Information reporting method, information transmission method, device, equipment and medium

Abstract

The invention provides an information reporting method, an information transmission method, devices, equipment and a medium. The method comprises the steps that performance monitoring related information for a first object is acquired, and the first object is an artificial intelligence model or an artificial intelligence function; performance monitoring related information is reported to network side equipment, and the performance monitoring related information comprises one or more of the following information: characteristic values of performance monitoring indexes at N moments; performance monitoring related information of each moment of the N moments; the characteristic values of the performance monitoring indexes at N moments correspond to moment related information, and N is a positive integer greater than 1; performing model reasoning on characteristic values of the corresponding performance monitoring indexes for multiple times; and time related information corresponding to the characteristic values of the corresponding performance monitoring indexes is subjected to model reasoning for many times. According to the application, the performance monitoring related information is reported to the network side equipment, so that the performance monitoring of the artificial intelligence model or the artificial intelligence function of the terminal equipment side by the network side can be realized.

Description

TECHNICAL FIELD

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

[0002] In new radio (NR) beam management, the existing technology determines the optimal transmission beam by measuring all beam information transmitted by a transmission end Tx, which requires measuring the reference signal (RS) of all beams, resulting in a large resource overhead. In artificial intelligence (AI) based beam management, only part of the beam information needs to be measured, and all beam information can be obtained through training of an AI model, greatly reducing the RS resource overhead.

[0003] As a data-driven algorithm, the generalization ability of an AI / machine learning (ML) model is limited, and different models may be deployed for different scenarios, i.e., each model is applicable to a limited scenario or configuration, and therefore the performance of the model needs to be monitored to ensure that the AI model used is matched with the current scenario or configuration to ensure system performance.

[0004] For a terminal device side model, performance monitoring may be at the base station side or at the terminal device side. When performance monitoring is at the base station side, the terminal device side needs to report information to enable the network side to perform performance monitoring, but there is no specific method for reporting what information. SUMMARY

[0005] The present application aims to provide an information reporting method, an information transmission method, an apparatus, a device and a medium to solve the problem of what information a terminal device reports to enable the network side to perform performance monitoring when performance monitoring is at the network side.

[0006] To achieve the above-mentioned purpose, in a first aspect, an embodiment of the present application provides an information reporting method applied to a terminal device, comprising:

[0007] obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0008] reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information:

[0009] characteristic values of performance monitoring indicators at N time instants;

[0010] performance monitoring related information at each of the N time instants;

[0011] the time-related information corresponding to the characteristic value of the performance monitoring index of the N time points, N being a positive integer greater than 1;

[0012] the characteristic value of the performance monitoring index corresponding to the multiple model inferences;

[0013] the time-related information corresponding to the characteristic value of the performance monitoring index corresponding to the multiple model inferences.

[0014] In some embodiments, the characteristic value includes one or more of a mean value, a maximum value, and a minimum value.

[0015] In some embodiments, the performance monitoring-related information for the first object is obtained by:

[0016] receiving first reference signals corresponding to M time points configured by the network-side device, one or more first reference signals corresponding to each time point;

[0017] performing measurement on the first reference signals corresponding to the M time points, M being a positive integer;

[0018] obtaining a prediction result for each time point of N time points according to the measurement results of the first reference signals corresponding to the M time points;

[0019] receiving second reference signals corresponding to the N time points configured by the network-side device, one or more second reference signals corresponding to each time point;

[0020] performing measurement on the second reference signals corresponding to the N time points;

[0021] obtaining the performance monitoring-related information for the first object according to the prediction result for each time point of the N time points and the measurement result for each time point of the N time points.

[0022] In some embodiments, the performance monitoring-related information is reported to the network-side device by:

[0023] reporting, to the network-side device, the number of time points correctly predicted or the correct prediction rate in the N time points, the characteristic value of the performance monitoring index of the N time points including the number of time points correctly predicted or the correct prediction rate in the N time points; and / or,

[0024] reporting, to the network-side device, first indication information, the first indication information being used to indicate whether the model inference for each time point of the N time points is correct, the performance monitoring-related information for each time point of the N time points including the first indication information.

[0025] In some embodiments, the prediction result of each of the N time points comprises K indexes predicted for each of the N time points; and the measurement result of each of the N time points comprises indexes of the maximum or strongest K reference signals measured at each of the N time points, K being a positive integer.

[0026] The performance monitoring related information of the first object is obtained according to the prediction result of each of the N time points and the measurement result of each of the N time points, comprising:

[0027] According to the K indexes predicted for each of the N time points and the indexes of the maximum or strongest K reference signals measured at each of the N time points, the number of correctly predicted time points or the correct prediction rate in the N time points is obtained; and / or,

[0028] According to the K indexes predicted for each of the N time points and the indexes of the maximum or strongest K reference signals measured at each of the N time points, first indication information is obtained, the first indication information being used to indicate whether the model inference of each of the N time points is correct.

[0029] In some embodiments, the first indication information is an N-bit bitmap.

[0030] In some embodiments, the measurement of the second reference signals corresponding to the N time points comprises:

[0031] The measurement of the second reference signals corresponding to the N time points obtains first RSRP measurement values of each of the N time points.

[0032] The first RSRP measurement values of each of the N time points are sorted to obtain indexes of the maximum or strongest K reference signals of each of the N time points.

[0033] In some embodiments, the reporting of the performance monitoring related information to the network side device comprises:

[0034] The reporting of the performance monitoring related information to the network side device comprises reporting one or more of the following information to the network side device:

[0035] A first average value, the first average value being an average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time points;

[0036] A maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time points;

[0037] a time corresponding to a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0038] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0039] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0040] In some embodiments, the prediction result of each of the N time instants comprises a RSRP prediction value of the target reference signal corresponding to each of the N time instants; and the measurement result of each of the N time instants comprises a RSRP measurement value of the target reference signal corresponding to each of the N time instants.

[0041] The performance monitoring related information of the first object is obtained according to the prediction result of each of the N time instants and the measurement result of each of the N time instants, comprising:

[0042] The difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants is calculated according to the RSRP prediction value of the target reference signal corresponding to each of the N time instants and the RSRP measurement value of the target reference signal corresponding to each of the N time instants.

[0043] According to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants, one or more of the following information is obtained:

[0044] a first mean value, the first mean value being a mean value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0045] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0046] a time corresponding to a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0047] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0048] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0049] In some embodiments, the performance monitoring related information comprises:

[0050] the measurement result of the first reference signal corresponding to the M time points is input into an artificial intelligence model to obtain the RSRP prediction value of the second reference signal corresponding to each of the N time points;

[0051] the RSRP prediction value of the target reference signal corresponding to each of the N time points is obtained according to the RSRP prediction value of the second reference signal corresponding to each of the N time points.

[0052] In some embodiments, the performance monitoring related information comprises:

[0053] the performance monitoring related information comprises one or more of the following information:

[0054] a maximum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or measurements within the first time period, P being a positive integer greater than 1;

[0055] a time point related information or an index value corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or measurements within the first time period;

[0056] a minimum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or measurements within the first time period;

[0057] a time point related information or an index value corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or measurements within the first time period;

[0058] In some embodiments, the performance monitoring related information comprises:

[0059] In some embodiments, the performance monitoring related information comprises:

[0060] the first reference signal configured by the network side device is received for one measurement;

[0061] measure the first reference signal;

[0062] obtain a RSRP prediction value for a target reference signal according to the measurement result of the first reference signal;

[0063] calculate a difference between the RSRP prediction value for the target reference signal and a RSRP measurement value for the target reference signal according to the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal;

[0064] after performing P times of measurements or measurements within a first time period, obtain one or more of the following information according to the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to each measurement:

[0065] a maximum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0066] time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0067] a minimum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0068] time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period.

[0069] In some embodiments, the target reference signal is determined by the first reference signal or the prediction result.

[0070] In some embodiments, the method further comprises:

[0071] receiving configuration information sent by a network side device, the configuration information being used to indicate content of the performance monitoring related information reported by the terminal device for a first object.

[0072] In a second aspect, the embodiments of the present application further provide an information transmission method applied to a network side device, comprising:

[0073] receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following information:

[0074] characteristic values of performance monitoring indexes at N time points;

[0075] performance monitoring related information of each of the N time points;

[0076] The characteristic of the performance monitoring index of the N time points is related to the time point information corresponding to the characteristic, and N is a positive integer greater than 1.

[0077] The characteristic value of the performance monitoring index corresponding to the multiple model inferences;

[0078] The time point information corresponding to the characteristic value of the performance monitoring index corresponding to the multiple model inferences.

[0079] In some embodiments, the characteristic value includes one or more of a mean value, a maximum value, and a minimum value.

[0080] In some embodiments, the method further includes:

[0081] The terminal device is configured with first reference signals corresponding to M time points, one or more first reference signals corresponding to each time point, and the first reference signals corresponding to the M time points are used to obtain predicted results of each of the N time points, and M is a positive integer.

[0082] The terminal device is configured with second reference signals corresponding to N time points, one or more second reference signals corresponding to each time point, and the second reference signals corresponding to the N time points are used to obtain measurement results of each of the N time points.

[0083] In some embodiments, the method further includes:

[0084] For each measurement in P measurements or within a first time period, the terminal device is configured with a first reference signal, and the first reference signal is used to obtain an RSRP prediction value for a target reference signal, and the target reference signal is determined by the first reference signal or the predicted result, and P is a positive integer greater than 1.

[0085] In some embodiments, the method further includes:

[0086] The configuration information is sent to the terminal device, and the configuration information is used to indicate the content of the performance monitoring related information of the first object reported by the terminal device.

[0087] In a third aspect, the embodiments of the present application also provide a terminal device, which includes a memory, a transceiver, and a processor, wherein the memory is used to store computer programs, the transceiver is used to transceive data under the control of the processor, and the processor is used to read program instructions in the memory and perform the following operations:

[0088] The performance monitoring related information of the first object is obtained, and the first object is an artificial intelligence model or an artificial intelligence function.

[0089] reporting the performance monitoring related information to the network side device, wherein the performance monitoring related information comprises one or more of the following information:

[0090] characteristic values of the performance monitoring indicators at the N time instants;

[0091] performance monitoring related information at each of the N time instants;

[0092] time instant related information corresponding to the characteristic values of the performance monitoring indicators at the N time instants, N being a positive integer greater than 1;

[0093] characteristic values of the performance monitoring indicators corresponding to the multiple model inferences;

[0094] time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences.

[0095] In some embodiments, the characteristic values comprise one or more of a mean value, a maximum value, and a minimum value.

[0096] In some embodiments, the operations further comprise:

[0097] receiving first reference signals corresponding to the M time instants configured by the network side device, one or more first reference signals corresponding to each time instant;

[0098] performing measurement on the first reference signals corresponding to the M time instants; wherein M is a positive integer;

[0099] obtaining predicted results at each of the N time instants according to the measurement results of the first reference signals corresponding to the M time instants;

[0100] receiving second reference signals corresponding to the N time instants configured by the network side device, one or more second reference signals corresponding to each time instant;

[0101] performing measurement on the second reference signals corresponding to the N time instants;

[0102] obtaining performance monitoring related information for the first object according to the predicted results at each of the N time instants and the measurement results at each of the N time instants.

[0103] In some embodiments, the operations further comprise:

[0104] reporting, to the network side device, a number of correctly predicted time instants or a correct prediction rate among the N time instants, wherein the characteristic values of the performance monitoring indicators at the N time instants comprise the number of correctly predicted time instants or the correct prediction rate among the N time instants; and / or,

[0105] reporting first indication information to the network side device, the first indication information being used to indicate whether the model inference of each of the N time instants is correct, the performance monitoring related information of each of the N time instants including the first indication information.

[0106] In some embodiments, the prediction result of each of the N time instants includes K indexes of the N time instants predicted, K being a positive integer; the measurement result of each of the N time instants includes indexes of the maximum or strongest K reference signals of each of the N time instants measured; and the operations further include:

[0107] obtaining, according to the K indexes of each of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each of the N time instants measured, the number of time instants correctly predicted or the prediction accuracy rate in the N time instants; and / or,

[0108] obtaining, according to the K indexes of each of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each of the N time instants measured, the first indication information, the first indication information being used to indicate whether the model inference of each of the N time instants is correct.

[0109] In some embodiments, the first indication information is an N-bit bitmap.

[0110] In some embodiments, the operations further include:

[0111] measuring the second reference signals corresponding to the N time instants to obtain first RSRP measurement values of each of the N time instants;

[0112] sorting the first RSRP measurement values of each of the N time instants to obtain indexes of the maximum or strongest K reference signals of each of the N time instants.

[0113] In some embodiments, the operations further include:

[0114] reporting one or more of the following information to the network side device:

[0115] a first average value, the first average value being an average value of differences between RSRP prediction values and RSRP measurement values of target reference signals corresponding to the N time instants;

[0116] a maximum value of differences between RSRP prediction values and RSRP measurement values of target reference signals corresponding to the N time instants;

[0117] information of a time corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0118] the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0119] the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

[0120] In some embodiments, the prediction result of each of the N time instants includes the RSRP prediction value of the target reference signal corresponding to each of the N time instants; the measurement result of each of the N time instants includes the RSRP measurement value of the target reference signal corresponding to each of the N time instants; and the operation further includes:

[0121] According to the RSRP prediction value of the target reference signal corresponding to each of the N time instants and the RSRP measurement value of the target reference signal corresponding to each of the N time instants, the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants is calculated.

