WO2024007172A1 - Channel estimation method and apparatus - Google Patents

Abstract

Disclosed in embodiments of the present application are a channel estimation method and apparatus. The method comprises: receiving a first demodulation reference signal (DMRS) sent by a network device on the basis of a first DMRS pattern, and according to the first DMRS, performing channel estimation on the basis of a channel estimation model. Therefore, terminal devices having different capabilities can support channel estimation based on an artificial intelligence technology, thereby effectively improving the accuracy of channel estimation, greatly increasing the success rate of decoding, effectively improving the spectrum efficiency of a communication system, and saving the pilot overhead of the system.

Description

Channel estimation method and device Technical field

The present application relates to the field of communication technology, and in particular, to a channel estimation method and device.

Background technique

With the continuous development and maturity of 5G and Artificial Intelligence (AI) technology, AI-assisted wireless communications are also gradually developing. For example, AI-assisted modulation and demodulation and radio frequency technology, including AI-assisted Channel State Information (CSI) feedback and AI-assisted beam management, can improve the speed and coverage of 5G networks, and improve the mobility and robustness of the system. Integrating AI technology into the design of wireless communication systems is also an important development direction of 6G in the future.

Contents of the invention

The first embodiment of the present application proposes a channel estimation method. The method is executed by a terminal device. The method includes: receiving a first DMRS sent by a network device based on a first demodulation reference signal DMRS pattern; according to the first DMRS pattern. A DMRS performs channel estimation based on the channel estimation model.

The second embodiment of the present application proposes a channel estimation method, which is executed by a network device. The method includes: sending a first DMRS to a terminal device based on a first demodulation reference signal DMRS pattern; the first DMRS Used for channel estimation based on the channel estimation model.

The third aspect embodiment of the present application proposes a channel estimation method, which is executed by a network device. The method includes: receiving a first DMRS sent by a terminal device based on a first demodulation reference signal DMRS pattern; A DMRS performs channel estimation based on the channel estimation model.

The fourth embodiment of the present application proposes a channel estimation method, which is executed by a terminal device. The method includes: sending a first DMRS to a network device based on a first demodulation reference signal DMRS pattern; the first DMRS Used for channel estimation based on the channel estimation model.

The fifth embodiment of the present application provides a channel estimation device, which includes:

A transceiver unit configured to receive the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern;

A processing unit configured to perform channel estimation based on the channel estimation model according to the first DMRS.

The sixth embodiment of the present application provides a channel estimation device, which includes:

A transceiver unit, configured to send the first DMRS to the terminal device based on the first demodulation reference signal DMRS pattern;

The first DMRS is used for channel estimation based on a channel estimation model.

The seventh embodiment of the present application provides a channel estimation device, which includes:

A transceiver unit configured to receive the first DMRS sent based on the first demodulation reference signal DMRS pattern sent by the terminal device;

A processing unit configured to perform channel estimation based on a channel estimation model according to the first DMRS.

The eighth embodiment of the present application provides a channel estimation device, which includes:

A transceiver unit configured to send the first DMRS to the network device based on the first demodulation reference signal DMRS pattern;

The first DMRS is used for channel estimation based on a channel estimation model.

The ninth aspect of the present application provides a communication device. The device includes a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory so that the The device performs the channel estimation method described in the embodiment of the first aspect, or performs the channel estimation method described in the embodiment of the second aspect.

The tenth embodiment of the present application provides a communication device. The device includes a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory so that the The device performs the channel estimation method described in the third embodiment, or performs the channel estimation method described in the fourth embodiment.

An eleventh aspect embodiment of the present application provides a communication device. The device includes a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to cause The device performs the channel estimation method described in the above-mentioned first aspect embodiment, or performs the channel estimation method described in the above-mentioned second aspect embodiment.

The twelfth embodiment of the present application provides a communication device. The device includes a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to cause The device performs the channel estimation method described in the third embodiment, or performs the channel estimation method described in the fourth embodiment.

The thirteenth embodiment of the present application proposes a computer-readable storage medium for storing instructions. When the instructions are executed, the channel estimation method described in the first embodiment is implemented, or the channel estimation method is implemented. The channel estimation method described in the above embodiment of the second aspect is implemented.

The fourteenth embodiment of the present application provides a computer-readable storage medium for storing instructions. When the instructions are executed, the channel estimation method described in the third embodiment is implemented, or the channel estimation method is implemented. The channel estimation method described in the above embodiment of the fourth aspect is implemented.

The fifteenth embodiment of the present application proposes a computer program that, when run on a computer, causes the computer to perform the channel estimation method described in the embodiment of the first aspect, or perform the channel estimation described in the embodiment of the second aspect. method.

The sixteenth embodiment of the present application proposes a computer program that, when run on a computer, causes the computer to perform the channel estimation method described in the embodiment of the third aspect, or perform the channel estimation described in the embodiment of the fourth aspect. method.

A channel estimation method and device provided by embodiments of the present application, by receiving the first DMRS sent by the network equipment based on the first demodulation reference signal DMRS pattern, and performing channel estimation based on the channel estimation model based on the first DMRS, so that different capabilities All terminal equipment can support channel estimation based on artificial intelligence technology, which effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application.

Description of the drawings

In order to more clearly explain the technical solutions in the embodiments of the present application or the background technology, the drawings required to be used in the embodiments or the background technology of the present application will be described below.

Figure 1 is a schematic architectural diagram of a communication system provided by an embodiment of the present application;

Figure 2 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 3 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 4 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 5 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 6 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 7 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 8 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 9 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 10 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 11 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application;

Figure 12 is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application;

Figure 13 is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application;

Figure 14 is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application;

Figure 15 is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application;

Figure 16 is a schematic structural diagram of another channel estimation device provided by an embodiment of the present application;

Figure 17 is a schematic structural diagram of a chip provided by an embodiment of the present disclosure.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the embodiments of the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of embodiments of the present application as detailed in the appended claims.

The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the embodiments of the present application. As used in the embodiments and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It will also be understood that the term "and/or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used to describe various information in the embodiments of this application, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the embodiments of the present application, the first information may also be called second information, and similarly, the second information may also be called first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "when" or "when" or "in response to determining."

The embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements throughout. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present application, but should not be construed as limiting the present application.

In order to better understand the channel estimation method disclosed in the embodiment of the present application, the communication system to which the embodiment of the present application is applicable is first described below.

Please refer to Figure 1. Figure 1 is a schematic architectural diagram of a communication system provided by an embodiment of the present application. The communication system may include but is not limited to one network device and one terminal device. The number and form of devices shown in Figure 1 are only for examples and do not constitute a limitation on the embodiments of the present application. In actual applications, two or more devices may be included. network equipment and two or more terminal devices. The communication system shown in Figure 1 includes a network device 101 and a terminal device 102 as an example.

It should be noted that the technical solutions of the embodiments of the present application can be applied to various communication systems. For example: Long Term Evolution (LTE) system, fifth-generation mobile communication system, 5G new air interface system, or other future new mobile communication systems.

The network device 101 in the embodiment of this application is an entity on the network side that is used to transmit or receive signals. For example, the network device 101 may be an evolved base station (Evolved NodeB, eNB), a transmission point (Transmission Reception Point, TRP), a next generation base station (Next Generation NodeB, gNB) in an NR system, or other base stations in future mobile communication systems. Base stations or access nodes in wireless fidelity (Wireless Fidelity, WiFi) systems, etc. The network device 101 in the embodiment of this application can be the network device itself, or it can be a network upper layer (Over The Top, OTT) server (OTT server) maintained by an operator, a base station manufacturer or a third party, or it can be an operation and maintenance management (OTT) server. Operation Administration and Maintenance (OAM), Location Management Function (LMF), etc. The embodiments of this application do not limit the specific technology and specific equipment form used by the network equipment. The network equipment provided by the embodiments of this application may be composed of a centralized unit (Central Unit, CU) and a distributed unit (Distributed Unit, DU). The CU may also be called a control unit (Control Unit), using CU-DU. The structure can separate the protocol layers of network equipment, such as base stations, and place some protocol layer functions under centralized control on the CU. The remaining part or all protocol layer functions are distributed in the DU, and the CU centrally controls the DU.

The terminal device 102 in the embodiment of this application is an entity on the user side that is used to receive or transmit signals, such as a mobile phone. Terminal equipment can also be called terminal equipment (terminal), user equipment (UE), mobile station (Mobile Station, MS), mobile terminal equipment (Mobile Terminal, MT), etc., or it can also be a reduced capability terminal equipment (RedCap UE), evolved reduced capability terminal equipment (eRedCap UE), etc. Terminal devices can be cars with communication functions, smart cars, mobile phones, wearable devices, tablets (Pad), computers with wireless transceiver functions, virtual reality (Virtual Reality, VR) terminal devices, augmented reality ( Augmented Reality (AR) terminal equipment, wireless terminal equipment in industrial control (Industrial Control), wireless terminal equipment in self-driving (Self-Driving), wireless terminal equipment in remote surgery (Remote Medical Surgery), smart grid ( Wireless terminal equipment in Smart Grid, wireless terminal equipment in Transportation Safety, wireless terminal equipment in Smart City, wireless terminal equipment in Smart Home, etc. The terminal device 102 in the embodiment of this application can be the terminal device itself, or it can be an OTT server maintained by a user equipment supplier (UE vendor), a chip manufacturer, or a third party. The embodiments of this application do not limit the specific technology and specific equipment form used by the terminal equipment.

With the continuous development and maturity of 5G and Artificial Intelligence (AI) technology, AI-assisted wireless communications are also gradually developing. For example, AI-assisted modulation and demodulation and radio frequency technology, including AI-assisted Channel State Information (CSI) feedback and AI-assisted beam management, can improve the speed and coverage of 5G networks, and improve the mobility and robustness of the system. Integrating AI technology into the design of wireless communication systems is also an important development direction of 6G in the future.

In typical AI application scenarios, such as image processing, autonomous driving, etc., the power consumption of AI algorithms can usually be evaluated through FLOPs/mW, FLOPs/W or GFLOPs/mW. Among them, FLOPs is the abbreviation of floating point operations, which means floating point operations. It can be understood as the amount of calculation and can be used to measure the complexity of the algorithm or model. GFLOPs are one billion floating point operations.

It can be understood that the power consumption of a communication device using an AI model to perform one inference = the computational complexity of the AI model (FLOPs)/the capability of the communication device (FLOPs/mW).

For communication equipment, FLOPs/mW, as a hardware capability, has a lot to do with the specific CPU process design and heat dissipation design. In some scenarios, when the computing energy consumption of the terminal is high, or the supported computing power is lower than a certain threshold, AI model training may not be completed quickly, resulting in model training having to be performed on the network side.

For AI-based downlink channel estimation methods, model training should be performed on the terminal side as much as possible. However, there may be some terminal devices that do not have the ability to train AI models and require network equipment to assist in training.

It can be understood that the communication system described in the embodiments of the present application is to more clearly illustrate the technical solutions of the embodiments of the present application, and does not constitute a limitation on the technical solutions provided by the embodiments of the present application. As those of ordinary skill in the art will know, With the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

The channel estimation method and device provided by this application will be introduced in detail below with reference to the accompanying drawings.

Please refer to Figure 2. Figure 2 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 2, the method may include the following steps:

Step 201: Receive the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern.

In this embodiment of the present application, the terminal device can receive a first demodulation reference signal (Demodulation Reference Signal, DMRS) sent by the network device. The first DMRS is sent by the network device based on a first DMRS pattern. After receiving the first DMRS, the terminal device can perform channel estimation based on the first DMRS based on the trained channel estimation model.

In some implementations, the terminal device can send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

Optionally, the first indication information may include at least one of the following: model training capability indication information of the terminal device; hardware processing capability information of the terminal device; computing capability information of the terminal device; power consumption capability of the terminal device. information.

Among them, the model training capability indication information of the terminal device can indicate whether the terminal device has the model training capability. The model training capability indication information may be at least 1 bit.

In some possible implementations, the terminal device can also send model inference capability indication information of the terminal device to the network device. The model inference capability indication information of the terminal device can indicate whether the terminal device has the ability to use a channel estimation model for model inference. ability. The model reasoning capability indication information may also be at least 1 bit.

