Information processing method and device and readable storage medium

Abstract

The invention discloses an information processing method and device and a readable storage medium, relates to the technical field of communication, and aims to improve system performance. The method comprises the steps that first CSI at a first moment is determined based on a first model, and input of the first model comprises first input; wherein the first input comprises one or more of the following items: information related to second CSI at a second moment reported by the terminal, and the second moment is a moment before the first moment; and information related to the third CSI at the first moment reported by the terminal. The embodiment of the invention can improve the system performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and in particular to an information processing method and device and a readable storage medium. BACKGROUND

[0002] In a New Radio (NR) system, a Downlink (DL) Type II codebook can obtain good Channel State Information (CSI) feedback accuracy. However, the feedback overhead of the DL Type II codebook is relatively large. Research shows that, based on an Artificial Intelligence (AI) / Machine Learning (ML) model, CSI compression feedback can reduce the overhead of CSI feedback under the same CSI feedback accuracy, or improve the accuracy of CSI feedback under the same CSI feedback overhead.

[0003] As shown in FIG. 1, it is a communication flow diagram of a terminal side and a network side of CSI feedback of a common AI model deployed on both sides: Figure 1

[0004] At the terminal side, the original channel information or the preprocessed channel information is compressed by using a terminal side AI / ML model (also referred to as a CSI generation part, or an AI / ML encoder, etc.), and the compressed information is quantized into binary bits and fed back to the base station;

[0005] At the network side (i.e., the network device side, the base station side), the binary bits fed back by the terminal side are dequantized, and then a base station side AI / ML model (also referred to as a CSI reconstruction part, or an AI / ML decoder, etc.) is used to recover the channel information.

[0006] The above-mentioned CSI compression feedback can be spatial-frequency domain or time-space-frequency domain. The spatial-frequency domain compression refers to the compression of channel information in the spatial domain and the frequency domain. The time-space-frequency domain refers to the compression of channel information in the time domain, the spatial domain and the frequency domain. Since the time-space-frequency domain compression utilizes the correlation in time, compared with the spatial-frequency domain compression, the CSI can be compressed to a smaller number of bits under the same performance.

[0007] When the time-space-frequency domain CSI compression is performed, the CSI reported by the UE can be lost. At this time, the model at the network side can not work normally, thereby affecting the system performance. SUMMARY​

[0008] Embodiments of the present application provide an information processing method, device and readable storage medium to improve system performance.

[0009] In a first aspect, embodiments of the present application provide an information processing method applied to a network device, comprising:

[0010] determining first CSI at a first time based on a first model, wherein an input of the first model comprises first input;

[0011] The first input comprises one or more of the following:

[0012] information related to second CSI reported by a terminal at a second time, the second time being a time before the first time;

[0013] information related to third CSI reported by the terminal at the first time.

[0014] Optionally, the information related to the second CSI reported by the terminal at the second time comprises:

[0015] the second CSI; or

[0016] information obtained by processing the second CSI.

[0017] Optionally, the information obtained by processing the second CSI comprises:

[0018] information obtained by dequantizing the second CSI; or

[0019] information obtained by predicting based on the second CSI; or

[0020] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0021] information obtained by predicting based on first target information, wherein the first target information comprises CSI obtained by predicting a target first output, the target first output being a first output obtained by processing first information as an input of the first model, the first information comprising:

[0022] information determined based on the second CSI and information determined based on fourth CSI; or

[0023] information determined based on the fourth CSI, wherein the fourth CSI is CSI reported by the terminal at a third time, the third time being a time before the second time.

[0024] Optionally, the information determined based on the fourth CSI comprises:

[0025] the fourth CSI; or

[0026] information obtained by dequantizing the fourth CSI; or

[0027] information obtained by prediction based on the fourth CSI; or

[0028] information obtained by prediction based on information obtained by dequantizing the fourth CSI.

[0029] Optionally, the second CSI comprises:

[0030] second CSI of one second time reported by the terminal, or second CSI of multiple second times reported by the terminal.

[0031] Optionally, the second CSI of one second time and the first time are located in a same time window; or

[0032] the multiple second times and the first time are located in a same time window.

[0033] Optionally, the information related to the third CSI of the first time reported by the terminal comprises:

[0034] information determined based on the third CSI of the first time reported by the terminal;

[0035] information determined based on the second CSI.

[0036] Optionally, the information determined based on the third CSI of the first time reported by the terminal comprises:

[0037] the third CSI of the first time reported by the terminal; or

[0038] information obtained by dequantizing the third CSI of the first time reported by the terminal.

[0039] Optionally, the information determined based on the second CSI comprises:

[0040] information obtained by dequantizing the second CSI; or

[0041] information obtained by prediction based on the second CSI; or

[0042] information obtained by prediction based on information obtained by dequantizing the second CSI; or

[0043] information obtained based on the second CSI, wherein the second CSI is obtained by predicting a target second output, the target second output being a second output obtained by taking the second information as an input of the first model and processing the second information by using the first model, and the second information comprises:

[0044] any one of the second CSI and information obtained by dequantizing the second CSI, and any one of the fifth CSI and information obtained by dequantizing the fifth CSI; or

[0045] the fifth CSI; or

[0046] information obtained by dequantizing the fifth CSI; or

[0047] information obtained by predicting based on the fifth CSI; or

[0048] information obtained by predicting based on information obtained by dequantizing the fifth CSI; or

[0049] information related to the fifth CSI;

[0050] wherein the fifth CSI is a CSI reported by the terminal at a fourth time point, and the fourth time point is a time point before the second time point.

[0051] Optionally, the information related to the fifth CSI comprises:

[0052] information determined based on the second CSI;

[0053] information determined based on the fifth CSI.

[0054] Optionally, the input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output.

[0055] if the first input comprises the second CSI, the second input comprises information determined based on the second CSI; or, if the first input comprises information obtained by processing the second CSI, the second input comprises information determined based on the second CSI;

[0056] the first output is used to indicate information determined based on the first CSI;

[0057] the second output is used to indicate information determined based on the third CSI and the second CSI.

[0058] Optionally, the input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output.

[0059] The information related to the third CSI reported by the terminal at the first time point comprises: the third CSI reported by the terminal at the first time point; or information obtained by processing the third CSI;

[0060] The second input comprises information determined based on the second CSI;

[0061] The first output is used for indicating information determined based on the first CSI;

[0062] The second output is used for indicating information determined based on the third CSI and the second CSI.

[0063] Optionally, the information obtained by processing the third CSI comprises:

[0064] Information obtained by dequantizing the third CSI.

[0065] Optionally, the second input comprises information determined based on the second CSI, which comprises:

[0066] Information obtained by dequantizing the second CSI; or

[0067] Information obtained by predicting based on the second CSI; or

[0068] Information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0069] Information obtained by predicting based on third target information, wherein the third target information comprises: CSI obtained by predicting a target third output, the target third output being a third output obtained by processing third information as input of the first model through the first model, the third information comprising:

[0070] The second CSI and information determined based on sixth CSI; or

[0071] Information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0072] Information obtained by predicting based on the second CSI and information determined based on the sixth CSI; or

[0073] Information obtained by predicting based on information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0074] The sixth CSI and information determined based on the sixth CSI; or

[0075] information obtained by processing the sixth CSI and information determined based on the sixth CSI; or

[0076] information related to the sixth CSI and information determined based on the sixth CSI;

[0077] The sixth CSI is CSI reported by the terminal at a fifth time point, and the fifth time point is a time point before the second time point.

[0078] Optionally, the first time point includes a time at which a first reference signal is transmitted in a first period, and the second time point includes a time at which a first reference signal is transmitted in an xth period before the first period, where x is an integer greater than or equal to 1.

[0079] In a second aspect, an embodiment of the present application provides an information processing method, applied to a network device, and including:

[0080] In a first time window, if a sixth time point CSI reported by a terminal is not detected, predicting a CSI at a target time point in the first time window based on a second model, the target time point including the sixth time point and / or a time point after the sixth time point;

[0081] The input of the second model includes:

[0082] a seventh CSI reported by the terminal at a time point before the target time point; or

[0083] information determined based on the seventh CSI reported by the terminal at the time point before the target time point.

[0084] Optionally, the information determined based on the seventh CSI reported by the terminal at the time point before the target time point includes:

[0085] the seventh CSI; or

[0086] information obtained by dequantizing the seventh CSI; or

[0087] information obtained by predicting based on the seventh CSI; or

[0088] information obtained by predicting based on information obtained by dequantizing the seventh CSI.

[0089] In a third aspect, an embodiment of the present application provides an information processing apparatus, applied to a network device, and including a memory, a transceiver, and a processor.

[0090] a memory for storing a computer program; a transceiver for transceiving data under control of the processor; a processor for reading the computer program in the memory and performing the following operations:

[0091] determining a first CSI at a first time based on a first model, wherein an input of the first model comprises a first input;

[0092] wherein the first input comprises one or more of:

[0093] information related to a second CSI reported by the terminal at a second time, the second time being a time before the first time;

[0094] information related to a third CSI reported by the terminal at the first time.

[0095] Optionally, the information related to the second CSI reported by the terminal at the second time comprises:

[0096] the second CSI; or

[0097] information obtained by processing the second CSI.

[0098] Optionally, the information obtained by processing the second CSI comprises:

[0099] information obtained by dequantizing the second CSI; or

[0100] information obtained by predicting based on the second CSI; or

[0101] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0102] information obtained by predicting based on first target information, wherein the first target information comprises: CSI obtained by predicting a target first output, the target first output being a first output obtained by processing first information as an input of the first model by the first model, the first information comprising:

[0103] information determined based on the second CSI and information determined based on fourth CSI; or

[0104] information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time, the third time being a time before the second time.

[0105] Optionally, the information determined based on the fourth CSI comprises:

[0106] the fourth CSI; or

[0107] information obtained by dequantizing the fourth CSI; or

[0108] information obtained by prediction based on the fourth CSI; or

[0109] information obtained by prediction based on information obtained by dequantizing the fourth CSI.

[0110] Optionally, the second CSI comprises:

[0111] second CSI of one second time reported by the terminal, or second CSI of multiple second times reported by the terminal.

[0112] Optionally, the second CSI of one second time and the first time are located in a same time window; or

[0113] the multiple second times and the first time are located in a same time window.

[0114] Optionally, information related to the third CSI of the first time reported by the terminal comprises:

[0115] information determined based on the third CSI of the first time reported by the terminal;

[0116] information determined based on the second CSI.