[0122] According to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants, one or more of the following information is obtained:

[0123] a first mean value, the first mean value being a mean value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0124] the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0125] information of a time corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0126] the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0127] information of a time corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

[0128] In some embodiments, the operation further includes:

[0129] inputting a measurement result of the first reference signal corresponding to the M time points into an artificial intelligence model to obtain an RSRP prediction value of the second reference signal corresponding to each of the N time points;

[0130] obtaining, according to the RSRP prediction value of the second reference signal corresponding to each of the N time points, an RSRP prediction value of a target reference signal corresponding to each of the N time points.

[0131] In some embodiments, the operations further include:

[0132] reporting one or more of the following information to the network side device:

[0133] a maximum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period, P being a positive integer greater than 1;

[0134] time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0135] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0136] time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0137] The feature value of the performance monitoring index corresponding to the multiple model inferences includes: the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period, and / or the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period.

[0138] In some embodiments, the operations further include:

[0139] receiving, for one measurement, a first reference signal configured by the network side device;

[0140] measuring the first reference signal;

[0141] obtaining, according to a measurement result of the first reference signal, an RSRP prediction value of a target reference signal;

[0142] According to the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal, a difference between the RSRP prediction value and the RSRP measurement value for the target reference signal is calculated;

[0143] After the P measurements or the measurements in the first time period are performed, according to the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to each measurement, one or more of the following information is obtained:

[0144] A maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P measurements or the measurements in the first time period;

[0145] Time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P measurements or the measurements in the first time period;

[0146] A minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P measurements or the measurements in the first time period;

[0147] Time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P measurements or the measurements in the first time period.

[0148] In some embodiments, the target reference signal is determined by the first reference signal or the prediction result.

[0149] In some embodiments, the operations further include:

[0150] Receiving configuration information sent by the network side device, the configuration information being used to indicate content of the performance monitoring related information for the first object reported by the terminal device.

[0151] In a fourth aspect, the embodiments of the present application further provide an information reporting apparatus, including:

[0152] An obtaining unit, configured to obtain performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0153] A reporting unit, configured to report the performance monitoring related information to a network side device, wherein the performance monitoring related information includes one or more of the following information:

[0154] Characteristic values of performance monitoring indicators at N time points;

[0155] Performance monitoring related information at each of the N time points;

[0156] the characteristic value of the performance monitoring index corresponding to the N time instants, N being a positive integer greater than 1;

[0157] the characteristic value of the performance monitoring index corresponding to the multiple model inferences;

[0158] the characteristic value of the performance monitoring index corresponding to the multiple model inferences.

[0159] In a fifth aspect, an embodiment of the present application further provides a network side device, comprising: a memory, a transceiver, and a processor; the memory is configured to store a computer program; the transceiver is configured to transceive data under the control of the processor; and the processor is configured to read program instructions in the memory and perform the following operations:

[0160] receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following information:

[0161] the characteristic value of the performance monitoring index corresponding to the N time instants;

[0162] the performance monitoring related information of each time instant of the N time instants;

[0163] the characteristic value of the performance monitoring index corresponding to the N time instants, N being a positive integer greater than 1;

[0164] the characteristic value of the performance monitoring index corresponding to the multiple model inferences;

[0165] the characteristic value of the performance monitoring index corresponding to the multiple model inferences.

[0166] In some embodiments, the characteristic value comprises one or more of a mean value, a maximum value, and a minimum value.

[0167] In some embodiments, the operations further comprise:

[0168] configuring the terminal device with first reference signals corresponding to M time instants, one or more first reference signals corresponding to each time instant, the first reference signals corresponding to the M time instants being used to obtain predicted results of each time instant of the N time instants, M being a positive integer;

[0169] configuring the terminal device with second reference signals corresponding to the N time instants, one or more second reference signals corresponding to each time instant, the second reference signals corresponding to the N time instants being used to obtain measurement results of each time instant of the N time instants.

[0170] In some embodiments, the operations further comprise:

[0171] For each measurement in P measurements or in a first time period, a first reference signal is configured for the terminal device, the first reference signal being used to obtain a RSRP prediction value for a target reference signal, the target reference signal being determined by the first reference signal or a prediction result, P being a positive integer greater than 1.

[0172] In some embodiments, the operations further include:

[0173] The configuration information is sent to the terminal device, the configuration information being used to indicate content of the performance monitoring related information reported by the terminal device for the first object.

[0174] In a sixth aspect, the embodiments of the present application further provide an information transmission device, comprising:

[0175] A first receiving unit is configured to receive performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following information:

[0176] Characteristic values of performance monitoring indexes at N time points;

[0177] Performance monitoring related information at each of the N time points;

[0178] Corresponding time point related information of characteristic values of performance monitoring indexes at the N time points, N being a positive integer greater than 1;

[0179] Characteristic values of performance monitoring indexes corresponding to multiple model inferences;

[0180] Corresponding time point related information of characteristic values of performance monitoring indexes corresponding to multiple model inferences.

[0181] In a seventh aspect, the embodiments of the present application further provide a processor readable storage medium, which stores a computer program, the computer program being used to make the processor execute steps of the information reporting method in the first aspect or steps of the information transmission method in the second aspect.

[0182] In an eighth aspect, the embodiments of the present application further provide a computer program product, which comprises computer instructions, the computer instructions being executed by a processor to implement steps in the information reporting method in the first aspect or steps in the information transmission method in the second aspect.

[0183] The above technical solutions of the present application have at least the following beneficial effects:

[0184] In the above-mentioned technical solution of the embodiment of the present application, by obtaining performance monitoring related information for the first object, the first object is an artificial intelligence model or an artificial intelligence function; the performance monitoring related information is reported to the network side device, wherein the performance monitoring related information includes one or more of the following information: characteristic values ​​of performance monitoring indicators at N moments; performance monitoring related information at each of the N moments; moment related information corresponding to the characteristic values ​​of performance monitoring indicators at N moments, N is a positive integer greater than 1; characteristic values ​​of performance monitoring indicators corresponding to multiple model inferences; moment related information corresponding to the characteristic values ​​of performance monitoring indicators corresponding to multiple model inferences. In this way, by reporting the above-mentioned performance monitoring related information to the network side device, the network side can realize performance monitoring of the artificial intelligence model or artificial intelligence function on the terminal device side. BRIEF DESCRIPTION OF THE DRAWINGS

[0185] Figure 1 This is a diagram of the basic reporting process for UE-side model performance monitoring;

[0186] Figure 2 A flowchart of an information reporting method according to an embodiment of the present application;

[0187] Figure 3 A schematic diagram of a bitmap of this application;

[0188] Figure 4 A flowchart of an information transmission method according to an embodiment of the present application;

[0189] Figure 5 This is a structural block diagram of a terminal device according to an embodiment of the present application;

[0190] Figure 6 This is a module diagram of the information reporting device according to an embodiment of the present application;

[0191] Figure 7 This is a structural block diagram of a network-side device according to an embodiment of the present application;

[0192] Figure 8 This is a module diagram of the information transmission device according to an embodiment of the present application. DETAILED DESCRIPTION

[0193] In the embodiments of this application, the term "and / or" describes the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0194] In the embodiments of the present application, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0195] With reference to the drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present application, and not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0196] In order to facilitate the understanding of the scheme of the present application, the related contents involved in the present application are introduced first.

[0197] The 3rd Generation Partnership Project (3GPP) has two sub-use cases for AI beam management, namely spatial beam prediction and temporal beam prediction. In spatial beam prediction, the input of the model is the reference signal received power (RSRP) of SetB beams, and the output is the optimal beam ID or RSRP of SetA beams. SetA is the set of all beams, and SetB is a subset of SetA. Alternatively, SetA is a set of narrow beams, and SetB is a set of wide beams. In temporal beam prediction, the input of the model is the RSRP of SetB beams at the past M time instants, and the output is the optimal beam ID or RSRP of SetA beams at the future N time instants.

[0198] The performance monitoring indicators can be beam prediction accuracy related key performance indicators (KPIs), such as Top-1 beam prediction accuracy and Top-K / 1 beam prediction accuracy. Top-1 beam prediction accuracy refers to the probability that the predicted Top-1 beam is the actual Top-1 beam, and Top-K / 1 beam prediction accuracy refers to the probability that the actual Top-1 beam is included in the predicted Top-K beams. The performance monitoring indicators can also be the difference between the measured and predicted values of the L1-RSRP of the predicted beam output by the model, such as the difference between the actual L1-RSRP and the predicted L1-RSRP of the predicted Top-K beams, or the difference between the actual L1-RSRP and the predicted L1-RSRP of Set B beams.

[0199] The basic performance monitoring process of the AI / ML model deployed on the UE side is as follows Figure 1performance monitoring, the performance monitoring result of the AI model or AI function in a period of time needs to be reported. If the existing UE reporting process is followed, the UE needs to report the result of a certain time or report the average value. In order to save the overhead, the UE can not need to report the measurement results of all beams at all times, but in order to guarantee the monitoring performance, the measurement and inference results of the Top-K beams at a certain time cannot be reported. On the other hand, in time-domain beam prediction, the AI model or AI function can predict the optimal beam at multiple time points at a time, and for one model monitoring, there is no method for the UE to report the performance monitoring indicators related to multiple predicted time points.

[0200] In summary, in the performance monitoring technology of the existing AI beam management BM-Case1 (spatial domain beam prediction) and BM-Case2 (time domain beam prediction) models, when the model is deployed on the UE side but needs to be monitored by the network side, there is no specific method for the specific content reported by the UE.

[0201] To solve the above technical problems, the embodiments of the present application provide an information reporting method, an information transmission method, a device, equipment and a medium, wherein the method and the device are based on the same application concept. Since the principles of the method and the device for solving the problem are similar, the implementation of the device and the method can be mutually referred to, and the repeated parts will not be described here.

[0202] As Figure 2 As shown in the flowchart of the information reporting method provided by the embodiments of the present application, the method is applied to a terminal device, that is, the method is executed by the terminal device. The method specifically includes the following steps.

[0203] Step 201, obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0204] Optionally, the artificial intelligence model is an AI beam management model, such as a spatial domain beam prediction model or a time domain beam prediction model.

[0205] Step 202, reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information includes one or more of the following information:

[0206] Characteristic values of performance monitoring indicators at N time points; here, the characteristic values of the performance monitoring indicators at the N time points can be the characteristic values of the performance monitoring indicators at the N time points for each monitoring.

[0207] Performance monitoring related information at each of the N time points;

[0208] The characteristic value of the performance monitoring index corresponding to the time-related information of the N time points, N being a positive integer greater than 1; here, the performance monitoring-related information of each time point of the N time points can be the performance monitoring-related information of each time point of the N time points of each monitoring.

[0209] The characteristic value of the performance monitoring index corresponding to the time-related information of the N time points, N being a positive integer greater than 1; here, the performance monitoring-related information of each time point of the N time points can be the performance monitoring-related information of each time point of the N time points of each monitoring.

[0210] The characteristic value of the performance monitoring index corresponding to the time-related information of the N time points, N being a positive integer greater than 1; here, the performance monitoring-related information of each time point of the N time points can be the performance monitoring-related information of each time point of the N time points of each monitoring.

[0211] It should be noted that the performance monitoring index can be a beam prediction accuracy related KPI, or the difference between the measured value and the predicted value of the L1-RSRP of the predicted beam output by the model.

[0212] Optionally, the characteristic value includes one or more of the mean value, the maximum value, and the minimum value.

[0213] In some embodiments, the performance monitoring-related information includes one or more of the following information:

[0214] The mean value of the performance monitoring index of the N time points;

[0215] The performance monitoring-related information of each time point of the N time points;

[0216] The maximum value of the performance monitoring index of the N time points;

[0217] The time-related information corresponding to the maximum value of the performance monitoring index of the N time points;

[0218] The minimum value of the performance monitoring index of the N time points;

[0219] The time-related information corresponding to the minimum value of the performance monitoring index of the N time points;

[0220] The maximum value of the performance monitoring index corresponding to the multiple model inferences;

[0221] The time-related information corresponding to the maximum value of the performance monitoring index corresponding to the multiple model inferences;

[0222] The minimum value of the performance monitoring index corresponding to the multiple model inferences;

[0223] The time-related information corresponding to the minimum value of the performance monitoring index corresponding to the multiple model inferences.

[0224] The information reporting method of the embodiments of the present application comprises: obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; and reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; and time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences. In this way, by reporting the above performance monitoring related information to the network side device, the performance monitoring of the artificial intelligence model or the artificial intelligence function on the terminal device side by the network side can be realized.

[0225] In some embodiments, the step 201 of obtaining performance monitoring related information for a first object comprises:

[0226] The step 2011a comprises: receiving first reference signals corresponding to M time instants configured by the network side device, one or more first reference signals corresponding to each time instant.