It can be understood that in various embodiments of the present application, the terminal device has model reasoning capabilities and can perform channel estimation based on a trained channel estimation model.

Optionally, the terminal device can be based on thresholds related to model training and inference configured by the network device or specified by the protocol, such as training delay threshold, training power consumption threshold, training calculation complexity threshold, etc. To determine whether it has model training capabilities; based on the inference delay threshold, inference power consumption threshold, and inference calculation complexity threshold, to determine whether it has model inference capabilities, and then report the model training capabilities of the terminal device Indicative information and/or model inference capability indication information. You can also directly judge whether you have model training capabilities based on your own capabilities, such as whether you have a graphics processor (Graphics Processing Unit, GPU), a neural network processor (Neural network Processing Unit, NPU), and power storage, etc., and Whether there is model inference capability, and then report the model training capability indication information and/or model inference capability indication information of the terminal device.

Optionally, the terminal device may directly send the model training capability indication information and model reasoning capability indication information of the terminal device to the network device to indicate whether the terminal device has model training capability and model reasoning capability. Alternatively, if the terminal device has the model training capability, indication information indicating that the terminal device has the model training capability can be sent to the network device and reported as having the model training capability; if the terminal device does not have the model training capability, the terminal device can directly report to the network device Send model reasoning capability indication information and report whether it has model reasoning capability. As an example, the terminal device may send indication information indicating that it has model training capabilities to the network device, reporting that the terminal device has model training capabilities, implicitly indicating that the terminal device also has model inference capabilities. As another example, the terminal device sends indication information indicating that it has the model inference capability to the network device, and reports that the terminal device has the model inference capability, which implicitly indicates that the terminal device does not have the model training capability.

In some embodiments, the terminal device can send hardware processing capability information, computing capability information, power consumption capability information, etc. to the network device, and the network device can determine some thresholds according to the delay requirements of the service or the protocol, such as the delay threshold. , power consumption threshold, computational complexity threshold, etc., to determine whether the terminal device has model training capabilities and model reasoning capabilities.

Optionally, the first indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information (UE Assistance Information, UAI); unlimited resource control (Radio Resource Control, RRC) Signaling; Medium Access Control (MAC) control element (Control Element, CE, or control unit); Uplink Control Information (UCI). The first indication information may also be sent through a physical uplink shared channel (Physical Uplink Shared Channel, PUSCH).

Step 202: According to the first DMRS, perform channel estimation based on the channel estimation model.

In this embodiment of the present application, the terminal device can perform channel estimation based on the received first DMRS based on the trained channel estimation model.

It can be understood that in the embodiment of the present application, the terminal device can directly use the received DMRS signal as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the DMRS, and use the channel estimate value as the channel The input of the estimation model is not limited by this application. The channel estimate at the DMRS estimated based on the DMRS can be estimated using the least squares method (least squares, LS), or the minimum mean square error method (minimum mean square error, MMSE). It can also be estimated using Other estimation algorithms, etc. are not limited by this application.

In the embodiment of this application, the training of the channel estimation model can be performed by the terminal device or the network device; the training can be performed using actual data or simulated data; the training can be performed offline or online. Conduct training.

In some implementations, the terminal device can receive the second DMRS sent by the network device based on the second DMRS pattern, and determine the training data of the channel estimation model based on the second DMRS.

Optionally, the terminal device can use the determined training data to train the channel estimation model.

Optionally, the terminal device can send the training data to the network device, and the network device uses the training data to train the channel estimation model.

Optionally, before determining the training data to be sent to the network device, the terminal device can also receive fourth indication information sent by the network device, where the fourth indication information is used to indicate the type of the training data. For example, it may be indicated that the training data is a received signal corresponding to the second DMRS, or it may be indicated that the training data is a channel estimate value estimated based on the second DMRS, and so on. The terminal device can determine what kind of training data the network device needs for model training according to the instructions of the fourth instruction information, and determine the training data according to the received second DMRS and send it to the network device.

In some embodiments, the terminal device can obtain the simulated signal received by the terminal device in the simulated channel, where the simulated signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulated channel, and the terminal device can obtain the simulated signal according to the simulated signal. , determine the simulation training data of the channel estimation model, and use the simulation training data to train the channel estimation model.

In some implementations, the channel estimation model is trained by the network device using simulation training data. The network device can also obtain the simulated signal received by the terminal device in the simulated channel, where the simulated signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulated channel. The network device can also determine the simulation training data of the channel estimation model based on the simulation signal, and use the simulation training data to train the channel estimation model.

In the embodiment of the present application, when the channel estimation model is trained by the network device, the terminal device can receive the trained channel estimation model sent by the network device.

In some implementations of the embodiments of this application, when the channel estimation model is trained by a terminal device, the terminal device can also send second indication information to the network device, where the second indication information is used to instruct the channel estimation model to be trained. Finish.

Optionally, the second indication information may also include at least one of the following information: capability information of the channel estimation model, and processing delay information of the channel estimation model.

Among them, the capability information of the channel estimation model refers to the capability of the channel estimation model compared with traditional channel estimation methods. For example, the model can use lower density DMRS compared with traditional patterns for channel estimation, or It can obtain higher-precision channel estimation results compared with traditional channel estimation methods, etc. The processing delay information of the channel estimation model refers to the processing delay of the terminal device when using the model, which can include the loading time of the model, the time of using the model for inference, etc.

The network device can determine that the channel estimation model has been trained based on the second indication information. At the same time, it can also obtain the capability information of the model and/or the processing delay information of the model, and can estimate the channel estimation model based on the capability information and the processing delay information. Terminal equipment performs reasonable scheduling.

Optionally, the second indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information UAI; unlimited resource control RRC signaling; media access control layer control element MAC CE ;Uplink control information UCI. The second indication information can also be sent through PUSCH.

In some embodiments, when the channel estimation model is trained by the terminal device, the terminal device can also receive third instruction information sent by the network device. The third instruction information is used to instruct the terminal device to start training of the channel estimation model. . The third indication information may be at least 1 bit.

In some embodiments, when the channel estimation model is trained by a terminal device, the terminal device may also directly start training the channel estimation model, or may start training the model after sending the first indication information for more than a preset time. train. The preset time may be configured by the network device or agreed or stipulated by the protocol.

In some embodiments, the network device may also send disabling signaling to the terminal device according to business needs and conditions to instruct the terminal device not to start model training.

In some embodiments, if the channel estimation model is trained using a supervised machine learning method, the terminal device can also receive the impulse signal sent by the network device and obtain the ideal channel estimation label of the channel based on the impulse signal. The channel estimation tag is used for the training of this channel estimation model.

In some embodiments, for the case where the channel estimation model is trained by the terminal device, if the channel estimation model is trained using a supervised machine learning method, the terminal device also needs to send auxiliary information to the network device to request the network device to issue impulse signal. The terminal device can obtain the ideal channel estimation label of the channel based on the impulse signal, and use the ideal channel estimation label to train the channel estimation model.

In some embodiments, the channel estimation model has the ability to use low-density DMRS for channel estimation, and the density of the first DMRS pattern is lower than the density of the second DMRS pattern. The second DMRS pattern may be a legacy DMRS pattern. Terminal equipment can use lower-density DMRS compared to legacy DMRS patterns, and obtain channel estimation results based on this channel estimation model.

In some embodiments, the channel estimation model has the capability of high-precision channel estimation results, and the density of the first DMRS pattern is the same as the density of the second DMRS pattern. Wherein, the second DMRS pattern can be a legacy DMRS pattern, and the terminal device can use DMRS with the same density as the legacy DMRS pattern, and obtain higher-precision channel estimation results based on the channel estimation model compared with traditional channel estimation methods.

In some implementations, the terminal device can also receive fifth indication information sent by the network device, where the fifth indication information is used to instruct the terminal device to perform channel estimation based on the channel estimation model. Only when the terminal device receives the fifth indication information, the trained channel estimation model will be enabled for channel estimation.

Optionally, the fifth indication information may be at least 1 bit of information, directly instructing the terminal device to enable the trained channel estimation model to perform channel estimation. The fifth indication information may also be the first DMRS pattern configuration sent by the network device to the terminal device to reduce pilot overhead.

In summary, by receiving the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern, and performing channel estimation based on the channel estimation model according to the first DMRS, terminal devices with different capabilities can support channels based on artificial intelligence technology. Estimation effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 3. Figure 3 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 3, the method may include the following steps:

Step 301: Send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

In this embodiment of the present application, the terminal device sends first indication information to the network device for reporting whether it has model training capabilities.

Optionally, the first indication information may include at least one of the following: model training capability indication information of the terminal device; hardware processing capability information of the terminal device; computing capability information of the terminal device; power consumption capability of the terminal device. information.

Among them, the model training capability indication information of the terminal device can indicate whether the terminal device has the model training capability. The model training capability indication information may be at least 1 bit. As an example, "0" can be used to represent that the terminal device does not have the model training capability, and "1" represents that the terminal device has the model training capability.

In some possible implementations, the terminal device can also send model inference capability indication information of the terminal device to the network device. The model inference capability indication information of the terminal device can indicate whether the terminal device has the ability to use a channel estimation model for model inference. ability. The model reasoning capability indication information may also be at least 1 bit. As an example, "0" can be used to represent that the terminal device does not have model reasoning capabilities, and "1" represents that the terminal device has model reasoning capabilities.

It can be understood that in various embodiments of the present application, the terminal device has model reasoning capabilities and can perform channel estimation based on a trained channel estimation model.

Optionally, the terminal device can be based on thresholds related to model training and inference configured by the network device or specified by the protocol, such as training delay threshold, training power consumption threshold, training calculation complexity threshold, etc. To determine whether it has model training capabilities; based on the inference delay threshold, inference power consumption threshold, and inference calculation complexity threshold, to determine whether it has model inference capabilities, and then report the model training capabilities of the terminal device Indicative information and/or model inference capability indication information. You can also directly determine whether you have model training capabilities and model reasoning capabilities based on your own capabilities, such as whether you have a GPU, NPU, and power storage, etc., and then report the model training capability indication information and/or model of the terminal device. Reasoning skills indicate information.

Optionally, the terminal device may directly send the model training capability indication information and model reasoning capability indication information of the terminal device to the network device to indicate whether the terminal device has model training capability and model reasoning capability. Alternatively, if the terminal device has the model training capability, indication information indicating that the terminal device has the model training capability can be sent to the network device and reported as having the model training capability; if the terminal device does not have the model training capability, the terminal device can directly report to the network device Send model reasoning capability indication information and report whether it has model reasoning capability. As an example, the terminal device may send indication information indicating that it has model training capabilities to the network device, reporting that the terminal device has model training capabilities, implicitly indicating that the terminal device also has model inference capabilities. As another example, the terminal device sends indication information indicating that it has the model inference capability to the network device, and reports that the terminal device has the model inference capability, which implicitly indicates that the terminal device does not have the model training capability.

In some embodiments, the terminal device can send hardware processing capability information, computing capability information, power consumption capability information, etc. to the network device, and the network device can determine some thresholds according to the delay requirements of the service or the protocol, such as the delay threshold. , power consumption threshold, computational complexity threshold, etc., to determine whether the terminal device has model training capabilities and model reasoning capabilities.

In some embodiments, the terminal device may have at least one model training capability and/or at least one model reasoning capability, and the first indication information can be used to determine the model training capability and/or model reasoning capability of the terminal device. As an example, the first indication information is model training capability indication information. The model training capability indication information is "00", which means that the terminal device does not have the model training capability. The model training capability indication information is "01", "02", " 03" all represent that the terminal device has model training capabilities. Different values represent different levels of model training capabilities. The larger the value, the stronger the training capability. Similarly, model reasoning capability indication information can also be reported in a similar manner. It can be understood that other methods can also be used to determine the model training capability of the terminal device, which is not limited in this application.

Optionally, the first indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information UAI; unlimited resource control RRC signaling; media access control layer control element MACCE; Uplink control information UCI. The first indication information may also be sent through PUSCH.

It can be understood that in this embodiment of the present application, the terminal device has model training capabilities and model reasoning capabilities.

Step 302: Receive the second DMRS sent by the network device based on the second DMRS pattern.