[0117] Optionally, the information determined based on the third CSI of the first time reported by the terminal comprises:

[0118] the third CSI of the first time reported by the terminal; or

[0119] information obtained by dequantizing the third CSI of the first time reported by the terminal.

[0120] Optionally, the information determined based on the second CSI comprises:

[0121] information obtained by dequantizing the second CSI; or

[0122] information obtained by prediction based on the second CSI; or

[0123] information obtained by prediction based on information obtained by dequantizing the second CSI; or

[0124] information obtained based on the second CSI, wherein the second target information comprises: CSI obtained by predicting a target second output, the target second output being a second output obtained by taking the second information as input of the first model and processing the second information by using the first model, and the second information comprising:

[0125] any one of the second CSI and information obtained by dequantizing the second CSI, and any one of the fifth CSI and information obtained by dequantizing the fifth CSI; or

[0126] the fifth CSI; or

[0127] information obtained by dequantizing the fifth CSI; or

[0128] information obtained by predicting based on the fifth CSI; or

[0129] information obtained by predicting based on information obtained by dequantizing the fifth CSI; or

[0130] information related to the fifth CSI;

[0131] wherein the fifth CSI is CSI reported by the terminal at a fourth time point, and the fourth time point is a time point before the second time point.

[0132] Optionally, the information related to the fifth CSI comprises:

[0133] information determined based on the second CSI;

[0134] information determined based on the fifth CSI.

[0135] Optionally, the input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output.

[0136] if the first input comprises the second CSI, the second input comprises information determined based on the second CSI; or, if the first input comprises information obtained by processing the second CSI, the second input comprises information determined based on the second CSI;

[0137] the first output is used to indicate information determined based on the first CSI;

[0138] the second output is used to indicate information determined based on the third CSI and the second CSI.

[0139] Optionally, the input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output.

[0140] The information related to the third CSI reported by the terminal at the first time point comprises: the third CSI reported by the terminal at the first time point; or information obtained by processing the third CSI;

[0141] The second input comprises information determined based on the second CSI;

[0142] The first output is used for indicating information determined based on the first CSI;

[0143] The second output is used for indicating information determined based on the third CSI and the second CSI.

[0144] Optionally, the information obtained by processing the third CSI comprises:

[0145] Information obtained by dequantizing the third CSI.

[0146] Optionally, the second input comprises information determined based on the second CSI, which comprises:

[0147] Information obtained by dequantizing the second CSI; or

[0148] Information obtained by predicting based on the second CSI; or

[0149] Information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0150] Information obtained by predicting based on third target information, wherein the third target information comprises: CSI obtained by predicting a target third output, the target third output being a third output obtained by taking third information as input of the first model and processing the first model, the third information comprising:

[0151] The second CSI and information determined based on sixth CSI; or

[0152] Information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0153] Information obtained by predicting based on the second CSI and information determined based on the sixth CSI; or

[0154] Information obtained by predicting based on information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0155] The sixth CSI and information determined based on the sixth CSI; or

[0156] information obtained by processing the sixth CSI and information determined based on the sixth CSI; or

[0157] information related to the sixth CSI and information determined based on the sixth CSI;

[0158] The sixth CSI is CSI reported by the terminal at a fifth time point, and the fifth time point is a time point before the second time point.

[0159] Optionally, the first time point includes a time at which a first reference signal is transmitted in a first period, and the second time point includes a time at which a first reference signal is transmitted in an xth period before the first period, where x is an integer greater than or equal to 1.

[0160] In a fourth aspect, an embodiment of the present application provides an information processing apparatus applied to a network device, including: a memory, a transceiver, and a processor.

[0161] The memory is configured to store a computer program; the transceiver is configured to transceive data under control of the processor; and the processor is configured to read the computer program in the memory and perform the following operations:

[0162] In a first time window, if a sixth time point CSI reported by a terminal is not detected, predicting a CSI at a target time point in the first time window based on a second model, the target time point including the sixth time point and / or a time point after the sixth time point;

[0163] The input of the second model includes:

[0164] a seventh CSI reported by the terminal at a time point before the target time point; or

[0165] information determined based on the seventh CSI reported by the terminal at the time point before the target time point.

[0166] Optionally, the information determined based on the seventh CSI reported by the terminal at the time point before the target time point includes:

[0167] the seventh CSI; or

[0168] information obtained by dequantizing the seventh CSI; or

[0169] information obtained by predicting based on the seventh CSI; or

[0170] information obtained by predicting based on information obtained by dequantizing the seventh CSI.

[0171] In a fifth aspect, an embodiment of the present application provides an information processing apparatus applied to a network device, comprising:

[0172] a first processing unit configured to determine first CSI at a first time based on a first model, wherein an input of the first model comprises first input;

[0173] The first input comprises one or more of the following:

[0174] information related to second CSI reported by a terminal at a second time, the second time being a time before the first time;

[0175] information related to third CSI reported by the terminal at the first time.

[0176] In a sixth aspect, an embodiment of the present application provides an information processing apparatus applied to a network device, comprising:

[0177] a first processing unit configured to, if no sixth time CSI reported by a terminal is detected within a first time window, predict CSI at a target time within the first time window based on a second model, the target time comprising the sixth time and / or a time after the sixth time;

[0178] The input of the second model comprises:

[0179] the terminal reports seventh CSI at a time before the target time; or

[0180] information determined based on the seventh CSI reported by the terminal at the time before the target time.

[0181] In a seventh aspect, an embodiment of the present application further provides a processor-readable storage medium, wherein the readable storage medium stores a computer program, and the computer program is executed by a processor to implement steps in the information processing method.

[0182] In an eighth aspect, an embodiment of the present application further provides a computer program product comprising computer instructions, and the computer instructions are executed by a processor to implement steps in the information processing method.

[0183] In the embodiment of the present application, the first input can be processed by the first model to determine the first CSI at the first time, so that the inference of the CSI can still be based on the first model, and the system performance can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0184] Figure 1 is a communication flow diagram of the terminal side and the network side of the CSI feedback of the AI model in the double-sided deployment;

[0185] Figure 2 Fig. 1 is a flow chart of an information processing method according to an embodiment of the present application;

[0186] Fig. 3(a) and Fig. 3(b) are processing diagrams according to an embodiment of the present application;

[0187] Fig. 4(a) and Fig. 4(b) are processing diagrams according to an embodiment of the present application;

[0188] Figure 5 Fig. 5 is a processing diagram according to an embodiment of the present application;

[0189] Figure 6 Fig. 6 is a processing diagram according to an embodiment of the present application;

[0190] Figure 7 Fig. 7 is a processing diagram according to an embodiment of the present application;

[0191] Figure 8 Fig. 8 is a flow chart of an information processing method according to an embodiment of the present application;

[0192] Figure 9 Fig. 9 is a processing diagram according to an embodiment of the present application;

[0193] Figure 10 Fig. 10 is a structure diagram of an information processing device according to an embodiment of the present application;

[0194] Figure 11 Fig. 11 is a structure diagram of an information processing device according to an embodiment of the present application;

[0195] Figure 12 Fig. 12 is a structure diagram of an information processing device according to an embodiment of the present application;

[0196] Figure 13 Fig. 13 is a structure diagram of an information processing device according to an embodiment of the present application. DETAILED DESCRIPTION

[0197] In the embodiments of the present application, the term "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship.

[0198] In the embodiments of the present application, the term "multiple" means two or more, and other quantifiers are similar.

[0199] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0200] The embodiments of the present application provide an information processing method and device to improve system performance.

[0201] The method and the device are based on the same application concept. Since the principles of the method and the device for solving problems are similar, the implementation of the device and the method can be referred to each other, and the repeated parts will not be described again.

[0202] Referring to Figure 2 , Figure 2 is a flowchart of an information processing method provided by the embodiments of the present application, applied to a network device. The network device may, for example, be a base station. As shown in Figure 2 , the following steps are included:

[0203] Step 201: determining a first CSI at a first time based on a first model, wherein the input of the first model includes a first input. The first input includes one or more of the following:

[0204] information related to second CSI reported by a terminal at a second time, the second time being a time before the first time;

[0205] information related to third CSI reported by the terminal at the first time.

[0206] The first model includes but is not limited to an AI model and an ML model. Optionally, the first time includes a time of transmitting a first reference signal in a first period, and the second time includes a time of transmitting a first reference signal in an xth period before the first period, where x is an integer greater than or equal to 1. Examples of the type of the first reference signal include a CSI reference signal (CSI Reference Signal, CSI-RS), a synchronization signal / physical broadcast channel signal block (Synchronization Signal and PBCH block, SSB), and a cell-specific reference signal (Cell-Specific Reference Signal, CRS).

[0207] It should be noted that, in the embodiments of the present application, the meaning of "related to" is a relatively broad definition, which can refer to information included in certain information having a direct correlation with the information, or can refer to information included in certain information not having a direct or obvious correlation with the information.

[0208] In the embodiments of the present application, the first model (such as an AI model or an ML model) can include a model constructed based on a Long Short-Term Memory (LSTM), a transformer, or a Convolutional Neural Network (CNN).

[0209] In the embodiments of the present application, the input of the first model can include multiple forms, and the output of the first model can also include multiple forms. In the following, the specific implementation process of the embodiments of the present application is described in combination with the first model of different structures.

[0210] In an implementation form of the embodiments of the present application, the input of the first model can include a first input and a second input, and the output of the first model can include a first output. The first output can be considered as the first CSI, or information obtained after processing based on the first CSI.

[0211] In an embodiment of the present application, the information related to the second CSI reported by the terminal at the second time includes the second CSI. That is, if the CSI reported by the terminal at the first time cannot be correctly decoded, the second CSI can be used as the first input, and the output can be obtained after processing by using the first model. Optionally, the second CSI can be the CSI reported by the terminal for a time before the first time, and the reported CSI can be correctly decoded by the network device.

[0212] In an embodiment of the present application, the information related to the second CSI reported by the terminal at the second time includes information obtained by processing the second CSI. The processing includes but is not limited to dequantization, prediction, information obtained based on dequantization and then prediction, and the like. In the embodiments of the present application, the manner of prediction is not limited.

[0213] Optionally, the information obtained by processing the second CSI includes:

[0214] information obtained by dequantizing the second CSI; or

[0215] information obtained by predicting based on the second CSI; or

[0216] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0217] information determined based on the second CSI and information determined based on fourth CSI; or

[0218] information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time point, and the third time point is a time point before the second time point.

[0219] information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time point, and the third time point is a time point before the second time point.

[0220] information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time point, and the third time point is a time point before the second time point.

