Method and apparatus for feedback channel state

Abstract

The embodiment of the present application provides a method and device for feeding back channel state, the method comprises the following steps: a terminal device receives first indication information from a network device; the terminal device reduces a first value of first information to a second value according to the first indication information, and the second value is used for determining channel state feedback information, wherein the first information comprises channel quality indication (CQI) information or signal to interference noise ratio (SINR) information. The accuracy of the channel state feedback information determined according to the second value is higher than that of the channel state feedback information determined according to the first value, so that the accuracy of the channel state feedback information can be improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of communications, and more particularly, to a method and apparatus for feeding back channel state. BACKGROUND

[0002] In a communication system, a network device (e.g., a base station) needs to obtain downlink channel information fed back by a terminal device, so as to perform precoding and other processing on data in the downlink direction. In the current feedback mechanism, the terminal device compresses channel information in the spatial domain or the frequency domain, for example, projects onto the discrete Fourier transform (DFT) basis in the spatial domain or the frequency domain, and describes the channel information in the original spatial domain and the original frequency domain based on the projection result.

[0003] However, this feedback manner adopts a fixed projection coordinate system, and the accuracy of information is lost to some extent, and with the increase of the spatial domain dimension, the frequency domain dimension, and the measurement accuracy requirement, the amount of feedback information required also increases.

[0004] Therefore, there is an urgent need for a method for feeding back channel state to improve the accuracy of feedback information. SUMMARY

[0005] Embodiments of the present application provide a method and apparatus for feeding back channel state, which can improve the accuracy of feedback information.

[0006] In a first aspect, a method for feeding back channel state is provided, which can be executed by a terminal device, or can also be executed by a chip or circuit arranged in the terminal device, which is not limited in the present application. For the convenience of description, the following will be described by taking the execution by a network device as an example.

[0007] The method for feeding back channel state comprises:

[0008] The terminal device receives first indication information from the network device; and the terminal device reduces a first value of first information to a second value according to the first indication information, the second value being used to determine channel state feedback information, wherein the first information comprises channel quality indicator (CQI) information or signal to interference and noise ratio (SINR) information.

[0009] According to the method for feeding back channel state provided in the embodiments of the present application, the terminal device reduces the first value of the CQI information or the SINR information to the second value according to the first indication information, and the channel state feedback information determined according to the second value is more accurate than the channel state feedback information determined according to the first value, thereby improving the accuracy of the channel state feedback information.

[0010] In some implementations of the first aspect, the method further includes: receiving, by the terminal device, a second probability value from the network device, the second probability value being used to determine whether to reduce the first value of the first information to the second value; and determining, by the terminal device, whether to reduce the first value of the first information to the second value according to the first indication information, including: determining, by the terminal device, the first probability value according to the first indication information; and determining, by the terminal device, whether to reduce the first value of the first information to the second value according to the first probability value and the second probability value.

[0011] In some implementations of the first aspect, the first indication information is a fallback indication information configured and delivered by the network device, the fallback indication information being used to instruct the terminal device to calculate the second value from the first value in the first information.

[0012] In some implementations of the first aspect, the terminal device determines whether to reduce the first value of the first information to the second value according to the first probability value and the second probability value, thereby avoiding the terminal device directly reducing the first value and improving the accuracy of the scheme.

[0013] In some implementations of the first aspect, the terminal device determines whether to reduce the first value of the first information to the second value according to the first probability value and the second probability value, including: the terminal device determines whether to reduce the first value of the first information to the second value according to the first probability value, the second probability value, and a first threshold value.

[0014] For example, in this case, the terminal device obtains multiple fallback values from the first indication information, and obtains multiple first probability values according to the multiple fallback values. The terminal device compares the second probability value sent by the network device to the terminal device with the first probability values respectively, to obtain a number N of the first probability values that are greater than or equal to the second probability value, and determines to perform fallback when the number N is greater than a first threshold value, determines not to perform fallback when the number N is less than the first threshold value, and determines to perform fallback or not to perform fallback when the number N is equal to the first threshold value.

[0015] Optionally, the first threshold value is predefined, or the first threshold value is sent by the network device to the terminal device, or the first threshold value is determined by negotiation between the terminal device and the network device.

[0016] In some implementations of the first aspect, the first indication information includes at least one of the following information: a fallback value, a cosine similarity of an AI model, and a normalized mean square error of the AI model, wherein the fallback value represents a difference between the first value and the second value, the AI model is used to determine the channel state feedback information, and the cosine similarity of the AI model and the normalized mean square error of the AI model are used to calculate the fallback value.

[0017] Based on the above scheme, there are many possible forms of the first indication information in the present application, which improves the flexibility of the scheme.

[0018] In some implementations of the first aspect, the square of the cosine similarity of the AI model is used to calculate the fallback value.

[0019] For example, in this case, the terminal device obtains the first indication information, which includes the cosine similarity of the AI model. The terminal device calculates the fallback value by calculating the square of the cosine similarity of the AI model. The terminal device reduces the square of the cosine similarity of the AI model to obtain a second value from the first value in the first information, and generates channel state feedback information based on the second value. The terminal device calculates the fallback value by using the square of the cosine similarity of the AI model, which simplifies the calculation process compared to other algorithms, reduces the complexity of the terminal device, and ensures the accuracy of the fallback value.

[0020] In some implementations of the first aspect, the first information is the CQI information, the first value is a first CQI value, and the second value is a second CQI value. The terminal device reduces the first value of the first information to the second value according to the first indication information, including: the terminal device determines the first CQI value according to the first SINR value, wherein the first SINR value is determined according to a channel feature vector, and the channel feature vector is used to represent information of a downlink channel; and the terminal device reduces the first CQI value of the CQI information to the second CQI value according to the first indication information.

[0021] In some implementations of the first aspect, the first information is the SINR information, the first value is a first SINR value, and the second value is a second SINR value. The terminal device reduces the first value of the first information to the second value according to the first indication information, including: the terminal device reduces the first SINR value of the SINR information to the second SINR value according to the first indication information; and after the terminal device reduces the first value of the first information to the second value according to the first indication information, the method further includes: the terminal device determines a second CQI value according to the second SINR value.

[0022] With reference to the first aspect, in some implementations of the first aspect, the first information is the SINR information, the first value is a first SINR value, the second value is a second SINR value, and the terminal device reduces the first value of the first information to the second value according to the first indication information includes that the terminal device reduces the first SINR value of the SINR information to the second SINR value according to the first indication information; after the terminal device reduces the first value of the first information to the second value according to the first indication information, the method further includes that the terminal device quantizes the second SINR value to generate a third SINR value according to a second indication information, the second indication information is from the network device, and the second indication information is used to indicate a quantization manner of the SINR information; and the terminal device determines the channel state feedback information according to the third SINR value.

[0023] The terminal device reduces the third SINR value directly, avoids the problem of information loss in the process of mapping the SINR value to the CQI value, and can enable the network device to obtain more accurate channel interference, thereby improving the accuracy of the resource configuration information issued by the network device for the terminal device. The resource configuration information can be information of a modulation and coding manner of the terminal device for downlink channel data transmission, can be time-frequency resource allocation information of the terminal device, or can be precoding information of the terminal device, and the type of the resource configuration information is not limited in the present application.

[0024] With reference to the first aspect, in some implementations of the first aspect, the second indication information includes at least one of the following: a quantization bit number, a quantization range, or a quantization step.

[0025] With reference to the first aspect, in some implementations of the first aspect, the method further includes that the terminal device receives a reference signal CSI-RS from the network device; and the terminal device determines the channel feature vector according to the CSI-RS.

[0026] With reference to the first aspect, in some implementations of the first aspect, after the terminal device reduces the first value of the first information to the second value according to the first indication information, the method further includes that the terminal device determines the channel state feedback information according to the second value; and the terminal device sends the channel state feedback information to the network device.

[0027] With reference to the first aspect, in some implementations of the first aspect, the first indication information is carried in radio resource control (RRC) configuration information.

[0028] Optionally, the first indication information can also be carried in high-layer signaling configuration information (for example, non-access stratum (NAS) signaling information) or configured through an application layer protocol.

[0029] In a second aspect, a method for feeding back channel state is provided, which can be executed by a network device or a chip or circuit arranged in the network device, and the present application does not limit this. For ease of description, the following describes an example of execution by the network device.

[0030] The method for feeding back channel state comprises:

[0031] The network device determines first indication information, which is used to instruct a terminal device to reduce a first value of first information to a second value; and the network device sends the first indication information to the terminal device, wherein the second value is used to determine channel state feedback information, and the first information comprises channel quality indicator (CQI) information or signal to interference noise ratio (SINR) information.

[0032] According to the method for feeding back channel state provided in the embodiments of the present application, the network device determines and sends first indication information to the terminal device, the first indication information instructs the terminal device to reduce a first value of CQI information or SINR information to a second value, and the second value is used to determine channel state feedback information with higher accuracy than channel state feedback information determined by the first value, thereby improving the accuracy of the channel state feedback information.

[0033] In combination with the second aspect, in some implementations of the second aspect, the method further comprises: the network device sends a second probability value to the terminal device, and the second probability value is used to determine whether to reduce the first value of the first information to the second value.

[0034] In the present application, the first probability value and the second probability value are used to determine whether to reduce the first value of the CQI information or the SINR information to the second value, which avoids the terminal device directly reducing the first value, thereby improving the accuracy of the scheme.

[0035] In combination with the second aspect, in some implementations of the second aspect, the first indication information further comprises at least one of the following information: a fallback value, a cosine similarity of an AI model, and a normalized mean square error of the AI model, wherein the fallback value represents a difference between the first value and the second value, the AI model is used to determine the channel state feedback information, and the cosine similarity of the AI model and the normalized mean square error of the AI model are used to calculate the fallback value.

[0036] Based on the above scheme, there are various possible forms of the first indication information in the present application, thereby improving the flexibility of the scheme.

[0037] In combination with the second aspect, in some implementations of the second aspect, a square of the cosine similarity of the AI model is used to calculate the fallback value.

[0038] With reference to the second aspect, in some implementations of the second aspect, the first information is the CQI information, the first value is a first CQI value, the second value is a second CQI value, the first CQI value is obtained according to a first SINR value, and the second CQI value is used to determine the channel state feedback information, wherein the first SINR value is determined according to a channel eigenvector, and the channel eigenvector is used to represent information of a downlink channel.

[0039] With reference to the second aspect, in some implementations of the second aspect, the first information is the SINR information, the first value is a first SINR value, the second value is a second SINR value, and the first indication information is used to instruct the terminal device to reduce the first value of the first information to the second value, including: the first indication information is used to instruct to reduce the first SINR value of the SINR information to the second SINR value, and the second CQI value is used to determine the channel state feedback information.

[0040] With reference to the second aspect, in some implementations of the second aspect, the first information is the SINR information, the first value is a first SINR value, the second value is a second SINR value, and the first indication information is used to instruct the terminal device to reduce the first value of the first information to the second value, including: the first indication information is used to instruct to reduce the first SINR value of the SINR information to the second SINR value, and the method further includes: the network device sends second indication information to the terminal device, and the second indication information is used to indicate a quantization manner of the SINR information.

[0041] It should be understood that the network device indicates the quantization manner of the SINR information by sending the second indication information, and the terminal device directly obtains a third SINR value by quantizing the second SINR value and sends the third SINR value to the network device. The network device receives the third SINR value, avoids the problem of information loss in the process of mapping the SINR value to the CQI value, and can make the network device obtain more accurate channel interference, and improve the accuracy of the resource configuration information issued by the network device for the terminal device.

[0042] With reference to the second aspect, in some implementations of the second aspect, the second indication information includes at least one of the following: a quantization bit number, a quantization range, or a quantization step.

[0043] With reference to the second aspect, in some implementations of the second aspect, the method further includes: the network device sends a reference signal CSI-RS to the terminal device, and the CSI-RS is used to determine a channel eigenvector.

[0044] With reference to the second aspect, in some implementations of the second aspect, the first indication information is carried in RRC configuration information.

[0045] In a third aspect, a device for feeding back channel state is provided, which comprises: a transceiver configured to receive first indication information from a network device; and a processor configured to reduce a first value of first information to a second value according to the first indication information, the second value being used to determine channel state feedback information, wherein the first information comprises channel quality indicator (CQI) information or signal to interference noise ratio (SINR) information.