In this embodiment of the present application, the terminal device can receive the second DMRS sent by the network device based on the second DMRS pattern, and can determine the training data of the channel estimation model based on the second DMRS, and use the training data to train the channel estimation model.

The second DMRS pattern may be a legacy DMRS pattern.

In this embodiment of the present application, the network device sends a reference signal to the terminal device based on the second DMRS pattern, and the terminal device collects actual data as training data based on the actual reference signal received to train the model.

Step 303: Determine training data for the channel estimation model according to the second DMRS.

In this embodiment of the present application, the terminal device can determine the training data of the channel estimation model based on the received second DMRS, and the training data is actual data obtained through actual channel transmission.

Optionally, in this embodiment of the present application, the terminal device may directly use the received signal of the second DMRS as training data for the channel estimation model, or may obtain a channel estimate value estimated based on the second DMRS, and convert the channel The estimated value is used as training data for the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc. can be used. This application There is no limit to this either.

In some embodiments, the terminal device uses a supervised machine learning method to train the channel estimation model. The terminal device can also receive the impulse signal sent by the network device, and can obtain the ideal channel label of the channel based on the impulse signal, using for the training of channel estimation models.

Optionally, the impulse signal can be transmitted using a semi-static scheduling method. The terminal device can also receive the configuration information of the impulse signal sent by the network device, which can include the transmission period of the impulse signal and the occupied time-frequency resources. etc. The impulse signal can also be transmitted using dynamic scheduling. The terminal device can also receive the scheduling information of the impulse signal sent by the network device, which can include the time-frequency domain resources occupied by the impulse signal transmission and the action indication domain of the DCI. etc. (such as specially used for scheduling impulse signals).

Step 304: Use training data to train the channel estimation model.

In this embodiment of the present application, the terminal device can use the training data determined in the previous steps to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiments of this application can be constructed and trained based on any machine learning method, such as convolutional neural networks (Convolutional Neural Networks, CNN), etc. This application does not Make restrictions.

In some embodiments, the terminal device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also use the ideal channel label of the channel obtained based on the received impulse signal to train the channel estimation model.

It should be noted that the impulse signal sent by the network device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

In some embodiments, the terminal device uses a supervised machine learning method to train the channel estimation model. The terminal device can also send auxiliary information to the network device. The auxiliary information is used to request the impulse signal. The network device receives the Auxiliary information, sending the impulse signal to the terminal device.

Optionally, the auxiliary information may be a request message, or may be instruction information indicating a training method of the channel estimation model (for example, indicating that the model is trained using a supervised/unsupervised machine learning method).

In some implementations, the terminal device can also receive third instruction information sent by the network device, where the third instruction information is used to instruct the terminal device to start training the model.

Optionally, the third indication information may be bit information with a value of "0" or "1", used to indicate whether the terminal device enables model training (for example, the third indication information is "0" to indicate disabling, The terminal device does not start model training, and the third indication information is "1" to indicate enablement, and the terminal device starts model training). The third indication information may also be an impulse signal sent by the network device, or a corresponding configuration of the impulse signal. Receiving the signal or the corresponding configuration of the signal indicates that the terminal device is enabled to start model training, and the terminal device can After receiving the impulse signal or the corresponding configuration of the impulse signal, training of the channel estimation model begins.

In some implementations, the terminal device can directly start the training of the channel estimation model, or start the training of the channel estimation model after sending the aforementioned first indication information for more than a preset time. The preset time may be configured by the network device, or may be pre-agreed or stipulated by the protocol.

It should be noted that in the embodiment of the present application, the training of the model can be performed online or offline, and this application does not limit this.

Step 305: Send second indication information to the network device, where the second indication information is used to indicate that the channel estimation model training is completed.

In this embodiment of the present application, after completing the training of the channel estimation model, the terminal device can also send second indication information to the network device, and the network device can determine that the channel estimation model training is completed based on the second indication information.

Optionally, in addition to model training end indication information, the second indication information may also include at least one of the following information: capability information of the channel estimation model, and processing delay information of the channel estimation model. Alternatively, the model training end indication information is implicitly indicated by at least one of the above two types of information.

Among them, the capability information of the channel estimation model refers to the capability of the channel estimation model compared with traditional channel estimation methods. For example, the model can use lower density DMRS compared with traditional patterns for channel estimation, or It can obtain higher-precision channel estimation results compared with traditional channel estimation methods, etc. The processing delay information of the channel estimation model refers to the processing delay of the terminal device when using the model, which can include the loading time of the model, the time of using the model for inference, etc.

The network device can determine that the channel estimation model has been trained based on the second indication information. At the same time, it can also obtain the capability information of the model and/or the processing delay/model complexity/model inference power consumption and other information of the model, and can Reasonably schedule the terminal device based on the capability information and processing delay information; and decide whether to enable the AI model based on the model's processing delay/model complexity/model inference power consumption and other information.

Optionally, the second indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information UAI; unlimited resource control RRC signaling; media access control layer control element MAC CE ;Uplink control information UCI. The second indication information may also be sent through PUSCH.

Step 306: Receive the first DMRS sent by the network device based on the first DMRS pattern.

In this embodiment of the present application, the terminal device can receive the first DMRS sent by the network device, and the first DMRS is sent by the network device based on the first DMRS pattern. The network device can determine the first DMRS pattern based on the capability of the channel estimation model. After receiving the first DMRS, the terminal device can perform channel estimation based on the first DMRS based on the trained channel estimation model.

In some embodiments, the channel estimation model has the ability to use low-density DMRS for channel estimation, and the density of the first DMRS pattern is lower than the density of the second DMRS pattern. The second DMRS pattern may be a legacy DMRS pattern. Terminal equipment can use lower-density DMRS compared to legacy DMRS patterns, and obtain channel estimation results based on this channel estimation model.

In some embodiments, the channel estimation model has the capability of high-precision channel estimation results, and the density of the first DMRS pattern is the same as the density of the second DMRS pattern. Wherein, the second DMRS pattern can be a legacy DMRS pattern, and the terminal device can use DMRS with the same density as the legacy DMRS pattern, and obtain higher-precision channel estimation results based on the channel estimation model compared with traditional channel estimation methods.

It can be understood that in the embodiment of the present application, the terminal device can directly use the received signal of the first DMRS as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the first DMRS, and convert the channel The estimated value is used as the input of the channel estimation model, and other data can also be obtained as input according to the configuration of the channel estimation model, which is not limited in this application. The channel estimation value is obtained based on the first DMRS. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application does not do this. Make restrictions.

Step 307: According to the first DMRS, perform channel estimation based on the channel estimation model.

In this embodiment of the present application, the terminal device can perform channel estimation based on the received first DMRS and the trained channel estimation model.

It can be understood that in the embodiment of the present application, the terminal device can directly use the received signal of the first DMRS as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the first DMRS, and convert the channel The estimated value serves as the input of the channel estimation model, and other data can also be obtained as input based on the configuration of the channel estimation model, which is not limited in this application. The channel estimation value is obtained based on the first DMRS. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application does not do this. Make restrictions.

In some implementations, the terminal device can also receive fifth indication information sent by the network device, where the fifth indication information is used to instruct the terminal device to perform channel estimation based on the channel estimation model. Only when the terminal device receives the fifth indication information, the trained channel estimation model will be enabled for channel estimation.

Optionally, the fifth indication information may be at least 1 bit of information, directly instructing the terminal device to enable the trained channel estimation model for channel estimation (for example, the fifth indication information is "0" to indicate disablement, and the terminal device does not enable it. The channel estimation model performs channel estimation, and the fifth indication information is "1" indicating enablement, and the terminal device enables the channel estimation model to perform channel estimation). The fifth indication information may also be the first DMRS pattern configuration sent by the network device to the terminal device to reduce pilot overhead.

It can be understood that if the terminal device does not enable the channel estimation model for channel estimation, the densities of the first DMRS pattern and the second DMRS pattern are the same, and the terminal device can use the traditional channel estimation algorithm to perform channel estimation based on the received DMRS. estimate.

In summary, by sending the first indication information to the network device, the first indication information is used to indicate whether the terminal device has model training capabilities, and receiving the second DMRS sent by the network device based on the second DMRS pattern. According to the second DMRS, Determine the training data of the channel estimation model, use the training data to train the channel estimation model, and send second indication information to the network device. The second indication information is used to indicate that the channel estimation model training is completed. The receiving network device based on the first DMRS According to the first DMRS sent by the pattern, channel estimation is performed based on the channel estimation model, so that terminal equipment with different capabilities can support channel estimation based on artificial intelligence technology, effectively improving the accuracy of channel estimation, thereby significantly improving The decoding success rate effectively improves the spectrum efficiency of the communication system and saves the pilot overhead of the system.

Please refer to Figure 4. Figure 4 is a schematic flowchart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 4, the method may include the following steps:

Step 401: Send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

In the embodiment of the present application, step 401 can be implemented in any manner among the embodiments of the present application, which is not limited by the embodiment of the present application and will not be described again.

Step 402: Obtain the simulated signal received by the terminal device in the simulated channel. The simulated signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulated channel.

In this embodiment of the present application, the terminal device can obtain the simulation signal received by the terminal device in the simulation channel, and the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel. And can determine the training data of the channel estimation model based on the simulation signal, and use the training data to train the channel estimation model.

The second DMRS pattern may be a legacy DMRS pattern.

In the embodiment of the present application, in the simulated channel model, the network device sends DMRS to the terminal device based on the second DMRS pattern. The terminal device can obtain the simulated signal received in the simulated channel, and determine the simulated data as training data to perform model development. train.

Step 403: Determine simulation training data for the channel estimation model based on the simulation signal.

In this embodiment of the present application, the terminal device can determine the training data of the channel estimation model based on the acquired simulation signal. The training data is simulation data transmitted in the simulation channel.

Optionally, in this embodiment of the present application, the terminal device may directly use the received simulated signal of the second DMRS as training data for the channel estimation model, or may obtain a channel estimate value estimated based on the simulated signal of the second DMRS. , the channel estimation value is used as the training data of the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc. can be used. This application There is no restriction on this either.

In some embodiments, the terminal device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also obtain the ideal channel label of the simulated channel for training of the channel estimation model.

Step 404: Use the simulation training data to train the channel estimation model.

In this embodiment of the present application, the terminal device can use the simulation training data determined in the previous steps to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In some embodiments, the terminal device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also obtain the ideal channel label of the simulated channel to train the channel estimation model. It can be understood that in the simulated channel model, the ideal channel label of the simulated channel can be obtained by establishing the channel parameters of the simulated channel model.

In some implementations, the terminal device can also receive third instruction information sent by the network device, where the third instruction information is used to instruct the terminal device to start training the model.

Optionally, the third indication information may be bit information with a value of "0" or "1", used to indicate whether the terminal device enables model training (for example, the third indication information is "0" to indicate disabling, The terminal device does not start model training, and the third indication information is "1" to indicate enablement, and the terminal device starts model training). The third instruction information may also be other information, instructing the terminal device to start model training in an implicit instruction manner.

In some implementations, the terminal device can directly start the training of the channel estimation model, or start the training of the channel estimation model after sending the aforementioned first indication information for more than a preset time. The preset time may be configured by the network device, or may be pre-agreed or stipulated by the protocol.

It should be noted that in the embodiment of the present application, the training of the model can be performed online or offline, and this application does not limit this.

In some implementations, the terminal device can also send instruction information of training data to the network device, and the instruction information of the training data is used to instruct the terminal device to perform model training based on simulation data or actual data. The network device may also configure or instruct the terminal device to use simulated data or actual data for model training explicitly or implicitly (for example, it may implicitly indicate to use actual data for model training by sending an impulse signal, etc.).

Step 405: Send second indication information to the network device. The second indication information is used to indicate that the channel estimation model training is completed.

Step 406: Receive the first DMRS sent by the network device based on the first DMRS pattern.

Step 407: According to the first DMRS, perform channel estimation based on the channel estimation model.

