also be applied to the network side, such as network devices (e.g., access network devices), modules (e.g., circuits, chips, or chip systems) within the network devices, or logical nodes, logical modules, or software capable of implementing all or part of the functions of the network devices.

[0006] The method includes: determining a first neural network model, the first neural network model being associated with a first reference signal density, the first neural network model being used to determine a first reference signal pattern; acquiring communication resources of the first reference signal in the first reference signal pattern; and receiving the first reference signal on a first channel based on the communication resources of the first reference signal.

[0007] Using the above method, the reference signal density can be used to characterize the reference signal pattern design model between the terminal device and the network device. That is, the reference signal density can be associated with the corresponding reference signal pattern design model (for example, the first reference signal density can be associated with the corresponding first neural network model). This allows the terminal device and the access network device to use the reference signal pattern design based on the neural network more efficiently, thereby ensuring the accuracy of the reconstructed channel information. At the same time, the selection of the neural network model can be completed by only obtaining the reference signal density, which helps to reduce signaling overhead and improve communication efficiency.

[0008] In one possible implementation, a first reference signal density is associated with a second neural network model, which is used to reconstruct channel information; the method further includes: determining first channel information corresponding to a first channel based on the second neural network model and the first reference signal.

[0009] In this way, the terminal device can reconstruct the first channel information using the second neural network model and the first reference signal.

[0010] In one possible implementation, determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal includes: determining the second channel information corresponding to the communication resources of the first reference signal based on the first reference signal; and processing the second channel information using the second neural network model to obtain the first channel information corresponding to the first channel.

[0011] In this way, after receiving the first reference signal p′, the terminal device can perform channel estimation based on the first reference signal to obtain the second channel information corresponding to the communication resources of the first reference signal on the first channel. (i.e., channel information of the first reference signal resource location), The channel estimation method f(·) includes, but is not limited to, the minimum mean squared error (MMSE) algorithm. Then, the second channel information... The information is input into the second neural network model for processing to obtain the complete channel information of the first channel, i.e., the first channel information, thereby realizing the reconstruction of the complete channel information.

[0012] In one possible implementation, acquiring the communication resources of the first reference signal in the first reference signal pattern includes: receiving first indication information, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern.

[0013] In this way, taking the network device as the model training end and the terminal device as the non-model training end as an example, the network device uses the first neural network model to determine the first reference signal pattern, which indicates the communication resources of the first reference signal; and then indicates the communication resources of the first reference signal in the first reference signal pattern to the terminal device to achieve the alignment of communication resources.

[0014] In one possible implementation, the method further includes receiving a second neural network model.

[0015] In this way, taking the network device as the model training end and the terminal device as the non-model training end as an example, the network device stores the second neural network model. After the network device determines the second neural network model, it will send the second neural network model to the terminal device so that the terminal device can use the second neural network model to realize channel reconstruction.

[0016] In one possible implementation, the method further includes receiving second indication information, which is used to indicate that the first neural network model and the second neural network model are updated.

[0017] In this way, taking the network device as the model training end and the terminal device as the non-model training end as an example, the network device can update the neural network model, which is beneficial to improving the flexibility and accuracy of the neural network model. At the same time, the network device will notify the terminal device that the first neural network model and the second neural network model have been updated to ensure information alignment between the network device and the terminal device.

[0018] In one possible implementation, the method further includes: sending first indication information, the first indication information being used to indicate communication resources of a first reference signal in a first reference signal pattern.

[0019] In this way, taking the terminal device as the model training end and the network device as the non-model training end as an example, the terminal device uses the first neural network model to determine the first reference signal pattern, which indicates the communication resources of the first reference signal; and then indicates the communication resources of the first reference signal in the first reference signal pattern to the network device to achieve the alignment of communication resources.

[0020] In one possible implementation, the method further includes: updating a first neural network model and a second neural network model; determining a change value corresponding to a first reference signal density based on the updated first neural network model and the updated second neural network model; and sending second indication information to indicate that the first neural network model and the second neural network model have been updated.

[0021] In this way, taking the terminal device as the model training end and the network device as the non-model training end as an example, the terminal device can update the neural network model, which helps to improve the flexibility and accuracy of the neural network model. At the same time, the terminal device will notify the network device that the first neural network model and the second neural network model have been updated to ensure information alignment between the network device and the terminal device.

[0022] In one possible implementation, the second indication information includes the change value corresponding to the first reference signal density.

[0023] This method allows terminal devices and network devices to adjust the density of the first reference signal in a timely manner.

[0024] In one possible implementation, updating the first neural network model and the second neural network model includes: obtaining a first time-domain unit, after which a first dataset used for training the first neural network model and the second neural network model is updated; and updating the first neural network model and the second neural network model based on the first dataset.

[0025] This approach helps improve the flexibility and accuracy of neural network models.

[0026] In one possible implementation, the method further includes: if the first neural network model is switched to the third neural network model and the second neural network model is switched to the fourth neural network model, and the first condition is met, then the first neural network model and the second neural network model are updated.

[0027] This approach helps improve the flexibility and accuracy of neural network models.

[0028] In one possible implementation, the first condition includes: the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model.

[0029] This approach helps ensure the accuracy of the first and second neural network models in subsequent applications.

[0030] In one possible implementation, the method further includes: sending a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0031] In this way, taking a terminal device as an example, the first reference signal density is selected by the terminal device from the reference signal densities supported by the network device, and the selected first reference signal density is informed to the network device to ensure information alignment.

[0032] In one possible implementation, the method further includes: receiving first information, the first information being used to indicate an identifier of a supported reference signal density or neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0033] In this way, taking a terminal device as an example, the network device can broadcast the identifiers of the reference signal density or neural network model it supports, and then the terminal device can select from them, which helps to improve the flexibility of reference signal density selection.

[0034] In one possible implementation, before receiving the first information, the method further includes sending a first request for an identifier of a supported reference signal density or neural network model.

[0035] In this way, taking a terminal device as an example, if the network device does not broadcast the identifier of the reference signal density or neural network model it supports, the terminal device can also actively request the identifier of the reference signal density or neural network model supported by the network device, which helps to improve the diversity of ways for the terminal device to obtain the identifier of the reference signal density or neural network model supported by the network device.

[0036] In one possible implementation, the method further includes sending third indication information, which is used to indicate the second reference signal density.

[0037] In this way, taking a terminal device as an example, if the current environmental characteristics change, the terminal device can also reselect the reference signal density, which helps to improve the flexibility of reference signal density selection.

[0038] In one possible implementation, the method further includes receiving fourth indication information, which is used to indicate the third reference signal density.

[0039] In this way, taking terminal devices as an example, network devices can also disregard the first reference signal density selected by the terminal device and reconfigure the reference signal density for the terminal device based on its own hardware support, environmental characteristics, feedback from the terminal device, etc., which helps to improve the flexibility of reference signal density selection.

[0040] In one possible implementation, the method further includes receiving a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0041] In this way, taking a terminal device as an example, the first reference signal density is selected by the network device from the reference signal densities supported by the network device, and the selected first reference signal density is informed to the terminal device to ensure information alignment.

[0042] In one possible implementation, the method further includes: sending capability information for indicating a reference signal pattern design that supports a neural network model, or an identifier of a supported reference signal density or neural network model; the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0043] In this way, taking terminal devices as an example, the terminal devices can report the identifiers of the reference signal densities or neural network models they support, and then the network devices can select from them, which helps to improve the flexibility of reference signal density selection.

[0044] In one possible implementation, the method further includes receiving fifth indication information, which is used to indicate a fourth reference signal density.

[0045] In this way, taking terminal devices as an example, if the current environmental characteristics change, the network devices can also reselect the reference signal density, which helps to improve the flexibility of reference signal density selection.

[0046] In one possible implementation, the method further includes sending a sixth indication message, which is used to indicate the fifth reference signal density.

[0047] In this way, taking terminal devices as an example, terminal devices can also disregard the first reference signal density selected by the network device and reconfigure the reference signal density for the network device based on their own hardware support, environmental characteristics, and feedback from the network device, which helps to improve the flexibility of reference signal density selection.

[0048] In one possible implementation, the first reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

[0049] Secondly, embodiments of this application provide an information acquisition method, which can be applied to a transmitting device. Exemplarily, the method can be applied to the network side, such as network devices (e.g., access network devices), modules (e.g., circuits, chips, or chip systems) within the network devices, or logical nodes, logical modules, or software capable of implementing all or part of the functions of the network devices. The method can also be applied to the terminal side, such as a terminal or a communication / processing module within the terminal, or circuits or chips in the terminal responsible for communication functions (e.g., modem chips, also known as baseband chips, or system-on-chip (SoC) chips containing modem cores, or system-in-package (SIP) chips), or circuits or chips in the terminal responsible for processing functions (e.g., graphics processing units (GPUs), artificial intelligence (AI) processors, or application-specific integrated circuits (ASICs)).

[0050] The method includes: determining a first neural network model, the first neural network model being associated with a first reference signal density, the first neural network model being used to determine a first reference signal pattern; acquiring communication resources of the first reference signal in the first reference signal pattern; and transmitting the first reference signal on a first channel based on the communication resources of the first reference signal.

[0051] In the embodiments of this application, the beneficial effects of possible implementations of the second aspect can be referred to the beneficial effects of possible implementations of the first aspect, and will not be repeated here.

[0052] In one possible implementation, the first reference signal density is associated with a second neural network model, which is used to reconstruct the channel information.

[0053] In one possible implementation, the method further includes: sending first indication information, the first indication information being used to indicate communication resources of a first reference signal in a first reference signal pattern.

[0054] In one possible implementation, the method further includes sending a second neural network model.

[0055] In one possible implementation, the method further includes: updating a first neural network model and a second neural network model; determining the change value corresponding to the first reference signal density based on the updated first neural network model and the updated second neural network model; and sending second indication information to indicate that the first neural network model and the second neural network model have been updated.

[0056] In one possible implementation, updating the first neural network model and the second neural network model includes: obtaining a first time-domain unit, after which a first dataset used for training the first neural network model and the second neural network model is updated; and updating the first neural network model and the second neural network model based on the first dataset.

[0057] In one possible implementation, the method further includes: if the first neural network model is switched to the third neural network model and the second neural network model is switched to the fourth neural network model, and the first condition is met, then the first neural network model and the second neural network model are updated.

[0058] In one possible implementation, the first condition includes: the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model.

[0059] In one possible implementation, acquiring the communication resources of the first reference signal in the first reference signal pattern includes: receiving first indication information, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern.

[0060] In one possible implementation, the method further includes receiving second indication information, which is used to indicate that the first neural network model and the second neural network model are updated.

[0061] In one possible implementation, the second indication information includes the change value corresponding to the first reference signal density.

[0062] In one possible implementation, the method further includes receiving a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0063] In one possible implementation, the method further includes: sending first information for indicating an identifier of a supported reference signal density or neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0064] In one possible implementation, before sending the first information, the method further includes receiving a first request for an identifier of a supported reference signal density or neural network model.

[0065] In one possible implementation, the method further includes receiving third indication information, which is used to indicate a second reference signal density.

[0066] In one possible implementation, the method further includes sending a fourth indication message, which is used to indicate a third reference signal density.

[0067] In one possible implementation, the method further includes: sending a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0068] In one possible implementation, the method further includes: receiving capability information for indicating a reference signal pattern design that supports a neural network model, or an identifier of a supported reference signal density or neural network model; the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0069] In one possible implementation, the method further includes sending a fifth indication message, which is used to indicate the fourth reference signal density.

[0070] In one possible implementation, the method further includes receiving sixth indication information, which is used to indicate the fifth reference signal density.

[0071] In one possible implementation, the first reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

[0072] Thirdly, embodiments of this application provide an information acquisition method that can be applied to a receiving device. Exemplarily, this method can be applied to the terminal side, such as a terminal or a communication / processing module within the terminal, or circuits or chips in the terminal responsible for communication functions (e.g., modem chips, also known as baseband chips, or system-on-chip (SoC) chips containing modem cores, or system-in-package (SIP) chips), or circuits or chips in the terminal responsible for processing functions (e.g., graphics processing unit (GPU), artificial intelligence (AI) processors, or application-specific integrated circuits (ASICs)). This method can also be applied to the network side, such as network devices (e.g., access network devices), modules (e.g., circuits, chips, or chip systems) within the network devices, or logical nodes, logical modules, or software capable of implementing all or part of the functions of the network devices.

[0073] The method includes: determining a second neural network model, the second neural network model being associated with a first reference signal density, the second neural network model being used to reconstruct channel information; receiving a first reference signal on a first channel; and determining first channel information corresponding to the first channel based on the second neural network model and the first reference signal.

[0074] Using the above method, the terminal device and network device can use reference signal density to characterize the channel reconstruction model. That is, the reference signal density can be associated with the corresponding channel reconstruction model (for example, the first reference signal density can be associated with the corresponding second neural network model). This allows the terminal device and network device to use the neural network-based channel reconstruction more efficiently, thereby ensuring the accuracy of the reconstructed channel information. At the same time, only the reference signal density needs to be obtained to complete the selection of the neural network model, which helps to reduce signaling overhead and improve communication efficiency.