[0046] With reference to the third aspect, in some implementations of the third aspect, the transceiver is further configured to receive a second probability value from the network device, the second probability value being used to determine whether to reduce the first value of the first information to the second value; and the processor is further configured to determine, according to the first indication information, whether to reduce the first value of the first information to the second value, including: determining, according to the first indication information, a first probability value, and determining, according to the first probability value and the second probability value, whether to reduce the first value of the first information to the second value.

[0047] With reference to the second aspect, in some implementations of the second aspect, the transceiver is configured to receive first indication information from the network device, the first indication information comprising a first threshold value, the first threshold value being used to determine whether to reduce the first value of the first information to the second value.

[0048] With reference to the third aspect, in some implementations of the third aspect, the first indication information further comprises at least one of the following information: a fallback value, a cosine similarity of an AI model, and a normalized mean square error of the AI model, wherein the fallback value represents a difference between the first value and the second value, the AI model is used to determine the channel state feedback information, and the cosine similarity of the AI model and the normalized mean square error of the AI model are used to calculate the fallback value.

[0049] With reference to the third aspect, in some implementations of the third aspect, a square of the cosine similarity of the AI model is used to calculate the fallback value.

[0050] With reference to the third aspect, in some implementations of the third aspect, when the first information is the CQI information, the first value is a first CQI value, and the second value is a second CQI value, the processor is further configured to reduce the first value of the first information to the second value according to the first indication information, including: determining, according to a first SINR value, the first CQI value, wherein the first SINR value is determined according to a channel eigenvector, and the channel eigenvector is used to represent information of a downlink channel; and reducing, according to the first indication information, the first CQI value of the CQI information to the second CQI value.

[0051] In some manners of implementation of the third aspect, in combination with the third aspect, when the first information is the SINR information, the first value is a first SINR value, and the second value is a second SINR value, the processing unit is further configured to reduce the first value of the first information to the second value according to the first indication information, including: the processing unit is configured to reduce the first SINR value of the SINR information to the second SINR value according to the first indication information; and after the processing unit reduces the first value of the first information to the second value according to the first indication information, the method further includes: the processing unit is configured to determine the second CQI value according to the second SINR value.

[0052] In some manners of implementation of the third aspect, in combination with the third aspect, when the first information is the SINR information, the first value is a first SINR value, and the second value is a second SINR value, the processing unit is further configured to reduce the first value of the first information to the second value according to the first indication information, including: the processing unit is configured to reduce the first SINR value of the SINR information to the second SINR value according to the first indication information; and after the processing unit reduces the first value of the first information to the second value according to the first indication information, the method further includes: the processing unit is configured to generate a third SINR value by quantizing the second SINR value according to second indication information, the second indication information is from the network device, and the second indication information is used to indicate a quantization manner of the SINR information; and the processing unit is configured to determine the channel state feedback information according to the third SINR value.

[0053] In some manners of implementation of the third aspect, in combination with the third aspect, the second indication information includes at least one of a quantization bit number, a quantization range, or a quantization step.

[0054] In some manners of implementation of the third aspect, in combination with the third aspect, the transceiver is further configured to receive, by the processing unit, a reference signal CSI-RS from the network device.

[0055] In some manners of implementation of the third aspect, in combination with the third aspect, the processing unit is further configured to determine the channel state feedback information according to the second value; and the transceiver is further configured to send, by the processing unit, the channel state feedback information to the network device.

[0056] In some manners of implementation of the third aspect, in combination with the third aspect, the first indication information is carried in RRC configuration information.

[0057] A fourth aspect provides an apparatus for feeding back a channel state, including: a processing unit configured to determine first indication information, the first indication information being used to reduce a first value of first information to a second value; and a transceiver configured to send the first indication information to a terminal device.

[0058] The second value is used to determine the channel state feedback information, and the first information includes channel quality indicator (CQI) information or signal to interference noise ratio (SINR) information.

[0059] With reference to the fourth aspect, in some implementations of the fourth aspect, the processing unit is configured to determine the first indication information, the first indication information including a first threshold value, the first threshold value being used to determine whether to reduce the first value of the first information to the second value.

[0060] With reference to the fourth aspect, in some implementations of the fourth aspect, the first indication information further includes at least one of the following: a fallback value, a cosine similarity of the AI model, and a normalized mean square error of the AI model, wherein the fallback value represents a difference between the first value and the second value, the AI model being used to determine the channel state feedback information, and the cosine similarity of the AI model and the normalized mean square error of the AI model being used to calculate the fallback value.

[0061] With reference to the fourth aspect, in some implementations of the fourth aspect, a square of the cosine similarity of the AI model is used to calculate the fallback value.

[0062] With reference to the fourth aspect, in some implementations of the fourth aspect, the first information is CQI information, the first value is a first CQI value, and the second value is a second CQI value, the first CQI value being obtained according to a first SINR value, the second CQI value being used to determine the channel state feedback information, wherein the first SINR value is determined according to a channel feature vector, and the channel feature vector is used to represent information of a downlink channel.

[0063] With reference to the fourth aspect, in some implementations of the fourth aspect, the first information is SINR information, the first value is a first SINR value, and the second value is a second SINR value, the first indication information being used to instruct the terminal device to reduce the first value of the first information to the second value, including: the first indication information being used to instruct to reduce the first SINR value of the SINR information to the second SINR value, the second SINR value being used to determine a second CQI value, and the second CQI value being used to determine the channel state feedback information.

[0064] With reference to the fourth aspect, in some implementations of the fourth aspect, the first information is SINR information, the first value is a first SINR value, and the second value is a second SINR value, the first indication information being used to instruct the terminal device to reduce the first value of the first information to the second value, including: the first indication information being used to instruct to reduce the first SINR value of the SINR information to the second SINR value, and the transceiver is further configured to: send, to the terminal device, second indication information, the second indication information being used to instruct a quantization manner of the SINR information.

[0065] In a possible implementation mode of the fourth aspect, the second indication information includes at least one of the following: a quantization bit number, a quantization range, or a quantization step.

[0066] In a possible implementation mode of the fourth aspect, the transceiver is further configured to send, to the terminal device, a reference signal (CSI-RS) used to determine a channel eigenvector used to represent information of a downlink channel.

[0067] The fifth aspect provides a communication apparatus, including: a processor coupled with a memory, the memory being configured to store programs or instructions, when the programs or instructions are executed by the processor, the communication apparatus is caused to execute the method in the first or second aspect and various possible implementation modes thereof.

[0068] Optionally, the processor is one or more, and the memory is one or more.

[0069] Optionally, the memory can be integrated with the processor, or the memory and the processor are separate devices.

[0070] Optionally, the forwarding device further includes a transmitter and a receiver.

[0071] The sixth aspect provides a communication system, the terminal device and the network device.

[0072] The seventh aspect provides a computer readable medium, the computer readable medium stores a computer program (also referred to as code or instructions), when the computer program is executed, the computer is caused to execute the method in any possible implementation mode of the first aspect or the second aspect.

[0073] The eighth aspect provides a chip system, including a memory configured to store a computer program and a processor configured to call and execute the computer program from the memory, so that the communication device installed with the chip system executes the method in any aspect of the first aspect or the second aspect and possible implementation modes thereof.

[0074] The chip system can include an input chip or interface for sending information or data, and an output chip or interface for receiving information or data. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 is a schematic diagram of a communication system to which the present application is applicable.

[0076] Figure 2is a method for feeding back a channel state provided by an embodiment of the present application.

[0077] Figure 3 is a CSI feedback process based on AI provided by an embodiment of the present application.

[0078] Figure 4 is another method for feeding back a channel state provided by an embodiment of the present application.

[0079] Figure 5 is still another method for feeding back a channel state provided by an embodiment of the present application.

[0080] Figure 6 is still another method for feeding back a channel state provided by an embodiment of the present application.

[0081] Figure 7 is still another method for feeding back a channel state provided by an embodiment of the present application.

[0082] Figure 8 is still another method for feeding back a channel state provided by an embodiment of the present application.

[0083] Figure 9 is a device 900 for feeding back a channel state provided by an embodiment of the present application.

[0084] Figure 10 is a device 1000 for feeding back a channel state provided by the present application. DETAILED DESCRIPTION

[0085] The technical solutions in the embodiments of the present application will be described below with reference to the drawings.

[0086] The technical solutions of this application embodiment can be applied to various communication systems, such as: Global System of Mobile communication (GSM) system, Code Division Multiple Access (CDMA) system, Wideband Code Division Multiple Access (WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, LTE Frequency Division Duplex (FDD) system, LTE Time Division Duplex (TDD) system, Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) system, 5th Generation (5G) system, or New Radio (NR), etc.

[0087] To facilitate understanding of the embodiments of this application, firstly... Figure 1 Taking the communication system shown as an example, the communication system applicable to the embodiments of this application will be described in detail. Figure 1 This is a schematic diagram of a communication system 100 for a channel state feedback method applicable to embodiments of this application. (See diagram below.) Figure 1 As shown, the communication system 100 may include at least one terminal device, such as Figure 1 The terminal device 120 is shown. Network device 110 and terminal device 120 can communicate via a wireless link. Each communication device, such as network device 110 or terminal device 120, can be configured with multiple antennas. For each communication device in the communication system 100, the configured multiple antennas may include at least one transmitting antenna for transmitting signals and at least one receiving antenna for receiving signals. Therefore, the communication devices in the communication system 100, such as network device 110 and terminal device 120, can communicate via multi-antenna technology.

[0088] It should be understood that Figure 1 This illustration is merely for the purpose of understanding the scenarios in which the channel state feedback method provided in this application can be applied, and does not constitute any limitation on the scope of protection of this application. The communication system 100 may also include other network devices or other terminal devices.Figure 1 The method of channel state feedback provided in the application can also be applied to other scenarios, such as a cellular system. The scenarios to which the application can be applied are not described herein.

[0089] The terminal device in the embodiments of the application can refer to a user equipment, an access terminal, a user unit, a user station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a terminal, a wireless communication device, a user agent or a user apparatus. The terminal device can also be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), a handheld device with a wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a wearable device, a terminal device in a future 5G network or a terminal device in a future evolved Public Land Mobile Network (PLMN), and the like. The specific form of the terminal device is not limited in the application.

[0090] It should be understood that, in the embodiments of the application, the terminal device can be an apparatus for implementing the functions of the terminal device, or an apparatus capable of supporting the terminal device to implement the functions, such as a chip system, which can be installed in the terminal. In the embodiments of the application, the chip system can be composed of a chip, or can include a chip and other discrete devices.

[0091] The network device in the embodiments of the present application can be a device for communicating with a terminal device. The network device can be a network device (Base Transceiver Station, BTS) in a Global System of Mobile communication (GSM) system or a Code Division Multiple Access (CDMA) system, can be a network device (NodeB, NB) in a Wideband Code Division Multiple Access (WCDMA) system, can be an evolved network device (eNB or eNodeB) in an LTE system, can be a wireless controller in a cloud radio access network (CRAN) scenario, or can be a relay station, an access point, a vehicle-mounted device, a wearable device, a network device in a future 5G network, or a network device in a future evolved PLMN network, and the like, which are not limited in the present application.

[0092] It should be understood that, in the embodiments of the present application, the network device can be an apparatus for implementing the function of the network device, or can be an apparatus capable of supporting the network device to implement the function, such as a chip system, which can be installed in the network device.

[0093] It should also be understood that the embodiments shown below do not particularly limit the specific structure of the execution subject of the method provided in the embodiments of the present application, as long as the execution subject can communicate according to the method provided in the embodiments of the present application by running a program in which the code of the method provided in the embodiments of the present application is recorded. For example, the execution subject of the method provided in the embodiments of the present application can be a terminal device or a network device, or a functional module in the terminal device or the network device that can call and execute the program.

[0094] In the following, without loss of generality, the method for channel state feedback provided in the embodiments of the present application is described in detail by taking the interaction between a network device and a terminal device as an example.

[0095] For the convenience of understanding the embodiments of the present application, several basic concepts involved in the embodiments of the present application are briefly described. The basic concepts described below are taken as examples of the basic concepts defined in the new radio (NR) protocol, but the embodiments of the present application are not limited to only being applicable to the NR system. Therefore, the standard names appearing when taking the NR system as an example for description are all functional descriptions, the specific names are not limited, and only represent the function of the device, which can be extended to other systems, such as the 2nd generation (2G), 3th generation (3G), 4th generation (4G), 5th generation (5G) or future communication system.