In the embodiment of the present application, steps 405 to 407 can be implemented in any manner in the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

In summary, by sending the first indication information to the network device, the first indication information is used to indicate whether the terminal device has model training capabilities, and the simulation signal received by the terminal device in the simulation channel is obtained. The simulation signal is the network device in The second DMRS sent based on the second DMRS pattern in the simulation channel determines the simulation training data of the channel estimation model based on the simulation signal, uses the simulation training data to train the channel estimation model, and sends the second instruction information to the network device , the second indication information is used to indicate that the channel estimation model training is completed, receive the first DMRS sent by the network device based on the first DMRS pattern, and perform channel estimation based on the channel estimation model according to the first DMRS, so that terminal devices with different capabilities Both can support channel estimation based on artificial intelligence technology, effectively improving the accuracy of channel estimation, thus greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 5. Figure 5 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 5, the method may include the following steps:

Step 501: Send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

In the embodiment of the present application, step 501 can be implemented in any manner among the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

It can be understood that in this embodiment of the present application, the channel estimation model is trained by the network device. The terminal device may not have the model training capability, or the terminal device may have the model training capability, but the network device may perform training based on business conditions, etc. , choose not to perform model training on the terminal device side.

In this embodiment of the present application, if the first indication information is the model training capability indication information of the terminal device, the terminal device may also need to send storage capability information, hardware processing capability information, computing capability information, power consumption capability information, etc. to the network device. , used by the network device to determine the channel estimation model that matches the terminal device. Alternatively, the terminal device can also send model recommendation information to the network device based on its own hardware capabilities, so that the network device determines a channel estimation model that matches the terminal device.

Step 502: Receive fourth indication information sent by the network device, where the fourth indication information is used to indicate the type of training data for the channel estimation model.

In this embodiment of the present application, the terminal device can receive fourth indication information sent by the network device, where the fourth indication information is used to indicate the type of the training data. For example, it may be indicated that the training data is a received signal corresponding to the second DMRS, or it may be indicated that the training data is a channel estimate value at the DMRS estimated based on the second DMRS, or other instructions may be indicated based on the configuration of the channel estimation model. Types of training data, etc. This application does not limit this.

The terminal device can determine what kind of training data the network device needs for model training according to the instructions of the fourth instruction information, and determine the training data according to the received second DMRS and send it to the network device.

Step 503: Receive the second DMRS sent by the network device based on the second DMRS pattern.

In this embodiment of the present application, the terminal device can receive the second DMRS sent by the network device based on the second DMRS pattern, and can determine the training data of the channel estimation model based on the second DMRS. The training data is required by the network device for model training. The data.

The second DMRS pattern may be a legacy DMRS pattern.

Step 504: Determine training data for the channel estimation model according to the second DMRS.

In this embodiment of the present application, the terminal device can determine the training data of the channel estimation model based on the received second DMRS, and the training data is actual data obtained through actual channel transmission.

In this embodiment of the present application, the terminal device can determine the training data based on the type of training data indicated by the received fourth indication information. For example, it is determined based on the fourth indication information that the training data is a received signal corresponding to the second DMRS, or it is determined based on the fourth indication information that the training data is a channel estimate estimated based on the second DMRS, and so on. Among them, the channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated by the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc., can be used. This application does not limit this.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also receive the impulse signal sent by the network device, and can obtain the ideal channel label of the channel based on the impulse signal, The ideal channel label and the training data are sent to the network device together, and the ideal channel label is used for training the channel estimation model.

Optionally, the impulse signal can be transmitted using a semi-static scheduling method. The terminal device can also receive the configuration information of the impulse signal sent by the network device, which can include the transmission period of the impulse signal and the occupied time-frequency resources. etc. The impulse signal can also be transmitted using dynamic scheduling. The terminal device can also receive the scheduling information of the impulse signal sent by the network device, which can include the time-frequency domain resources occupied by the impulse signal transmission and the role indication of the DCI. domain (e.g. dedicated to scheduling impulse signals).

In some embodiments, the terminal device can also receive configuration information or instruction information for reporting training data sent by the network device, which is used to configure or instruct the terminal device to report the training data period, the dimensions of the training data to be reported, and the amount of training data to be reported. , as well as the time-frequency resources used to report training data, etc.

It can be understood that if the network device uses a supervised machine learning method to train the channel estimation model, the terminal device also needs to send the ideal channel label obtained by receiving the impulse signal together with the training data to the network device. For example, a sample contains The {training input value, label} is {received signal at DMRS, data and ideal channel estimate at DMRS} or {channel estimate at DMRS, data and ideal channel estimate at DMRS}.

Step 505: Send the training data to the network device, where the training data is used to train the channel estimation model.

In this embodiment of the present application, the terminal device can send the determined training data to the network device, and the network device uses the training data to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also send an ideal channel label of the channel to the network device, where the ideal channel label is based on the received information from the network device. The impulse signal sent is obtained.

It should be noted that the impulse signal sent by the network device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

In some embodiments, the terminal device can also report training data according to the configuration information or instruction information received from the training data report sent by the network device, according to the period of reporting training data configured or indicated therein, and the dimensions of the reported training data. The number of data, as well as the time-frequency resources used to report training data, etc., are sent to the training data and/or the ideal channel label.

Step 506: Receive the trained channel estimation model sent by the network device.

In this embodiment of the present application, the terminal device can receive the trained channel estimation model sent by the network device.

Optionally, the network device can send the message to the terminal device through RRC signaling, a new wireless signaling bearer (Signaling Radio Bearer, SRB) or a channel identified by a unique logical channel identifier (Logical Channel Identify, LCID). The trained channel estimation model.

In this embodiment of the present application, the network device can use the actual channel data sent by the terminal device as training data to train the channel estimation model. The network device can perform model training by itself, or it can perform model training through an OTT server, OAM, or LMF.

In this embodiment of the present application, the model trained by the network device may be determined based on the model recommendation information reported by the terminal device; it may also be a model determined by the network device itself that matches the capabilities of the terminal device (for example, a network The device is determined based on the storage capacity, hardware processing capability information, computing capability information, power consumption capability information, etc. reported by the terminal device, and is a model that matches the terminal device); it can also be a network device that is not based on the capabilities of the terminal device. definite.

It should be noted that in the embodiment of the present application, the training of the model can be performed online or offline, and this application does not limit this.

Step 507: Receive the first DMRS sent by the network device based on the first DMRS pattern.

Step 508: According to the first DMRS, perform channel estimation based on the channel estimation model.

In the embodiment of the present application, steps 507 to 508 can be implemented in any manner in the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

In summary, by sending the first indication information to the network device, the first indication information is used to indicate whether the terminal device has model training capabilities, and the fourth indication information sent by the network device is received, the fourth indication information is used to indicate channel estimation. The type of training data of the model is to receive the second DMRS sent by the network device based on the second DMRS pattern, determine the training data of the channel estimation model according to the second DMRS, and send the training data to the network device, and the training data is used for the The channel estimation model is trained, the trained channel estimation model is received from the network device, and the first DMRS sent by the network device based on the first DMRS pattern is received. According to the first DMRS, channel estimation is performed based on the channel estimation model to enable different capabilities. All terminal equipment can support channel estimation based on artificial intelligence technology, which effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 6. Figure 6 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 6, the method may include the following steps:

Step 601: Send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

In the embodiment of the present application, step 601 can be implemented in any manner among the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

It can be understood that in this embodiment of the present application, the channel estimation model is trained by the network device. The terminal device may not have the model training capability, or the terminal device may have the model training capability, but the network device may perform training based on business conditions, etc. , choose not to perform model training on the terminal device side.

In this embodiment of the present application, if the first indication information is the model training capability indication information of the terminal device, the terminal device may also need to send storage capability information, hardware processing capability information, computing capability information, power consumption capability information, etc. to the network device. , used by the network device to determine the channel estimation model that matches the terminal device. Alternatively, the terminal device can also send model recommendation information to the network device based on its own hardware capabilities, so that the network device determines a channel estimation model that matches the terminal device.

Step 602: Receive the trained channel estimation model sent by the network device.

In this embodiment of the present application, the terminal device can receive the trained channel estimation model sent by the network device.

Optionally, the network device may send the trained channel estimation model to the terminal device in the form of RRC signaling, new wireless signaling carrying SRB, or a channel identified by a unique LCID.

In this embodiment of the present application, the network device can obtain the simulated signal received by the terminal device in the simulated channel, and the simulated signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulated channel. And can determine the training data of the channel estimation model based on the simulation signal, and use the training data to train the channel estimation model.

The second DMRS pattern may be a legacy DMRS pattern.

In the embodiment of the present application, in the simulated channel model, the network device sends DMRS to the terminal device based on the second DMRS pattern. The terminal device can obtain the simulated signal received in the simulated channel, and determine the simulated data as training data to perform model development. train.

Optionally, in this embodiment of the present application, the network equipment may directly use the received simulated signal of the second DMRS as training data for the channel estimation model, or may obtain the channel estimate estimated based on the simulated signal of the second DMRS. value, the channel estimation value is used as training data for the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc. can be used. This application There is no restriction on this either.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the network device can also obtain the ideal channel label of the simulated channel to train the channel estimation model. It can be understood that in the simulated channel model, the ideal channel label of the simulated channel can be obtained by establishing the channel parameters of the simulated channel model.

In this embodiment of the present application, the network device may perform model training by itself, or may perform model training through an OTT server, OAM, or LMF.

In this embodiment of the present application, the model trained by the network device may be determined based on the model recommendation information reported by the terminal device; it may also be a model determined by the network device itself that matches the capabilities of the terminal device (for example, a network The device is determined based on the storage capacity, hardware processing capability information, computing capability information, power consumption capability information, etc. reported by the terminal device, and is a model that matches the terminal device); it can also be a network device that is not based on the capabilities of the terminal device. definite.

It should be noted that in the embodiment of the present application, the training of the model can be performed online or offline, and this application does not limit this.

Step 603: Receive the first DMRS sent by the network device based on the first DMRS pattern.

Step 604: According to the first DMRS, perform channel estimation based on the channel estimation model.

In the embodiment of the present application, steps 603 to 604 can be implemented in any manner in the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

In summary, by sending the first indication information to the network device, the first indication information is used to indicate whether the terminal device has model training capabilities, receiving the trained channel estimation model sent by the network device, and the receiving network device based on the first DMRS According to the first DMRS sent by the pattern, channel estimation is performed based on the channel estimation model, so that terminal equipment with different capabilities can support channel estimation based on artificial intelligence technology, effectively improving the accuracy of channel estimation, thereby significantly improving The decoding success rate effectively improves the spectrum efficiency of the communication system and saves the pilot overhead of the system.

Please refer to Figure 7, which is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by a network device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 7, the method may include the following steps:

Step 701: Send a first DMRS to the terminal device based on the first DMRS pattern, where the first DMRS is used for channel estimation based on the channel estimation model.

In this embodiment of the present application, the network device can send the first DMRS to the terminal device based on the first DMRS pattern. After receiving the first DMRS, the terminal device can perform channel estimation based on the first DMRS based on the trained channel estimation model.

In some implementations, the network device can receive the first indication information sent by the terminal device, and the network device can determine whether the terminal device has the model training capability based on the first indication information.

Optionally, the first indication information may include at least one of the following: model training capability indication information of the terminal device; hardware processing capability information of the terminal device; computing capability information of the terminal device; power consumption capability of the terminal device. information.

Among them, the model training capability indication information of the terminal device can indicate whether the terminal device has the model training capability. The model training capability indication information may be at least 1 bit.

In some possible implementations, the terminal device can also send model inference capability indication information of the terminal device to the network device. The model inference capability indication information of the terminal device can indicate whether the terminal device has the ability to use a channel estimation model for model inference. ability. The model reasoning capability indication information may also be at least 1 bit.

It can be understood that in various embodiments of the present application, the terminal device has model reasoning capabilities and can perform channel estimation based on a trained channel estimation model.

Optionally, the first indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information UAI; unlimited resource control RRC signaling; media access control layer control element MACCE; Uplink control information UCI. The first indication information may also be sent through the physical uplink shared channel PUSCH.

In this embodiment of the present application, the terminal device can perform channel estimation based on the received first DMRS based on the trained channel estimation model.

It can be understood that in the embodiment of the present application, the terminal device can directly use the received DMRS signal as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the DMRS, and use the channel estimate value as the channel The input of the estimation model is not limited by this application. Estimating the channel based on the DMRS obtains the channel estimate at the DMRS. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application There are no restrictions either.