[0221] the fourth CSI; or

[0222] information obtained by dequantizing the fourth CSI; or

[0223] information obtained by predicting based on the fourth CSI; or

[0224] information obtained by predicting based on information obtained by dequantizing the fourth CSI.

[0225] Similarly, how to predict based on the fourth CSI is not limited herein.

[0226] In the above description, the second CSI can include a second CSI reported by the terminal at one second time point, or second CSIs reported by the terminal at multiple second time points. Optionally, the second CSI at the one second time point and the first time point are located in a same time window; or the multiple second time points and the first time point are located in a same time window.

[0227] In the above embodiment, the information determined based on the second CSI includes:

[0228] information obtained by dequantizing the second CSI; or

[0229] information obtained by predicting based on the second CSI; or

[0230] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0231] information obtained based on the second CSI, wherein the second target information comprises: CSI obtained by predicting a target second output, wherein the target second output is a second output obtained by taking the second information as an input of the first model and processing the second information through the first model, and the second information comprises:

[0232] any one of the second CSI and information obtained by dequantizing the second CSI, and any one of the fifth CSI and information obtained by dequantizing the fifth CSI; or

[0233] the fifth CSI; or

[0234] information obtained by dequantizing the fifth CSI; or

[0235] information obtained by predicting based on the fifth CSI; or

[0236] information obtained by predicting based on information obtained by dequantizing the fifth CSI; or

[0237] information related to the fifth CSI;

[0238] wherein the fifth CSI is CSI reported by the terminal at a fourth time point, and the fourth time point is a time point before the second time point.

[0239] Optionally, the information related to the fifth CSI comprises:

[0240] information determined based on the second CSI; and information determined based on the fifth CSI.

[0241] Optionally, in the information related to the fifth CSI, the information determined based on the second CSI can comprise information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI, etc. The information determined based on the fifth CSI can comprise information obtained by dequantizing the fifth CSI; or information obtained by predicting based on the fifth CSI; or information obtained by predicting based on information obtained by dequantizing the fifth CSI, etc.

[0242] As shown in FIG. 3(a), which is a schematic diagram of an embodiment of the present application, the simplified schematic diagram of FIG. 3(a) can be as shown in FIG. 3(b). In FIG. 3(a), for example, for the terminal side, inference is performed by the AI model on the terminal side (schematically shown as UE-side AI model in the figure); for the network device, inference is performed by the AI model on the network device side (schematically shown as NW-side AI model in the figure). For the terminal side, the CSI output at the current time or the dequantization information of the CSI can be used as the input at the subsequent time after the current time for the network device side. For the network device side, when performing inference at the current time by using the first model, the first input is the CSI fed back by the terminal at the current time, and the second input is the CSI fed back by the terminal at the previous time of the current time. The current time can be understood as the time at which inference is performed by using the first model.

[0243] For example, as can be seen from FIG. 3(b), for inference at the current time, the inputs of the network device are respectively the CSI fed back by the terminal at the previous time (or the dequantization information of the CSI at the previous time), and the CSI fed back by the terminal at the current time (or the dequantization information of the CSI at the current time), which can be considered as being located in the same time window.

[0244] As shown in FIG. 4(a), which is a schematic diagram of an embodiment of the present application, the simplified schematic diagram of FIG. 4(a) can be as shown in FIG. 4(b). In FIG. 4(a), for example, for the terminal side, inference is performed by the AI model on the terminal side (schematically shown as UE-side AI model in the figure); for the network device, inference is performed by the AI model on the network device side (schematically shown as NW-side AI model in the figure). For the terminal side, the CSI output at the current time or the dequantization information of the CSI can be used as the input at the subsequent multiple times after the current time for the network device side. For the network device side, when performing inference at the current time by using the first model, the first input is the CSI fed back by the terminal at the current time, and the second input is the CSI fed back by the terminal at the multiple times before the current time.

[0245] For example, as can be seen from FIG. 4(b), for inference at the current time, the inputs are respectively the CSI fed back by the terminal at the previous multiple times (or the dequantization information of the CSI at the previous multiple times), and the CSI fed back by the terminal at the current time (or the dequantization information of the CSI at the current time), which can be considered as being located in the same time window.

[0246] In conjunction with Figure 3(a), for the AI ​​model on the network device side, except for some special moments (reset moments, such as time t), the AI ​​model of the AI ​​model on the network device side includes a first input, a second input, and a first output. When the CSI fed back (or reported) by the terminal can be correctly decoded, taking time t+1 as an example, the first input is the CSI of the current moment fed back by the terminal (CSI FB (t+1)) (or, it can be the CSI at the current moment fed back by the terminal (CSI FB (t+1)) dequantization information), the second input is the CSI (CSI) of the previous moment fed back by the terminal FB (t)) (or, it can be the CSI of the previous moment fed back by the terminal (CSI FB (t)) dequantized information), the first output includes the CSI (CSI) of the current moment reconstructed by the network device output (t+1)).

[0247] In the case where the CSI fed back by the terminal cannot be correctly decoded, for example, as shown in FIG3(a), at time n, the network device fails to receive the CSI fed back by the terminal at time n (i.e., the CSI fed back by the terminal). FB (n) lost), the network device will send the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) (or the CSI at the previous moment fed back by the terminal (CSI FB (n-1)) is used as the first input to perform AI model inference and obtain the reconstructed CSI (CSI) at time n. output (n)). Among them, the CSI of the previous moment fed back by the terminal (CSI FB (n-1)) is an example of the aforementioned second CSI.

[0248] That is, in this case, when the CSI fed back by the terminal cannot be decoded correctly, the network device uses the CSI (CSI FB (n-1)) (or the CSI at the previous moment fed back by the terminal (CSI FB (n-1)) dequantized information) instead of the current CSI that should have been fed back (originally CSI FB (n)), perform AI model inference, and obtain the CSI (CSI) at time n reconstructed by the network device output (n)).

[0249] Combined with Figure 3 (a), when the CSI fed back by the terminal cannot be decoded correctly, for example, at time n, the network device fails to receive the CSI fed back by the terminal at time n (i.e., the CSI fed back by the terminal FB(n) lost), the network device will send the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) is used to predict the CSI (CSI predict (n)) (or the CSI (CSI) at the previous moment fed back by the terminal FB (n-1)) is used as the first input and the CSI (CSI) of the previous moment fed back by the terminal is used as the first input. FB (n-1)) (or, it can be the CSI of the previous moment fed back by the terminal (CSI FB (n-1)) is used as the second input to perform AI model inference and obtain the reconstructed CSI (CSI) at time n. output (n)). Among them, the CSI of the previous moment fed back by the terminal (CSI FB (n-1)) is an example of the aforementioned second CSI.

[0250] That is, in this case, when the CSI fed back by the terminal cannot be decoded correctly, the network device uses the CSI (CSI FB (n-1)) is used to predict the CSI (CSI predict (n)) (or the CSI (CSI) at the previous moment fed back by the terminal FB (n-1)) is used to predict the information obtained by dequantizing the information) instead of the CSI at the current moment that should have been used for the final feedback (originally CSI FB (n)), perform AI model inference, and obtain the CSI (CSI) at time n reconstructed by the network device output (n)).

[0251] Combined with Figure 3 (a), when the CSI fed back by the terminal cannot be decoded correctly, for example, at time n, the network device fails to receive the CSI fed back by the terminal at time n (i.e., the CSI fed back by the terminal FB (n) loss), the network device uses the CSI obtained by predicting the output at a certain moment before the current moment as the first input, and the CSI (CSI) fed back by the terminal at the previous moment FB (n-1)) (or, it can be the CSI of the previous moment fed back by the terminal (CSI FB (n-1)) is used as the second input to perform AI model inference and obtain the reconstructed CSI (CSI) at time n. output (n)).

[0252] For example, the output at a certain moment before the current moment may be the output at the moment before the current moment. The output at the previous moment may be inferred based on the CSI at the previous moment fed back by the terminal and the CSI at the moment before the previous moment fed back by the terminal (in the case where the CSI at the previous moment fed back by the terminal can be correctly decoded), or it may be inferred based on the CSI at the moment before the previous moment fed back by the terminal (in the case where the CSI at the previous moment fed back by the terminal cannot be correctly decoded).

[0253] For example, in this case, it is assumed that the CSI (CSI FB (n-1)) is correctly decoded (i.e., an example of the second CSI), then the target first output can be considered as the CSI (CSI) of the previous moment fed back by the terminal. FB (n-1)) and the CSI fed back by the terminal at time n-2 (CSI FB (n-2)) is used as the input of the first model and is obtained after being processed by the first model; Assume that the CSI (CSI) at the previous moment of the terminal feedback FB (n-1)) is not correctly decoded (i.e., an example of the second CSI), the target first output can be considered as the CSI (CSI) fed back by the terminal at time n-2 FB (n-2)) is used as the input of the first model and is obtained after being processed by the first model. The terminal here is the CSI (CSI) fed back at time n-2 FB (n-2)) can be used as an example of the fourth CSI.

[0254] Since the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) may not be accurately decoded, so instead, the terminal can feed back the CSI (CSI) at time n-2. FB (n-2)), can be used as an example of the fifth CSI, thereby replacing the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) perform the above-mentioned processing.

[0255] That is, for this situation, at time n+1, when the CSI fed back by the terminal cannot be decoded correctly, the network device will decode the CSI (CSI FB (n+1)) The CSI fed back by the terminal at the current moment FB (n) is replaced with the corresponding input of the moment before the current moment as the input of the AI ​​model, and the AI ​​model is inferred to obtain the CSI (CSI) at the moment n+1 reconstructed by the base station. output (n+1)).

[0256] In conjunction with Figure 4(a), for the AI ​​model on the network device side, except for some special moments (reset moments, such as time t), the AI ​​model of the AI ​​model on the network device side includes a first input, a second input, and a first output. When the CSI fed back (or reported) by the terminal can be correctly decoded, taking time t+1 as an example, the first input is the CSI of the current moment fed back by the terminal (CSI FB (t+1)) (or, it can be the CSI at the current moment fed back by the terminal (CSI FB (t+1)) dequantized information), the second input is the CSI of one or more moments before the current moment fed back by the terminal (or it can also be the dequantized information of the CSI of one or more moments before the current moment fed back by the terminal), the first output includes the CSI of the current moment reconstructed by the network device (CSI output (t+1)). That is, in this embodiment, the CSI or its dequantized information fed back by the terminal at each moment can be used as the second input at each moment after that moment.

