Method, apparatus and storage medium for determining a compression model for compressing channel state information

Abstract

The present disclosure relates to a method, device and storage medium for determining a compression model for compressing channel state information. The method comprises: receiving, by a terminal, model information sent by a network device, the model information comprising first model parameters of a decompression model used by the network device to decompress channel state information sent by the terminal; and determining, by the terminal, a compression model used by the terminal to compress channel state information according to the model information. The terminal receives the model information sent by the network device, and obtains the compression model used to compress channel state information according to the model information. Since the model information comprises the first model parameters of the decompression model of the network device, the terminal can obtain the compression model used in combination with the decompression model of the network device according to the first model parameters and private data of the terminal manufacturer or chip manufacturer. In this process, the network device does not need to know the parameters of the compression model deployed by the terminal, thereby ensuring the privacy of the compression model.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of communication, and in particular, to a method and device for determining a compression model for compressing channel state information and a storage medium. BACKGROUND

[0002] In the field of wireless communication, channel state information (CSI) is the channel property of a communication link. It describes the attenuation factor of a signal on each transmission path, i.e., the value of each element in the channel gain matrix, such as signal scattering (Scattering), environmental attenuation (fading, multipath fading or shadowing fading), distance attenuation (power decay of distance), etc. CSI can enable the communication system to adapt to the current channel conditions and provide high reliability and high rate communication in a multi-antenna system.

[0003] In the related art, AI (Artificial Intelligence) model-based CSI enhancement has become an industry trend. However, the AI model deployment and transmission method for CSI enhancement in the related art cannot meet the needs of different manufacturers for model parameter and model data privacy. SUMMARY

[0004] To overcome the problems in the related art, the present disclosure provides a method, device and storage medium for determining a compression model for compressing channel state information. The compression model is used for AI-based reporting of channel state information.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining a compression model for compressing channel state information is provided, applied to a terminal, and the method comprises:

[0006] receiving model information sent by a network device, the model information comprising first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal;

[0007] determining, according to the model information, a compression model for compressing channel state information by the terminal.

[0008] According to a second aspect of an embodiment of the present disclosure, a method for determining a compression model for compressing channel state information is provided, applied to a network device, and the method comprises:

[0009] transmit model information, the model information including first model parameters of a decompression model used by the network device to decompress channel state information transmitted by the terminal, the model information being used by the terminal to determine a compression model used to compress channel state information.

[0010] According to a third aspect of embodiments of the present disclosure, a device for determining a compression model used to compress channel state information is provided, and applied to a terminal, the device comprising:

[0011] a receiving module configured to receive model information transmitted by a network device, the model information including first model parameters of a decompression model used by the network device to decompress channel state information transmitted by the terminal;

[0012] a determining module configured to determine, according to the model information, a compression model used by the terminal to compress channel state information.

[0013] According to a fourth aspect of embodiments of the present disclosure, a device for determining a compression model used to compress channel state information is provided, and applied to a network device, the device comprising:

[0014] a transmitting module configured to transmit model information, the model information including first model parameters of a decompression model used by the network device to decompress channel state information transmitted by the terminal, the model information being used by the terminal to determine a compression model used to compress channel state information.

[0015] According to a fifth aspect of embodiments of the present disclosure, a device for determining a compression model used to compress channel state information is provided, and comprising:

[0016] a processor;

[0017] a memory for storing processor-executable instructions;

[0018] wherein the processor is configured to:

[0019] receive model information transmitted by a network device, the model information including first model parameters of a decompression model used by the network device to decompress channel state information transmitted by the terminal;

[0020] determine, according to the model information, a compression model used by the terminal to compress channel state information.

[0021] According to a sixth aspect of embodiments of the present disclosure, a device for determining a compression model used to compress channel state information is provided, and comprising:

[0022] a processor;

[0023] a memory for storing processor-executable instructions;

[0024] The processor is configured to:

[0025] send model information, the model information comprising first model parameters of a decompression model used by the network device to decompress channel state information sent by the terminal, the model information being used by the terminal to determine a compression model used to compress the channel state information.

[0026] According to a seventh aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, and the computer readable storage medium stores computer program instructions. The computer program instructions are executed by a processor to implement the steps of the method in any one of the first aspect of the present disclosure, or the steps of the method in any one of the second aspect of the present disclosure.

[0027] In the technical solution provided by the embodiments of the present disclosure, the terminal accepts the model information sent by the network device, and obtains the compression model used to compress the CSI according to the model information. Since the model information comprises the first model parameters of the decompression model of the network device, the terminal can obtain the compression model used in combination with the decompression model of the network device according to the first model parameters and the private data of the terminal manufacturer or the chip manufacturer, and the network device does not need to know the parameters of the compression model deployed by the terminal in the process, thereby ensuring the privacy of the compression model.

[0028] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above described and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following detailed description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0030] Figure 1A is a schematic diagram of a network system architecture in the related art.

[0031] Figure 1A is a schematic diagram of another network system architecture in the related art.

[0032] Figure 2 is a flowchart illustrating a method of determining a compression model used to compress channel state information according to an exemplary embodiment.

[0033] Figure 3 is a flowchart illustrating a method of determining a compression model used to compress channel state information according to an exemplary embodiment.

[0034] Figure 4 is a flowchart illustrating a method of determining a compression model used to compress channel state information according to an exemplary embodiment.

[0035] Figure 5 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0036] Figure 6 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0037] Figure 7 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0038] Figure 8 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0039] Figure 9 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0040] Figure 10 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0041] Figure 11 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0042] Figure 12 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0043] Figure 13 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0044] Figure 14 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0045] Figure 15 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0046] Figure 16 is a flowchart of a method of determining a compression model for compressing channel state information according to an example embodiment.

[0047] Figure 17is a flowchart of a method of determining a compression model for compressing channel state information according to an exemplary embodiment.

[0048] Figure 18 is a flowchart of a method of determining a compression model for compressing channel state information according to an exemplary embodiment.

[0049] Figure 19 is a flowchart of a method of determining a compression model for compressing channel state information according to an exemplary embodiment.

[0050] Figure 20 is a flowchart of a method of determining a compression model for compressing channel state information according to an exemplary embodiment.

[0051] Figure 21 is an interaction diagram of a method of determining a compression model for compressing channel state information according to an exemplary embodiment.

[0052] Figure 22 is a block diagram of an apparatus for determining a compression model for compressing channel state information according to an exemplary embodiment.

[0053] Figure 23 is a block diagram of an apparatus for determining a compression model for compressing channel state information according to an exemplary embodiment.

[0054] Figure 24 is a block diagram of a terminal according to an exemplary embodiment.

[0055] Figure 25 is a block diagram of a network device according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] Embodiments of the present disclosure are described in detail below with reference to the attached drawings, which are meant to be exemplary and not limiting.

[0057] In the field of wireless communication, channel state information (CSI) is the channel property of a communication link. It describes the attenuation factor of a signal on each transmission path, i.e., the value of each element in the channel gain matrix, such as signal scattering (Scattering), environmental attenuation (fading, multipath fading or shadowing fading), power decay of distance, etc. CSI can enable the communication system to adapt to the current channel conditions and provide high reliability and high rate communication in a multi-antenna system.

[0058] In the related art, AI (Artificial Intelligence) model-based CSI enhancement has become an industry trend. However, the AI model deployment and transmission method for CSI enhancement in the related art cannot meet the needs of different manufacturers for model parameter and model data privacy.

[0059] To solve the above problems, the embodiments of the present disclosure provide a method, device and storage medium for determining a compression model for compressing channel state information.

[0060] First, the implementation environment of the embodiments of the present disclosure is introduced as follows:

[0061] The embodiments of the present disclosure can be applied to a 4G (fourth generation mobile communication system) evolution system, such as a long term evolution (LTE) system, or a 5G (fifth generation mobile communication system) system, such as a new radio access technology (New RAT) access network, a cloud radio access network (CRAN), and other communication systems.

[0062] Figure 1A An exemplary system architecture diagram to which the embodiments of the present disclosure are applicable is shown. It should be understood that the embodiments of the present disclosure are not limited to Figure 1A In the system shown, in addition, Figure 1A The device in the system can be hardware, or functionally divided software, or a combination of the two. For example, Figure 1A As shown, the system architecture provided by the embodiments of the present disclosure includes a terminal, a base station, a mobility management network element, a session management network element, a user plane network element, and a data network (DN). The terminal communicates with the DN through the base station and the user plane network element.

[0063] Among them Figure 1AThe network element shown in the middle can be a network element in a 4G architecture, and can also be a network element in a 5G architecture.

[0064] A data network (DN) provides a data transmission service for a user, and can be a Protocol Data Unit (PDN) network, such as an internet, an IP Multi-media Service (IMS), and the like.

