Data processing method and related device

Abstract

The invention provides a data processing method and a related device, in the method, a data sending device can obtain Nt groups of first data based on Nt groups of observation data in St groups of observation data corresponding to a tth time unit, and send the Nt groups of first data, so that a data receiving device compensates prediction data corresponding to the tth time unit; wherein the data in the St group of observation data is more than the data in the Nt group of observation data, i.e., the Nt group of first data is obtained based on part of the data in the St group of observation data; the prediction data corresponding to the tth time unit is obtained by the p prediction data corresponding to the previous p time units. According to the scheme, the prediction accuracy can be improved, and the transmission overhead is saved.

Description

Data processing method and related device Technical Field

[0001] The present application relates to the field of wireless communications, and in particular to a data processing method and related devices. Background Art

[0002] The application scenarios of wireless communication technology are becoming increasingly diverse, but new scenarios also bring more and more transmission overhead.

[0003] Currently, there are proposals to use historical data to predict future data, thereby reducing data transmission and saving transmission overhead. For example, the temporal correlation of data can be exploited to predict future data through time series prediction, such as autoregressive (AR) prediction.

[0004] However, during the data prediction process, errors accumulate over time, leading to distorted predictions. Therefore, compensation can be applied to the predicted data. However, this often requires feedback of complete observation data, which incurs significant transmission overhead.

[0005] Summary of the Invention

[0006] The present application provides a data processing method and related devices to improve the accuracy of predicted data while avoiding huge transmission overhead.

[0007] First, the present application provides a data processing method, which can be executed by a data sending device (hereinafter referred to as the sending device). Unless otherwise specified, the sending device in this application may refer to the sending device itself (for example, a network device, a terminal device), or a component in the sending device (for example, a chip, a chip system, or a processor, etc.), or it may refer to a logical module or software that can realize all or part of the functions of the sending device.

[0008] Exemplarily, the method includes: obtaining N t The first data of the group, the N t The first data of the group is based on the S corresponding to the t-th time unit t N in the set of observation data t The N t The first data of the group, the N t The first group of data is used to compensate the predicted data corresponding to the t-th time unit, where the predicted data is obtained based on p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

[0009] On the second aspect, a data processing method is provided, which can be executed by a data receiving device (referred to as the receiving device). Unless otherwise specified, the receiving device in this application can refer to the receiving device itself (for example, a network device, a terminal device), or a component in the receiving device (for example, a chip, a chip system, or a processor, etc.), or it can also be a logic module or software that can realize all or part of the functions of the receiving device.

[0010] Exemplarily, the method includes: receiving N t The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N sets of observation data t Determined by a set of observation data; based on the N t A first group of data is obtained, and the predicted data corresponding to the t-th time unit is compensated to obtain compensated data; the predicted data is obtained based on p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

[0011] Among them, S t The group of observation data may be obtained by grouping a set of observation data corresponding to the t-th time unit (hereinafter referred to as the observation data set), and the observation data corresponding to the t-th time unit may refer to data observed in the t-th time unit. Each group of observation data may include one or more observation data.

[0012] N t Can be less than S t The value of N t The set of observation data is S t A portion of the observation data, or N t The set of observation data is S t A subset of the observation data. In other words, the S t There are more data in the set of observations than the N t The data in the set of observation data, or in other words, the S t The number of observations in the set is greater than N t The number of data in the set of observation data. If each set of observation data includes one observation data, it can also be said that the S t The number of groups of observation data is greater than the N t The number of groups of observation data; if each group of observation data includes multiple observation data, it can also be called the S t The total amount of data in the set of observation data is greater than the N t The total amount of data for a group of observations.

[0013] N t The first data set is based on the N t The data to be transmitted is determined by the set of observation data. t The first data of the group can be N t The set of observation data itself can also be based on the N t This application does not limit this.

[0014] The sending device can t The first data is sent to another device communicating with the device, for example, a data receiving device (hereinafter referred to as a receiving device). The receiving device can be a network device or a terminal device. In other words, the communication can be between a network device and a terminal device, or between terminal devices, which is not limited in this application.

[0015] By adding N t The first group of data is sent to the receiving device, which can facilitate the receiving device to compensate for the predicted data corresponding to the t-th time unit based on the data. The predicted data corresponding to the t-th time unit can refer to the data predicted by the t-th time unit. The predicted data corresponding to the t-th time unit can be obtained based on historical data prediction. The historical data can, for example, include the predicted data of one or more time units before the t-th time unit, which are denoted as p in this article. P can be a predefined constant and p is a positive integer. The p time units can, for example, include the (t-p-a+1)th time unit to the (t-a)th time unit, and a can be a predefined constant. The p time units can be continuous p time units or discontinuous p time units, for example, they can be determined based on different sampling rates.

[0016] Based on the above scheme, the data sending device selects a part of the data from the set of observed data for data transmission, so that the data receiving device can compensate the predicted data based on the received data, thereby reconstructing the complete data. This makes the compensated data closer to the observed data, which also makes the predicted data output by the receiving device closer to the observed data, thereby improving the accuracy of the predicted data. Compared with sending the entire set of observed data, this method selects and sends the data in the set, which can avoid the huge transmission overhead caused by the transmission of all the data in the set; compared with not reporting the observed data, this method can use the part of the data selected to be sent to compensate for the predicted data, which is conducive to obtaining more accurate predicted data. Therefore, an effective compromise is achieved between transmission overhead and data accuracy.

[0017] In this application, since the data sending device selects a part of the data from the observation data set for data transmission, it is like compressing the entire observation data set. Therefore, for the convenience of explanation, this article calls this transmission method compressed transmission.

[0018] In combination with the first aspect or the second aspect, in some possible implementations, a is 1.

[0019] Due to the temporal correlation of data, when predicting historical data for the tth time unit, the closer the corresponding time node is to the tth time unit, the more accurate the data will be. Therefore, the p time units can be, for example, the p consecutive time units before the tth time unit, that is, including the (t-p)th time unit to the (t-1)th time unit. This allows for more accurate prediction data.

[0020] In combination with the first aspect or the second aspect, in some possible implementations, the N t The first data set is used to determine compensation data, and the compensation data is used to compensate the prediction data corresponding to the t-th time unit to obtain compensated data.

[0021] That is, the receiving device can t The first data of the group is used to determine the compensation data corresponding to the t-th time unit, and then the prediction data corresponding to the t-th time unit is compensated to obtain the compensated data corresponding to the t-th time unit.

[0022] More specifically, in some possible implementations of the second aspect, the t The first data set is used to compensate the predicted data corresponding to the t-th time unit to obtain compensated data, including: the N t The first data is used to determine compensation data; and the compensated data is obtained based on the compensation data and the predicted data corresponding to the t-th time unit.

[0023] In combination with the first aspect or the second aspect, in some possible implementations, the compensated data is the sum of the predicted data and the compensated data corresponding to the t-th time unit.

[0024] Determining the sum of the predicted data and the compensated data corresponding to the t-th time unit as the compensated data is only one possible design. For example, the compensated data can also be obtained by weighted summation of the predicted data and the compensated data corresponding to the t-th time unit. This application includes but is not limited to this.

[0025] In one example, the compensation data is recorded as ΔV(t), and the compensation data ΔV(t) satisfies: Wherein, K(t) is the Kalman gain matrix corresponding to the t-th time unit, For the N t The first data of the group is the predicted data corresponding to the t-th time unit, and the O(t) is based on the N t The observation data of the group is t The position in the set of observation data is determined.

[0026] The above parameters are all parameters corresponding to the t-th time unit. For example, ΔV(t) represents the compensation data corresponding to the t-th time unit.

[0027] Since K(t) is the Kalman gain matrix, the compensation data in this example can be obtained through KF prediction. K(t) can be iteratively updated based on the prediction covariance accumulated from the most recent prediction and historical predictions, as well as the covariance of noise introduced by compression and transmission.

[0028] N t The first data of the group is represented by a vector, which can be written as The predicted data corresponding to the t-th time unit is identified by a vector, which can be recorded as Assume S t The set of observation data includes L data, the vector and can be vectors of length L. t Vector of the first data of the group In addition to N t Other data besides the group prediction data are in S t The value of the corresponding position in the group prediction data is 0.

[0029] O(t) can be understood as being able to describe N t The observation data of the group is in S t The observation matrix of the position in the set of observation data can be a matrix of L×L dimensions. t In the set of observations, the rows and columns correspond to N t The value of the position where the group observation data is located can be 1, and the value of other positions can be 0.

[0030] is formed by transforming the vector Each element in the vector The vector obtained by subtracting the elements at corresponding positions in .

[0031] Based on the formula satisfied by the above compensation data ΔV(t), and the relationship between the compensated data and the predicted data and compensated data corresponding to the t-th time unit, the compensated data corresponding to the t-th time unit can be obtained. satisfy:

[0032] In the present application, KF prediction is an example of a prediction method for obtaining compensation data, and the present application includes but is not limited to this. For example, the compensation data can also be obtained based on a derivative algorithm of KF prediction, such as extended KF, adaptive KF, time-delayed KF, etc., or, it can also be calculated based on an artificial intelligence (AI) model, such as a recurrent neural network (RNN) model, a long short-term memory (LSTM) model, a transformer model, etc., or, it can also be obtained based on other types of filtering algorithms, such as arithmetic mean filtering, sliding average filtering, etc., which will not be repeated here.

[0033] In combination with the first aspect or the second aspect, in some possible implementations, the N t The observation data of the group is t Positions in the set of observations are determined based on a predefined first pattern.

[0034] The first pattern may be one of the predefined one or more patterns. Each pattern may be used to indicate N t The observation data of the group is in S t A distribution of positions in a set of observation data, and as t changes, N t The observation data of the group is in S t The position within a group of observations may also vary.

[0035] Determine N based on the pattern t The group observation data can be easily aligned with N by the sending device and the receiving device through simple signaling interaction. t The observation data of the group is in S t The position in the set of observation data, which facilitates the determination of the above-mentioned observation matrix.

[0036] In combination with the first aspect or the second aspect, in some possible implementations, the N t The observation data of the group is t The positions in the set of observations are determined based on a predefined first selection rule.

[0037] The first selection rule may be one of one or more predefined selection rules. Each selection rule may be defined based on a certain objective function and a corresponding performance indicator. For example, the objective function may be a distance function, and the performance indicator may be a distance threshold; for another example, the objective function may be a similarity function, and the performance indicator may be a similarity threshold, etc., which will not be further described.

[0038] Determine N based on predefined selection rules t The sending device can select N based on a more reasonable objective function. t A set of observation data, so that the selected N t The group of observation data is used to compensate the predicted data to achieve better results.

[0039] In N t The observation data of the group is t When the position in the set of observation data is determined based on a predefined first selection rule, optionally, in some possible implementations of the first aspect, the method further includes: sending first information, the first information being used to indicate the N t The observation data of the group is t The position in the set of observations.

[0040] Accordingly, in some possible implementations of the second aspect, the method further includes: receiving first information, the first information being used to indicate the N t The observation data of the group is t The position in the set of observations.

[0041] By indicating N to the receiving device t The observation data of the group is in S t The position in the set of observation data can facilitate the receiving device to perform data compensation. For example, in the formula shown above, the N t The observation data of the group is in S t The locations in the set of observations can be used to determine the observation matrix.

[0042] In combination with the first aspect, in some possible implementations of the first aspect, the method further includes: receiving second information, the second information being used to indicate the N t The method of selecting the set of observation data, the N t The selection methods of group observation data include: selection based on predefined patterns, or selection based on predefined selection rules.

[0043] Accordingly, in combination with the second aspect, in some possible implementations of the second aspect, the method further includes: sending second information, the second information being used to indicate the N tThe method of selecting the set of observation data, the N t The selection methods of group observation data include: selection based on predefined patterns, or selection based on predefined selection rules.

[0044] That is, the sending device t Select N from the set of observation data t The selection method in the group observation data can be instructed by the receiving device to the sending device.

[0045] Of course, the sending device from S t Select N from the set of observation data t The selection method in the group observation data can also be predefined by the protocol, which can reduce signaling overhead.

[0046] Optionally, the second information is also used to indicate the first pattern.

[0047] For example, there are multiple predefined patterns, and the receiving device can indicate the first pattern to the sending device. The indication information of the first pattern can be carried in the second information. In this way, the sending device and the receiving device can determine the selected N based on the same pattern. t The number of groups of observation data (that is, N t ) and position.

[0048] If the sending device selects N based on predefined selection rules t Since the receiving device cannot know in advance how many groups of observation data the sending device can select based on a certain selection rule, and the selected observation data in S t The position in the set of observation data, so the sending device can t The observation data of the group is in S t The position in the group observation data is notified to the receiving device through the above-mentioned first information.

[0049] Optionally, the second information is also used to indicate the N t The observation data of the group is t A feedback method for the position in the group observation data includes: feedback through a bitmap or feedback through a combination identifier.

[0050] That is, the receiving device can indicate the feedback mode to the sending device, and the indication information of the feedback mode can also be carried in the second information. In this way, the receiving device can interpret the received first information based on the indicated feedback mode. Of course, the receiving device indicates N based on what feedback mode t The observation data of the group is in S t The location of the group observation data can also be predefined by the protocol.

[0051] One feedback method is to indicate N through a bitmap. t The observation data of the group is in S t The location of the group observation data. The bitmap can include t S t bits, the value of each bit is used to indicate whether the corresponding set of observation data is selected, or to indicate whether the corresponding set of observation data belongs to N t Group observation data.

[0052] Indicate N through a bitmap t The observation data of the group is in S t The location of the group observation data is relatively simple and convenient to implement.

[0053] Another feedback method is to indicate N by combining the flags t The observation data of the group is in S t The position of the group of observation data. This combination identifier can be used to indicate the combination of multiple groups of observation data. In this solution, t The identifier corresponding to the combination of the group observation data is recorded as a first combination identifier, for example, and the first combination identifier may be one of a plurality of combination identifiers, each of which is used to indicate a combination of a plurality of groups of data.

[0054] The first combination identifier is one of a plurality of combination identifiers, and the plurality of combination identifiers can be used with different N t The combination of group observation data corresponds, and each combination identifier can correspond to a combination.

[0055] In the case where the second information indicates that the feedback mode is feedback through a bitmap, the above-mentioned first information can be used to indicate the bitmap; in the case where the second information indicates feedback through a combined identifier, the above-mentioned first information can be used to indicate the N t The first combination identifier corresponding to the group of observation data.

[0056] Indicated by combining the markings N t The observation data of the group is in S t The location of group observation data has less signaling overhead than bitmap.

[0057] In combination with the first aspect or the second aspect, in some possible implementations, the N t The first data set is the N t set of observation data; or, the N t The first data of the group is N t Group residual data, the N t Data in group residual data For the N t Group observation data With Nt Data in group prediction data The difference; among them, Indicates the N t The i-th data in the n-th group of residual data, Indicates the N t The i-th data in the n-th group of observation data, Indicates the N t The i-th data in the n-th group of prediction data, the N t The set of predicted data is S t Part of the data in the group of prediction data, and the nth group of prediction data is in the N t The position of the group of predicted data is the same as the nth group of observed data in the N t The position of the observation data in the group is the same, the S t The group of prediction data includes the prediction data corresponding to the t-th time unit; 1≤i≤I n , 1≤n≤N t , I n represents the number of data included in the nth group of residual data, i, I n and n are both positive integers.

[0058] N obtained by the sending device t The first data of the group is N t The set of observation data is still N t The group residual data may be indicated by the receiving device, or may be predefined by a protocol, or may be determined by the sending device, which is not limited in this application.

[0059] If N t The first data of the group is N t If N sets of observation data are available, the sending device does not need to make predictions, which reduces the amount of calculation required by the sending device and saves power. t The first data of the group is N t If a set of residual data is required, the sending device and the receiving device need to make synchronous predictions, but compared with the observation data, the amount of residual data is smaller, which can further reduce the transmission overhead.

[0060] If the N t The first data of the group is N t The set of observation data is still N t The group of residual data is indicated by the receiving device. Optionally, in some possible implementations of the first aspect, the method further includes: receiving eighth information, the eighth information being used to indicate the N t The first data set is the N t The set of observation data is still the N tGroup residual data.

[0061] Accordingly, in some possible implementations of the second aspect, the method further includes: sending eighth information, the eighth information being used to indicate the N t The first data set is the N t The set of observation data is still the N t Group residual data.

[0062] If the N t The first data of the group is N t The set of observation data is still N t The group of residual data is determined by the sending device. Optionally, in some possible implementations of the first aspect, the method further includes: sending eighth information, the eighth information being used to indicate the N t The first data set is the N t The set of observation data is still the N t Group residual data.

[0063] Accordingly, in some possible implementations of the second aspect, the method further includes: receiving eighth information, the eighth information being used to indicate the N t The first data set is the N t The set of observation data is still the N t Group residual data.

[0064] In combination with the first aspect or the second aspect, in some possible implementations, the S t Any two sets of observation data include the same number of data.

[0065] In other words, the S t Group observation data is obtained by evenly dividing the observation data set. The number of data included in each group of observation data, or the number of groups into which the observation data set is divided, can be indicated by the receiving device or determined by the terminal device. Since the number of data in the observation data set is known in advance by the terminal device, once either the number of groups of observation data or the number of data included in each group of observation data is determined, the other can be inferred.

[0066] Optionally, in some possible implementations of the first aspect, the method further includes: receiving fourth information, where the fourth information is used to indicate one or more of the following: t or the S t The number of data included in each set of observation data. That is, the receiving device instructs the sending device how to divide the set of observation data.