[0227] Optionally, the first reference signals are reference signals for a first beam set. The first beam set can be a SetB beam, where SetA is a set of all beams and SetB is a subset of SetA, or SetA is a set of narrow beams and SetB is a set of wide beams.

[0228] The step 2012a comprises: measuring the first reference signals corresponding to the M time instants, M being a positive integer.

[0229] Here, the measurement results of the first reference signals corresponding to the M time instants correspond to the input of the AI model.

[0230] The step 2013a comprises: obtaining prediction results at each of N time instants according to the measurement results of the first reference signals corresponding to the M time instants.

[0231] Specifically, the measurement results of the first reference signals corresponding to the M time instants are input into the AI model to obtain the prediction results at each of the N time instants. The prediction results at each of the N time instants can comprise K indexes predicted at each of the N time instants, can comprise RSRP prediction values of target reference signals corresponding to each of the N time instants, and can comprise RSRP prediction values for the target reference signals.

[0232] Here, the K indices of each of the N time instants predicted can refer to the indices of the top-K reference signals at each of the N time instants predicted. It should be understood that the top-K reference signals refer to the K reference signals with the largest or strongest signal strength (such as RSRP).

[0233] In the case where the AI model is an AI beam management model, the reference signals are transmitted in a certain beam direction, that is, the reference signals have a corresponding relationship with the beams, and the indices of the top-K reference signals at each of the N time instants predicted can also be understood as the indices of the top-K beams (Top-K beams) at each of the N time instants predicted.

[0234] Optionally, the target reference signal is determined by the first reference signal or the prediction result. In the case where the AI model is an AI beam management model, the target reference signal determined by the first reference signal or the prediction result can be understood as including reference signals for the first beam set (SetB beams), or the target reference signal includes reference signals corresponding to the top-K beams determined by the prediction result.

[0235] Step 2014a, receiving the second reference signals corresponding to the N time instants configured by the network side device, one or more second reference signals corresponding to each time instant;

[0236] Optionally, the second reference signals are reference signals for the second beam set, reference signals for the first beam set, or reference signals for the top-K beams at each of the N time instants. The second beam set can be SetA beams, and the first beam set can be SetB beams. Here, SetA is a set of all beams, and SetB is a subset of SetA. Alternatively, SetA is a set of narrow beams, and SetB is a set of wide beams.

[0237] Step 2015a, measuring the second reference signals corresponding to the N time instants;

[0238] Step 2016a, obtaining performance monitoring related information for the first object according to the prediction result of each of the N time instants and the measurement result of each of the N time instants.

[0239] It should be noted that the measurement result of each of the N time instants is a result obtained by performing the above step 2015a.

[0240] Here, the above embodiments are applicable to the scenario where the AI model is a time-domain beam prediction model.

[0241] As an optional implementation manner one, the above step 202, reporting the performance monitoring related information to the network side device, comprises:

[0242] (1) reporting the number of correctly predicted time instants or the prediction accuracy in the N time instants to the network side device, wherein the characteristic values of the performance monitoring indicators of the N time instants include the number of correctly predicted time instants or the prediction accuracy in the N time instants; here, the number of correctly predicted time instants or the prediction accuracy in the N time instants is included in the mean value of the performance monitoring indicators of the N time instants.

[0243] (2) reporting first indication information to the network side device, wherein the first indication information is used to indicate whether the model inference of each time instant of the N time instants is correct, and the performance monitoring related information of each time instant of the N time instants includes the first indication information; wherein the relationship between (1) and (2) is "and / or".

[0244] Optionally, the first indication information is an N-bit bitmap.

[0245] In this implementation, the performance monitoring indicator is a beam prediction accuracy related KPI.

[0246] Based on this, in an optional embodiment, the prediction result of each time instant of the N time instants includes K indexes of the N time instants predicted; that is, the AI model outputs the predicted indexes. Here, the K indexes of the N time instants predicted can refer to the indexes of the maximum or strongest K reference signals of each time instant of the N time instants predicted. In the case of an AI beam management model, the reference signal is transmitted in a certain beam direction, that is, the reference signal has a corresponding relationship with the beam, and then the indexes of the maximum or strongest K reference signals of each time instant of the N time instants predicted can also be understood as the indexes of the maximum or strongest K beams (Top-K beams) of each time instant of the N time instants predicted.

[0247] The measurement result of each time instant of the N time instants includes the indexes of the maximum or strongest K reference signals of each time instant of the N time instants measured, K being a positive integer;

[0248] Correspondingly, the above step 2016a obtains the performance monitoring related information for the first object according to the prediction result of each time instant of the N time instants and the measurement result of each time instant of the N time instants, including:

[0249] I) obtaining the number of correctly predicted time instants or the prediction accuracy in the N time instants according to the K indexes of each time instant of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each time instant of the N time instants measured;

[0250] Specifically, by comparing the K indexes obtained by prediction with the K indexes obtained by measurement, the number of time instants or the prediction accuracy of the N time instants is obtained.

[0251] II) obtaining first indication information according to the K indexes of each time instant of the N time instants obtained by prediction and the indexes of the maximum or strongest K reference signals of each time instant of the N time instants obtained by measurement, the first indication information being used to indicate whether the model inference of each time instant of the N time instants is correct.

[0252] Specifically, by comparing the K indexes obtained by prediction with the K indexes obtained by measurement, it is determined whether the model inference of each time instant of the N time instants is correct, and then indicated by the first indication information.

[0253] The relationship between I) and II) is "and / or".

[0254] It should be noted that by reporting the number of time instants or the prediction accuracy of the N time instants or the first indication information, the reporting overhead can be reduced.

[0255] In order to obtain the indexes of the maximum or strongest K reference signals of each time instant of the N time instants obtained by measurement, correspondingly, the step 2015a, measuring the second reference signals corresponding to the N time instants, comprises:

[0256] Measuring the second reference signals corresponding to the N time instants to obtain the first RSRP measurement values of each time instant of the N time instants;

[0257] Here, since each time instant corresponds to one or more second reference signals, it should be understood that the first RSRP measurement values of each time instant of the N time instants obtained by measurement, wherein the first RSRP measurement value of each time instant refers to the RSRP measurement value of the one or more second reference signals corresponding to each time instant.

[0258] The first RSRP measurement values of each time instant of the N time instants are sorted to obtain the indexes of the maximum or strongest K reference signals of each time instant of the N time instants.

[0259] The following describes reporting the number of time instants of the N time instants by an embodiment.

[0260] Embodiment one

[0261] In this embodiment, the AI / ML model on the UE side is a time-domain beam prediction model, the input of the model is the measurement results of M time instants SetB beams, and the output is the index of the optimal beam (Top-K beam) in the N time instants SetA beams.

[0262] 1. The base station configures the UE to report the number of predicted correct moments among the N predicted moments. For example, the base station instructs the UE to report 1 bit to indicate the number of predicted correct moments, where:

[0263] 2. The base station sends the reference signal of the SetB beam at M times;

[0264] 3. The UE measures the reference signals of the SetB beam received at M moments to obtain the L1-RSRP of the SetB beam. The L1-RSRP of the SetB beam is used as the input of the AI ​​model for inference to obtain the index of the optimal beam in the SetA beam at N predicted moments.

[0265] Here, a SetB beam is a beam set. It should be understood that the reference signal for a SetB beam refers to the reference signal for a beam set, where one beam corresponds to one reference signal. The reference signal for a SetB beam at M time instants corresponds to one reference signal at each time instant, meaning that each time instant corresponds to one or more reference signals.

[0266] The UE measures the reference signal of the SetB beam at M moments to obtain the L1-RSRP. Specifically, the UE obtains M groups of L1-RSRP, with one group of L1-RSRP corresponding to each moment.

[0267] 4. The base station sends the reference signals of the SetA beam at N moments predicted by the UE to the UE. The UE measures the reference signals of the SetA beam at N moments, obtains the RSRP measurement values ​​of the SetA beam at N moments, and sorts the RSRP measurement values ​​of SetA at N moments to obtain the index of the optimal beam among the SetA beams at N moments.

[0268] It should be noted that the understanding of the reference signal of the SetA beam can be referred to the understanding of the reference signal of the SetB beam in the above step 2; the understanding of the reference signal of the SetA beam at N moments can be referred to the understanding of the reference signal of the SetB beam at M moments in the above step 2.

[0269] 5. The UE compares the prediction result in step 3 with the measurement result in step 4 to determine whether the optimal beam at the N time moments is correctly predicted, and obtains the number of correctly predicted time moments among the N predicted time moments. Furthermore, the UE can obtain the prediction accuracy based on the number of correctly predicted time moments and N.

[0270] If N=4, the UE reports 3 bits to indicate the number of correctly predicted moments, as shown in Table 1 below:

[0271] Table 1

[0272]

[0273]

[0274] Suppose that the number of correct prediction time is 3 in N = 4 time, UE reports 011.

[0275] 6、UE reports the number of correct prediction time in N predicted time.

[0276] 7、After the base station receives the UE's report information, it can calculate the beam prediction accuracy of this monitoring to be 75%, and the base station can also average the UE's multiple reporting results to calculate the beam prediction accuracy. If the accuracy is low, the base station can instruct the UE to deactivate the AI model or AI function.

[0277] The implementation process of reporting the first indication information is described below by way of example.

[0278] Example Two

[0279] In this embodiment, the UE-side AI / ML model is a time-domain beam prediction model, and the input and output of the AI / ML model are the same as in Example One.

[0280] 1、The base station configures the UE to report N bits, and the i-th bit represents the prediction result of the i-th time in N predicted time, wherein 0 represents prediction error and 1 represents prediction correct.

[0281] 2、The base station sends the reference signal of the SetB beam of M time;

[0282] 3、The UE measures the L1-RSRP of the SetB beam based on the received reference signal of the SetB beam of M time, and the L1-RSRP of the SetB beam is used as the AI model input for inference to obtain the index of the optimal beam in the SetA beam of N predicted time.

[0283] 4、The base station sends the reference signal of the SetA beam of N predicted time to the UE, and the UE measures the RSRP of the SetA beam of N time, sorts the RSRP measurement value of the SetA of N time, and obtains the index of the optimal beam in the SetA beam of N time.

[0284] 5、The UE compares the prediction result in step 3 and the measurement result in step 4 to determine whether the optimal beam of N time is predicted correctly, and obtains whether the model inference of each time of N predicted time is correct.

[0285] Suppose that the number of correct prediction time is 3 in N = 4 time, UE reports 011.

[0286] 6. The UE reports a 4-bit bitmap (first indication information) to indicate whether the model inference at each of the N moments is accurate. Figure 3 As shown, 0 represents an incorrect prediction and 1 represents a correct prediction.

[0287] 7. After receiving the UE's report, the base station can calculate that the beam prediction accuracy for this monitoring period is 75%, and that the beam prediction at the second moment is incorrect. The base station can average the UE's multiple reports to calculate the beam prediction accuracy. If the accuracy is low, the base station can instruct the UE to deactivate the AI ​​model or AI function.

[0288] As an optional implementation method 2, the above step 202, reporting the performance monitoring related information to the network side device, includes:

[0289] Report one or more of the following information to the network device:

[0290] A first mean value, where the first mean value is the mean of the differences between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to N moments;

[0291] The maximum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments;

[0292] Time information corresponding to the maximum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N time moments;

[0293] The minimum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments;

[0294] The time information corresponding to the minimum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N time moments.

[0295] Here, the characteristic values ​​of the performance monitoring indicators at N moments include one or more of the first mean, the maximum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments, and the minimum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments.

[0296] The moment-related information corresponding to the characteristic values ​​of the performance monitoring indicators at N moments includes the moment-related information corresponding to the maximum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments, and / or the moment-related information corresponding to the minimum value of the difference between the RSRP predicted value and the RSRP measured value of the target reference signal corresponding to the N moments.

[0297] In this implementation, the performance monitoring indicator is a difference between a predicted value and a measured value of RSRP of a predicted beam output by the model.

[0298] To obtain the above-mentioned reporting information, as an optional implementation three, the prediction result of each of the N time instants includes a predicted value of RSRP of the target reference signal corresponding to the each of the N time instants; and the measurement result of each of the N time instants includes a measured value of RSRP of the target reference signal corresponding to the each of the N time instants.

[0299] Correspondingly, the step 2016a obtains performance monitoring related information for the first object according to the prediction result of each of the N time instants and the measurement result of each of the N time instants, including:

[0300] According to the predicted value of RSRP of the target reference signal corresponding to each of the N time instants and the measured value of RSRP of the target reference signal corresponding to each of the N time instants, a difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to each of the N time instants is calculated.

[0301] According to the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to each of the N time instants, one or more of the following information is obtained:

[0302] a first mean value, the first mean value being a mean value of the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to the N time instants;

[0303] a maximum value of the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to the N time instants;

[0304] time instant related information corresponding to the maximum value of the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to the N time instants;

[0305] a minimum value of the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to the N time instants;

[0306] time instant related information corresponding to the minimum value of the difference between the predicted value of RSRP and the measured value of RSRP of the target reference signal corresponding to the N time instants.