[0257] In the case where the CSI fed back by the terminal cannot be correctly decoded, for example, as shown in FIG4(a), at time n, the network device fails to receive the CSI fed back by the terminal at time n (i.e., the CSI fed back by the terminal). FB (n) is lost), the network device takes the CSI of multiple moments before the current moment fed back by the terminal (or the dequantized information of the CSI of multiple moments before the current moment fed back by the terminal) as the first input, performs inference on the AI ​​model, and obtains the reconstructed CSI (CSI) at moment n. output (n)). Here, the CSI of multiple time moments before the current time moment fed back by the terminal is an example of the second CSI.

[0258] That is, in this case, when the CSI fed back by the terminal cannot be correctly decoded, the network device uses the CSI of the previous multiple moments fed back by the terminal (or the dequantized information of the CSI of the previous multiple moments fed back by the terminal) to replace the CSI of the current moment that should have been fed back by the terminal (the CSI that should have been fed back originally). FB (n)), perform AI model inference, and obtain the CSI (CSI) at time n reconstructed by the network device output (n)).

[0259] Combined with Figure 4 (a), when the CSI fed back by the terminal cannot be decoded correctly, for example, at time n, the network device fails to receive the CSI fed back by the terminal at time n (i.e., the CSI fed back by the terminal FB (n) lost), the network device will send the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) is used to predict the CSI (CSIpredict (n)) (or information obtained by predicting dequantization information of the CSI (CSI FB (n-1)) fed back by the terminal) as the first input, the CSI fed back by the terminal at multiple time points before the current time point as the second input, performs inference of the AI model, and obtains the reconstructed CSI (CSI output (n)) at time point n. Here, the CSI (CSI FB (n-1)) fed back by the terminal is an example of the second CSI.

[0260] That is, in this case, in the case where the CSI fed back by the terminal cannot be correctly decoded, the network device substitutes the CSI (CSI predict (n)) obtained by predicting the CSI fed back by the terminal at multiple time points before the current time point (or information obtained by predicting dequantization information of the CSI fed back by the terminal at multiple time points before the current time point) for the CSI (CSI FB (n)) that should have been used at the current time point, performs inference of the AI model, and obtains the CSI (CSI output (n)) reconstructed by the network device at time point n.

[0261] In combination with FIG. 4(a), in the case where the CSI fed back by the terminal cannot be correctly decoded, for example, at time point n, the network device fails to receive the CSI (CSI FB (n) fed back by the terminal at time point n (i.e., the CSI (CSI output (n)) fed back by the terminal is lost), the network device substitutes the CSI obtained by predicting based on the output at a certain time point before the current time point as the first input, the CSI fed back by the terminal at multiple time points before the current time point (or dequantization information of the CSI fed back by the terminal at multiple time points before the current time point) as the second input, performs inference of the AI model, and obtains the reconstructed CSI (CSI output (n)) at time point n.

[0262] That is, in this case, at time point n+1, in the case where the CSI fed back by the terminal cannot be correctly decoded, the network device substitutes the CSI (CSI FB (n+1)) fed back by the terminal at the current time point, the CSI (CSI FB (n) fed back by the terminal at the current time point for the corresponding input at the time point one time before the current time point as the input of the AI model, performs inference of the AI model, and obtains the CSI (CSI output (n+1)) reconstructed by the base station at time point n+1.

[0263] For example, in this case, it is assumed that the CSI (CSI FB(n-1)) is correctly decoded (i.e., an example of the second CSI), then the target first output can be considered as the CSI (CSI) of the previous moment fed back by the terminal. FB (n-1)), the terminal feeds back CSI at time n-2 (CSI FB (n-2)), the terminal is the CSI fed back at time n-3 (CSI FB (n-3)) etc. are used as the input of the first model and are obtained after being processed by the first model; Assume that the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) is not correctly decoded (i.e., an example of the second CSI), the target first output can be considered as the CSI (CSI) fed back by the terminal at time n-2 FB (n-2)), the terminal is the CSI fed back at time n-3 (CSI FB (n-3)) etc. are used as the input of the first model and are obtained after being processed by the first model. The terminal here is the CSI (CSI) fed back at time n-2 FB (n-2)), the terminal is the CSI fed back at time n-3 (CSI FB (n-3)) etc. can be used as an example of the fourth CSI.

[0264] In one embodiment of the present application, the information related to the third CSI at the first moment reported by the terminal includes: information determined based on the third CSI at the first moment reported by the terminal; and information determined based on the second CSI. Optionally, because the second moment is before the first moment, the second CSI and the third CSI may be different.

[0265] Optionally, the information determined based on the third CSI at the first moment reported by the terminal includes the third CSI at the first moment reported by the terminal; or information obtained by dequantizing the third CSI at the first moment reported by the terminal. The meaning of the information determined based on the second CSI may refer to the description of the preceding embodiment.

[0266] In conjunction with FIG3 (a), when the CSI fed back by the terminal cannot be correctly decoded, that is, as shown in FIG3 (a), at time n, the network device fails to receive the CSI fed back by the terminal at time n (that is, the CSI fed back by the terminal FB (n) loss), the network device will be based on the CSI (CSI) of the previous moment fed back by the terminal. FB (n-1)) (or the CSI at the previous moment fed back by the terminal (CSI FB The output (CSI) obtained by using the dequantized information of (n-1)) as the first input to perform inference on the AI ​​model out (n-1)) is output as the CSI at time n.

[0267] At time n+1, in the case that the CSI fed back by the terminal can be correctly decoded, the first input of the first model is the CSI fed back by the terminal at the current time (CSI FB (t+1)) (or the dequantization information of the CSI fed back by the terminal at the current time (CSI FB (t+1)), and the second input is information determined based on the CSI at a time before the current time fed back by the terminal, i.e., information determined based on the second CSI.

[0268] For example, since the inference at time n+1 needs to use the CSI at time n fed back by the terminal (or the dequantization information of the CSI at time n fed back by the terminal), but the CSI at time n fed back by the terminal fails to be received, at this time, the inference can be performed based on the CSI at a time before time n (such as time n-1, assuming that the CSI at time n-1 fed back by the terminal can be correctly decoded) as the second input.

[0269] Similarly, in combination with FIG. 3(b), since the inference at time n+1 needs to use the CSI at time n fed back by the terminal (or the dequantization information at time n fed back by the terminal), but the CSI at time n fed back by the terminal fails to be received, at this time, the inference can be performed based on the CSI at a plurality of times before time n (such as time n-1, n-2, n-3, etc., assuming that the CSI at time n-1, n-2, n-3 fed back by the terminal can be correctly decoded) as the second input.

[0270] II. In an implementation form of the embodiment of the present application, the input of the first model can include a first input and a second input, and the output includes a first output and a second output. The first input can include the CSI at the current time fed back by the terminal, the second input can include information determined based on the CSI at a time before the current time fed back by the terminal (which can be considered as historical accumulated information), the first output can be considered as the first CSI or information obtained by processing based on the first CSI, etc., and the second output includes the information determined based on the CSI at the current time fed back by the terminal and the CSI at a time before the current time fed back by the terminal (which can be considered as historical accumulated information).

[0271] In one embodiment of the present application, the first input includes the second CSI, and the second input includes information determined based on the second CSI; or, the first input includes information obtained by processing the second CSI, and the second input includes information determined based on the second CSI. In this embodiment, the information obtained by processing the second CSI includes information obtained by dequantizing the second CSI. The first output is used to indicate information determined based on the first CSI; and the second output is used to indicate information determined based on the third CSI and the second CSI.

[0272] For example, Figure 5 As shown in the figure, except for some special moments (reset moments, such as time t), the AI ​​model of the network device (shown as the NW-side AI model in the figure) includes two inputs and two outputs. When the CSI fed back by the terminal can be correctly decoded, taking time t+1 as an example, the first input includes the CSI of the current moment fed back by the terminal (CSI FB (t+1)), the second input includes historical accumulated information O determined based on the CSI of the historical moments before the current moment fed back by the terminal NW (t); The first output includes the CSI (CSI) of the current moment reconstructed by the network device output (t+1)), the second output includes the historical accumulated information O determined by the network device based on the CSI fed back by the terminal at the current moment and the CSI fed back at the historical moment NW (t+1).

[0273] In the case that the CSI fed back by the terminal cannot be decoded correctly, that is, Figure 5 As shown, at time n, the network device fails to receive the CSI at time n fed back by the terminal (the CSI fed back by the terminal FB (n) is lost), the network device will send the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) as the first input, the historical accumulated information O determined based on the CSI of the historical moment fed back by the terminal previously obtained NW (n-1) is used as the second input for AI model reasoning, and the CSI (CSI) at time n reconstructed by the network device is obtained. output (n)) and historical accumulated information O NW (n). Here, the CSI (CSI) of the previous moment fed back by the terminal FB (n-1)) can be used as an example of the second CSI, and the CSI at the historical moment of terminal feedback can be used as an example of the sixth CSI.

[0274] That is, in this case, when the CSI fed back by the terminal cannot be decoded correctly, the network device uses the CSI (CSIFB (n-1)) (or dequantization information of the CSI (CSI FB (n-1)) fed back by the terminal at the previous time) instead of the CSI (CSI FB (n)) that should have been fed back by the terminal at the current time, performs inference of the AI model, and obtains the CSI (CSI output (n)) reconstructed by the network device.

[0275] In this way, even if some CSI information fed back by the terminal at some time is lost, the network device can still perform inference of the CSI information at subsequent times by using the AI model, thereby improving system performance.

[0276] Optionally, the AI algorithm at the terminal side and / or the AI algorithm at the network device side are reset every T (T ≥ 1) times, or the AI algorithm at the terminal side and / or the AI algorithm at the network device side are reset based on an agreement (for example, a protocol agreement, the terminal indicating a reset time to the network device, the network device indicating a reset time to the terminal, etc.). The period between two resets can be referred to as a detection window. The above method can be performed in each detection window.

[0277] In an embodiment of the present application, the first input includes information obtained by processing the second CSI, and the second input includes information determined based on the second CSI. The first output is used to indicate information determined based on the first CSI, and the second output is used to indicate information determined based on the third CSI and the second CSI. In this embodiment, the information obtained by processing the second CSI includes:

[0278] information obtained by predicting based on the second CSI; or

[0279] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0280] information obtained by predicting based on third target information, wherein the third target information includes: CSI obtained by predicting a target third output, the target third output being a third output obtained by processing third information as the input of the first model, and the third information including:

[0281] the second CSI and information determined based on the sixth CSI; or

[0282] information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0283] information obtained by predicting the second CSI and information determined based on the sixth CSI; or

[0284] information obtained by predicting information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0285] the sixth CSI and information determined based on the sixth CSI; or

[0286] information obtained by processing the sixth CSI and information determined based on the sixth CSI; or

[0287] information related to the sixth CSI and information determined based on the sixth CSI;

[0288] The sixth CSI is the CSI at the fifth moment reported by the terminal, and the fifth moment is a moment before the second moment.