[0067] Accordingly, in some possible implementations of the second aspect, the method further includes: sending fourth information, where the fourth information is used to indicate one or more of the following: t or the S t The number of data included in each set of observation data. That is, the receiving device instructs the sending device how to divide the set of observation data.

[0068] That is, the receiving device instructs the sending device how to divide the observation data set.

[0069] Optionally, in some possible implementations of the first aspect, the method further includes: sending fourth information, where the fourth information is used to indicate one or more of the following: t or the S t The number of data included in each set of observation data. That is, the sending device notifies the receiving device how to divide the set of observation data.

[0070] Accordingly, in some possible implementations of the second aspect, the method further includes: receiving fourth information, where the fourth information is used to indicate one or more of the following: t or the S t The number of data included in each set of observation data. That is, the receiving device instructs the sending device how to divide the set of observation data.

[0071] That is, the sending device instructs the receiving device how to divide the observation data set.

[0072] Through the signaling interaction of the fourth information mentioned above, the sending device and the receiving device can respectively divide the observation data and the prediction data of the t-th time unit based on the same division rule.

[0073] In combination with the first aspect or the second aspect, in some possible implementations, the S t The group observation data is obtained by dividing the observation data set based on at least two pre-configured parameters: t The starting position, ending position or number of data included in each group of observation data in the observation data set, wherein the observation data set is the set of observation data corresponding to the t-th time unit.

[0074] In other words, the S t The group observation data may be obtained by evenly or unevenly dividing the observation data set, which is not limited in this application.

[0075] Optionally, in some possible implementations of the first aspect, the method further includes: receiving third information, the third information being used to indicate the S t At least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0076] Accordingly, in some possible implementations of the second aspect, the method further includes: sending third information, wherein the third information is used to indicate that the S t At least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0077] That is, the receiving device instructs the sending device how to divide the observation data set.

[0078] Optionally, the method further includes: sending third information, wherein the third information is used to indicate the S t At least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0079] That is, the sending device instructs the receiving device how to divide the observation data set.

[0080] Through the signaling interaction of the third information mentioned above, the sending device and the receiving device can respectively divide the observation data and the prediction data of the t-th time unit based on the same division rule.

[0081] Furthermore, one of the sending device and the receiving device may also determine whether to divide the observation data set, or in other words, whether to adopt the compression transmission method provided by this solution.

[0082] In some possible implementations of the first aspect, the method further includes: sending seventh information, where the seventh information is used to indicate whether the observation data set is divided into multiple groups of observation data.

[0083] Accordingly, in some possible implementations of the second aspect, the method further includes: receiving seventh information, where the seventh information is used to indicate whether the set of observation data is divided into multiple groups of observation data.

[0084] That is, the sending device instructs the receiving device whether to divide the observation data set, or whether to perform compressed transmission.

[0085] In some possible implementations of the first aspect, the method further includes: receiving seventh information, where the seventh information is used to indicate whether the set of observation data is divided into multiple groups of observation data.

[0086] Accordingly, in some possible implementations of the second aspect, the method further includes: sending seventh information, where the seventh information is used to indicate whether the observation data set is divided into multiple groups of observation data.

[0087] That is, the receiving device instructs the sending device whether to divide the observation data set, or whether to perform compressed transmission.

[0088] By exchanging the seventh information, the sending device and the receiving device can perform data transmission and reception processing based on the same data processing method. For example, when compressed transmission is used, the sending device can select a portion of the observed data set for compressed transmission, and the receiving device can compensate for a portion of the predicted data based on the received data.

[0089] In combination with the first aspect or the second aspect, in some possible implementations, the tth time unit is one of T time units, where T is an integer greater than 1 and the T time units are periodic, or T is 1.

[0090] Optionally, in some possible implementations of the first aspect, the method further includes: receiving sixth information, where the sixth information is used to indicate the T time units.

[0091] Accordingly, in some possible implementations of the second aspect, the method further includes: sending sixth information, where the sixth information is used to indicate the T time units.

[0092] That is, the receiving device can indicate to the sending device at which time nodes the observation data needs to be fed back.

[0093] In one possible case, T is an integer greater than 1, and the T time units are multiple time units. The T time units may be arranged at equal intervals in the time domain.

[0094] For example, the T time units are periodic, or static. Accordingly, the sixth information may be used to indicate the period length of the T time units. Optionally, the sixth information may also be used to indicate the start time of the T time units.

[0095] For another example, the T time units are semi-persistent (SP), or in other words, semi-static. Accordingly, the sixth information may be used to indicate the period length of the T time units and the effective time of the T time units.

[0096] Another possible situation is that T is 1, and the T time units may be non-periodic, or dynamic.

[0097] Correspondingly, the sixth information is used to indicate the time of the T time units.

[0098] That is, the sending device can observe data at the time nodes where T time units are located, without having to observe the data continuously, thereby saving power consumption caused by data observation.

[0099] In combination with the first aspect or the second aspect, in some possible implementations, the prediction data is obtained through AR prediction.

[0100] AR prediction is only an example, and the prediction data can also be obtained through other methods, such as derivative schemes of AR prediction, such as moving average (MA) prediction, autoregressive integrated moving average (ARIMA) prediction, etc., which will not be described in detail.

[0101] In combination with the first aspect or the second aspect, in some possible implementations, the predicted data is obtained by prediction by a prediction model, and the method also includes: performing model training based on training samples to obtain the prediction model, and the training samples include: multiple signals for training and observation data of the multiple signals, or multiple observation data.

[0102] That is, the model training process can be implemented by the sending device or the receiving device. The sample data used for training can be multiple observation data or multiple signals and their corresponding observation data. This application does not limit this.

[0103] Optionally, in the case of sending a device to perform model training, in some possible implementations of the first aspect, the method further includes: sending model parameters, where the model parameters are used to construct the prediction model.

[0104] Accordingly, in some possible implementations of the second aspect, the method further includes: receiving model data, where the model data is used to construct the prediction model.

[0105] Exemplarily, the prediction model is an AR model, and the model parameters include: a transfer matrix for the AR prediction, an initial value for the AR prediction, a first covariance matrix and a second covariance matrix, wherein the first covariance is obtained based on the error introduced by noise, and the second covariance matrix is ​​determined based on the error of the most recent AR prediction.

[0106] Optionally, in the case where the receiving device performs model training, in some possible implementations of the second aspect, the method further includes: sending model parameters, where the model parameters are used to construct the prediction model.

[0107] Accordingly, in some possible implementations of the first aspect, the method further includes: receiving model data, where the model data is used to construct the prediction model.

[0108] Exemplarily, the prediction model is an AR model, and the model parameters include: a transfer matrix for the AR prediction, a first covariance matrix and a second covariance matrix, the first covariance is obtained based on the error introduced by noise, and the second covariance matrix is ​​determined based on the error of the most recent AR prediction.

[0109] In combination with the first aspect, in some possible implementations of the first aspect, the predicted data is obtained by prediction by a prediction model; the method also includes: sending ninth information, the ninth information is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs the model training.

[0110] Accordingly, in combination with the second aspect, in some possible implementations of the second aspect, the prediction data is obtained by prediction by a prediction model; the method also includes: receiving ninth information, the ninth information is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs the model training.

[0111] That is, the receiving device indicates the parameters for model training to the sending device.

[0112] In combination with the first aspect, in some possible implementations of the first aspect, the predicted data is obtained by prediction by a prediction model; the method also includes: receiving ninth information, the ninth information is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs the model training.

[0113] Accordingly, in combination with the second aspect, in some possible implementations of the second aspect, the predicted data is obtained by prediction by a prediction model; the method also includes: sending ninth information, the ninth information is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs the model training.

[0114] That is, the sending device indicates the parameters for model training to the receiving device.

[0115] The device performing model training may be a receiving device or a sending device, and the ninth information may be used to indicate which device is used for model training.

[0116] One or more of the parameters for model training listed above may be indicated by the ninth information, and the remaining items may be predefined by the protocol, or all of the parameters for model training listed above may be predefined by the protocol, and this application does not limit this.

[0117] In a third aspect, the present application provides a data processing device, comprising modules or units for implementing the method in the first aspect and any possible implementation of the first aspect. Each module or unit can implement the corresponding function by executing a computer program.

[0118] In a fourth aspect, the present application provides a data processing device, comprising a processor, wherein the processor is configured to execute the data processing method described in the first aspect and any possible implementation of the first aspect.

[0119] The apparatus may further include a memory for storing instructions and data. The memory is coupled to the processor, and when the processor executes the instructions stored in the memory, the methods described in the above aspects may be implemented. The apparatus may further include a communication interface for communicating between the apparatus and other devices. Exemplarily, the communication interface may be a transceiver, circuit, bus, module, or other type of communication interface.

[0120] Illustratively, the apparatus in the third aspect or the fourth aspect is a terminal device, or a component in the terminal device, such as a chip, a chip system, a processor, etc.

[0121] In a fifth aspect, the present application provides a chip system comprising at least one processor for supporting the implementation of the functions involved in the above-mentioned first aspect and any possible implementation of the first aspect, for example, receiving or processing the data and / or information involved in the above-mentioned method.

[0122] In one possible design, the chip system further includes a memory, which is used to store program instructions and data, and the memory is located inside or outside the processor.

[0123] The chip system can be composed of chips, or can include chips and other discrete devices.

[0124] In a sixth aspect, the present application provides a data processing device, comprising modules or units for implementing the method in the second aspect and any possible implementation of the second aspect. Each module or unit can implement the corresponding function by executing a computer program.

[0125] In a seventh aspect, the present application provides a data processing device, comprising a processor, wherein the processor is configured to execute the data processing method described in the second aspect and any possible implementation of the second aspect.

[0126] The apparatus may further include a memory for storing instructions and data. The memory is coupled to the processor, and when the processor executes the instructions stored in the memory, the methods described in the above aspects may be implemented. The apparatus may further include a communication interface for communicating between the apparatus and other devices. Exemplarily, the communication interface may be a transceiver, circuit, bus, module, or other type of communication interface.

[0127] Illustratively, the apparatus in the sixth or seventh aspect is a network device, or a component in a network device, such as a chip, a chip system, a processor, etc.

[0128] In an eighth aspect, the present application provides a chip system comprising at least one processor for supporting the implementation of the functions involved in the above-mentioned second aspect and any possible implementation of the second aspect, for example, receiving or processing the data and / or information involved in the above-mentioned method.

[0129] In one possible design, the chip system further includes a memory, which is used to store program instructions and data, and the memory is located inside or outside the processor.

[0130] The chip system can be composed of chips, or can include chips and other discrete devices.

[0131] In a ninth aspect, the present application provides a computer-readable storage medium comprising a computer program, which, when executed on a computer, enables the computer to implement the method in the first or second aspect and any possible implementation of the first or second aspect.

[0132] In the tenth aspect, the present application provides a computer program product, which includes: a computer program (also referred to as code, or instructions), which, when run, enables a computer to execute the method in the first or second aspect and any possible implementation of the first or second aspect.

[0133] In the eleventh aspect, an embodiment of the present application provides a communication system, including a sending device and a receiving device for the aforementioned data.

[0134] The third to eleventh aspects of this application correspond to the technical solutions of the first and second aspects of this application. The beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0135] FIG1 is a schematic diagram of the architecture of a communication system applicable to the method provided in this application;

[0136] FIG2 is a schematic flow chart of the data processing method provided by the present application;

[0137] FIG3 is a schematic diagram of evenly dividing an observation data set according to an embodiment of the present application;

[0138] FIG4 is a schematic diagram of uneven partitioning of an observation data set provided by an embodiment of the present application;

[0139] FIG5 is a schematic diagram of fourth information provided in an embodiment of the present application;

[0140] FIG6 is a schematic diagram of third information provided in an embodiment of the present application;

[0141] FIG. 7 is a diagram of a first pattern provided in an embodiment of the present application and a diagram of a method for selecting N based on the first pattern. t Schematic diagram of group observation data;

[0142] FIG8 is a schematic diagram of selecting observation data based on a first selection rule provided in an embodiment of the present application;

[0143] FIG9 is a schematic diagram of second information provided in an embodiment of the present application;

[0144] FIG10 is a schematic diagram of control signaling provided in an embodiment of the present application;

[0145] Figures 11 and 12 are diagrams of the terminal device obtaining the N t Schematic diagram of the first data set;

[0146] Figure 13 is a schematic diagram of AR prediction;

[0147] Figure 14 is a schematic diagram of KF prediction;

[0148] 15 and 16 are schematic diagrams of network equipment performing data compensation;

[0149] FIG17 is a schematic diagram of T time units provided in an embodiment of the present application;

[0150] FIG18 is a schematic diagram of sixth information provided in an embodiment of the present application;

[0151] Figures 19 to 22 are schematic flow charts of model training provided in embodiments of the present application;

[0152] FIG23 is a schematic diagram of ninth information provided in an embodiment of the present application;

[0153] 24 and 25 are schematic diagrams of a data processing device provided in an embodiment of the present application;

[0154] FIG26 is a schematic diagram of the structure of a terminal device provided in an embodiment of the present application;

[0155] Figure 27 is a structural diagram of the network device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0156] The technical solution in this application will be described below with reference to the accompanying drawings.

[0157] The technical solution in this application will be described below with reference to the accompanying drawings.

[0158] For ease of understanding, the following points are first explained:

[0159] First, since there are many letters involved in this application, the following is a brief explanation of the letters involved.

[0160] N t : The number of groups of the first data, N t is less than S t A positive integer;

[0161] S t : The total number of groups obtained by dividing the observation data set, S t is a positive integer greater than 1;

[0162] L: The number of observations in the data set, or S t The total number of data contained in the group observation data, L is a positive integer greater than 1;

[0163] t: one of T time units, T is a positive integer;

[0164] p: the order of AR prediction, or the order of AR model;

[0165] S corresponding to the t-th time unit t The vector representation of the set of observation data, in the embodiment of the present application, assume that the S t The number of data contained in the group of observation data is L, so The length is L;

[0166] N corresponding to the t-th time unit t The vector representation of the set of observation data, in the embodiment of the present application, for the convenience of calculation, N t The vector representation of the group observation data is designed to be the same as S t The vectors of the group observation data represent vectors of equal length, that is, the length is L;

[0167] N corresponding to the t-th time unit tThe vector representation of the residual data set, in the embodiment of the present application, for the convenience of calculation, N t The vector representation of the group residual data is designed to be the same as S t The vectors of the group observation data represent vectors of equal length, that is, the length is L;

[0168] The vector representation of the predicted data corresponding to the t-th time unit. In the embodiment of the present application, for the convenience of calculation, the S t The number of data contained in the group prediction data is related to S t The number of observations in each group is the same, which is L, so The length is L;

[0169] O(t): the observation matrix corresponding to the t-th time unit;

[0170] K(t): Kalman gain matrix corresponding to the t-th time unit.

[0171] Second, in an embodiment of the present application, for different types of observation data, the definition of the quantity (or number) of observation data may be different. For example, for type II codebook feedback in channel state information (CSI), the number of observation data may refer to the number of at least one weighting coefficient corresponding to at least one spatial domain vector and at least one frequency domain vector obtained by dual-domain compression. For another example, for perception data, the number of observation data may refer to the number of data points in the perceived point cloud. The number of observation data can be determined based on the business or application scenario and will not be listed here. The definition of the number of predicted data is similar and will not be repeated here.

[0172] Third, to facilitate a clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. For example, the first information and the second information are merely used to distinguish different information and do not limit their order or the number of signaling messages. Those skilled in the art will understand that the words "first" and "second" do not limit the number or execution order, and the words "first" and "second" do not necessarily mean that they are different.

[0173] Fourth, the "sending" and "receiving" in the embodiments of the present application indicate the direction of signal transmission. For example, "sending information to a network device" can be understood as the destination end of the information being the network device, which can include direct sending through the air interface, and also includes indirect sending through the air interface by other units or modules. "Receiving information from a network device" can be understood as the source end of the information being the network device, which can include direct receiving from the network device through the air interface, and also includes indirect receiving from the network device through the air interface from other units or modules. "Sending" can also be understood as the "output" of the chip interface, and "receiving" can also be understood as the "input" of the chip interface.

[0174] Fifth, "at least one" means one or more, and "more" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b and c can mean: a, or b, or c, or a and b, or a and c, or b and c, or a, b and c, where a, b, c can be single or multiple.

[0175] Sixth, "predefinition" or "preconfiguration" can be achieved by pre-saving corresponding codes, tables or other methods that can be used to indicate relevant information in a device (for example, including a terminal device and a network device). This application does not limit its specific implementation method. Among them, "saving" can mean saving in one or more memories. The one or more memories can be set separately or integrated in an encoder or decoder, a processor, or a communication device. The one or more memories can also be partially set separately and partially integrated in a decoder, a processor, or a communication device. The type of memory can be any form of storage medium, which is not limited by this application.

[0176] Seventh, in the embodiments of the present application, descriptions such as "when...", "in the case of...", "if" and "if" all mean that under certain objective circumstances, the device (such as the terminal device or network device described below) will make corresponding processing, which is not a time limit, and does not require the device (such as the terminal device or network device described below) to have a judgment action when implementing it, nor does it mean that there are other limitations.

[0177] The technical solution provided in this application can be applied to various communication systems, such as: long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD) system, universal mobile telecommunication system (UMTS), world-wide interoperability for microwave access (WiMAX) communication system, fifth generation (5G) communication system, and future sixth generation (6G) communication system. Of course, the technical solution provided in this application can also be applied to other possible communication systems, for example, to the Internet of Things (IoT) network, a wireless local area network system that supports the 802.11 series of protocols, and can also be applied to a wireless personal area network system based on ultra wide band (UWB), and can also be applied to a sensing system, and can also be applied to vehicle to everything (V2X), machine type communication (MTC), machine-to-machine information interaction (long term evolution-machine, LTE-M), machine to machine (M2M) communication, vehicle to vehicle (V2V) communication, vehicle network communication (long term evolution-vehicle, LTE-V), satellite communication system, etc. The above-mentioned communication system applicable to this application is only an example, and the communication system applicable to this application is not limited to this. They are uniformly explained here and will not be repeated below.