[0307] To implement the above-mentioned implementation two and implementation three, the prediction result of each of the N time instants needs to be obtained, which is to obtain the predicted value of RSRP of the target reference signal corresponding to each of the N time instants.

[0308] As an optional implementation manner four, the step 2013a comprises:

[0309] inputting the measurement results of the first reference signals corresponding to the M time points into an artificial intelligence model to obtain RSRP prediction values of second reference signals corresponding to the N time points;

[0310] obtaining RSRP prediction values of target reference signals corresponding to the N time points according to the RSRP prediction values of the second reference signals corresponding to the N time points.

[0311] It should be noted that, in the actual execution of the method, the terminal device first executes the implementation manner four, i.e., the step of obtaining the prediction results of the N time points, then executes the implementation manner three, i.e., the step of obtaining the reporting information, and finally executes the implementation manner two, i.e., the step of reporting the obtained reporting information to the network side device.

[0312] Optionally, the target reference signal is determined by the first reference signal or the prediction result. In the case where the AI model is an AI beam management model, the target reference signal determined by the first reference signal or the prediction result can be understood as that the target reference signal includes reference signals for the first beam set (SetB beam), or the target reference signal includes reference signals corresponding to the maximum or strongest K beams determined by the prediction result. Therefore, there are two execution processes including the above-mentioned implementation manners two, three and four, which are as follows:

[0313] Process one:

[0314] Step 1-1, in the case where the target reference signal is the first reference signal, inputting the measurement results of the first reference signals corresponding to the M time points into an artificial intelligence model to obtain RSRP prediction values of second reference signals corresponding to the N time points;

[0315] Step 1-2, obtaining RSRP prediction values of the first reference signals corresponding to the N time points according to the RSRP prediction values of the second reference signals corresponding to the N time points;

[0316] For example, the first reference signal is a reference signal for a SetB beam, and the second reference signal is a reference signal for a SetA beam. The measurement results of the reference signals of the SetB beams corresponding to the M time points are input into the AI model to obtain the RSRP prediction values of the reference signals of the SetA beams corresponding to the N time points. Since the SetB beams are a subset of the SetA beams, the RSRP prediction values of the reference signals of the SetB beams corresponding to the N time points can be obtained from the RSRP prediction values of the reference signals of the SetA beams corresponding to the N time points.

[0317] Step 1-3, in the case that the prediction results of the N time points each include the RSRP prediction values of the first reference signals corresponding to the N time points each, and the measurement results of the N time points each include the RSRP measurement values of the first reference signals corresponding to the N time points each, the difference between the RSRP prediction value and the RSRP measurement value of the first reference signal corresponding to each time point is calculated according to the RSRP prediction value of the first reference signal corresponding to each time point and the RSRP measurement value of the first reference signal corresponding to each time point;

[0318] Step 1-4, according to the difference between the RSRP prediction value and the RSRP measurement value of the first reference signal corresponding to each time point, one or more of the following information is obtained:

[0319] a first mean value, the first mean value being a mean value of the differences between the RSRP prediction values and the RSRP measurement values of the first reference signals corresponding to the N time points;

[0320] a maximum value of the differences between the RSRP prediction values and the RSRP measurement values of the first reference signals corresponding to the N time points;

[0321] time related information corresponding to the maximum value of the differences between the RSRP prediction values and the RSRP measurement values of the first reference signals corresponding to the N time points;

[0322] a minimum value of the differences between the RSRP prediction values and the RSRP measurement values of the first reference signals corresponding to the N time points;

[0323] time related information corresponding to the minimum value of the differences between the RSRP prediction values and the RSRP measurement values of the first reference signals corresponding to the N time points.

[0324] Step 1-5, one or more of the following information is reported to the network side device:

[0325] a first mean value of differences between RSRP prediction values and RSRP measurement values of the first reference signal corresponding to the N time instants;

[0326] a maximum value of differences between RSRP prediction values and RSRP measurement values of the first reference signal corresponding to the N time instants;

[0327] time information corresponding to a maximum value of differences between RSRP prediction values and RSRP measurement values of the first reference signal corresponding to the N time instants;

[0328] a minimum value of differences between RSRP prediction values and RSRP measurement values of the first reference signal corresponding to the N time instants;

[0329] time information corresponding to a minimum value of differences between RSRP prediction values and RSRP measurement values of the first reference signal corresponding to the N time instants.

[0330] Flow II:

[0331] Step 2-1, in the case where the target reference signal includes the maximum or strongest K reference signals, input the measurement results of the first reference signal corresponding to the M time instants into the artificial intelligence model to obtain RSRP prediction values of the second reference signal corresponding to each of the N time instants;

[0332] Step 2-2, according to the RSRP prediction values of the second reference signal corresponding to each of the N time instants, obtain RSRP prediction values of the maximum or strongest K reference signals corresponding to each of the N time instants;

[0333] For example, the first reference signal is a reference signal for SetB beams, and the second reference signal is a reference signal for SetA beams. The measurement results of the reference signal of the SetB beams corresponding to the M time instants are input into the AI model to obtain RSRP prediction values of the reference signal of the SetA beams corresponding to each of the N time instants. The RSRP prediction values of the reference signal of the SetA beams corresponding to each of the N time instants are sorted to obtain RSRP prediction values of the maximum or strongest K reference signals corresponding to each of the N time instants.

[0334] Step 2-3, in the case that the prediction result at each of the N time instants comprises RSRP prediction values of the maximum or strongest K reference signals corresponding to the each of the N time instants, and the measurement result at each of the N time instants comprises RSRP measurement values of the maximum or strongest K reference signals corresponding to the each of the N time instants, the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to each of the N time instants is calculated according to the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to each of the N time instants;

[0335] Step 2-4, one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to each of the N time instants:

[0336] a first mean value, the first mean value being a mean value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0337] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0338] time related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0339] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0340] time related information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants.

[0341] Step 2-5, one or more of the following information is reported to the network side device:

[0342] a first mean value, the first mean value being a mean value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0343] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0344] time related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the maximum or strongest K reference signals corresponding to the N time instants;

[0345] a minimum value of a difference between the RSRP measurement value and the RSRP prediction value of the maximum or strongest K reference signals corresponding to the N moments;

[0346] time information corresponding to a minimum value of a difference between the RSRP measurement value and the RSRP prediction value of the maximum or strongest K reference signals corresponding to the N moments.

[0347] The following embodiments illustrate the implementation process of the UE performing information reporting when the target reference signal is the first reference signal.

[0348] Embodiment Three

[0349] In this embodiment, the AI / ML model on the UE side is a time-domain beam prediction model, and the input of the model is the measurement result of the SetB beam at M moments, and the output is the L1-RSRP prediction value of the SetA beam at N moments.

[0350] 1. The base station configures the UE to report the average value of the difference between the L1-RSRP prediction value and the L1-RSRP measurement value of the SetB beam at N moments.

[0351] 2. The base station sends the reference signals of the SetB beam at M moments.

[0352] 3. The UE measures the received reference signals of the SetB beam at M moments to obtain the L1-RSRP of the SetB beam, which is used as the input of the AI model for inference to obtain the predicted L1-RSRP prediction value of the SetA beam corresponding to each moment at N moments. Assuming that the SetB beam is a subset of the SetA beam, the L1-RSRP prediction value of the SetB beam corresponding to each moment at N moments can be obtained.

[0353] 4. The base station sends the reference signals of the SetB beam predicted by the UE at N moments, and the UE measures the reference signals of the SetB beam at N moments to obtain the L1-RSRP measurement value of the SetB corresponding to each moment at N moments (corresponding to the RSRP measurement value of the target reference signal corresponding to each moment at N moments, and the target reference signal is the first reference signal, i.e., the reference signal of the SetB beam).

[0354] 5. The UE calculates the difference between the L1-RSRP measurement value and the prediction value of the SetB corresponding to each moment at N moments, and obtains the average value of the difference between the L1-RSRP measurement value and the prediction value of the SetB at N moments by averaging the differences at N moments.

[0355] 6、UE reports the average of the difference between the L1-RSRP measurement value and the predicted value of SetB at N time points to the base station.

[0356] 7、The base station can average the reporting results of the UE multiple times as a monitoring index, and if the absolute value of the average is greater than 1dB, it is considered that the performance of the AI model or AI function is poor, and the UE can be instructed to deactivate the AI model or AI function.

[0357] Embodiment Four

[0358] In this embodiment, the UE-side AI / ML model is a time-domain beam prediction model, and the input and output of the AI / ML model are the same as in Embodiment Three.

[0359] 1、The base station configures the UE to report the maximum and minimum values of the difference between the L1-RSRP prediction value and the L1-RSRP measurement value of SetB beams at N time points, as well as the corresponding time information.

[0360] 2、The base station sends reference signals of SetB beams at M time points;

[0361] 3、The UE measures the received reference signals of SetB beams at M time points to obtain the L1-RSRP of SetB beams, which is used as the input of the AI model for inference to obtain the predicted L1-RSRP prediction value of SetA beams corresponding to each time point at N time points. Assuming that SetB beams are a subset of SetA beams, the L1-RSRP prediction value of SetB beams corresponding to each time point at N time points can be obtained.

[0362] 4、The base station sends the reference signals of SetB beams predicted by the UE at N time points, and the UE measures the reference signals of SetB beams at N time points to obtain the L1-RSRP measurement value of SetB corresponding to each time point at N time points (corresponding to the RSRP measurement value of the target reference signal at each time point at N time points, the target reference signal being the first reference signal, i.e. the reference signal for SetB beams);

[0363] 5、The UE calculates the difference between the L1-RSRP measurement value and the predicted value of SetB corresponding to each time point at N time points, and the calculation result is that the difference at the first time point is the maximum, and the difference at the third time point is the minimum. Assuming that 4 bits are used to represent the difference between the L1-RSRP measurement value and the predicted value of SetB, and 2 bits are used to represent which time point the reported difference corresponds to.

[0364] The UE reports the content as shown in Table 2 below:

[0365] Table 2

[0366] Reporting bit Meaning 1101 Maximum difference between RSRP measurement and prediction 0010 Minimum difference between RSRP measurement and prediction 00 Time corresponding to maximum difference 10 Time corresponding to minimum difference

[0367] 6、Base station can know the accuracy of the predicted optimal beam at N time points according to the reporting result of UE, such as considering that the difference is greater than a certain threshold, the prediction is incorrect. The base station can average the reporting results of the UE multiple times as a monitoring index, so as to judge the performance of the AI model or AI function on the UE side.

[0368] The following will illustrate the implementation process of UE performing information reporting in the case that the target reference signal includes the maximum or strongest K reference signals through examples.

[0369] Example five

[0370] In this embodiment, the AI / ML model on the UE side is a time domain beam prediction model, and the input and output of the AI / ML model are the same as those in example three.

[0371] 1、The base station configures the content reported by the UE as the average of the difference between the predicted L1-RSRP value and the measured L1-RSRP value of the Top-2 beams in the predicted N time points.

[0372] 2、The base station sends the reference signals of the SetB beams at M time points;

[0373] 3、The UE measures the L1-RSRP of the SetB beams based on the received reference signals of the SetB beams at M time points, and the L1-RSRP of the SetB beams is used as the input of the AI model for inference to obtain the predicted L1-RSRP value of the SetA corresponding to each time point in the predicted N time points. The predicted L1-RSRP values of the SetA corresponding to each time point in the predicted N time points are sorted to obtain the indexes of the Top-2 beams at each time point in the predicted N time points.

[0374] 4、The UE reports the indexes of the Top-2 beams at each time point in the predicted N time points to the base station.

[0375] 5、The base station sends the reference signals of the Top-2 beams at each time point in the predicted N time points, and the UE measures the L1-RSRP of the Top-2 beams at each time point in the N time points to obtain the L1-RSRP measurement value of the Top-2 beams at each time point in the N time points (corresponding to the RSRP measurement value of the target reference signal corresponding to each time point in the N time points, the target reference signal includes the maximum or strongest K reference signals);

[0376] 6、The UE calculates the difference between the L1-RSRP measurement value and the predicted value of the Top-2 beams at each time point, averages the differences in the N time points to obtain the average of the differences between the L1-RSRP measurement value and the predicted value of the Top-2 beams in the N time points, and reports it to the base station.

[0377] 7、The base station can average the multiple reporting results of the UE and use the average as a monitoring indicator. If the absolute value of the average is greater than 1 dB, it is considered that the performance of the AI model or AI function is poor, and the UE can be instructed to deactivate the AI model or AI function.

[0378] Embodiment six

[0379] In this embodiment, the UE-side AI / ML model is a time-domain beam prediction model, and the input and output of the AI / ML model are the same as in Embodiment Three.

[0380] 1、The base station configures the UE to report the maximum value of the difference between the predicted L1-RSRP value and the measured L1-RSRP value of the Top-2 beams at the predicted N time points.

[0381] 2、The base station sends reference signals of the SetB beams at M time points;

[0382] 3、The UE measures the L1-RSRP of the SetB beams based on the received reference signals of the SetB beams at M time points, and uses the L1-RSRP of the SetB beams as AI model input for inference to obtain the predicted L1-RSRP value of the SetA at each of the predicted N time points. The predicted L1-RSRP values of the SetA at each of the predicted N time points are sorted to obtain the index of the Top-2 beams at each of the predicted N time points.