[0065] Referring to Figure 1B The system architecture diagram of 5G shown in the middle: the mobility management network element can include an access and mobility management function (AMF) in 5G. The mobility management network element is responsible for access and mobility management of a terminal in a mobile network. The AMF is responsible for terminal access and mobility management, NAS message routing, session management function (SMF) selection, and the like. The AMF can serve as an intermediate network element to transmit session management messages between the terminal and the SMF.

[0066] The session management network element is responsible for forwarding path management, such as issuing a message forwarding policy to a user plane network element, and instructing the user plane network element to process and forward messages according to the message forwarding policy. The session management network element can be an SMF in 5G (as shown in the middle), which is responsible for session management, such as session creation / modification / deletion, user plane network element selection, and allocation and management of user plane tunnel information. Figure 1B

[0067] The user plane network element can be a user plane function (UPF) in a 5G architecture, as shown in the middle. The UPF is responsible for message processing and forwarding. Figure 1B

[0068] The system architecture provided by the embodiments of the present disclosure can further include a data management network element for processing terminal device identification, access authentication, registration, and mobility management. In a 5G communication system, the data management network element can be a unified data management (UDM) network element.

[0069] ​​The system architecture provided by the embodiments of the present disclosure can further include a policy control function entity (PCF) or a policy and charging control function entity (PCRF). The PCF or PCRF is responsible for policy control decision and flow-based charging control.

[0070] The system architecture provided by the embodiments of the present disclosure can further include a network storage network element for maintaining real-time information of all network function services in the network. In the 5G communication system, the network storage network element can be a network repository function (NRF) network element. The network repository network element can store information of many network elements, such as information of SMF, information of UPF, information of AMF, etc. The network elements such as AMF, SMF, and UPF in the network can be connected to the NRF. On the one hand, the network elements can register their own network element information to the NRF. On the other hand, other network elements can obtain the information of the network elements that have been registered from the NRF. Other network elements (such as AMF) can request the NRF to obtain optional network elements according to the network element type, data network identifier, and unknown area information. If the domain name system (DNS) server is integrated in the NRF, the corresponding selection function network element (such as AMF) can request the NRF to obtain the other network elements (such as SMF) to be selected.

[0071] The base station, as a specific implementation form of an access network (AN), can also be referred to as an access node. If it is in the form of wireless access, it is referred to as a radio access network (RAN), such as a 5G RAN. Figure 1BThe access node can be a base station in a global system for mobile communication (GSM) system or a code division multiple access (CDMA) system, a base station (NodeB) in a wideband code division multiple access (WCDMA) system, an evolved node B (eNB or eNodeB) in an LTE system, or a base station device in a 5G network, a small base station device, a wireless access node (WiFi AP), a worldwide interoperability for microwave access base station (WiMAX BS), and the like, and the disclosure is not limited thereto.

[0072] The terminal can also be referred to as an access terminal, a user equipment (UE), a subscriber unit, a subscriber station, a mobile station, a mobile, a remote station, a remote terminal, a mobile device, a user terminal, a wireless communication device, a user agent, or a user device, and the like. Figure 1B The terminal can 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 wireless communication function, a computing device, or other processing device connected to a wireless modem, an in-vehicle device, a wearable device, an Internet of Things terminal device, such as a fire detection sensor, a smart water meter / electricity meter, a factory monitoring device, and the like.

[0073] The above functions can be network elements in a hardware device, software functions running on a dedicated hardware, or virtualized functions instantiated on a platform (e.g., a cloud platform).

[0074] In the embodiments of the present disclosure, a scheme is adopted, in which a model is deployed on the base station and the terminal respectively, the CSI is compressed by a compression model and executed on the terminal side, the CSI is decompressed by a decompression model and executed on the base station side, and the decompression model is used in combination with the compression model. The model parameters (such as the first model parameter, the second model parameter and the third model parameter described below) involved in the embodiments of the present disclosure can include configuration variables inside the model and / or configuration variables outside the model (i.e., model hyperparameters). The embodiments of the present disclosure do not limit the types of the decompression model and the compression model. For example, for a neural network model, the variables configured inside the model can include a calculation parameter matrix of each neuron node, and the hyperparameters can be a learning step length of the trained neural network. For a support vector machine, the variables configured inside the model can be support vectors, and the hyperparameters can be a sigma parameter of the support vector machine. Taking the neural network as an example, in a possible implementation manner, the model parameters in the embodiments of the present disclosure can specifically include at least one of the following parameters: a model type, a learning step length, a calculation parameter matrix of each neuron node, a padding value used for padding the calculation parameter matrix, a bias of each neuron node, and an activation function of each neuron node.

[0075] Figure 2 FIG. 1 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a terminal. Figure 2 As shown in FIG. 1, the method includes the following steps.

[0076] S201, the terminal receives model information sent by a network device, the model information including first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0077] S202, the terminal determines, according to the model information, a compression model of the terminal for compressing the channel state information.

[0078] In the embodiments of the present disclosure, the first model parameters are used to indicate a compression model of the terminal corresponding to the decompression model.

[0079] As shown in FIG. 2, the network device can be an access network device (such as a base station) as shown in FIG. 2, or can be another network logical entity in a core network. It should be understood that the communication network includes an access network (such as the base station shown in FIG. 2), a bearer network and a core network (such as the mobility management network element, the session management network element and the NRF, the AMF shown in FIG. 2). Figure 1A Figure 1A Figure 1A Figure 1B The first model parameters of the decompression model can be obtained by the base station through training, and the training manner of the decompression model is not limited in the present disclosure.

[0080] ​​​It can be understood that the first model parameter is used to indicate a compression model corresponding to the decompression model; the terminal can obtain the compression model used for compressing the CSI which can be used in combination with the decompression model based on at least the first model parameter of the decompression model. For example, the terminal can train the compression model corresponding to the decompression model based on the first model parameter and the private data collected by the terminal.

[0081] In an example, the model information can be sent by a base station through RRC signaling, and the network device can be the base station. Alternatively, in another example, the model information can also be sent by an AMF (Access and Mobility Management Function) in a core network through NAS (Non-access stratum) layer signaling, and the network device can be the AMF.

[0082] In an example, the model information can be sent in a case where the decompression model deployed in the network device is changed, can be sent in response to the terminal accessing the base station, or can be periodically sent by the network device at a certain period. When the compression model in the terminal is changed, the decompression model deployed in the network device is not affected, that is, the decompression model of the network device is not affected by the compression model of the terminal, thereby effectively ensuring that the complexity of the decompression model deployed in the network device does not increase.

[0083] In another example, the compression model used by the terminal to compress the channel state information satisfies at least one of the following model performance conditions: the mean square error or the normalized mean square error of the compression model used in combination with the decompression model is less than a first preset threshold; the cosine similarity of the compression model used in combination with the decompression model is greater than a second preset threshold; the square of the cosine similarity of the compression model used in combination with the decompression model is greater than a third preset threshold; and the signal-to-noise ratio of the compression model used in combination with the decompression model is greater than a fourth preset threshold.

[0084] In some possible implementations, the compression model used by the terminal to compress the channel state information needs to satisfy multiple conditions above at the same time. If the compression model cannot satisfy any one of the multiple conditions, the terminal can further adjust the parameters of the compression model, or cause the network device to re-send the model information of the compression model, thereby ensuring that the compression model obtained based on the model information sent by the terminal can reliably compress the channel state information.

[0085] In the embodiments of the present disclosure, the terminal accepts the model information sent by the network device, and obtains the compression model for compressing the channel state information according to the model information. Since the first model parameter of the decompression model of the network device is included in the model information, the terminal can obtain the compression model for joint use with the decompression model of the network device according to the first model parameter and the private data of the terminal manufacturer or chip manufacturer, and the network device does not need to know the parameters of the compression model deployed by the terminal in the process, thereby ensuring the privacy of the compression model.

[0086] Figure 3 is a flow chart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a terminal, as shown in Figure 3 The method comprises:

[0087] S301, the terminal receives model information sent by a network device, the model information comprising a first model parameter of a decompression model and a second model parameter of a compression model corresponding to the decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0088] S302, the terminal determines a compression model for compressing channel state information according to the model information.

[0089] It can be understood that due to the performance differentiation of different terminal manufacturers or terminal chip manufacturers, the terminal itself has different capabilities (the capabilities of the terminal itself may, for example, include the hardware capabilities of the terminal, such as its computing power and the like, and the capability information may also include the current state information of the terminal, such as load information and the like), in step S302, the terminal can determine whether to obtain the compression model according to the first model parameter or to obtain the compression model according to the second model parameter according to its own capabilities or other parameters. In another example, the model information further comprises configuration information of the network device (such as a base station), and the terminal can determine whether to use the second model parameter or to obtain the compression model according to the first model parameter according to the configuration information. The configuration information can at least be used to indicate whether the terminal directly uses the compression model corresponding to the second model parameter or whether the second model parameter is adjusted.

[0090] With the present solution, by receiving the model information comprising the first model parameter of the decompression model and the second model parameter of the compression model corresponding to the decompression model, the terminal can obtain the compression model according to the first model parameter, or can obtain the compression model according to the second model parameter, and can ensure that the terminals of different terminal manufacturers or terminal chip manufacturers can effectively obtain the compression model.