[0178] The terminal device may also be referred to as user equipment (UE), access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile equipment (ME), user terminal, terminal, wireless communication device, user agent or user device.

[0179] A terminal device is a device with wireless transceiver capabilities. It communicates with one or more core network (CN) devices (also called core devices) via access network equipment (or access devices) within the radio access network (RAN). Terminal devices can be deployed on land, indoors or outdoors, handheld or in vehicles; on water (such as ships); or in the air (such as aircraft, balloons, and satellites). In the embodiments of the present application, a terminal device may also be referred to as user equipment (UE), and may be a mobile phone, a mobile station (MS), a tablet computer, a computer with wireless transceiver capabilities, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical care, a wireless terminal device in smart grids, a wireless terminal device in transportation safety, a wireless terminal device in smart cities, a wireless terminal device in smart homes, a subscriber unit, a cellular phone, a wireless data card, a personal digital assistant (PDA), a tablet computer, a laptop computer, an MTC terminal device, a drone, etc. The terminal device may include various handheld devices with wireless communication capabilities, vehicle-mounted devices, wearable devices, computing devices, or other processing devices connected to a wireless modem. Optionally, the terminal device can be a handheld device (handset) with wireless communication function, a terminal device in the Internet of Things or the Internet of Vehicles, a terminal device of any form in 5G and communication systems evolved after 5G, etc., and this application does not limit this.

[0180] Furthermore, terminal devices can also be end devices in an IoT system, also known as IoT nodes. IoT is a crucial component of future information technology development. Its primary technical feature is connecting objects to the network through communication technologies, thereby enabling intelligent networks that interconnect humans and machines, and objects and things. Connections can be achieved through broadband or narrowband (NB) technology. IoT technology, for example, utilizes narrowband technology to achieve massive connections, deep coverage, and power-saving terminals.

[0181] In addition, terminal devices can also include sensors such as smart printers, train detectors, and gas stations. Their main functions include collecting data, receiving control information and downlink data from network devices, and sending electromagnetic waves to transmit uplink data to network devices.

[0182] In the embodiments of the present application, the device for realizing the function of the terminal device can be a terminal device, or a device capable of supporting the terminal device to realize the function, such as a chip system, which can be installed in the terminal device or used in combination with the terminal device. In the embodiments of the present application, the chip system can be composed of a chip, or it can include a chip and other discrete devices. In the embodiments of the present application, only the terminal device is used as an example for description, and the embodiments of the present application are not limited to the solutions of the embodiments of the present application.

[0183] An access network device can be any device with wireless transceiver capabilities and capable of communicating with a terminal device, such as a RAN node that connects a terminal device to a wireless network. Currently, some examples of RAN nodes include: NodeB, evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, auxiliary station, multi-standard radio (motor slide retainer, MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. The base station can be a macro base station, a micro base station, a relay node, a donor node or the like, or a combination thereof, or a wireless controller in a cloud radio access network (CRAN) scenario, a node in an open radio access network (O-RAN or ORAN) scenario, etc. The base station can also refer to a communication module, modem or chip used to be set in the aforementioned device or apparatus. The base station can also be a mobile switching center and a device that performs the base station function in D2D, V2X, and M2M communications, a network side device in a 6G network, a device that performs the base station function in future communication systems, etc. The base station can support networks with the same or different access technologies. Optionally, the RAN node can also be a server, a wearable device, a vehicle or an on-board device, etc. For example, the access network device in the vehicle to everything (V2X) technology can be a road side unit (RSU). The embodiments of the present application do not limit the specific technology and specific device form adopted by the network equipment.In some deployments, the network devices mentioned in the embodiments of the present application may include a CU, a DU, or both a CU and a DU, or a control plane CU node (central unit-control plane (CU-CP)), a user plane CU node (central unit-user plane (CU-UP)), and a DU node. For example, the network devices may include a gNB-CU-CP, a gNB-CU-UP, and a gNB-DU.

[0184] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes implementing portions of the base station's functionality. For example, a RAN node can be a CU, DU, CU-CP, CU-UP, or RU. The CU and DU can be separate or included in the same network element, such as the BBU. The RU can be included in a radio frequency device or radio unit, such as an RRU, AAU, or RRH.

[0185] In one possible design, the processing unit for implementing baseband functions in the BBU is called a baseband high layer (BBH) unit, and the processing unit for implementing baseband functions in the RRU / AAU / RRH is called a baseband low layer (BBL) unit.

[0186] In different systems, CU (or CU-CP and CU-UP), DU or RU may also have different names, but those skilled in the art can understand their meanings. For example, in the ORAN system, CU may also be called O-CU (Open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, CU-UP may also be called O-CU-UP, and RU may also be called O-RU. Any unit of CU (or CU-CP, CU-UP), DU and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0187] In the embodiments of the present application, the device for implementing the functions of the network device can be a network device; it can also be a device that can support the network device to implement the functions, such as a chip system, a hardware circuit, a software module, or a hardware circuit and a software module. The device can be installed in the network device or used in conjunction with the network device. In the embodiments of the present application, only the device for implementing the functions of the network device is used as an example to illustrate, and does not constitute a limitation on the solutions of the embodiments of the present application.

[0188] The network device and / or terminal device can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; it can also be deployed on the water surface; it can also be deployed on aircraft, balloons and satellites in the air. The embodiments of this application do not limit the scenarios in which the network device and the terminal device are located. In addition, the terminal device and the network device can be hardware devices, or they can be software functions running on dedicated hardware, software functions running on general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities including dedicated or general-purpose hardware devices and software functions. This application does not limit the specific forms of the terminal device and the network device.

[0189] To support artificial intelligence (AI) technology in wireless networks, AI nodes can be introduced into the network. The AI ​​nodes can be AI network elements or AI modules.

[0190] Optionally, the AI ​​node can be deployed in one or more of the following locations in the communication system: access network equipment, terminal equipment, or core network equipment. Alternatively, the AI ​​node can be deployed separately, for example, in a location other than any of the above devices, such as a host or cloud server in an over-the-top (OTT) system. The AI ​​node can communicate with other devices in the communication system, such as one or more of the following: network equipment, terminal equipment, or core network elements.

[0191] This application does not limit the number of AI nodes. For example, when there are multiple AI nodes, the multiple AI nodes can be divided based on function, such as different AI nodes are responsible for different functions.

[0192] AI nodes can be independent devices, or they can be integrated into the same device to implement different functions. They can also be network elements in hardware devices, software functions running on dedicated hardware, or virtualized functions instantiated on a platform (for example, a cloud platform). This application does not limit the specific form of the above-mentioned AI nodes.

[0193] Figure 1 is a schematic diagram of the architecture of a communication system 100 applicable to the method provided in this application. As shown in Figure 1, the communication system 100 includes a wireless access network 10 and a core network 20. Optionally, the communication system 100 also includes the Internet 30. The wireless access network 10 may include at least one wireless access network device (such as 11a and 11b in Figure 1) and may also include at least one terminal device (such as 12a-12j in Figure 1). The terminal device can be connected to the wireless access network device in a wireless manner. Terminal devices and terminal devices, as well as wireless access network devices and wireless access network devices, can be connected to each other in a wired or wireless manner. Figure 1 is only a schematic diagram. The communication system 100 may also include other network devices, such as wireless relay devices and wireless backhaul devices, which are not shown in Figure 1.

[0194] The wireless access network device can be a base station deployed in the air, such as a satellite base station 11a, or a base station deployed indoors, such as a micro base station or indoor station 11b. It should be understood that this application does not limit the specific technology or device form used by the wireless access network device. For ease of description, the following description uses a base station as an example of a wireless access network device.

[0195] The terminal device can be a terminal device deployed in the air, such as the helicopter or drone 12i in Figure 1; it can also be a terminal device deployed on the ground, such as the mobile phones 12a, 12e, 12f and 12j, vehicle 12b, computer 11b, printer 12h, etc. in Figure 1.

[0196] Optionally, the terminal device can also be used to act as a base station. For example, the UE can act as a scheduling entity that provides sidelink signals between terminal devices in vehicle-to-everything (V2X), device-to-device (D2D), or peer-to-peer (P2P) scenarios.

[0197] Base stations and terminal devices can be fixed or mobile. They can be deployed on land, indoors or outdoors, handheld or vehicle-mounted; on water; or in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the application scenarios of base stations and terminal devices.

[0198] The roles of base stations and terminal devices can be relative. For example, the helicopter or drone 12i in Figure 1 can be configured as a mobile base station. To terminal devices 12j that access the wireless access network 10 via 12i, terminal device 12i is a base station. However, to base station 11a, 12i is a terminal device, meaning that communication between 11a and 12i occurs via a wireless air interface protocol. Of course, communication between 11a and 12i can also occur via a base station-to-base station interface protocol. In this case, 12i is also a base station relative to 11a. Therefore, base stations and terminal devices can be collectively referred to as communication devices. 11a, 11b, and 12a-12j in Figure 1 can be referred to as communication devices having their respective corresponding functions, such as communication devices having base station functions or communication devices having terminal functions.

[0199] In the embodiments of the present application, the functions of the base station may also be performed by a module (such as a chip) in the base station, or by a control subsystem that includes the base station function. The control subsystem that includes the base station function here may be a control center in the application scenarios of the above-mentioned terminals such as smart grid, industrial control, intelligent transportation, and smart city. The functions of the terminal device may also be performed by a module (such as a chip) in the terminal device, or by a device that includes the terminal device function. This application does not limit this.

[0200] As the application scenarios of wireless communication technologies become increasingly diverse, data transmission for new scenarios brings more and more transmission overhead.

[0201] To reduce transmission overhead, some proposals have been proposed to use historical data to predict future data, thereby reducing data transmission and saving transmission overhead. For example, time series forecasting, such as AR forecasting, can be used to predict future data by leveraging the temporal correlation of data. However, due to the accumulation of errors during the forecasting process, the accuracy of the forecasted data may decrease over time. Therefore, observational data can be used to compensate for the forecasted data. However, using observational data for compensation often requires the feedback of complete observational data, which incurs significant transmission overhead.

[0202] In view of this, the present application provides a method in which a data sending device selects a portion of data from a set of observation data to send, so that the data receiving device can compensate for the predicted data obtained by its own prediction based on the received portion of data, thereby reconstructing the complete data. Compared to sending the entire set of observation data, this method compresses the set for transmission, which can avoid the large transmission overhead caused by transmitting all the data in the set; compared to not reporting the observation data, this method can use the selected portion of data to compensate for the predicted data, which is conducive to obtaining more accurate predicted data. Therefore, an effective compromise is achieved between transmission overhead and data accuracy.

[0203] The inventors of the present application have found through simulation that if the observation data is selectively fed back, the performance improvement brought about is relatively obvious. For example, if the channel conditions are fixed: the signal-to-noise ratio (SNR) is 0 decibel (dB), the number of base station antennas is 4, and the number of subbands is 8. Then, under the same distortion rate, the compressed transmission scheme provided by the present application (that is, a scheme in which part of the data is selected from the observation data set and feedback is based on the part of the data) is adopted. Compared with the scheme in which the entire observation data set is fed back, the CSI compression rate can be increased by 1 / 3. For another example, when the average compression rate is similar, the compressed transmission scheme proposed in the present application and the type II codebook compression scheme adopted in the current standard are respectively used for CSI feedback, and the cosine similarity with the measured CSI is improved by 8%.

[0204] The technical solution in this application will be described below with reference to the accompanying drawings.

[0205] Figure 2 is a schematic flow chart of the data processing method provided by the present application. Figure 2 describes the data processing method provided by the present application by taking the interaction between a terminal device and a network device as an example. As an example, the terminal device can be a data sending device, and the network device can be a data receiving device. The method provided by the present application is not limited to the data sending device and the data receiving device. For example, the data sending device can be a network device, and the data receiving device can be a terminal device; or, both the data sending device and the data receiving device can be terminal devices, and so on.

[0206] In addition, Figure 2 only describes the method by taking the interaction between the terminal device and the network device as an example, and does not limit the execution subject of the method. In this application, unless otherwise specified, the terminal device can refer to the terminal device itself or a component in the terminal device, such as a chip, a chip system, a processor, etc. in the terminal device, or a logic module or software that can realize all or part of the functions of the terminal device; the network device can refer to the network device itself or a component in the network device, such as a chip, a chip system, a processor, etc. in the network device, or a logic module or software that can realize all or part of the functions of the network device. This application does not limit this.

[0207] The method 200 shown in FIG. 2 may include steps 210 to 290 . Each step in the method 200 is described in detail below.

[0208] In step 210, the terminal device divides the observation data set of the t-th time unit into S t Group observation data.

[0209] In this application, for the convenience of distinction and explanation, the set of observation data corresponding to the t-th time unit is called the observation data set, which may include multiple observation data. By dividing the observation data set of the t-th time unit, the S of the t-th time unit can be obtained. t In other words, the S t The group of observation data is obtained by dividing multiple data observed in the t-th time unit, and each group of observation data may include one or more observation data.

[0210] Exemplarily, the observation data in the observation data set may be perception data or imaging data, such as environment reconstruction data, point cloud data, and two-dimensional (2D) or three-dimensional (3D) imaging data; or, the observation data in the observation data set may be distributed AI data or edge AI data, such as AI model data, gradient update data, feature information extracted by a neural network, or, the observation data in the observation data set may be observation data of a channel, such as a channel matrix, channel information, or channel state information (CSI). This application includes but is not limited to this.

[0211] For the convenience of explanation, this paper assumes that the number of observation data included in the observation data set of the t-th time unit is L, L is an integer greater than 1, and the L observation data are represented by a vector, which can be obtained as a vector of length L Divide the L observation data into St Group observation data, S t It is also an integer greater than 1.

[0212] One possible design is that the terminal device divides the observation data set of the t-th time unit evenly into S t group of observation data, then the S t Any two sets of observation data include the same number of data.

[0213] For example, in this embodiment, the set of observation data corresponding to the t-th time unit includes L observation data. If L is S t The number of observation data included in each group of observation data obtained by uniform division is L / S t ; If L is not S t If it is an integer multiple of , S can be obtained by padding (such as zero padding) t The number of data included in each set of observation data is obtained by dividing the filled observation data set evenly. The obtained S t Any two sets of observation data include the same number of data.

[0214] For example, FIG3 is a schematic diagram of evenly dividing the observation data set provided by an embodiment of the present application. FIG3 assumes that S t is 4, that is, the number of data contained in each set of observation data L / S t . The vector of length L of the t-th time unit Evenly divided into 4 vectors of equal length, which can be expressed as and The length of each vector is L / S t . Thus, 4 groups of observation data can be obtained after even division.

[0215] Or, in another implementation, where L is not S t In the case of an integer multiple of As the former (S t -1) The number of data included in each group of observation data, the remaining The data is taken as the S t A set of observation data is obtained before (S t -1) The number of data included in any two groups of observation data is equal, and the number of data included in each group of observation data is No. S t The number of data included in the group observation data is Another possible design is that the terminal device divides the observation data set of the t-th time unit non-uniformly into S t Group observation data.

[0216] Divide the observation data set of the t-th time unit into S t group of observation data, the S t The number of observation data contained in each group of observation data can be equal or unequal. t Each set of observation data in the set of observation data has at least two parameters as follows, which divides the observation data set of the t-th time unit into S t Group Observation Data: The starting position, ending position, or number of data included in each group of observation data within the observation data set. Simply put, by predetermining at least two of the starting position, ending position, or length of each group of observation data, the position of each group of observation data within the observation data set can be determined.

[0217] For example, the L data in the observation data set are numbered sequentially from the first data to the last data, for example, the first data is numbered 0 and the last data is numbered L-1. The starting position of each set of observation data can be identified by the number of the first data in the set of L data, and the ending position of each set of observation data can be identified by the number of the last data in the set of L data. The number of data contained in each set of observation data can be obtained by subtracting the numbers of the last data and the first data in the set of L data, respectively, and then adding 1. The numbers defined for the L data above are only examples, and this application does not limit the specific size of the numbers.

[0218] FIG4 is a schematic diagram of an embodiment of the present application providing an uneven partitioning of an observation data set. FIG4 assumes that S t is 4, the terminal device is based on S t At least two parameters of each set of observation data in the set of observation data: the starting position, the ending position or the number of data included in the observation data set, the vector of length L of the t-th time unit Divided into 4 groups of observation data, represented as and The four groups of observation data may or may not overlap, and the number of data included in the four groups of observation data may or may not be equal, which is not limited in this application.

[0219] Whether the terminal device divides the observation data set uniformly or non-uniformly may be predefined by the protocol, indicated by the network device, or determined by the terminal device, and this application does not limit this.

[0220] If the network device instructs the terminal device to divide the observation data set evenly, the method further includes: the network device sends fourth information to the terminal device, or the terminal device receives fourth information from the network device, the fourth information is used to instruct the observation data set of the t-th time unit to be divided into S t When the number of observation data sets is fixed, by indicating S t or the number of data included in each set of observations, the other can be inferred. t or S t At least one of the number of data included in each group of observation data in the group of observation data can be used to determine how to divide the observation data set.

[0221] Exemplarily, the fourth information specifically indicates one or more of the following: t or S t The number of data included in each set of observation data in the set of observation data. The terminal device can evenly divide the observation data set according to the parameter indicated by the fourth information.