[0096] 1. Precoding technology:

[0097] The network device can process the to-be-sent signal by means of a precoding matrix matched with the channel state in the case of known channel state, so that the to-be-sent signal after precoding is adapted to the channel, thereby enabling the receiving device to suppress the signal influence between multiple receiving devices and maximize the signal to interference plus noise ratio (SINR) of the received signal. Therefore, through the precoding processing of the to-be-sent signal, the quality of the received signal (for example, the signal to interference plus noise ratio (SINR) and the like) is improved. Therefore, by using the precoding technology, the transmitting device and multiple receiving devices can transmit on the same time-frequency resource, that is, the MU-MIMO is realized.

[0098] It should be understood that the related description of the precoding technology in the present application is only an example for the convenience of understanding, and is not used to limit the protection scope of the embodiments of the present application. In the specific implementation process, the transmitting device can also perform precoding in other ways. For example, in the case where the channel information (for example, but not limited to, the channel matrix) cannot be obtained, a pre-set precoding matrix or a weighting processing manner is used for precoding. For the sake of brevity, the specific content is not described herein.

[0099] 2. Channel reciprocity:

[0100] In a time division duplexing (TDD) mode, uplink and downlink channels transmit signals on different time domain resources of the same frequency domain resource. Within a relatively short time (e.g., a coherence time of channel propagation), it can be considered that the channels experienced by signals on the uplink and downlink channels are the same, and the uplink and downlink channels can be equivalently obtained. This is the reciprocity of the uplink and downlink channels. Based on the reciprocity of the uplink and downlink channels, a network device can measure an uplink channel according to an uplink reference signal, such as a sounding reference signal (SRS). The network device can estimate a downlink channel according to the uplink channel, and thus can determine a precoding matrix for downlink transmission.

[0101] However, in a frequency division duplexing (FDD) mode, because the frequency band interval of the uplink and downlink channels is much larger than a coherence bandwidth, the uplink and downlink channels do not have complete reciprocity, and a precoding matrix for downlink transmission determined by using the uplink channel can not be adapted to the downlink channel. In an FDD system, a network device can perform downlink precoding by relying on channel state information (CSI) fed back to the network device by a terminal device. Specifically, a basic flowchart of CSI feedback performed by a network device and a terminal device is as shown in Figure 2

[0102] Figure 2 A method for channel state feedback provided by the present application is shown in a schematic flowchart. The feedback process includes the following steps: S210, a network device sends a reference signal to a terminal device, or a terminal device receives a reference signal from a network device. Specific descriptions of the reference signal can be found in the relevant introduction below. The terminal device receives the reference signal and can perform channel measurement based on the reference signal, and calculates a final channel state feedback information. The channel state feedback information includes CSI information of a downlink channel, as shown in Figure 2 The method flowchart also includes S220: the terminal device generates channel state feedback information based on the received reference signal. Specifically, the terminal device also needs to send the channel state feedback information to the network device, as shown in Figure 2 The method flowchart also includes S230: the terminal device sends the obtained channel state feedback information to the network device.

[0103] Optionally, in the method flowchart S240 shown in Figure 2 When the network device learns the channel state feedback information, the network device can send resource configuration information to the terminal device based on the channel state feedback information.

[0104] Optionally, Figure 2 ​The method shown also includes S250: the network device sends resource configuration information to the terminal device according to the first indication information, and the resource configuration information includes a modulation and coding scheme (MCS). The modulation and coding scheme (MCS) is used to indicate the modulation and coding scheme of the terminal device in the transmission of the downlink channel data.

[0105] Specifically, the network device determines the number of streams for transmitting data to the terminal device according to the rank indicator (RI) fed back by the terminal device; the network device determines the modulation order and the code rate of channel coding for transmitting data to the terminal device according to the channel quality indicator (CQI) fed back by the terminal device; and the network device determines the precoding for transmitting data to the terminal device according to the precoding matrix indicator (PMI) fed back by the terminal device.

[0106] 3. Reference signal (RS):

[0107] The RS can also be referred to as a pilot, a reference sequence, etc. In the embodiments of the present application, the reference signal can be a reference signal for channel measurement. For example, the reference signal can be a channel state information reference signal (CSI-RS) for downlink channel measurement, or a sounding reference signal (SRS) for uplink channel measurement. It should be understood that the reference signals listed above are only examples and should not constitute any limitation on the present application. The present application does not exclude the possibility of defining other reference signals in future protocols to achieve the same or similar functions.

[0108] 4. Precoding matrix indicator (PMI):

[0109] The PMI can be used to indicate a precoding matrix. The precoding matrix can be determined by the terminal device based on the channel matrix of each frequency domain unit, for example, or can be the precoding matrix in the predefined codebook that is closest to the precoding matrix determined based on the channel matrix. The precoding matrix can be determined by the terminal device through channel estimation or based on channel reciprocity. However, it should be understood that the specific method of determining the precoding matrix by the terminal device is not limited to the above, and the specific implementation can refer to the prior art. For the sake of brevity, they will not be listed one by one here.

[0110] For example, the precoding matrix can be obtained by singular value decomposition (SVD) of the channel matrix or the covariance matrix of the channel matrix, or can also be obtained by eigenvalue decomposition (EVD) of the covariance matrix of the channel matrix. It should be understood that the above-mentioned ways of determining the precoding matrix are only examples and should not constitute any limitation on the present application. The way of determining the precoding matrix can refer to the prior art, and for the sake of brevity, will not be listed one by one here.

[0111] The precoding matrix determined by the terminal device can be referred to as a precoding matrix to be fed back, or a precoding matrix to be reported. The terminal device can indicate the precoding matrix to be fed back through the PMI, so that the network device recovers the precoding matrix based on the PMI. The precoding matrix recovered by the network device based on the PMI can be the same as or similar to the precoding matrix to be fed back.

[0112] The higher the approximation degree of the precoding matrix determined by the network device according to the PMI and the precoding matrix determined by the terminal device in the downlink channel measurement, the more suitable the precoding matrix determined by the network device for data transmission is to the downlink channel, and thus the transmission quality of the signal can be improved.

[0113] It should be understood that the PMI is only a name and should not constitute any limitation on the present application. The present application does not exclude the possibility of defining other names of signaling for the same or similar functions in future protocols.

[0114] It should be noted that the method provided by the embodiments of the present application can be used by the network device to determine the precoding matrix corresponding to one or more frequency domain units based on the feedback of the terminal device. The precoding matrix determined by the network device can be directly used for downlink data transmission; or can be subjected to some beamforming methods, such as zero forcing (ZF), regularized zero-forcing (RZF), minimum mean-squared error (MMSE), maximum signal-to-leakage-and-noise (SLNR), etc., to obtain the precoding matrix finally used for downlink data transmission. The present application does not make any limitation thereon. In the absence of any specific description, the precoding matrix involved in the following can refer to the precoding matrix determined based on the method provided by the present application.

[0115] 5. Modulation and Coding Scheme (MCS):

[0116] The configuration of rate in LTE and NR is achieved by MCS index value. The MCS takes the factors affecting the communication rate as the column of the table, and takes the MCS index as the row, to form a rate table. Therefore, each MCS index actually corresponds to a physical transmission rate under a set of parameters. CSI is channel state information, which is a kind of information fed back by the terminal device to the network device, mainly including three parts: CQI, RI and PMI. Among them, CQI is the feedback information sent by the terminal device to the network device through the downlink channel. There are 15 values of CQI specified in the current standard, which are represented by 4 bits. Each CQI corresponds to a different modulation mode and code rate, as shown in the following Table 1:

[0117] Table 1

[0118]

[0119] The determination of CQI is generally based on the Signal to Interference-plus-noise ratio (SINR) value. The terminal device obtains the channel measurement information and interference information by measuring the relevant resources of the downlink CSI measurement, and further calculates the SINR. According to the SINR, the terminal device can determine the CQI level to be reported by looking up the SINR-BLER table stored by itself. An important information in determining CQI is the SINR value. The calculation of SINR value is related to channel quality, interference information, noise power, terminal device receiving algorithm, etc. In multiple-input-multiple-output (MIMO), the SINR value is also related to the precoding of the network device. If the network device precoding can match the channel (for example: the amplitude is large after the channel is multiplied by the precoding), the SINR value received by the terminal device can be improved. Therefore, when calculating the SINR value, the precoding assumption needs to be determined.

[0120] In NR, the standard specifies the precoding assumption under different CSI feedback modes, that is, the standard defines different precoding matrix W determination methods. Among them, the terminal device takes the matrix calculation formula as an example in the CQI calculation process:

[0121]

[0122] It can be seen that the formula measures and calculates the channel state, wherein [x] is the reference signal sent by the network device, W(i) is the precoding matrix, and [y] is the signal obtained by the terminal device after precoding. When the terminal device does not perform precoding assumption, the precoding matrix W(i) is associated with the PMI; when the terminal device performs precoding assumption, the precoding matrix W is associated with the downlink channel information measured by the terminal device.

[0123] For example, when the wideband and subband PMI information are included in the CSI feedback content, the precoding assumption during CQI calculation is the precoding matrix corresponding to the reported PMI. When the CSI feedback content only contains wideband PMI information, the precoding assumption during CQI calculation is a wideband precoding matrix multiplied by a random subband precoding matrix. When the CSI feedback content does not contain PMI information, the precoding assumption during CQI calculation is a unit matrix. In summary, which precoding assumption is used is predefined. Under this assumption, when subsequent data transmission is assumed, the precoding matrix used by the network device is the precoding matrix corresponding to the PMI reported by the terminal device, and the SINR value calculated by the terminal device is basically close to the SINR value during subsequent data transmission, under the condition that the downlink channel and interference do not change much.

[0124] It should be understood that the network device determines the MCS allocated to the terminal device based on the CQI value reported by the terminal device. When the SINR value calculated by the terminal device when reporting the CQI is basically close to the SINR value during subsequent transmission, the error probability is generally low under the corresponding MCS.

[0125] 6、Artificial intelligence (AI):

[0126] The wave of artificial intelligence (AI) is sweeping the world, and many words such as artificial intelligence, machine learning, and deep learning are constantly lingering in our ears. The present application relates to deep learning in AI, which is a learning process that uses deep neural networks to solve feature expression. Neural networks generally include two stages of training and testing. Training is the process of extracting model parameters from training data using a neural network model. Testing is to run the test data using the trained model (neural network model + model parameters) and view the results. Among them, the data involved in the training process is abstracted and formed into a usable framework, which is called a training framework. The present application mainly relates to the application of the neural network model of AI in the CQI feedback process, and the AI neural network architecture is not limited in the present application.

[0127] 7、AI encoder and decoder:

[0128] In current research on channel state feedback (CSI) in large-scale MIMO systems, the sparsity of the channel in the time, frequency, and spatial domains, as described above, can be utilized to compress the CSI to be fed back. The compressed CSI is then transmitted to the network device via a feedback link, where the network device reconstructs the original sparse channel matrix from the feedback values. This CSI feedback method is based on the autoencoder and decoder architecture in AI networks, performing AI-based encoding (optionally, AI-based or non-AI quantization) at the terminal device. Typically, the AI-encoded channel is a floating-point number and requires further quantization.

[0129] Optionally, AI-based quantization is used, where the entire neural network comprises three parts: an AI encoder, an AI quantizer, and an AI decoder. The AI ​​quantizer and AI encoder are sent together to the terminal device. After AI quantization, the terminal device sends the quantized bits to the network device. On the network device side, an AI dequantizer (which can have the same structural parameters as the AI ​​quantizer but is the inverse operation of the AI ​​quantizer) and an AI decoder are configured.

[0130] Optionally, non-AI-based quantization, such as equal-length quantization or uniform quantization, can be used: uniform quantization is performed on the range where the encoded floating-point number is located, and the network device reconstructs the data based on AI decoding.

[0131] It should be understood that the dimension of the information input to the encoder is usually inconsistent with the dimension of the output information after encoding. This indicates that the encoder compresses the information. In this embodiment, the ratio of the dimensions between the encoder output information and the input information is referred to as the encoder's compression ratio, compression percentage, or compression efficiency. For example, if the encoder input information is 4 bits and the encoder output information is 2 bits, then the encoder's compression ratio is 2 / 4 = 1 / 2.