[0091] Figure 4 ​is a flow chart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a terminal, such as Figure 4 As shown, the method comprises:

[0092] S401, the terminal receives model information sent by the network device, the model information comprising first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0093] S402, the terminal trains a compression model corresponding to the decompression model according to the first model parameters.

[0094] In step S402, the terminal can train the compression model based on at least the private data of the terminal manufacturer or chip manufacturer and the first model parameters, and the specific training method of the compression model is not limited in the present disclosure.

[0095] In an example, the model information further comprises second model parameters of the compression model corresponding to the decompression model. Before step S402 is performed, the terminal can determine whether to perform step S402 according to its capability information, i.e., the terminal determines whether it has the capability to train the compression model corresponding to the decompression model according to the first model parameters according to its capability information. The capability information of the terminal can include the hardware capability of the terminal, such as its computing power and the like, and the capability information can also include the current state information of the terminal, such as load information and the like.

[0096] In another example, the terminal can also determine whether to perform step S402 according to the configuration information of the base station, i.e., the terminal determines whether it needs to train the compression model corresponding to the decompression model according to the first model parameters according to the base station configuration information. In the case where the model information further comprises second model parameters of the compression model corresponding to the decompression model, the configuration information can also indicate whether the terminal needs to directly use the second model parameters. In the absence of the base station configuration information, the terminal can decide whether to directly use the second model parameters or to obtain the compression model according to the first model parameters according to its own capability or other information.

[0097] In another example, if the compression model obtained according to step S402 does not meet any or at least one of the model performance conditions described in the above embodiments, the terminal can retrain the compression model corresponding to the decompression model according to the first model parameters.

[0098] According to the scheme, the terminal receives the first model parameter in the model information sent by the network device, and trains the compression model for compressing the channel state information according to the first model parameter. Since the training process of the compression model and / or the compression model parameter is transparent to the network device, the privacy of the compression model is effectively ensured.

[0099] Figure 5 is a flow chart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a terminal, as shown in Figure 5 The method comprises:

[0100] S501, the terminal receives model information sent by the network device, the model information comprising a first model parameter of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0101] S502, the terminal sends an acquisition request to the server and acquires a compression model corresponding to the decompression model sent by the server.

[0102] Specifically, the training of the model relies on the corresponding server on the terminal side, such as the server of a chip company. After the terminal acquires the first model parameter, the terminal sends the first model parameter to the corresponding server on the terminal side. After the server completes the training of the compression model, the server sends the compression model to the terminal.

[0103] The server can be a third-party server, a server on the network side, such as a server in the core network, or a server on the terminal side, such as a server of a chip manufacturer or a server of a terminal manufacturer, and the present disclosure does not limit the server.

[0104] It can be understood that the acquisition request at least comprises the first model parameter of the decompression model. The acquisition request can also comprise training data for training the compression model, or the server stores the training data for training the compression model.

[0105] In an example, the terminal can also train a compression model corresponding to the decompression model locally on the terminal. The terminal sends an acquisition request to the server to request the server to train the compression model, or the training locally on the terminal can be determined according to the capability information of the terminal. For example, if the current load of the terminal is greater than a preset threshold, the terminal can send an acquisition request to the server to request the server to train the compression model. If the current load of the terminal is less than the preset threshold, the terminal can train locally.

[0106] In another example, the model information further includes second model parameters of a compression model corresponding to the decompression model. Before step S502 is performed, the terminal can determine whether to perform step S502 according to its capability information, that is, the terminal determines whether the terminal has the capability to train the compression model corresponding to the decompression model according to the first model parameters according to its capability information. If the terminal does not have the capability to train the compression model corresponding to the decompression model according to the first model parameters, the terminal can obtain the compression model according to the second model parameters. In addition, whether to perform S502 can also be determined according to the private needs of the terminal. For example, if the terminal has no private needs, the terminal can not perform step S502, but directly obtain the compression model according to the second model parameters.

[0107] In another example, the terminal can determine whether to directly use the second model parameters or obtain the compression model according to the first model parameters according to the configuration information of the base station (which can be included in the model information). In the absence of the configuration information of the base station, the terminal can determine whether to directly use the second model parameters or obtain the compression model according to the first model parameters according to its own capability or other information.

[0108] In yet another example, if the compression model obtained according to steps S501 and S502 does not meet any or at least one of the model performance conditions described in the above embodiments, the terminal can again send an acquisition request to the server and obtain an updated compression model corresponding to the decompression model.

[0109] With the above scheme, by sending an acquisition request to the server to make the server train the compression model, so that the terminal obtains the compression model for compressing the channel state information, the terminal can train the compression model in the case that the terminal cannot train the compression model locally or the terminal temporarily cannot support the training of the compression model, which effectively ensures that the terminal of different terminal manufacturers or chip manufacturers can obtain the compression model through training while ensuring the privacy of the compression model.

[0110] Figure 6 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a terminal, as shown in Figure 6 The method includes:

[0111] S601, the terminal receives model information sent by a network device, the model information at least including second model parameters of a compression model corresponding to a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0112] S602, the terminal optimizes the second model parameters to obtain third model parameters.

[0113] S603, the terminal establishes a compression model for compressing channel state information according to the third model parameter.

[0114] In step S602, the second model parameter can be adjusted based on private data of a terminal manufacturer or a chip manufacturer in the terminal, and the adjustment manner of the second model parameter in the disclosure is not limited.

[0115] In an example, step S602 and step S603 can be determined by the terminal according to its capability information, for example, in a case where it is determined that the capability information of the terminal represents that the computing power of the terminal does not have the ability to train a compression model, but has the ability to adjust the second model parameter, then the step S602 and the step S603 can be executed.

[0116] In another example, if the compression model obtained according to step S602 does not meet any one or at least one of the model performance conditions described in the above embodiments, the terminal can further optimize the third model parameter or re-optimize the second model parameter, and obtain a compression model corresponding to the decompression model.

[0117] In an implementation, the model information in step S601 can include the second model parameter of the compression model corresponding to the decompression model. In another implementation, the model information in step S601 can include the first model parameter of the decompression model and the second model parameter of the compression model corresponding to the decompression model.

[0118] With the above scheme, the terminal receives the second model parameter of the compression model corresponding to the decompression model sent by the network device, and further adjusts the second model parameter to obtain the third model parameter and establish a compression model according to the third model parameter, which can make the adjusted compression model obtained by adjusting the second model parameter sent by the network device in the case of insufficient terminal capability, and the parameters of the compression model are transparent to the network device, which can effectively ensure the privacy of the compression model.

[0119] Figure 7 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a terminal, as shown in Figure 7 The method comprises:

[0120] S701, the terminal receives model information sent by a network device, the model information at least including a second model parameter of a compression model corresponding to a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0121] S702, the terminal establishes a compression model for compressing channel state information according to the second model parameter.

[0122] In an example, step S702 can be determined by the terminal according to its capability information, for example, in the case where the capability information of the terminal represents that the computing power of the terminal does not have the ability to train the compressed model, and does not have the ability to adjust the second model parameter of the second model, then step S702 can be performed, so that the terminal directly applies the model parameter of the compressed model sent by the network device.

[0123] In another example, if the compressed model obtained according to step S702 does not meet any one or at least one of the model performance conditions described in the above embodiments, the terminal can cause the network device to resend the model information, and establish the compressed model according to the second model parameter in the model information newly sent by the network device.

[0124] In an implementation manner, the model information in step S701 can include the second model parameter of the compressed model corresponding to the decompression model. In another implementation manner, the model information in step S701 can include the first model parameter of the decompression model and the second model parameter of the compressed model corresponding to the decompression model.

[0125] By using the above scheme, the terminal receives the second model parameter of the compressed model corresponding to the decompression model sent by the network device, and directly establishes the compressed model for compressing the channel state information by using the second model parameter, so that in the case where the terminal capability is insufficient, the compressed model can be obtained by receiving and applying the second model parameter sent by the network device.

[0126] It is worth noting that in the case where the model information includes the first model parameter of the decompression model and the second model parameter of the compressed model corresponding to the decompression model, steps S402, S502, S602 to S603, and S702 in the above embodiments can be selectively performed according to the capability information of the terminal.

[0127] For example, if the current load of the terminal is higher than a first preset threshold, step S702 can be selected to be performed, if the current load of the terminal is lower than a second preset threshold, step S402 can be selected to be performed, and if the current load of the terminal is higher than the second preset threshold and lower than the first preset threshold, steps S602 to S603 can be selected to be performed. Alternatively, in the case where the current load of the terminal is lower than the second preset threshold and the computing power of the terminal is less than a preset computing power threshold, steps S502 to S503 can be performed. The disclosure does not make specific limitation on which way the terminal specifically selects to obtain the compressed model based on the model information sent by the network device.