[0222] Of course, the network device may not indicate the number of groups S t or S t The number of data included in each group of observation data in the group observation data, and how to evenly divide the observation data set can be determined by the terminal device itself. In this case, the terminal device can divide the number of groups S t or S t The network device is informed of at least one item of the quantity of data included in each set of observation data.

[0223] If the terminal device itself determines to divide the observation data set evenly, the method further includes: the terminal device sends fourth information to the network device, or the network device receives the fourth information from the terminal device, the fourth information is used to indicate that the observation data set of the t-th time unit is divided into S t The network device can determine S according to the fourth information. t The number and location of data included in each group of observations.

[0224] In the case where the terminal device supports multiple partitioning modes such as uniform partitioning and non-uniform partitioning of the observation data set, the third information may further indicate the partitioning mode. Optionally, the fourth information is also used to indicate the partitioning mode, which is, for example, uniform partitioning.

[0225] Figure 5 is a schematic diagram of the fourth information provided by an embodiment of the present application. Figure 5 (a), (b) and (c) respectively show three possible forms of the fourth information.

[0226] In FIG5(a), the fourth information includes two information elements, which are used to indicate the following two items: a division method and a number of division groups, wherein the division method can be a uniform division, and the number of division groups can be the number of groups S for dividing the observation data set. t .

[0227] In FIG5(b), the fourth information includes two information elements, which are used to indicate the following two items: the division method and the number of data included in each group of observation data, wherein the division method can be uniform division and the number of data included in each group of observation data is S. t The number of data included in each set of observation data. For example, if the number of data included in any two sets of observation data is the same, this field may indicate a value, that is, the number of data included in each set of observation data; or, if the number of data included in the two sets of observation data is different, this field may indicate a value, such as the value of the preceding (S t -1) The number of data included in each set of observation data, which can also indicate two values, such as the first (S t -1) The number of data included in each set of observation data, and the number of t The number of data included in the set of observation data, or it can also indicate S t The values ​​are S t The number of data points included in each set of observations.

[0228] In FIG5(c), the fourth information includes three elements, each of which indicates the following: the division method, the number of division groups, and the amount of data included in each group of observation data. The contents indicated by each element can be found in the previous description of FIG5(a) and (b), and will not be repeated here.

[0229] If the network device instructs the terminal device to perform non-uniform division on the observation data set, the method further includes: the network device sends third information to the terminal device, or the terminal device receives the third information from the network device, the third information is used to instruct S t At least two of the following parameters of each set of observation data in the set of observation data: the starting position, the ending position or the number of data included in the observation data set. The terminal device may divide the observation data set according to the parameters indicated by the third information.

[0230] If the terminal device determines to divide the observation data set unevenly, the method further includes: the terminal device sends a third information to the network device, or the network device receives the third information from the terminal device, the third information is used to indicate S tAt least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0231] The network device can determine S according to the third information. t The number and location of data contained in each set of observations.

[0232] In the case where the terminal device supports multiple partitioning modes such as uniform partitioning and non-uniform partitioning of the observation data set, the third information may further indicate the partitioning mode. Optionally, the third information is also used to indicate the partitioning mode, which is, for example, non-uniform partitioning.

[0233] Figure 6 is a schematic diagram of the third information provided by an embodiment of the present application. Figure 6 (a), (b), (c) and (d) respectively show four possible forms of the third information.

[0234] In FIG6 (a), the third information includes three information elements, which are used to indicate the following three items: the division mode, S t The starting position of each set of observation data and S t The end position of each set of observation data. The division method can be non-uniform division; the starting position and the end position of each set of observation data can be the numbers of the first data and the last data in each set of observation data in the observation data set; S t The starting position of each set of observation data can be, S t The number of the first observation data in each group of observation data is arranged in a certain order to form an array, for example, an array arranged in order from the first group to the last group; S t The end position of each set of observation data can be, S t The array is formed by arranging the numbers of the last observation data in each group of observation data in a certain order, for example, the array is formed by arranging them in order from the first group to the last group.

[0235] In FIG6(b), the third information includes three information elements, which are used to indicate the following three items: the division mode, S t The starting position of each set of observation data and S t The number of data included in each group of observation data, where the division method and S t The starting position of each group of observation data can be found in the above description in conjunction with FIG6 (a), which will not be repeated here. tThe starting position of each set of observation data is an array composed of the first data in each set of observation data and the number of the first data in the observation data set arranged in a certain order. t The number of data included in each set of observation data can be, S t The number of data included in each group of observation data is arranged in a certain order to form an array, for example, the array is arranged in order from the first group to the last group.

[0236] In FIG6(c), the third information includes three information elements, which are used to indicate the following three items: division mode, S t The number of data included in each set of observation data and S t The end position of each group of observation data. The content indicated by each information element can be found in the above description in conjunction with Figure 6 (a) and (b), and will not be repeated here.

[0237] In FIG6(d), the third information includes four cells, which respectively indicate the following four items: division mode, S t The starting position of each set of observation data, S t The number of data included in each set of observation data and S t The end position of each group of observation data. The content indicated by each information element can be found in the above description in conjunction with Figure 6 (a) and (b), and will not be repeated here.

[0238] Whether the terminal device divides the observation data set can be predefined by the protocol. For example, the protocol can predefine the use of the compression transmission scheme provided by this scheme for data feedback. In this case, the network device and the terminal device can determine the need to divide the observation data set based on this.

[0239] Alternatively, whether the terminal device divides the observation data set may be instructed by the network device or determined by the terminal device itself.

[0240] Optionally, before step 210, the method further includes: the network device sends seventh information to the terminal device, or the terminal device receives seventh information from the network device, where the seventh information is used to indicate whether to divide the observation data set of the tth time unit into multiple groups of observation data.

[0241] In this implementation, the terminal device can determine whether to divide the observation data set of the t-th time unit into multiple groups of observation data based on the seventh information from the network device. For example, the network device can determine whether it is necessary to adopt a compressed transmission scheme for data feedback based on the current bandwidth usage. For example, when the current bandwidth usage is large, it can be determined to adopt a compressed transmission scheme for data feedback; when the current bandwidth usage is small, it can be determined to feedback complete data instead of adopting a compressed transmission scheme for data feedback. This application does not limit the factors considered by the network device to determine whether to divide the observation data set.

[0242] Optionally, before step 210, the method further includes: the terminal device sending seventh information to the network device, or the network device receiving seventh information from the terminal device, the seventh information indicating whether the observation data set is divided into multiple groups of observation data.

[0243] In this implementation, the terminal device independently determines whether to divide the observation data set and notifies the network device. That is, the terminal device independently determines whether to use a compressed transmission scheme for data feedback. For example, the terminal device can determine based on the amount of data in the observation data set. For example, when the amount of data is large, it determines to use a compressed transmission scheme for data feedback. When the amount of data is small, it determines that the complete data can be fed back without using a compressed transmission scheme for data feedback. This application does not limit the factors considered by the terminal device in determining whether to divide the observation data set.

[0244] In one example, a network device may determine whether to divide an observation data set and how to divide the observation data set. In this case, any one of the third information or the fourth information may be carried in the same control signaling as the seventh information, and the network device may send the control signaling to the terminal device so that the terminal device divides the observation data set based on the control signaling. The control information may include, but is not limited to, any of the information elements exemplified in (a), (b), and (c) in FIG. 5 , and any one of (a), (b), (c), and (d) in FIG. 6 , as well as information elements for indicating the division of the observation data set. For the sake of brevity, the accompanying drawings are not provided.

[0245] In step 220, the terminal device obtains N t The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N in the set of observation data t Determined by a set of observation data.

[0246] The S t The number of observations in a group is greater than Nt The number of observations in a set of observations. When each set of observations contains only one observation, the number of observations can be the number of groups of observations. In this case, it can also be called S t The number of groups of observation data is greater than N t The number of groups of observation data; when each group of observation data includes one or more observation data, S t The number of data included in the group of observations is greater than N t The number of observations included in the set, or S t The amount of data included in the group of observations is greater than N t The amount of data included in the set of observations.

[0247] The N t The set of observation data is just S t A subset of the set of observation data, or a part of the above observation data set. t N is determined by a set of observation data t The first data of the group can also be understood as a representation of part of the data in the observation data set, rather than a representation of all the data.

[0248] Optionally, step 220 specifically includes:

[0249] Step 2201: The terminal device receives data from S t Select N from the set of observation data t set of observational data; and

[0250] Step 2202: The terminal device uses the N t The set of observation data determines N t Group 1 data.

[0251] The following is a detailed description of steps 2201 and 2202.

[0252] In step 2201, the N t The observation data of the group is t The position in the set of observation data is determined based on a predefined first pattern; or the N t The observation data of the group is t The positions in the set of observations are determined based on a predefined first selection rule.

[0253] The following is a detailed explanation based on the drawings and selection rules.

[0254] One possible scenario is that N t The observation data of the group is t The position in the set of observation data is determined based on the predefined first pattern. After the terminal device determines the first pattern, it can tThe observation data of the group is in S t The distribution of the positions in the group observation data, from S t Select N from the set of observation data t Group observation data.

[0255] The first pattern is one of at least one predefined or preconfigured pattern, and each pattern in the at least one pattern is used to indicate: N t The observation data of the group is in S t A distribution of the positions in a set of observation data. And, as t changes, N t and S t It can be fixed or change with the change of t, N t The observation data of the group is in S t The position in the group observation data may also change with the change of t, which is not limited in this application.

[0256] In one example, the N indicated by the first pattern t The observation data of the group is t The position in the set of observation data is expressed as N t is the polling granularity, as t increases in the S t That is, as t increases, by selecting N t group of observation data, S t Group observation data traversal. Among them, N t and S t Can be fixed.

[0257] For example, FIG7 is a diagram of a first pattern provided in an embodiment of the present application and a diagram of a first pattern selected based on N. t Schematic diagram of a group of observation data. As shown in the figure, assuming S t is 6, N t =2, that is, the observation data set is divided into 6 groups of observation data, and 2 groups of observation data are selected from the 6 groups of observation data. As can be seen from the first pattern shown in Figure 7 (a), in the first time unit, the positions of the 2 groups of observation data are S t The first and second groups of observation data in the group of observation data; in the second time unit, the positions of the two groups of observation data are S t The third and fourth groups of observation data in the group of observation data; in the third time unit, the positions of the two groups of observation data selected by polling are distributed in S t The 5th and 6th groups of observation data are included in the group of observation data.

[0258] Based on the first pattern, we can obtain the observation data selected by the terminal device at the three time units t1, t2, and t3 as shown in Figure 7 (b). As shown in the figure, the observation data set corresponding to the t1th time unit is recorded as The 6 groups of observation data obtained by uniform division are: to According to the first figure, the two groups of observation data selected in the t1th time unit are the first and second groups of the six groups of observation data. Therefore, the two groups of observation data selected in the t1th time unit are and The observation data set corresponding to the t2th time unit is recorded as The 6 groups of observation data obtained by uniform division are: to According to the first figure, the two groups of observation data selected in the t2th time unit are the third and fourth groups of the six groups of observation data. Therefore, the two groups of observation data selected in the t2th time unit are and The observation data set corresponding to the t2th time unit is recorded as The 6 groups of observation data obtained by uniform division are: to According to the first figure, the two groups of observation data selected at the t3th time unit are the 5th and 6th groups of observation data among the 6 groups of observation data. Therefore, the two groups of observation data selected at the t3th time unit are and

[0259] In the first pattern shown in (a) of Figure 7, the polling granularity is 2, and the position of the selected observation data in the observation data set changes with each change of t. Figure 7 (a) is only an example and should not constitute any limitation to this application. Those skilled in the art can also make simple changes based on the first pattern shown in (a) of Figure 7. For example, a repetition factor R (R is a positive integer) can be introduced into the first pattern, and the repetition factor can be used to describe how many consecutive time units the position of the observation data selected in the observation data set remains unchanged. For example, assuming that R is 1, the observation data selected in each time unit as shown in (b) of Figure 7 can be obtained; for another example, assuming that R is 2, it means that the position of the observation data selected in every 2 consecutive time units in the observation data set remains unchanged. For the sake of brevity, the accompanying drawings are not illustrated. For different values ​​of R, N observations in different time units can be performed based on the same pattern. t Selection of observation data.

[0260] In addition, polling is only one possible design, and this application does not limit the specific form of the pattern.

[0261] Another possible situation is that N t The observation data of the group is t The position in the set of observation data is determined based on a predefined first selection rule. After determining the first selection rule, the terminal device selects the position from S t Select N from the set of observation data t The first selection rule may be one of at least one predefined selection rule.

[0262] Optionally, the first selection rule includes an objective function. In one example, the objective function is a distance function, which can be used to calculate the distance between the observed data and the target value, which can be specifically represented by the cosine similarity, Euclidean distance, etc. The first selection rule can be, for example, the N distance rankings that are ranked high based on the distance function. t A set of observation data. The target value may be the predicted data corresponding to the same time unit and the same location as the observation data used to calculate the distance. For example, the terminal device may obtain the predicted data for the same time unit through a prediction algorithm and then calculate the distance between each set of observation data and the predicted data at the corresponding location.

[0263] Optionally, the first selection rule includes a performance indicator, which may be, for example, a threshold value of a performance parameter having a significant impact on performance obtained through AI training.

[0264] In one example, the performance indicator includes a proportional fairness coefficient. The proportional fairness coefficient R i satisfy:

[0265] The performance parameters may vary for different application scenarios and application requirements. For example, the performance parameters may include, but are not limited to, cosine similarity in CSI compression, normalized mean squared error (NMSE) in image reconstruction, reconstruction accuracy in point cloud compression, and the like.

[0266] The content included in the at least one selection rule exemplified above is only an example and should not constitute any limitation to this application. This application does not limit the specific content of the at least one selection rule.

[0267] For example, FIG8 is a schematic diagram of selecting observation data based on the first selection rule provided by the embodiment of the present application. Taking the first selection rule including the distance function as an example, by calculating the cosine similarity between the observation data and the predicted data, the top N t Set prediction data as the selected N t Figure 8 Assume that St is 6, N t =2, that is, 2 groups of observation data are selected from 6 groups of observation data. Assume that the cosine similarity between the i-th group of observation data and the i-th group of prediction data corresponding to the t-th time unit is expressed as d i (t), then by calculating the cosine similarity between each set of observation data in the 6 sets of observation data and its corresponding set of prediction data, the arrangement can be obtained by sorting the 6 sets of observation data in each time unit.

[0268] As shown in the figure, when t is t1, that is, in the t1th time unit, the cosine similarity of the 6 groups of observation data and the corresponding predicted data is ranked as follows: d1(t1)>d5(t1)>d3(t1)>…>d6(t1), so the first group of observation data and the fifth group of observation data can be selected from the 6 groups of observation data, as shown in the figure. and When t is t2, that is, in the t2th time unit, the cosine similarity of the 6 groups of observation data and the corresponding predicted data is ranked as follows: d5(t2)>d3(t2)>d6(t2)>…>d2(t2), so the 3rd and 5th groups of observation data can be selected from the 6 groups of observation data, as shown in the figure. and When t is t3, that is, in the t3th time unit, the cosine similarity of the 6 groups of observation data and the corresponding predicted data is ranked as follows: d1(t3)>d2(t3)>d3(t3)>…>d4(t3), so the first group of observation data and the second group of observation data can be selected from the 6 groups of observation data, as shown in the figure and

[0269] Select N based on predefined selection rules t In the case of group prediction data, N t The value of can be fixed or can change with the change of t, which is not limited in this application. Since the network device may not know in advance the number and location of the observation data groups selected by the terminal device in each time unit, the terminal device can feedback the number and location of the observation data groups selected in each time unit to the terminal device.

[0270] How does the terminal device select N? t The group observation data may be predefined by a protocol, or may be indicated by a network device, or may be determined by the terminal device itself.

[0271] Optionally, the method further includes: the network device sends second information to the terminal device, or the terminal device receives second information from the network device, the second information is used to indicate Nt The method of selecting a group of observation data, the N t The selection methods of group observation data include: selection based on predefined patterns, or selection based on predefined selection rules.

[0272] That is, the network device instructs the terminal device N t The method of selecting the observation data set.

[0273] Furthermore, if the network device instructs the terminal device to select N based on a predefined pattern t When there are multiple predefined patterns for group observation data, the network device can also instruct the terminal device on which pattern to select. For example, the second information is also used to indicate the first pattern. Of course, the terminal device can also independently determine which pattern to select based on and feedback it to the network device.

[0274] If the network device instructs the terminal device to select N based on a predefined selection rule t The network device can also instruct the terminal device to t The observation data of the group is in S t A feedback method for the position in the group observation data includes: feedback through a bitmap or feedback through a combination identifier.

[0275] By indicating the feedback mode, it is convenient for both the terminal device and the network device to generate and interpret the N-signal based on the same feedback mode. t The observation data of the group is in S t The information of the position in the set of observation data is the first information described below.

[0276] As mentioned above, the terminal device has t The observation data of the group is in S t The feedback methods for the position in the group observation data include: bitmap feedback or combined identifier feedback. The following describes these two feedback methods in detail.

[0277] Feedback via bitmap: The bitmap may include S t bits, and S t Each bit can be used to indicate whether the corresponding set of observation data is selected, or in other words, each bit can be used to indicate whether the corresponding set of observation data is included in the N t group of observation data.

[0278] Taking the example of selecting observation data shown in the above text and Figure 8 as an example, in the t1th time unit, the terminal device selects the 1st group of observation data and the 5th group of observation data from the 6 groups of observation data, and the corresponding bitmap can be expressed as "100010"; in the t2th time unit, the terminal device selects the 3rd group of observation data and the 5th group of observation data from the 6 groups of observation data, and the corresponding bitmap can be expressed as "001010"; in the t1th time unit, the terminal device selects the 1st group of observation data and the 2nd group of observation data from the 6 groups of observation data, and the corresponding bitmap can be expressed as "110000".