[0075] Furthermore, in the embodiments of this application, the beneficial effects of possible implementations of the third aspect can be found in the beneficial effects of possible implementations of the first aspect, and will not be repeated here.

[0076] In one possible implementation, the first reference signal density is associated with a first neural network model, the first neural network model being used to determine the first reference signal pattern; the method further includes: acquiring communication resources of the first reference signal in the first reference signal pattern; receiving the first reference signal on a first channel, including: receiving the first reference signal on the first channel based on the communication resources of the first reference signal.

[0077] In one possible implementation, determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal includes: determining the second channel information corresponding to the communication resources of the first reference signal based on the first reference signal; and processing the second channel information using the second neural network model to obtain the first channel information corresponding to the first channel.

[0078] In one possible implementation, acquiring the communication resources of the first reference signal in the first reference signal pattern includes: receiving first indication information, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern.

[0079] In one possible implementation, the method further includes receiving a second neural network model.

[0080] In one possible implementation, the method further includes: sending first indication information, the first indication information being used to indicate communication resources of a first reference signal in a first reference signal pattern.

[0081] In one possible implementation, the method further includes receiving second indication information, which is used to indicate that the first neural network model and the second neural network model are updated.

[0082] In one possible implementation, the method further includes: sending first indication information, the first indication information being used to indicate communication resources of a first reference signal in a first reference signal pattern.

[0083] In one possible implementation, the method further includes: updating a first neural network model and a second neural network model; determining a change value corresponding to a first reference signal density based on the updated first neural network model and the updated second neural network model; and sending second indication information to indicate that the first neural network model and the second neural network model have been updated.

[0084] In one possible implementation, the second indication information includes the change value corresponding to the first reference signal density.

[0085] In one possible implementation, updating the first neural network model and the second neural network model includes: obtaining a first time-domain unit, after which a first dataset used for training the first neural network model and the second neural network model is updated; and updating the first neural network model and the second neural network model based on the first dataset.

[0086] In one possible implementation, the method further includes: if the first neural network model is switched to the third neural network model and the second neural network model is switched to the fourth neural network model, and the first condition is met, then the first neural network model and the second neural network model are updated.

[0087] In one possible implementation, the first condition includes: the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model.

[0088] In one possible implementation, the method further includes: sending a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0089] In one possible implementation, the method further includes: receiving first information, the first information being used to indicate an identifier of a supported reference signal density or neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0090] In one possible implementation, before receiving the first information, the method further includes sending a first request for an identifier of a supported reference signal density or neural network model.

[0091] In one possible implementation, the method further includes sending third indication information, which is used to indicate the second reference signal density.

[0092] In one possible implementation, the method further includes receiving fourth indication information, which is used to indicate the third reference signal density.

[0093] In one possible implementation, the method further includes receiving a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0094] In one possible implementation, the method further includes: sending capability information for indicating a reference signal pattern design that supports a neural network model, or an identifier of a supported reference signal density or neural network model; the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0095] In one possible implementation, the method further includes receiving fifth indication information, which is used to indicate a fourth reference signal density.

[0096] In one possible implementation, the method further includes sending a sixth indication message, which is used to indicate the fifth reference signal density.

[0097] In one possible implementation, the first reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

[0098] Fourthly, embodiments of this application provide an information acquisition method, which can be applied to a transmitting device. Exemplarily, the method can be applied to the network side, such as network devices (e.g., access network devices), modules (e.g., circuits, chips, or chip systems) within the network devices, or logical nodes, logical modules, or software capable of implementing all or part of the functions of the network devices. The method can also be applied to the terminal side, such as a terminal or a communication / processing module within the terminal, or circuits or chips in the terminal responsible for communication functions (e.g., modem chips, also known as baseband chips, or system-on-chip (SoC) chips containing modem cores, or system-in-package (SIP) chips), or circuits or chips in the terminal responsible for processing functions (e.g., graphics processing units (GPUs), artificial intelligence (AI) processors, or application-specific integrated circuits (ASICs)).

[0099] The method includes: determining a second neural network model, the second neural network model being associated with a first reference signal density, the second neural network model being used to reconstruct channel information; and transmitting a first reference signal on a first channel.

[0100] In the embodiments of this application, the beneficial effects of possible implementations of the fourth aspect can be referred to the beneficial effects of possible implementations of the third aspect, and will not be repeated here.

[0101] In one possible implementation, the first reference signal density is associated with a first neural network model, the first neural network model being used to determine the first reference signal pattern; the method further includes: acquiring communication resources of the first reference signal in the first reference signal pattern; transmitting the first reference signal on a first channel, including: transmitting the first reference signal on the first channel based on the communication resources of the first reference signal.

[0102] In one possible implementation, the method further includes: sending first indication information, the first indication information being used to indicate communication resources of a first reference signal in a first reference signal pattern.

[0103] In one possible implementation, the method further includes sending a second neural network model.

[0104] In one possible implementation, the method further includes: updating a first neural network model and a second neural network model; determining a change value corresponding to a first reference signal density based on the updated first neural network model and the updated second neural network model; and sending second indication information to indicate that the first neural network model and the second neural network model have been updated.

[0105] In one possible implementation, updating the first neural network model and the second neural network model includes: obtaining a first time-domain unit, after which a first dataset used for training the first neural network model and the second neural network model is updated; and updating the first neural network model and the second neural network model based on the first dataset.

[0106] In one possible implementation, the method further includes: if the first neural network model is switched to the third neural network model and the second neural network model is switched to the fourth neural network model, and the first condition is met, then the first neural network model and the second neural network model are updated.

[0107] In one possible implementation, the first condition includes: the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model.

[0108] In one possible implementation, acquiring the communication resources of the first reference signal in the first reference signal pattern includes: receiving first indication information, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern.

[0109] In one possible implementation, the method further includes receiving second indication information, which is used to indicate that the first neural network model and the second neural network model are updated.

[0110] In one possible implementation, the second indication information includes the change value corresponding to the first reference signal density.

[0111] In one possible implementation, the method further includes receiving a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0112] In one possible implementation, the method further includes: sending first information for indicating an identifier of a supported reference signal density or neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0113] In one possible implementation, before sending the first information, the method further includes receiving a first request for an identifier of a supported reference signal density or neural network model.

[0114] In one possible implementation, the method further includes receiving third indication information, which is used to indicate a second reference signal density.

[0115] In one possible implementation, the method further includes sending a fourth indication message, which is used to indicate a third reference signal density.

[0116] In one possible implementation, the method further includes: sending a first reference signal density or a first identifier, the first identifier being used to indicate the first reference signal density.

[0117] In one possible implementation, the method further includes: receiving capability information for indicating a reference signal pattern design that supports a neural network model, or an identifier of a supported reference signal density or neural network model; the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

[0118] In one possible implementation, the method further includes sending a fifth indication message, which is used to indicate the fourth reference signal density.

[0119] In one possible implementation, the method further includes receiving sixth indication information, which is used to indicate the fifth reference signal density.

[0120] In one possible implementation, the first reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

[0121] Fifthly, embodiments of this application provide a communication device for executing the method in any one of the first to fourth aspects or any possible implementation of any one of the first to fourth aspects. The communication device includes modules having the capability to execute the method in any one of the first to fourth aspects or any possible implementation of any one of the first to fourth aspects.

[0122] Sixthly, embodiments of this application provide a communication device including a processing circuit for executing a method from any one of the first to fourth aspects or any possible implementation thereof. The processing circuit executes a program stored in a memory, and when the program is executed, the method described in any one of the first to fourth aspects or any possible implementation thereof is executed.

[0123] In one possible implementation, the memory is located outside the aforementioned communication device.

[0124] In one possible implementation, the memory is located within the aforementioned communication device.

[0125] In this embodiment, the processing circuitry and memory can also be integrated into a single device; that is, the processing circuitry and memory can be integrated together. For example, the communication device can be a chip.

[0126] In one possible implementation, the communication device further includes a transceiver circuit for receiving information (or inputting information) or sending information (or outputting information).

[0127] In a seventh aspect, embodiments of this application provide a communication device, which includes a processing circuit and a transceiver circuit. The processing circuit can be a logic circuit, and the transceiver circuit can be an interface circuit. The logic circuit and the interface circuit are coupled. The interface circuit is used to input and / or output information, and the logic circuit is used to execute a method in any one of the first to fourth aspects or any possible implementation of any one of the first to fourth aspects.

[0128] Eighthly, embodiments of this application provide a chip including a processing circuit and an interface circuit, the processing circuit and the interface circuit being coupled; the interface circuit is used for inputting and / or outputting information, and the processing circuit is used for executing code instructions to cause the method shown in any of the first to fourth aspects or any possible implementation thereof to be executed.

[0129] Ninthly, embodiments of this application provide a computer-readable storage medium for storing a computer program that, when run on a computer, causes the methods shown in any of the first to fourth aspects or any possible implementation thereof to be executed.

[0130] In a tenth aspect, embodiments of this application provide a computer program product that, when run on a computer, causes the methods shown in any of the first to fourth aspects or any possible implementations above to be executed.

[0131] In one aspect, this application provides a communication system including a terminal device and an access network device. The terminal device is used to perform the method shown in the first aspect or any possible implementation thereof, and the access network device is used to perform the method shown in the second aspect or any possible implementation thereof.

[0132] In a twelfth aspect, this application provides a communication system comprising a terminal device and an access network device. The terminal device is configured to perform the method shown in the third aspect or any possible implementation thereof, and the access network device is configured to perform the method shown in the fourth aspect or any possible implementation thereof. Attached Figure Description

[0133] Figure 1 This is a schematic diagram of the architecture of a communication system provided in an embodiment of this application;

[0134] Figure 2A This is a schematic diagram of a possible application framework in a communication system provided in an embodiment of this application;

[0135] Figure 2B This is a schematic diagram of another possible application framework in a communication system provided in the embodiments of this application;

[0136] Figure 3 This is a schematic diagram of a feedforward neural network provided in an embodiment of this application;

[0137] Figure 4A This is a schematic diagram of a block-shaped pilot arrangement provided in an embodiment of this application;

[0138] Figure 4B This is a schematic diagram of a comb-shaped pilot arrangement provided in an embodiment of this application;

[0139] Figure 4C This is a schematic diagram of a lattice-type pilot arrangement provided in an embodiment of this application;

[0140] Figure 4D This is a schematic diagram of a reference signal pattern design based on a neural network model provided in an embodiment of this application;

[0141] Figure 5 This is a flowchart illustrating an information acquisition method provided in an embodiment of this application;

[0142] Figure 6A This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0143] Figure 6B This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0144] Figure 6C This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0145] Figure 6D This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0146] Figure 6E This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0147] Figure 7 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0148] Figure 8 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0149] Figure 9 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application;

[0150] Figure 10 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0151] Figure 11 This is a schematic diagram of another communication device provided in an embodiment of this application;

[0152] Figure 12 This is a schematic diagram of another communication device provided in an embodiment of this application. Detailed Implementation

[0153] To facilitate understanding of the technical solution of this application, the application will be further described below with reference to the accompanying drawings.

[0154] The terms "first" and "second," etc., used in the specification, claims, and drawings of this application are used only to distinguish different objects and not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0155] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0156] In this application, "at least one (item)" refers to one or more, "more than one" refers to two or more, "at least two (items)" refers to two or three or more, and "and / or" is used to describe the relationship between related objects, indicating that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. "Or" indicates that there can be two relationships, such as only A exists and only B exists; when A and B are not mutually exclusive, it can also mean that there are three relationships, such as only A exists, only B exists, and both A and B exist simultaneously. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items. For example, at least one (item) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c".

[0157] In this application, "send" and "receive" indicate the direction of signal transmission. For example, "send information to XX" can be understood as the destination of the information being XX, which can include direct transmission via the air interface or indirect transmission via the air interface from other units or modules. "Receive information from YY" can be understood as the source of the information being YY, which can include direct reception from YY via the air interface or indirect reception from YY via the air interface from other units or modules. "Send" can also be understood as the "output" of a chip interface, and "receive" can also be understood as the "input" of a chip interface. In other words, sending and receiving can occur between devices, such as between network devices and terminal devices, or within a device, such as between components, modules, chips, software modules, or hardware modules within the device via buses, traces, or interfaces.

[0158] To better understand the embodiments of this application, the communication system involved in the embodiments of this application will be described below:

[0159] The method provided in this application can be applied to various communication systems, such as: wireless local area network (WLAN) communication systems, wireless fidelity (Wi-Fi) systems, multiple-in multiple-out (MIMO) communication systems, long-term evolution (LTE) systems, internet of things (IoT) systems, narrowband internet of things (NB-IoT) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, fourth-generation (4G) systems, fifth-generation (5G) systems, or new radio (NR) systems, and other future communication systems, such as sixth-generation (6G) systems. Among these, IoT networks may include, but are not limited to, vehicle-to-everything (V2X) networks. The communication methods in V2X systems can be collectively referred to as vehicle-to-everything (V2X), where X can represent anything. For example, V2X can include vehicle-to-vehicle (V2V) communication, vehicle-to-infrastructure (V2I) communication, vehicle-to-pedestrian (V2P) communication, or vehicle-to-network (V2N) communication. The method provided in this application also supports communication systems that integrate multiple wireless technologies. For example, it can be applied to systems that integrate non-terrestrial networks (NTN) with terrestrial mobile communication networks, such as drones, satellite communication systems, and high-altitude platform station (HAPS) communication. Additionally, it can be applied to low-frequency (sub-6GHz) and high-frequency (above 6GHz) communication scenarios. It is understood that the system architecture described in this application is for the purpose of more clearly illustrating the technical solutions of this application and does not constitute a limitation on the technical solutions provided in this application.