[0132] It should also be understood that the dimension of the information input to the decoder is usually inconsistent with the dimension of the information output after decoding. This can be understood as the encoder decompressing the information. In this embodiment, the ratio of the dimensions between the decoder's input and output information is referred to as the decoder's decompression rate, decompression percentage, or decompression efficiency, etc. For example, if the decoder's input information is 2 bits and the decoder's output information is 4 bits, then the decoder's decompression rate is 2 / 4 = 1 / 2.

[0133] Specifically, such as Figure 3 As shown, Figure 3is a CSI feedback process based on AI provided by an embodiment of the present application. The network device and the terminal device are jointly trained to obtain an encoder and a decoder. The encoder is deployed in the terminal device, and the decoder is deployed in the network device. The terminal device measures downlink channel information according to a downlink reference signal. The downlink channel information is encoded by the encoder to obtain compressed code word information that needs to be fed back to the air interface. The compressed code word information is fed back to the network device through a feedback link. After the network device receives the compressed code word information, the reconstructed information of the channel information fed back by the terminal device is obtained by decoding through the locally deployed decoder. The CSI feedback method based on AI requires the terminal device to compress and feed back the full channel information, and the feedback amount is large. If only the AI-based method is considered to compress and transmit the channel information, for example, the PMI information is compressed, the problem of CQI calculation and the problem of precoding assumption in the CQI calculation process are not considered. Therefore, the present application provides a method for feeding back channel state, so as to solve the problem of CQI calculation and improve the accuracy of feedback information.

[0134] 8. AI-based CQI feedback

[0135] Specifically, as shown in Figure 3 , Figure 3 is a CQI feedback process based on AI provided by an embodiment of the present application. In the embodiment of the present application, the framework including the encoder and the decoder can be referred to as an AI model or an AI architecture, and no limitation is made to this.

[0136] The network device and the terminal device are jointly trained to obtain an encoder and a decoder. The encoder compresses and encodes the input information into a vector or a matrix, and inputs the vector or the matrix output by the encoder into the decoder to restore the original input. In the AI model architecture, the network structure of the encoder and the decoder is not constrained, and AI models such as a convolutional neural network (CNN) and a fully connected neural network (FCNN) can be used. The AI model used in the present application is not described here.

[0137] It should be noted that the encoder is deployed in the terminal device, and the decoder is deployed in the network device. The terminal device measures downlink channel information according to a downlink reference signal. The downlink channel information is encoded by the encoder to obtain encoded information, which is fed back to the network device through a feedback link. After the network device receives the encoded information, the reconstructed information of the channel information fed back by the terminal device is obtained by decoding through the locally deployed decoder.

[0138] It should be further noted that the training process of the encoder and the decoder involved in the AI-based CQI feedback process is relatively complex, and the training of the encoder and the decoder (optionally, the quantizer and the dequantizer) is generally completed in an offline state, and then the trained encoder and the decoder are applied to the encoding and decoding of the online channel.

[0139] It should be further noted that the encoder mentioned in the present application can also include a quantizer, and the decoder can include a dequantizer. The present application does not limit this.

[0140] When the terminal device calculates the CQI, in the pre-coding assumption process using the feature vector, the channel feature vector will be lost in the AI encoding process of the terminal device, which will cause the CQI value calculated by the terminal device to be too high, and the channel feedback information generated by the terminal device will have a large error. The network device determines to allocate a matching MCS mode to the terminal device, which is related to the channel feedback information sent by the terminal device. Therefore, when the terminal device reports inaccurate channel feedback information, the MCS allocated by the network device may not match, resulting in a low throughput of the terminal device.

[0141] As shown in Figure 4 , the method shown in Figure 4 is a feedback channel state architecture based on an AI model. The encoder and the decoder trained jointly by the network device and the terminal device are deployed on the terminal device side and the network device side, respectively. Figure 4

[0142] One possible implementation of the AI model required by the terminal device and the network device is that the network device is responsible for the management of the encoder and the decoder, that is, the network device has the entire AI model, and sends the encoder in the AI model to the terminal device. The sending method can be configured through RRC, other high-layer signaling, or through an application layer protocol, which is not limited in the present application.

[0143] Among them, Figure 4 The method shown in

[0144] S410, S410 is the same as S210 described above, and the specific description can be referred to the description of S210 in the above Figure 2 for the sake of brevity, which will not be repeated here.

[0145] Further, after receiving the reference signal, the terminal device can generate channel state feedback information according to the reference signal, Figure 4 The method flow shown in ​

[0146] Specifically, the terminal device obtains a feature vector of a downlink channel according to a received reference signal;

[0147] Then, the terminal device encodes the obtained feature vector of the downlink channel based on an encoder to obtain an encoded result;

[0148] Finally, the encoded result is quantized to generate channel state feedback information.

[0149] As an example, the channel information obtained by the terminal device according to the received reference signal is a channel matrix of Nr x Nt x M, where Nr is the number of antennas of the terminal device, Nt is the number of antennas of the base station, and M is the number of resource blocks (RBs) or the number of subcarriers.

[0150] The terminal device obtains a feature vector from the obtained channel information, which is a matrix of R x Nt x M, where R is the rank of the channel, and the AI encoder encodes the matrix of R x Nt x M as input.

[0151] Suppose the AI encoder is a fully connected layer, and the output of the fully connected layer is a D-dimensional vector. After the feature vector passes through the fully connected layer, it can be compressed into a D-dimensional vector. Further, the D-dimensional vector is input into an AI quantizer, which contains a matrix of L x D dimensions. Among the L vectors, find a vector closest to the input D-dimensional vector in terms of distance (e.g., Euclidean distance), and the index of the vector is taken as the quantized channel state feedback information, which is sent by the terminal device to the network device.

[0152] Wherein, the AI quantizer can be trained together with the AI encoder and the AI decoder.

[0153] It should be understood that the channel state feedback information is used to feedback the channel state, wherein the channel state feedback can be understood as the terminal device feeding back the measured channel between the network device and the terminal device.

[0154] In order to facilitate understanding, a specific example is used to illustrate how to feedback the channel state.

[0155] For example, the channel state feedback is to feed back the channel between the network device and the terminal device measured by the terminal device. Taking the network device with 32 antennas, the terminal device with 2 antennas, and the measurement bandwidth of 100 RBs as an example. When the terminal device measures the reference signal sent by the network device, a 32*2 channel matrix can be measured on each reference signal through channel estimation, and therefore the total size of the measured channel is 32*2*100. Since the channel has correlation in the frequency domain, a plurality of reference signals can be divided into a sub-band in the frequency domain and uniformly reported. For example, the size of the sub-band is 4 RBs / sub-band, and therefore the final matrix to be reported is a 32*2*25 complex matrix. If the correlation of the channel in the space is considered, the rank of the channel can be relatively low, for example, the rank of the above matrix can be 1, and therefore only the strongest eigenvector of the matrix in each sub-band needs to be reported without reporting the full channel information. Therefore, the channel information to be reported is a 32*25 matrix (singular value decomposition / eigenvalue decomposition is performed on each sub-band to obtain a 32*1 eigenvector). To further reduce the overhead, the terminal device maps each eigenvector to a predefined codebook and uses PMI to describe each eigenvector and report. In the AI-based CSI feedback mode, the eigenvector can be encoded by AI, and the AI decoder on the network device side can be used to recover.

[0156] Optionally, in the CSI reporting configuration issued by the network device, the ID of the configuration, the content reported, and the like are defined. Different CSI reporting configurations can configure the reporting of PMI and the non-reporting of PMI. For example, an RRC information element for reporting AIPMI is added. When AI CSI reporting is used, the related configuration about AI can be added in the report Quantity configuration, indicating that the reporting of PMI is based on the AI mode.

[0157] After the terminal device generates the channel state feedback information, the terminal device can send the channel state feedback information to the network device, as shown in the method flow. Figure 4 The method flow shown also includes:

[0158] S430, the terminal device sends the channel state feedback information to the network device, or the network device receives the channel state feedback information from the terminal device.

[0159] After the network device receives the channel state feedback information, the network device can determine the resources required by the downlink channel based on the channel state feedback information. Figure 4 The method flow shown also includes:

[0160] S440, the network device determines the resources required by the downlink channel and generates resource configuration information.

[0161] As a possible implementation manner, the channel state feedback information includes generated PMI information, RI information and CQI information of the terminal device, wherein the PMI information is information encoded by an AI encoder in the terminal device.

[0162] The channel feedback information includes PMI information, RI information and CQI information reported by the terminal device, wherein the PMI information is information encoded by an AI encoder in the terminal device.

[0163] In this implementation manner, the network device receives the channel state feedback information including RI information and CQI information and PMI information encoded by an AI encoder in the terminal device, and then obtains the key information of the downlink channel by decoding the information by a decoder. The key information of the downlink channel can include precoding information and / or rank information of the downlink channel, and quality information of the downlink channel.

[0164] As another possible implementation manner, the network device determines the RB to be allocated according to the CQI information on different frequency band resources, and further determines the MCS information on the corresponding RB. In addition, the network device can determine the precoding and rank required by the downlink channel according to the PMI information and RI information on the corresponding RB.

[0165] The required resources of the downlink channel are used to generate the resource configuration information.

[0166] Further, the network device generates the resource configuration information according to the downlink channel information and sends it to the terminal device. Figure 4 The method flowchart also includes:

[0167] S450, the network device sends the resource configuration information to the terminal device, or the terminal device receives the resource configuration information from the network device.

[0168] The resource configuration information includes the MCS information required by the network device to allocate for the corresponding downlink channel of the terminal device.

[0169] The method described in the above Figure 4 After the terminal device encodes and quantizes the obtained channel feature vector by an encoder, the channel state feedback information is obtained. The terminal device sends the channel state feedback information to the network device. The network device decodes the received channel state feedback information. Figure 4 The method described in the above solves the problem of transmission accuracy of the channel state feedback information and the problem of low matching degree of the MCS allocated by the network device for the terminal device and the channel.

[0170] Figure 5A method for channel state feedback provided by an embodiment of the present application is shown in a schematic flowchart. The method can include the following steps:

[0171] S510, the network device sends a reference signal to the terminal device, or the terminal device receives the reference signal sent by the network device.

[0172] It should be understood that the transmission mode of the reference signal can be divided into beamformed and non-beamformed.

[0173] Mode one: in the non-beamformed transmission mode, the network device does not precode the to-be-transmitted reference signal, which is equivalent to omnidirectional transmission;

[0174] Mode two: in the beamformed transmission mode, the network device pre-codes the to-be-transmitted reference signal before transmission, and the terminal device measures the downlink channel when receiving the reference signal. The measured downlink channel has superimposed pre-coding information.

[0175] It should also be understood that in an implementable beamformed reference signal transmission scenario, the transmitting end transmits the reference signal in the pre-coding mode based on compression sensing to the terminal device. The reference signal based on compression sensing is superimposed on the same time-frequency resource for transmission, which can save time-frequency resources. The receiving end can recover the superimposed reference signal through an iterative algorithm to estimate the complete channel. Since the estimated channel information in the beamformed reference signal mode already contains pre-coding, the terminal device does not need to make additional pre-coding assumptions, and the terminal device is equivalent to using a unit matrix pre-coding assumption.

[0176] It should also be understood that the above step S510 is an optional step. For example, the terminal device performs channel estimation and obtains a channel feature vector based on a cell-common signal (such as a synchronization signal), and does not need to obtain the channel feature vector through reference information.

[0177] Further, after receiving the above reference signal, the terminal device can obtain a downlink channel feature vector based on the reference signal.

[0178] It should be understood that the terminal device obtains downlink channel information, i.e., CSI information, based on the reference signal measurement. The CSI information mainly includes pre-coding matrix indication (PMI), rank indication (RI), channel quality indication (CQI), and other parameters, which are used to describe the downlink channel information. Among them, PMI is used to describe the channel feature vector; RI is used to describe the rank of the channel; and CQI is used to describe the overall quality of the channel, in addition to the useful signal quality, also including the interference condition of the channel.

[0179] The terminal device calculates a first SINR value according to the obtained downlink channel feature vector as a precoding assumption, Figure 5 The method flowchart also includes: S520, the terminal device calculates a first value based on the channel feature vector as a precoding assumption.