[0383] 4、The UE reports the index of the Top-2 beams at each of the predicted N time points to the base station.

[0384] 5、The base station sends reference signals of the Top-2 beams at each of the predicted N time points, and the UE measures the L1-RSRP of the Top-2 beams at each of the N time points to obtain the L1-RSRP measurement value of the Top-2 beams at each of the N time points (corresponding to the RSRP measurement value of the target reference signal at each of the N time points, and the target reference signal includes the maximum or strongest K reference signals);

[0385] 6、The UE calculates the difference between the L1-RSRP measurement value and the predicted value of the Top-2 beams at each time point, and the calculation result is that the difference at the second time point is the largest.

[0386] The UE reports the following Table 3:

[0387] Table 3

[0388] Reporting bit Meaning 1101 Maximum difference between RSRP measurement and prediction 01 Time corresponding to maximum difference

[0389] 7、Base station can know the accuracy of the predicted optimal beam at the N time according to the reporting result of UE, such as considering that the difference is greater than a certain threshold, the prediction is incorrect. The base station can average the reporting results of the UE multiple times as a monitoring index, so as to judge the performance of the AI model or AI function on the UE side.

[0390] As an optional implementation mode five, the step 202 of reporting the performance monitoring related information to the network side device comprises:

[0391] The step 202 of reporting the performance monitoring related information to the network side device comprises:

[0392] A maximum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period, P being a positive integer greater than 1; here, the first time period is indicated by the network side device, indicating a period of time.

[0393] A maximum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period, P being a positive integer greater than 1; here, the first time period is indicated by the network side device, indicating a period of time.

[0394] A minimum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period.

[0395] A minimum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period.

[0396] The feature value of the performance monitoring index corresponding to the multiple model inferences comprises: a maximum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period, and / or a minimum value of a difference between a predicted RSRP value and a measured RSRP value corresponding to the P times of measurement or measurement in the first time period.

[0397] Here, the above implementation mode is applicable to the scenario where the AI model is an airspace beam prediction model.

[0398] Based on this, as an optional implementation mode six, the step 201 of obtaining the performance monitoring related information for the first object comprises:

[0399] The step 2011b receives a first reference signal configured by the network side device for one measurement.

[0400] Optionally, the first reference signal is a reference signal for a first beam set. Wherein, the first beam set can be SetB beams, where SetA is a set of all beams, and SetB is a subset of SetA; or SetA is a set of narrow beams, and SetB is a set of wide beams.

[0401] Step 2012b, measuring the first reference signal;

[0402] Step 2013b, obtaining a RSRP prediction value for a target reference signal according to the measurement result of the first reference signal;

[0403] Step 2014b, calculating a difference between the RSRP prediction value for the target reference signal and a RSRP measurement value for the target reference signal according to the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal;

[0404] Step 2015b, after performing P times of measurements or measurements within a first time period, obtaining one or more of the following information according to the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to each measurement:

[0405] A maximum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0406] Time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0407] A minimum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0408] Time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period.

[0409] It should be noted that in the actual execution steps of the method, the terminal device first executes the implementation mode six, i.e., the step of obtaining the reported information, and then executes the implementation mode five, i.e., the step of reporting the obtained reported information to the network side device.

[0410] Optionally, the target reference signal is determined by the first reference signal or the prediction result. In the case that the AI model is an AI beam management model, the target reference signal is determined by the first reference signal or the prediction result, which can be understood as that the target reference signal includes the reference signal for the first beam set (SetB beam), or the target reference signal includes the reference signal corresponding to the maximum or strongest K beams determined by the prediction result. Therefore, there are two execution flows including the above-mentioned implementation mode five and implementation mode six, which are as follows:

[0411] Flow one:

[0412] Step 1a, in the case that the target reference signal is the first reference signal, for one measurement, the first reference signal configured by the network side device is received;

[0413] Step 2a, the first reference signal is measured;

[0414] Step 3a, according to the measurement result of the first reference signal, an RSRP prediction value for the first reference signal is obtained;

[0415] Here, the measurement result of the first reference signal is input into the AI model to obtain the RSRP prediction value for the first reference signal.

[0416] For example, the first reference signal is the reference signal for the SetB beam, and the measurement result of the reference signal for the SetB beam is taken as the input of the AI model to obtain the RSRP prediction value for the reference signal of the SetA beam; it is known that the SetB beam is a subset of the SetA beam, and based on the predicted RSRP prediction value for the reference signal of the SetA beam, the RSRP prediction value for the reference signal of the SetB beam can be obtained.

[0417] Step 4a, according to the RSRP prediction value for the first reference signal and the RSRP measurement value for the first reference signal, a difference value between the RSRP prediction value and the RSRP measurement value for the first reference signal is calculated;

[0418] Step 5a, after performing P times of measurements or measurements within a first time period, according to the difference value between the RSRP prediction value and the RSRP measurement value for the first reference signal corresponding to each measurement, one or more of the following information is obtained:

[0419] The maximum value of the difference value between the RSRP prediction value and the RSRP measurement value for the first reference signal corresponding to P times of measurements or measurements within a first time period;

[0420] a time point related information or an index value of a measurement corresponding to a maximum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period;

[0421] a minimum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period;

[0422] a time point related information or an index value of a measurement corresponding to a minimum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period.

[0423] Flow II:

[0424] Step 1b, in the case that the target reference signals include the maximum or strongest K reference signals, receiving the first reference signal configured by the network side device for one measurement;

[0425] Step 2b, measuring the first reference signal;

[0426] Step 3b, obtaining the RSRP prediction value for the maximum or strongest K reference signals according to the measurement result of the first reference signal;

[0427] Here, the measurement result of the first reference signal is input to the AI model to obtain the RSRP prediction value for the maximum or strongest K reference signals.

[0428] For example, the first reference signal is a reference signal for SetB beam, the measurement result of the reference signal for SetB beam is taken as the input of the AI model to obtain the RSRP prediction value for the reference signal for SetA beam; based on the predicted RSRP prediction value for the reference signal for SetA beam, the RSRP prediction value for the maximum or strongest K beams (corresponding reference signals) of SetA beam is obtained.

[0429] Step 4b, calculating the difference between the RSRP prediction value and the RSRP measurement value for the maximum or strongest K reference signals according to the RSRP prediction value for the maximum or strongest K reference signals and the RSRP measurement value for the maximum or strongest K reference signals;

[0430] Here, the RSRP measurement value for the maximum or strongest K reference signals can be obtained by the following steps:

[0431] The UE reports the indexes of the maximum or strongest K beams of the SetA beams to the network side device; then receives the maximum or strongest K reference signals corresponding to the SetA beams (i.e., the reference signals of the maximum or strongest K beams of the SetA beams) sent by the network side device, and then the UE performs measurement on the maximum or strongest K reference signals to obtain the RSRP measurement values of the maximum or strongest K reference signals.

[0432] After the P times of measurement or the measurement within the first time period is performed, one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value corresponding to each measurement for the maximum or strongest K reference signals:

[0433] The maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P times of measurement or the measurement within the first time period for the maximum or strongest K reference signals;

[0434] The time related information or the index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P times of measurement or the measurement within the first time period for the maximum or strongest K reference signals;

[0435] The minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P times of measurement or the measurement within the first time period for the maximum or strongest K reference signals;

[0436] The time related information or the index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P times of measurement or the measurement within the first time period for the maximum or strongest K reference signals.

[0437] The implementation process of reporting the characteristic values of the performance monitoring indicators of the multiple times of model inference is described below through an embodiment.

[0438] Embodiment Seven

[0439] In this embodiment, the AI / ML model on the UE side is a spatial beam prediction model, the input of the model is the measurement result of the SetB beam, and the output is the RSRP prediction value of the SetA beam. By sorting the RSRP prediction value of the SetA beam, the index corresponding to the optimal beam at each time in the predicted N times of time is obtained.

[0440] 1. The base station configures the UE to periodically report the maximum and minimum values of the difference between the RSRP prediction value and the RSRP measurement value of the SetB beam within the performance monitoring time T. T is the reporting period, i.e., the UE reports the maximum and minimum values of the difference between the RSRP prediction value and the RSRP measurement value of the SetB beam within the reporting interval.

[0441] 2. The base station transmits the reference signal of the SetB beam;

[0442] 3. The UE measures the received reference signal of the SetB beam to obtain the RSRP of the SetB beam, which is used as the input of the AI model for inference to obtain the predicted RSRP of the SetA beam. Assuming that the SetB is a subset of the SetA, the predicted RSRP of the SetB can be obtained; the UE calculates the difference between the predicted RSRP and the measured RSRP of the SetB;

[0443] 4. Steps 2 and 3 are repeated until the measurement within the time T indicated by the base station is performed;

[0444] 5. Assuming that the base station transmits the reference signal corresponding to the SetB beam P times within the time T, the UE respectively takes the RSRP of the reference signal corresponding to the SetB beam P times as the input of the model for inference to obtain P groups of predicted RSRP of the SetA beam. Assuming that the SetB is a subset of the SetA, the predicted RSRP of the SetB beam can be obtained from the inference result of the UE. According to the measurement and inference result, the UE obtains the maximum and minimum of the difference between the predicted RSRP and the measured RSRP of the P SetB beams, and reports them to the base station. It should be noted that in addition to reporting the maximum and minimum of the difference between the predicted RSRP and the measured RSRP of the P SetB beams, the time information corresponding to the maximum and minimum of the difference or the corresponding monitoring times can also be reported.

[0445] 6. The base station calculates the monitoring index according to the reporting result of the UE, such as taking the average of the multiple reporting results of the UE as the monitoring index, so as to judge the performance of the AI model or AI function. If the performance is poor, the UE can be instructed to deactivate the AI model or AI function.

[0446] In some embodiments, the method of the present application further comprises:

[0447] receiving the configuration information sent by the network side device, wherein the configuration information is used to indicate the content of the performance monitoring related information reported by the terminal device for the first object.

[0448] Here, the terminal device reports the content required by the network side device according to the indication of the configuration information sent by the network side device, so as to meet the performance monitoring requirements of the network side device.

[0449] The information reporting method of the embodiments of the present application comprises: obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; and reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; and time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences. In this way, by reporting the above performance monitoring related information to the network side device, the performance monitoring of the artificial intelligence model or the artificial intelligence function on the terminal device side by the network side can be realized.

[0450] As shown in Figure 4 FIG. 1 is a flowchart of an information transmission method provided by the embodiments of the present application, which is applied to a network side device, i.e., the method is executed by the network side device. The method specifically comprises the following steps:

[0451] In step 401, performance monitoring related information reported by a terminal device is received, wherein the performance monitoring related information comprises one or more of the following information:

[0452] characteristic values of performance monitoring indexes at N time instants;

[0453] performance monitoring related information at each of the N time instants;

[0454] time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1;

[0455] characteristic values of performance monitoring indexes corresponding to multiple model inferences;

[0456] time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences.

[0457] It should be noted that the method executed by the network side device corresponds to the method executed by the terminal device side, and the understanding of related concepts and steps can refer to the terminal device side, which will not be repeated here.

[0458] In some embodiments, the characteristic values comprise one or more of a mean value, a maximum value and a minimum value.

[0459] In some embodiments, the method of the present application further comprises:

[0460] configure the terminal device with first reference signals corresponding to M time instants, each time instant corresponding to one or more first reference signals, the first reference signals corresponding to the M time instants being used to obtain prediction results for each of N time instants, M being a positive integer;

[0461] configure the terminal device with second reference signals corresponding to N time instants, each time instant corresponding to one or more second reference signals, the second reference signals corresponding to the N time instants being used to obtain measurement results for each of the N time instants.

[0462] In some embodiments, the method of the present application further comprises:

[0463] for each of P measurements or within a first time period, configure the terminal device with a first reference signal, the first reference signal being used to obtain an RSRP prediction value for a target reference signal, the target reference signal being determined by the first reference signal or the prediction result, P being a positive integer greater than 1.

[0464] In some embodiments, the method of the present application further comprises:

[0465] send configuration information to the terminal device, the configuration information being used to indicate the content of the performance monitoring related information reported by the terminal device for a first object.

[0466] Here, the network side device sends configuration information to the terminal device, so that the terminal device reports the content required by the network side device in accordance with the indication of the configuration information, thereby meeting the performance monitoring requirements of the network side device.

[0467] The information transmission method of the embodiments of the present application receives the performance monitoring related information reported by the terminal device, wherein the performance monitoring related information includes one or more of the following information: characteristic values of performance monitoring indicators at N time instants; performance monitoring related information at each of the N time instants; time related information corresponding to characteristic values of performance monitoring indicators at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indicators corresponding to multiple model inferences; time related information corresponding to characteristic values of performance monitoring indicators corresponding to multiple model inferences, so that by receiving the above performance monitoring related information reported by the terminal device, the performance monitoring of the artificial intelligence model or artificial intelligence function on the terminal device side by the network side can be realized.