[0160] Figure 1 This is a schematic diagram of the architecture of a communication system applicable to embodiments of this application. For example... Figure 1As shown, the communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. RAN 100 includes at least one RAN node (e.g., ...). Figure 1 110a and 110b, collectively referred to as 110) and at least one terminal (such as Figure 1 RAN 100, denoted as RAN 120a-120j, is collectively referred to as RAN 120. RAN 100 may also include other RAN nodes, such as wireless relay equipment and / or wireless backhaul equipment. Figure 1 (Not shown in the image). Terminal 120 is connected to RAN node 110 wirelessly. RAN node 110 is connected to core network 200 wirelessly or via wired connection. The core network equipment in core network 200 and RAN node 110 in RAN 100 can be different physical devices, or they can be the same physical device integrating core network logical functions and radio access network logical functions.

[0161] RAN 100 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G mobile communication systems, or future-oriented evolution systems. RAN 100 can also be an open access network (O-RAN or ORAN), a cloud radio access network (CRAN), or a wireless fidelity (WiFi) system. RAN 100 can also be a communication system that integrates two or more of the above systems.

[0162] RAN node 110, sometimes also referred to as access network equipment, network device, RAN entity, or access node, constitutes part of the communication system and is used to help terminals achieve wireless access. Multiple RAN nodes 110 in communication system 10 can be of the same type or different types. In some scenarios, the roles of RAN node 110 and terminal 120 are relative, for example... Figure 1 Network element 120i can be a helicopter or a drone, and it can be configured as a mobile base station. For terminals 120j that access RAN 100 through network element 120i, network element 120i is a base station; however, for base station 110a, network element 120i is a terminal. RAN node 110 and terminal 120 are sometimes referred to as communication devices, for example... Figure 1 Network elements 110a and 110b can be understood as communication devices with base station functions, while network elements 120a-120j can be understood as communication devices with terminal functions.

[0163] In one possible scenario, a RAN node can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a base station in a future mobile communication system, or an access node in a WiFi system, etc. Figure 1 110a), micro base stations or indoor stations (such as Figure 1 The RAN node can be a relay node or donor node (as described in section 110b), or a wireless controller in a CRAN scenario. Optionally, the RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). All or part of the functions of the RAN node in this application can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node can also be equipped with communication modules, circuits, or chips that perform corresponding communication functions. The RAN node can also be configured with program instructions for performing corresponding communication functions and corresponding program instructions. The RAN node in this application can also be a logical node, logical module, or software capable of implementing all or part of the RAN node functions.

[0164] In another possible scenario, multiple RAN nodes collaborate to assist the terminal in achieving wireless access, with different RAN nodes each implementing a portion of the base station's functions. For example, RAN nodes can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0165] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, this application uses CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules.

[0166] A terminal can be a device or module that accesses the aforementioned communication system and has corresponding communication functions. A terminal can also be called a terminal device, user equipment (UE), mobile station, mobile terminal, etc. Terminals can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), extended reality (ER), virtual reality (VR), augmented reality (AR), industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, etc. Terminals can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices (such as smartwatches, smart bracelets, pedometers, etc.), vehicles, drones, helicopters, airplanes, ships, robots, robotic arms, smart home devices, transportation vehicles with wireless communication capabilities, communication modules, etc. The embodiments of this application do not limit the device form of the terminal. Terminals typically contain communication modules, circuits, or chips that perform the corresponding communication functions. They may also contain program instructions configured to perform these functions.

[0167] Core network equipment refers to the equipment in the core network that provides service support for terminal devices. It is primarily responsible for registration, call setup, billing, mobility management, providing user connectivity, managing users, and carrying out service delivery, data processing, and routing. Core network equipment can correspond to different devices in different communication systems. For example, in a 4G communication system, it may correspond to one or more of the following: Mobility Management Entity (MME), Serving Gateway (S-GW), etc. Similarly, in a 5G communication system, it may correspond to one or more of the following: Access and Mobility Management Function (AMF) network elements, Session Management Function (SMF) network elements, User Plane Function (UPF) network elements, etc. In next-generation or future communication systems, it may correspond to one or more network elements, devices, or entities that provide service support for terminal devices.

[0168] It should be noted that, Figure 1 The communication system shown is not limited to the terminal equipment, access network equipment and core network equipment shown in the figure, but may also include other equipment not shown in the figure. These will not be listed here.

[0169] To support artificial intelligence (AI) technology in wireless networks, AI nodes may be introduced into the network for model training, model inference, and other purposes.

[0170] AI nodes can be deployed in one or more of the following locations within the communication system: access network nodes (RAN nodes), terminal devices, or core network devices. Alternatively, AI nodes can be deployed independently, for example, in a location other than any of the aforementioned devices, such as in the host or cloud server of an over-the-top (OTT) system. AI nodes can communicate with other devices in the communication system, which can be one or more of the following: network devices, terminal devices, or core network elements.

[0171] It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, they can be divided based on function, such as different AI nodes being responsible for different functions.

[0172] It can also be understood that AI nodes can be independent devices, or they can be integrated into the same device to achieve different functions. Alternatively, they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the aforementioned AI nodes.

[0173] AI nodes can be AI network elements or AI modules.

[0174] Figure 2A This is a schematic diagram of a possible application framework in a communication system provided in an embodiment of this application. For example... Figure 2A As shown, network elements in a communication system are connected via interfaces (e.g., NG, Xn) or air interfaces. These network element nodes, such as core network equipment, access network nodes (RAN nodes), terminals, or one or more devices in operations administration and maintenance (OAM), are equipped with one or more AI modules (for clarity, ...). Figure 2A (Only one is shown in the image). An access network node can be a single RAN node or can include multiple RAN nodes, such as a CU and a DU. The CU and / or DU can also be equipped with one or more AI modules. The CU can also be split into CU-CP and CU-UP, and one or more AI modules can be set in the CU-CP and / or CU-UP.

[0175] AI modules are used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. The models of AI modules can achieve different functions depending on the parameter configurations. The models of AI modules can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or biases in the activation function), input parameters (e.g., the type and / or dimension of the input parameters), or output parameters (e.g., the type and / or dimension of the output parameters). The biases in the activation function can also be referred to as the biases of the neural network.

[0176] In one example, the neural network mentioned above could be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), or a generative adversarial network (GAN).

[0177] Deep Neural Networks (DNNs) are artificial neural network architectures with multiple layers of nonlinear transformation units stacked in a hierarchical structure to form deep computational models. Compared to shallow neural networks, deep neural networks have more hidden layers, allowing the network model to capture more complex data structures and higher-level abstract features.

[0178] A CNN is a deep neural network with a convolutional structure. A CNN contains a feature extractor consisting of convolutional layers and subsampling layers. This feature extractor can be viewed as a filter, and the convolution process can be seen as performing convolution between a trainable filter and an input image or a convolutional feature map.

[0179] RNN is a type of recursive neural network that takes sequence data as input, recursively moves along the direction of sequence evolution, and connects all nodes (recurrent units) in a chain-like manner.

[0180] GAN is a deep learning model. It consists of a generator and a discriminator, and is trained through adversarial learning. Its purpose is to estimate the potential distribution of data samples and generate new data samples.

[0181] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0182] Figure 2B This is a schematic diagram of another possible application framework in a communication system provided in the embodiments of this application. For example... Figure 2B As shown, the communication system includes a RAN intelligent controller (RIC). For example, the RIC could be... Figure 2A The AI ​​module shown is used to implement AI-related functions. RICs include near-real-time RICs (near-RT RICs) and non-real-time RICs (non-RT RICs). Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency in the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency in the order of tens of milliseconds.

[0183] Near real-time (NRT) RICs are used for model training and inference. For example, they are used to train AI models and then use those models for inference. NRT RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, and / or RUs) and / or terminals. This information can be used as training data or inference data. NRT RICs can deliver inference results to RAN nodes and / or terminals. Inference results can be exchanged between CUs and DUs, and / or between DUs and RUs. For example, a NRT RIC delivers an inference result to a DU, which then forwards it to an RU.

[0184] Non-real-time RICs are also used for model training and inference. For example, they are used to train AI models and then use those models for inference. Non-real-time RICs can obtain network-side and / or terminal-side information from RAN nodes (e.g., CUs, CU-CPs, CU-UPs, DUs, and / or RUs) and / or terminals. This information can be used as training data or inference data, and the inference results can be delivered to RAN nodes and / or terminals. Inference results can be exchanged between CUs and DUs, and / or between DUs and RUs; for example, a non-real-time RIC delivers inference results to a DU, which then forwards them to an RU.

[0185] Near real-time RICs and non-real-time RICs can also be configured as separate network elements. Near real-time RICs and non-real-time RICs can also be part of other devices. For example, near real-time RICs can be set in RAN nodes (e.g., CU, DU), while non-real-time RICs can be set in OAM, cloud servers, core network devices, or other network devices.

[0186] To facilitate understanding of the solutions provided in the embodiments of this application, the relevant concepts involved in the embodiments of this application are introduced below:

[0187] 1. Neural Network Model

[0188] Machine learning is an important branch of artificial intelligence and a technological means to achieve artificial intelligence. Among the many research directions in machine learning, neural network algorithms, due to their ability to infinitely approximate any continuous function given by the universal approximation theorem, have become a highly promising technique, capable of accurately abstracting and modeling complex high-dimensional problems. Deep neural networks and deep learning have already achieved significant results in applications such as image processing, speech processing, and natural language processing.

[0189] Deep neural networks (DNNs) are typically multi-layered structures. The model of a neural network is similar to how brain cells transmit neural signals. Neural network models are described based on mathematical models of neurons. Increasing the depth and width of a neural network can improve its expressive power, providing more powerful information extraction and abstract modeling capabilities for complex systems.

[0190] Generally, neural networks can be divided into three main types: feedforward neural networks, inverse neural networks, and graph neural networks. Taking a feedforward neural network (FNN) as an example... Figure 3 As shown, a feedforward neural network is a simple type of neural network, also known as a multilayer perceptron (MLP). Different neurons belong to different layers, consisting of an input layer, hidden layers, and an output layer. The input layer receives the information to be processed; the hidden layers extract features to varying degrees (DNNs typically contain more than one hidden layer); and the output layer maps the extracted features to the desired output information. The circular symbols represent the neurons in each layer. The connection method and activation function used in each layer determine the neural network's expression function.

[0191] Feedforward neural networks (DNNs) include activation functions (such as sigmoid and tanh functions), loss functions (such as mean squared error loss and cross-entropy loss), and optimization algorithms (such as back propagation (BP)). Considering different needs, DNNs can also be constructed in various ways. For example, recurrent neural networks (RNNs) and convolutional deep neural networks (CNNs) can be used. It's important to note that a loss function is a real-valued function that measures the degree of inconsistency between the network model's output value f(x) and the target value Y. It can be represented by L(Y, f(x)). The optimization objective is usually to minimize the loss function; the smaller the loss value, the better the model performance. Of course, if the loss function is defined as -L(Y, f(x)), the problem can be transformed into maximizing the loss value.

[0192] 2. Multiple-input multiple-output (MIMO)

[0193] MIMO technology refers to the use of multiple transmit and receive antennas at both the transmitting and receiving ends, enabling signals to be transmitted and received through these antennas, thereby improving communication quality. It makes full use of spatial resources, achieving multiple transmissions and receptions through multiple antennas, and can significantly increase system channel capacity without increasing spectrum resources or antenna transmit power.

[0194] Specifically, the transmitting device performs bit mapping on the multi-layer information bits to be sent to the receiving device, mapping the modulated signal onto one or more transmission layers to generate one or more data streams (also known as layer mapping). The transmitting device then maps this one or more data streams onto antenna ports, performing resource mapping on each antenna port to generate orthogonal frequency division multiplexing (OFDM) symbols and transmit them. To support the simultaneous transmission of multiple data streams, corresponding antenna ports are provided for each layer; that is, each layer corresponds to at least one antenna port.

[0195] 3. Antenna Port

[0196] An antenna port is a logical concept, which can be understood as a logical port used for transmission. There is no direct correspondence between an antenna port and a physical antenna. Antenna ports are typically associated with a reference signal, and their meaning can be understood as a transmit / receive interface on the channel through which the reference signal passes. Therefore, in some cases, an antenna port can also be a reference signal port or a pilot port. For low-frequency systems, an antenna port may correspond to one or more antenna elements that jointly transmit the reference signal; the receiver can treat them as a whole without distinguishing between individual elements. For high-frequency systems, an antenna port may correspond to a beam; similarly, the receiver only needs to treat this beam as an interface without distinguishing between individual elements.