[0289] In this case, since the second input can be considered as historical accumulated information, which is obtained by combining the CSI reported by the terminal at the current moment and the CSI reported by the terminal at historical moments before the current moment, the CSI reported by the terminal at historical moments before the current moment can be considered as the sixth CSI here.

[0290] For example, Figure 6 As shown, except for some special moments (reset moments, such as time t), the AI ​​model at the network device includes two inputs and two outputs. When the CSI fed back by the terminal can be correctly decoded, taking time t+1 as an example, the first input includes the CSI of the current moment fed back by the terminal (CSI FB (t+1)), the second input includes historical accumulated information O determined based on the CSI of the historical moment fed back by the terminal. NW (t); The first output includes the CSI (CSI) of the current moment reconstructed by the network device output (t+1)), the second output includes the historical accumulated information O determined by the network device based on the CSI at the current moment and the CSI at the previous moment fed back by the terminal. NW (t+1).

[0291] When the CSI fed back by the terminal cannot be decoded correctly, such as Figure 6 As shown, at time n, the network device fails to receive the CSI at time n fed back by the terminal (the CSI fed back by the terminal FB (n) lost), the network device will predict the CSI (CSI) based on the previous CSI fed back by the terminal. predict (n)) as the first input, the historical accumulated information O determined based on the CSI of the historical moment fed back by the terminal previously obtainedNW (n-1) as the second input to perform inference of the AI model, to obtain the CSI (CSI output (n)) reconstructed by the network device at time n and the historical accumulated information O NW (n).

[0292] In the embodiments of the present application, the CSI predict (n) can be obtained by prediction based on the CSI output by the terminal at time n-1 and before time n, for example, based on CSI FB (n-1) and CSI FB (n-2), etc.

[0293] Here, the CSI (CSI FB (n-1)) and CSI FB (n-2) fed back by the terminal can be an example of the second CSI.

[0294] That is, in this case, when the CSI fed back by the terminal cannot be correctly decoded, the network device uses the CSI (CSI FB (n-1)) predicted based on the CSI fed back by the terminal at the previous time to obtain the CSI (CSI predict (n)) (or information obtained by predicting the dequantization information of the CSI (CSI FB (n-1)) fed back by the terminal at the previous time) to replace the CSI (CSI FB (n)) that should have been used, performs inference of the AI model, and obtains the CSI (CSI output (n)) reconstructed by the network device at time n.

[0295] Optionally, the AI algorithm on the terminal side and / or the AI algorithm on the network device side is reset every T (T≥1) times, or the AI algorithm on the terminal side and / or the AI algorithm on the network device side is reset based on an agreement (for example, a protocol agreement, the terminal indicating a reset time to the network device, the network device indicating a reset time to the terminal, etc.). The period between two resets can be referred to as a detection window. The above method can be performed in each detection window.

[0296] In one embodiment of the present application, the first input includes third CSI reported by the terminal at a first moment; or information obtained by processing the third CSI, optionally information obtained by dequantizing the third CSI; the second input includes information determined based on the second CSI. The first output indicates information determined based on the first CSI; the second output indicates information determined based on the third CSI and the second CSI. The meaning of the information determined based on the second CSI can be found in the description of the previous embodiment.

[0297] For example, Figure 7 As shown, except for some special moments (reset moments, such as time t), the AI ​​model at the network device includes two inputs and two outputs. When the CSI fed back by the terminal can be correctly decoded, taking time t+1 as an example, the first input includes the CSI of the current moment fed back by the terminal (CSI FB (t+1)), the second input includes historical accumulated information O determined based on the CSI of the historical moment fed back by the terminal. NW (t); The first output includes the CSI (CSI) of the current moment reconstructed by the network device output (t+1)), the second output includes the historical accumulated information O determined by the network device based on the CSI at the current moment and the CSI at the previous moment fed back by the terminal. NW (t+1).

[0298] When the CSI fed back by the terminal cannot be decoded correctly, such as Figure 7 As shown, at time n, the network device fails to receive the CSI at time n fed back by the terminal (the CSI fed back by the terminal FB (n) is lost), the network device will output the AI ​​model obtained at the last moment (CSI out (n-1)) is output as the CSI at time n, that is, the network device considers the CSI out (n) = CSI out (n-1).

[0299] At time n+1, if the CSI fed back by the terminal can be decoded correctly, the network device will receive the CSI (CSI FB (n+1)) as the first input, and the historical accumulated information O determined based on the CSI feedback of the terminal historical moment is obtained previously. NW (n-1) is used as the second input for AI model reasoning, and the CSI (CSI) at time n+1 reconstructed by the network device is obtained. output (n+1)), and historical accumulated information O NW (n+1).

[0300] Here, the CSI fed back by the terminal at the current moment (CSI FB (n+1)) can be taken as an example of the third CSI.

[0301] That is, for this case, at the moment n+1, in the case that the CSI fed back by the terminal cannot be correctly decoded, the network device replaces the CSI fed back by the terminal at the current moment (CSI FB (n+1)) with the CSI fed back by the terminal at the previous moment of the current moment (CSI FB (n) to perform inference of the AI model on the corresponding input at the previous moment of the current moment, to obtain the CSI reconstructed by the base station at the moment n+1 (CSI output (n+1)).

[0302] Optionally, the AI algorithm on the terminal side and / or the AI algorithm on the network device side are reset once every T (T≥1) moments, or the AI algorithm on the terminal side and / or the AI algorithm on the network device side are reset based on an agreement (for example, a protocol agreement, the terminal indicating a reset moment to the network device, the network device indicating a reset moment to the terminal, etc.). The period between two resets can be referred to as a detection window. The above method can be performed in each detection window.

[0303] Referring to Figure 8 , Figure 8 is a flowchart of an information processing method provided by an embodiment of the present application, and is applied to a network device. The network device can be, for example, a base station. As shown in Figure 8 , the method comprises the following steps:

[0304] Step 801: within a first time window, if a sixth moment CSI reported by a terminal is not detected, predicting a CSI at a target moment within the first time window based on a second model, the target moment including the sixth moment and / or a moment after the sixth moment;

[0305] The input of the second model comprises:

[0306] a seventh CSI reported by the terminal for a moment before the target moment; or

[0307] information determined based on the seventh CSI reported by the terminal for the moment before the target moment.

[0308] The first model includes but is not limited to an AI model, an ML model or other forms of models.

[0309] Optionally, the information determined based on the seventh CSI reported by the terminal for the moment before the target moment comprises:

[0310] the seventh CSI; or

[0311] information obtained by dequantizing the seventh CSI; or

[0312] information obtained by prediction based on the seventh CSI; or

[0313] information obtained by prediction based on information obtained by dequantizing the seventh CSI.

[0314] In the embodiments of the present application, the manner of prediction is not limited. The first time window can be considered as an arbitrary time window, and the size thereof can be set as needed.

[0315] In an implementation form of the embodiments of the present application, the input of the first model can include a first input and a second input, and the output includes a first output and a second output. The first input can include the CSI of the current moment fed back by the terminal, the second input can include information determined based on the CSI of the moment before the current moment fed back by the terminal (which can be considered as historical accumulated information), the first output can be considered as the first CSI or information obtained by processing based on the first CSI, and the second output includes information determined based on the CSI of the current moment fed back by the terminal and the CSI of the moment before the current moment fed back by the terminal (which can be considered as historical accumulated information).

[0316] For example, as shown in FIG. 6, the network device side resets every T (T≥1) moments, and the period between two resets is called a detection window. In each detection window, in addition to the reset moment, such as moment t, the AI model of the network device includes two inputs and two outputs. Figure 9

[0317] In the case that the CSI fed back by the terminal can be correctly decoded, taking moment t+1 as an example, the first input includes the CSI of the current moment fed back by the terminal (CSI FB (t+1)), the second input includes historical accumulated information O NW (t) determined based on the CSI of the historical moment before the current moment fed back by the terminal; the first output includes the CSI of the current moment reconstructed by the network device (CSI output (t+1)), and the second output includes historical accumulated information O NW (t+1) determined by the network device based on the CSI of the current moment and the historical moment fed back by the terminal.

[0318] For a time window, assuming that the CSI fed back by the terminal before moment n can be correctly decoded, the network device takes the CSI of moment n fed back by the terminal (CSI FB (n)) and historical accumulated information O NW (n-1) determined based on the CSI of the historical moment fed back by the terminal as the input of the AI model, and obtains the CSI of moment n (CSI​output (n))).

[0319] At time n, if the network device fails to detect the CSI of time n fed back by the terminal (the CSI fed back by the terminal is lost), the network device can use the information related to the CSI fed back by the terminal at a time before time n (which can be an example of the seventh CSI) (or information determined based on the CSI fed back by the terminal at a time before time n) to predict the CSI at time n and / or a time after time n in the time window from time n to the next time when the AI algorithm is reset (to the next time window) (that is, the time window from time n and / or a time after time n). FB (n) is lost), the network device can use the information related to the CSI fed back by the terminal at a time before time n (which can be an example of the seventh CSI) (or information determined based on the CSI fed back by the terminal at a time before time n) to predict the CSI at time n and / or a time after time n in the time window from time n to the next time when the AI algorithm is reset (to the next time window) (that is, the time window from time n and / or a time after time n).

[0320] Optionally, the information related to the CSI fed back by the terminal at a time before time n can include:

[0321] The CSI fed back by the terminal at a time before time n, or information obtained by prediction based on the CSI fed back by the terminal at a time before time n, or dequantization information of the CSI fed back by the terminal at a time before time n, or information obtained by prediction based on the dequantization information of the CSI fed back by the terminal at a time before time n.

[0322] In the embodiments of the present application, the first input can be processed through the first model to determine the first CSI at the first time, so that the inference of the CSI can still be based on the first model, and the system performance can be improved.