[0180] Optionally, the terminal device receives a reference signal sent by the network device;

[0181] Then, the terminal device obtains a channel feature vector of a downlink channel according to the reference signal;

[0182] Finally, the terminal device calculates a first value of the first information based on the obtained channel feature vector of the downlink channel as a precoding assumption.

[0183] After the terminal device obtains the first value, it will calculate a second value according to first indication information sent by the network device. For example, Figure 5 The method flowchart also includes:

[0184] S530, the network device sends first indication information to the terminal device, or the terminal device receives first indication information from the network device.

[0185] S540, the terminal device reduces the first value of the first information to a second value according to the first indication information.

[0186] The second value is used to determine channel state feedback information. For example, the terminal device can generate channel state feedback information according to the second value.

[0187] The terminal device reduces the first value of the first information to a second value according to the first indication information, which can also be described as the terminal device reduces the first information from the first value to the second value according to the first indication information.

[0188] Optionally, the first indication information is a fallback value of the channel state feedback configuration sent by the network device, and the terminal device calculates the first value in the first information according to the channel feature vector as a precoding assumption, and calculates the second value based on the fallback value sent by the network device.

[0189] Optionally, the first indication information is AI model information of the channel state feedback configuration sent by the network device, and the terminal device further calculates the required fallback value according to the model parameter, and reduces the first value in the first information by the calculated fallback value to obtain the second value.

[0190] Optionally, the first information is CQI information, the first value in the first information is a first CQI value, and the second value is a second CQI value.

[0191] Optionally, the first information is SINR information, where the first value is the first SINR value and the second value is the second SINR value.

[0192] As described above, one possible implementation is that the first indication information can be a fallback value directly configured by the network device and sent to the terminal device.

[0193] In this implementation, the terminal device receives a backoff value in the first indication information. For example, if the network device configures the SINR backoff value to be one of {0.5, 1, 1.5, 2}, after receiving the first indication information, the terminal device directly calculates the second SINR based on the backoff value in the first indication information. The specific code in the RRC cell is shown below:

[0194]

[0195] As mentioned above, as another possible implementation, the first indication information is the AI ​​model information configured by the network device.

[0196] In this implementation, the first instruction information received by the terminal device from the network device is AI model information. For example, the network device can configure the AI ​​model information in the CSIreport information element in the above code, or it can be sent along with the AI ​​model as supplementary information.

[0197] Optionally, the network device may send the first indication information in the following ways: through RRC signaling configuration, physical layer signaling configuration (such as downlink control information), application layer protocol configuration, medium access control element (MAC-CE), or other higher layer signaling configuration. This application does not limit this.

[0198] It should be noted that AI model information is used to measure the degree of matching between the AI ​​model and the original channel information in channel information recovery. Specifically, AI model information may include information such as the cosine similarity of the AI ​​model or the normalized mean square error of the AI ​​model. This application does not limit the content of the AI ​​model information used herein.

[0199] After determining the second value following the fallback, the terminal device can generate channel state feedback information based on the second value. It should be noted that this channel state feedback information is determined by the second value from the first information.

[0200] As one possible approach, we take the AI ​​model information as the first indication information, which includes the cosine similarity of the AI ​​model, as an example. The cosine similarity is defined as follows:

[0201]

[0202] wherein, W is the original channel matrix, is the recovered channel matrix. The range interval of the index p is [0, 1], the closer to 1, the closer the channel matrix decoded by the network device decoder to the original channel, and the better the channel information recovery performance.

[0203] Further, the terminal device determines the fallback value according to the model performance information in the first indication information sent by the network device.

[0204] Specifically, taking the rank of the channel matrix as 1 as an example, when the terminal device calculates the SINR corresponding to the CQI, the terminal device adopts the channel eigenvector as the assumption of precoding, and then the SINR can be expressed as:

[0205]

[0206] wherein, g is the receiving vector adopted by the terminal device, W is the channel eigenvector, and I+n represents the interference and noise power after multiplying the receiving vector. Further, the SVD decomposition is performed on H: H = USV, or the eigenvalue decomposition is performed on HH': HH' = USU', wherein H' is the conjugate matrix of H. s is the maximum singular value / eigenvalue of the H matrix, and U is the left eigenvector of the H matrix.

[0207] Then, the network device recovers the AI-encoded channel eigenvector W and uses it as the subsequent precoding vector. Assuming that the interference and noise in the channel do not change in a short time, the SINR received by the terminal device can be expressed as:

[0208]

[0209] wherein, is the recovered eigenvector of the AI encoder. According to the definition of the cosine similarity, combined with the fact that the modulus of the eigenvector is 1, the rightmost equation can be further obtained.

[0210] Alternatively, it can be seen from the above formula that the second value in the channel state information actually received by the network device has a loss of the square of the cosine similarity compared to the first value calculated by the terminal device. Therefore, the p 2 in the above formula can be taken as the fallback value when calculating the downlink channel CQI. That is, after the terminal device calculates the first value, the corresponding fallback value p 2 is multiplied, the terminal device determines the SINR value of the new downlink channel and maps to obtain the new CQI value, that is, the second value.

[0211] When Rank > 1, the corresponding formula can also be obtained similarly, for example, the backoff value is expressed as follows:

[0212]

[0213] wherein p ji is the cosine similarity, a ij is an element in the matrix obtained by multiplying the reception matrix and the left eigenvector of the channel, and s is the eigenvector of the channel. is the SINR value calculated by the terminal device based on the eigenvector of the channel for precoding assumption. The cosine similarity p ji can constitute a symmetric matrix, indicating the cosine similarity between the recovered eigenvector i and the actual eigenvector j of the channel.

[0214] As another possible implementation manner, the cosine similarity exemplified in the embodiments of the present application can also be a statistical mean. For example, the cosine similarity mean of the AI model. In fact, the performance of the AI model is not completely the same under different test data. Therefore, the cosine similarity of the AI model actually has the probability distribution characteristic, and when the variance of the probability distribution density is small, it is appropriate to use the statistical mean to express the performance of the cosine similarity.

[0215] However, when the variance of the probability distribution density is large, using only the statistical mean to determine the backoff value can cause the backoff value calculated by the terminal device to be too small. For example, when the cosine similarity recovered by the AI model is mu-sigma, which is less than the statistical mean mu, the backoff value calculated by the terminal device according to the model parameter is too small, which can cause the CQI value reported by the terminal device to be still too large.

[0216] As another possible implementation manner, the embodiments of the present application propose to indicate the distribution characteristics of the performance of the AI model, taking the following two indicating manners as examples: the terminal device uses multiple samples of the probability distribution, such as the cosine similarities corresponding to the positions of mu, mu-sigma, mu-2sigma, etc.; the network device directly indicates the type, mean and / or variance, etc. information of the distribution of the terminal device.

[0217] It should be understood that, in an ideal case, the cosine similarity matrix is a unit matrix. The recovered eigenvector is consistent with the actual eigenvector.

[0218] It should also be understood that in the manner in which the network device configures the first indication information of the AI model, the network device configures the cosine similarity information of the AI model in the first indication information and sends it to the terminal device under different ranks. After receiving the first indication information sent by the network device, the terminal device calculates the fallback value according to the cosine similarity of the AI model under different rank values in the first indication information. The terminal device reduces the first value in the first information to the second value according to the calculated fallback value. The terminal device sends the channel state feedback information generated by the second value to the network device.

[0219] The present application proposes two implementation manners of CQI fallback. There are other ways to determine the CQI value, which are not limited by the present application.

[0220] Manner one: the terminal device directly performs fallback on the SINR value, and the terminal device determines the corresponding CQI, i.e., the second CQI, according to the remapping of the fallback SINR value. For example, the first information is SINR information, the first value is the first SINR value, and the second value is the second SINR value. The terminal device reduces the first SINR to the second SINR according to the first indication information, and the terminal device maps the second CQI according to the second SINR value.

[0221] Manner two: the terminal device first maps the original calculated SINR value to the original CQI, i.e., the first CQI. The terminal device obtains the new CQI, i.e., the second CQI, by lowering the first CQI by several levels according to the first CQI and the first indication information. For example, the first information is CQI information, the first value is the first CQI value, and the second value is the second CQI value. The terminal device determines the first CQI value according to the first SINR value, wherein the first SINR value is determined by the terminal device according to the channel feature vector. The terminal device reduces the first CQI to the second CQI according to the first indication information.

[0222] As a possible implementation manner, the terminal device calculates the SINR value and maps a CQI value to be reported, and judges whether the CQI mismatch condition is possible to occur. According to the above fallback value calculation manner, the fallback value threshold value under the calculated CQI level can be obtained, and the threshold value is calculated through (CQI threshold value + fallback value). Thus, according to the calculated threshold value, the fallback range under each CQI level can be obtained. For example, when the calculated SINR value falls within the fallback range, it indicates that the SINR value is reduced (for example, the actual CQI level is 2), and according to the calculated SINR, the CQI mapping falls within the SINR range of other CQI levels (for example, the CQI level to be reported is 3), that is, the CQI level mismatch condition occurs. In order to solve the above CQI level mismatch condition, the terminal device first judges whether the originally calculated SINR value falls within the fallback range, and if it falls within the fallback range, the terminal device performs fallback on the CQI value to be reported by the terminal device according to the CQI level predefined by the network device. For example, the terminal device is configured with a fallback level of 1, and the terminal device reports the CQI value to be reported according to the calculated fallback value decreased by one level according to the configured fallback level 1.

[0223] It should be noted that the calculation of the CQI depends on the SINR value calculated by the terminal device.

[0224] It should also be noted that the first value of the first information in the embodiments of the present application can also not be subjected to fallback. For specific examples, please refer to the method shown in the present application Figure 6 , which will not be repeated here.

[0225] After the terminal device calculates the second CQI according to the first indication information, the terminal device can generate channel state feedback information according to the second CQI, and the terminal device sends the channel state feedback information to the network device. Optionally, Figure 5 The method flow shown also includes:

[0226] S550, the terminal device sends the channel state feedback information to the network device, or in other words, the network device receives the channel state feedback information from the terminal device.

[0227] After the network device receives the channel state feedback information, the network device can judge the resources required by the downlink channel based on the channel state feedback information. Optionally, Figure 5 The method flow shown also includes:

[0228] S560, the network device generates resource configuration information according to the channel state feedback information. Or in other words, the network device judges the resources required by the downlink channel. S560 is the same as S440 described above, and the specific description can be referred to the description of S440 in the above Figure 4 , in order to avoid repetition, which will not be repeated here.

[0229] Optionally, Figure 5 The method flow shown also includes:

[0230] S570, the network device sends resource configuration information to the terminal device, or the terminal device receives resource configuration information from the network device.

[0231] It should be understood that the resource configuration information can include the MCS resource information and the like that the network device needs to allocate for the downlink channel corresponding to the terminal device.

[0232] The method described above Figure 5 The method described above Figure 4 The method described above adds the step of the network device sending first indication information, and the terminal device performs backoff calculation on the first value of the first information calculated initially according to the received first indication information. The terminal device generates channel state feedback information according to the second value recalculated, improves the accuracy of the channel state feedback information determined by the terminal device, and also improves the matching degree of the resource allocated by the network device for the terminal device.

[0233] It should be noted that when the cosine similarity of the AI model is represented based on the distribution characteristics, the terminal device can determine multiple backoff values. For example, the terminal device can obtain multiple cosine similarities according to the sampling of the cosine similarity distribution of the AI model, so as to calculate multiple backoff values according to a predefined formula. The predefined formula can be the square of the cosine similarity or the formula for calculating the backoff value when the rank>1. When the terminal device uses different backoff values, the probability of the occurrence of CQI mismatch is also different. For example, when the terminal device calculates a smaller backoff value, the network device can not appear CQI mismatch; when the terminal device calculates a larger backoff value, the network device can appear CQI mismatch.

[0234] It should also be noted that in order to ensure that the resource allocated by the network device according to the channel state feedback information reported by the terminal device can be 100% matched, the terminal device can back off the first value of the first information more, so that the terminal device generates a lower second CQI. In this case, although the network device has a lower bit error rate, the throughput of the terminal device will also be lower. The present application proposes a flexible mechanism based on probability for the above problems, that is, even if the network device side can appear mismatch, if the probability of this situation is relatively small, the terminal device can still use a smaller backoff value, so that the terminal device can report a higher CQI level as much as possible.