[0197] 4. Reference signal

[0198] Currently at the physical layer, uplink communication includes the transmission of uplink physical channels and uplink signals. Uplink physical channels include the random access channel (PRACH), physical uplink control channel (PUCCH), and physical uplink shared channel (PUSCH), etc. Uplink signals include the sounding reference signal (SRS), the PUCCH de-modulation reference signal (PUCCH-DMRS), the PUSCH-DMRS, the phase noise tracking reference signal (PTRS), the uplink positioning signal, etc. Downlink communication includes the transmission of downlink physical channels and downlink signals. The downlink physical channels include the physical broadcast channel (PBCH), the physical downlink control channel (PDCCH), and the physical downlink shared channel (PDSCH); the downlink signals include the primary synchronization signal (PSS) / secondary synchronization signal (SSS), the downlink control channel demodulation reference signal (PDCCH-DMRS), the downlink data channel demodulation reference signal (PDSCH-DMRS), the phase noise tracking signal (PTRS), the channel status information reference signal (CSI-RS / CSIRS), the cell reference signal (CRS) (not present in NR), the time / frequency tracking reference signal (TRS) (not present in LTE), and the LTE / NR positioning signal (positioning RS), etc.

[0199] Reference signals, also known as pilot signals, are known signals provided by the transmitter to the receiver for channel estimation, channel measurement, channel sounding, or channel demodulation, ensuring the stability and reliability of communication. Uplink reference signals may include, for example, channel sounding signals (SRS), PUCCH-DMRS, PUSCH-DMRS, PTRS, uplink positioning signals, etc.; downlink reference signals may include, for example, synchronization signal blocks (SSB), physical downlink control channel-demodulation reference signals (PDCCH-DMRS), PDSCH-DMRS, PTRS, CSI-RS, CRS, time / frequency domain tracking reference signals (TRS) in NR, and downlink positioning signals (positioning RS), etc.

[0200] 5. Reference signal pattern

[0201] A reference signal pattern, also known as a pilot pattern, refers to the arrangement of a reference signal sequence in the time, frequency, and spatial domains (i.e., any combination of time, frequency, and antenna ports). Essentially, the reference signal pattern indicates how or in which communication resources the reference signal should be inserted. For example, the index corresponding to the i-th reference signal is... k is the number of reference signals; the index set I is represented as I = {I} 1 ,I 2 ,...,I k The design of the reference signal pattern is to address different types of fading channels and ensure the accuracy of channel estimation. Different reference signal arrangements should be used for different types of fading channels. The transmitting device can transmit the reference signal on the communication resources indicated in the reference signal pattern, and correspondingly, the receiving device will also receive the reference signal on the communication resources indicated in the reference signal pattern. This allows the receiving device to use the reference signal for channel estimation (such as the minimum mean squared error (MMSE) algorithm) to subsequently reconstruct complete channel information. The MMSE algorithm optimizes parameter estimation by minimizing the mean squared error between the estimated and actual values.

[0202] Taking pilot patterns as an example, pilot patterns include block pilots, comb pilots, and lattice pilots. Among them, such as... Figure 4AThe diagram shows a block-type pilot arrangement. In this type, OFDM symbols (referred to here as pilot symbols) are periodically transmitted for channel estimation. All subcarriers on each pilot symbol are used as pilots, and time-domain interpolation is performed using these pilots to estimate the channel along the time axis. For example... Figure 4B The diagram shows a comb-type pilot arrangement. In this type, pilot signals are periodically placed on the subcarriers of each OFDM symbol, and then frequency domain interpolation is performed using these pilot signals to estimate the channel along the frequency axis. For example... Figure 4C The diagram shows a lattice-type pilot arrangement. In this type, pilots are inserted along both the time and frequency axes at a given period. The pilots are distributed along the time and frequency axes, making channel estimation more convenient for interpolation in the time / frequency domain.

[0203] With technological advancements, deep neural networks can also be applied to the physical layer of wireless communication (such as MIMO). For example, neural network models can be used to design reference signal patterns, and the output reference signal pattern can effectively select communication resources that contribute to the reconstruction of complete channel information. Figure 4D As shown, Figure 4D This diagram illustrates a reference signal pattern design based on a neural network model. Sc represents frequency domain resources, Symbol represents time domain resources, and Port represents spatial domain resources. The dots in the reference signal pattern represent the locations of the communication resources occupied by the reference signal, i.e., the locations of the time domain, frequency domain, and spatial domain resources. Transmitting the reference signal on these communication resources is more beneficial for the receiver to subsequently reconstruct the complete channel information. Furthermore, by way of example, a neural network model can be used for channel reconstruction. The complete channel information can be reconstructed using the channel information on the communication resources of the reference signal, which is more conducive to ensuring the accuracy of the reconstructed channel information.

[0204] However, current reference signal pattern design or channel reconstruction based on neural networks lacks compatibility with air interfaces, and there is a lack of corresponding technical details and specifications supporting the interface and process design. The so-called air interface, also known as the wireless interface, refers to the interface between the terminal device and the access network device, abbreviated as Uu interface. Therefore, how to use pilot pattern design or channel reconstruction based on neural networks between the terminal device and the access network device to ensure the accuracy of the reconstructed channel information is a problem that urgently needs to be solved.

[0205] Therefore, in order to efficiently utilize reference signal pattern design or channel reconstruction based on neural networks between terminal devices and access network devices, and to ensure the accuracy of reconstructed channel information, this application provides an information acquisition method and a communication device. The information acquisition method and communication device provided in the embodiments of this application will be further described in detail below.

[0206] I. Downlink Transmission (Access network equipment is the transmitting end equipment, and terminal equipment is the receiving end equipment)

[0207] Scenario 1: Associating the first neural network model with the first reference signal density.

[0208] Figure 5 This is a flowchart illustrating an information acquisition method provided in an embodiment of this application. For example... Figure 5 As shown, the information acquisition method includes the following steps S501 to S503. Figure 5 The method shown can be implemented by the aforementioned terminal devices and access network devices. Alternatively, Figure 5 The device that performs the method shown can be a chip in a terminal device or a chip in an access network device; however, this application does not limit the implementation of such a method. Figure 5 The method will be explained using terminal equipment and access network equipment as the main implementers.

[0209] It is understood that this application uses terminal equipment and access network equipment as examples to illustrate the execution of the interaction, but this application does not limit the execution subject of the interaction. For example, the method executed by the terminal equipment in this application can also be implemented by the communication / processing module in the terminal equipment or the circuit or chip (such as a modem chip (also known as a baseband chip), or a SoC chip / SIP chip containing a modem core, or a GPU / AI processor / ASIC) in the terminal equipment responsible for communication / processing functions; the method executed by the access network equipment in this application can also be implemented by the module (such as a circuit, chip or chip system, etc.) in the access network equipment, or a logical node, logical module or software that can implement all or part of the functions of the access network equipment.

[0210] S501. The terminal equipment and the access network equipment determine a first neural network model, which is associated with a first reference signal density. The first neural network model is used to determine the first reference signal pattern.

[0211] In this embodiment, the terminal device or access network device can train a corresponding reference signal pattern design model based on different reference signal densities. This reference signal pattern design model can be used to determine the reference signal pattern. Therefore, the trained reference signal pattern design model and the reference signal density are correlated; specifically, it can be considered that the reference signal pattern design model and the reference signal density correspond one-to-one, and the reference signal density can be used to characterize the reference signal pattern design model. That is, the corresponding reference signal pattern design model can be associated with the reference signal density.

[0212] For example, taking a first neural network model (i.e., a design model for a certain reference signal pattern) as an example, the first neural network model is trained based on a first reference signal density (i.e., a certain reference signal density), and the first neural network model is used to determine the first reference signal pattern. Therefore, the first neural network model and the first reference signal density are related, and the first reference signal density can be used to characterize the first neural network model. That is, the terminal device and the access network device can be associated with the corresponding first neural network model through the first reference signal density.

[0213] In one possible implementation, a first reference signal density is associated with a second neural network model, which is used to reconstruct channel information. This can be understood as the terminal device or access network device training corresponding reference signal pattern design models and channel reconstruction models based on different reference signal densities. The channel reconstruction model can then be used to reconstruct channel information. Therefore, the trained reference signal pattern design model and channel reconstruction model are correlated with the reference signal density; specifically, they can be considered to have a one-to-one correspondence. The reference signal density can be used to characterize the reference signal pattern design model and channel reconstruction model; that is, the reference signal density can be linked to the corresponding reference signal pattern design model and channel reconstruction model.

[0214] For example, taking a first neural network model (i.e., a design model for a certain reference signal pattern) and a second neural network model (i.e., a channel reconstruction model) as examples, the first neural network model is trained based on a first reference signal density (i.e., a certain reference signal density) and is used to determine the first reference signal pattern; the second neural network model is also trained based on the first reference signal density (i.e., a certain reference signal density) and is used for the channel reconstruction model. Therefore, the first neural network model and the second neural network model are related to the first reference signal density. The first reference signal density can be used to characterize the first neural network model and the second neural network model, that is, the terminal device and the access network device can be associated with the corresponding first neural network model and the second neural network model through the first reference signal density.

[0215] It should be noted that the reference signal density here (such as the first reference signal density) can be characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports. Specifically, time-domain reference signal density refers to the distribution density of the reference signal in the time domain; frequency-domain reference signal density refers to the distribution density of the reference signal in the frequency domain; and spatial-domain reference signal density refers to the distribution density of the reference signal in the spatial domain.

[0216] Optionally, the spatial reference signal density may include the number of selected antenna ports and / or the total number of antenna ports. Optionally, the spatial reference signal density may be the percentage of the number of selected antenna ports in the total number of antenna ports.

[0217] To more easily indicate the various reference signal pattern design models and channel reconstruction models, each reference signal pattern design model and channel reconstruction model can be identified. As shown in Table 1, Table 1 is a correlation table between a neural network model and a reference signal density provided in an embodiment of this application. It is assumed that the reference signal density includes the time-domain reference signal density and the frequency-domain reference signal density (k), the number of selected antenna ports (p), and the total number of antenna ports (P). When the neural network model is identified as 1, it indicates that the reference signal pattern design model and channel reconstruction model identified as 1 have the following corresponding reference signal densities: both the time-domain and frequency-domain reference signal densities (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128. When the neural network model is identified as 2, it indicates that the reference signal pattern design model and channel reconstruction model identified as 2 have the following corresponding reference signal densities: both the time-domain and frequency-domain reference signal densities (k) are 8, the number of selected antenna ports (p) is 32, and the total number of antenna ports (P) is 128. When the neural network model is labeled as 3, it indicates that the reference signal pattern design model and channel reconstruction model are labeled as 3. The corresponding reference signal densities include: time-domain reference signal density and frequency-domain reference signal density (k) are both 12, the number of selected antenna ports (p) is 32, and the total number of antenna ports (P) is 128.

[0218] Table 1

[0219]

[0220] As shown in Table 2, Table 2 is another correlation table between a neural network model and a reference signal density provided in the embodiments of this application. It is assumed that the reference signal density includes the time-domain reference signal density and the frequency-domain reference signal density (k), and the percentage of the number of selected antenna ports in the total number of antenna ports (p / P). When the neural network model is identified as 1, it represents the reference signal pattern design model and channel reconstruction model identified as 1. The corresponding reference signal density includes: both the time-domain and frequency-domain reference signal densities (k) are 4, and the percentage of the number of selected antenna ports in the total number of antenna ports (p / P) is 1 / 16. When the neural network model is identified as 2, it represents the reference signal pattern design model and channel reconstruction model identified as 2. The corresponding reference signal density includes: both the time-domain and frequency-domain reference signal densities (k) are 8, and the percentage of the number of selected antenna ports in the total number of antenna ports (p / P) is 1 / 8. When the neural network model is labeled as 3, it indicates that the reference signal pattern design model and channel reconstruction model are labeled as 3. The corresponding reference signal densities include: the time domain reference signal density and the frequency domain reference signal density (k) are both 12, and the proportion (p / P) of the number of selected antenna ports in the total number of antenna ports is 1 / 16.

[0221] Table 2

[0222]

[0223] For example, taking Table 1 as an example, assuming that the first reference density includes: the time-domain reference signal density and the frequency-domain reference signal density (k) are both 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128, then the reference signal pattern design model and the channel reconstruction model corresponding to the first reference density are identified as 1. The reference signal pattern design model identified as 1 can be called the first neural network model, and the channel reconstruction model identified as 1 can be called the second neural network model. In this way, the first reference signal density is associated with the corresponding first neural network model and the second neural network model.

[0224] Additionally, it should be noted that the relationship between the neural network model and the reference signal density can be predefined by the protocol standard, or it can be established by the training end and then communicated to the other end. This is not limited here. For example, if the terminal device trains corresponding reference signal pattern design models and channel reconstruction models based on different reference signal densities (i.e., the terminal device is the model training end), thereby establishing the relationship between the neural network model and the reference signal density (as shown in Tables 1 and 2), then the terminal device also needs to inform the access network device of this relationship in advance. This ensures that the terminal device and the access network device align this relationship, facilitating efficient information exchange between them using low-overhead signaling. Similarly, if the access network device trains the corresponding reference signal pattern design model and channel reconstruction model based on different reference signal densities (that is, the access network device is the model training end), thereby constructing the correlation between the neural network model and the reference signal density (as shown in Tables 1 and 2), then the access network device also needs to inform the terminal device of this correlation in advance, so that the terminal device and the access network device can align the correlation, thereby facilitating the subsequent information exchange between the terminal device and the access network device using low-overhead signaling.