[0279] Furthermore, in order to save transmission overhead, the bitmap can also be converted into a higher-base value for indication, and each value can correspond to an identifier. In this way, when the data dimension is high, for example, the observation data is point cloud data, the bitmap can be compressed and transmitted, thereby saving signaling overhead and improving transmission efficiency.

[0280] Feedback by combination identification: In this embodiment, the t The combination identifier corresponding to the combination of the group observation data is recorded as the first combination identifier. The first combination identifier can be one of multiple combination identifiers. The multiple combination identifiers can be N t There is a one-to-one correspondence between the various combinations of observation data.

[0281] For S t Group observation data selection N t In the case of a group of observation data, the bit length can be The combination identifier is used to uniquely identify each possible combination. Indicates rounding up. Table 1 below shows S t 6, N t When the value is 2, this is an example of the correspondence between multiple combination identifiers and multiple combinations.

[0282] Table 1

[0283] Table 1 is only an example and should not constitute any limitation to this application. For example, the corresponding relationship between each combination and combination identifier shown in Table 1 can be adjusted. For another example, in S t and N t In the case of other values, a correspondence between multiple combinations and multiple combination identifiers similar to Table 1 can be obtained. For example, the bit length of the combination identifier can be longer or shorter, and the number of combinations can be more or less. For the sake of brevity, examples are not given here one by one.

[0284] Since N t It can change with the change of t, so each time the N selected by the combined identification feedback is tWhen the observation data is set, N t Network devices and terminal devices can also pre-store N t A corresponding relationship table between the combination identifier and the combination under different values ​​is provided, so that both parties can generate and interpret the combination identifier based on the same corresponding relationship table.

[0285] Figure 9 is a schematic diagram of the second information provided by an embodiment of the present application. Figure 9 (a) and (b) respectively show two possible forms of the second information.

[0286] In (a) of Figure 9 , the second information includes two fields, each used to indicate the following two items: a selection method and a pattern. The selection method may be based on a predefined pattern selection, and the pattern may be the first pattern.

[0287] In FIG9( b ), the second information includes two fields, which are used to indicate the following two items: selection method and feedback method. The selection method can be based on predefined selection rules, and the feedback method can be feedback through bitmap or combination identifier.

[0288] The above is named based on predefined patterns and based on predefined selection rules to distinguish two different selection methods. Such naming is only an example and should not constitute any limitation. For example, selection based on predefined patterns can also be called static selection, and selection based on predefined selection rules can also be called dynamic selection. For example, the two can also be distinguished by the first selection method and the second selection method. This application includes but is not limited to this.

[0289] Similarly, feedback through bitmap and feedback through combination identifier are also named to distinguish two different feedback methods. Such naming is only an example and should not constitute any limitation. For example, these two feedback methods can also be distinguished by the first feedback method and the second feedback method. This application includes but is not limited to this.

[0290] Furthermore, the second information and the seventh information described above, any one of the third information or the fifth information, can be carried in the same control signaling as the seventh information and the second information. Figure 10 is a schematic diagram of the control signaling provided in an embodiment of the present application. As shown in the figure, the control signaling may include, for example, a partition indication field and a selection indication. The partition indication field can be used to carry the seventh information and the third information, or can be used to carry the seventh information and the fourth information, and the selection indication field can be used to carry the second information. For specific examples of the third information, the fourth information and the second information, please refer to the description in conjunction with Figures 5, 6 and 9 above, and no further details will be given.

[0291] In step 2202, the terminal device may, based on the N selected in step 2201,t The set of observation data determines N t Group 1 data.

[0292] In one possible implementation, the N t The first data of the group is N t Group observation data.

[0293] That is, the terminal device can directly use part of the observation data in the observation data set as the N to be transmitted to the network device. t The N groups of first data are represented by vectors, for example, In order to t The forecast data of the groups are distinguished.

[0294] As an example, suppose S t is 4, N t is 2, and N t The set of observation data is S t Group 1 and Group 3 in the group observation data,

[0295] Then we can get the N t The vector representation of the first data set is:

[0296] For ease of understanding, FIG11 shows the terminal device obtaining the N t As shown in the figure, after operation ①, the terminal device completes the observation data set. The data is divided into 4 groups, and 4 groups of observation data are obtained. to Operation ① can correspond to step 210 above; after operation ②, the terminal device completes the selection of 4 sets of observation data and obtains 2 sets of first data and Operation ② may correspond to obtaining N in step 220. t The specific process of each operation can be found in the relevant instructions of the corresponding steps and will not be repeated here.

[0297] Considering that noise is inevitably introduced in the process of observing and compressing data, the N t Group First Data It can be further expressed as:

[0298] in, represents the real data corresponding to the t-th time unit, represents the noise introduced by observation and compression corresponding to the t-th time unit, which can be considered as independent and identically distributed zero-mean Gaussian noise. For the convenience of distinction and explanation, the covariance matrix of the noise introduced by observation and compression is denoted as the first covariance matrix below.

[0299] Assume that N t The first data set is N obtained by evenly dividing t The first covariance matrix Σ0 can satisfy the following:

[0300] in, represents the variance of the Gaussian distribution, Indicates the dimension The unit array.

[0301] O(t) can be understood as being able to describe N t The observation data of the group is in S t The observation matrix of the position in the set of observation data can be a matrix of dimension L×L. t In the set of observations, the rows and columns correspond to N t The value of the position of the observation data of the group can be 1, and the other positions can be 0. That is, from the data in the prediction data set, only the data with N t The predicted data for the same position of the group of observation data are taken as 0, and the predicted data for other positions are taken as 0.

[0302] N t The first data set is N obtained by evenly dividing t The corresponding observation matrix O(t) can satisfy the following:

[0303] In this observation matrix, except for the two unit matrices on the diagonal corners, which correspond to the positions of the first group of observation data and the third group of observation data respectively, the others are zero matrices.

[0304] It should be noted that the Kalman gain matrix and the observation matrix illustrated in this article are based on the following assumptions: the observation data set includes L data in total, and the four groups of observation data obtained by uniform division are the above S t Groups of observation data, of which the first and third groups of observation data are selected N t group of observation data. If the S t The observation data set is obtained through non-uniform division, then the number of data included in the observation data set, S t The number of data included in each set of observation data is used to obtain block diagonal matrices of different dimensions, and then combined with the selected N tThe observation data of group N t The positions in the group of observation data are spliced ​​together to obtain the Kalman gain matrix and the observation matrix.

[0305] Optionally, the N t The first data of the group is N t Group residual data.

[0306] The residual data can be used to describe the difference between the observed data and the predicted data. In this embodiment, the observed data set and the predicted data set are corresponding, and both are sets of data corresponding to the same time unit (for example, the t-th time unit). Among them, the predicted data set can be a set of data obtained by prediction. Taking the predicted data set and the observed data set corresponding to the t-th time unit as an example, the number of data in the predicted data set is equal to the number of data in the observed data set, and the observed data set can be divided into S t The set of observation data can also be divided into S t The number of data in the nth group of observation data is equal to the number of data in the nth group of prediction data.

[0307] In N t In the set of residual data, the nth set of residual data is obtained by subtracting the observation data and the predicted data corresponding to the same position in the nth set of observation data and the nth set of predicted data. n can be 1 to N t For example, the i-th data in the n-th group of observations is recorded as The i-th data in the n-th group of predicted data is recorded as The i-th data in the n-th group of residual data is recorded as but i can be from 1 to L n Integer value in, L n Indicates the number of data contained in the nth group of residual data (or the nth group of predicted data, or the nth group of observed data), L n Is a positive integer.

[0308] Similar to the observed data set, in this paper, the predicted data set corresponding to the t-th time unit can be represented by a vector of length L, for example, The residual data corresponding to the t-th time unit can be represented by a vector of length L, for example,

[0309] As an example, suppose S t is 4, N t is 2, and N t The set of observation data is S t The first and third groups of observation data are obtained, then Nt The vector representation of the first data set is: In the formula Can meet: Indicates N t The nth group of data in the first group of data, Indicates N t The nth group of data in the group prediction data, Indicates N t The nth group of data in the group residual data.

[0310] Alternatively, in order to reduce the dimension of matrix calculation, the residual data can be recorded as The length can be L n Indicates N t The length of the nth group of observations in the set of observations.

[0311] For ease of understanding, FIG12 shows the terminal device obtaining the N t As shown in the figure, after operation ①, the terminal device completes the observation data set. The data is divided into 4 groups, and 4 groups of observation data are obtained. to Operation ① can correspond to step 210 above; after operation ②, the terminal device completes the selection of 4 sets of observation data and obtains 2 sets of first data and Operation ② may correspond to obtaining N in step 220. t After operation ③, the terminal device obtains the difference between the two groups of prediction data and the prediction data of the corresponding position. and Operation ③ may correspond to obtaining N in step 220. t The implementation method of group residual data. The specific process of each operation can be found in the relevant instructions of the corresponding steps and will not be repeated here.

[0312] Comparing the two forms of the first data mentioned above, it can be found that if the first data is observation data, the terminal device can directly select N from the observation data set. t The first data is a set of observations, without requiring additional operations. If the first data is residual data, however, the terminal device still needs to perform prediction to obtain predicted data and then calculate the residual data. In comparison, the former does not require the terminal device to perform prediction, which can save processing power. However, since the observed data itself is larger than the residual data, the transmission overhead is greater than the latter. In other words, the residual data can compress the observed data to a greater extent, thereby further reducing transmission overhead.

[0313] Optionally, the method further includes: step 230, the terminal device obtains a set of prediction data corresponding to the tth time unit.

[0314] The terminal device can perform data prediction based on an existing prediction method to obtain a predicted data set corresponding to the t-th time unit. For example, the prediction method can be AR prediction, or a derivative algorithm of AR prediction, such as MA prediction, ARIMA prediction, etc., or AI prediction, which will not be listed here. This application does not limit the prediction method.

[0315] As an example, the prediction method is AR prediction, and the prediction data corresponding to the t-th time unit can be obtained based on p prediction data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

[0316] Because AR prediction exploits the temporal correlation of data, using historical data to predict future data, in this embodiment, the closer the historical moment is to the tth time unit, the more accurate the prediction. Therefore, a can be 1. When a is 1, the predicted data for the tth time unit is based on the p predicted data corresponding to the (t-p)th time unit to the (t-1)th time unit.

[0317] AR prediction is based on the autocorrelation between the values ​​of the historical time series of the data to be reconstructed at different times, so as to establish a regression equation for prediction and complete data reconstruction. Among them, the historical time series is a sequence composed of data from multiple historical moments. Therefore, the larger the value of p, that is, the larger the order of AR prediction, which means that more historical data is used for prediction, that is, more abundant historical data can be obtained for prediction, and the results of AR prediction are more accurate. At the same time, the amount of data for AR prediction also increases, and the amount of calculation also increases. In this embodiment, the value of p can be predefined by the protocol, or determined by the network device itself, or notified by the terminal device, or can also be set manually. This application does not impose any restrictions on this.

[0318] Assuming that the prediction method is AR prediction, if the prediction data set of the t-th time unit is represented by the vector Indicates that satisfy:

[0319] Where p represents the order of AR prediction, Represents the data of the (t-τ-a+1)th time unit, and a is a constant. That is to say, the network device can predict the predicted data of the tth time node based on the historical data of p time nodes. Among them, the data of the (t-τ-a+1)th time unit can also be obtained by prediction, that is, the predicted data of the first p time units are used as historical data to predict the predicted data of the next time unit. As time goes by, the first p time units are also advancing, and through continuous iterative updates, the predicted data of the most recent time unit is obtained. Among them, a can be 1, and when a is 1, Formula 1 can be simplified to: That is, the predicted data of the t-th time node is predicted based on the historical data of the p time nodes from the (t-1)th time node to the (t-p)th time node before the t-th time node. τ Represents the state transfer matrix, corresponding to different values ​​of τ, A τ Can be expressed as A1, A1, ..., A p . Change A1, A1,...,A p Splicing to get matrix A o , A o =[A1A2…A p ], which can be obtained by solving the Yule-Walker equation as follows:

[0320] In formula 3, represents Gaussian noise, and I represents the unit matrix.

[0321] Formula 4 is used to calculate the autocorrelation of M (M is a positive integer) training signals. Where M represents the number of training signals, and i can take integer values ​​from 0 to M-1. represents the nth data in M ​​training signals, express The conjugate transpose of Express request expected value.

[0322] Solving formula 4 yields The R(i) corresponding to each value of i in the equation is obtained by formula 3. Substituting into formula 2, we can get [A1A2…A p ]. Substituting formula 2 into formula 1, we get:

[0323] In formula 5, It represents the matrix obtained by concatenating the vectors corresponding to the prediction data sets corresponding to the p time units from the tth time unit to the (t-p+1)th time unit in the order of time from near to far. The matrix represents the concatenation of the vectors corresponding to the prediction data sets corresponding to the p time units (t-1) to (t-p) in order of time from near to far. It can be seen that the prediction data set for the tth time unit can be obtained based on the prediction data set for the (t-1)th to (t-p)th time units.

[0324] Figure 13 is a schematic diagram of AR prediction. If the module used to perform AR prediction is regarded as a model, such as an AR model, the prediction data output by the AR model each time is affected by the prediction data output by the previous AR prediction. In other words, the prediction data output by the AR model in the previous AR prediction is affected by the state transition matrix A. τ The role is to generate the prediction data for the next AR prediction.

[0325] In addition, AR prediction may introduce errors. The covariance matrix of the error introduced by each AR prediction is Σ ○ satisfy:

[0326] For the convenience of distinction and explanation, the covariance matrix of the error introduced by each AR prediction is hereinafter referred to as the second covariance matrix.

[0327] As time goes by, the error introduced by each AR prediction will continue to accumulate. For the t-th time unit, the covariance matrix P of the error introduced by the corresponding AR prediction is ○ (t) satisfies: P ○ (t) = A ○ P ○ (t-1)A ○H +Σ ○ …………Formula 7.

[0328] Among them, Σ ○ A represents the covariance matrix of the error introduced by the AR prediction performed at the t-th time unit, ○ P ○ (t-1)A ○H It represents the covariance matrix of the cumulative error of the AR prediction before the tth time. It can be seen that as time goes by, the covariance matrix of the accumulated error of the AR prediction can be continuously iteratively updated.

[0329] For the convenience of distinction and explanation, the following will be the sum of the error covariance matrix of each AR prediction and the covariance matrix of the cumulative error of the previous AR prediction (that is, the P calculated by formula 7 above) ○ (t)) is recorded as the third covariance matrix.

[0330] The above description is merely for ease of understanding, using AR prediction as one possible implementation method for obtaining prediction data, and details the process of obtaining the prediction data set for the t-th time unit. However, this does not constitute any limitation on this application. In other prediction methods, the prediction data set for the t-th time unit can also be obtained using other formulas or algorithms, which will not be further described here.

[0331] In step 2202, the terminal device t N determined by a set of observation data t The first data of the group is N t The set of observation data is still N t The group residual data may be predefined by a protocol, or may be determined by one of the terminal device and the network device and then notified to the other end.

[0332] If it is determined by the terminal device, optionally, the method further includes: the terminal device sends eighth information to the network device, or the network device receives the eighth information from the terminal device, the eighth information is used to indicate N t The first data of the group is N t The set of observation data is still N t The residual data of the group, or N t The data type of the first data in the group.

[0333] If it is determined by the network device, optionally, the method further includes: the network device sends eighth information to the terminal device, or the terminal device receives the eighth information from the network device, the eighth information is used to indicate N t The first data of the group is N t The set of observation data is still N t The residual data of the group, or N t The data type of the first data in the group.

[0334] As an example, if the eighth information is sent by the network device to the terminal device, the eighth information can be carried in the same control signaling as the third information or one of the fourth information, the second information and the seventh information mentioned above, and sent to the terminal device.

[0335] In step 240, the terminal device sends N t Accordingly, the network device receives N from the terminal device. t Group 1 data.

[0336] The Nt The first set of data can be used to compensate for the predicted data corresponding to the t-th time unit. Here, the predicted data can refer to data predicted by the network device. Therefore, the network device needs to perform data prediction to obtain a set of predicted data corresponding to the t-th time unit.

[0337] It should be noted that the N shown in step 220 above t Example of group first data and In the example, it is represented as a vector of length L just for the convenience of calculation. In the actual transmission process, since the network equipment can determine N t The observation data of the group is in S t The position in the group of observation data, so the terminal device can send out the non-zero sub-array, such as in the above example, in and or in and As a result, the transmission overhead can be greatly reduced.

[0338] Optionally, the method further includes: step 250, the network device obtains a prediction data set corresponding to the tth time unit.

[0339] The network device may also perform prediction based on an existing prediction method to obtain a set of prediction data corresponding to the t-th time unit. For example, the prediction method may be AR prediction, or a derivative algorithm of AR prediction, such as MA prediction, ARIMA prediction, etc., or AI prediction, which will not be listed here. This application does not limit the prediction method.

[0340] As previously mentioned, when the terminal device feeds back residual data as the first data, it also needs to obtain a set of predicted data corresponding to the tth time unit. In this embodiment, the network device and the terminal device can obtain this predicted data based on the same prediction method. Therefore, the process by which the network device obtains the set of predicted data corresponding to the tth time unit is similar to that by the terminal device. The specific process of step 250 can be found in the relevant description of step 230 above and will not be repeated here.

[0341] In step 260, the network device t The first data of the group is used to compensate the predicted data corresponding to the t-th time unit to obtain compensated data.