[0235] Specifically, assuming that the SINR range between CQI grades is Y dB, and the backoff range is X dB. Considering that the distribution of the calculated SINR is uniform and equally probable, the probability of CQI mismatch can be expressed as:

[0236]

[0237] Assuming that in the cosine similarity distribution, the probability of obtaining a certain cosine similarity is Prho, then the probability of obtaining the corresponding backoff value is also Prho. Therefore, in this backoff case, the probability of network equipment CQI mismatch can be expressed as:

[0238] P mismatch =P backoff ×Prho

[0239] If the probability of CQI mismatch calculated by the terminal device under different cosine similarity samplings is less than or equal to the threshold value given by the network device, it means that the probability of network device CQI mismatch is small, so in this case, the terminal device can perform CQI mapping according to the calculated SINR value, and the terminal device does not back off the SINR value. Otherwise, the backoff value can be determined according to the cosine similarity sampling value corresponding to the maximum mismatch probability, rather than simply selecting the average value of the cosine similarity as the backoff value.

[0240] Optionally, the determination of the backoff value is closely related to the AI model, and the configuration, update and backoff value of the AI model may not be synchronized. For example, as described above, the AI model is transmitted through other high-layer signaling / application layer protocol, and the configuration of the backoff value is configured in the RRC as the CSI reporting configuration. Therefore, the model may be replaced, and the backoff value may not be replaced, so a default processing method is needed to deal with this situation.

[0241] As a possible implementation, the default processing method can include:

[0242] Method one: the default backoff value of the terminal device is 0, that is, no backoff is performed;

[0243] Method two: the terminal device does not use the method based on the channel feature vector to calculate the CQI, but uses the fixed precoding assumption method to calculate the CQI, which can avoid the CQI mismatch.

[0244] The essence of the above two methods is based on the replacement of the AI model of the terminal device, and the default processing method of the feedback of the backoff value, and other processing methods are not limited by the present application.

[0245] According to the above channel state feedback method, Figure 6Another method for channel state feedback is shown in the schematic flowchart.

[0246] As shown in the method shown in Figure 6 The method can include steps 610 to 650.

[0247] S610 and S620 are the same as S510 and S520 in Figure 5 For brevity, they will not be repeated here.

[0248] S630, the network device sends the first indication information and the second probability value to the terminal device, or the terminal device receives the first indication information and the second probability value sent by the network device.

[0249] As a possible implementation, the network device configures the first indication information for the terminal device through RRC signaling, the configuration content indicates "AIPMI-CQI", and in this field, the first indication information configures distribution-related parameters about the AI model (for example: cosine similarity of the AI model, normalized mean square error of the AI model), and the second probability value, the configured information for channel state feedback is used for the terminal device to judge whether the second value needs to be calculated back. For example, it can be implemented through the following code:

[0250]

[0251] When the network device configures the first indication information, it also sends the second probability value to the terminal device. The second probability value can also be a CQI mismatch tolerance probability value. The AI model cannot recover the original feature vector with 100% accuracy, so there is a certain probability that the CQI calculated by the terminal device is higher than the actual channel CQI. That is, the CQI received by the network device may not match. The first probability given by the network device is the upper limit of the probability of allowing this situation to occur.

[0252] It should be understood that in the process of mapping SINR value to CQI, the SINR gap between different CQI levels is not completely the same, so the probability of mismatch on different CQI levels is also not completely the same.

[0253] The network device sends the first indication information and the second probability value to the terminal device, and accordingly, the terminal device receives the first indication information and the second probability value sent by the network device, and optionally, Figure 6 The method shown in

[0254] S640, the terminal device judges whether to perform fallback calculation on the first value according to the received first indication information and the second probability value. The terminal device further generates channel state feedback information according to the judgment result.

[0255] Specifically, in order to determine whether to perform backoff, the terminal device needs to obtain a plurality of backoff values according to the first indication information, and obtain a plurality of first probability values according to the plurality of backoff values. For the convenience of understanding, the first probability value can be referred to as a CQI mismatch probability value.

[0256] The terminal device compares the plurality of CQI mismatch probability values with the second probability value respectively, wherein the number of CQI mismatch probability values greater than or greater than or equal to the second probability value is N.

[0257] In the case that the N is greater than the first threshold value, the terminal device determines to perform backoff; in the case that the N is less than the first threshold value, the terminal device determines not to perform backoff; in the case that the N is equal to the first threshold value, the terminal device can determine to perform backoff or not to perform backoff.

[0258] Exemplarily, the first threshold value is predefined; or, the first threshold value is sent by the network device to the terminal device; or, the first threshold value is determined by negotiation between the terminal device and the network device. The present application does not make any limitation in this regard.

[0259] For example, the terminal device obtains a plurality of backoff values (such as 0.36dB, 0.6dB, 0.86dB shown in Table 2) according to the first indication information, and obtains a plurality of CQI mismatch probability values (such as 0.12, 0.097, 0.057 shown in Table 2) according to the plurality of backoff values as shown in Table 2. In the case that the second probability value received by the terminal device is 0.1 and the first threshold value is 1, the number N of CQI mismatch probability values greater than the second probability value (such as 0.12) in the plurality of CQI mismatch probability values is 1, the value of the N is equal to the first threshold value, and the terminal device can determine not to perform backoff, or the terminal device can also determine to perform backoff.

[0260] For example, the terminal device obtains a plurality of backoff values (such as 0.36dB, 0.6dB, 0.86dB shown in Table 2) according to the first indication information, and obtains a plurality of CQI mismatch probability values (such as 0.12, 0.097, 0.057 shown in Table 2) according to the plurality of backoff values as shown in Table 2. In the case that the second probability value received by the terminal device is 0.08 and the first threshold value is 1, according to Table 2, the number N of CQI mismatch probability values greater than the second probability value (such as 0.12, 0.097) in the plurality of CQI mismatch probability values is 2, the value of the N is greater than the first threshold value, and the terminal device determines to perform backoff.

[0261] In the case of a certain backoff value, the CQI mismatch probability value = the probability of falling into the backoff range * the cumulative probability of the sampling point. The specific values are shown in the following table:

[0262] Table 2 mismatch probability calculation example

[0263] Fall back value Probability of falling into fall back range Cumulative probability of sampling points CQI mismatch probability value 0.36 dB (rho = 0.92) 0.24 0.5 0.12 0.6 dB (rho = 0.87) 0.39 0.25 0.097 0.86 dB (rho = 0.82) 0.57 0.1 0.057

[0264] Optionally, the selection of the backoff value can use the backoff value corresponding to the maximum probability of mismatch. For example, the maximum CQI mismatch probability value in Table 2 is 0.12, and the terminal device should select 0.36 dB as the backoff value for calculation.

[0265] As another implementable manner, if the network device configures the configuration information of the probability density distribution of the cosine similarity of the AI model, the terminal device can calculate a backoff value threshold according to the threshold of the network device allowing the mismatch to occur and the range of the SINR corresponding to the current CQI level. The terminal device can obtain the lower bound of the cosine similarity square according to the inverse operation of the pre-defined formula, so as to determine the position of the cosine similarity in the performance distribution and the corresponding cumulative probability.

[0266] For example, the network device configures the mismatch threshold as 0.1, and the terminal device calculates the SINR range corresponding to the CQI level as 2 dB. According to the backoff range of 0.2 dB in the backoff indication information issued by the network device, the cosine similarity square at least needs to be 0.95, taking rank 1 as an example. Further, the terminal device determines the cumulative probability of the cosine similarity square being greater than 0.95 according to the distribution of the cosine similarity.

[0267] If the cumulative probability is greater than the second probability value, that is, the probability of the cosine similarity of the AI model being greater than the lower bound of the cosine similarity, and is higher than a certain threshold value, no backoff is performed, otherwise, the backoff is performed. Optionally, the backoff can also be performed when the cumulative probability is greater than or equal to the second probability value.

[0268] The terminal device determines and generates the channel state feedback information according to step 640, and sends the channel state feedback information to the network device, or the network device receives the channel state feedback information sent by the terminal device.

[0269] Specifically, when the terminal device determines that the first value does not back off, the channel state feedback information is generated according to the first value calculated in step 620.

[0270] Optionally, when the first value is the SINR value, the CQI needs to be further mapped.

[0271] Optionally, when the terminal device determines that the first value needs to back off according to step 640, the channel state feedback information is generated after the backoff of the first value calculated in step 620.

[0272] Optionally, after the terminal device generates the channel state feedback information, the terminal device sends the channel state feedback information to the network device, such as Figure 6The method shown in the method further comprises:

[0273] S650, the terminal device sends channel state feedback information to the network device, or the network device receives the channel state feedback information sent by the terminal device.

[0274] S650 to S670 can be referred to Figure 5 S550 to S570 in the method, for brevity, will not be repeated here.

[0275] Figure 6 The method shown in the method is further combined with the distribution in the AI model information to determine the required fallback value, and the terminal device judges whether the first value needs to be calculated according to the second probability value sent by the network device and the first probability value obtained by the terminal device according to the first indication information. Compared with the above other embodiments, the terminal device determines whether the first value needs to be calculated, which avoids the error caused by the terminal device relying on a single statistical value to calculate the fallback value, and also improves the accuracy of the feedback information.

[0276] Figure 7 is another example of a specific example of the channel state feedback method in the present application.

[0277] S710 is the same as Figure 5 S510 in the method, for brevity, will not be repeated here.

[0278] S720, the terminal device acquires the channel feature vector according to the reference signal and generates the first SINR value.

[0279] Optionally, the terminal device acquires the channel feature vector according to the reference signal, and adopts a unit matrix as a precoding assumption, and the terminal device calculates the first SINR value by using its own receiver algorithm and / or a typical receiver algorithm (such as MRC algorithm).

[0280] Optionally, when the network device adopts the beamformed mode to send the reference signal, the network device has already precoded the sent reference signal, so when the terminal device adopts the unit matrix to calculate the first SINR value, it can be considered that the precoding information has been considered in the channel matrix, and no additional precoding matrix assumption is needed.

[0281] Optionally, when the network device adopts the non-beamformed mode to send the reference signal, the terminal device calculates the first SINR value by taking the channel feature vector as a precoding assumption. Figure 6 S620 in the method, for brevity, will not be repeated here.

[0282] Specifically, the terminal device acquires a channel feature vector, and calculates a first SINR value by using a self-receiver algorithm and / or a typical receiver algorithm. If the terminal device uses the self-receiver algorithm to obtain the SINR value, the SINR value obtained by the terminal device is the first SINR value. If the terminal device uses the self-receiver algorithm and the typical receiver algorithm to obtain different SINR values, the terminal device can take the SINR value obtained by the typical receiver algorithm as a SINR reference value, and send a difference between the SINR value obtained by the self-receiver algorithm and the SINR reference value to the network device. At this time, the network device obtains the same SINR reference value as the terminal device by using the typical receiver algorithm, and analyzes the difference between the SINR reference value and the SINR value reported by the terminal device to obtain the channel state information. The terminal device sends the SINR value directly, instead of sending the CQI, thereby avoiding the information loss in the process of mapping the SINR value to generate the CQI, and enabling the network device to determine the channel state more accurately.

[0283] Optionally, the typical receiver algorithm can be configured by the network device and sent to the terminal device; or the typical receiver algorithm is predefined; or the typical receiver algorithm is determined by negotiation between the terminal device and the network device. The present application does not limit this.

[0284] S730, the terminal device receives the first indication information and the second indication information from the network device, or the network device sends the first indication information or the second indication information to the terminal device.

[0285] It should be noted that the first indication information and the second indication information can be sent by one message, for example, the first indication information and the second indication information are carried in one message; or the first indication information and the second indication information can be sent by different messages, for example, the first indication information is carried in message #1, and the second indication information is carried in message #2.

[0286] S740, the terminal device determines the second SINR value and the first SINR value according to the first indication information. Figure 5 The second value determined in S540 in S540 is the same as that in S540, and for the sake of brevity, it will not be repeated again.

[0287] S750, the terminal device generates a third SINR value according to the second indication information and the second SINR value.