[0225] Based on the above, taking the first reference signal density as an example, the terminal device and the access network device can associate the first neural network model and the second neural network model through the first reference signal density. Therefore, before determining the first neural network model and the second neural network model, the key lies in how the terminal device and the access network device obtain the first reference signal density. The following details the different scenarios in which the terminal device and the access network device obtain the first reference signal density:

[0226] Case 1: The first reference signal density is selected by the terminal device from the reference signal densities supported by the access network device.

[0227] Method 1: For example Figure 6A As shown, it includes the following steps s11 and s12.

[0228] s11. The access network device broadcasts first information, which indicates the identifier of the supported reference signal density or neural network model. Accordingly, the terminal device receives the first information from the access network device.

[0229] s12. The terminal device sends a first reference signal density or a first identifier to the access network device, the first identifier indicating the first reference signal density. Accordingly, the access network device receives the first reference signal density or the first identifier from the terminal device.

[0230] Here, the first reference signal density is one of the reference signal densities supported by the access network device, and the first identifier is one of the identifiers of the neural network model supported by the access network device. This indicates that the terminal device has selected the first reference signal density from the reference signal densities supported by the access network device, or that the terminal device has selected the first identifier from the identifiers of the neural network model supported by the access network device. This approach improves the flexibility of reference signal density selection.

[0231] Method 2: For example Figure 6A As shown, it includes the following steps s21 to s23.

[0232] s21. The terminal device sends a first request to the access network device, the first request being for an identifier of a supported reference signal density or neural network model. Accordingly, the terminal device receives the first request from the access network device.

[0233] s22. The access network device sends first information to the terminal device, the first information being used to indicate the identifier of the supported reference signal density or neural network model. Accordingly, the terminal device receives the first information from the access network device.

[0234] s23. The terminal device sends a first reference signal density or a first identifier to the access network device, the first identifier indicating the first reference signal density. Accordingly, the access network device receives the first reference signal density or the first identifier from the terminal device.

[0235] Here, the first reference signal density is one of the reference signal densities supported by the access network device, and the first identifier is one of the identifiers of the neural network model supported by the access network device. This indicates that the terminal device has selected the first reference signal density from the reference signal densities supported by the access network device, or that the terminal device has selected the first identifier from the identifiers of the neural network model supported by the access network device. This approach improves the flexibility of reference signal density selection.

[0236] Of course, if the current environmental characteristics change, in one possible implementation, the terminal device can also reselect the reference signal density based on its own decoding and channel conditions (such as signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR), reference signal received power (RSRP), etc.). That is, the terminal device sends a third indication information to the access network device, which indicates the second reference signal density. Correspondingly, the access network device receives the third indication information from the terminal device. This approach improves the flexibility of reference signal density selection.

[0237] In another possible implementation, the access network device may not adopt the first reference signal density selected by the terminal device. Instead, it can configure a reference signal density for the terminal device based on its own hardware support, environmental characteristics, and feedback from the terminal device. Specifically, the access network device sends a fourth indication message to the terminal device, which indicates the third reference signal density. Correspondingly, the terminal device receives the fourth indication message from the access network device. This approach improves the flexibility of reference signal density selection.

[0238] Case 2: The first reference signal density is selected by the access network equipment from the reference signal densities supported by the terminal equipment.

[0239] In one possible implementation, such as Figure 6B As shown, the steps s31 and s32 are included:

[0240] s31. The terminal device sends capability information to the access network device. This capability information indicates support for a reference signal pattern design assisted by a neural network model, or an identifier of the supported reference signal density or neural network model. Accordingly, the access network device receives the capability information from the terminal device.

[0241] s32. The access network device sends a first reference signal density or a first identifier to the terminal device, the first identifier indicating the first reference signal density. Accordingly, the terminal device receives the first reference signal density or the first identifier from the access network device.

[0242] Wherein, the first reference signal density is one of the reference signal densities supported by the terminal device, and the first identifier is one of the identifiers of the neural network model supported by the terminal device. This indicates that the access network device selected the first reference signal density from the reference signal densities supported by the terminal device, or that the access network device selected the first identifier from the identifiers of the neural network model supported by the terminal device. This approach improves the flexibility of reference signal density selection.

[0243] Of course, if the current environmental characteristics change, in one possible implementation, the access network device can also reselect the reference signal density based on its own decoding and channel conditions (such as signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR), and reference signal received power (RSRP). That is, the access network device sends a fifth indication message to the terminal device, which indicates the fourth reference signal density. Correspondingly, the terminal device receives the fifth indication message from the access network device. This approach improves the flexibility of reference signal density selection.

[0244] In another possible implementation, the terminal device may not adopt the first reference signal density selected by the access network device. Instead, it can reconfigure the reference signal density for the access network device based on its own hardware support, environmental characteristics, and feedback from the access network device. Specifically, the terminal device sends a sixth indication message to the access network device, which indicates the fifth reference signal density. Correspondingly, the access network device receives the sixth indication message from the terminal device. This approach improves the flexibility of reference signal density selection.

[0245] S502, The terminal equipment and the access network equipment acquire the communication resources of the first reference signal in the first reference signal pattern.

[0246] In this embodiment, the access network device can be a model training terminal or a non-model training terminal; correspondingly, the terminal device can be a non-model training terminal or a model training terminal. The following provides specific explanations for different scenarios:

[0247] Scenario A: The access network device is the model training end, and the terminal device is not the model training end.

[0248] In one possible implementation, the specific implementation of the terminal device and the access network device acquiring the communication resources of the first reference signal in the first reference signal pattern may include the following steps s41 and s42, such as... Figure 6C As shown.

[0249] s41. The access network device uses a first neural network model to determine a first reference signal pattern, which indicates the communication resources of the first reference signal.

[0250] s42. The access network device sends first indication information to the terminal device, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern. Accordingly, the terminal device receives the first indication information from the access network device.

[0251] Since the access network device serves as the model training end, and it stores the second neural network model, after determining the second neural network model, in one possible implementation, the method further includes step s43: the access network device sends the second neural network model to the terminal device. Correspondingly, the terminal device receives the second neural network model from the access network device so that it can subsequently utilize this second neural network model to achieve channel reconstruction.

[0252] Of course, the access network device can also update the first and second neural network models to improve their flexibility and accuracy. It will also notify the terminal device of these updates to ensure information alignment between the access network device and the terminal device. Since updating the first and second neural network models will change the corresponding first reference signal density, to ensure consistency between the neural network models stored in the access network device and the terminal device and the first reference signal density, the access network device will further send the change value corresponding to the first reference signal density to the terminal device, allowing the terminal device to adjust the first reference signal density in a timely manner.

[0253] Specifically, in one possible implementation, the access network device updates the first neural network model and the second neural network model; then, based on the updated first neural network model and the updated second neural network model, it determines the change value corresponding to the first reference signal density; further, the access network device sends second indication information to the terminal device, which indicates that the first neural network model and the second neural network model have been updated. Accordingly, the terminal device receives the second indication information from the access network device. Optionally, the second indication information includes the change value corresponding to the first reference signal density.

[0254] For example, taking Table 1 as an example, assume that the first reference density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128; the identifier associated with the first reference signal density to the corresponding neural network model is 1, i.e., the first neural network model and the second neural network model. After the access network device updates the first neural network model and the second neural network model, based on the updated first neural network model and the updated second neural network model, it determines that the change value corresponding to the total number of antenna ports is ΔP = +32. This means that the total number of antenna ports has increased by 32. Then the adjusted first reference signal density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 160. Furthermore, the access network device will also send the change value corresponding to the total number of antenna ports (i.e., ΔP = +32) to the terminal device through the second indication information, so that the terminal device can also adjust the first reference signal density in a timely manner.

[0255] For example, taking Table 1 as an example, assume that the first reference density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128; the identifier associated with the first reference signal density to the corresponding neural network model is 1, namely the first neural network model and the second neural network model. After the access network device updates the first neural network model and the second neural network model, based on the updated first neural network model and the updated second neural network model, it determines that the change value corresponding to the time-domain reference signal density and the frequency-domain reference signal density is Δk = +1. This means that both the time-domain reference signal density and the frequency-domain reference signal density have increased by 1. Then the adjusted first reference signal density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 5, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 160. Furthermore, the access network device will also send the change values ​​(i.e., Δk = +1) corresponding to the time-domain reference signal density and the frequency-domain reference signal density to the terminal device through the second indication information, so that the terminal device can also adjust the first reference signal density in a timely manner.

[0256] Optionally, the specific implementation methods for the access network device to update the first neural network model and the second neural network model include, but are not limited to, the following:

[0257] Method 1: Update based on dataset.

[0258] For example, the access network device acquires a first time-domain unit. After the first time-domain unit, a first dataset used for training the first neural network model and the second neural network model is updated. Then, the first neural network model and the second neural network model are updated based on the first dataset. This can be understood as follows: after the first time-domain unit, the first dataset used for training the first neural network model and the second neural network model is updated. The access network device needs to use this first dataset to retrain the first neural network model and the second neural network model to update them, which helps improve the flexibility and accuracy of the neural network model.

[0259] Method 2: Update based on model parameters.

[0260] For example, the model parameters of the first neural network model and the model parameters of the second neural network model may include parameters such as time-domain reference signal density, frequency-domain reference signal density, number of selected antenna ports, and total number of antenna ports. The access network device can update the first neural network model and the second neural network model by adjusting the model parameters of the first neural network model and the second neural network model, which is beneficial to improving the flexibility and accuracy of the neural network model.

[0261] In addition, access network devices will also proactively update the first and second neural network models, including but not limited to the following situations:

[0262] Case a: When the first neural network model is switched to the third neural network model, and the second neural network model is switched to the fourth neural network model, if the first condition is met, the access network device needs to update the first neural network model and the second neural network model.

[0263] The first condition can be that the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model. This can be understood as follows: initially, the optimal neural network model determined based on environmental characteristics and other information is the first neural network model and the second neural network model. That is, the channel information reconstructed using the first and second neural network models should have the highest accuracy. However, after switching from the first neural network model to the third neural network model, and from the second neural network model to the fourth neural network model, the reconstructed channel information actually has higher accuracy. This means that the first and second neural network models need to be updated in a timely manner to ensure the accuracy of their subsequent use.

[0264] The first condition could also be that the channel quality corresponding to the first channel determined by the fourth neural network model is greater than the channel quality corresponding to the first channel determined by the first neural network model. Of course, the first condition could also be other conditions, which are not limited here.

[0265] Case b: If the usage time of the first neural network model and the second neural network model reaches a preset threshold, the access network device needs to update the first neural network model and the second neural network model.

[0266] This can be understood as follows: when the usage time of the first neural network model and the second neural network model reaches a preset threshold, it means that the accuracy of the first neural network model and the second neural network model will decrease. At this time, it is necessary to update the first neural network model and the second neural network model in a timely manner to ensure the accuracy of model processing.

[0267] Scenario B: The access network device is not the model training end, and the terminal device is the model training end.

[0268] In one possible implementation, the specific implementation of the terminal device and the access network device acquiring the communication resources of the first reference signal in the first reference signal pattern may include the following steps s51 and s52, such as... Figure 6D As shown.

[0269] s51. The terminal device uses a first neural network model to determine a first reference signal pattern, which indicates the communication resources of the first reference signal.

[0270] s52. The terminal device sends first indication information to the access network device, the first indication information being used to indicate the communication resources of the first reference signal in the first reference signal pattern. Correspondingly, the access network device receives the first indication information from the terminal device.

[0271] Furthermore, the terminal device can update the first and second neural network models and notify the access network device of the update to ensure information alignment between the access network device and the terminal device. Since the first reference signal density also changes after updating the first and second neural network models, to ensure consistency between the neural network models stored in the access network device and the terminal device and the first reference signal density, the terminal device will further send the change value corresponding to the first reference signal density to the access network device, enabling the access network device to adjust the first reference signal density in a timely manner.

[0272] Specifically, in one possible implementation, the terminal device updates the first neural network model and the second neural network model; then, based on the updated first neural network model and the updated second neural network model, it determines the change value corresponding to the first reference signal density; further, the terminal device sends second indication information to the access network device, which indicates that the first neural network model and the second neural network model have been updated. Accordingly, the access network device receives the second indication information from the terminal device. Optionally, the second indication information includes the change value corresponding to the first reference signal density.

[0273] For example, taking Table 1 as an example, assume that the first reference density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128; the identifier associated with the first reference signal density to the corresponding neural network model is 1, i.e., the first neural network model and the second neural network model. After the terminal device updates the first neural network model and the second neural network model, based on the updated first neural network model and the updated second neural network model, it determines that the change value corresponding to the total number of antenna ports is ΔP = +32. This means that the total number of antenna ports has increased by 32. Then the adjusted first reference signal density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 160. Furthermore, the terminal device will also send the change value corresponding to the total number of antenna ports (i.e., ΔP = +32) to the access network device through the second indication information, so that the access network device can also adjust the first reference signal density in a timely manner.