In the embodiment of this application, the training of the channel estimation model can be performed by the terminal device or the network device; the training can be performed using actual data or simulated data; the training can be performed offline or online. Conduct training.

In some implementations, the network device can send a second DMRS to the terminal device based on the second DMRS pattern, where the second DMRS is used to determine training data for the channel estimation model.

Optionally, the terminal device can use the determined training data to train the channel estimation model.

Optionally, the terminal device can send the training data to the network device, and the network device uses the training data to train the channel estimation model.

Optionally, before the terminal device determines the training data to be sent to the network device, the network device can also send fourth indication information to the terminal device, where the fourth indication information is used to indicate the type of the training data. For example, it may be indicated that the training data is a received signal corresponding to the second DMRS, or it may be indicated that the training data is a channel estimate value estimated based on the second DMRS, and so on. The terminal device can determine what kind of training data the network device needs for model training according to the instructions of the fourth instruction information, and determine the training data according to the received second DMRS and send it to the network device.

In some implementations, the channel estimation model is trained by the terminal device using simulation training data. The terminal device can obtain the simulation signal received by the terminal device in the simulation channel, where the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel, and the terminal device can determine the channel estimation model based on the simulation signal. Simulate training data, and use the simulation training data to train the channel estimation model.

In some implementations, the channel estimation model is trained by the network device using simulation training data. The network device can also obtain the simulated signal received by the terminal device in the simulated channel, where the simulated signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulated channel. The network device can also determine the simulation training data of the channel estimation model based on the simulation signal, and use the simulation training data to train the channel estimation model.

In this embodiment of the present application, when the channel estimation model is trained by a network device, the network device can send the trained channel estimation model to the terminal device.

In some implementations of the embodiments of this application, when the channel estimation model is trained by a terminal device, the network device can also receive second indication information sent by the terminal device, and the second indication information is used to indicate the channel estimation model. Training completed.

Optionally, in addition to model training end indication information, the second indication information may also include at least one of the following information: capability information of the channel estimation model, and processing delay information of the channel estimation model. Alternatively, the model training end indication information is implicitly indicated by at least one of the above two types of information.

Among them, the capability information of the channel estimation model refers to the capability of the channel estimation model compared with traditional channel estimation methods. For example, the model can use lower density DMRS compared with traditional patterns for channel estimation, or It can obtain higher-precision channel estimation results compared with traditional channel estimation methods, etc. The processing delay information of the channel estimation model refers to the processing delay of the terminal device when using the model, which can include the loading time of the model, the time of using the model for inference, etc.

The network device can determine that the channel estimation model has been trained based on the second indication information. At the same time, it can also obtain the model's capability information and/or the model's processing delay information/model complexity/model inference power consumption, etc., and can Reasonably schedule the terminal device based on the capability information and processing delay information; and decide whether to enable the AI model based on the model's processing delay/model complexity/model inference power consumption and other information.

Optionally, the second indication information may be included in at least one of the following signaling: capability reporting signaling (UE capability); user assistance information UAI; unlimited resource control RRC signaling; media access control layer control element MAC CE ;Uplink control information UCI. The second indication information may also be sent through PUSCH.

In some embodiments, when the channel estimation model is trained by the terminal device, the network device can also send third instruction information to the terminal device, where the third instruction information is used to instruct the terminal device to start training the channel estimation model. The third indication information may be at least 1 bit.

In some embodiments, when the channel estimation model is trained by a terminal device, the terminal device may also directly start training the channel estimation model, or may start training the model after sending the first indication information for more than a preset time. train. The preset time may be configured by the network device or agreed or stipulated by the protocol.

In some embodiments, the network device may also send disabling signaling to the terminal device according to business needs and conditions to instruct the terminal device not to start model training.

In some embodiments, if the channel estimation model is trained using a supervised machine learning method, the network device can also send an impulse signal to the terminal device, and the terminal device can obtain the ideal channel estimation label of the channel based on the impulse signal. The ideal channel estimation label is used for the training of this channel estimation model.

In some embodiments, for the case where the channel estimation model is trained by the terminal device, if the channel estimation model is trained using a supervised machine learning method, the network device can also receive auxiliary information sent by the terminal device to request the network device to download Send impulse signal. The terminal device can obtain the ideal channel estimation label of the channel based on the impulse signal, and use the ideal channel estimation label to train the channel estimation model.

In some embodiments, the channel estimation model has the ability to use low-density DMRS for channel estimation, and the density of the first DMRS pattern is lower than the density of the second DMRS pattern. The second DMRS pattern may be a legacy DMRS pattern. Terminal equipment can use lower-density DMRS compared to legacy DMRS patterns, and obtain channel estimation results based on this channel estimation model.

In some embodiments, the channel estimation model has the capability of high-precision channel estimation results, and the density of the first DMRS pattern is the same as the density of the second DMRS pattern. Wherein, the second DMRS pattern can be a legacy DMRS pattern, and the terminal device can use DMRS with the same density as the legacy DMRS pattern, and obtain higher-precision channel estimation results compared with traditional channel estimation methods based on the channel estimation model.

In some implementations, the network device can also send fifth indication information to the terminal device, where the fifth indication information is used to instruct the terminal device to perform channel estimation based on the channel estimation model. Only when the terminal device receives the fifth indication information, the trained channel estimation model will be enabled for channel estimation.

Optionally, the fifth indication information may be at least 1 bit of information, directly instructing the terminal device to enable the trained channel estimation model to perform channel estimation. The fifth indication information may also be the first DMRS pattern configuration sent by the network device to the terminal device to reduce pilot overhead.

In summary, by sending the first DMRS to the terminal device based on the first DMRS pattern, the first DMRS is used for channel estimation based on the channel estimation model, so that terminal devices with different capabilities can support channel estimation based on artificial intelligence technology, effectively improving It improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

In addition, optionally, in various embodiments of the present application, for online model training, model updates can also be performed, and a model update cycle can be defined. Before the model update cycle ends, the terminal device needs to complete a new round of model training and/or model testing, and re-deploy the retrained model. Among them, in some implementations, considering that model loading such as model redeployment may take a certain amount of time, during this time the terminal device can use the original model for reasoning according to its own storage capacity (for example, the terminal device can store at least two data at the same time. model, and the new model will not overwrite the original model), or traditional methods can be used for channel estimation (for example, the new model stored in the terminal device will overwrite the original model). This time can be configured or indicated by the network device or specified by the protocol.

The model update cycle can be configured by the network device; or the terminal device reports the shortest model update cycle supported, and the network device configures a reasonable model update cycle based on the report of the terminal device; or the network device configures the model update cycle based on the update cycle reported by the terminal device. Enable or disable the online model training function (considering that if the model update (re-development) takes too long, the model trained based on the channel data at the beginning of the update may no longer be able to meet the current channel environment, etc., so you can Disable the model training function to avoid meaningless work); or, the protocol stipulates the shortest model training/update cycle, and the terminal device decides whether to perform online model training and update based on its actual situation.

Please refer to Figure 8. Figure 8 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by a network device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 8, the method may include the following steps:

Step 801: Receive the first DMRS sent by the terminal device based on the first DMRS pattern.

In this embodiment of the present application, the network device can receive the first DMRS sent by the terminal device, and the first DMRS is sent by the terminal device based on the first DMRS pattern. After receiving the first DMRS, the network device can perform uplink channel estimation based on the first DMRS based on the trained channel estimation model.

In some embodiments, the network device can receive the second DMRS sent by the terminal device based on the second DMRS pattern, and can determine the training data of the channel estimation model based on the second DMRS, and use the training data to train the channel estimation model. The second DMRS pattern may be a legacy DMRS pattern.

Optionally, in this embodiment of the present application, the network device may directly use the received signal of the second DMRS as training data for the channel estimation model, or may obtain the channel estimate value estimated based on the second DMRS, and convert the channel The estimated value is used as training data for the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimation is performed based on the second DMRS to obtain the channel estimation value. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application does not Make restrictions.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model. The network device can also send instruction information to the terminal device. The instruction information is used to instruct the terminal device to send an impulse signal. The network device receives the terminal device. The impulse signal sent by the device can obtain the ideal channel label of the channel based on the impulse signal and train the channel estimation model.

The transmission period and occupied time-frequency resources of the impulse signal may be configured by the network device, or may be dynamically scheduled and instructed by the network device.

It should be noted that the impulse signal sent by the terminal device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

In some implementations, the network device can obtain the simulated signal received by the network device in the simulated channel, where the simulated signal is the second DMRS sent by the terminal device in the simulated channel based on the second DMRS pattern. The network device can determine the simulation training data of the channel estimation model based on the simulation signal, and use the simulation training data to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In this embodiment of the present application, the channel estimation model can be trained using actual data or simulated data; it can be trained offline or online.

It can be understood that, in this embodiment of the present application, the first DMRS pattern and the second DMRS pattern are both determined by the terminal device based on the configuration and/or instructions of the network device.

Step 802: According to the first DMRS, perform channel estimation based on the channel estimation model.

In this embodiment of the present application, the network device can perform channel estimation based on the received first DMRS based on the trained channel estimation model.

It can be understood that in the embodiment of the present application, the network device can directly use the received DMRS signal as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the DMRS and use the channel estimate value as the channel The input of the estimation model is not limited by this application. The channel estimation value is obtained by estimating the channel based on the DMRS. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application does not limit this. .

In some embodiments, the channel estimation model has the ability to use low-density DMRS for channel estimation, and the density of the first DMRS pattern is lower than the density of the second DMRS pattern. The second DMRS pattern may be a legacy DMRS pattern. Network equipment can use lower-density DMRS compared to legacy DMRS patterns, and obtain uplink channel estimation results based on this channel estimation model. The network device can configure or instruct the terminal device to reduce the density of DMRS patterns based on its own trained model capabilities.

In some embodiments, the channel estimation model has the capability of high-precision channel estimation results, and the density of the first DMRS pattern is the same as the density of the second DMRS pattern. Wherein, the second DMRS pattern can be a legacy DMRS pattern, and the network device can use DMRS with the same density as the legacy DMRS pattern. Based on the channel estimation model, a higher-precision uplink channel estimation result can be obtained compared with the traditional channel estimation method.

In some embodiments, for online model training and model updating, if the network device uses traditional methods for uplink channel estimation when performing model updates, the network device has configured or instructed the terminal device to reduce the density of DMRS patterns. , it may also be necessary to configure or indicate high-density DMRS patterns (such as legacy DMRS patterns) to the terminal equipment.

It can be understood that after the network device completes the training of the channel estimation model, it can also flexibly choose whether to use the channel estimation model for channel estimation based on actual conditions. If the network device uses the channel model for channel estimation, it can configure or indicate a DMRS pattern with reduced density compared to the traditional pattern to the terminal device based on the capabilities of the model, or configure or indicate a legacy DMRS pattern to the terminal device. If the network device does not use this channel model for channel estimation, it can configure or indicate a high-density DMRS pattern (such as a legacy DMRS pattern) to the terminal device.

In summary, by receiving the first DMRS sent by the terminal device based on the first DMRS pattern, and performing channel estimation based on the channel estimation model according to the first DMRS, the network device can perform uplink channel estimation based on artificial intelligence technology, effectively improving the channel The accuracy of the estimation can greatly improve the success rate of decoding, effectively improve the spectrum efficiency of the communication system, and save the pilot overhead of the system.

Please refer to Figure 9. Figure 9 is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in the embodiment of the present application is executed by the network device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 9, the method may include the following steps:

Step 901: Receive the second DMRS sent by the terminal device based on the second DMRS pattern.

In this embodiment of the present application, the network device can receive the second DMRS sent by the terminal device based on the second DMRS pattern, and can determine the training data of the channel estimation model based on the second DMRS, and use the training data to train the channel estimation model. The second DMRS pattern may be a legacy DMRS pattern.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model. The network device can also send instruction information to the terminal device. The instruction information is used to instruct the terminal device to send an impulse signal. The network device receives the terminal device. The impulse signal sent by the device can obtain the ideal channel label of the channel based on the impulse signal and train the channel estimation model.

The transmission period and occupied time-frequency resources of the impulse signal may be configured by the network device, or may be dynamically scheduled and instructed by the network device.

It should be noted that the impulse signal sent by the terminal device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

Step 902: Determine training data for the channel estimation model according to the second DMRS.