[0323] The technical solutions provided by the embodiments of the present application can be applied to various systems, especially 5G systems. For example, the applicable systems can be global system of mobile communication (GSM) systems, code division multiple access (CDMA) systems, wideband code division multiple access (WCDMA) general packet radio service (GPRS) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, long term evolution advanced (LTE-A) systems, universal mobile system (UMTS), worldwide interoperability for microwave access (WiMAX) systems, 5G new radio (NR) systems, and the like. Among these various systems, there are terminal devices and network devices. The system can also include a core network part, such as an evolved packet system (EPS), a 5G system (5GS), and the like.

[0324] The terminal device to which the embodiments of the present application relate can refer to a device that provides voice and / or data connectivity to a user, a handheld device having a wireless connection function, or other processing devices connected to a wireless modem, etc. In different systems, the name of the terminal device can also be different, for example, in the 5G system, the terminal device can be called user equipment (User Equipment, UE). The wireless terminal device can communicate with one or more core networks (Core Network, CN) through a radio access network (Radio Access Network, RAN). The wireless terminal device can be a mobile terminal device, such as a mobile phone (also known as a "cellular" phone) and a computer with a mobile terminal device, for example, it can be a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device that exchanges language and / or data with a radio access network. For example, personal communication service (Personal Communication Service, PCS) phones, cordless phones, session initiation protocol (Session Initiated Protocol, SIP) phones, wireless local loop (Wireless Local Loop, WLL) stations, personal digital assistants (Personal Digital Assistant, PDA) and the like. The wireless terminal device can also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile, a remote station, an access point, a remote terminal, an access terminal, a user terminal, a user agent, a user device, which is not limited in the embodiments of the present application.

[0325] The network device related to the embodiments of the present application can be a base station, which can include multiple cells serving terminals. According to different application scenarios, the base station can also be referred to as an access point, or can be a device in an access network that communicates with wireless terminal devices through one or more sectors over an air interface, or other names. The network device can be used to exchange received air frames and Internet Protocol (IP) packets as a router between wireless terminal devices and the rest of the access network, which can include an Internet Protocol (IP) communication network. The network device can also coordinate the management of the properties of the air interface. For example, the network device related to the embodiments of the present application can be a network device (Base Transceiver Station, BTS) in the Global System for Mobile Communications (GSM) or Code Division Multiple Access (CDMA), or a network device (NodeB) in Wide-band Code Division Multiple Access (WCDMA), or an evolved network device (evolutional Node B, eNB or e-NodeB) in a long term evolution (LTE) system, or a 5G base station (gNB) in a next generation system, or a Home evolved Node B (HeNB), a relay node, a femto, a pico, etc., which are not limited in the embodiments of the present application. In some network structures, the network device can include a centralized unit (CU) node and a distributed unit (DU) node, and the centralized unit and the distributed unit can also be geographically separated.

[0326] Network devices and terminal devices can each use one or more antennas for Multiple Input Multiple Output (MIMO) transmission. MIMO transmission can be either Single User MIMO (SU-MIMO) or Multi User MIMO (MU-MIMO). Depending on the configuration and number of antenna combinations, MIMO transmission can be 2D-MIMO, 3D-MIMO, FD-MIMO, or Massive-MIMO. It can also use diversity transmission, precoding, or beamforming.

[0327] like Figure 10 As shown, the information processing apparatus of an embodiment of the present application is applied to a network device, including: a processor 1000, configured to read a program in a memory 1020, and execute the following process:

[0328] Determining a first CSI at a first moment based on a first model, wherein an input of the first model includes a first input;

[0329] The first input includes one or more of the following:

[0330] Information related to second CSI at a second moment reported by the terminal, where the second moment is a moment before the first moment;

[0331] Information related to the third CSI at the first moment reported by the terminal.

[0332] The transceiver 1010 is configured to receive and send data under the control of the processor 1000 .

[0333] Among them, Figure 10 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 1000 and memory represented by memory 1020. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are all well known in the art and, therefore, will not be described further herein. The bus interface provides an interface. The transceiver 1010 may be a plurality of components, i.e., including a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The processor 1000 is responsible for managing the bus architecture and general processing, and the memory 1020 may store data used by the processor 1000 when performing operations.

[0334] The processor 1000 can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or a complex programmable logic device (CPLD). The processor can also be a multi-core architecture.

[0335] The processor 1000 is responsible for managing the bus architecture and general processing, and the memory 1020 can store data used by the processor 1000 when performing operations.

[0336] Optionally, the information related to the second CSI reported by the terminal at the second time point includes:

[0337] The second CSI; or

[0338] Information obtained by processing the second CSI.

[0339] Optionally, the information obtained by processing the second CSI includes:

[0340] Information obtained by dequantizing the second CSI; or

[0341] Information obtained by predicting based on the second CSI; or

[0342] Information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0343] Information obtained by predicting based on first target information, wherein the first target information includes: CSI obtained by predicting a target first output, the target first output being a first output obtained by processing first information as input of the first model through the first model, and the first information including:

[0344] Information determined based on the second CSI and information determined based on fourth CSI; or

[0345] Information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time point, and the third time point is a time point before the second time point.

[0346] Optionally, the information determined based on the fourth CSI includes:

[0347] The fourth CSI; or

[0348] information obtained by dequantizing the fourth CSI; or

[0349] information obtained by prediction based on the fourth CSI; or

[0350] information obtained by prediction based on information obtained by dequantizing the fourth CSI.

[0351] Optionally, the second CSI comprises:

[0352] second CSI of one second time reported by the terminal, or second CSI of multiple second times reported by the terminal.

[0353] Optionally, the second CSI of one second time and the first time are located in a same time window; or

[0354] the multiple second times and the first time are located in a same time window.

[0355] Optionally, information related to the third CSI of the first time reported by the terminal comprises:

[0356] information determined based on the third CSI of the first time reported by the terminal;

[0357] information determined based on the second CSI.

[0358] Optionally, the information determined based on the third CSI of the first time reported by the terminal comprises:

[0359] the third CSI of the first time reported by the terminal; or

[0360] information obtained by dequantizing the third CSI of the first time reported by the terminal.

[0361] Optionally, the information determined based on the second CSI comprises:

[0362] information obtained by dequantizing the second CSI; or

[0363] information obtained by prediction based on the second CSI; or

[0364] information obtained by prediction based on information obtained by dequantizing the second CSI; or

[0365] information obtained by prediction based on second target information, wherein the second target information comprises: CSI obtained by prediction of a target second output, the target second output being a second output obtained by processing second information as input of the first model by the first model, the second information comprising:

[0366] any one of the second CSI and information obtained by dequantizing the second CSI, and any one of the fifth CSI and information obtained by dequantizing the fifth CSI; or

[0367] the fifth CSI; or

[0368] information obtained by dequantizing the fifth CSI; or

[0369] information obtained by predicting based on the fifth CSI; or

[0370] information obtained by predicting based on information obtained by dequantizing the fifth CSI; or

[0371] information related to the fifth CSI;

[0372] The fifth CSI is a CSI reported by the terminal at a fourth time point, and the fourth time point is a time point before the second time point.

[0373] Optionally, the information related to the fifth CSI includes:

[0374] information determined based on the second CSI;

[0375] information determined based on the fifth CSI.

[0376] Optionally, the input of the first model further includes a second input, and the output of the first model includes a first output and a second output.

[0377] If the first input includes the second CSI, the second input includes information determined based on the second CSI; or, if the first input includes information obtained by processing the second CSI, the second input includes information determined based on the second CSI.

[0378] The first output is used to indicate information determined based on the first CSI.

[0379] The second output is used to indicate information determined based on the third CSI and the second CSI.

[0380] Optionally, the input of the first model further includes a second input, and the output of the first model includes a first output and a second output.

[0381] The information related to the third CSI reported by the terminal at a first time point includes: the third CSI reported by the terminal at the first time point; or information obtained by processing the third CSI;

[0382] The second input includes information determined based on the second CSI.

[0383] The first output is used to indicate information determined based on the first CSI.

[0384] The second output is used to indicate information determined based on the third CSI and the second CSI.

[0385] Optionally, the information obtained by processing the third CSI includes:

[0386] Information obtained by dequantizing the third CSI.

[0387] Optionally, the second input includes information determined based on the second CSI, including:

[0388] Information obtained by dequantizing the second CSI; or

[0389] Information obtained by predicting based on the second CSI; or

[0390] Information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0391] Information obtained by predicting based on third target information, wherein the third target information includes: CSI obtained by predicting a target third output, the target third output being a third output obtained by taking third information as input of the first model and processing the third information through the first model, the third information including:

[0392] The second CSI and information determined based on sixth CSI; or

[0393] Information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0394] Information obtained by predicting the second CSI and information determined based on the sixth CSI; or

[0395] Information obtained by predicting based on information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0396] The sixth CSI and information determined based on the sixth CSI; or

[0397] Information obtained by processing the sixth CSI and information determined based on the sixth CSI; or

[0398] Information related to the sixth CSI and information determined based on the sixth CSI.

[0399] The sixth CSI is the CSI at the fifth moment reported by the terminal, and the fifth moment is a moment before the second moment.

[0400] Optionally, the first moment includes the time when the first reference signal is transmitted in the first period, and the second moment includes the time when the first reference signal is transmitted in the x-th period before the first period, where x is an integer greater than or equal to 1.

[0401] It should be noted here that the above-mentioned device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0402] like Figure 11 As shown, the information processing apparatus according to an embodiment of the present application is applied to a network device, including: a processor 1100 configured to read a program in a memory 1120 and execute the following process:

[0403] Within the first time window, if the CSI at the sixth moment reported by the terminal is not detected, predicting the CSI at a target moment within the first time window based on the second model, where the target moment includes the sixth moment and / or a moment after the sixth moment;

[0404] The input of the second model includes:

[0405] The seventh CSI reported by the terminal for a time before the target time; or

[0406] Information determined based on seventh CSI reported by the terminal for a time before the target time.

[0407] The transceiver 1110 is configured to receive and send data under the control of the processor 1100 .

[0408] Among them, Figure 11 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 1100 and memory represented by memory 1120. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are all well known in the art and, therefore, will not be described further herein. The bus interface provides an interface. The transceiver 1110 may be a plurality of components, i.e., a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The processor 1100 is responsible for managing the bus architecture and general processing, and the memory 1120 may store data used by the processor 1100 when performing operations.

[0409] The processor 1100 can be a central processor (CPU), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or a complex programmable logic device (CPLD). The processor can also be a multi-core architecture.

[0410] The processor 1100 is responsible for managing the bus architecture and general processing, and the memory 1120 can store data used by the processor 1100 when performing operations.