[0468] As shown in Figure 5 The embodiments of the present application also provide a terminal device, which comprises: a memory 520, a transceiver 500, and a processor 510; the memory 520 is used to store program instructions; the transceiver 500 is used to transceive data under the control of the processor 510; the processor 510 is used to read the program instructions in the memory 520 and perform the following operations:

[0469] obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0470] reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information:

[0471] feature values of performance monitoring indexes at N time instants;

[0472] performance monitoring related information at each of the N time instants;

[0473] time instant related information corresponding to the feature values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1;

[0474] feature values of performance monitoring indexes corresponding to multiple model inferences;

[0475] time instant related information corresponding to the feature values of the performance monitoring indexes corresponding to the multiple model inferences.

[0476] wherein, in the Figure 5 The bus architecture can include any number of interconnecting buses and bridges, depending on the specific application of the processor 510, and can link such various circuits as the processor 510 and the memory 520. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, all of which are well known in the art, and therefore, will not be described any further. The bus interface provides an interface to the transceiver 500. The transceiver 500 can be a plurality of elements, including a transmitter and a receiver, that provide means for communicating with various other apparatus over a transmission medium, including a wireless channel, a wired channel, optical cable, and the like. The user interface 530 can also be an interface to other means for allowing a user or operator to interact with the apparatus, including but not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.

[0477] The processor 510 is responsible for managing the bus architecture and general processing, and the memory 520 can store data used by the processor 510 in executing operations.

[0478] Optionally, the processor 510 can be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or a CPLD (Complex Programmable Logic Device), and the processor 510 can also adopt a multi-core architecture.

[0479] The processor 510 is configured to execute any of the methods provided by the embodiments of the present application by invoking stored program instructions of the memory. The processor 510 and the memory 520 can also be physically arranged separately.

[0480] In some embodiments, the characteristic value includes one or more of a mean value, a maximum value, and a minimum value.

[0481] In some embodiments, the operations further include:

[0482] receiving first reference signals corresponding to M time instants configured by the network-side device, each time instant corresponding to one or more first reference signals;

[0483] measuring the first reference signals corresponding to the M time instants; wherein M is a positive integer;

[0484] obtaining a prediction result of each of N time instants according to a measurement result of the first reference signals corresponding to the M time instants;

[0485] receiving second reference signals corresponding to the N time instants configured by the network-side device, each time instant corresponding to one or more second reference signals;

[0486] measuring the second reference signals corresponding to the N time instants;

[0487] obtaining performance monitoring related information for the first object according to the prediction result of each of the N time instants and a measurement result of each of the N time instants.

[0488] In some embodiments, the operations further include:

[0489] reporting, to the network-side device, a number of correctly predicted time instants or a correct prediction rate in the N time instants, and a characteristic value of a performance monitoring indicator of the N time instants includes the number of correctly predicted time instants or the correct prediction rate in the N time instants; and / or,

[0490] reporting first indication information to the network side device, the first indication information being used to indicate whether the model inference of each of the N time instants is correct, the performance monitoring related information of each of the N time instants including the first indication information.

[0491] In some embodiments, the prediction result of each of the N time instants includes K indexes of the N time instants predicted, K being a positive integer; the measurement result of each of the N time instants includes indexes of the maximum or strongest K reference signals of each of the N time instants measured; and the operations further include:

[0492] obtaining, according to the K indexes of each of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each of the N time instants measured, the number of time instants correctly predicted or the prediction accuracy rate in the N time instants; and / or,

[0493] obtaining, according to the K indexes of each of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each of the N time instants measured, the first indication information, the first indication information being used to indicate whether the model inference of each of the N time instants is correct.

[0494] In some embodiments, the first indication information is an N-bit bitmap.

[0495] In some embodiments, the operations further include:

[0496] performing measurement on the second reference signals corresponding to the N time instants to obtain first RSRP measurement values of each of the N time instants;

[0497] sorting the first RSRP measurement values of each of the N time instants to obtain indexes of the maximum or strongest K reference signals of each of the N time instants.

[0498] In some embodiments, the operations further include:

[0499] reporting one or more of the following information to the network side device:

[0500] a first average value, the first average value being an average value of differences between RSRP prediction values and RSRP measurement values of target reference signals corresponding to the N time instants;

[0501] a maximum value of differences between RSRP prediction values and RSRP measurement values of target reference signals corresponding to the N time instants;

[0502] information of a time corresponding to a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0503] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0504] information of a time corresponding to a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

[0505] In some embodiments, the prediction result of each of the N time instants includes a RSRP prediction value of the target reference signal corresponding to the each of the N time instants; the measurement result of each of the N time instants includes a RSRP measurement value of the target reference signal corresponding to the each of the N time instants; and the operations further include:

[0506] a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants is calculated according to the RSRP prediction value of the target reference signal corresponding to each of the N time instants and the RSRP measurement value of the target reference signal corresponding to each of the N time instants;

[0507] one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N time instants:

[0508] a first mean value, the first mean value being a mean value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0509] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0510] information of a time corresponding to a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0511] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0512] information of a time corresponding to a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

[0513] In some embodiments, the operations further include:

[0514] inputting a measurement result of the first reference signal corresponding to the M time points into an artificial intelligence model to obtain an RSRP prediction value of the second reference signal corresponding to each of the N time points;

[0515] obtaining, according to the RSRP prediction value of the second reference signal corresponding to each of the N time points, an RSRP prediction value of a target reference signal corresponding to each of the N time points.

[0516] In some embodiments, the operations further include:

[0517] reporting one or more of the following information to the network side device:

[0518] a maximum value of a difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period, P being a positive integer greater than 1;

[0519] time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0520] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0521] time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period;

[0522] The feature value of the performance monitoring index corresponding to the multiple model inferences includes: the maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period, and / or the minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to P times of measurement or measurement within the first time period.

[0523] In some embodiments, the operations further include:

[0524] receiving, for one measurement, a first reference signal configured by the network side device;

[0525] measuring the first reference signal;

[0526] obtaining, according to a measurement result of the first reference signal, an RSRP prediction value of a target reference signal;

[0527] According to the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal, a difference between the RSRP prediction value and the RSRP measurement value for the target reference signal is calculated.

[0528] After performing P times of measurements or measurements within a first time period, according to the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to each measurement, one or more of the following information is obtained:

[0529] The maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to P times of measurements or measurements within a first time period;

[0530] The maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to P times of measurements or measurements within a first time period corresponds to time-related information or an index value of the measurement;

[0531] The minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to P times of measurements or measurements within a first time period;

[0532] The minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to P times of measurements or measurements within a first time period corresponds to time-related information or an index value of the measurement.

[0533] In some embodiments, the target reference signal is determined by a first reference signal or a prediction result.

[0534] In some embodiments, the operations further include:

[0535] Receiving configuration information sent by the network side device, the configuration information being used to indicate the content of the performance monitoring related information for the first object reported by the terminal device.

[0536] The terminal device of the embodiments of the present application, by obtaining the performance monitoring related information for the first object, the first object being an artificial intelligence model or an artificial intelligence function; reporting the performance monitoring related information to the network side device, wherein the performance monitoring related information includes one or more of the following information: characteristic values of performance monitoring indicators at N time points; performance monitoring related information at each time point at N time points; time-related information corresponding to the characteristic values of the performance monitoring indicators at N time points, N being a positive integer greater than 1; characteristic values of performance monitoring indicators corresponding to multiple model inferences; time-related information corresponding to the characteristic values of the performance monitoring indicators corresponding to multiple model inferences, so that by reporting the above performance monitoring related information to the network side device, the performance monitoring of the network side to the artificial intelligence model or the artificial intelligence function on the terminal device side can be realized

[0537] As Figure 6 shown, the embodiments of the present application also provide an information reporting device, comprising:

[0538] An acquisition unit 601 is configured to acquire performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0539] A reporting unit 602 is configured to report the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information:

[0540] Characteristic values of performance monitoring indicators at N time instants;

[0541] Performance monitoring related information at each of the N time instants;

[0542] Time instant related information corresponding to the characteristic values of the performance monitoring indicators at the N time instants, N being a positive integer greater than 1;

[0543] Characteristic values of performance monitoring indicators corresponding to multiple model inferences;

[0544] Time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences.

[0545] In some embodiments, the characteristic values include one or more of a mean value, a maximum value, and a minimum value.

[0546] In some embodiments, the acquisition unit 601 is specifically configured to:

[0547] receive first reference signals corresponding to M time instants configured by the network side device, one or more first reference signals corresponding to each time instant;

[0548] measure the first reference signals corresponding to the M time instants; wherein M is a positive integer;

[0549] obtain predicted results at each of the N time instants according to measurement results of the first reference signals corresponding to the M time instants;

[0550] receive second reference signals corresponding to N time instants configured by the network side device, one or more second reference signals corresponding to each time instant;

[0551] measure the second reference signals corresponding to the N time instants;

[0552] obtain performance monitoring related information for the first object according to the predicted results at each of the N time instants and measurement results at each of the N time instants.

[0553] In some embodiments, the reporting unit 602 is specifically configured to:

[0554] report, to the network-side device, the number of correctly predicted time instants or the prediction accuracy among the N time instants, and the characteristic values of the performance monitoring indicators of the N time instants include the number of correctly predicted time instants or the prediction accuracy among the N time instants; and / or,

[0555] report, to the network-side device, first indication information, and the performance monitoring related information of each time instant of the N time instants includes the first indication information, where the first indication information is used to indicate whether the model inference of each time instant of the N time instants is correct.

[0556] In some embodiments, the prediction result of each time instant of the N time instants includes K indexes of the N time instants predicted, and the measurement result of each time instant of the N time instants includes indexes of the maximum or strongest K reference signals of each time instant of the N time instants measured, where K is a positive integer; accordingly, the obtaining unit 601 is specifically configured to:

[0557] obtain, according to the K indexes of each time instant of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each time instant of the N time instants measured, the number of correctly predicted time instants or the prediction accuracy among the N time instants; and / or,

[0558] obtain, according to the K indexes of each time instant of the N time instants predicted and the indexes of the maximum or strongest K reference signals of each time instant of the N time instants measured, first indication information, where the first indication information is used to indicate whether the model inference of each time instant of the N time instants is correct.

[0559] In some embodiments, the first indication information is an N-bit bitmap.

[0560] In some embodiments, the obtaining unit 601 is specifically configured to:

[0561] measure the second reference signals corresponding to the N time instants to obtain first RSRP measurement values of each time instant of the N time instants;

[0562] sort the first RSRP measurement values of each time instant of the N time instants to obtain indexes of the maximum or strongest K reference signals of each time instant of the N time instants.

[0563] In some embodiments, the reporting unit 602 is specifically configured to:

[0564] report, to the network-side device, one or more of the following information:

[0565] a first average value, the first average value being an average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0566] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0567] time instant related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0568] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0569] time instant related information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

[0570] In some embodiments, the prediction result of each time instant of the N time instants comprises the RSRP prediction value of the target reference signal corresponding to the each time instant of the N time instants; the measurement result of each time instant of the N time instants comprises the RSRP measurement value of the target reference signal corresponding to the each time instant of the N time instants; accordingly, the obtaining unit 601 is specifically configured to:

[0571] the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each time instant is calculated according to the RSRP prediction value of the target reference signal corresponding to each time instant of the N time instants and the RSRP measurement value of the target reference signal corresponding to each time instant of the N time instants;

[0572] one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each time instant:

[0573] a first average value, the first average value being an average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0574] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0575] time instant related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0576] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants;

[0577] an instant corresponding to a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N instants.

[0578] In some embodiments, the obtaining unit 601 is specifically configured to:

[0579] input the measurement result of the first reference signal corresponding to the M instants into an artificial intelligence model to obtain the RSRP prediction value of the second reference signal corresponding to each of the N instants;

[0580] obtain the RSRP prediction value of the target reference signal corresponding to each of the N instants according to the RSRP prediction value of the second reference signal corresponding to each of the N instants.

[0581] In some embodiments, the reporting unit 602 is specifically configured to:

[0582] report one or more of the following information to the network side device:

[0583] a maximum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period, P being a positive integer greater than 1;

[0584] instant information or an index value of measurement corresponding to a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period;

[0585] a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period;

[0586] instant information or an index value of measurement corresponding to a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period;

[0587] The feature value of the performance monitoring index corresponding to the multiple model inferences includes: a maximum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period, and / or a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period.

[0588] In some embodiments, the obtaining unit 601 is specifically configured to:

[0589] for one measurement, receive the first reference signal configured by the network side device;

[0590] measure the first reference signal;

[0591] obtain a RSRP prediction value for a target reference signal according to the measurement result of the first reference signal;

[0592] calculate a difference between the RSRP prediction value and the RSRP measurement value for the target reference signal according to the RSRP prediction value and the RSRP measurement value for the target reference signal;

[0593] after performing P times of measurements or measurements within a first time period, obtain one or more of the following information according to the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to each measurement:

[0594] a maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0595] time related information or an index value of the measurement corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0596] a minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period;

[0597] time related information or an index value of the measurement corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to the P times of measurements or the measurements within the first time period.

[0598] In some embodiments, the target reference signal is determined by the first reference signal or the prediction result.

[0599] In some embodiments, the apparatus of the present application further comprises:

[0600] a second receiving unit, configured to receive configuration information sent by a network side device, the configuration information being used to indicate content of performance monitoring related information for a first object reported by the terminal device.