[0342] As mentioned above, the N tThe first set of data can be used to determine the compensation data corresponding to the t-th time unit, and the compensation data corresponding to the t-th time unit can be used to compensate the prediction data corresponding to the t-th time unit. Accordingly, step 260 can specifically include:

[0343] Step 2601: The network device t The first data of the group determines the compensation data corresponding to the t-th time unit;

[0344] In step 2602 , the network device obtains compensated data based on the compensation data corresponding to the t-th time unit and the predicted data corresponding to the t-th time unit.

[0345] The following is a detailed description of step 2601 and step 2602.

[0346] In step 2601, the network device may t Set the first data and determine N t Group prediction data and N t The difference between the two groups of observation data, and then the N t Compensation is performed based on the set of forecast data.

[0347] For example, the network device may determine compensation data based on KF prediction. If the compensation data is represented by a vector, for example, ΔV(t), then ΔV(t) satisfies: or,

[0348] in, and N t The two forms of the first data of the group correspond to N in the previous step 2202. t Group observation data and N t Group residual data, is the predicted data corresponding to the t-th time unit.

[0349] O(t) is the observation matrix, and the relevant description can be found in the previous step 2022, which will not be repeated here.

[0350] As mentioned before, N t The observation data of the group is in S t The position in the set of observations can be indicated by a bitmap or a combination identifier, or in the case of uniform partitioning, the N t The observation data of the group is in S t The position of the group observation data can also be determined by the number of groups N t and / or the number of data in each set of observation data, so the network device can know N in advance. t The observation data of the group is in St The position in the set of observation data, that is, the observation matrix can be generated.

[0351] For example, in the above example, a bitmap is used to indicate N t The observation data of the group is in S t The position of the group observation data, the bitmap can be, for example There are L bits in total, from low to high, they are L / 4 "1", L / 4

[0352] "0", L / 4 "1"s, and L / 4 "0". This bitmap can also be indicated by higher-base values, which will not be used as an example here. From this, we can determine that the first L / 4 rows and the first L / 4 columns of the observation matrix are the identity matrix, the (L / 2+1)th row to the 3L / 4th row and the (L / 2+1)th column to the 3L / 4th column are the identity matrix, and the rest of the positions are all 0.

[0353] K(t) is the Kalman gain matrix corresponding to the t-th time unit, which can be obtained through existing technologies.

[0354] For example, assuming that the above prediction data is obtained through AR prediction, K(t) may satisfy:

[0355] K(t)=P AR (t)O H (t)(O(t)P AR (t)O H (t)+Σ o ) -1 …………Formula 8.

[0356] Among them, P AR (t) from the matrix Can meet:

[0357] Where * indicates a non-zero irrelevant term.

[0358] According to the above formula 7, the third covariance matrix of AR prediction can be expressed as:

[0359] In formula 8 Used to compare with the updated P based on KF prediction in the following text ○ (t) distinguish and define, Represents the covariance matrix of the error after AR prediction, P ○ (t) represents the error covariance matrix after the KF prediction is updated based on the AR prediction. Both are essentially the third covariance matrix. Compared with P in the previous formula 7 ○(t) is the same. Since the covariance matrix of the error introduced by the AR prediction corresponding to the t-th time unit can be calculated, the result obtained is as shown in Formula 9. From this, we can get P AR (t), which can be substituted into Formula 8 to obtain K(t).

[0360] According to formula 8, K(t) has the following form:

[0361] Here K1(t) and K3(t) represent the non-zero submatrices in K(t).

[0362] After KF prediction, we can Update as follows: P KF (t)=(IK(t)O(t))P AR (t)…………Formula 11,

[0363] In formula 12, P ○ (t) represents the updated third covariance matrix. Comparing Formula 9 and Formula 12, we can find that P in AR (t) in P ○ (t) is replaced by P KF (t), but is obtained through KF prediction. That is, through KF prediction, the third covariance matrix can be updated at the same time. The covariance matrix of the error introduced by the AR prediction after each KF prediction update can be used to calculate the Kalman gain matrix used for the next KF prediction. That is, the P obtained by formula 11 is ○ (t) as P ○ Substituting (t-1) into formula 10, we get Substituting into Formula 9 and Formula 8, we can obtain K(t) for the t-th time unit. This completes an iterative update of K(t).

[0364] In the above iterative update process of K(t), P in Formula 10 ○ The initial value of (t-1) can be the covariance matrix of the noise introduced by observation and compression Thereafter, the third covariance matrix may be iteratively updated according to Formulas 8 to 12 provided above.

[0365] In step 2602, when the compensation data for the t-th time unit is used to compensate the predicted data for the t-th time unit, it can be used to compensate the predicted data corresponding to the same position in the predicted data set. The compensated data obtained through compensation is a complete data set of dimension L. This compensated data is output in the form of a predicted data set and can therefore also be referred to as a compensated predicted data set. The implementation process of step 260 above can also be referred to as a process of reconstructing the predicted data set.

[0366] One possible design is that the compensated data corresponding to the t-th time unit is the sum of the compensated data corresponding to the t-th time unit and the predicted data corresponding to the t-th time unit.

[0367] If the compensated data (or reconstructed data) corresponding to the t-th time unit is represented by a vector, for example, but Can meet:

[0368] The network device can determine the compensated data corresponding to the t-th time unit based on one of Formula 13 and Formula 14. For example, when the terminal device sends N t The first data of the group is N t When the terminal device sends N t The first data of the group is N t When the residual data is collected, the network device can determine it based on Formula 14.

[0369] Of course, the above formulas 13 and 14 are only examples. For example, by substituting formulas 13 and 14 into formula 1, the following transformations of formulas 13 and 14 can be obtained:

[0370] or,

[0371] For example, to simplify the calculation, we can perform dimensionality reduction on Formula 13 to obtain the following transformation:

[0372] Indicates N t The nth group of observations in the set of observations, Indicates S t The prediction data of the group is t A set of prediction data with the same position as the nth set of observation data in the set of observation data. Assume that the length of the nth set of observation data is L n , then the dimension of the observation matrix K'(t) also becomes The dimension of K'(t) here is different from that of K(t) above, but the meaning is the same.

[0373] Similarly, in Formula 14 Can also be replaced with The dimensions can be

[0374] In this case, Formula 14 can be transformed as follows:

[0375] Dimensionality reduction can simplify calculations, reduce the computational complexity of the device, and reduce power consumption.

[0376] Those skilled in the art can perform other mathematical transformations on the formula based on the same concept, which are not listed here for the sake of brevity.

[0377] In addition, the compensated data is not necessarily the sum of the compensated data and the observed data. For example, it can also be a weighted sum, etc., which will not be listed here.

[0378] Figure 14 is a schematic diagram of KF prediction. If the module used to perform KF prediction is regarded as a model, such as the KF model, the N feedback from the terminal device is t The first data of the group is the input of the KF model. The KF model can compensate the AR prediction data output by the AR model by calculating the Kalman gain matrix, and obtain and output the compensated data.

[0379] For ease of understanding, FIG15 and FIG16 respectively illustrate the process of data compensation performed by the network device.

[0380] In the process shown in Figure 15, the network device receives N t The first data of the group is N t The network device can perform data compensation based on the above formula 13 or its variant. t Group observation data After operation ④, we can get N t Group residual data The N t Group residual data The data in After operation ⑤, the network device can obtain the compensated data, operation ④ and operation ⑤ may correspond to the above step 260 using formula 13 or its variation to perform data compensation implementation, wherein the specific process of operation ④ can refer to step 220 to obtain N t The implementation method of group residual data will not be described in detail.

[0381] In the process shown in Figure 16, the network device receives N t The first data of the group is N t The network device can perform data compensation based on the above formula 14 or its variant. As shown in the figure, the network device receives N t After the residual data is grouped, the compensated data can be obtained through operation ⑥. Operation ⑥ corresponds to the implementation method of using formula 14 or its variation to perform data compensation in step 260 above, and will not be repeated here.

[0382] It should be noted that, in some embodiments, the implementation process of step 2601 may be referred to as KF prediction, and in other embodiments, the processes of steps 2601 and 2602 may be referred to as KF prediction. This application does not limit the steps included in KF prediction. In addition, obtaining compensation data through KF prediction is only one possible method and should not constitute any limitation to this application. For example, compensation data can also be obtained based on a derivative algorithm of KF prediction, such as extended KF, adaptive KF, time-delay KF, etc., or it can also be calculated based on an AI model, such as an RNN model, an LSTM model, a transformer model, etc., which will not be repeated here.

[0383] Whether the network device performs data compensation based on Formula 13 or Formula 14 can be determined according to the form of the first data sent by the terminal device in step 240. The terminal device feeds back N in step 240. t The set of observation data is still N t The group residual data may be predefined by a protocol, or may be indicated by a network device, or may be determined by the terminal device itself.

[0384] Optionally, the method further includes: the network device sends eighth information to the terminal device, or the terminal device receives eighth information from the network device, the eighth information is used to indicate N t The first data of the group is N t The set of observation data is still N t Group residual data.

[0385] In other words, the network device can instruct the terminal device through the eighth information what form of data to feed back.

[0386] Optionally, the method further includes: the terminal device sends eighth information to the network device, or the network device receives eighth information from the terminal device, the eighth information is used to indicate N t The first data of the group is N t The set of observation data is still N t Group residual data.

[0387] In other words, the terminal device can notify the network device of the form of data fed back through the eighth information.

[0388] Whether it is feedback N t Group observation data or feedback N t A set of residual data can directly or indirectly inform the network device of this N t The group observation data is collected, making it easier for network devices to perform data compensation.

[0389] From another perspective, N t The transmission of group observation data and N t The transmission of the residual data of the group can also be regarded as two different transmission modes, for example, recorded as the first transmission mode and the second transmission mode. The first transmission mode or the second transmission mode, the first transmission mode is the feedback N t The second transmission mode is to send N t Transmission mode of group residual data.

[0390] Therefore, it can also be said that the eighth information is used to indicate N t The transmission mode of the group observation data includes: a first transmission mode or a second transmission mode, the first transmission mode is feedback N t The second transmission mode is to send N t Transmission mode of group residual data.

[0391] When the same compression rate is used, the terminal device can obtain more accurate reconstruction performance by using the second transmission mode compared to the first transmission mode. Alternatively, when the same reconstruction performance index is used, the terminal device can save transmission resources by using the second transmission mode compared to the first transmission mode.

[0392] The compression ratio can be expressed as the ratio of the compressed data size to the uncompressed data size. For example, the CSI compression ratio can be expressed as the ratio of the bit length of the compressed CSI (or the CSI actually transmitted) to the bit length of the observed CSI (or the CSI desired for feedback). In other words, when using the same compression ratio and given a fixed number of transmission resources, the second transmission mode can indicate more information than the first transmission mode, thereby facilitating more accurate reconstruction performance.

[0393] Terminal device obtains and feedbacks N tThe operation of obtaining the first data set may be performed once or multiple times, which may be determined by the configuration of the network device. That is, the t-th time unit may be determined based on the configuration of the network device. For example, the network device may configure T time units through signaling, and the t-th time unit is one of the T time units, where T may be greater than 1 or equal to 1. In other words, the terminal device may obtain and feedback N t The operation of the first data of the group can also be performed multiple times to obtain and feedback N t Operation on the first data of the group.

[0394] Optionally, the method further includes: the network device sending sixth information to the terminal device, or the terminal device receiving sixth information from the network device, the sixth information being used to indicate T time units. That is, the network device configures T time units for the terminal device through the sixth information.

[0395] In one possible case, T is an integer greater than 1, and the T time units are multiple time units. The T time units may be arranged at equal intervals in the time domain.

[0396] The network device is in the first mode within the T time units and can be in the second mode throughout the entire time domain. The terminal device is in the first mode within the T time units and can be in the second mode throughout the entire time domain (for example, N t The first data of the group is N t group of residual data), or is neither in the first mode nor in the second mode (such as N t The first data of the group is N t group residual data).

[0397] For example, the first mode may be a KF prediction mode, and the second mode may be an AR prediction mode. That is, the network device may continuously perform AR prediction in the second mode and perform KF prediction in the first mode. Accordingly, when the terminal device is in the second mode, the terminal device may continuously perform AR prediction in the second mode and obtain and feedback N in the first mode. t Group 1 data.

[0398] FIG17 is a schematic diagram of T time units provided in an embodiment of the present application. As shown in the figure, the T time units include t1 to t T , t1 to t T They are arranged at equal intervals in the time domain. For example, t1 is the first time slot in the time domain, t2 is the sixth time slot in the time domain, t3 is the eleventh time slot in the time domain, and so on. There are 5 time slots between every two adjacent time units in the T time units.

[0399] In each of these T time units, the terminal device and the network device are in the first mode, and in other time units, such as the time unit between t1 and t2, the time unit between t2 and t3, and so on (not marked in the figure), the terminal device and the network device are in the second mode.

[0400] For example, the T time units are periodic, or in other words, static. Accordingly, the sixth information may be used to indicate the period length of the T time units.

[0401] The cycle length may specifically refer to the time length between every two adjacent time units in the T time units, or in other words, the time offset between every two adjacent time units.

[0402] Optionally, the sixth information is also used to indicate the start time of the T time units.

[0403] By indicating the start time of the T time units, the terminal device can determine when to start entering the T time units. Of course, the network device may not indicate the start time of the T time units. In this case, the terminal device can assume that it has entered the T time units since receiving the sixth information.

[0404] For another example, the T time units are semi-persistent (SP), or in other words, semi-static. Accordingly, the sixth information may be used to indicate the period length of the T time units and the effective time of the T time units.

[0405] The effective time of T time units specifically means that the first mode is entered for T time units within the effective time range, and the first mode is not entered outside the effective time range.

[0406] The sixth information may be carried in, for example, high-layer signaling (such as an RRC message), that is, the time for the terminal device to enter the first mode is statically or semi-statically configured through high-layer signaling.

[0407] The terminal device may also enter the first mode in some or all of the T time units based on the sampling rate. For example, if the sampling rate is 1, the terminal device may enter the first mode in each of the T time units; if the sampling rate is 0.5, the terminal device may enter the first mode once every other time unit in the T time units. And so on.

[0408] The sampling rate may be configured by the network device, determined by the terminal device itself, or predefined by the protocol, and this application does not limit this.

[0409] Another possible situation is that T is 1, and the T time units may be non-periodic, or dynamic. Accordingly, the sixth information is used to indicate the time of the T time units.

[0410] That is, the terminal device can obtain and feedback N in T time units. t The first data of the group can be collected without having to perform the operation continuously, thereby saving power consumption caused by data observation.

[0411] The sixth information may be carried in physical layer signaling (such as DCI), for example. That is, the time for the terminal device to enter the first mode is dynamically configured through physical layer signaling.

[0412] Corresponding to the three different time domain behaviors of periodic, semi-persistent, and aperiodic, the network device can indicate them using different identifiers. The configurations corresponding to the three different time domain behaviors are referred to as configurations of different prediction rules. The first prediction rule indicates configuring T time units of periodicity, the second prediction rule indicates configuring T time units of semi-persistentity, and the third prediction rule indicates configuring T time units of aperiodicity. The network device can further carry an identifier for indicating the prediction rule in the sixth information to indicate the three different configurations.

[0413] Alternatively, the network device and the terminal device enter the first mode based on which of the three time domain behaviors mentioned above can also be predefined by the protocol.

[0414] Figure 18 is a schematic diagram of the sixth information provided by an embodiment of the present application. The sixth information shown in (a), (b) and (c) of Figure 18 can be used to indicate periodic, semi-persistent and non-periodic T time units.

[0415] In (a) in Figure 18, the sixth information includes three fields, which are used to indicate the following three items: prediction rule, cycle length and start time, where the prediction rule is the first prediction rule, the cycle length is the time length between every two time units in T time units, and the start time is the start time of T time units.

[0416] In (b) in Figure 18, the sixth information includes three fields, which are used to indicate the following three items: prediction rule, cycle length and effective time, where the prediction rule is the second prediction rule, the cycle length is the time length between every two time units in T time units, and the effective time is the effective time of T time units.

[0417] In (c) of FIG18 , the sixth information includes two fields, one for indicating a prediction rule and the other for indicating a time, wherein the prediction rule is the third prediction rule and the time is the time of the T time units.

[0418] Based on the above technical solution, the terminal device selects a part of the data from the set of observation data for data transmission, so that the network device can compensate the predicted data based on the received data, thereby reconstructing the complete data. This makes the compensated data closer to the observation data, which also makes the predicted data output by the network device closer to the observation data, thereby improving the accuracy of the predicted data. Compared with sending the entire set of observation data, this method selects and sends the data in the set, which can avoid the huge transmission overhead caused by the transmission of all the data in the set; compared with not reporting the observation data, this method can use the part of the data selected to be sent to compensate for the predicted data, which is conducive to obtaining more accurate predicted data. Therefore, an effective compromise is achieved between transmission overhead and data accuracy.

[0419] In addition, network equipment can configure data feedback time for terminal devices according to demand, so that terminal devices feedback data when there is demand and do not feedback data when there is no demand, thereby saving transmission overhead to a greater extent.

[0420] In the embodiment of the present application, the predicted data set may be obtained through model prediction. For the convenience of distinction and description, the model used to obtain the predicted data set is estimated as a prediction model, which may be, for example, an AR model.

[0421] The prediction model may be obtained through model training. The model training may be performed by at least one of the network device or the terminal device. The model training may be an operation performed before the prediction model is put into use, such as an operation performed before step 210 above, or an operation performed during the process of the prediction model being put into use, for example, although the terminal device may be able to t The first data of the group feeds back the observation data to the network device, but as time goes by, the prediction error will still accumulate at a relatively slow rate. Therefore, during the online use process, the network device can also determine whether the model needs to be retrained based on certain preset indicators. The preset indicators can, for example, be the prediction deviation of the reconstructed data, etc. This application does not limit this.