[0274] For example, taking Table 1 as an example, assume that the first reference density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 4, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 128; the identifier associated with the first reference signal density to the corresponding neural network model is 1, namely the first neural network model and the second neural network model. After the terminal device updates the first neural network model and the second neural network model, based on the updated first neural network model and the updated second neural network model, it determines that the change value of the time-domain reference signal density and the frequency-domain reference signal density is Δk = +1. This means that both the time-domain reference signal density and the frequency-domain reference signal density have increased by 1. Then the adjusted first reference signal density includes: both the time-domain reference signal density and the frequency-domain reference signal density (k) are 5, the number of selected antenna ports (p) is 16, and the total number of antenna ports (P) is 160. Furthermore, the terminal device will also send the change values ​​(i.e., Δk = +1) corresponding to the time-domain reference signal density and the frequency-domain reference signal density to the access network device through the second instruction information, so that the access network device can also adjust the first reference signal density in a timely manner.

[0275] Optionally, the specific implementation methods for updating the first neural network model and the second neural network model by the terminal device include, but are not limited to, the following:

[0276] Method (1): Update based on dataset.

[0277] For example, the terminal device acquires a first time-domain unit, after which a first dataset used for training the first neural network model and the second neural network model is updated; then, the first neural network model and the second neural network model are updated based on the first dataset. This can be understood as follows: after the first time-domain unit, the first dataset used for training the first neural network model and the second neural network model is updated, and the terminal device needs to retrain the first neural network model and the second neural network model using the first dataset to update the first neural network model and the second neural network model, which helps to improve the flexibility and accuracy of the neural network model.

[0278] Method (2): Update based on model parameters.

[0279] For example, the model parameters of the first neural network model and the model parameters of the second neural network model may include parameters such as time-domain reference signal density, frequency-domain reference signal density, number of selected antenna ports, and total number of antenna ports. The access network device can update the first neural network model and the second neural network model by adjusting the model parameters of the first neural network model and the second neural network model, which is beneficial to improving the flexibility and accuracy of the neural network model.

[0280] In addition, the terminal device will also proactively update the first neural network model and the second neural network model, specifically including but not limited to the following situations:

[0281] Case (a): If the first neural network model is switched to the third neural network model and the second neural network model is switched to the fourth neural network model, and the first condition is met, the terminal device needs to update the first neural network model and the second neural network model.

[0282] The first condition can be that the accuracy of the third channel information corresponding to the first channel determined based on the fourth neural network model is greater than the accuracy of the first channel information corresponding to the first channel determined based on the first neural network model. This can be understood as follows: initially, the optimal neural network model determined based on environmental characteristics and other information is the first neural network model and the second neural network model. The channel information reconstructed using the first and second neural network models should have the highest accuracy. However, after switching from the first neural network model to the third neural network model, and from the second neural network model to the fourth neural network model, the reconstructed channel information actually has higher accuracy. This means that the first and second neural network models need to be updated in a timely manner to ensure the accuracy of their subsequent use.

[0283] The first condition could also be that the channel quality corresponding to the first channel determined by the fourth neural network model is greater than the channel quality corresponding to the first channel determined by the first neural network model. Of course, the first condition could also be other conditions, which are not limited here.

[0284] Case (b): If the usage time of the first neural network model and the second neural network model reaches a preset threshold, the terminal device needs to update the first neural network model and the second neural network model.

[0285] This can be understood as follows: when the usage time of the first neural network model and the second neural network model reaches a preset threshold, it means that the accuracy of the first neural network model and the second neural network model will decrease. At this time, it is necessary to update the first neural network model and the second neural network model in a timely manner to ensure the accuracy of model processing.

[0286] S503. The access network device transmits the first reference signal to the terminal device on the first channel based on the communication resources of the first reference signal. Correspondingly, the terminal device receives the first reference signal from the access network device on the first channel based on the communication resources of the first reference signal.

[0287] In this embodiment of the application, after the terminal device and the access network device obtain the communication resources of the first reference signal in the first reference signal pattern, the access network device can send the first reference signal to the terminal device on the first channel based on the communication resources of the first reference signal. Correspondingly, the terminal device can receive the first reference signal from the access network device on the same communication resources.

[0288] In one possible implementation, after the terminal device receives the first reference signal, the method further includes: the terminal device determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal. That is, the terminal device can reconstruct the first channel information using the second neural network model and the first reference signal.

[0289] Optionally, the specific implementation of the terminal device determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal may include the following steps s61 and s62, such as... Figure 6E As shown.

[0290] s61. The terminal device determines the second channel information corresponding to the communication resources of the first reference signal based on the first reference signal.

[0291] s62. The terminal device uses the second neural network model to process the second channel information to obtain the first channel information corresponding to the first channel.

[0292] In a specific implementation, after receiving the first reference signal p′, the terminal device can perform channel estimation based on the first reference signal to obtain the second channel information corresponding to the communication resources of the first reference signal on the first channel. (i.e., channel information of the first reference signal resource location), Among them, the channel estimation method f(·) includes, but is not limited to, the minimum mean square error (MMSE) algorithm. Then, the second channel information... The information is input into the second neural network model for processing to obtain the complete channel information of the first channel, i.e., the first channel information, thereby realizing the reconstruction of the complete channel information.

[0293] It should be noted that during the process of the access network device transmitting the first reference signal to the terminal device based on the communication resources of the first reference signal on the first channel, data transmission may also be included. Of course, the access network device may also transmit data to the terminal device after the terminal device has reconstructed the complete channel information; this is not limited here.

[0294] It can be seen that, based on Figure 5The described method allows terminal devices and access network devices to use reference signal density to characterize reference signal pattern design models. That is, the reference signal density can be associated with the corresponding reference signal pattern design model (for example, the first reference signal density can be associated with the corresponding first neural network model). This enables terminal devices and access network devices to use reference signal pattern designs based on neural networks more efficiently, thereby ensuring the accuracy of reconstructed channel information. At the same time, the selection of neural network models can be completed by obtaining only the reference signal density, which helps to reduce signaling overhead and improve communication efficiency.

[0295] Scenario 2: Correlate the second neural network model with the first reference signal density.

[0296] Figure 7 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application. For example... Figure 7 As shown, the information acquisition method includes the following steps S701 to S703. Figure 7 The method shown can be implemented by the aforementioned terminal devices and access network devices. Alternatively, Figure 7 The device that performs the method shown can be a chip in a terminal device or a chip in an access network device; however, this application does not limit the implementation of such a method. Figure 7 The method will be explained using terminal equipment and access network equipment as the main implementers.

[0297] It is understood that this application uses terminal equipment and access network equipment as examples to illustrate the execution of the interaction, but this application does not limit the execution subject of the interaction. For example, the method executed by the terminal equipment in this application can also be implemented by the communication / processing module in the terminal equipment or the circuit or chip (such as a modem chip (also known as a baseband chip), or a SoC chip / SIP chip containing a modem core, or a GPU / AI processor / ASIC) in the terminal equipment responsible for communication / processing functions; the method executed by the access network equipment in this application can also be implemented by the module (such as a circuit, chip or chip system, etc.) in the access network equipment, or a logical node, logical module or software that can implement all or part of the functions of the access network equipment.

[0298] S701, the terminal equipment and the access network equipment determine the second neural network model, which is associated with the first reference signal density, and the second neural network model is used to reconstruct channel information.

[0299] In this embodiment, the terminal device or access network device can train a corresponding channel reconstruction model based on different reference signal densities. This channel reconstruction model can be used to reconstruct channel information. Therefore, the trained channel reconstruction model and the reference signal density are correlated. Specifically, it can be considered that the channel reconstruction model and the reference signal density are in one-to-one correspondence, and the reference signal density can be used to characterize the channel reconstruction model. That is, the corresponding channel reconstruction model can be associated with the reference signal density.

[0300] For example, taking a second neural network model (i.e., a certain channel reconstruction model) as an example, the second neural network model is trained based on the first reference signal density (i.e., a certain reference signal density), and the second neural network model is used to reconstruct channel information. Therefore, the second neural network model and the first reference signal density are related, and the first reference signal density can be used to characterize the second neural network model. That is, the terminal device and the access network device can be associated with the corresponding second neural network model through the first reference signal density.

[0301] In one possible implementation, a first reference signal density is associated with a first neural network model, which is used to determine the first reference signal pattern. This can be understood as the terminal device or access network device training corresponding reference signal pattern design models and channel reconstruction models based on different reference signal densities. The reference signal pattern design model can be used to determine the reference signal pattern. Therefore, the trained reference signal pattern design model and channel reconstruction model are correlated with the reference signal density. Specifically, it can be considered that the reference signal pattern design model and channel reconstruction model correspond one-to-one with the reference signal density, and the reference signal density can be used to characterize the reference signal pattern design model and channel reconstruction model; that is, the reference signal density can be associated with the corresponding reference signal pattern design model and channel reconstruction model.

[0302] For example, taking a first neural network model (i.e., a design model for a certain reference signal pattern) and a second neural network model (i.e., a channel reconstruction model) as examples, the first neural network model is trained based on a first reference signal density (i.e., a certain reference signal density) and is used to determine the first reference signal pattern; the second neural network model is also trained based on the first reference signal density (i.e., a certain reference signal density) and is used for the channel reconstruction model. Therefore, the first neural network model and the second neural network model are related to the first reference signal density. The first reference signal density can be used to characterize the first neural network model and the second neural network model, that is, the terminal device and the access network device can be associated with the corresponding first neural network model and the second neural network model through the first reference signal density.

[0303] It should be noted that the reference signal density (such as the first reference signal density) here can be characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports. Specifically, time-domain reference signal density refers to the distribution density or intensity of the reference signal in the time domain; frequency-domain reference signal density refers to the distribution density or intensity of the reference signal in the frequency domain; and spatial-domain reference signal density refers to the distribution density or intensity of the reference signal in the spatial domain.

[0304] Optionally, the spatial reference signal density may include the number of selected antenna ports and / or the total number of antenna ports. Optionally, the spatial reference signal density may be the percentage of the number of selected antenna ports in the total number of antenna ports.

[0305] To more easily indicate the various reference signal pattern design models and channel reconstruction models, each model can be labeled. See Tables 1 and 2 above for details, which will not be elaborated upon here.

[0306] Based on the above, taking the first reference signal density as an example, the terminal device and the access network device can associate the first neural network model and the second neural network model through the first reference signal density. Therefore, the key lies in how the terminal device and the access network device obtain the first reference signal density. For different scenarios of how the terminal device and the access network device obtain the first reference signal density, please refer to the descriptions of scenarios 1 and 2 in step S501 above, which will not be repeated here.

[0307] S702. The access network device sends a first reference signal to the terminal device on the first channel. Correspondingly, the terminal device receives the first reference signal from the access network device on the first channel.

[0308] In one possible implementation, the method further includes: the terminal device and the access network device acquiring the communication resources of the first reference signal in the first reference signal pattern.

[0309] Optionally, the specific implementation of the access network device sending the first reference signal to the terminal device on the first channel can be: sending the first reference signal on the first channel based on the communication resources of the first reference signal. Correspondingly, the specific implementation of the terminal device receiving the first reference signal from the access network device on the first channel can be: receiving the first reference signal on the first channel based on the communication resources of the first reference signal.

[0310] Since the access network device can be either a model training end or a non-model training end, the corresponding terminal device can be either a non-model training end or a model training end. Therefore, for different situations, the methods by which the terminal device and the access network device obtain the communication resources of the first reference signal in the first reference signal pattern can refer to the description of step S502 above, and will not be repeated here.

[0311] S703, The terminal device determines the first channel information corresponding to the first channel based on the second neural network model and the first reference signal.

[0312] In this embodiment of the application, after receiving the first reference signal, the terminal device can use the second neural network model and the first reference signal to reconstruct the first channel information.

[0313] In one possible implementation, the specific implementation of the terminal device determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal may include the description of steps s61 and s62 above, which will not be repeated here.

[0314] It can be seen that, based on Figure 7 The described method allows terminal devices and access network devices to use reference signal density to characterize the channel reconstruction model. That is, the reference signal density can be associated with the corresponding channel reconstruction model (for example, the first reference signal density can be associated with the corresponding second neural network model). This enables terminal devices and access network devices to use neural network-based channel reconstruction more efficiently, thereby ensuring the accuracy of the reconstructed channel information. At the same time, the selection of the neural network model can be completed by obtaining only the reference signal density, which helps to reduce signaling overhead and improve communication efficiency.

[0315] II. Uplink Transmission (The terminal device is the transmitting device, and the access network device is the receiving device)

[0316] Scenario 1: Associating the first neural network model with the first reference signal density.

[0317] Figure 8 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application. For example... Figure 8 As shown, the information acquisition method includes the following steps S801 to S803. Figure 8 The method shown can be implemented by the aforementioned terminal devices and access network devices. Alternatively, Figure 8 The device that performs the method shown can be a chip in a terminal device or a chip in an access network device; however, this application does not limit the implementation of such a method. Figure 8 The method will be explained using terminal equipment and access network equipment as the main implementers.

[0318] It is understood that this application uses terminal equipment and access network equipment as examples to illustrate the execution of the interaction, but this application does not limit the execution subject of the interaction. For example, the method executed by the terminal equipment in this application can also be implemented by the communication / processing module in the terminal equipment or the circuit or chip (such as a modem chip (also known as a baseband chip), or a SoC chip / SIP chip containing a modem core, or a GPU / AI processor / ASIC) in the terminal equipment responsible for communication / processing functions; the method executed by the access network equipment in this application can also be implemented by the module (such as a circuit, chip or chip system, etc.) in the access network equipment, or a logical node, logical module or software that can implement all or part of the functions of the access network equipment.