[0128] Figure 8is a flow chart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a terminal, such as Figure 8 As shown in the figure, the method comprises:

[0129] S801, the terminal receives model information sent by the network device, the model information comprising first model parameters of a plurality of decompression models; or the model information comprising first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models; wherein the decompression model is used by the network device to decompress channel state information sent by the terminal.

[0130] S802, the terminal determines model parameters of a target model from the model information according to at least capability information of the terminal and / or frequency band information used by the terminal, the target model being a decompression model or a compression model.

[0131] S803, the terminal determines a compression model for compressing channel state information according to the model parameters of the target model.

[0132] It can be understood that since the terminal accessing the network device corresponds to a terminal manufacturer or a chip manufacturer which can be completely different, the capability and the frequency band used by the terminal corresponding to different terminal manufacturers or chip manufacturers can also be completely different. Therefore, the network device can deploy a plurality of decompression models corresponding to different capabilities and frequency bands to ensure that the compressed channel state information sent by the terminal with different capabilities and frequency bands can be reliably decompressed.

[0133] The first model parameters of the plurality of decompression models or the first model parameters of the plurality of decompression models and the second model parameters of the plurality of compression models corresponding to the plurality of decompression models in the model information can be determined according to the capability information and / or the frequency band information reported by the terminal. For example, if the capability information reported by the terminal indicates that the terminal does not have the capability of model training, the network device can send model information comprising the first model parameters of the plurality of decompression models and the second model parameters of the plurality of compression models corresponding to the plurality of decompression models.

[0134] In step S802, the terminal can determine the decompression model and / or the compression model corresponding to the terminal based on the frequency band information used by the terminal, and then determine whether the target model is a decompression model or a compression model according to whether the terminal has the capability of compression model training.

[0135] For example, in the case where the model information only comprises the first model parameters of a first decompression model corresponding to a first frequency band and a second decompression model corresponding to a second frequency band, if the frequency band information used by the terminal indicates that the terminal uses the first frequency band, it can be determined that the target model is the first decompression model.

[0136] For example, when the model information includes first model parameters of a first decompression model corresponding to a first frequency band and second model parameters of a first compression model corresponding to the first decompression model, and first model parameters of a second decompression model corresponding to a second frequency band and second model parameters of a second compression model corresponding to the second decompression model, if the frequency band information used by the terminal indicates that the terminal uses the first frequency band and the capability information of the terminal indicates that the terminal does not have the capability of training a compression model, it can be determined that the target model of the terminal is the first compression model.

[0137] Specifically, if the target model is a decompression model, the step S803 can be that the terminal establishes the compression model according to the first model parameters of the decompression model, where the compression model can be locally trained by the terminal or trained by the server. If the target model is a compression model, the terminal can determine, according to the capability information, whether the terminal directly establishes the compression model according to the second model parameters of the target model or optimizes the second model parameters to obtain third model parameters and establishes the compression model based on the third model parameters.

[0138] With the above scheme, the network device transmits the model parameters of multiple decompression models, or the model parameters of multiple decompression models and multiple compression models corresponding to the multiple decompression models, so that the terminal determines the target model according to the capability information and / or the frequency band information, and determines the compression model of the terminal based on the parameters of the target model, which ensures that the channel state information can be effectively compressed for terminals with different capabilities and frequency bands, and enables the network terminal to reliably decompress the compressed channel state information transmitted by the terminals.

[0139] Figure 9 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a terminal, as shown in Figure 9 The method includes:

[0140] S901, the terminal receives model information transmitted by the network device, where the model information includes first model parameters of multiple decompression models; or the model information includes first model parameters of multiple decompression models and second model parameters of multiple compression models corresponding to the multiple decompression models one by one; where the decompression model is used by the network device to decompress channel state information transmitted by the terminal.

[0141] S902, the terminal determines, from the model information, model parameters of a target model according to at least capability information of the terminal and / or frequency band information used by the terminal, where the target model is a decompression model or a compression model.

[0142] S903, the terminal determines, according to the model parameters of the target model, a compression model of the terminal for compressing channel state information.

[0143] S904: The terminal reports the target model determined by the terminal to the network device.

[0144] It is worth noting that the above-mentioned step S904 can be executed after step S903 is executed, or before step S903 is executed, or after step S902 is executed and simultaneously with step S903. This disclosure does not limit this.

[0145] It should be noted that step S903 is an exemplary method of determining the model parameters of the target model from the model information through the terminal capability information and / or the frequency band information used by the terminal; of course, those skilled in the art can understand that the terminal can also determine the model parameters of the target model from the model information through other parameters, which will not be repeated here.

[0146] By adopting the above scheme, after determining the target model, the target model determined by the terminal is reported to the network device, so that after receiving the target model reported by the terminal, the network device can determine the decompression model corresponding to the channel state information sent by the terminal based on the target model, so that the network device can correctly select the decompression model and use it in conjunction with the compression model in the terminal to ensure that the channel state information sent by the terminal can be effectively decompressed.

[0147] Figure 10 is a flowchart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, which is applied to a terminal, such as Figure 10 As shown, the method includes:

[0148] S1001. The terminal receives model information sent by a network device, where the model information includes first model parameters of multiple decompression models; or, the model information includes first model parameters of multiple decompression models and second model parameters of compression models corresponding one-to-one to the multiple decompression models; wherein the decompression model is used by the network device to decompress channel state information sent by the terminal.

[0149] S1002: The terminal determines model parameters of a target model from the model information based at least on the capability information of the terminal and / or the frequency band information used by the terminal, where the target model is a decompression model or a compression model.

[0150] S1003: The terminal reports the terminal's capability information and / or the frequency band information used by the terminal to the network device. The terminal's capability information and / or the frequency band information used by the terminal are used by the network device to determine the target model determined by the terminal.

[0151] For example, in a case where the model information only includes the first model parameters of the first decompression model corresponding to the first frequency band and the second model parameters of the second decompression model corresponding to the second frequency band, if the network device determines, according to the frequency band information reported by the terminal, that the frequency band information indicates that the terminal uses the first frequency band, the network device can determine that the first decompression model selected by the terminal is the target model.

[0152] For another example, in a case where the model information includes the first model parameters of the first decompression model corresponding to the first frequency band and the second model parameters of the first compression model corresponding to the first decompression model, and the first model parameters of the second decompression model corresponding to the second frequency band and the second model parameters of the second compression model corresponding to the second decompression model, if the network device determines, according to the frequency band information reported by the terminal, that the frequency band information indicates that the terminal uses the first frequency band, and the capability information reported by the terminal indicates that the terminal does not have the capability of training the compression model, the network device can determine that the target model selected by the terminal is the first compression model.

[0153] Further, the network device can determine, according to the target model selected by the terminal, a decompression model for decompressing the channel state information sent by the terminal.

[0154] It should be noted that the model parameters of the target model are determined from the model information by the capability information of the terminal and / or the frequency band information used by the terminal in step S1002; of course, those skilled in the art can understand that the terminal can also determine the model parameters of the target model from the model information by other parameters, which will not be described here.

[0155] By the above scheme, after the target model is determined, the capability information of the terminal and / or the frequency band information used by the terminal are reported to the network device, so that the network device determines the target model determined by the terminal, and the network device can determine the decompression model corresponding to the channel state information sent by the terminal according to the target model after receiving the target model reported by the terminal, so as to ensure that the channel state information sent by the terminal can be effectively decompressed.

[0156] Figure 11 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a terminal, as shown in Figure 11 The method includes:

[0157] S1101, the terminal receives model information sent by the network device, the model information including first model parameters of a plurality of decompression models; or, the model information including first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one; wherein the decompression model is used by the network device to decompress channel state information sent by the terminal.

[0158] S1102, the terminal determines, from the model information, a model parameter of a target model according to at least the capability information of the terminal and / or the frequency band information used by the terminal, the target model being a decompression model or a compression model.

[0159] S1103, the terminal reports, to the network device, identification information of the target model.

[0160] It can be understood that the identification information is used to uniquely identify each decompression model in the model information.

[0161] Corresponding to each group of associated decompression models and compression models, the same identification information can be used or different identification information can be used, which can be represented by a model ID, for example. For example, for a first decompression model and a first compression model corresponding to the first decompression model, “0001” can be used as the corresponding identification information of both, or for the first decompression model, “0010” can be used as the corresponding identification information thereof, and for the first compression model, “0011” can be used as the corresponding identification information thereof. The present disclosure does not limit the setting manner of the identification information.

[0162] It should be noted that in step S1102, the capability information of the terminal and / or the frequency band information used by the terminal are exemplary parameters used by the terminal to determine the model parameter of the target model from the model information. Of course, those skilled in the art can understand that the terminal can also determine the model parameter of the target model from the model information by using other parameters, which will not be described herein.

[0163] By using the above scheme, after the target model is determined, the identification information of the target model is reported to the network device, so that the network device determines the target model determined by the terminal. After the network device determines the target model reported by the terminal, the decompression model corresponding to the channel state information sent by the terminal can be determined according to the target model, so as to ensure that the channel state information sent by the terminal can be effectively decompressed.