[0288] Specifically, when the network device configures the first indication information for the terminal device, the first indication information content is configured as "AIPMI-SINR", indicating that the terminal device reports the PMI based on the AI mode and reports the SINR value. The second indication information configured by the network device includes the quantization mode of the SINR, which can be implemented by the following code:

[0289]

[0290] Optionally, the second indication information sent by the network device indicates a third SINR value directly sent by the terminal device to the network device. The third SINR value can be one SINR value or a difference value based on a SINR reference value. The SINR reference value is obtained based on some typical receiver algorithm, for example, the SINR value calculated by the terminal device using the MRC receiving algorithm is used as the SINR reference value. When the terminal device sends the SINR reference value, the range of the SINR value is relatively small, and therefore the accuracy of the SINR value is higher under the same overhead.

[0291] Optionally, the second indication information sent by the network device includes SINR quantization configuration information, the terminal device quantizes the second SINR value to obtain a third SINR value according to the SINR quantization configuration information, and sends the third SINR value to the network device.

[0292] Optionally, the second indication information sent by the network device includes at least one of the following: quantization range, quantization bit number or quantization step. The quantization range, quantization bit number and quantization step all belong to the SINR quantization configuration information. This embodiment illustrates the quantization methods as follows:

[0293] Method one: the quantization bit number, quantization interval and step are all fixed as the standard definition, and in this case, the network device does not need to additionally give the quantization related configuration in the RRC configuration.

[0294] Method two: the quantization bit number is fixed, the network device configures the SINR range reported by the terminal device through the RRC configuration, and in this case, a uniform step is used. In this case, the second indication information includes the fixed quantization bit number configured by the network device to the terminal device, and specifically, the terminal device quantizes the SINR value using the uniform step according to the fixed quantization bit number to obtain the reported SINR range.

[0295] Method three: the quantization range is fixed, the network device configures the bit number used by the terminal device for reporting through the RRC configuration, and in this case, a uniform step is used. In this case, the second indication information includes the fixed SINR value range configured by the network device to the terminal device, and specifically, the terminal device quantizes the SINR value using the uniform step according to the fixed SINR value range to obtain the reported bit number.

[0296] Of course, there are other quantization methods, which are not limited in the present application.

[0297] According to the different quantization methods configured by the network device to the terminal device, the terminal device quantizes the SINR value to be reported and then sends it to the network device.

[0298] Optionally, the second indication information indicates that the terminal device reports an absolute SINR value, and the terminal device directly sends the third SINR value obtained by quantizing the second SINR value to the network device.

[0299] Optionally, the second indication information indicates that the terminal device reports a SINR gap value, and the terminal device calculates a SINR gap value based on the SINR value calculated by the receiver algorithm of the terminal device and the SINR value calculated by the typical receiver algorithm, and sends the SINR gap value to the network device as the third SINR value.

[0300] The terminal device sends the third SINR value to the network device, or the network device receives the third SINR value sent by the terminal device, as described in the method in Figure 7 The method described in

[0301] The terminal device sends the third SINR value to the network device, or the network device receives the third SINR value sent by the terminal device, as described in the method in Figure 7 The method described in

[0302] The network device determines the channel interference condition based on the third SINR value sent by the terminal device, and further determines the resource configuration information.

[0303] Specifically, if the terminal device reports a SINR value to the network device, the network device determines the channel interference condition by combining the received SINR value with the AI encoded channel information (for example, PMI information). If the terminal device reports a SINR gap value to the network device, the network device obtains a SINR value based on the typical receiver algorithm, and calculates the SINR value under the receiver of the terminal device based on the SINR gap value reported by the terminal device, to determine the channel interference condition. The network device determines the MCS resource allocation information for the terminal device by combining the channel condition, the channel interference condition, and the precoding condition of the network device.

[0304] Optionally, after the network device determines the resource configuration information, the network device sends the resource configuration information to the terminal device, as described in the method in Figure 7 The method described in

[0305] The network device sends the resource configuration information to the terminal device, or the terminal device receives the resource configuration information sent by the network device.

[0306] Figure 7The terminal device reports the SINR in the manner shown in the embodiment, instead of reporting the CQI value, and there is no need to map the SINR value to obtain the CQI value. Since the CQI value is discrete and the SINR value is continuous, part of information may be lost in the process of mapping the SINR to the CQI value. Therefore, by using the method that the terminal device directly reports the SINR value, the network device can obtain more accurate channel interference, and the accuracy of the resource configuration information issued by the network device for the terminal device is improved.

[0307] Figure 8 is another method for channel state feedback provided by the embodiment.

[0308] S810 and with Figure 5 S510 in the embodiment, which will not be described here for brevity.

[0309] S820, the terminal device determines a first channel feature vector according to the obtained reference signal. The terminal device encodes the first channel feature vector to generate a third channel feature vector, and decodes the third channel feature vector by using an AI decoder to generate a second channel feature vector.

[0310] Specifically, the terminal device encodes the first channel feature vector by using an AI encoder to obtain a third channel feature vector, and decodes the third channel feature vector by using an AI decoder to generate a second channel feature vector, where the second channel feature vector is used to calculate a first CQI.

[0311] Optionally, the AI encoder and / or the AI decoder can be configured by the network device in advance for the terminal device, or can be obtained by the terminal device itself.

[0312] Optionally, the AI decoder is a low-complexity decoder, and the AI decoder is used for determining a precoding matrix assumption in the calculation of the first CQI, so that the network device and the terminal device have the same precoding assumption.

[0313] S830, the terminal device generates a first CQI value according to the second channel feature vector.

[0314] Specifically, the terminal device decodes the third channel feature vector obtained after encoding by using a low-complexity decoder, and generates the first CQI according to the second channel feature vector obtained after decoding and a precoding assumption.

[0315] S840, the terminal device sends the third channel feature vector and the first CQI to the network device. Correspondingly, the network device receives the third channel feature vector and the first CQI sent by the terminal device.

[0316] The network device receives the third channel feature vector and the first CQI sent by the terminal device, further determines the channel state and the interference of the channel, and generates resource configuration information, that is Figure 8 The method further includes:

[0317] S850, the network device generates the resource configuration information according to the third channel feature vector and the first CQI.

[0318] Specifically, the network device further decodes the precoding matrix according to the third channel feature vector sent by the terminal device;

[0319] Then, the network device uses the same low-complexity decoder as the terminal device to decode the obtained third channel feature vector and the first CQI to obtain a rough channel feature vector, and determines the channel interference according to the obtained rough feature vector and the first CQI.

[0320] Optionally, the network device further generates and sends the resource configuration information according to the precoding matrix and the determined channel interference. Figure 8 The method further includes:

[0321] S860, the network device sends the resource configuration information to the terminal device, or the terminal device receives the resource configuration information from the network device.

[0322] The resource configuration information includes the MCS information that the network device needs to allocate for the downlink channel corresponding to the terminal device.

[0323] Figure 8 The first CQI and the third channel feature vector sent by the terminal device to the network device ensure that the terminal device and the network device have the same precoding assumption, and the network device can more accurately determine the channel interference according to the received first CQI and the third channel feature vector obtained after encoding, thereby solving the problem that the network device allocates resources with low matching degree for the terminal device according to the CQI with large error. By maintaining the same precoding assumption between the network device and the terminal device, the network device can obtain accurate channel interference, thereby improving the accuracy of the resource configuration information sent by the network device for the terminal device.

[0324] In the above method embodiment, the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application. And it is possible that not all operations in the above method embodiment need to be performed.

[0325] It should be understood that the terminal device and / or the network device in the above method embodiments can perform some or all of the steps in the examples, and these steps or operations are only examples, and the embodiments of the present application can also include performing other operations or variations of various operations.

[0326] It should also be understood that in various embodiments of the present application, the terms and / or descriptions of different embodiments can be consistent and can be referred to each other if there is no special description and logical conflict, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0327] The above is described in combination with Figures 4-8 The method for feeding back the channel state provided by the embodiments of the present application is described in detail below, and the above is described in combination with Figures 9-10 The device for feeding back the channel state provided by the embodiments of the present application is described in detail.

[0328] The above is described in combination with Figure 9 and Figure 10 The device for feeding back the channel state provided by the embodiments of the present application is described in detail. It should be understood that the description of the device embodiments corresponds to the description of the method embodiments, therefore, the content not described in detail can be referred to the above method embodiments, and some content will not be described again for the sake of brevity.

[0329] The embodiments of the present application can divide the functional modules of the transmitting end device or the receiving end device according to the above method examples, for example, each functional module can be divided according to each function, or two or more functions can be integrated in one processing module. The above integrated module can be realized in the form of hardware or in the form of software functional module. It should be noted that the division of the modules in the embodiments of the present application is illustrative, and is only a logical function division, and another division mode can be used in actual implementation. The following will be described taking the division of each functional module corresponding to each function as an example.

[0330] Figure 9 is an example of the information transmission device 900 provided by the present application. The above Figures 4 to 8 Any device involved in any method in the above Figure 9 The device for feeding back the channel state shown in

[0331] It should be understood that the information transmission device 900 can be an entity device, a component (for example, an integrated circuit, a chip, etc.) of an entity device, or a functional module in an entity device.

[0332] As Figure 9As shown, the device 900 for channel state feedback includes one or more processors 910. The processor 910 can invoke an interface to implement the receiving and sending functions. The interface can be a logical interface or a physical interface, which is not limited. For example, the interface can be a transceiver circuit, an input / output interface, or an interface circuit. The transceiver circuit, the input / output interface, or the interface circuit for implementing the receiving and sending functions can be separate or integrated together. The transceiver circuit or the interface circuit described above can be used for reading and writing of codes / data, or the transceiver circuit or the interface circuit described above can be used for transmission or transfer of signals.

[0333] Optionally, the interface can be implemented through a transceiver. Optionally, the information transmission device 900 can further include a transceiver 930. The transceiver 930 can also be referred to as a transceiving unit, a transceiver, a transceiving circuit, etc., for implementing the transceiving function.

[0334] Optionally, the device 900 for channel state feedback can further include a memory 920. The specific deployment position of the memory 920 is not limited in the embodiments of the present application. The memory can be integrated in the processor, or independent of the processor. For the case that the device 900 for channel state feedback does not include the memory, the device 900 for channel state feedback has the processing function, and the memory can be deployed at other positions (such as a cloud system).

[0335] The processor 910, the memory 920, and the transceiver 930 communicate with each other through internal connection paths to transfer control and / or data signals.

[0336] It can be understood that, although not shown, the device 900 for channel state feedback can further include other apparatuses, such as an input apparatus, an output apparatus, a battery, etc.

[0337] Optionally, in some embodiments, the memory 920 can store execution instructions for executing the method of the embodiments of the present application. The processor 910 can execute the instructions stored in the memory 920 to complete the steps of the method execution shown below in combination with other hardware (such as the transceiver 930). The specific working process and beneficial effects can be referred to the description in the method embodiments.

[0338] The method disclosed by the embodiments of the present application can be applied to the processor 910 or implemented by the processor 910. The processor 910 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the method can be completed by the integrated logic circuit or the instruction in the form of software of the hardware in the processor. The processor mentioned above can be a general processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor and the like. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory (RAM), a flash memory, a read-only memory (ROM), a programmable read-only memory or an electrically erasable programmable memory, a register or the like. The storage medium is located in the memory, and the processor reads the instruction in the memory and combines the hardware to complete the steps of the above method.

[0339] It is to be understood that the memory 920 can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. Where the nonvolatile memory is a read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. The volatile memory can be a random access memory (RAM) used as external cache memory. By way of example, and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM). It is to be noted that the memory described herein is intended to include, without being limited to, these and any other suitable types of memory.

[0340] Figure 10 is a schematic block diagram of an apparatus 1000 for feeding back channel state provided by the present application.

[0341] Optionally, the apparatus 1000 for feeding back channel state can be a general purpose computer device or a chip in a general purpose computer device, and the embodiments of the present application do not limit this. As shown in the figure, the apparatus for feeding back channel state includes a processing unit 1010 and a transceiver unit 1020. Figure 10

[0342] Specifically, the apparatus 1000 for feeding back channel state can be any device involved in the present application, and can implement the functions that the device can implement. It should be understood that the apparatus 1000 for feeding back channel state can be an entity device, a component of an entity device (for example, an integrated circuit, a chip, etc.), and also can be a functional module in an entity device.