[0411] Optionally, the information determined based on the seventh CSI reported by the terminal at the time point before the target time point includes:

[0412] The seventh CSI; or

[0413] Information obtained by dequantizing the seventh CSI; or

[0414] Information obtained by predicting based on the seventh CSI; or

[0415] Information obtained by predicting based on information obtained by dequantizing the seventh CSI.

[0416] It should be noted that the above device provided by the embodiments of the present application can realize all the method steps realized by the above method embodiments, and can achieve the same technical effects. Therefore, the same parts and beneficial effects of the method embodiments will not be described in detail.

[0417] As shown in Figure 12 The information processing device of the embodiments of the present application is applied to a network device, and includes:

[0418] The first processing unit 1201 is configured to determine a first CSI at a first time point based on a first model, wherein the input of the first model includes a first input;

[0419] The first input includes one or more of the following:

[0420] Information related to a second CSI reported by a terminal at a second time point, the second time point being a time point before the first time point;

[0421] Information related to a third CSI reported by the terminal at the first time point.

[0422] Optionally, the information related to the second CSI reported by the terminal at the second time comprises:

[0423] the second CSI; or

[0424] information obtained by processing the second CSI.

[0425] Optionally, the information obtained by processing the second CSI comprises:

[0426] information obtained by dequantizing the second CSI; or

[0427] information obtained by predicting based on the second CSI; or

[0428] information obtained by predicting based on the information obtained by dequantizing the second CSI; or

[0429] information obtained by predicting based on first target information, wherein the first target information comprises: CSI obtained by predicting a target first output, the target first output being a first output obtained by processing first information as input of the first model through the first model, and the first information comprising:

[0430] information determined based on the second CSI and information determined based on fourth CSI; or

[0431] information determined based on the fourth CSI, wherein the fourth CSI is a CSI reported by the terminal at a third time, and the third time is a time before the second time.

[0432] Optionally, the information determined based on the fourth CSI comprises:

[0433] the fourth CSI; or

[0434] information obtained by dequantizing the fourth CSI; or

[0435] information obtained by predicting based on the fourth CSI; or

[0436] information obtained by predicting based on the information obtained by dequantizing the fourth CSI.

[0437] Optionally, the second CSI comprises:

[0438] second CSI of one second time reported by the terminal, or second CSI of multiple second times reported by the terminal.

[0439] Optionally, the second CSI of one second time and the first time are located in the same time window; or

[0440] The plurality of second time instants and the first time instant are located in a same time window.

[0441] Optionally, the information related to the third CSI of the first time instant reported by the terminal comprises:

[0442] information determined based on the third CSI of the first time instant reported by the terminal;

[0443] information determined based on the second CSI.

[0444] Optionally, the information determined based on the third CSI of the first time instant reported by the terminal comprises:

[0445] the third CSI of the first time instant reported by the terminal; or

[0446] information obtained after dequantization of the third CSI of the first time instant reported by the terminal.

[0447] Optionally, the information determined based on the second CSI comprises:

[0448] information obtained after dequantization of the second CSI; or

[0449] information obtained based on prediction of the second CSI; or

[0450] information obtained based on prediction of information obtained after dequantization of the second CSI; or

[0451] information obtained based on prediction of second target information, wherein the second target information comprises: CSI obtained based on prediction of a target second output, the target second output being a second output obtained after processing of second information by the first model, the second information comprising:

[0452] any one of the second CSI and information obtained after dequantization of the second CSI, and any one of the fifth CSI and information obtained after dequantization of the fifth CSI; or

[0453] the fifth CSI; or

[0454] information obtained after dequantization of the fifth CSI; or

[0455] information obtained based on prediction of the fifth CSI; or

[0456] information obtained based on prediction of information obtained after dequantization of the fifth CSI; or

[0457] information related to the fifth CSI;

[0458] The fifth CSI is a fourth-time CSI reported by the terminal, and the fourth time is a time before the second time.

[0459] Optionally, the information related to the fifth CSI includes:

[0460] information determined based on the second CSI;

[0461] information determined based on the fifth CSI.

[0462] Optionally, the input of the first model further includes a second input, and the output of the first model includes a first output and a second output.

[0463] If the first input includes the second CSI, the second input includes information determined based on the second CSI; or, if the first input includes information obtained by processing the second CSI, the second input includes information determined based on the second CSI.

[0464] The first output is used to indicate information determined based on the first CSI.

[0465] The second output is used to indicate information determined based on the third CSI and the second CSI.

[0466] Optionally, the input of the first model further includes a second input, and the output of the first model includes a first output and a second output.

[0467] The information related to the third CSI reported by the terminal at the first time includes: the third CSI reported by the terminal at the first time; or information obtained by processing the third CSI;

[0468] The second input includes information determined based on the second CSI.

[0469] The first output is used to indicate information determined based on the first CSI.

[0470] The second output is used to indicate information determined based on the third CSI and the second CSI.

[0471] Optionally, the information obtained by processing the third CSI includes:

[0472] Information obtained by dequantizing the third CSI.

[0473] Optionally, the second input includes information determined based on the second CSI, including:

[0474] information obtained by dequantizing the second CSI; or

[0475] information obtained by predicting based on the second CSI; or

[0476] information obtained by predicting based on information obtained by dequantizing the second CSI; or

[0477] information obtained by predicting based on third target information, wherein the third target information comprises: CSI obtained by predicting a target third output, the target third output being a third output obtained by processing third information as input of the first model through the first model, the third information comprising:

[0478] the second CSI and information determined based on sixth CSI; or

[0479] information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0480] information obtained by predicting the second CSI and information determined based on the sixth CSI; or

[0481] information obtained by predicting based on information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or

[0482] the sixth CSI and information determined based on the sixth CSI; or

[0483] information obtained by processing the sixth CSI and information determined based on the sixth CSI; or

[0484] information related to the sixth CSI and information determined based on the sixth CSI.

[0485] The sixth CSI is CSI reported by the terminal at a fifth moment, and the fifth moment is a moment before the second moment.

[0486] Optionally, the first moment comprises a time at which a first reference signal is transmitted in a first period, and the second moment comprises a time at which a first reference signal is transmitted in an xth period before the first period, wherein x is an integer greater than or equal to 1.

[0487] It should be noted that the above apparatus provided by the embodiments of the present application can realize all the method steps achieved by the above method embodiments, and achieve the same technical effects. Therefore, the same parts and beneficial effects of the method embodiments will not be described in detail.

[0488] AsFigure 13 The information processing device provided by the embodiment of the present application is applied to a network device and comprises:

[0489] The first processing unit 1301 is configured to, if no sixth-time CSI reported by a terminal is detected within a first time window, predict CSI of a target time within the first time window based on a second model, wherein the target time comprises the sixth time and / or a time after the sixth time.

[0490] The input of the second model comprises:

[0491] a seventh CSI reported by the terminal for a time before the target time; or

[0492] information determined based on the seventh CSI reported by the terminal for the time before the target time.

[0493] Optionally, the information determined based on the seventh CSI reported by the terminal for the time before the target time comprises:

[0494] the seventh CSI; or

[0495] information obtained by dequantizing the seventh CSI; or

[0496] information obtained by predicting based on the seventh CSI; or

[0497] information obtained by predicting based on the information obtained by dequantizing the seventh CSI.

[0498] It should be noted that the above device provided by the embodiment of the present application can realize all the method steps realized by the above method embodiments and achieve the same technical effects, and thus the same parts and beneficial effects of the method embodiments will not be described in detail.

[0499] It should be noted that the division of units in the embodiment of the present application is illustrative, and is only a logical function division. In actual implementation, another division mode can be used. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0500] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a processor-readable storage medium. Based on such an understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the present application or the whole or part of the technical solutions can be embodied in the form of a computer software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0501] The embodiments of the present application also provide a communication device, including a memory, a processor, and a program stored in the memory and executable on the processor, and the processor implements the steps in the information processing method when executing the program.

[0502] The embodiments of the present application also provide a computer program product, including computer instructions executable by a processor to implement the various processes of the above-mentioned information processing method embodiments and achieve the same technical effects. To avoid repetition, details are not repeated here.

[0503] The embodiments of the present application also provide a processor-readable storage medium, which stores a program executable by a processor to implement the various processes of the above-mentioned information processing method embodiments and achieve the same technical effects. To avoid repetition, details are not repeated here. The readable storage medium can be any available medium or data storage device accessible by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state disk (SSD), etc.).

[0504] It should be noted that, in the present document, the terms "comprises / comprising" or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0505] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and the necessary general hardware platform, of course, they can also be realized by hardware, but in many cases the former is a better embodiment. According to such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a number of instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present application.

[0506] The embodiments of the present application are described above in combination with the accompanying drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative and not restrictive. Those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the scope protected by the claims.

Claims

1. An information processing method characterized by comprising: The method is applied to a network device, and comprises: determining first channel state information (CSI) at a first time based on a first model, wherein an input of the first model comprises a first input; wherein the first input comprises one or more of: information related to second CSI reported by a terminal at a second time, the second time being a time prior to the first time; information related to third CSI reported by the terminal at the first time.

2. The method of claim 1, wherein, The information related to the second CSI reported by the terminal at the second time comprises: the second CSI; or information obtained by processing the second CSI.

3. The method of claim 2, wherein, The information obtained by processing the second CSI comprises: information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI; or information obtained by predicting based on first target information, wherein the first target information comprises: CSI obtained by predicting a target first output, the target first output being a first output obtained by processing first information as an input of the first model by using the first model, the first information comprising: information determined based on the second CSI and information determined based on fourth CSI; or information determined based on the fourth CSI, wherein the fourth CSI is CSI reported by the terminal at a third time, the third time being a time prior to the second time.

4. The method of claim 3, wherein, The information determined based on the fourth CSI comprises: the fourth CSI; or information obtained by dequantizing the fourth CSI; or information obtained by predicting based on the fourth CSI; or information obtained by predicting based on information obtained by dequantizing the fourth CSI.

5. The method according to any one of claims 1 to 4, characterized in that, The second CSI comprises: second CSI reported by the terminal at one second time, or second CSI reported by the terminal at multiple second times.

6. The method of claim 5, wherein: the second CSI at the one second time and the first time are located in a same time window; or the multiple second times and the first time are located in a same time window.

7. The method of claim 1, wherein, The information related to the third CSI reported by the terminal at the first time comprises: information determined based on the third CSI reported by the terminal at the first time; information determined based on the second CSI.

8. The method of claim 7, wherein, The information determined based on the third CSI reported by the terminal at the first time comprises: the third CSI reported by the terminal at the first time; or information obtained by dequantizing the third CSI reported by the terminal at the first time.