[0164] Figure 12 is a flowchart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a terminal, as shown in Figure 12 The method comprises the following steps.

[0165] S1201, the terminal receives model information sent by a network device, the model information comprising first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal.

[0166] S1202, the terminal determines, according to the model information, a compression model used by the terminal to compress channel state information.

[0167] S1203, the terminal determines an enabling time of the compression model according to a preset time length in the communication protocol or a preset time length configured by the network device, the preset time length being a time length relative to a specific time.

[0168] The preset time length configured by the network device can be sent to the terminal through the model information or sent to the terminal through other signaling.

[0169] It can be understood that the communication protocol is a protocol predefined for communication between the terminal and the network device, and the network device can also determine the enabling time of the compression model of the terminal and its own decompression model according to the preset time length in the communication protocol or the preset time length configured by the network device.

[0170] In an example, the specific time is an ending time of a PDSCH in which the terminal receives the model information.

[0171] In another example, the specific time is a time of a last symbol of an uplink resource in which the terminal feeds back HARQ-ACK information for the PDSCH including the model information.

[0172] The determination manner of the preset time length corresponding to different compression models can be different.

[0173] For example, the preset time length can be X unit time lengths corresponding to the determination manner of the compression model in steps S402 and S502 in the above embodiment; the preset time length can be Y unit time lengths corresponding to the determination manner of the compression model in steps S602 to S603 in the above embodiment; and the preset time length can be Z unit time lengths corresponding to the determination manner of the compression model in step S702 in the above embodiment. The sizes of X, Y and Z can be predefined in the communication protocol or configured by the network device. X can be greater than Y and Y can be greater than Z, or X, Y and Z can be equal, which is not limited in the disclosure.

[0174] The unit time length can be a time length corresponding to one slot, one symbol or one subframe, which is not limited in the disclosure.

[0175] For example, if the ending time of the PDSCH in which the terminal receives the model information is a time corresponding to slot n, and the unit time length is a slot, where slot represents a slot and n is the number of the slot, if the terminal adopts the determination manner of the compression model in steps S402 to S403, the enabling time of the compression model can be a time corresponding to slot n+X, that is, the compression model is enabled at a time corresponding to the slot with the number n+X.

[0176] By adopting the foregoing scheme, the terminal and the network device can synchronously enable the compression model and the decompression model based on the preset time length specified in the communication protocol or configured in the network device and the specific time corresponding to the model information, thereby ensuring that the network device can reliably decompress the compressed channel state information sent by the terminal, and ensuring the communication quality between the network device and the terminal.

[0177] It should be noted that the foregoing embodiments performed by the terminal can be implemented alone or in combination, and the embodiments of the present disclosure do not limit this.

[0178] Figure 13 is a flowchart of a method for determining a compression model for compressing channel state information according to an exemplary embodiment, applied to a network device, such as a base station. Figure 13 As shown in the figure, the method comprises the following steps.

[0179] S1301, the network device sends model information, the model information comprising first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by a terminal, the model information being used by the terminal to determine a compression model for compressing channel state information.

[0180] In an example, the model information can further comprise first model parameters of a plurality of decompression models; or, the model information comprises first model parameters of a plurality of decompression models, and second model parameters of a compression model corresponding to each of the plurality of decompression models.

[0181] In an example, the model information can be sent by a base station through RRC signaling, and in this case, the network device can be the base station. Alternatively, in another example, the model information can be sent by an AMF (Access and Mobility Management Entity) in a core network through NAS (Non-access stratum) layer signaling, and in this case, the network device can be the AMF.

[0182] In another example, the compression model used by the terminal to compress channel state information satisfies at least one of the following model performance conditions: the mean square error or the normalized mean square error of the compression model and the decompression model used jointly is less than a first preset threshold; the cosine similarity of the compression model and the decompression model used jointly is greater than a second preset threshold; the square of the cosine similarity of the compression model and the decompression model used jointly is greater than a third preset threshold; the signal-to-noise ratio of the compression model and the decompression model used jointly is greater than a fourth preset threshold.

[0183] The model information can be sent in the case that the decompression model deployed in the network device changes, or can be sent in response to the terminal accessing the base station.

[0184] It is worth noting that sending the model information by the network device to enable the terminal to determine the compression model for compressing the channel state information can ensure that the network device has the dominant right of the compression model of the terminal, and in the case that the decompression model deployed on the network device side is updated, the compression model deployed on the terminal side can be effectively updated.

[0185] In an example, the network device can also be configured with a preset time length, or the preset time length can be specified in the communication protocol between the network device and the terminal, so that the terminal determines the enabling time of the compression model according to the preset time length. Moreover, the network device can determine the enabling time of the decompression model corresponding to the compression model used by the terminal according to the preset time length. The preset time length is a time length relative to a specific time. In this way, the terminal and the network device can synchronously enable the corresponding compression model and decompression model to achieve reliable compression and decompression of the channel state information.

[0186] In the embodiments of the present disclosure, the network device sends the model information to the terminal, so that the terminal obtains the compression model for compressing the channel state information according to the model information. Since the model information includes the first model parameter of the decompression model of the network device, the terminal can obtain the compression model for joint use with the decompression model of the network device according to the first model parameter and the private data of the terminal manufacturer or chip manufacturer. In this process, the network device does not need to know the parameters of the compression model deployed by the terminal, thereby ensuring the privacy of the compression model.

[0187] Figure 14 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, such as Figure 14 As shown in the figure, the method comprises:

[0188] S1401, the network device sends model information, the model information including a first model parameter of a decompression model and a second model parameter of a compression model corresponding to the decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal, and the model information being used by the terminal to determine a compression model for compressing channel state information.

[0189] The network device whether to send the second model parameter can be determined according to the capability information reported by the terminal or the terminal group. In addition, after receiving the model information, the terminal can determine whether to use the first model parameter to determine the decompression model or use the second model parameter to determine the compression model according to its own capability or other parameters. In addition, after receiving the model information, the terminal can determine whether to use the second model parameter to determine the compression model or use the third model parameter obtained by optimizing the second model parameter, and establish the compression model for compressing the channel state information according to the third model parameter.

[0190] By adopting the above scheme, by sending the first model parameter and the second model parameter to the terminal, not only can the terminal with model training capability obtain the compression model according to the first model parameter, but also the terminal without model training capability can obtain the compression model according to the second model parameter, so that the terminals of different terminal manufacturers or terminal chip manufacturers can effectively obtain the compression model.

[0191] Figure 15 is a flow chart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, as shown in Figure 15 The method comprises the following steps.

[0192] S1501, the network device broadcasts model information, the model information comprising a first model parameter of a decompression model, the decompression model being used by the network device to decompress channel state information sent by a terminal, and the model information being used by the terminal to determine a compression model for compressing channel state information.

[0193] In an example, the model information can further comprise a second model parameter of a compression model corresponding to the decompression model.

[0194] In another example, the model information can further comprise first model parameters of a plurality of decompression models; or the model information comprises first model parameters of a plurality of decompression models, and second model parameters of compression models corresponding to the plurality of decompression models one by one.

[0195] By adopting the above scheme, the network device sends the model information in a broadcast manner, so that the terminal within the broadcast range of the network device can receive the model information, and then determine the compression model for compressing the channel state information according to the model information.

[0196] Figure 16 is a flow chart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, as shown in Figure 16 The method comprises the following steps.

[0197] S1601, the network device sends model information to the terminal in a unicast manner, the model information comprises first model parameters of a decompression model, the decompression model is used for decompression of channel state information sent by the terminal, and the model information is used for the terminal to determine a compression model used for compression of the channel state information.

[0198] It can be understood that the model information sent in the unicast manner can be different for different terminals. For example, first model information can be sent to a first terminal in a unicast manner, and second model information can be sent to a second terminal in a unicast manner.

[0199] In an example, the model information can further comprise second model parameters of a compression model corresponding to the decompression model.

[0200] In another example, the model information can further comprise first model parameters of a plurality of decompression models; or the model information comprises first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one.

[0201] With the above scheme, the model information is sent to the terminal in a unicast manner, the model information can be sent more specifically, the network resources can be prevented from being occupied by too much redundant information in the sent information, and the network resource utilization rate can be effectively improved.

[0202] Figure 17 is a flowchart of a method for determining a compression model used for compression of channel state information according to an example embodiment, applied to a network device, as shown in Figure 17 The method comprises the following steps.

[0203] S1701, the network device sends model information to a terminal group in a multicast manner, the model information comprises first model parameters of a decompression model, the decompression model is used for decompression of channel state information sent by the terminal, and the model information is used for the terminal to determine a compression model used for compression of the channel state information.

[0204] It can be understood that the model information sent in the multicast manner can be different for different terminal groups. For example, first model information can be sent to terminals in a first terminal group in a multicast manner, and second model information can be sent to terminals in a second terminal group in a multicast manner.