[0422] Optionally, the method further includes: the terminal device performs model training based on training samples to obtain the prediction model, and the training samples include: multiple signals for training and observation data obtained based on the multiple signals, or multiple observation data.

[0423] Optionally, the method further includes: the network device performs model training based on training samples to obtain the prediction model, and the training samples include: multiple signals for training and observation data obtained based on the multiple signals, or multiple observation data.

[0424] In one example, the signal used for training is a reference signal, and the observation data obtained based on these multiple signals is CSI. That is, the network device can send multiple reference signals for training to the terminal device, and the terminal device can perform channel measurement based on each received reference signal to obtain multiple CSI corresponding to these multiple reference signals. Alternatively, the terminal device can reuse the reference signals sent by the network device during the wireless communication process to perform channel measurement to obtain CSI. Therefore, the training samples may include or exclude signals used for training.

[0425] The CSI measured by the terminal device can be used as observation data and input into the prediction model. The prediction model can predict the CSI based on the prediction algorithm to obtain predicted data. The model training device (such as a network device and / or terminal device) performs model training based on the observation data and predicted data to obtain a prediction model.

[0426] Whether the model training is performed by the terminal device or the network device can be configured by the network device or predefined by the protocol.

[0427] Optionally, the method also includes: the network device sends ninth information to the terminal device, or the terminal device receives ninth information from the network device, where the ninth information is used to indicate model training, as well as one or more of the following: the order of the prediction model, the number of training signals, or the device that performs model training.

[0428] When the predicted data differs significantly from the observed data, the network device may retrain the prediction model. In this case, the ninth information instruction may be used to instruct the model training.

[0429] If the protocol predefines that one or both of the terminal device or the network device performs model training, the ninth information may not indicate the device that performs model training.

[0430] In the case where the execution device of the model training is configured by a network device, the ninth information may further indicate the device that executes the model training.

[0431] If the ninth information indicates that the device performing model training is a network device, the number of training signals may be further indicated. Since training signals are used for model training but are not necessary for model training, the number of training signals may not be indicated if the training samples do not include training signals.

[0432] If the ninth information indicates that the device performing model training is a terminal device, the order of the prediction model and / or the number of training signals may be further indicated. The order of the prediction model is specifically the order of the prediction model, such as the order of the AR model, the order of the ARMA model, and so on. Of course, when the prediction model is another model, the order of the model can also be replaced by the parameters of other prediction models. The training signal is used for model training, but it is not necessary for model training. Therefore, when the training sample does not include the training signal, the number of the training signal may not be indicated.

[0433] Based on the different devices performing model training, multiple training modes can be defined. For example, the multiple training modes include: a mode in which the network device performs model training without feedback on model parameters; a mode in which the network device performs model training with feedback on model parameters; a mode in which the terminal device performs model training; and a mode in which the network device and the terminal device perform model training. Model parameters are parameters used to build the model, and they vary slightly in different training modes. Since the above-mentioned multiple training modes will be described in detail below, they will not be detailed here.

[0434] The device for performing model training indicated by the ninth information above can also be replaced by a training mode. In other words, the ninth information is used to indicate the model training and one or more of the following: the order of the prediction model, the number of training signals, or the training mode. For ease of understanding, the following uses AR model training as an example and illustrates the processes of the above-mentioned various training modes with reference to the accompanying figures.

[0435] First training mode: a mode in which the network device performs model training without feeding back model parameters.

[0436] Figure 19 is a schematic flowchart of the model training provided in an embodiment of the present application.

[0437] 1901. The network device sends ninth information, which is used to indicate model training, and one or more of the following: the training mode is the first training mode, the order of AR prediction, or the number of signals used for training.

[0438] 1902. The network device sends a training signal to the terminal device.

[0439] 1903. The terminal device performs measurement based on the training signal to obtain observation data.

[0440] 1904. The terminal device sends observation data to the network device. Correspondingly, the network device receives the observation data from the terminal device.

[0441] 1905. The network device performs AR prediction based on the measured value to obtain prediction data.

[0442] 1906. The network device performs model training based on the observation data and prediction data to obtain an AR model.

[0443] Second training mode: a mode in which the network device performs model training and feeds back model parameters.

[0444] Figure 20 is another schematic flowchart of model training provided in an embodiment of the present application.

[0445] 2001. The network device sends ninth information, which is used to indicate model training, and one or more of the following: the training mode is the second training mode, the order of AR prediction, or the number of signals used for training.

[0446] In 2002, the network device sends a training signal to the terminal device.

[0447] In 2003, the terminal device performs measurements based on the training signal to obtain observation data.

[0448] In 2004, the terminal device sends observation data to the network device. Correspondingly, the network device receives the observation data from the terminal device.

[0449] In 2005, network equipment performed AR prediction based on the measured values ​​to obtain prediction data.

[0450] In 2006, network equipment conducted model training based on observation data and prediction data to obtain the AR model.

[0451] In 2007, the network device sends the model parameters to the terminal device. Correspondingly, the terminal device receives the model parameters from the network device.

[0452] In the second training mode, the model parameters can be used by the terminal device to construct the AR model. Exemplarily, the model parameters include: a transfer matrix for AR prediction, a first covariance matrix, and a second covariance matrix, where the first covariance matrix is ​​obtained based on the error introduced by noise, and the second covariance matrix is ​​determined based on the error of the most recent AR prediction.

[0453] The transfer matrix used for AR prediction can be obtained through training. The first covariance matrix can be obtained through statistics. For example, the first covariance matrix can be decomposed into an observation error covariance matrix and a data compression error covariance matrix. The former can be obtained by counting the changes in the transmitted pilot signal affected by noise, and the latter can be obtained by counting the differences between the compressed data and the uncompressed data. The second covariance matrix can be calculated using Formula 6.

[0454] The third training mode: a mode in which the model training is performed by the terminal device.

[0455] Figure 21 is another schematic flowchart of the model training provided in an embodiment of the present application.

[0456] 2101. The network device sends ninth information, which is used to indicate model training, and one or more of the following: the training mode is the third training mode, the order of AR prediction, or the number of signals used for training.

[0457] 2102. The network device sends a training signal to the terminal device.

[0458] 2103. The terminal device performs measurement based on the training signal to obtain observation data.

[0459] 2104. The terminal device performs AR prediction and obtains prediction data.

[0460] 2105. The terminal device performs model training based on the observation data and the prediction data to obtain an AR model.

[0461] 2106. The terminal device sends the model parameters to the network device. Correspondingly, the network device receives the model parameters from the terminal device.

[0462] In the third training mode, the model parameters can be used by the network device to construct the AR model. Exemplarily, the model parameters include:

[0463] The transfer matrix for AR prediction, the initial value for AR prediction, the first covariance matrix, and the second covariance matrix are obtained based on the error introduced by noise, and the second covariance matrix is ​​determined based on the error of the most recent AR prediction. For the relevant description of the first covariance matrix and the second covariance matrix, please refer to the above and will not be repeated here.

[0464] The initial value for AR prediction may be, for example, p prediction data. The method for obtaining the transfer matrix, the first covariance matrix, and the second covariance matrix for AR prediction has been described in the second training mode, and will not be repeated here.

[0465] The fourth training mode is a mode in which model training is performed by network devices and terminal devices.

[0466] Figure 22 is another schematic flowchart of the model training provided in an embodiment of the present application.

[0467] 2201. The network device sends ninth information, which is used to indicate model training, and one or more of the following: the training mode is the fourth training mode, the order of AR prediction, or the number of signals used for training.

[0468] 2202. The network device sends a training signal to the terminal device.

[0469] 2203. The terminal device performs measurement based on the training signal to obtain observation data.

[0470] 2204. The terminal device sends the observation data to the network device. Correspondingly, the network device receives the observation data from the terminal device.

[0471] 2205. The terminal device and the network device respectively perform AR prediction to obtain prediction data.

[0472] 2206. The terminal device and the network device perform model training based on the observation data and the prediction data respectively to obtain an AR model.

[0473] In the four training modes provided above, several possible situations are shown in which the device performing model training is a terminal device, a network device, and a terminal device and a network device. In actual application, any of the above modes can be used according to needs.

[0474] Figure 23 is a schematic diagram of the ninth information provided by an embodiment of the present application. As shown in the figure, the ninth information includes four fields, which are used to indicate the following four items: model training, the order of the model, the number of training signals, and the training mode. Among them, performing model training means notifying the terminal device that the information transmitted this time is used for model training; the order of the model is, for example, the order of AR prediction; the number of training signals is the number of signals required for model training; the training mode is, for example, any one of the four training modes mentioned above, that is, indicating whether the model training is performed by the terminal device or the network device.

[0475] The ninth information shown in Figure 23 is only an example. In actual applications, the ninth information may include two or more of the above four fields to indicate model training, as well as one or more of the order of the model, the number of training signals or the training mode.

[0476] Through the above model training, the network device can obtain a prediction model, which is applied in the aforementioned step 250. The terminal device can also obtain a prediction model, which is applied in the aforementioned step 230.

[0477] The method provided by the embodiment of the present application is described in detail above with reference to a plurality of drawings. The device provided by the embodiment of the present application is described below with reference to the drawings.

[0478] Figures 24 to 27 are schematic diagrams of possible devices provided in embodiments of the present application. These devices can be used to implement the functions of the terminal device or network device in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0479] FIG24 is a schematic block diagram of a data processing device according to an embodiment of the present application. As shown in FIG24 , the device 2400 includes a transceiver module 2410 and a processing module 2420 .

[0480] One possible design is that the apparatus 2400 is used to implement the functions of the terminal device in the method embodiment shown in Figure 2. For example, the apparatus 1600 may correspond to the terminal device in Figure 2.

[0481] Exemplarily, the processing module 2420 is configured to send N t The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N in the set of observation data t Optionally, the device 2400 further includes a transceiver module 2410, and the processing module 2420 can be used to obtain N t The first data of the group is sent to the network device for compensating the predicted data corresponding to the t-th time unit, where the predicted data is obtained based on the p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

[0482] Optionally, the transceiver module 2410 is further configured to send a first message, where the first message is configured to indicate N t The observation data of the group is t The position in the set of observations.

[0483] Optionally, the transceiver module 2410 is further configured to receive second information, the second information being configured to indicate the N t The method of selecting the set of observation data, the N t The selection methods of group observation data include: selection based on predefined patterns, or selection based on predefined selection rules.

[0484] In one example, the second information is used to indicate the N t In the case where the set of observation data is selected based on the predefined pattern, the second information is also used to indicate the first pattern among the predefined multiple patterns. In another example, when the second information is used to indicate the N t When the set of observation data is selected based on the predefined selection rule, the second information is also used to indicate the selection of the N t The observation data of the group is in S t A feedback method for the position in the group observation data includes: feedback through a bitmap or feedback through a combination identifier.

[0485] Optionally, the transceiver module 2410 is further configured to send or receive third information, wherein the third information is configured to indicate the S tAt least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0486] Optionally, the S t The number of data included in any two groups of observation data in the group of observation data is equal; the transceiver module 2410 is also used to send or receive fourth information, the fourth information is used to indicate one or more of the following: the S t or the S t The number of observations in each set of observations.

[0487] Optionally, the transceiver module 2410 is further configured to send or receive eighth information, where the eighth information is configured to indicate N t The first data of the group is N t The set of observation data is still N t Group residual data.

[0488] Optionally, the transceiver module 2410 is further configured to send or receive seventh information, where the seventh information is configured to indicate whether the observation data set is divided into multiple groups of observation data.

[0489] Optionally, the transceiver module 2410 is further configured to receive sixth information indicating T time units. The tth time unit is one of the T time units, where T is an integer greater than 1 and the T time units are periodic, or T is 1.

[0490] Optionally, the transceiver module 2410 is also used to send or receive ninth information, which is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs model training.

[0491] A more detailed description of the above-mentioned transceiver module 2410 and processing module 2420 can be directly obtained by referring to the relevant description in the embodiment shown in Figure 2, and will not be repeated here.

[0492] It should be noted that the device 2400 may include a sending module or a receiving module, depending on whether the device 2400 performs the sending action and the receiving action in the above solution. It is understandable that since the device 2400 has a communication function, it can also be called a communication device.

[0493] Another possible design is that the apparatus 2400 is used to implement the functions of the network device in the method embodiment shown in Figure 2. For example, the apparatus 2400 may correspond to the network device in Figure 2.

[0494] For example, the transceiver module 2410 is used to receive Nt The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N in the set of observations t Optionally, the device 2400 further includes a processing module 2420, which can be used to determine the N t The first data of the group is used to compensate the predicted data corresponding to the t-th time unit to obtain compensated data, where the predicted data is obtained based on p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

[0495] Optionally, the transceiver module 2410 is further configured to receive first information, the first information being used to indicate N t The observation data of the group is t The position in the set of observations.

[0496] Optionally, the transceiver module 2410 is further configured to send a second message, the second message being configured to indicate the N t The method of selecting a group of observation data, the N t Methods for selecting a group of observation data include: selecting based on a predefined pattern, or selecting based on a predefined selection rule.

[0497] In one example, the second information is used to indicate the N t In the case where the set of observation data is selected based on the predefined pattern, the second information is also used to indicate the first pattern among the predefined multiple patterns. In another example, when the second information is used to indicate the N t When the set of observation data is selected based on the predefined selection rule, the second information is also used to indicate the selection of the N t The observation data of the group is in S t A feedback method for the position in the group observation data includes: feedback through a bitmap or feedback through a combination identifier.

[0498] Optionally, the transceiver module 2410 is further configured to send or receive third information, wherein the third information is configured to indicate the S t At least two of the following parameters of each set of observation data in the set of observation data: a starting position, an ending position or the number of included data in the set of observation data.

[0499] Optionally, the S t The number of data included in any two groups of observation data in the group of observation data is equal; the transceiver module 2410 is also used to send or receive fourth information, the fourth information is used to indicate one or more of the following: the S t or the S tThe number of observations in each set of observations.

[0500] Optionally, the transceiver module 2410 is further configured to send or receive eighth information, where the eighth information is configured to indicate N t The first data of the group is N t The set of observation data is still N t Group residual data.

[0501] Optionally, the transceiver module 2410 is further configured to send or receive seventh information, where the seventh information is configured to indicate whether the observation data set is divided into multiple groups of observation data.

[0502] Optionally, the transceiver module 2410 is further configured to send sixth information, where the sixth information is configured to indicate T time units. The tth time unit is one of the T time units, where T is an integer greater than 1, and the T time units are periodic, or T is 1.

[0503] Optionally, the transceiver module 2410 is also used to send or receive ninth information, which is used to indicate model training, and one or more of the following: the order of the prediction model, the number of training signals of the prediction model, or the device that performs model training.

[0504] A more detailed description of the above-mentioned transceiver module 2410 and processing module 2420 can be directly obtained by referring to the relevant description in the embodiment shown in Figure 2, and will not be repeated here.

[0505] It should be noted that the device 2400 may include a sending module or a receiving module, depending on whether the device 2400 performs the sending action and the receiving action in the above solution. It is understandable that since the data processing device 2400 has a communication function, it can also be called a communication device.

[0506] Figure 25 is another schematic block diagram of a data processing device provided in an embodiment of the present application. As shown in Figure 25, device 2500 includes one or more processors 2510. The processor 2510 can be a general-purpose processor or a dedicated processor. For example, it can be a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the device (such as a terminal device, network device, or chip), execute software programs, and process software program data.

[0507] Optionally, in one design, the processor 2510 may include a program (also referred to as code or instructions), which may be executed on the processor 2510 to cause the apparatus 2500 to perform the method performed by the terminal device or network device in the above method embodiment. In another possible design, the apparatus 2500 includes a circuit (not shown in FIG. 25 ) configured to implement the functions of the terminal device or network device in the above method embodiment.

[0508] Exemplarily, the processor 2510 may be configured to execute a computer program or instruction in a memory to implement the steps performed by a terminal device or a network device in the method embodiment shown in the embodiment shown in FIG. 2 .

[0509] Optionally, the device 2500 may include one or more memories 2520 on which programs (sometimes also referred to as codes or instructions) are stored. The programs can be run on the processor 2510, so that the device 2500 executes the method executed by the terminal device or network device in the above embodiment.

[0510] Optionally, the processor 2510 and / or the memory 2520 may include artificial intelligence (AI).

[0511] An AI (AI) module is provided, wherein the AI ​​module is configured to implement AI-related functions. The AI ​​module may be implemented using software, hardware, or a combination of software and hardware. For example, the AI ​​module may include a wireless intelligent controller (RIC) module. For example, the AI ​​module may be a near-real-time RIC or a non-real-time RIC.

[0512] Optionally, data may be stored in the processor 2510 and / or the memory 2520. The processor and memory may be provided separately or integrated together.

[0513] Optionally, the apparatus 2500 may further include a communication interface 2530. The processor 2510 may also be sometimes referred to as a processing unit, which controls the apparatus (e.g., a RAN node or terminal). The communication interface 2530 may also be sometimes referred to as a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, etc., which is used to implement the transceiver function of the apparatus.

[0514] Optionally, the apparatus 2500 further includes a communication interface 2530. The processor 2510 and the communication interface 2530 are coupled to each other. It is understood that the communication interface 2530 may be a transceiver or an input / output interface.

[0515] It can be understood that since the data processing device 2500 has a communication function, it can also be called a communication device.