[0601] The information reporting device provided in the embodiments of the present application acquires performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; and reports the performance monitoring related information to a network side device, wherein the performance monitoring related information includes one or more of the following information: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; and time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences. In this way, by reporting the above performance monitoring related information to the network side device, the network side can perform performance monitoring on the artificial intelligence model or the artificial intelligence function on the terminal device side.

[0602] It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division. In actual implementation, another division manner 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 exist physically, 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.

[0603] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solutions of the present application, essentially or in part, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several 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 methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and various other media that can store program codes.

[0604] It should be noted that the above device provided in the embodiments of the present application can implement all the method steps achieved by the method embodiments, and can achieve the same technical effects. Therefore, the same parts and beneficial effects of the method embodiments in the embodiments will not be described in detail.

[0605] In some embodiments of the present application, a processor-readable storage medium is also provided, which stores program instructions for causing the processor to perform the following steps:

[0606] obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function;

[0607] reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information includes one or more of the following information:

[0608] characteristic values of performance monitoring indicators at N time instants;

[0609] performance monitoring related information at each of the N time instants;

[0610] time instant related information corresponding to the characteristic values of the performance monitoring indicators at the N time instants, N being a positive integer greater than 1;

[0611] characteristic values of performance monitoring indicators corresponding to multiple model inferences;

[0612] time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences.

[0613] The program is executed by the processor to achieve all the implementation manners of the above-mentioned method embodiments applied to the terminal device side as shown in Figure 2 For the sake of brevity, the above-mentioned method embodiments applied to the terminal device side will not be described here again.

[0614] As shown in Figure 7 The embodiments of the present application also provide a network side device, which comprises a memory 720, a transceiver 700, and a processor 710: the memory 720 is used to store computer programs; the transceiver 700 is used to transceive data under the control of the processor 710; the processor 710 is used to read program instructions in the memory 720 and perform the following operations:

[0615] receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information includes one or more of the following information:

[0616] characteristic values of performance monitoring indicators at N time instants;

[0617] performance monitoring related information at each of the N time instants;

[0618] time instant related information corresponding to the characteristic values of the performance monitoring indicators at the N time instants, N being a positive integer greater than 1;

[0619] characteristic values of performance monitoring indicators corresponding to multiple model inferences;

[0620] The performance monitoring index corresponding to the multiple model inferences corresponds to the time-related information of the characteristic value.

[0621] In the formula, the bus architecture can include any number of interconnected buses and bridges, which link together various circuits of the processor 710, which represents one or more processors, and the memory 720, which represents memory. Figure 7 The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and thus will not be further described herein. The bus interface provides an interface. The transceiver 700 can be a plurality of elements, i.e., including a transmitter and a receiver, which provide units for communicating with various other devices on transmission media, including wireless channels, wired channels, optical cables, and the like transmission media.

[0622] The processor 710 is responsible for managing the bus architecture and general processing, and the memory 720 can store data used by the processor 710 when performing operations.

[0623] Optionally, the processor 710 can be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a CPLD (Complex Programmable Logic Device), and the processor 710 can also adopt a multi-core architecture.

[0624] The processor 710 is used to execute any of the methods provided by the embodiments of the application according to the executable instructions obtained by invoking the program instructions stored in the memory. The processor 710 and the memory 720 can also be physically arranged separately.

[0625] In some embodiments, the characteristic value includes one or more of the mean value, the maximum value, and the minimum value.

[0626] In some embodiments, the operations further include:

[0627] The terminal device is configured with M time instants corresponding to first reference signals, each time instant corresponding to one or more first reference signals, and the M time instants corresponding to the first reference signals are used to obtain prediction results at each of the N time instants, M being a positive integer;

[0628] The terminal device is configured with N time instants corresponding to second reference signals, each time instant corresponding to one or more second reference signals, and the N time instants corresponding to the second reference signals are used to obtain measurement results at each of the N time instants.

[0629] In some embodiments, the operations further include:

[0630] configuring, for each of P measurements or within a first time period, the terminal device with a first reference signal, the first reference signal being used to obtain a RSRP prediction value for a target reference signal, the target reference signal being determined by the first reference signal or the prediction result, P being a positive integer greater than 1.

[0631] In some embodiments, the operations further include:

[0632] sending, to the terminal device, configuration information, the configuration information being used to indicate content of performance monitoring related information reported by the terminal device for a first object.

[0633] The network side device of the embodiments of the present application receives the performance monitoring related information reported by the terminal device, wherein the performance monitoring related information includes one or more of the following information: characteristic values of performance monitoring indicators at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to characteristic values of performance monitoring indicators at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indicators corresponding to multiple model inferences; time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences, so that by receiving the above performance monitoring related information reported by the terminal device, the performance monitoring of the artificial intelligence model or artificial intelligence function on the terminal device side by the network side can be realized.

[0634] As shown in Figure 8 The embodiments of the present application also provide an information transmission device, which includes:

[0635] A first receiving unit 801 is configured to receive performance monitoring related information reported by a terminal device, wherein the performance monitoring related information includes one or more of the following information:

[0636] Characteristic values of performance monitoring indicators at N time instants;

[0637] Performance monitoring related information at each of the N time instants;

[0638] Time instant related information corresponding to characteristic values of performance monitoring indicators at the N time instants, N being a positive integer greater than 1;

[0639] Characteristic values of performance monitoring indicators corresponding to multiple model inferences;

[0640] Time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences.

[0641] In some embodiments, the feature values include one or more of a mean value, a maximum value, and a minimum value.

[0642] In some embodiments, the apparatus further includes:

[0643] a first processing unit configured to configure, for the terminal device, first reference signals corresponding to M time instants, each time instant corresponding to one or more first reference signals, the first reference signals corresponding to the M time instants being used to obtain a prediction result for each of N time instants, M being a positive integer;

[0644] a second processing unit configured to configure, for the terminal device, second reference signals corresponding to the N time instants, each time instant corresponding to one or more second reference signals, the second reference signals corresponding to the N time instants being used to obtain a measurement result for each of the N time instants.

[0645] In some embodiments, the apparatus further includes:

[0646] a third processing unit configured to configure, for the terminal device, a first reference signal for each measurement in P measurements or in a first time period, the first reference signal being used to obtain a predicted value of RSRP for a target reference signal, the target reference signal being determined by the first reference signal or the prediction result, P being a positive integer greater than 1.

[0647] In some embodiments, the apparatus further includes:

[0648] a first sending unit configured to send, to the terminal device, configuration information, the configuration information being used to indicate content of performance monitoring related information reported by the terminal device for a first object.

[0649] The information transmission apparatus according to an embodiment of the present application receives performance monitoring related information reported by a terminal device, wherein the performance monitoring related information includes one or more of the following information: feature values of performance monitoring indicators at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to feature values of performance monitoring indicators at the N time instants, N being a positive integer greater than 1; feature values of performance monitoring indicators corresponding to multiple model inferences; and time instant related information corresponding to feature values of performance monitoring indicators corresponding to the multiple model inferences. In this way, by receiving the performance monitoring related information reported by the terminal device, the network side can perform performance monitoring on an artificial intelligence model or an artificial intelligence function on the terminal device side.

[0650] It should be noted that the division of the unit in the embodiments of the present application is illustrative, and is only a logical function division. In actual implementation, another division manner 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 separately, 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.

[0651] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product 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 methods 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 program code storage media.

[0652] It should be noted that the above device provided by the embodiments of the present application can realize all method steps achieved by the above method embodiments, and can achieve the same technical effects. Here, the same parts and beneficial effects of the method embodiments in the embodiments will not be described in detail.

[0653] In some embodiments of the present application, a processor-readable storage medium is also provided, which stores program instructions for causing the processor to perform the following steps:

[0654] Receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information includes one or more of the following information:

[0655] Characteristic values of performance monitoring indexes at N time instants;

[0656] Performance monitoring related information at each of the N time instants;

[0657] Characteristic values of performance monitoring indexes at N time instants;

[0658] Characteristic values of performance monitoring indexes corresponding to multiple model inferences;

[0659] The performance monitoring index corresponding to the multiple model inferences corresponds to the characteristic value of the moment-related information.

[0660] The program, when executed by a processor, can implement all implementation manners of the above-mentioned method embodiments applied to the network side device as shown in Figure 4 The above-mentioned method embodiments applied to the network side device as shown in

[0661] In some embodiments of the present application, a computer program product is also provided, which includes computer instructions, and the computer instructions, when executed by a processor, implement each process of the method embodiments as shown in Figure 2 or Figure 4 The above-mentioned method embodiments applied to the network side device as shown in

[0662] The technical solutions provided by the embodiments of the present application can be applied to various systems, such as a 5G system and above. For example, the applicable system can be a Global System of Mobile communication (GSM) system, a Code Division Multiple Access (CDMA) system, a Wideband Code Division Multiple Access (WCDMA) General Packet Radio Service (GPRS) system, a Long Term Evolution (LTE) system, an LTE Frequency Division Duplex (FDD) system, an LTE Time Division Duplex (TDD) system, a Long Term Evolution Advanced (LTE-A) system, a Universal Mobile Telecommunication System (UMTS), a Worldwide interoperability for Microwave Access (WiMAX) system, a 5G New Radio (NR) system, and the like. Among the 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.

[0663] 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.

[0664] 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 evolutional network device (evolutional Node B, eNB or e-NodeB) in the Long Term Evolution (LTE) system, or a 5G base station (gNB) in the 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.

[0665] The network device and the terminal device can each use one or more antennas for multi-input multi-output (MIMO) transmission, which can be single-user MIMO (SU-MIMO) or multiple-user MIMO (MU-MIMO). According to the form and number of root antenna combinations, the MIMO transmission can be 2D-MIMO, 3D-MIMO, FD-MIMO, or massive-MIMO, and can also be diversity transmission or precoding transmission or beamforming transmission, etc.

[0666] Those skilled in the art will understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage, etc.) containing computer-usable program code.

[0667] The present application is described with reference to flowcharts and / or block diagrams according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer executable instructions. These computer executable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in the flowcharts and / or block diagrams.

[0668] These processor executable instructions can also be stored in a processor readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the processor readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that implements the functions specified in the flowcharts and / or block diagrams.

[0669] These processor executable instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 the steps of a function specified in one or more blocks.

[0670] It is clear that many modifications and changes can be made to the application without departing from the spirit and scope of the application. It is therefore intended that such modifications and changes be included within the scope of the application as measured by the claims and their equivalents.

Claims

1. An information reporting method, applied to a terminal device, characterized in that: The method comprises: obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; reporting the performance monitoring related information to a network side device, wherein the performance monitoring related information comprises one or more of the following information: characteristic values of performance monitoring indicators at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indicators at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indicators corresponding to multiple model inferences; time instant related information corresponding to the characteristic values of the performance monitoring indicators corresponding to the multiple model inferences.

2. The method of claim 1, wherein, The characteristic values comprise one or more of a mean value, a maximum value, and a minimum value.

3. The method of claim 1, wherein, The obtaining of the performance monitoring related information for the first object comprises: receiving first reference signals corresponding to M time instants configured by the network side device, one or more first reference signals corresponding to each time instant; performing measurement on the first reference signals corresponding to the M time instants; wherein M is a positive integer; obtaining prediction results at each of the N time instants according to the measurement results of the first reference signals corresponding to the M time instants; receiving second reference signals corresponding to the N time instants configured by the network side device, one or more second reference signals corresponding to each time instant; performing measurement on the second reference signals corresponding to the N time instants; obtaining the performance monitoring related information for the first object according to the prediction results at each of the N time instants and the measurement results at each of the N time instants.

4. The method according to any one of claims 1 to 3, characterized in that, The reporting of the performance monitoring related information to the network side device comprises: reporting, to the network side device, a number of time instants correctly predicted or a correct prediction rate among the N time instants, wherein the characteristic values of the performance monitoring indicators at the N time instants comprise the number of time instants correctly predicted or the correct prediction rate among the N time instants; and / or reporting, to the network side device, first indication information, wherein the first indication information is used to indicate whether model inference at each of the N time instants is correct, and wherein the performance monitoring related information at each of the N time instants comprises the first indication information.

5. The method of claim 3, wherein, The prediction results at each of the N time instants comprise K indexes predicted at each of the N time instants, and the measurement results at each of the N time instants comprise indexes of maximum or strongest K reference signals measured at each of the N time instants, K being a positive integer; The obtaining of the performance monitoring related information for the first object according to the prediction results at each of the N time instants and the measurement results at each of the N time instants comprises: obtaining, according to the K indexes predicted at each of the N time instants and the indexes of the maximum or strongest K reference signals measured at each of the N time instants, the number of time instants correctly predicted or the correct prediction rate among the N time instants; and / or According to the K indexes of each of the N moments obtained by prediction and the indexes of the maximum or strongest K reference signals of each of the N moments obtained by measurement, first indication information is obtained, the first indication information being used to indicate whether the model inference of each of the N moments is correct.