[0205] In an example, the model information can further comprise second model parameters of a compression model corresponding to the decompression model.

[0206] In another example, the model information can further include first model parameters of a plurality of decompression models; or the model information includes first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one.

[0207] By using the above scheme, the model information is sent to the terminal group in a multicast manner, the model information can be sent more specifically, the network resources can be avoided from being occupied by too much redundant information in the sent information, and the network resource utilization can be effectively improved.

[0208] Figure 18 is a flow chart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, such as Figure 17 As shown in the figure, the method includes the following steps.

[0209] S1801, the network device sends model information to terminals in a broadcast manner for terminals not accessing the network device, and sends the model information to terminals or a terminal group having completed access to the network device in a unicast or multicast manner, the model information includes first model parameters of decompression models, the decompression models are used by the network device to decompress channel state information sent by the terminals, and the model information is used by the terminals to determine a compression model for compressing channel state information.

[0210] The broadcast model information and the unicast or multicast model information can be different. For example, the broadcast model information can include first model parameters corresponding to all decompression models deployed in the network device. The unicast or multicast model information can include only first model parameters corresponding to part of the decompression models deployed in the network device.

[0211] In an example, the model information can further include second model parameters of compression models corresponding to the decompression models.

[0212] In another example, the model information can further include first model parameters of a plurality of decompression models; or the model information includes first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one.

[0213] By using the above scheme, for terminals not accessing the network device, the model information can be sent to terminals in its broadcast range in a broadcast manner, so that the terminals in the broadcast range can determine a compression model for compressing channel state information according to the model information, and the terminals in the broadcast range can communicate with the network device according to the corresponding compression model. For terminals having accessed the network, the model information can be sent more specifically in a unicast or multicast manner, the network resources can be avoided from being occupied by too much redundant information in the sent information, and the network resource utilization can be effectively improved.

[0214] Figure 19 is a flow chart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, such as Figure 19 as shown, the method comprises:

[0215] S1901, the network device sends model information in a unicast or groupcast manner in response to terminal capability information and / or frequency band information sent by a terminal or a terminal group, the model information including first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal, the model information being used by the terminal to determine a compression model for compressing channel state information.

[0216] In an example, the network device sends different model information for terminals or terminal groups with different terminal capability information and / or frequency band information used by the terminals or terminal groups.

[0217] For example, if the terminal frequency band information sent by the terminal group indicates that the terminals in the terminal group use a first frequency band, and the capability information indicates that the terminals have the capability of model training, the network device can send model information including first model parameters of a decompression model corresponding to the first frequency band to the terminal group in a groupcast manner. Further, if the capability information indicates that some terminals in the terminal group are currently overloaded and cannot perform model training, the model information sent in the groupcast manner can further include second model parameters of a compression model corresponding to the decompression model corresponding to the first frequency band, so that the terminal can obtain the compression model by tuning the second model parameters.

[0218] In another example, if the frequency band information sent by the terminal group indicates that the terminals in the terminal group use a second frequency band, the network device can send model information including first model parameters of a decompression model corresponding to the second frequency band to the terminal group in a groupcast manner. Alternatively, if the frequency band information indicates that the terminals in the terminal group use a first frequency band and a second frequency band, the network device can send model information including first model parameters of a decompression model corresponding to the first frequency band and first model parameters of a decompression model corresponding to the second frequency band to the terminal group in a groupcast manner.

[0219] In yet another example, if the frequency band information sent by the terminal indicates that the terminal uses a second frequency band, and the capability information indicates that the terminal currently does not have the capability of model training, the network device can send model information including first model parameters of a decompression model corresponding to the second frequency band and second model parameters of a compression model corresponding to the decompression model to the terminal in a unicast manner.

[0220] In an example, the model information can further include second model parameters of a compression model corresponding to the decompression model.

[0221] In another example, the model information further includes first model parameters of the plurality of decompression models; or the model information includes first model parameters of the plurality of decompression models and second model parameters of the plurality of compression models corresponding to the plurality of decompression models one by one.

[0222] With the above scheme, the network device receives the capability information and / or the frequency band information of the terminal or the terminal group that has accessed the network device, and transmits the model information in a unicast or multicast manner, so that the model information can be transmitted more targetedly according to the capability and the used frequency band of the terminal or the terminal group, and the communication resources can be avoided from being occupied by too much redundant information in the transmitted information, and the communication resource utilization rate can be effectively improved.

[0223] Figure 20 is a flowchart of a method for determining a compression model for compressing channel state information according to an example embodiment, applied to a network device, as shown in Figure 20 The method includes the following steps.

[0224] S2001, the network device transmits model information, the model information including first model parameters of a plurality of decompression models; or the model information including first model parameters of the plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one, the decompression model being used for decompressing channel state information transmitted by a terminal, and the model information being used for the terminal to determine a compression model for compressing channel state information.

[0225] S2002, the network device receives a target model reported by the terminal, the target model being determined by the terminal from the model information, and model parameters of the target model being used by the terminal to determine a compression model for compressing channel state information.

[0226] It can be understood that, in the case that the model information includes first model parameters of the plurality of decompression models; or the model information includes first model parameters of the plurality of decompression models and second model parameters of the plurality of compression models corresponding to the plurality of decompression models one by one, the terminal will only use the model parameters corresponding to one decompression model or compression model to obtain the compression model. Therefore, if the network device and the terminal need to jointly use the corresponding compression model and decompression model, the target model selected by the terminal needs to be determined, so as to ensure that the network device can effectively decompress the channel state information.

[0227] In an example, the terminal can report identification information of the target model, so that the network device can determine the target model determined by the terminal according to the identification information.

[0228] In another example, the terminal can report the capability information of the terminal and / or the frequency band information used by the terminal, so that the network device determines the target model determined by the terminal according to the capability information and / or the frequency band information.

[0229] By using the above scheme, the network device can determine the decompression model corresponding to the compression model used by the terminal to decompress the compressed channel state information when the network device receives the compressed channel state information sent by the terminal, so as to effectively guarantee the communication quality between the terminal and the network device.

[0230] Figure 21 Fig. 1 is an interaction diagram of a method for determining a compression model for compressing channel state information according to an example embodiment, as shown in the figure, the method comprises the following steps. Figure 21

[0231] S2101. The network device sends model information to the terminal.

[0232] The model information can include first model parameters of a plurality of decompression models, or the model information includes first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one.

[0233] The model information can be sent in a multicast manner, and the terminal is one of the terminals in the terminal group, or the model information can be sent in a broadcast manner.

[0234] S2102. The terminal determines a target model from the model information according to the capability information of the terminal and / or the frequency band information used by the terminal.

[0235] S2103. The terminal reports the capability information of the terminal and / or the frequency band information used by the terminal to the network device.

[0236] S2104. The network device determines the target model determined by the terminal according to the capability information and / or the frequency band information reported by the terminal.

[0237] In some other optional embodiments, the terminal can report identification information of the target model determined by the terminal in step S2103, so that the network device determines the target model determined by the terminal according to the identification information.

[0238] S2105. The terminal determines a compression model for compressing channel state information according to model parameters of the target model.

[0239] The target model can be a compression model or a decompression model.

[0240] ​In a case that the target model is the decompression model, the terminal can train the compression model according to the first model parameter of the decompression model. The step of training the compression model can be performed by the server.

[0241] In a case that the target model is the compression model, the terminal can directly apply the second model parameter corresponding to the compression model, or the terminal can apply the second model parameter after optimization, that is, establish the compression model according to the third model parameter obtained by optimization.

[0242] It can be understood that the training of the compression model by the terminal itself or the training of the compression model by the server can be determined according to the capability information of the terminal. For example, if the current load of the terminal is too high to train itself, the terminal can send a model obtaining request to the server, so that the compression model trained by the server is obtained.

[0243] Similarly, directly applying the second model parameter of the compression model corresponding to the target model or constructing the compression model after optimizing the second model can also be determined according to the capability information of the terminal.

[0244] S2106, determining an enabling time of the compression model according to a preset time length in the communication protocol or a preset time length configured by the network device.

[0245] The preset time length can be a time length relative to a specific time. The specific time includes an ending time of the PDSCH in which the terminal receives the model information, and / or a time of a last symbol of an uplink resource in which the terminal feeds back HARQ-ACK information of the PDSCH including the model information.

[0246] S2107, enabling the compression model to compress the channel state information at the enabling time.

[0247] It can be understood that the network device can synchronously enable a decompression model corresponding to the compression model at the enabling time. Or, the decompression model remains in an enabled state in the network device. The network device can determine the decompression model corresponding to the compressed channel state information according to the received compressed channel state information, to decompress the compressed channel state information.

[0248] It should be noted that the foregoing various embodiments performed by the network device can be implemented alone or in combination. The embodiments of the present disclosure do not limit this.

[0249] Those skilled in the art can understand that the foregoing embodiments performed by the terminal and the embodiments performed by the network device are corresponding to each other, so that the corresponding operations are necessarily performed on the other side for the steps described on one side. For example, the network device broadcasts the model information or sends the model information to the terminal by unicast or sends the model information to the terminal group by multicast; and the terminal necessarily receives the model information by the corresponding mode.