[0319] S801, the terminal equipment and the access network equipment determine a first neural network model, which is associated with a first reference signal density, and the first neural network model is used to determine the first reference signal pattern.

[0320] S802, The terminal equipment and the access network equipment acquire the communication resources of the first reference signal in the first reference signal pattern.

[0321] S803. The terminal device sends a first reference signal to the access network device on the first channel based on the communication resources of the first reference signal. Correspondingly, the access network device receives the first reference signal from the terminal device on the first channel based on the communication resources of the first reference signal.

[0322] In the embodiments of this application, the main difference between steps S801 to S803 and steps S501 to S503 is the interchange of the execution subject, that is, the "terminal device" and the "access network device" are interchanged. The specific implementation of steps S801 to S803 can be referred to the specific implementation of steps S501 to S503 above, and will not be repeated here.

[0323] It can be seen that, based on Figure 8 The described method allows terminal devices and access network devices to use reference signal density to characterize reference signal pattern design models. That is, the reference signal density can be associated with the corresponding reference signal pattern design model (for example, the first reference signal density can be associated with the corresponding first neural network model). This enables terminal devices and access network devices to use reference signal pattern designs based on neural networks more efficiently, thereby ensuring the accuracy of reconstructed channel information. At the same time, the selection of neural network models can be completed by obtaining only the reference signal density, which helps to reduce signaling overhead and improve communication efficiency.

[0324] Scenario 2: Correlate the second neural network model with the first reference signal density.

[0325] Figure 9 This is a flowchart illustrating another information acquisition method provided in an embodiment of this application. For example... Figure 9 As shown, the information acquisition method includes the following steps S901 to S903. Figure 9 The method shown can be implemented by the aforementioned terminal devices and access network devices. Alternatively, Figure 9 The device that performs the method shown can be a chip in a terminal device or a chip in an access network device; however, this application does not limit the implementation of such a method. Figure 9 The method will be explained using terminal equipment and access network equipment as the main implementers.

[0326] It is understood that this application uses terminal equipment and access network equipment as examples to illustrate the execution of the interaction, but this application does not limit the execution subject of the interaction. For example, the method executed by the terminal equipment in this application can also be implemented by the communication / processing module in the terminal equipment or the circuit or chip (such as a modem chip (also known as a baseband chip), or a SoC chip / SIP chip containing a modem core, or a GPU / AI processor / ASIC) in the terminal equipment responsible for communication / processing functions; the method executed by the access network equipment in this application can also be implemented by the module (such as a circuit, chip or chip system, etc.) in the access network equipment, or a logical node, logical module or software that can implement all or part of the functions of the access network equipment.

[0327] S901, the terminal equipment and the access network equipment determine the second neural network model, which is associated with the first reference signal density, and the second neural network model is used to reconstruct channel information.

[0328] S902, the terminal device sends a first reference signal to the access network device on the first channel. Correspondingly, the access network device receives the first reference signal from the terminal device on the first channel.

[0329] S903, The access network device determines the first channel information corresponding to the first channel based on the second neural network model and the first reference signal.

[0330] In the embodiments of this application, the main difference between steps S901 to S903 and steps S701 to S703 is the interchange of the execution subject, that is, the "terminal device" and the "access network device" are interchanged. The specific implementation of steps S901 to S903 can refer to the specific implementation of steps S701 to S703 above, and will not be repeated here.

[0331] It can be seen that, based on Figure 9The described method allows terminal devices and access network devices to use reference signal density to characterize the channel reconstruction model. That is, the reference signal density can be associated with the corresponding channel reconstruction model (for example, the first reference signal density can be associated with the corresponding second neural network model). This enables terminal devices and access network devices to use neural network-based channel reconstruction more efficiently, thereby ensuring the accuracy of the reconstructed channel information. At the same time, the selection of the neural network model can be completed by obtaining only the reference signal density, which helps to reduce signaling overhead and improve communication efficiency.

[0332] The apparatus provided in the embodiments of this application will be described below.

[0333] This application divides the device into functional modules according to the above method embodiments. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated modules can be implemented in hardware or as software functional modules. It should be noted that the module division in this application is illustrative and only represents one logical functional division; other division methods may be used in actual implementation. The following will combine... Figures 10 to 12 The apparatus of the embodiments of this application is described in detail.

[0334] Figure 10 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application, such as... Figure 10 As shown, the communication device includes a processing module 1001 and a transceiver module 1002. The transceiver module 1002 can implement corresponding communication functions, and the processing module 1001 is used to implement corresponding processing functions. The transceiver module 1002 can also be referred to as an interface, a communication interface, or a communication module, etc.

[0335] In some embodiments of this application, the communication device can be used to perform the actions performed by the terminal device in the above method embodiments. In this case, the communication device can be the terminal device itself or a chip or functional module configurable within the terminal device. The transceiver module 1002 is used to perform transceiver-related operations of the terminal device in the above method embodiments, and the processing module 1001 is used to perform processing-related operations of the terminal device in the above method embodiments.

[0336] For example, the processing module 1001 can be used to determine a first neural network model, which is associated with a first reference signal density, and the first neural network model is used to determine a first reference signal pattern;

[0337] The processing module 1001 can also be used to acquire the communication resources of the first reference signal in the first reference signal pattern;

[0338] The transceiver module 1002 can be used to receive the first reference signal based on the communication resources of the first reference signal on the first channel.

[0339] Reuse Figure 10 In other embodiments of this application, the communication device can be used to perform the actions performed by the access network device in the above method embodiments. In this case, the communication device can be the access network device itself or a chip or functional module configurable within the access network device. The transceiver module 1002 is used to perform transceiver-related operations of the access network device in the above method embodiments, and the processing module 1001 is used to perform processing-related operations of the access network device in the above method embodiments.

[0340] For example, the processing module 1001 can be used to determine a first neural network model, which is associated with a first reference signal density, and the first neural network model is used to determine a first reference signal pattern;

[0341] The processing module 1001 can also be used to acquire the communication resources of the first reference signal in the first reference signal pattern;

[0342] The transceiver module 1002 can be used to transmit the first reference signal based on the communication resources of the first reference signal on the first channel.

[0343] Reuse Figure 10 In other embodiments of this application, the communication device can be used to perform the actions performed by the terminal device in the above method embodiments. In this case, the communication device can be the terminal device itself or a chip or functional module configurable within the terminal device. The transceiver module 1002 is used to perform transceiver-related operations of the terminal device in the above method embodiments, and the processing module 1001 is used to perform processing-related operations of the terminal device in the above method embodiments.

[0344] For example, the processing module 1001 can be used to determine a second neural network model, which is associated with a first reference signal density, and the second neural network model is used to reconstruct channel information;

[0345] The transceiver module 1002 can be used to receive a first reference signal on the first channel;

[0346] The processing module 1001 can also be used to determine the first channel information corresponding to the first channel based on the second neural network model and the first reference signal.

[0347] Reuse Figure 10In other embodiments of this application, the communication device can be used to perform the actions performed by the access network device in the above method embodiments. In this case, the communication device can be the access network device itself or a chip or functional module configurable within the access network device. The transceiver module 1002 is used to perform transceiver-related operations of the access network device in the above method embodiments, and the processing module 1001 is used to perform processing-related operations of the access network device in the above method embodiments.

[0348] For example, the processing module 1001 can be used to determine a second neural network model, which is associated with a first reference signal density, and the second neural network model is used to reconstruct channel information;

[0349] The transceiver module 1002 can be used to transmit a first reference signal on the first channel.

[0350] The embodiments of this application and the method embodiments shown above are based on the same concept and have the same technical effects. For the specific principles, please refer to the description of the embodiments shown above, which will not be repeated here.

[0351] For example, the transceiver module 1002 may include a radio frequency module, an antenna module, etc. For example, the transceiver module 1002 may include a pin module, etc.

[0352] Optionally, in the above embodiments, the communication device may further include a storage module, which can be used to store instructions and / or data. The processing module 1001 can read the instructions and / or data in the storage module to enable the device to implement the aforementioned method embodiments. For example, the storage module may also store the first neural network model, the second neural network model, the reference signal density, etc., as shown above.

[0353] For details regarding the terms or steps of the neural network model, reference signal, reference signal density, reference signal pattern, etc. in each sub-block in the above embodiments, please refer to the description in the above method embodiments, and they will not be described in detail here.

[0354] The specific descriptions of the transceiver module and processing module shown in the above embodiments are merely examples. For the specific functions or execution steps of the transceiver module and processing module, please refer to the above method embodiments, which will not be described in detail here.

[0355] The apparatus of the embodiments of this application has been described above. The possible product forms of the described apparatus are described below. Any device possessing the above-described features... Figure 10 Any form of product that incorporates the functionality of the described device falls within the protection scope of the embodiments of this application. The following description is merely illustrative and does not limit the product form of the device in the embodiments of this application to this specific example.

[0356] In one possible implementation, Figure 10 In the communication device shown, the processing module 1001 can be one or more processing circuits, and the transceiver module 1002 can be a transceiver circuit, or the transceiver module 1002 can also be a transmitting module and a receiving module. The transmitting module can be a transmitting circuit, and the receiving module can be a receiving circuit, which are integrated into one device, such as a transceiver circuit. In the embodiments of this application, the processing circuit and the transceiver circuit can be coupled, etc., and the connection method of the processing circuit and the transceiver circuit is not limited in the embodiments of this application. In the process of performing the above method, the process of sending information in the above method can be the process of the processing circuit outputting the above information. When outputting the above information, the processing circuit outputs the above information to the transceiver circuit so that the transceiver circuit can transmit (or output). After the above information is output by the processing circuit, it may need to undergo other processing before reaching the transceiver circuit. Similarly, the process of receiving information in the above method can be the process of the processing circuit receiving the input above information. When the processing circuit receives the input information, the transceiver circuit receives the above information and inputs it into the processing circuit. Furthermore, after the transceiver circuit receives the aforementioned information, the information may need to undergo further processing before being input into the processing circuit.

[0357] Figure 11 This is a schematic diagram of another communication device provided in an embodiment of this application. For example... Figure 11 As shown, the communication device 110 includes one or more processing circuits 1120 and transceiver circuits 1110.

[0358] In some embodiments of this application, the communication device can be used to perform the steps, methods, or functions performed by the terminal device described above, such as the processing circuit 1120 being used to perform... Figure 10 The transceiver circuit 1110 can be used to perform the functions or steps implemented by the processing module 1001 shown. Figure 10 The transceiver module 1002 shown illustrates the functions or steps implemented by this module. For detailed descriptions of the processing circuit 1120 and the transceiver circuit 1110, please refer to [link / reference needed]. Figure 10 Alternatively, the method embodiments shown above will not be described in detail here.

[0359] In other embodiments of this application, the apparatus is used to perform the steps, methods, or functions performed by the access network device described above. For example, the processing circuit 1120 may be used to perform, for example... Figure 10 The transceiver circuit 1110 can be used to perform the functions or steps implemented by the processing module 1001 shown. Figure 10 The transceiver module 1002 shown illustrates the functions or steps implemented by this module. For detailed descriptions of the processing circuit 1120 and the transceiver circuit 1110, please refer to [link / reference needed]. Figure 10 Alternatively, the method embodiments shown above will not be described in detail here.

[0360] For example, the processing circuitry may be one or more processors, or all or part of the circuitry within one or more processors. The transceiver circuitry may be a transceiver, an input / output circuit, or an interface circuit, etc.

[0361] For example, in Figure 11 In various implementations of the illustrated apparatus, the transceiver circuitry may include a receiver for performing a receiving function (or operation) and a transmitter for performing a transmitting function (or operation). The transceiver circuitry is also used for communicating with other devices / appliances via a transmission medium.

[0362] Optionally, the communication device 110 may further include one or more memories 1130 for storing program instructions and / or data. The memories 1130 are coupled to the processing circuitry 1120. The coupling in this embodiment is an indirect coupling or communication connection between devices, units, or modules, and can be electrical, mechanical, or other forms, used for information exchange between devices, units, or modules. The processing circuitry 1120 may operate in conjunction with the memories 1130. The processing circuitry 1120 may execute the program instructions stored in the memories 1130. Optionally, at least one of the aforementioned memories may be included in the processing circuitry.

[0363] This application embodiment does not limit the specific connection medium between the transceiver circuit 1110, the processing circuit 1120, and the memory 1130. This application embodiment... Figure 11 The memory 1130, processing circuit 1120, and transceiver circuit 1110 are connected via a bus 1140. Figure 11 The connections between other components are shown in bold and are for illustrative purposes only, not as limiting information. The bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0364] In the embodiments of this application, the processing circuit may be a general-purpose processing circuit, a digital signal processing circuit, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processing circuit may be a microprocessor circuit or any conventional processing circuit, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processing circuit, or being executed by a combination of hardware and software modules in the processing circuit, etc.