In this embodiment of the present application, the network device can determine the training data of the channel estimation model based on the received second DMRS. The training data is actual data obtained through actual channel transmission.

Optionally, in this embodiment of the present application, the network device may directly use the received signal of the second DMRS as training data for the channel estimation model, or may obtain the channel estimate value estimated based on the second DMRS, and convert the channel The estimated value is used as training data for the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimation is performed based on the second DMRS to obtain the channel estimation value. The least square method LS can be used for estimation, the minimum mean square error method MMSE can also be used for estimation, and other estimation algorithms can also be used. This application does not Make restrictions.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model. The network device can also send instruction information to the terminal device. The instruction information is used to instruct the terminal device to send an impulse signal. The network device receives the terminal device. The impulse signal sent by the device.

It should be noted that the impulse signal sent by the terminal device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

Step 903: Use the training data to train the channel estimation model.

In this embodiment of the present application, the network device can use the training data determined in the previous steps to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the network device can also use the ideal channel label of the channel obtained based on the received impulse signal to train the channel estimation model.

Step 904: Receive the first DMRS sent by the terminal device based on the first DMRS pattern.

Step 905: According to the first DMRS, perform channel estimation based on the channel estimation model.

In the embodiment of the present application, steps 904 to 905 can be implemented in any manner in the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

In summary, by receiving the second DMRS sent by the terminal device based on the second DMRS pattern, determining the training data of the channel estimation model based on the second DMRS, and using the training data to train the channel estimation model, the receiving terminal device is based on the first DMRS. According to the first DMRS sent by the DMRS pattern, channel estimation is performed based on the channel estimation model, allowing the network equipment to perform uplink channel estimation based on artificial intelligence technology, effectively improving the accuracy of the channel estimation, thereby greatly improving the decoding efficiency. Success rate, effectively improving the spectrum efficiency of the communication system and saving the pilot overhead of the system.

Please refer to Figure 10, which is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by a network device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 10, the method may include the following steps:

Step 1001: Obtain the simulated signal received by the network device in the simulated channel. The simulated signal is the second DMRS sent by the terminal device in the simulated channel based on the second DMRS pattern.

In this embodiment of the present application, the network device can obtain the simulation signal received by the network device in the simulation channel, and the simulation signal is the second DMRS sent by the terminal device based on the second DMRS pattern in the simulation channel. And can determine the training data of the channel estimation model based on the simulation signal, and use the training data to train the channel estimation model.

The second DMRS pattern may be a legacy DMRS pattern.

In the embodiment of the present application, in the simulated channel model, the terminal device sends DMRS to the network device based on the second DMRS pattern. The network device can obtain the simulated signal received in the simulated channel, and determine the simulated data as training data to perform model development. train.

Step 1002: Determine simulation training data for the channel estimation model based on the simulation signal.

In this embodiment of the present application, the network device can determine the training data of the channel estimation model based on the acquired simulation signal. The training data is simulation data transmitted in the simulation channel.

Optionally, in this embodiment of the present application, the network device may directly use the received simulated signal of the second DMRS as training data for the channel estimation model, or may obtain the channel estimate value estimated based on the simulated signal of the second DMRS. , the channel estimation value is used as the training data of the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc. can be used. This application There is no limit to this either.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the network device can also obtain the ideal channel label of the simulated channel for training of the channel estimation model.

Step 1003: Use the simulation training data to train the channel estimation model.

In this embodiment of the present application, the network device can use the simulation training data determined in the previous steps to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the network device can also obtain the ideal channel label of the simulated channel to train the channel estimation model. It can be understood that in the simulated channel model, the ideal channel label of the simulated channel can be obtained by establishing the channel parameters of the simulated channel model.

Step 1004: Receive the first DMRS sent by the terminal device based on the first DMRS pattern.

Step 1005: According to the first DMRS, perform channel estimation based on the channel estimation model.

In the embodiment of the present application, steps 1004 to 1005 can be implemented in any manner in the embodiments of the present application. The embodiment of the present application does not limit this and will not be described again.

In summary, by obtaining the simulation signal received by the network device in the simulation channel, the simulation signal is the second DMRS sent by the terminal device based on the second DMRS pattern in the simulation channel, and based on the simulation signal, the simulation of the channel estimation model is determined training data, use the simulation training data to train the channel estimation model, receive the first DMRS sent by the terminal device based on the first DMRS pattern, and perform channel estimation based on the channel estimation model according to the first DMRS, so that the network device can perform based on The uplink channel estimation of artificial intelligence technology effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 11, which is a schematic flow chart of a channel estimation method provided by an embodiment of the present application. It should be noted that the channel estimation method in this embodiment of the present application is executed by the terminal device. This method can be executed independently or in conjunction with any other embodiment of the present application. As shown in Figure 11, the method may include the following steps:

Step 1101: Send a first DMRS to the network device based on the first DMRS pattern, where the first DMRS is used for channel estimation based on the channel estimation model.

In this embodiment of the present application, the terminal device can send the first DMRS to the network device based on the first DMRS pattern. After receiving the first DMRS, the network device can perform uplink channel estimation based on the first DMRS based on the trained channel estimation model. The first DMRS pattern is determined by the terminal device based on the configuration and/or instruction of the network device.

In some embodiments, the terminal device can send a second DMRS to the network device based on the second DMRS pattern, and the network device can determine training data of the channel estimation model based on the second DMRS, and use the training data to train the channel estimation model. The second DMRS pattern may be a legacy DMRS pattern. The second DMRS pattern is also determined by the terminal device based on the configuration and/or instructions of the network device.

Optionally, in this embodiment of the present application, the network device may directly use the received signal of the second DMRS as training data for the channel estimation model, or may obtain the channel estimate value estimated based on the second DMRS, and convert the channel The estimated value is used as training data for the channel estimation model, and other training data can also be obtained based on the configuration of the channel estimation model, which is not limited in this application. The channel estimate value at the DMRS obtained by estimating the channel based on the second DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc. can be used. This application There is no limit to this either.

In some embodiments, the network device uses a supervised machine learning method to train the channel estimation model, and the terminal device can also receive instruction information sent by the network device. The instruction information is used to instruct the terminal device to send an impulse signal. The network device receives the impulse signal sent by the terminal device, and can obtain the ideal channel label of the channel based on the impulse signal, and perform training of the channel estimation model.

The transmission period and occupied time-frequency resources of the impulse signal may be configured by the network device, or may be dynamically scheduled and instructed by the network device.

It should be noted that the impulse signal sent by the terminal device may be sent separately from the second DMRS or may be sent together with the second DMRS, which is not limited in this application.

In some implementations, the network device can obtain the simulated signal received by the network device in the simulated channel, where the simulated signal is the second DMRS sent by the terminal device in the simulated channel based on the second DMRS pattern. The network device can determine the simulation training data of the channel estimation model based on the simulation signal, and use the simulation training data to train the channel estimation model.

In this embodiment of the present application, the channel estimation model can be trained using a supervised machine learning method or an unsupervised machine learning method.

It should be noted that the channel estimation model in the embodiment of the present application can be constructed and trained based on any machine learning method, such as a convolutional neural network (CNN), etc. This application is not limited to this.

In this embodiment of the present application, the channel estimation model can be trained using actual data or simulated data; it can be trained offline or online.

In this embodiment of the present application, the network device can perform channel estimation based on the received first DMRS based on the trained channel estimation model.

It can be understood that in the embodiment of the present application, the network device can directly use the received DMRS signal as the input of the channel estimation model, or can obtain the channel estimate value estimated based on the DMRS and use the channel estimate value as the channel The input of the estimation model is not limited by this application. The channel estimate value at the DMRS obtained by estimating the channel based on the DMRS can be estimated using the least square method LS, or the minimum mean square error method MMSE, or other estimation algorithms, etc., this application There are no restrictions either.

In some embodiments, the channel estimation model has the ability to use low-density DMRS for channel estimation, and the density of the first DMRS pattern is lower than the density of the second DMRS pattern. The second DMRS pattern may be a legacy DMRS pattern. Network equipment can use lower-density DMRS compared to legacy DMRS patterns, and obtain uplink channel estimation results based on this channel estimation model. The network device can configure or instruct the terminal device to reduce the density of DMRS patterns based on its own trained model capabilities.

In some embodiments, the channel estimation model has the capability of high-precision channel estimation results, and the density of the first DMRS pattern is the same as the density of the second DMRS pattern. Wherein, the second DMRS pattern can be a legacy DMRS pattern, and the network device can use DMRS with the same density as the legacy DMRS pattern. Based on the channel estimation model, a higher-precision uplink channel estimation result can be obtained compared with the traditional channel estimation method.

In some embodiments, for online model training and model updating, if the network device uses traditional methods for uplink channel estimation when performing model updates, the network device has configured or instructed the terminal device to reduce the density of DMRS patterns. , it may also be necessary to configure or indicate high-density DMRS patterns (such as legacy DMRS patterns) to the terminal equipment. The terminal device sends DMRS according to the configuration or instructions of the network device.

In summary, by sending the first DMRS to the network device based on the first DMRS pattern, the first DMRS is used for channel estimation based on the channel estimation model, so that the network device can perform uplink channel estimation based on artificial intelligence technology, effectively improving channel estimation. accuracy, thus greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Corresponding to the channel estimation methods provided by the above-mentioned embodiments, this application also provides a channel estimation device. Since the channel estimation device provided by the embodiments of this application corresponds to the methods provided by the above-mentioned embodiments, the channel estimation method is The implementation of the method is also applicable to the channel estimation device provided in the following embodiments, and will not be described in detail in the following embodiments.

Please refer to Figure 12, which is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application.

As shown in Figure 12, the channel estimation device 1200 includes: a transceiver unit 1210 and a processing unit 1220, where:

The transceiver unit 1210 is configured to receive the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern;

The processing unit 1220 is configured to perform channel estimation based on the channel estimation model according to the first DMRS.

Optionally, the transceiver unit 1210 is also used for:

First indication information is sent to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

Optionally, the transceiver unit 1210 is also used for:

Receive the second DMRS sent by the network device based on the second DMRS pattern;

According to the second DMRS, training data for the channel estimation model is determined.

Optionally, the processing unit 1220 is also used to:

The training data is used to train the channel estimation model.

Optionally, the transceiver unit 1210 is also used for:

The training data is sent to the network device, and the training data is used to train the channel estimation model.

Optionally, the processing unit 1220 is also used to:

Obtain the simulation signal received by the terminal device in the simulation channel, and the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel;

According to the simulation signal, determine the simulation training data of the channel estimation model;

The simulation training data is used to train the channel estimation model.

Optionally, the transceiver unit 1210 is also used to:

Receive the trained channel estimation model sent by the network device.

Optionally, the transceiver unit 1210 is also used for:

Second indication information is sent to the network device, where the second indication information is used to indicate that the channel estimation model training is completed.

Optionally, the second indication information includes at least one of the following: capability information of the channel estimation model, and processing delay information of the channel estimation model.

Optionally, the transceiver unit 1210 is also used for:

Third instruction information sent by the network device is received, where the third instruction information is used to instruct the terminal device to start training of the channel estimation model.

Optionally, the transceiver unit 1210 is also used for:

Receive fourth indication information sent by the network device, where the fourth indication information is used to indicate the type of the training data.

Optionally, the density of the first DMRS pattern is lower than the density of the second DMRS pattern;

The capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.

Optionally, the density of the first DMRS pattern is the same as the density of the second DMRS pattern;

Among them, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.

Optionally, the transceiver unit 1210 is also used for:

Receive impulse signals sent by the network device;

According to the impulse signal, determine the ideal channel estimation label of the channel, and the ideal channel estimation label is used for training of the channel estimation model;

Among them, the channel estimation model is trained using a supervised machine learning device.

Optionally, the transceiver unit 1210 is also used for:

Send auxiliary information to the network device, the auxiliary information being used to request the impulse signal;

Wherein, the terminal device uses the ideal channel estimation label to train the channel estimation model.

Optionally, the first indication information includes at least one of the following:

Model training capability indication information of the terminal device;

Hardware processing capability information of the terminal device;

Computing capability information of the terminal device;

Power consumption capability information of the terminal device.