[0516] When apparatus 2500 is used to implement the method shown in FIG2 , processor 2510 is used to perform the functions of the processing unit described above, and communication interface 2530 is used to perform the functions of the transceiver module described above. Whether communication interface 2530 is used for sending or receiving can be determined by whether it is used to perform a sending action or a receiving action in the solution implemented by apparatus 2500.

[0517] When the apparatus 2500 is a chip implemented in a terminal device, the chip implements the functions of the terminal device in the above-described method embodiments. The chip of the terminal device receives signals from other modules in the terminal device (such as a radio frequency module or antenna), which may be signals sent by a network device to the terminal device; or the chip of the terminal device sends signals to other modules in the terminal device (such as a radio frequency module or antenna), which may be signals sent by the terminal device to a network device.

[0518] When the apparatus 2500 is a chip used in a network device, the chip implements the functions of the network device in the above method embodiment. The chip of the network device receives a signal from another module in the network device (such as a radio frequency module or antenna), and the signal may be sent by the terminal device to the network device; or the chip of the network device sends a signal to another module in the network device (such as a radio frequency module or antenna), and the signal may be sent by the network device to the terminal device.

[0519] It is understood that when the apparatus 2500 is a terminal device or a network device, the communication interface 2530 may be a transceiver, specifically including a transmitter and a receiver, where the transmitter is used to transmit signals and the receiver is used to receive signals. When the apparatus 2500 is a chip used in a terminal device or a network device, the communication interface 2530 may be an input / output circuit, where the input circuit can be used for receiving and the output interface can be used for transmitting.

[0520] Figure 26 is a schematic diagram of the structure of a terminal device provided in an embodiment of the present application. As shown in Figure 26, the terminal device 2600 can be applied to the system shown in Figure 1 to perform the functions of the terminal device in the method embodiment shown in Figure 2. As shown in the figure, the terminal device 2600 includes a processor 2601 and a transceiver 2602. Optionally, the terminal device 2600 also includes a memory 2603. The processor 2601, the transceiver 2602, and the memory 2603 can communicate with each other through an internal connection path to transmit control and / or data signals. The memory 2603 is used to store a computer program, and the processor 2601 is used to call and run the computer program from the memory 2603 to control the transceiver 2602 to transmit and receive signals. Optionally, the terminal device 2600 may also include an antenna 2604 for transmitting the uplink data or uplink control signaling output by the transceiver 2602 via a wireless signal.

[0521] The processor 2601 and the memory 2603 may be combined into a processing device, and the processor 2601 is configured to execute the program code stored in the memory 2603 to implement the aforementioned functions. In a specific implementation, the memory 2603 may also be integrated into the processor 2601 or independent of the processor 2601. The processor 2601 may correspond to the processing module in FIG. 24 or the processor in FIG. 25 .

[0522] The transceiver 2602 may correspond to the transceiver module in FIG. 24 or the communication interface in FIG. 25 , and may also be referred to as a transceiver unit. The transceiver 2602 may include a receiver (or receiver, receiving circuit) and a transmitter (or transmitter, transmitting circuit). The receiver is used to receive signals, and the transmitter is used to transmit signals.

[0523] It should be understood that terminal device 2600 shown in FIG26 is capable of implementing the various processes involved in the terminal device in the method embodiment shown in FIG2 . The operations and / or functions of the various modules in terminal device 2600 are respectively for implementing the corresponding processes in the above method embodiment. For details, please refer to the description of the above method embodiment; to avoid repetition, detailed descriptions are omitted here.

[0524] The processor 2601 can be used to execute the actions implemented within the terminal device described in the previous method embodiments, while the transceiver 2602 can be used to execute the actions of the terminal device sending to or receiving from the network device described in the previous method embodiments. For details, please refer to the description of the previous method embodiments and will not be repeated here.

[0525] Optionally, the terminal device 2600 may further include a power supply 2605 for providing power to various devices or circuits in the terminal device.

[0526] In addition, in order to make the functions of the terminal device more complete, the terminal device 2600 can also include one or more of an input unit 2606, a display unit 2607, an audio circuit 2608, a camera 2609 and a sensor 2610, and the audio circuit can also include a speaker 2608a, a microphone 2608b, etc.

[0527] Figure 27 is a schematic diagram of the structure of a network device provided in an embodiment of the present application, for example, a base station. Base station 2700 can be used in the system shown in Figure 1 to perform the functions of the terminal device in the method embodiment shown in Figure 2. As shown in the figure, base station 2700 may include one or more of the following: one or more (DU+RU) units 2710 and one or more CUs 2720. CU 2720 can communicate with the next generation core (NG core). The DU may include at least one antenna 2711, at least one radio frequency unit 2712, at least one processor 2713, and at least one memory 2714. The DU portion is primarily used for transmitting and receiving radio frequency signals, converting radio frequency signals into baseband signals, and performing partial baseband processing. CU 2720 may include at least one processor 2722 and at least one memory 2721. CU 2720 and the DU may communicate via an interface. The control plane (CP) interface may be an Fs-C, such as F1-C, and the user plane (UP) interface may be an Fs-U, such as F1-U. The DU and RU can work together to implement the functions of the physical (PHY) layer. A DU can be connected to one or more RUs. The functions of the DU and RU can be configured in various ways according to the design. For example, the DU is configured to implement the baseband function, and the RU is configured to implement the mid-RF function. For another example, the DU is configured to implement the high-layer functions in the PHY layer, and the RU is configured to implement the low-layer functions and RF functions in the PHY layer. The high-layer functions in the PHY layer may include a part of the functions of the PHY layer, which is closer to the MAC layer, and the low-layer functions in the PHY layer may include another part of the functions of the PHY layer, which is closer to the mid-RF side.

[0528] The CU 2720 is primarily used for baseband processing and base station control. The DU and CU 2720 may be physically located together or physically separated, i.e., a distributed base station. The CU 2720 is the control center of the base station and may correspond to the processing module in FIG. 24 or the processor in FIG. 25 , and may also be referred to as a processing unit. It is primarily used to perform baseband processing functions. For example, the CU 2720 may be used to control the base station to execute the operational procedures for access network devices in the above-described method embodiments.

[0529] Specifically, baseband processing on the CU and DU can be divided according to the protocol layers of the wireless network. For example, the functions of the packet data convergence protocol (PDCP) layer and above are set in the CU, while the functions of the protocol layers below PDCP, such as the RLC layer and the MAC layer, are set in the DU. For another example, the CU implements the functions of the RRC layer and the PDCP layer, while the DU implements the functions of the RLC layer, the MAC layer, and the PHY layer.

[0530] In addition, optionally, the base station 2700 may include one or more radio frequency units (RUs), one or more DUs, and one or more CUs. The DU may include at least one processor 2713 and at least one memory 2714, the RU may include at least one antenna 2711 and at least one radio frequency unit 2712, and the CU may include at least one processor 2722 and at least one memory 2721.

[0531] In one example, the CU 2720 may be composed of one or more single boards, and the multiple single boards may jointly support a wireless access network with a single access indication (such as a 5G network), or may respectively support wireless access networks with different access standards (such as an LTE network, a 5G network, or other networks). The memory 2721 and the processor 2722 may serve one or more single boards. That is, a memory and a processor may be separately set on each single board. It is also possible that multiple single boards share the same memory and processor. In addition, necessary circuits may be provided on each single board. The DU may be composed of one or more single boards, and the multiple single boards may jointly support a wireless access network with a single access indication (such as a 5G network), or may respectively support wireless access networks with different access standards (such as an LTE network, a 5G network, or other networks). The memory 2714 and the processor 2713 may serve one or more single boards. That is, a memory and a processor may be separately set on each single board. It is also possible that multiple single boards share the same memory and processor. In addition, necessary circuits may be provided on each single board.

[0532] It should be understood that base station 2700 shown in Figure 27 is capable of implementing the various processes involving network devices in the method embodiment shown in Figure 2 . The operations and / or functions of the various modules in base station 2700 are respectively for implementing the corresponding processes in the aforementioned method embodiment. For details, please refer to the description of the aforementioned method embodiment; to avoid repetition, detailed descriptions are omitted here.

[0533] It should be understood that the base station 2700 shown in Figure 27 is only one possible architecture of a network device and should not constitute any limitation to this application. The method provided in this application is applicable to network devices of other architectures. For example, network devices including CU, DU, and AAU. This application does not limit the specific architecture of the network device.

[0534] It should be understood that FIG27 is merely an example and not a limitation, and the network device may not rely on the structure shown in FIG27. For example, the network device may include an AAU, a CU, and / or a DU, or a BBU and an adaptive radio unit (ARU). This application is not limited to this.

[0535] The CU and / or DU described above can be used to perform the actions implemented within the network device described in the previous method embodiments, while the AAU can be used to perform the actions described in the previous method embodiments in which the network device sends to or receives from the terminal device. For details, please refer to the description in the previous method embodiments and will not be repeated here.

[0536] It should be noted that the above method embodiments can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed by hardware integrated logic circuits in the processor or by software instructions.

[0537] The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0538] The steps of the method disclosed in the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0539] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0540] The present application also provides a chip system, which includes at least one processor for supporting the implementation of the functions of the terminal device or the network device involved in any one of the above method embodiments, for example, sending, receiving or processing the data and / or information involved in the above method.

[0541] In one possible design, the chip system further includes a memory, which is used to store program instructions and data, and the memory is located inside or outside the processor.

[0542] The chip system can be composed of chips, or can include chips and other discrete devices.

[0543] The present application also provides a computer program product, which includes: a computer program (also referred to as code, or instructions). When the computer program is run, the method executed by the terminal device in the embodiment shown in Figure 2 is executed, or the method executed by the network device is executed.

[0544] The present application also provides a computer-readable storage medium storing a computer program (also referred to as code or instructions). When the computer program is executed, the method executed by the terminal device in the embodiment shown in FIG2 is executed, or the method executed by the network device is executed.

[0545] The present application also provides a communication system, which includes the aforementioned terminal device and network device.

[0546] The methods provided in the above embodiments can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, they can be implemented in whole or in part in the form of a computer program product. The computer program product may include one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic disk), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0547] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0548] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0549] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0550] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0551] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0552] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk, or an optical disk.

Claims

1. A data processing method, characterized in that: include: Get N t The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N in the set of observations t Determined by group observation data; Send the N t The first data of the group, the N t The first group of data is used to compensate the predicted data corresponding to the t-th time unit, and the predicted data is obtained based on p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

2. The method according to claim 1, characterized in that The a is 1.

3. The method according to claim 1 or 2, characterized in that The N t The first group of data is used to compensate the predicted data corresponding to the t-th time unit, including: The N t The first group of data is used to determine the compensation data, and the compensation data is used to compensate the prediction data corresponding to the t-th time unit to obtain compensated data.

4. The method according to claim 3, characterized in that The compensated data is the sum of the predicted data and the compensated data corresponding to the t-th time unit.

5. The method according to claim 3 or 4, characterized in that The compensation data ΔV(t) satisfies: Wherein, K(t) is the Kalman gain matrix corresponding to the t-th time unit, For the N t The first data of the group is the predicted data corresponding to the t-th time unit, and O(t) is based on N t The observation data of the group is in the S t The position in the set of observations is determined.

6. The method according to any one of claims 1 to 5, characterized in that The N t The observation data of the group is in the S t The positions in the set of observations are determined based on a predefined first pattern; or The N t The observation data of the group is in the S t The positions in the set of observations are determined based on a predefined first selection rule.

7. The method according to claim 6, characterized in that, The N t The observation data of the group is in the S t The position in the set of observation data is determined based on the first selection rule; The method further comprises: Sending first information; the first information is used to indicate the N t The observation data of the group is in the S t The position of the group of observations.

8. The method according to claim 6 or 7, characterized in that The method further comprises: Receive second information, wherein the second information is used to indicate the N t The method of selecting a group of observation data, the N t The selection method of each group of observation data includes: selection based on a predefined pattern, or selection based on a predefined selection rule.

9. The method according to any one of claims 1 to 8, characterized in that The N t The first data set is the N t set of observational data; or, The N t The first data of the group is N t The residual data of the group, N t Data in group residual data For the N t Data of group observation data With N t Data in group prediction data The difference; among them, Indicates the N t The i-th data in the n-th group of residual data, Indicates the N t The i-th data in the n-th group of observation data, Indicates the N t The i-th data in the n-th group of prediction data, the N t The group prediction data is S t The nth group of prediction data is part of the data in the N t The position of the group of predicted data in the Nth group of observed data t The observation data of the group are in the same position, and the S t The group prediction data includes the prediction data corresponding to the t-th time unit; 1≤i≤I n , 1≤n≤N t , I n represents the number of data included in the nth group of residual data, i, I n and n are both positive integers.

10. The method according to any one of claims 1 to 9, characterized in that The S t Any two sets of observation data in the set of observation data include the same number of data; or, The S t The group observation data is obtained by dividing the observation data set based on at least two pre-configured parameters as follows: t The starting position, the ending position or the number of data included in each group of observation data in the group of observation data in the observation data set, wherein the observation data set is the set of observation data corresponding to the t-th time unit.

11. The method according to claim 10, characterized in that The method further comprises: Send or receive third information, the third information is used to indicate the S t At least two of the following parameters of each set of observation data in the set of observation data: the starting position, the ending position or the number of data included in the set of observation data; or The S t Any two groups of observation data in the groups of observation data include an equal number of data; the method further includes: Send or receive fourth information, the fourth information is used to indicate one or more of the following: t or the S t The number of observations in each set of observations.

12. The method according to any one of claims 1 to 11, characterized in that The t-th time unit is one of T time units, wherein T is an integer greater than 1, and the T time units are periodic, or T is 1; The method further comprises: Sixth information is received, where the sixth information is used to indicate the T time units.

13. The method according to any one of claims 1 to 12, characterized in that The predicted data is obtained through autoregression AR prediction.

14. A data processing method, characterized in that: include: Receive N t The first data of the group, the N t The first data of the group is based on the S corresponding to the tth time unit t N in the set of observations t Determined by group observation data; Based on the N t The first data of the group is used to compensate the predicted data corresponding to the t-th time unit to obtain compensated data; the predicted data is obtained based on p predicted data corresponding to the (t-p-a+1)-th time unit to the (t-a)-th time unit, where p and a are constants.

15. The method of claim 14, wherein: The a is 1.

16. The method according to claim 14 or 15, characterized in that Based on the N t The first data is grouped, and the predicted data corresponding to the t-th time unit is compensated to obtain compensated data, including: The N t The first data of the group determines the compensation data; Compensated data is obtained based on the compensation data and the prediction data corresponding to the t-th time unit.

17. The method according to claim 16, characterized in that The compensated data is the sum of the predicted data corresponding to the t-th time unit and the compensated data.

18. The method according to claim 16 or 17, characterized in that The compensation data ΔV(t) satisfies: Wherein, K(t) is the Kalman gain matrix corresponding to the t-th time unit, For the N t The first data of the group is the predicted data corresponding to the t-th time unit, and O(t) is based on N t The observation data of the group is in the S t The position in the set of observations is determined.

19. The method according to any one of claims 14 to 18, characterized in that The method further comprises: Receive first information, the first information is used to indicate the N t The observation data of the group is in the S t The position of the group of observations.

20. The method of claim 19, wherein: The method further comprises: Sending second information, wherein the second information is used to indicate the N t The method of selecting a group of observation data, the N t The selection method of the group observation data includes: selection based on a predefined pattern, or selection based on a predefined selection rule.

21. The method according to any one of claims 14 to 20, characterized in that The N t The first data set is the N t set of observational data; or, The N t The first data of the group is N t The residual data of the group, N t Data in group residual data For the N t Data of group observation data With N t Data in group prediction data The difference; among them, Indicates the N t The i-th data in the n-th group of residual data, Indicates the N t The i-th data in the n-th group of observation data, Indicates the N t The i-th data in the n-th group of prediction data, the N t The group prediction data is S t The nth group of prediction data is part of the data in the N t The position of the group of predicted data in the Nth group of observed data t The observation data of the group are in the same position, and the S t The group prediction data includes the prediction data corresponding to the t-th time unit; 1≤i≤I n , 1≤n≤N t , I n represents the number of data included in the nth group of residual data, i, I n and n are both positive integers.

22. The method according to any one of claims 14 to 21, characterized in that The S t Any two sets of observation data in the set of observation data include the same number of data; or, The S t The group observation data is obtained by dividing the observation data set based on at least two pre-configured parameters as follows: t The starting position, the ending position or the number of data included in each set of observation data in the set, wherein the observation data set is the set of observation data corresponding to the t-th time unit.

23. The method of claim 22, wherein: The method further comprises: Send or receive third information, the third information is used to indicate the S t At least two of the following parameters of each set of observation data in the set of observation data: the starting position, the ending position or the number of data included in the set of observation data; or The S t Any two of the groups of observations include an equal number of data; and The method further comprises: Send or receive fourth information, the fourth information is used to indicate one or more of the following: t or the S t The number of observations in each set of observations.

24. The method according to any one of claims 14 to 23, characterized in that The t-th time unit is one of T time units, where T is an integer greater than 1, and the T time units are periodic, or T is 1; The method further comprises: Sixth information is received, where the sixth information is used to indicate the T time units.

25. The method according to any one of claims 14 to 24, characterized in that The predicted data is obtained through autoregression AR prediction.

26. A data processing device, characterized in that: comprising a processor and a memory, wherein: The memory is used to store computer programs; The processor is configured to call the computer program so that the method according to any one of claims 1 to 13 is executed, or the method according to any one of claims 14 to 25 is executed.

27. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 13 is executed, or the method according to any one of claims 14 to 25 is executed.

28. A computer program product, characterized in that The invention comprises a computer program, which, when being executed, causes the method according to any one of claims 1 to 13 to be performed, or causes the method according to any one of claims 14 to 25 to be performed.

Images