[0250] The device of the embodiment of the present disclosure is described below. It should be noted that the functions of various modules have been described in detail in the embodiments of the method, and will not be described in detail here.

[0251] Figure 22 A block diagram of a device 22 for determining a compression model for compressing channel state information according to an exemplary embodiment is shown, the device 22 is applied to a terminal, and the device 22 includes:

[0252] The receiving module 2201 is configured to receive model information sent by a network device, the model information including first model parameters of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal;

[0253] The determining module 2202 is configured to determine, according to the model information, a compression model used by the terminal to compress the channel state information.

[0254] Optionally, the model information further includes second model parameters of a compression model corresponding to the decompression model.

[0255] Optionally, the determining module 2202 includes:

[0256] The training sub-module is configured to train the compression model corresponding to the decompression model according to the first model parameters.

[0257] Optionally, the determining module 2202 includes:

[0258] The local training module is configured to locally train the compression model corresponding to the decompression model at the terminal;

[0259] The server training module is configured to send an acquisition request to a server and acquire the compression model corresponding to the decompression model sent by the server.

[0260] Optionally, the determining module 2202 includes:

[0261] The tuning sub-module is configured to tune the second model parameters to obtain third model parameters.

[0262] The first establishing submodule is configured to establish, according to the third model parameter, a compression model for compressing channel state information.

[0263] Optionally, the determining module 2202 includes:

[0264] The second establishing submodule is configured to establish, according to the second model parameter, a compression model for compressing channel state information.

[0265] Optionally, the compression model used by the terminal for compressing channel state information satisfies at least one of the following model performance conditions:

[0266] A mean square error or a normalized mean square error in a joint use case of the compression model and a decompression model is less than a first preset threshold value;

[0267] A cosine similarity in the joint use case of the compression model and the decompression model is greater than a second preset threshold value;

[0268] A square of the cosine similarity in the joint use case of the compression model and the decompression model is greater than a third preset threshold value;

[0269] A signal-to-noise ratio in the joint use case of the compression model and the decompression model is greater than a fourth preset threshold value.

[0270] Optionally, the model information includes first model parameters of a plurality of decompression models; or the model information includes the first model parameters of the plurality of decompression models, and second model parameters of a plurality of compression models corresponding to the plurality of decompression models in one-to-one correspondence.

[0271] Optionally, the determining module 2202 includes:

[0272] The first determining submodule is configured to determine, from the model information, a model parameter of a target model according to at least capability information of the terminal and / or frequency band information used by the terminal, the target model being a decompression model or a compression model;

[0273] The second determining submodule is configured to determine, according to the model parameter of the target model, a compression model used by the terminal for compressing channel state information.

[0274] Optionally, the apparatus 22 further includes:

[0275] The reporting module is configured to report, to a network device, the target model determined by the terminal.

[0276] Optionally, the reporting module includes:

[0277] The first reporting submodule is configured to report, to the network device, identification information of the target model.

[0278] The second reporting submodule is configured to report the capability information of the terminal and / or the frequency band information used by the terminal to the network device, and the capability information of the terminal and / or the frequency band information used by the terminal are used by the network device to determine the target model determined by the terminal.

[0279] Optionally, the apparatus 22 further comprises:

[0280] The enabling module is configured to determine the enabling time of the compression model according to a preset time length in the communication protocol or a preset time length configured by the network device, and the preset time length is a time length relative to a specific time.

[0281] Optionally, the specific time includes an ending time of a PDSCH in which the terminal receives the model information, and / or a time of a last symbol of an uplink resource in which the terminal feeds back HARQ-ACK information of the PDSCH including the model information.

[0282] Figure 23 A block diagram of an apparatus 23 for determining a compression model for compressing channel state information according to an example embodiment is shown, the apparatus 23 is applied to a network device, and the apparatus 23 comprises:

[0283] The sending module 2301 is configured to send model information, the model information including a first model parameter of a decompression model, the decompression model being used by the network device to decompress channel state information sent by the terminal, and the model information being used by the terminal to determine a compression model for compressing channel state information.

[0284] Optionally, the model information further includes a second model parameter of a compression model corresponding to the decompression model.

[0285] Optionally, the sending module 2301 comprises:

[0286] The first broadcasting submodule is configured to broadcast the model information; or,

[0287] The unicast submodule is configured to send the model information to the terminal in a unicast manner; or,

[0288] The groupcast submodule is configured to send the model information to a terminal group in a groupcast manner.

[0289] Optionally, the sending module 2301 comprises:

[0290] The second sending submodule is configured to send the model information to a terminal that has not accessed the network device in a broadcast manner;

[0291] The second sending submodule is configured to send the model information to a terminal or a terminal group that has completed access to the network device in a unicast or groupcast manner.

[0292] Optionally, the second sending sub-module comprises:

[0293] The third sending sub-module is configured to, for a terminal or a terminal group that has completed access to the network device, send model information in a unicast or groupcast manner in response to terminal capability information and / or frequency band information sent by the terminal or the terminal group.

[0294] Optionally, the network device sends different model information for terminals or terminal groups with different terminal capability information and / or used frequency band information.

[0295] Optionally, the model information comprises first model parameters of a plurality of decompression models; or the model information comprises first model parameters of a plurality of decompression models and second model parameters of a plurality of compression models corresponding to the plurality of decompression models one by one.

[0296] Optionally, the apparatus 23 further comprises:

[0297] The first receiving module is configured to receive a target model determined by a terminal and reported by the terminal, the target model being determined by the terminal from the model information, and model parameters of the target model being used by the terminal to determine a compression model used by the terminal to compress channel state information.

[0298] Optionally, the network device is a base station, and the sending module 2301 comprises:

[0299] The RRC signaling sending module is configured to send the model information through RRC signaling.

[0300] Optionally, the network device is an AMF network element in a core network, and the sending module 2301 comprises:

[0301] The NAS signaling sending module is configured to send the model information through NAS signaling.

[0302] As to the apparatus in the above embodiments, the specific manners in which various modules perform operations have been described in detail in the embodiments of the method, and thus will not be described here in detail.

[0303] The present disclosure also provides a computer readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the steps of the method for determining a compression model used to compress channel state information according to any one of the preceding method embodiments provided by the present disclosure.

[0304] Figure 24 is a block diagram of a terminal 2400 according to an exemplary embodiment. For example, the terminal 2400 can be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0305] Referring toFigure 24 The terminal 2400 can include one or more of the following components: a processing component 2402, a memory 2404, a power supply component 2406, a multimedia component 2408, an audio

[0306] The processing component 2402 usually controls overall operations of the terminal 2400, such as operations associated with displaying, making phone calls, data communications, camera operations and recording operations. The processing component 2402 can include one or more processors 2420 to execute instructions to complete all or part of steps of the above methods. In addition, the processing component 2402 can include one or more modules to facilitate

[0307] The memory 2404 is configured to store various types of data to support operations of the terminal 2400. Examples of these data include instructions for any applications or methods operating on the terminal 2400, contact data, phonebook data, messages, pictures, videos, and the like. The memory 2404 can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic storage, flash memory, magnetic or optical disks.

[0308] The power supply component 2406 provides power for the various components of the terminal 2400. The power supply component 2406 can include a power management system, one or more power sources, and other components associated with generating, managing and distributing power for the terminal 2400.

[0309] The multimedia component 2408 includes a screen providing an output interface between the terminal 2400 and a user. In some embodiments, the screen can include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors for sensing a touch, a slide and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touching or a sliding action, but also detect duration and pressure related to the touching or sliding action. In some embodiments, the multimedia component 2408 includes a front camera and / or a rear camera. When the terminal 2400 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front and rear camera can be a fixed optical lens system or have a focal length and optical zooming capability.

[0310] The audio component 2410 is configured to output and / or input an audio signal. For example, the audio component 2410 includes a microphone (MIC) configured to receive an external audio signal when the terminal 2400 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 2404 or transmitted via the communication component 2416. In some embodiments, the audio component 2410 further includes a speaker for outputting an audio signal.

[0311] The I / O interface 2412 provides an interface between the processing component 2402 and peripheral interface modules, which can be a keypad, a click wheel, buttons, and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button, and a lock button.

[0312] The sensor component 2414 includes one or more sensors for providing status assessments of various aspects of the terminal 2400. For example, the sensor component 2414 can detect an open / closed position of the terminal 2400, relative positioning of components, such as a display and a keypad of the terminal 2400, a change in position of the terminal 2400 or a component of the terminal 2400, presence or absence of user contact with the terminal 2400, an orientation or acceleration / deceleration / g-force and a temperature change of the terminal 2400. The sensor component 2414 can include a proximity sensor configured to detect presence of a nearby object without any physical touch. The sensor component 2414 can further include a light sensor such as a CMOS or CCD image sensor for use in an imaging application. In some embodiments, the sensor component 2414 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0313] The communication component 2416 is configured to facilitate wired or wireless communication between the terminal 2400 and other devices. The terminal 2400 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 2416 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 2416 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0314] In an exemplary embodiment, terminal 2400 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above methods.