[0365] In this application embodiment, the memory may include, but is not limited to, non-volatile memory such as hard disk drive (HDD) or solid-state drive (SSD), random access memory (RAM), erasable programmable read-only memory (EPROM), read-only memory (ROM), or compact disc read-only memory (CD-ROM), etc. Memory is any storage medium capable of carrying or storing program code having instruction or data structure forms, and capable of being read and / or written by a computer (such as the device shown in this application), but is not limited to this. The memory in this application embodiment may also be a circuit or any other device capable of implementing storage functions, used to store program instructions and / or data.

[0366] For example, the processing circuit 1120 is mainly used to process communication protocols and communication data, control the entire device, execute software programs, and process the data of the software programs. The memory 1130 is mainly used to store software programs and data. The transceiver circuit 1110 may include a control circuit and an antenna. The control circuit is mainly used for converting baseband signals to radio frequency signals and processing radio frequency signals. The antenna is mainly used for transmitting and receiving radio frequency signals in the form of electromagnetic waves. Input / output devices, such as touch screens, displays, and keyboards, are mainly used to receive user input data and output data to the user.

[0367] When the device is powered on, the processing circuit 1120 can read the software program in the memory 1130, interpret and execute the instructions of the software program, and process the data of the software program. When data needs to be transmitted wirelessly, the processing circuit 1120 performs baseband processing on the data to be transmitted and outputs the baseband signal to the radio frequency (RF) circuit. The RF circuit then performs RF processing on the baseband signal and transmits the RF signal outward in the form of electromagnetic waves through the antenna. When data is sent to the device, the RF circuit receives the RF signal through the antenna, converts the RF signal into a baseband signal, and outputs the baseband signal to the processing circuit 1120. The processing circuit 1120 converts the baseband signal into data and processes the data.

[0368] In another implementation, the radio frequency circuit and antenna can be set up independently of the processing circuit that performs baseband processing. For example, in a distributed scenario, the radio frequency circuit and antenna can be arranged remotely, independent of the device.

[0369] The apparatus shown in the embodiments of this application may also have a higher... Figure 11This application does not limit the use of other components or other related elements. The methods performed by the processing circuit and transceiver circuit shown above are merely examples; the specific steps performed by the processing circuit and transceiver circuit can be found in the methods described above.

[0370] In another possible implementation Figure 10 In the illustrated device, the processing module 1001 can be one or more logic circuits, and the transceiver module 1002 can be an input / output interface, or a communication interface, or an interface circuit, or an interface, etc. Alternatively, the transceiver module 1002 can also be a transmitting module and a receiving module. The transmitting module can be an output interface, and the receiving module can be an input interface. The transmitting module and the receiving module are integrated into one module, such as an input / output interface.

[0371] Figure 12 This is a schematic diagram of another communication device provided in an embodiment of this application. For example... Figure 12 As shown, Figure 12 The communication device shown includes logic circuit 1201 and interface circuit 1202. That is, the processing module 1001 can be implemented using logic circuit 1201, and the transceiver module 1002 can be implemented using interface circuit 1202. The logic circuit 1201 can be a chip, processing circuit, integrated circuit, or system-on-chip (SoC) chip, etc., and the interface circuit 1202 can be a communication interface, input / output interface, pins, etc. For example, Figure 12 The above-mentioned communication device is used as an example of a chip, which includes a logic circuit 1201 and an interface circuit 1202.

[0372] In this embodiment, the logic circuit and the interface can also be coupled to each other. The specific connection method between the logic circuit and the interface is not limited in this embodiment. For example, the logic circuit 1201 can be used to perform... Figure 10 The interface circuit 1202 can be used to execute the functions or steps implemented by the processing module 1001 shown. Figure 10 The transceiver module 1002 shown illustrates the functions or steps implemented by this module. For detailed explanations of the logic circuit 1201 and the interface circuit 1202, please refer to [link / reference needed]. Figure 10 Alternatively, the method embodiments shown above will not be described in detail here.

[0373] The apparatus shown in the embodiments of this application can be implemented in hardware or software, and the embodiments of this application do not limit this.

[0374] This application also provides a communication system, which includes a terminal device and an access network device, and the terminal device and the access network device can be used to perform the methods in any of the foregoing embodiments.

[0375] In addition, this application also provides a computer program for implementing the operations and / or processes performed by various devices in the method provided in this application.

[0376] This application also provides a computer-readable storage medium storing computer code that, when executed on a computer, causes the computer to perform the operations and / or processes performed by the various devices in the methods provided in this application.

[0377] This application also provides a computer program product comprising computer code or a computer program that, when run on a computer, causes the operations and / or processes performed by various entities in the method provided in this application to be executed.

[0378] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or modules, or they may be electrical, mechanical, or other forms of connection.

[0379] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the technical effects of the solutions provided in the embodiments of this application.

[0380] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0381] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0382] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An information acquisition method, characterized in that, The method includes: A first neural network model is determined, which is associated with a first reference signal density, and the first neural network model is used to determine the first reference signal pattern. Obtain the communication resources of the first reference signal in the first reference signal pattern; The first reference signal is received on the first channel based on the communication resources of the first reference signal.

2. The method according to claim 1, characterized in that, The first reference signal density is associated with a second neural network model, which is used to reconstruct channel information; the method further includes: The first channel information corresponding to the first channel is determined based on the second neural network model and the first reference signal.

3. The method according to claim 2, characterized in that, The step of determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal includes: Based on the first reference signal, determine the second channel information corresponding to the communication resources of the first reference signal; The second channel information is processed using the second neural network model to obtain the first channel information corresponding to the first channel.

4. The method according to claim 2 or 3, characterized in that, The step of acquiring the communication resources of the first reference signal in the first reference signal pattern includes: Receive first indication information, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

5. The method according to any one of claims 2-4, characterized in that, The method further includes: Receive the second neural network model.

6. The method according to any one of claims 2-5, characterized in that, The method further includes: Receive second indication information, which is used to indicate that the first neural network model and the second neural network model have been updated.

7. The method according to claim 2 or 3, characterized in that, The method further includes: Send a first indication message, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

8. The method according to claim 7, characterized in that, The method further includes: Update the first neural network model and the second neural network model; Based on the updated first neural network model and the updated second neural network model, the change value corresponding to the first reference signal density is determined; Send a second indication message, which is used to indicate that the first neural network model and the second neural network model have been updated.

9. The method according to claim 6 or 8, characterized in that, The second indication information includes the change value corresponding to the first reference signal density.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: Send the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

11. The method according to claim 10, characterized in that, The method further includes: Receive first information, the first information being used to indicate a supported reference signal density or an identifier of a neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

12. The method according to any one of claims 1-9, characterized in that, The method further includes: Receive the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

13. The method according to claim 12, characterized in that, The method further includes: Send capability information, which is used to indicate a reference signal pattern design that supports a neural network model, or a supported reference signal density or an identifier of a neural network model; the first reference signal density is one of the supported reference signal densities, and the first identifier is one of the identifiers of the supported neural network model.

14. The method according to any one of claims 1-13, characterized in that, The reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

15. An information acquisition method, characterized in that, The method includes: A first neural network model is determined, which is associated with a first reference signal density, and the first neural network model is used to determine the first reference signal pattern. Obtain the communication resources of the first reference signal in the first reference signal pattern; The first reference signal is transmitted on the first channel based on the communication resources of the first reference signal.

16. The method according to claim 15, characterized in that, The first reference signal density is associated with a second neural network model, which is used to reconstruct channel information.

17. The method according to claim 16, characterized in that, The method further includes: Send a first indication message, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

18. The method according to claim 16 or 17, characterized in that, The method further includes: Send the second neural network model.

19. The method according to any one of claims 16-18, characterized in that, The method further includes: Update the first neural network model and the second neural network model; Based on the updated first neural network model and the updated second neural network model, the change value corresponding to the first reference signal density is determined; Send a second indication message, which is used to indicate that the first neural network model and the second neural network model have been updated.

20. The method according to claim 16, characterized in that, The step of acquiring the communication resources of the first reference signal in the first reference signal pattern includes: Receive first indication information, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

21. The method according to claim 20, characterized in that, The method further includes: Receive second indication information, which is used to indicate that the first neural network model and the second neural network model have been updated.

22. The method according to claim 19 or 21, characterized in that, The second indication information includes the change value corresponding to the first reference signal density.

23. The method according to any one of claims 1-22, characterized in that, The method further includes: Receive the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

24. The method according to claim 23, characterized in that, The method further includes: Send first information, the first information being used to indicate a supported reference signal density or an identifier of a neural network model, the first reference signal density being one of the supported reference signal densities, and the first identifier being one of the identifiers of a supported neural network model.

25. The method according to any one of claims 1-22, characterized in that, The method further includes: Send the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

26. The method according to claim 25, characterized in that, The method further includes: The system receives capability information, which indicates a reference signal pattern design that supports a neural network model, or an identifier of a supported reference signal density or neural network model; the first reference signal density is one of the supported reference signal densities, and the first identifier is one of the identifiers of a supported neural network model.

27. The method according to any one of claims 1-26, characterized in that, The reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

28. An information acquisition method, characterized in that, The method includes: A second neural network model is determined, which is associated with the first reference signal density, and the second neural network model is used to reconstruct channel information; Receive the first reference signal on the first channel; The first channel information corresponding to the first channel is determined based on the second neural network model and the first reference signal.

29. The method according to claim 28, characterized in that, The first reference signal density is associated with a first neural network model, which is used to determine the first reference signal pattern; the method further includes: Obtain the communication resources of the first reference signal in the first reference signal pattern; Receiving the first reference signal on the first channel includes: The first reference signal is received on the first channel based on the communication resources of the first reference signal.

30. The method according to claim 29, characterized in that, The step of determining the first channel information corresponding to the first channel based on the second neural network model and the first reference signal includes: Based on the first reference signal, determine the second channel information corresponding to the communication resources of the first reference signal; The second channel information is processed using the second neural network model to obtain the first channel information corresponding to the first channel.

31. The method according to claim 29 or 30, characterized in that, The step of acquiring the communication resources of the first reference signal in the first reference signal pattern includes: Receive first indication information, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

32. The method according to any one of claims 28-31, characterized in that, The method further includes: Receive the second neural network model.

33. The method according to claim 29 or 30, characterized in that, The method further includes: Send a first indication message, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

34. The method according to any one of claims 28-33, characterized in that, The method further includes: Send the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

35. The method according to any one of claims 28-33, characterized in that, The method further includes: Receive the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

36. The method according to any one of claims 28-35, characterized in that, The reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

37. An information acquisition method, characterized in that, The method includes: A second neural network model is determined, which is associated with the first reference signal density, and the second neural network model is used to reconstruct channel information; The first reference signal is transmitted on the first channel.

38. The method according to claim 37, characterized in that, The first reference signal density is associated with a first neural network model, which is used to determine the first reference signal pattern; the method further includes: Obtain the communication resources of the first reference signal in the first reference signal pattern; The step of transmitting the first reference signal on the first channel includes: The first reference signal is transmitted on the first channel based on the communication resources of the first reference signal.

39. The method according to claim 38, characterized in that, The method further includes: Send a first indication message, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

40. The method according to any one of claims 37-39, characterized in that, The method further includes: Send the second neural network model.

41. The method according to claim 38, characterized in that, The step of acquiring the communication resources of the first reference signal in the first reference signal pattern includes: Receive first indication information, which is used to indicate the communication resources of the first reference signal in the first reference signal pattern.

42. The method according to any one of claims 37-41, characterized in that, The method further includes: Receive the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

43. The method according to any one of claims 37-41, characterized in that, The method further includes: Send the first reference signal density or the first identifier, wherein the first identifier is used to indicate the first reference signal density.

44. The method according to any one of claims 37-43, characterized in that, The reference signal density is characterized by at least one of the following: time-domain reference signal density, frequency-domain reference signal density, spatial-domain reference signal density, the number of selected antenna ports, or the total number of antenna ports.

45. A communication device, characterized in that, It includes a module for performing the method as described in any one of claims 1-14, or a module for performing the method as described in any one of claims 15-27, or a module for performing the method as described in any one of claims 28-36, or a module for performing the method as described in any one of claims 37-44.

46. ​​A communication device, characterized in that, The device includes a processing circuit and a transceiver circuit, the transceiver circuit being used to input and / or output information, the processing circuit being used to perform the method as described in any one of claims 1-14, or the processing circuit being used to perform the method as described in any one of claims 15-27, or the processing circuit being used to perform the method as described in any one of claims 28-36, or the processing circuit being used to perform the method as described in any one of claims 37-44.

47. A chip, characterized in that, It includes a processing circuit and an interface circuit, the processing circuit and the interface circuit being coupled; the interface circuit is used for inputting and / or outputting information, and the processing circuit is used for executing code instructions to cause the method of any one of claims 1-14 to be executed, or to cause the method of any one of claims 15-27 to be executed, or to cause the method of any one of claims 28-36 to be executed, or to cause the method of any one of claims 37-44 to be executed.

48. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when executed, performs the method as claimed in any one of claims 1-14, or the method as claimed in any one of claims 15-27, or the method as claimed in any one of claims 28-36, or the method as claimed in any one of claims 37-44.

49. A computer program product, characterized in that, When the computer program product is executed, the method as described in any one of claims 1-14 is executed, or the method as described in any one of claims 15-27 is executed, or the method as described in any one of claims 28-36 is executed, or the method as described in any one of claims 37-44 is executed.

Images