Optionally, the transceiver unit 1210 is also used for:

Receive fifth instruction information sent by the network device, where the fifth instruction information is used to instruct the terminal device to perform channel estimation based on the channel estimation model.

The channel estimation device of this embodiment can receive the first DMRS sent by the network equipment based on the first demodulation reference signal DMRS pattern, and perform channel estimation based on the channel estimation model according to the first DMRS, so that terminal equipment with different capabilities can support Channel estimation based on artificial intelligence technology effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 13, which is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application.

As shown in Figure 13, the channel estimation device 1300 includes: a transceiver unit 1310, where:

The transceiver unit 1310 is configured to send the first DMRS to the terminal device based on the first demodulation reference signal DMRS pattern;

The first DMRS is used for channel estimation based on the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Receive first indication information sent by the terminal device, where the first indication information is used to indicate whether the terminal device has model training capabilities.

Optionally, the transceiver unit 1310 is also used to:

Send a second DMRS to the terminal device based on the second DMRS pattern;

The second DMRS is used to determine training data for the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Receive the training data sent by the terminal device;

The training data is used to train the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Obtain the simulation signal received by the terminal device in the simulation channel, and the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel;

According to the simulation signal, determine the simulation training data of the channel estimation model;

The simulation training data is used to train the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Receive the trained channel estimation model sent by the network device.

Optionally, the transceiver unit 1310 is also used to:

Receive second indication information sent by the terminal device, where the second indication information is used to indicate that the channel estimation model training is completed.

Optionally, the second indication information includes at least one of the following: capability information of the channel estimation model, and processing delay information of the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Third instruction information is sent to the terminal device, where the third instruction information is used to instruct the terminal device to start training of the channel estimation model.

Optionally, the transceiver unit 1310 is also used to:

Fourth indication information is sent to the terminal device, where the fourth indication information is used to indicate the type of the training data.

Optionally, the density of the first DMRS pattern is lower than the density of the second DMRS pattern;

The capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.

Optionally, the density of the first DMRS pattern is the same as the density of the second DMRS pattern;

Among them, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.

Optionally, the transceiver unit 1310 is also used to:

Send an impulse signal to the terminal device;

The impulse signal is used to determine the ideal channel estimation label of the channel, and the ideal channel estimation label is used for training of the channel estimation model;

Among them, the channel estimation model is trained using a supervised machine learning device.

Optionally, the transceiver unit 1310 is also used to:

Receive auxiliary information sent by the terminal device, where the auxiliary information is used to request the impulse signal;

Wherein, the terminal device uses the ideal channel estimation label to train the channel estimation model.

Optionally, the first indication information includes at least one of the following:

Model training capability indication information of the terminal device;

Hardware processing capability information of the terminal device;

Computing capability information of the terminal device;

Power consumption capability information of the terminal device.

Optionally, the transceiver unit 1310 is also used to:

Send fifth instruction information to the terminal device, where the fifth instruction information is used to instruct the terminal device to perform channel estimation based on the channel estimation model.

The channel estimation device of this embodiment can send the first DMRS to the terminal device based on the first DMRS pattern. The first DMRS is used for channel estimation based on the channel estimation model, so that terminal devices with different capabilities can support artificial intelligence technology. The channel estimation effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 14, which is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application.

As shown in Figure 14, the channel estimation device 1400 includes: a transceiver unit 1410 and a processing unit 1420, where:

The transceiver unit 1410 is configured to receive the first DMRS sent by the terminal device based on the first demodulation reference signal DMRS pattern;

The processing unit 1420 is configured to perform channel estimation based on the channel estimation model according to the first DMRS.

Optionally, the transceiver unit 1410 is also used to:

Receive the second DMRS sent by the terminal device based on the second DMRS pattern;

According to the second DMRS, determine the training data of the channel estimation model;

The training data is used to train the channel estimation model.

Optionally, the transceiver unit 1410 is also used to:

Send instruction information to the terminal device, where the instruction information is used to instruct the terminal device to send an impulse signal;

Receive the impulse signal sent by the terminal device;

According to the impulse signal, determine the ideal channel estimation label of the channel, and the ideal channel estimation label is used for training of the channel estimation model;

Among them, the channel estimation model is trained using a supervised machine learning device.

Optionally, the processing unit 1420 is also used to:

Obtain the simulation signal received by the network device in the simulation channel, and the simulation signal is the second DMRS sent by the terminal device based on the second DMRS pattern in the simulation channel;

According to the simulation signal, determine the simulation training data of the channel estimation model;

The simulation training data is used to train the channel estimation model.

Optionally, the density of the first DMRS pattern is lower than the density of the second DMRS pattern;

The capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.

Optionally, the density of the first DMRS pattern is the same as the density of the second DMRS pattern;

Among them, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.

The channel estimation device of this embodiment can receive the first DMRS sent by the terminal device based on the first DMRS pattern, and perform channel estimation based on the channel estimation model according to the first DMRS, so that the network device can perform uplink channel estimation based on artificial intelligence technology. Estimation effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

Please refer to Figure 15, which is a schematic structural diagram of a channel estimation device provided by an embodiment of the present application.

As shown in Figure 15, the channel estimation device 1500 includes: a transceiver unit 1510, where:

The transceiver unit 1510 is configured to send the first DMRS to the network device based on the first demodulation reference signal DMRS pattern;

The first DMRS is used for channel estimation based on the channel estimation model.

Optionally, the transceiver unit 1510 is also used for:

Send a second DMRS to the network device based on the second DMRS pattern;

The second DMRS is used to determine training data for the channel estimation model.

Optionally, the transceiver unit 1510 is also used for:

Receive instruction information sent by the network device, where the instruction information is used to instruct the terminal device to send an impulse signal;

Send the impulse signal to the network device;

According to the impulse signal, determine the ideal channel estimation label of the channel, and the ideal channel estimation label is used for training of the channel estimation model;

Among them, the channel estimation model is trained using a supervised machine learning device.

Optionally, the density of the first DMRS pattern is lower than the density of the second DMRS pattern;

The capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.

Optionally, the density of the first DMRS pattern is the same as the density of the second DMRS pattern;

Among them, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.

The channel estimation device of this embodiment can send the first DMRS to the network device based on the first DMRS pattern. The first DMRS is used for channel estimation based on the channel estimation model, so that the network device can perform uplink channel estimation based on artificial intelligence technology. , effectively improves the accuracy of channel estimation, thereby greatly improving the success rate of decoding, effectively improving the spectrum efficiency of the communication system, and saving the pilot overhead of the system.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory, so that the device executes the steps shown in Figure 2 to The method shown in the embodiment of Figure 6.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory, so that the device executes the implementation in Figure 7 The method shown in the example.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory, so that the device executes the steps shown in Figure 8 to The method shown in the embodiment of Figure 10.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and a memory. A computer program is stored in the memory. The processor executes the computer program stored in the memory, so that the device executes the implementation in Figure 11 The method shown in the example.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to The methods shown in the embodiments of Figures 2 to 6 are executed.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to The method shown in the embodiment of Figure 7 is executed.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to The methods shown in the embodiments of Figures 8 to 10 are executed.

In order to implement the above embodiments, embodiments of the present application also provide a communication device, including: a processor and an interface circuit. The interface circuit is used to receive code instructions and transmit them to the processor. The processor is used to run the code instructions to The method shown in the embodiment of Figure 11 is executed.

Please refer to FIG. 16 , which is a schematic structural diagram of another channel estimation apparatus provided by an embodiment of the present disclosure. The channel estimation device 1600 may be a network device, a terminal device, a chip, a chip system, or a processor that supports the network device to implement the above method, or a chip, a chip system, or a processor that supports the terminal device to implement the above method. or processor etc. The device can be used to implement the method described in the above method embodiment. For details, please refer to the description in the above method embodiment.

The channel estimation device 1600 may include one or more processors 1601. The processor 1601 may be a general-purpose processor or a special-purpose processor, or the like. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data. The central processor can be used to control the channel estimation device (such as base station, baseband chip, terminal equipment, terminal equipment chip, DU or CU, etc.) and execute the computer Program, a computer program that processes data.

Optionally, the channel estimation device 1600 may also include one or more memories 1602, on which a computer program 1603 may be stored. The processor 1601 executes the computer program 1603, so that the channel estimation device 1600 performs the steps described in the above method embodiments. method. The computer program 1603 may be solidified in the processor 1601, in which case the processor 1601 may be implemented by hardware.

Optionally, the memory 1602 may also store data. The channel estimation device 1600 and the memory 1602 can be set up separately or integrated together.

Optionally, the channel estimation device 1600 may also include a transceiver 1605 and an antenna 1606. The transceiver 1605 may be called a transceiver unit, a transceiver, a transceiver circuit, etc., and is used to implement transceiver functions. The transceiver 1605 may include a receiver and a transmitter. The receiver may be called a receiver or a receiving circuit, etc., used to implement the receiving function; the transmitter may be called a transmitter, a transmitting circuit, etc., used to implement the transmitting function.

Optionally, the channel estimation device 1600 may also include one or more interface circuits 1607. The interface circuit 1607 is used to receive code instructions and transmit them to the processor 1601 . The processor 1601 executes code instructions to cause the channel estimation device 1600 to perform the method described in the above method embodiment.

In one implementation, the processor 1601 may include a transceiver for implementing receiving and transmitting functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuits, interfaces or interface circuits used to implement the receiving and transmitting functions can be separate or integrated together. The above-mentioned transceiver circuit, interface or interface circuit can be used for reading and writing codes/data, or the above-mentioned transceiver circuit, interface or interface circuit can be used for signal transmission or transfer.

In one implementation, the channel estimation device 1600 may include a circuit, and the circuit may implement the functions of sending, receiving, or communicating in the foregoing method embodiments. The processors and transceivers described in this disclosure may be implemented on integrated circuits (ICs), analog ICs, radio frequency integrated circuits (RFICs), mixed signal ICs, application specific integrated circuits (ASICs), printed circuit boards ( printed circuit board (PCB), electronic equipment, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), n-type metal oxide-semiconductor (NMOS), P-type Metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

The channel estimation device described in the above embodiments may be a network device or a terminal device, but the scope of the channel estimation device described in this disclosure is not limited thereto, and the structure of the channel estimation device may not be limited by Figures 12-15. The channel estimation device may be a stand-alone device or may be part of a larger device. For example, the channel estimation device can be:

(1) Independent integrated circuit IC, or chip, or chip system or subsystem;

(2) A collection of one or more ICs. Optionally, the IC collection may also include storage components for storing data and computer programs;

(3)ASIC, such as modem;

(4) Modules that can be embedded in other devices;

(5) Receivers, terminal equipment, intelligent terminal equipment, cellular phones, wireless equipment, handheld devices, mobile units, vehicle-mounted equipment, network equipment, cloud equipment, artificial intelligence equipment, etc.;

(6) Others, etc.

For the case where the channel estimation device can be a chip or a chip system, please refer to the schematic structural diagram of the chip shown in Figure 17. The chip shown in Figure 17 includes a processor 1701 and an interface 1702. The number of processors 1701 may be one or more, and the number of interfaces 1702 may be multiple.

For the case where the chip is used to implement the functions of the network device in the embodiment of the present disclosure:

Interface 1702 for code instructions and transmission to the processor;

The processor 1701 is configured to run code instructions to perform the methods shown in Figure 2 to Figure 6, or to perform the method shown in Figure 11.

For the case where the chip is used to implement the functions of the terminal device in the embodiment of the present disclosure:

Interface 1702 for code instructions and transmission to the processor;

The processor 1701 is configured to run code instructions to perform the method as shown in Figure 7, or to perform the methods as shown in Figures 8 to 10.

Optionally, the chip also includes a memory 1703, which is used to store necessary computer programs and data.

Those skilled in the art can also understand that the various illustrative logical blocks and steps listed in the embodiments of the present disclosure can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented in hardware or software depends on the specific application and overall system design requirements. Those skilled in the art can use various methods to implement the functions for each specific application, but such implementation should not be understood as exceeding the scope of protection of the embodiments of the present disclosure.

Embodiments of the present disclosure also provide a communication system that includes a channel estimation device as a terminal device and a channel estimation device as a network device in the aforementioned embodiment of FIGS. 8-9 , or the system includes the device in the aforementioned embodiment of FIG. 10 A channel estimation device as a terminal device and a channel estimation device as a network device.