[0343] In a possible design, the apparatus 1000 for feeding back channel state can be a terminal device (such as the terminal device 120) in the above method embodiments, and also can be a chip for implementing the functions of the terminal device (such as the terminal device 120) in the above method embodiments.​

[0344] For example, the transceiver is configured to receive first indication information from the network device, the first indication information comprising channel quality indicator (CQI) information or signal-to-interference ratio (SINR) information; and the processor is configured to reduce a first value of the first information to a second value according to the first indication information.

[0345] Optionally, the transceiver is further configured to receive a second probability value from the network device; and the processor is further configured to determine the first probability value according to the first indication information, and determine the first value of the first information to be reduced to the second value according to the first probability value and the second probability value.

[0346] It should also be understood that when the apparatus 1000 for feeding back channel state is a terminal device (e.g., the terminal device 120), the transceiver 1020 in the apparatus 1000 for feeding back channel state can be implemented through a communication interface (e.g., a transceiver or an input / output interface), and the processor 1010 in the apparatus 1000 for feeding back channel state can be implemented through at least one processor, which can correspond to the processor 910 shown in FIG. 9, for example. Figure 9

[0347] Optionally, the apparatus 1000 for feeding back channel state can further include a storage unit, which can be configured to store instructions or data, and the processor can invoke the instructions or data stored in the storage unit to implement corresponding operations.

[0348] It should be understood that the specific process in which each unit performs the corresponding steps described above has been described in detail in the method embodiments, and thus will not be described here again for the sake of brevity.

[0349] In another possible design, the apparatus 1000 for feeding back channel state can be the network device (e.g., the network device 110) in the above method embodiments, or can be a chip for implementing the functions of the network device (e.g., the network device 110) in the above method embodiments.

[0350] For example, the transceiver is configured to send first indication information to the terminal device, the first indication information comprising channel quality indicator (CQI) information or signal-to-interference ratio (SINR) information.

[0351] It should also be understood that when the apparatus 1000 for feeding back channel state is a network device 101, the transceiver 1020 in the apparatus 1000 for feeding back channel state can be implemented through a communication interface (e.g., a transceiver or an input / output interface), which can correspond to the communication interface 930 shown in FIG. 9, for example, and the processor 1010 in the apparatus 1000 for feeding back channel state can be implemented through at least one processor, which can correspond to the processor 910 shown in FIG. 9, for example. Figure 9 Figure 9 ​​​

[0352] Optionally, the apparatus 1000 for feeding back channel state further includes a storage unit, which can be used to store instructions or data, and the processing unit can invoke the instructions or data stored in the storage unit to implement corresponding operations.

[0353] It should be understood that the specific process of each unit performing the corresponding steps described above has been described in detail in the method embodiments described above, and for the sake of brevity, it will not be repeated here.

[0354] It should be understood that the specific process of each unit performing the corresponding steps described above has been described in detail in the method embodiments described above, and for the sake of brevity, it will not be repeated here.

[0355] In addition, in the present application, the apparatus 1000 for feeding back channel state is presented in the form of functional modules. The "module" here can refer to an application specific integrated circuit (ASIC), a circuit, a processor and a memory executing one or more software or firmware programs, an integrated logic circuit, and / or other devices that can provide the above functions. In a simple embodiment, those skilled in the art can think that the apparatus 1000 can adopt the form shown in the figure. The processing unit 1010 can be implemented by the processor 910 shown in the figure. Optionally, if the computer device shown in the figure includes the memory 900, the processing unit 1010 can be implemented by the processor 910 and the memory 900. The transceiving unit 1020 can be implemented by the transceiver 930 shown in the figure. The transceiver 930 includes a receiving function and a sending function. Specifically, the processor is implemented by executing the computer program stored in the memory. Optionally, when the apparatus 1000 is a chip, then the functions and / or implementation processes of the transceiving unit 1020 can also be implemented by pins or circuits, etc. Optionally, the memory can be a storage unit in the chip, such as a register, a cache, etc., and the storage unit can also be a storage unit outside the chip in the apparatus for feeding back channel state, such as the memory 920 shown in the figure, or also can be a storage unit deployed in other systems or devices, not in the computer device. Figure 10 Figure 9 Figure 9 Figure 9 Figure 9

[0356] ​​​​​Various aspects or features of this application can be implemented as methods, apparatus, or articles of manufacture using standard programming and / or engineering techniques. For example, computer-readable media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes), optical discs (e.g., compact discs (CDs), digital versatile discs (DVDs), etc.), smart cards, and flash memory devices (e.g., erasable programmable read-only memory (EPROMs), cards, sticks, or key drives, etc.). Furthermore, the various storage media described herein may represent one or more devices and / or other machine-readable media for storing information. The term "machine-readable medium" may include, but is not limited to, various other media capable of storing, containing, and / or carrying instructions and / or data.

[0357] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: a computer program or a set of instructions, which, when executed on a computer, causes the computer to perform... Figures 4 to 8 The method of any one of the embodiments shown.

[0358] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing a program or a set of instructions, which, when executed on a computer, causes the computer to perform... Figures 4 to 8 The method of any one of the embodiments shown.

[0359] According to the method provided in the embodiments of this application, this application also provides a communication system, which includes the aforementioned apparatus or device.

[0360] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0361] It should also be understood that the term "and / or" as used herein merely describes associated objects, and can exist in three forms: A and / or B, A or B, and A and B. In addition, the character " / " generally represents an "or" relationship between the front and rear associated objects.

[0362] It should also be understood that the terms "first", "second" and the like introduced in the embodiments of the present application are only intended to distinguish different objects, for example, to distinguish different "information", or "devices", or "units". The specific objects and the corresponding relationship between different objects should be determined according to their functions and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0363] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0364] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for feedback channel state, applied to a terminal device or a chip of a terminal device, characterized in that, The method includes: Receive the first instruction information from the network device; Based on the first indication information, the first value of the first information is reduced to a second value, and the second value is used to determine the channel state feedback information. The first indication information includes a backoff value, which represents the difference between the first value and the second value. The first information includes channel state indication (CQI) information or signal-to-interference-plus-noise ratio (SINR) information.

2. The method according to claim 1, characterized in that, The method further includes: Receive a second probability value from the network device, the second probability value being used to determine whether to reduce the first value of the first information to the second value; Determining whether to reduce the first value of the first information to a second value based on the first indication information includes: A first probability value is determined based on the first indication information; Based on the first probability value and the second probability value, the first value of the first information is determined to be reduced to the second value.

3. The method according to claim 2, characterized in that, Determining to reduce the first value of the first information to a second value based on the first probability value and the second probability value includes: Based on the first probability value, the second probability value, and the first threshold, the first value of the first information is determined to be reduced to the second value.

4. The method according to any one of claims 1 to 3, characterized in that, The first indication information also includes at least one of the following: The cosine similarity of the AI ​​model and the normalized mean square error of the AI ​​model. The AI ​​model is used to determine the channel state feedback information, and the cosine similarity and normalized mean square error of the AI ​​model are used to calculate the backoff value.

5. The method according to claim 4, characterized in that, The square of the cosine similarity of the AI ​​model is used to calculate the backoff value.

6. The method according to any one of claims 1 to 3, characterized in that, The first information is the CQI information, the first value is the first CQI value, and the second value is the second CQI value. Reducing the first value of the first information to a second value according to the first instruction information includes: The first CQI value is determined based on the first SINR value, wherein the first SINR value is determined based on the channel feature vector, and the channel feature vector is used to represent downlink channel information; The first CQI value of the CQI information is reduced to the second CQI value according to the first instruction information.

7. The method according to any one of claims 1 to 3, characterized in that, The first information is the SINR information, where the first value is a first SINR value and the second value is a second SINR value. Reducing the first value of the first information to a second value according to the first instruction information includes: According to the first instruction information, the first SINR value of the SINR information is reduced to the second SINR value; After reducing the first value of the first information to the second value according to the first indication information, the method further includes: The second CQI value is determined based on the second SINR value.

8. The method according to any one of claims 1 to 3, characterized in that, The first information is the SINR information, where the first value is a first SINR value and the second value is a second SINR value. Reducing the first value of the first information to a second value according to the first instruction information includes: According to the first instruction information, the first SINR value of the SINR information is reduced to the second SINR value; After reducing the first value of the first information to the second value according to the first indication information, the method further includes: A third SINR value is generated by quantizing the second SINR value according to the second indication information, wherein the second indication information comes from the network device and is used to indicate the quantization method of the SINR information; The channel state feedback information is determined based on the third SINR value.

9. The method according to claim 8, characterized in that, The second instruction information includes at least one of the following: Quantization bit count, quantization range, or quantization step size.

10. The method according to claim 6, characterized in that, The method further includes: Receive reference signal CSI-RS from the network device; The channel feature vector is determined based on the CSI-RS.

11. The method according to any one of claims 1 to 3, characterized in that, After reducing the first value of the first information to the second value according to the first indication information, the method further includes: The channel state feedback information is determined based on the second value; Send the channel status feedback information to the network device.

12. The method according to any one of claims 1 to 3, characterized in that, The first indication information is carried in the Radio Resource Control (RRC) configuration information.

13. A method for feedback of channel status, applied to a network device or a chip in a network device, characterized in that, The method includes: Determine first indication information, which is used to instruct the terminal device to reduce the first value of the first information to a second value; Send the first instruction information to the terminal device. The first indication information includes a backoff value, which represents the difference between the first value and the second value. The second value is used to determine channel state feedback information. The first information includes channel state indication (CQI) information or signal-to-interference-plus-noise ratio (SINR) information.

14. The method according to claim 13, characterized in that, The method further includes: A second probability value is sent to the terminal device, the second probability value being used to determine whether to reduce the first value of the first information to the second value.

15. The method according to claim 13 or 14, characterized in that, The first indication information also includes at least one of the following: The cosine similarity of the AI ​​model and the normalized mean square error of the AI ​​model. The AI ​​model is used to determine the channel state feedback information, and the cosine similarity and normalized mean square error of the AI ​​model are used to calculate the backoff value.

16. The method according to claim 15, characterized in that, The square of the cosine similarity of the AI ​​model is used to calculate the backoff value.

17. The method according to claim 13 or 14, characterized in that, The first information is the CQI information, where the first value is a first CQI value and the second value is a second CQI value. The first CQI value is obtained based on a first SINR value, and the second CQI value is used to determine the channel state feedback information. The first SINR value is determined based on the channel feature vector, which is used to represent downlink channel information.

18. The method according to claim 13 or 14, characterized in that, The first information is the SINR information, where the first value is a first SINR value and the second value is a second SINR value. The first indication information is used to instruct the terminal device to reduce the first value of the first information to a second value, including: The first indication information is used to indicate that the first SINR value of the SINR information be reduced to the second SINR value. The second SINR value is used to determine the second CQI value, and the second CQI value is used to determine the channel state feedback information.

19. The method according to claim 13 or 14, characterized in that, The first information is the SINR information, where the first value is a first SINR value and the second value is a second SINR value. The first indication information is used to instruct the terminal device to reduce the first value of the first information to a second value, including: The first indication information is used to indicate that the first SINR value of the SINR information be reduced to the second SINR value. The method further includes: Send a second indication message to the terminal device, the second indication message being used to indicate the quantization method of the SINR information.

20. The method according to claim 19, characterized in that, The second instruction information includes at least one of the following: Quantization bit count, quantization range, or quantization step size.

21. The method according to claim 17, characterized in that, The method further includes: A reference signal CSI-RS is sent to the terminal device. The CSI-RS is used to determine the channel feature vector, which is used to represent downlink channel information.

22. An apparatus for feedback of channel state, characterized in that, include: A transceiver unit is configured to receive first indication information from a network device, wherein the first indication information is used to indicate that a first value of the first information is reduced to a second value. A processing unit is configured to execute a computer program stored in a memory to cause the apparatus to perform the method according to any one of claims 1 to 21.

23. A communication device, characterized in that, The device includes a processor connected to a memory for storing a computer program, the processor executing the computer program stored in the memory to cause the device to perform the method as described in any one of claims 1 to 21.

24. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions that, when executed by a processor, cause the method as described in any one of claims 1 to 21 to be performed.

25. A computer program product containing instructions, characterized in that, When the instructions are executed on a computer, the method described in any one of claims 1 to 21 is performed.

Images