6. The method of claim 4, wherein, The first indication information is an N-bit bitmap.

7. The method of claim 5, wherein, The measurement of the second reference signals corresponding to the N moments comprises: The measurement of the second reference signals corresponding to the N moments obtains first RSRP measurement values of each of the N moments; The first RSRP measurement values of each of the N moments are sorted to obtain indexes of the maximum or strongest K reference signals of each of the N moments.

8. The method according to any one of claims 1 to 3, characterized in that, The reporting of the performance monitoring related information to the network side device comprises: The reporting of one or more of the following information to the network side device comprises: a first average value, the first average value being an average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; time related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; time related information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments.

9. The method of claim 3, wherein, The prediction result of each of the N moments comprises an RSRP prediction value of the target reference signal corresponding to each of the N moments; and the measurement result of each of the N moments comprises an RSRP measurement value of the target reference signal corresponding to each of the N moments; The obtaining of the performance monitoring related information for the first object according to the prediction result of each of the N moments and the measurement result of each of the N moments comprises: a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N moments is calculated; one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each of the N moments: a first average value, the first average value being an average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; time related information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N moments; a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; time instant information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

10. The method of claim 9, wherein, The obtaining of the prediction result of each of the N time instants from the measurement result of the first reference signal corresponding to the M time instants comprises: inputting the measurement result of the first reference signal corresponding to the M time instants into an artificial intelligence model to obtain the RSRP prediction value of the second reference signal corresponding to each of the N time instants; obtaining the RSRP prediction value of the target reference signal corresponding to each of the N time instants from the RSRP prediction value of the second reference signal corresponding to each of the N time instants.

11. The method of claim 1, wherein, The reporting of the performance monitoring related information to the network side device comprises: reporting one or more of the following information to the network side device: a maximum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period, P being a positive integer greater than 1; time instant information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period, or an index value of the measurement; a minimum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period; time instant information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period, or an index value of the measurement; The feature value of the performance monitoring indicator corresponding to the multiple model inferences comprises: the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period, and / or the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period.

12. The method of claim 11, wherein, The obtaining of the performance monitoring related information for the first object comprises: receiving a first reference signal configured by the network side device for one measurement; performing measurement on the first reference signal; obtaining a RSRP prediction value for a target reference signal from the measurement result of the first reference signal; calculating a difference between the RSRP prediction value and the RSRP measurement value for the target reference signal from the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal; After performing P measurements or measurements within a first time period, obtaining one or more of the following information from the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to each measurement: a maximum value of a difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P measurements or measurements within a first time period; a time point corresponding to a maximum value of a difference between a predicted value of a reference signal received power (RSRP) of a target reference signal and a measured value of the RSRP corresponding to P measurements or measurements within a first time period, or an index value of the measurement; a minimum value of a difference between a predicted value of a reference signal received power (RSRP) of a target reference signal and a measured value of the RSRP corresponding to P measurements or measurements within a first time period; a time point corresponding to a minimum value of a difference between a predicted value of a reference signal received power (RSRP) of a target reference signal and a measured value of the RSRP corresponding to P measurements or measurements within a first time period, or an index value of the measurement.

13. The method according to any one of claims 8 to 12, characterized in that, The target reference signal is determined by a first reference signal or a prediction result.

14. The method of claim 1, wherein, The method further comprises: receiving configuration information sent by a network side device, the configuration information being used to indicate content of performance monitoring related information reported by the terminal device for a first object.

15. An information transmission method applied to a network side device, comprising: comprises: receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following information: a characteristic value of a performance monitoring index at N time points; performance monitoring related information at each of the N time points; time point related information corresponding to a characteristic value of a performance monitoring index at N time points, N being a positive integer greater than 1; a characteristic value of a performance monitoring index corresponding to multiple model inferences; time point related information corresponding to a characteristic value of a performance monitoring index corresponding to multiple model inferences.

16. The method of claim 15, wherein, The characteristic value comprises one or more of a mean value, a maximum value, and a minimum value.

17. The method according to claim 15 or 16, characterized in that, The method further comprises: configuring, for the terminal device, a first reference signal corresponding to M time points, one or more first reference signals corresponding to each time point, the first reference signal corresponding to the M time points being used to obtain a prediction result at each of the N time points, M being a positive integer; configuring, for the terminal device, a second reference signal corresponding to N time points, one or more second reference signals corresponding to each time point, the second reference signal corresponding to the N time points being used to obtain a measurement result at each of the N time points.

18. The method of claim 15 or 16, wherein, The method further comprises: configuring, for the terminal device, a first reference signal for each of P measurements or measurements within a first time period, the first reference signal being used to obtain a predicted value of a reference signal received power (RSRP) of a target reference signal, the target reference signal being determined by a first reference signal or a prediction result, P being a positive integer greater than 1.

19. The method of claim 15 or 16, wherein, The method further comprises: sending, to the terminal device, configuration information, the configuration information being used to indicate content of performance monitoring related information reported by the terminal device for a first object.

20. A terminal device, comprising: comprises: a memory, a transceiver, and a processor: the memory is used to store program instructions; the transceiver is used to transceive data under control of the processor; the processor is used to read program instructions in the memory and perform the following operations: obtaining performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; reporting, to a network side device, the performance monitoring related information, wherein the performance monitoring related information comprises one or more of the following information: a characteristic value of a performance monitoring index at N time points; performance monitoring related information at each of the N time points; time point related information corresponding to a characteristic value of a performance monitoring index at N time points, N being a positive integer greater than 1; a characteristic value of a performance monitoring index corresponding to multiple model inferences; time point related information corresponding to a characteristic value of a performance monitoring index corresponding to multiple model inferences. The characteristic value of the performance monitoring index corresponding to the time-related information of the N moments, N being a positive integer greater than 1; The characteristic value of the performance monitoring index corresponding to the multiple model inferences; The characteristic value of the performance monitoring index corresponding to the multiple model inferences corresponds to the time-related information of the N moments.

21. The terminal device of claim 20, wherein, The characteristic value includes one or more of the mean value, the maximum value, and the minimum value.

22. The terminal device of claim 20, wherein, The operations further include: Receiving the first reference signals corresponding to the M moments configured by the network side device, one or more first reference signals corresponding to each moment; Measuring the first reference signals corresponding to the M moments; M being a positive integer; Obtaining the prediction result of each moment of the N moments according to the measurement result of the first reference signals corresponding to the M moments; Receiving the second reference signals corresponding to the N moments configured by the network side device, one or more second reference signals corresponding to each moment; Measuring the second reference signals corresponding to the N moments; Obtaining the performance monitoring related information for the first object according to the prediction result of each moment of the N moments and the measurement result of each moment of the N moments.

23. The terminal device of any one of claims 20 to 22, wherein, The operations further include: Reporting the number of correctly predicted moments or the correct prediction rate of the N moments to the network side device, the characteristic value of the performance monitoring index of the N moments including the number of correctly predicted moments or the correct prediction rate of the N moments; and / or, Reporting the first indication information to the network side device, the first indication information being used to indicate whether the model inference of each moment of the N moments is correct, the performance monitoring related information of each moment of the N moments including the first indication information.

24. The terminal device of claim 22, wherein, The prediction result of each moment of the N moments includes the K indexes of each moment of the N moments predicted, the measurement result of each moment of the N moments including the indexes of the maximum or strongest K reference signals of each moment of the N moments measured, K being a positive integer; the operations further include: Obtaining the number of correctly predicted moments or the correct prediction rate of the N moments according to the K indexes of each moment of the N moments predicted and the indexes of the maximum or strongest K reference signals of each moment of the N moments measured; and / or, Obtaining the first indication information according to the K indexes of each moment of the N moments predicted and the indexes of the maximum or strongest K reference signals of each moment of the N moments measured, the first indication information being used to indicate whether the model inference of each moment of the N moments is correct.

25. The terminal device of claim 23, wherein, The first indication information is an N-bit bitmap.

26. The terminal device of claim 24, wherein, The operations further include: Measuring the second reference signals corresponding to the N moments to obtain the first RSRP measurement value of each moment of the N moments; Sorting the first RSRP measurement value of each moment of the N moments to obtain the indexes of the maximum or strongest K reference signals of each moment of the N moments.

27. The terminal device according to any one of claims 20 to 22, characterized by, The operations further include: Reporting one or more of the following information to the network side device: a first average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; time information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; time information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

28. The terminal device of claim 22, wherein, The prediction result of each time instant of the N time instants includes the RSRP prediction value of the target reference signal corresponding to each time instant of the N time instants; the measurement result of each time instant of the N time instants includes the RSRP measurement value of the target reference signal corresponding to each time instant of the N time instants; and the operation further includes: calculating the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each time instant according to the RSRP prediction value of the target reference signal corresponding to each time instant of the N time instants and the RSRP measurement value of the target reference signal corresponding to each time instant of the N time instants; obtaining one or more of the following information according to the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to each time instant: a first average value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; time information corresponding to the maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants; time information corresponding to the minimum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to the N time instants.

29. The terminal device of claim 28, wherein, The operation further includes: inputting the measurement result of the first reference signal corresponding to the M time instants into an artificial intelligence model to obtain the RSRP prediction value of the second reference signal corresponding to each time instant of the N time instants; obtaining the RSRP prediction value of the target reference signal corresponding to each time instant of the N time instants according to the RSRP prediction value of the second reference signal corresponding to each time instant of the N time instants.

30. The terminal device of claim 20, wherein, The operation further includes: reporting one or more of the following information to the network side device: a maximum value of the difference between the RSRP prediction value and the RSRP measurement value of the target reference signal corresponding to P times of measurement or measurement within a first time period, P being a positive integer greater than 1; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; The feature values of the performance monitoring indicators corresponding to the multiple model inferences include: a maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period, and / or a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period.

31. The terminal device of claim 21, wherein, The operations further include: receiving a first reference signal configured by the network side device for one measurement; performing measurement on the first reference signal; obtaining an RSRP prediction value for a target reference signal according to the measurement result of the first reference signal; calculating the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal according to the RSRP prediction value for the target reference signal and the RSRP measurement value for the target reference signal; After performing the P measurements or the measurements within the first time period, one or more of the following information is obtained according to the difference between the RSRP prediction value and the RSRP measurement value for the target reference signal corresponding to each measurement: a maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; a maximum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period; a minimum value of the difference between the RSRP prediction value and the RSRP measurement value corresponding to the P measurements or the measurements within the first time period.

32. The terminal device of any one of claims 27 to 31, wherein, The target reference signal is determined by the first reference signal or the prediction result.

33. The terminal device of claim 20, wherein, The operations further include: receiving configuration information sent by the network side device, the configuration information being used to indicate the content of the performance monitoring related information for the first object reported by the terminal device.

34. An information reporting apparatus, characterized by comprising: including: an obtaining unit, configured to obtain performance monitoring related information for a first object, the first object being an artificial intelligence model or an artificial intelligence function; a reporting unit, configured to report the performance monitoring related information to a network side device, wherein the performance monitoring related information includes one or more of the following information: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences.

35. A network-side device, comprising: comprising: a memory, a transceiver, and a processor; the transceiver, configured to transceive data under control of the processor; the processor, configured to read program instructions in the memory and perform the following operations: receiving performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences.

36. The network-side device of claim 35, wherein, The characteristic values comprise one or more of a mean value, a maximum value, and a minimum value.

37. The network-side device of claim 35 or 36, characterized by The operations further comprise: configuring the terminal device with first reference signals corresponding to M time instants, one or more first reference signals corresponding to each time instant, the first reference signals corresponding to the M time instants being used to obtain predicted results at each of the N time instants, M being a positive integer; configuring the terminal device with second reference signals corresponding to the N time instants, one or more second reference signals corresponding to each time instant, the second reference signals corresponding to the N time instants being used to obtain measurement results at each of the N time instants.

38. The network-side device of claim 35 or 36, characterized by The operations further comprise: configuring the terminal device with a first reference signal for each measurement in P measurements or in a first time period, the first reference signal being used to obtain an RSRP prediction value for a target reference signal, the target reference signal being determined by the first reference signal or a predicted result, P being a positive integer greater than 1.

39. The network-side device of claim 35 or 36, characterized by The operations further comprise: sending configuration information to the terminal device, the configuration information being used to instruct the terminal device to report content of performance monitoring related information for a first object.

40. An information transmission apparatus, characterized by comprising: comprising: a first receiving unit, configured to receive performance monitoring related information reported by a terminal device, wherein the performance monitoring related information comprises one or more of the following: characteristic values of performance monitoring indexes at N time instants; performance monitoring related information at each of the N time instants; time instant related information corresponding to the characteristic values of the performance monitoring indexes at the N time instants, N being a positive integer greater than 1; characteristic values of performance monitoring indexes corresponding to multiple model inferences; time instant related information corresponding to the characteristic values of the performance monitoring indexes corresponding to the multiple model inferences.

41. A processor-readable storage medium, comprising: The processor readable storage medium stores a computer program, which is used to make the processor perform steps of the information reporting method in any one of claims 1 to 14, or perform steps of the information transmission method in any one of claims 15 to 19.

42. A computer program product, characterised in that, Computer program product comprising computer instructions which, when executed by a processor, implement the steps of the information reporting method according to any one of claims 1 to 14, or implement the steps of the information transmission method according to any one of claims 15 to 19.

Images