The present disclosure also provides a readable storage medium on which instructions are stored, and when the instructions are executed by a computer, the functions of any of the above method embodiments are implemented.

The present disclosure also provides a computer program product, which, when executed by a computer, implements the functions of any of the above method embodiments.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. A computer program product includes one or more computer programs. When a computer program is loaded and executed on a computer, processes or functions according to embodiments of the present disclosure are generated in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable device. The computer program may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be transmitted from a website, computer, server or data center via a wireline (e.g. Coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means to transmit to another website, computer, server or data center. Computer-readable storage media can be any available media that can be accessed by a computer or a data storage device such as a server, data center, or other integrated media that contains one or more available media. Available media may be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., high-density digital video discs (DVD)), or semiconductor media (e.g., solid state disks (SSD)) )wait.

Those of ordinary skill in the art can understand that the first, second, and other numerical numbers involved in this disclosure are only for convenience of description and are not used to limit the scope of the embodiments of the disclosure, nor to indicate the order.

At least one in the present disclosure can also be described as one or more, and the plurality can be two, three, four or more, and the present disclosure is not limited. In the embodiment of the present disclosure, for a technical feature, the technical feature is distinguished by “first”, “second”, “third”, “A”, “B”, “C” and “D” etc. The technical features described in "first", "second", "third", "A", "B", "C" and "D" are in no particular order or order.

The corresponding relationships shown in each table in this disclosure can be configured or predefined. The values of the information in each table are only examples and can be configured as other values, which is not limited by this disclosure. When configuring the correspondence between information and each parameter, it is not necessarily required to configure all the correspondences shown in each table. For example, in the table in this disclosure, the corresponding relationships shown in some rows may not be configured. For another example, appropriate deformation adjustments can be made based on the above table, such as splitting, merging, etc. The names of the parameters shown in the titles of the above tables may also be other names understandable by the communication device, and the values or expressions of the parameters may also be other values or expressions understandable by the communication device. When implementing the above tables, other data structures can also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables or hash tables. wait.

Predefinition in this disclosure may be understood as definition, pre-definition, storage, pre-storage, pre-negotiation, pre-configuration, solidification, or pre-burning.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each specific application, but such implementations should not be considered to be beyond the scope of this disclosure.

Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be described again here.

It should be understood that the various forms of processes shown above may be used, with steps reordered, added or deleted. For example, each step described in the embodiment of the present disclosure can be executed in parallel, sequentially, or in a different order. As long as the desired results of the technical solution disclosed in the present invention can be achieved, there is no limitation here.

The above-mentioned specific embodiments do not constitute a limitation on the scope of the present invention. It will be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions are possible depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims (50)

1. A channel estimation method, characterized in that the method is executed by a terminal device, and the method includes:

   Receive the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern;

   According to the first DMRS, channel estimation is performed based on a channel estimation model.
2. The method of claim 1, further comprising:

   Send first indication information to the network device, where the first indication information is used to indicate whether the terminal device has model training capabilities.
3. The method of claim 2, further comprising:

   Receive the second DMRS sent by the network device based on the second DMRS pattern;

   According to the second DMRS, training data for the channel estimation model is determined.
4. The method of claim 3, further comprising:

   The channel estimation model is trained using the training data.
5. The method of claim 3, further comprising:

   Send the training data to the network device.
6. The method of claim 2, further comprising:

   Obtain the simulation signal received by the terminal device in the simulation channel, where the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel;

   Determine simulation training data for the channel estimation model according to the simulation signal;

   The channel estimation model is trained using the simulation training data.
7. The method according to claim 2 or 5, characterized in that, the method further includes:

   Receive the trained channel estimation model sent by the network device.
8. The method according to claim 4 or 6, characterized in that, the method further includes:

   Send second indication information to the network device, where the second indication information is used to indicate that training of the channel estimation model is completed.
9. The method according to claim 8, wherein the second indication information includes at least one of the following: capability information of the channel estimation model, and processing delay information of the channel estimation model.
10. The method according to claim 4 or 6, characterized in that, the method further includes:

    Receive third instruction information sent by the network device, where the third instruction information is used to instruct the terminal device to start training of the channel estimation model.
11. The method of claim 5, further comprising:

    Receive fourth indication information sent by the network device, where the fourth indication information is used to indicate the type of the training data.
12. The method according to any one of claims 4-7, characterized in that,

    The density of the first DMRS pattern is lower than the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.
13. The method according to any one of claims 4-7, characterized in that,

    The density of the first DMRS pattern is the same as the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.
14. The method according to claim 4 or 5, characterized in that the method further includes:

    Receive impulse signals sent by the network device;

    Determine an ideal channel estimation label of the channel according to the impulse signal, and the ideal channel estimation label is used for training of the channel estimation model;

    Wherein, the channel estimation model is trained using a supervised machine learning method.
15. The method of claim 14, further comprising:

    Send auxiliary information to the network device, where the auxiliary information is used to request the impulse signal;

    Wherein, the terminal device uses the ideal channel estimation label to train the channel estimation model.
16. The method according to claim 2, characterized in that the first indication information includes at least one of the following:

    Model training capability indication information of the terminal device;

    Hardware processing capability information of the terminal device;

    Computing capability information of the terminal device;

    Power consumption capability information of the terminal device.
17. The method according to any one of claims 1-16, characterized in that the method further includes:

    Receive fifth instruction information sent by the network device, where the fifth instruction information is used to instruct the terminal device to perform channel estimation based on the channel estimation model.
18. A channel estimation method, characterized in that the method is executed by a network device, and the method includes:

    Send the first DMRS to the terminal device based on the first demodulation reference signal DMRS pattern;

    The first DMRS is used for channel estimation based on a channel estimation model.
19. The method of claim 18, further comprising:

    Receive first indication information sent by the terminal device, where the first indication information is used to indicate whether the terminal device has model training capabilities.
20. The method of claim 19, further comprising:

    Send a second DMRS to the terminal device based on the second DMRS pattern;

    The second DMRS is used to determine training data of the channel estimation model.
21. The method of claim 20, further comprising:

    Receive the training data sent by the terminal device;

    The channel estimation model is trained using the training data.
22. The method of claim 19, further comprising:

    Obtain the simulation signal received by the terminal device in the simulation channel, where the simulation signal is the second DMRS sent by the network device based on the second DMRS pattern in the simulation channel;

    Determine simulation training data for the channel estimation model according to the simulation signal;

    The channel estimation model is trained using the simulation training data.
23. The method according to claim 21 or 22, characterized in that, the method further includes:

    Receive the trained channel estimation model sent by the network device.
24. The method according to claim 19 or 20, characterized in that the method further includes:

    Receive second indication information sent by the terminal device, where the second indication information is used to indicate completion of training of the channel estimation model.
25. The method of claim 24, wherein the second indication information includes at least one of the following: capability information of the channel estimation model, and processing delay information of the channel estimation model.
26. The method according to claim 19 or 20, characterized in that the method further includes:

    Send third instruction information to the terminal device, where the third instruction information is used to instruct the terminal device to start training of the channel estimation model.
27. The method according to claim 21, characterized in that, the method further includes:

    Send fourth indication information to the terminal device, where the fourth indication information is used to indicate the type of the training data.
28. The method according to any one of claims 19-22, characterized in that,

    The density of the first DMRS pattern is lower than the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.
29. The method according to any one of claims 19-22, characterized in that,

    The density of the first DMRS pattern is the same as the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.
30. The method according to claim 20 or 21, characterized in that, the method further includes:

    Send an impulse signal to the terminal device;

    The impulse signal is used to determine an ideal channel estimation label of the channel, and the ideal channel estimation label is used for training of the channel estimation model;

    Wherein, the channel estimation model is trained using a supervised machine learning method.
31. The method of claim 30, further comprising:

    Receive auxiliary information sent by the terminal device, where the auxiliary information is used to request the impulse signal;

    Wherein, the terminal device uses the ideal channel estimation label to train the channel estimation model.
32. The method according to claim 19, characterized in that the first indication information includes at least one of the following:

    Model training capability indication information of the terminal device;

    Hardware processing capability information of the terminal device;

    Computing capability information of the terminal device;

    Power consumption capability information of the terminal device.
33. The method according to any one of claims 18-32, characterized in that the method further includes:

    Send fifth instruction information to the terminal device, where the fifth instruction information is used to instruct the terminal device to perform channel estimation based on the channel estimation model.
34. A channel estimation method, characterized in that the method is executed by a network device, and the method includes:

    receiving the first DMRS sent by the terminal device based on the first demodulation reference signal DMRS pattern;

    According to the first DMRS, channel estimation is performed based on a channel estimation model.
35. The method of claim 34, further comprising:

    Receive the second DMRS sent by the terminal device based on the second DMRS pattern;

    Determine training data for the channel estimation model according to the second DMRS;

    The channel estimation model is trained using the training data.
36. The method of claim 35, further comprising:

    Send instruction information to the terminal device, where the instruction information is used to instruct the terminal device to send an impulse signal;

    Receive the impulse signal sent by the terminal device;

    Determine an ideal channel estimation label of the channel according to the impulse signal, and the ideal channel estimation label is used for training of the channel estimation model;

    Wherein, the channel estimation model is trained using a supervised machine learning method.
37. The method of claim 34, further comprising:

    Obtain the simulation signal received by the network device in the simulation channel, where the simulation signal is the second DMRS sent by the terminal device based on the second DMRS pattern in the simulation channel;

    Determine simulation training data for the channel estimation model according to the simulation signal;

    The channel estimation model is trained using the simulation training data.
38. The method according to any one of claims 35-37, characterized in that,

    The density of the first DMRS pattern is lower than the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.
39. The method according to any one of claims 35-37, characterized in that,

    The density of the first DMRS pattern is the same as the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.
40. A channel estimation method, characterized in that the method is executed by a terminal device, and the method includes:

    Send the first DMRS to the network device based on the first demodulation reference signal DMRS pattern;

    The first DMRS is used for channel estimation based on a channel estimation model.
41. The method of claim 40, further comprising:

    Send a second DMRS to the network device based on the second DMRS pattern;

    The second DMRS is used to determine training data of the channel estimation model.
42. The method of claim 41, further comprising:

    Receive instruction information sent by the network device, where the instruction information is used to instruct the terminal device to send an impulse signal;

    Send the impulse signal to the network device;

    Determine an ideal channel estimation label of the channel according to the impulse signal, and the ideal channel estimation label is used for training of the channel estimation model;

    Wherein, the channel estimation model is trained using a supervised machine learning method.
43. The method according to claim 41 or 42, characterized in that,

    The density of the first DMRS pattern is lower than the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the ability to use low-density DMRS for channel estimation.
44. The method according to claim 41 or 42, characterized in that,

    The density of the first DMRS pattern is the same as the density of the second DMRS pattern;

    Wherein, the capability information of the channel estimation model is the capability of having high-precision channel estimation results.
45. A channel estimation device, characterized in that the device includes:

    A transceiver unit configured to receive the first DMRS sent by the network device based on the first demodulation reference signal DMRS pattern;

    A processing unit configured to perform channel estimation based on the channel estimation model according to the first DMRS.
46. A channel estimation device, characterized in that the device includes:

    A transceiver unit, configured to send the first DMRS to the terminal device based on the first demodulation reference signal DMRS pattern;

    The first DMRS is used for channel estimation based on a channel estimation model.
47. A channel estimation device, characterized in that the device includes:

    A transceiver unit configured to receive the first DMRS sent based on the first demodulation reference signal DMRS pattern sent by the terminal device;

    A processing unit configured to perform channel estimation based on a channel estimation model according to the first DMRS.
48. A channel estimation device, characterized in that the device includes:

    A transceiver unit configured to send the first DMRS to the network device based on the first demodulation reference signal DMRS pattern;

    The first DMRS is used for channel estimation based on a channel estimation model.
49. A communication device, characterized in that the device includes a processor and a memory, a computer program is stored in the memory, and the processor executes the computer program stored in the memory, so that the device executes the claims The method described in any one of 1 to 44.
50. A computer-readable storage medium for storing instructions, which when executed, enables the method according to any one of claims 1 to 44 to be implemented.

Images