[0315] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 2404 including instructions, which can be executed by the processor 2420 of the terminal 2400 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0316] In another exemplary embodiment, a computer program product is also provided, which includes a computer program that can be executed by a programmable device and has a code portion for performing the above-mentioned method of determining a compression model for compressing channel state information when executed by the programmable device.

[0317] Figure 25 25 is a block diagram of a network device according to an exemplary embodiment. For example, the network device 2500 may be provided as a base station, or may be provided as other network logic entities in a core network. Figure 25 Network device 2500 includes a processing component 2522, which further includes one or more processors and memory resources represented by memory 2532 for storing instructions executable by processing component 2522, such as applications. The applications stored in memory 2532 may include one or more modules, each corresponding to a set of instructions. In addition, processing component 2522 is configured to execute instructions to perform the steps of the method for determining a compression model for compressing channel state information provided in the above method embodiment.

[0318] The network device 2500 can also include a power supply component 2526 configured to perform power management for the device 2500, a wired or wireless network interface 2550 configured to connect the network device 2500 to a network, and an input output (I / O) interface 2558. The network device 2500 can operate on an operating system stored in the memory 2532, such as Windows Server TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM , or the like.

[0319] In another exemplary embodiment, there is also provided a computer program product comprising a computer program being executable by a programmable apparatus, the computer program having code portions for performing the above-described method of determining a compression model for compressing channel state information when executed by the programmable apparatus.

[0320] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure. Variations and modifications of the embodiments disclosed herein can be made based on the description set forth herein, without departing from the scope and spirit of the disclosure. The application is intended to cover any variations, uses or adaptations of the disclosure including such departures from the

[0321] It is to be understood that the disclosure is not limited to the precise construction that has been described and shown in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the disclosure. The scope of the disclosure is limited only by the claims that follow.

Claims

1. A method for determining a compression model for compressing channel state information, characterized in that: Applied to a terminal, the method includes: receiving model information sent by a network device, the model information including a first model parameter of a decompression model, the decompression model being used by the network device to decompress the channel state information sent by the terminal; A compression model used by the terminal to compress channel state information is determined according to the model information and the private data of the terminal.

2. The method according to claim 1, characterized in that The model information further includes second model parameters of a compression model corresponding to the decompression model.

3. The method according to claim 1 or 2, characterized in that The determining, based on the model information and the private data of the terminal, a compression model used by the terminal to compress the channel state information includes: A compression model corresponding to the decompression model is obtained by training according to the first model parameters and the privatized data.

4. The method according to claim 3, characterized in that The training, based on the first model parameters and the privatized data, to obtain a compression model corresponding to the decompression model includes: Training the terminal locally to obtain a compression model corresponding to the decompression model; or, Send an acquisition request to the server, and acquire the compression model corresponding to the decompression model sent by the server.

5. The method according to claim 2, characterized in that The determining, based on the model information and the private data of the terminal, a compression model used by the terminal to compress the channel state information includes: Tuning the second model parameters according to the privatized data to obtain third model parameters; A compression model for compressing channel state information is established according to the third model parameters.

6. The method according to claim 2, characterized in that The determining, according to the model information, a compression model used by the terminal to compress the channel state information includes: A compression model for compressing channel state information is established according to the second model parameters.

7. The method according to claim 1, characterized in that The compression model used by the terminal to compress the channel state information satisfies at least one of the following model performance conditions: A mean square error or a normalized mean square error when the compression model and the decompression model are used in combination is less than a first preset threshold; The cosine similarity when the compression model and the decompression model are used together is greater than a second preset threshold; The square of the cosine similarity when the compression model and the decompression model are used together is greater than a third preset threshold; A signal-to-noise ratio when the compression model and the decompression model are used in combination is greater than a fourth preset threshold.

8. The method according to claim 1, characterized in that The model information includes first model parameters of multiple decompressed models; or, the model information includes first model parameters of multiple decompressed models and second model parameters of compressed models corresponding one-to-one to the multiple decompressed models.

9. The method according to claim 8, characterized in that The determining, based on the model information and the private data of the terminal, a compression model used by the terminal to compress the channel state information includes: determining, from the model information, model parameters of a target model based at least on the capability information of the terminal and / or information on a frequency band used by the terminal, the target model being a decompression model or a compression model; A compression model used by the terminal to compress channel state information is determined according to the model parameters of the target model and the private data of the terminal.

10. The method according to claim 9, characterized in that The method further comprises: Reporting the target model determined by the terminal to the network device.

11. The method according to claim 10, characterized in that The reporting of the target model determined by the terminal to the network device includes: reporting identification information of the target model to the network device; or, The capability information of the terminal and / or the frequency band information used by the terminal is reported to the network device, and the capability information of the terminal and / or the frequency band information used by the terminal are used by the network device to determine the target model determined by the terminal.

12. The method according to claim 1, characterized in that The method further comprises: The activation time of the compression model is determined according to a preset duration in the communication protocol or a preset duration configured by the network device, where the preset duration is a duration relative to a specific moment.

13. The method according to claim 11, characterized in that The specific moment includes the end moment when the terminal receives the PDSCH containing the model information, and / or the moment when the terminal feeds back the last symbol of the uplink resource of the HARQ-ACK information for the PDSCH including the model information.

14. A method for determining a compression model for compressing channel state information, characterized in that: Applied to a network device, the method includes: Send model information, the model information including a first model parameter of a decompression model, the decompression model being used by the network device to decompress the channel state information sent by the terminal, and the model information being used by the terminal to determine a compression model for compressing the channel state information in combination with the privatized data of the terminal.

15. The method according to claim 14, characterized in that The model information further includes second model parameters of a compression model corresponding to the decompression model.

16. The method according to claim 14, characterized in that The sending model information includes: broadcasting the model information; or, Send the model information to the terminal in a unicast manner; or The model information is sent to the terminal group in a multicast manner.

17. The method according to claim 14, characterized in that The sending model information includes: Sending the model information to terminals that are not connected to the network device by broadcasting; The model information is sent to the terminal or terminal group that has completed access to the network device in a unicast or multicast manner.

18. The method according to claim 14, characterized in that The sending of the model information to the terminal that has completed access to the network device by unicast or multicast includes: For a terminal or a terminal group that has completed access to the network device, the model information is sent via unicast or multicast in response to the terminal capability information and / or frequency band information sent by the terminal or the terminal group.

19. The method according to claim 18, characterized in that For terminals or terminal groups with different terminal capability information and / or used frequency band information, the model information sent by the network device is different.

20. The method according to claim 14, wherein The model information includes first model parameters of multiple decompressed models; or, the model information includes first model parameters of multiple decompressed models and second model parameters of compressed models corresponding one-to-one to the multiple decompressed models.

21. The method according to claim 20, characterized in that The method further comprises: A target model determined by the terminal and reported by the receiving terminal is received, where the target model is determined by the terminal from the model information, and model parameters of the target model are used by the terminal to determine a compression model used by the terminal to compress channel state information.

22. The method according to any one of claims 14 to 21, characterized in that The network device is a base station, and the sending model information includes: The model information is sent via RRC signaling.

23. The method according to any one of claims 14 to 21, characterized in that The network device is an AMF network element in the core network, and the sending model information includes: The model information is sent via NAS signaling.

24. An apparatus for determining a compression model for compressing channel state information, characterized in that Applied to a terminal, the device includes: a receiving module configured to receive model information sent by a network device, wherein the model information includes a first model parameter of a decompression model, and the decompression model is used by the network device to decompress the channel state information sent by the terminal; The determination module is configured to determine a compression model used by the terminal to compress channel state information according to the model information and the private data of the terminal.

25. An apparatus for determining a compression model for compressing channel state information, characterized in that Applied to network equipment, the device includes: The sending module is configured to send model information, wherein the model information includes a first model parameter of a decompression model, the decompression model is used by the network device to decompress the channel state information sent by the terminal, and the model information is used by the terminal to determine the compression model for compressing the channel state information in combination with the privatized data of the terminal.

26. An apparatus for determining a compression model for compressing channel state information, characterized in that include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: receiving model information sent by a network device, the model information including a first model parameter of a decompression model, the decompression model being used by the network device to decompress the channel state information sent by the terminal; A compression model used by the terminal to compress channel state information is determined according to the model information and private data of the terminal.

27. An apparatus for determining a compression model for compressing channel state information, characterized in that include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: Send model information, the model information including a first model parameter of a decompression model, the decompression model being used by a network device to decompress channel state information sent by a terminal, and the model information being used by the terminal to determine a compression model for compressing the channel state information in combination with the terminal's privatized data.

28. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 13 or the steps of the method described in any one of claims 14 to 23 are implemented.

Images