9. The method according to claim 3 or 7, characterized in that, The information determined based on the second CSI comprises: information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI; or information obtained based on the second target information, wherein the second target information comprises: CSI obtained by predicting a target second output, the target second output being a second output obtained by processing second information as input of the first model by using the first model, the second information comprising: any one of the second CSI and information obtained by dequantizing the second CSI, and any one of the fifth CSI and information obtained by dequantizing the fifth CSI; or the fifth CSI; or information obtained by dequantizing the fifth CSI; or information obtained by predicting based on the fifth CSI; or information obtained by predicting based on information obtained by dequantizing the fifth CSI; or information related to the fifth CSI; wherein the fifth CSI is CSI reported by the terminal at a fourth time point, and the fourth time point is a time point before the second time point.

10. The method of claim 9, wherein, The information related to the fifth CSI comprises: information determined based on the second CSI; information determined based on the fifth CSI.

11. The method of claim 2, wherein, The input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output. If the first input comprises the second CSI, the second input comprises information determined based on the second CSI; or if the first input comprises information obtained by processing the second CSI, the second input comprises information determined based on the second CSI; the first output is used to indicate information determined based on the first CSI; the second output is used to indicate information determined based on the third CSI and the second CSI.

12. The method of claim 1, wherein, The input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output. The information related to the third CSI reported by the terminal at the first time point comprises: the third CSI reported by the terminal at the first time point; or information obtained by processing the third CSI; the second input comprises information determined based on the second CSI; the first output is used to indicate information determined based on the first CSI; the second output is used to indicate information determined based on the third CSI and the second CSI.

13. The method of claim 12, wherein, The information obtained by processing the third CSI comprises: information obtained by dequantizing the third CSI.

14. The method according to any one of claims 11-13, characterized in that, The second input comprises information determined based on the second CSI, comprising: information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI; or information obtained by predicting based on third target information, wherein the third target information comprises: CSI obtained by predicting a target third output, the target third output being a third output obtained by processing third information as input of the first model by using the first model, the third information comprising: The second CSI and information determined based on the sixth CSI; or Information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or Information obtained by predicting the second CSI and information determined based on the sixth CSI; or Information obtained by dequantizing the second CSI and information obtained by predicting the second CSI and information determined based on the sixth CSI; or The sixth CSI and information determined based on the sixth CSI; or Information obtained by processing the sixth CSI and information determined based on the sixth CSI; or Information related to the sixth CSI and information determined based on the sixth CSI. The sixth CSI is CSI reported by the terminal at a fifth time point, and the fifth time point is a time point before the second time point.

15. The method of claim 1, wherein, The first time point includes a time at which a first reference signal is transmitted in a first period, and the second time point includes a time at which a first reference signal is transmitted in an xth period before the first period, where x is an integer greater than or equal to 1.

16. An information processing method characterized by comprising: Applied to a network device, comprising: In a first time window, if a sixth time point CSI reported by a terminal is not detected, a CSI at a target time point in the first time window is predicted based on a second model, the target time point includes the sixth time point and / or a time point after the sixth time point; Wherein the input of the second model includes: The seventh CSI reported by the terminal at a time point before the target time point; or Information determined based on the seventh CSI reported by the terminal at a time point before the target time point.

17. The method of claim 16, wherein, The information determined based on the seventh CSI reported by the terminal at a time point before the target time point, includes: The seventh CSI; or Information obtained by dequantizing the seventh CSI; or Information obtained by predicting based on the seventh CSI; or Information obtained by predicting based on information obtained by dequantizing the seventh CSI.

18. An information processing apparatus comprising: Applied to a network device, comprising: a memory, a transceiver, a processor: The memory is used to store computer programs; the transceiver is used to transceive data under the control of the processor; the processor is used to read the computer programs in the memory and perform the following operations: Determine the first CSI at the first time point based on the first model, wherein the input of the first model includes the first input; Wherein the first input includes one or more of the following: Information related to the second CSI reported by the terminal at the second time point, the second time point is a time point before the first time point; Information related to the third CSI reported by the terminal at the first time point.

19. The apparatus of claim 18, wherein, The information related to the second CSI reported by the terminal at the second time point, includes: The second CSI; or Information obtained by processing the second CSI.

20. The apparatus of claim 19, wherein, The information obtained by processing the second CSI, includes: Information obtained by dequantizing the second CSI; or Information obtained by predicting based on the second CSI; or Information obtained by predicting based on information obtained by dequantizing the second CSI; or information determined based on the second CSI and information determined based on fourth CSI; or information determined based on the fourth CSI, wherein the fourth CSI is CSI of a third time point reported by the terminal, and the third time point is a time point before the second time point. The information determined based on the fourth CSI comprises:

21. The apparatus of claim 20, wherein, The fourth CSI; or information obtained by dequantizing the fourth CSI; or information obtained by predicting based on the fourth CSI; or information obtained by predicting based on information obtained by dequantizing the fourth CSI. The second CSI comprises:

22. The apparatus of any one of claims 18-21, wherein, second CSI of a second time point reported by the terminal, or second CSI of multiple second time points reported by the terminal.

23. The apparatus of claim 22, wherein the second CSI of the second time point and the first time point are located in a same time window; or the multiple second time points and the first time point are located in a same time window. The information related to the third CSI of the first time point reported by the terminal comprises:

24. The apparatus of claim 18, wherein, information determined based on the third CSI of the first time point reported by the terminal; information determined based on the second CSI. The information determined based on the third CSI of the first time point reported by the terminal comprises:

25. The apparatus of claim 24, wherein, The third CSI of the first time point reported by the terminal; or information obtained by dequantizing the third CSI of the first time point reported by the terminal. The information determined based on the second CSI comprises:

26. The apparatus of claim 20 or 24, wherein, information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI; or information obtained by predicting based on second target information, wherein the second target information comprises: CSI obtained by predicting target second output, wherein the target second output is second output obtained by processing second information as input of the first model through the first model, and the second information comprises: any one of the second CSI and information obtained by dequantizing the second CSI, and any one of fifth CSI and information obtained by dequantizing the fifth CSI; or The fifth CSI; or information obtained by dequantizing the fifth CSI; or information obtained by predicting based on the fifth CSI; or information obtained by predicting based on information obtained by dequantizing the fifth CSI; or information related to the fifth CSI; wherein the fifth CSI is CSI of a fourth time point reported by the terminal, and the fourth time point is a time point before the second time point. ​ 27. The apparatus of claim 26, wherein, The information related to the fifth CSI comprises: information determined based on the second CSI; information determined based on the fifth CSI.

28. The apparatus of claim 19, wherein, The input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output; if the first input comprises the second CSI, the second input comprises information determined based on the second CSI; or, if the first input comprises information obtained by processing the second CSI, the second input comprises information determined based on the second CSI; the first output is used for indicating information determined based on the first CSI; the second output is used for indicating information determined based on the third CSI and the second CSI.

29. The apparatus of claim 18, wherein, The input of the first model further comprises a second input, and the output of the first model comprises a first output and a second output; The information related to the third CSI reported by the terminal at the first time comprises: the third CSI reported by the terminal at the first time; or information obtained by processing the third CSI; the second input comprises information determined based on the second CSI; the first output is used for indicating information determined based on the first CSI; the second output is used for indicating information determined based on the third CSI and the second CSI.

30. The apparatus of claim 29, wherein, The information obtained by processing the third CSI comprises: information obtained by dequantizing the third CSI.

31. The device of any one of claims 28-30, wherein, The second input comprises information determined based on the second CSI, comprising: information obtained by dequantizing the second CSI; or information obtained by predicting based on the second CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI; or information obtained by predicting based on third target information, wherein the third target information comprises: CSI obtained by predicting a target third output, the target third output being a third output obtained by taking third information as the input of the first model and processing the first model, the third information comprising: the second CSI and information determined based on sixth CSI; or information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or information obtained by predicting the second CSI and information determined based on the sixth CSI; or information obtained by predicting based on information obtained by dequantizing the second CSI and information determined based on the sixth CSI; or the sixth CSI and information determined based on the sixth CSI; or information obtained by processing the sixth CSI and information determined based on the sixth CSI; or information related to the sixth CSI and information determined based on the sixth CSI; wherein the sixth CSI is a CSI reported by the terminal at a fifth time, and the fifth time is a time before the second time.

32. The apparatus of claim 18, wherein, The first time comprises a time at which a first reference signal is transmitted in a first period, and the second time comprises a time at which a first reference signal is transmitted in an xth period before the first period, where x is an integer greater than or equal to 1.

33. An information processing apparatus for enabling a user to efficiently search for a desired information, comprising: The network device comprises a memory, a transceiver, and a processor. The memory is configured to store a computer program; the transceiver is configured to transceive data under control of the processor; and the processor is configured to read the computer program in the memory and perform the following operations: In a first time window, if a sixth time CSI reported by a terminal is not detected, a CSI at a target time in the first time window is predicted based on a second model, the target time comprising the sixth time and / or a time after the sixth time; wherein the input of the second model comprises: a seventh CSI reported by the terminal at a time before the target time; or information determined based on the seventh CSI reported by the terminal at the time before the target time.

34. The apparatus of claim 33, wherein, The information determined based on the seventh CSI reported by the terminal at the time before the target time comprises: the seventh CSI; or information obtained by dequantizing the seventh CSI; or information obtained by prediction based on the seventh CSI; or information obtained by prediction based on the information obtained by dequantizing the seventh CSI.

35. An information processing apparatus for enabling a user to efficiently search for and retrieve information, the apparatus comprising: The network device comprises: a first processing unit configured to determine a first CSI at a first time based on a first model, wherein the input of the first model comprises a first input; wherein the first input comprises one or more of the following: information related to a second CSI reported by a terminal at a second time, the second time being a time before the first time; or information related to a third CSI reported by the terminal at the first time.

36. An information processing apparatus for enabling a user to efficiently search for a desired information, comprising: The network device comprises: a first processing unit configured to, in a first time window, if a sixth time CSI reported by a terminal is not detected, predict a CSI at a target time in the first time window based on a second model, the target time comprising the sixth time and / or a time after the sixth time; wherein the input of the second model comprises: a seventh CSI reported by the terminal at a time before the target time; or information determined based on the seventh CSI reported by the terminal at the time before the target time.

37. A processor-readable storage medium, comprising: The processor readable storage medium stores a computer program, and the computer program is configured to cause the processor to perform the method according to any one of claims 1 to 17.

Images