Method, apparatus and related device for adjusting beam measurement reporting quantity

Abstract

The application provides a method and device for adjusting the number of beam measurement reports and related equipment, and relates to the technical field of artificial intelligence. After a first communication device receives feedback information fed back by a second communication device based on N beams at a first time, the application determines the minimum number M of reference signal receiving powers that need to be included in the feedback information fed back by the second communication device according to a set error tolerance value and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal receiving powers in subsequent feedback, thereby reducing the cost of resources.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and specifically to a method, apparatus, and related equipment for adjusting the number of beam measurement reports. Background Technology

[0002] Artificial intelligence (AI) technology has revolutionized the wireless network field, effectively overcoming the technical limitations of traditional networks and achieving reduced operating costs and improved network performance. Currently, AI and machine learning (ML) technologies are applied to beam management to reduce system overhead and latency, and improve the accuracy of beam selection. Generally, taking spatial downlink beam prediction as an example, during the training phase, the network (NW) scans the beams, and the user equipment (UE) provides the reference signal received power (RSRP). The network-side model determines the optimal beam ID. After training, the network only sends sparse beams for scanning. The UE provides the RSRP of the scanned small number of beams, and the network performs beam inference based on the AI ​​model, sending the predicted Top-K beams to the UE for scanning. The UE then scans these beams to ultimately determine the optimal beam ID and reports it to the network. In related technologies, the number of beam measurements reported by the UE is set to a fixed value. However, setting the number of beam measurements reported to a fixed value has limitations, resulting in high resource overhead. Summary of the Invention

[0003] This invention provides a method, apparatus, and related equipment for adjusting the number of beam measurement reports, which solves the problem of high resource overhead when reporting beam measurement results on a second communication device in the prior art.

[0004] To solve the above problems, the present invention is implemented as follows:

[0005] In a first aspect, this application provides a method for adjusting the number of beam measurement reports, applied to a first communication device, the method comprising:

[0006] After the first communication device sends K beams to the second communication device, it receives feedback information reported by the second communication device at a first moment. The feedback information includes N reference signal received powers from K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams. K is a positive integer and N is a positive integer less than or equal to K.

[0007] Obtain a set error tolerance value, wherein the set error tolerance value is used to represent the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation;

[0008] Based on the N reference signal received power and the set error tolerance value, determine the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device, where M is a positive integer less than or equal to K;

[0009] The M is sent to the second communication device.

[0010] Optionally, after the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at a first moment, the method further includes:

[0011] Obtain a training dataset, which includes X training samples. Each training sample includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the best beam. The best beam is the beam with the strongest reference signal receiving power between the second communication device and the first communication device. X is a positive integer.

[0012] The training dataset is augmented to obtain Y training samples, where Y is a positive integer greater than X;

[0013] The initial machine learning model deployed on the first communication device is trained based on the Y training samples to obtain the machine learning model.

[0014] Optionally, after sending the M to the second communication device, the method further includes:

[0015] The received power of M reference signals sent by the second communication device at a second time, where the second time is a time after the first time;

[0016] The received power of the M reference signals is input into the machine learning model to obtain the probability value of each beam in all beams transmitted between the first communication device and the second communication device. The probability value is used to represent the probability that each beam in all beams transmitted between the first communication device and the second communication device is the optimal beam.

[0017] Optionally, determining the minimum number M of reference signal received power to be included in the feedback information reported by the second communication device based on the N reference signal received power and the set error tolerance value includes:

[0018] Based on the N reference signal received power and the set error tolerance value, a query is performed in the target lookup table to obtain the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device;

[0019] The target lookup table includes multiple reference signal received power groups, multiple different error tolerance values, and the minimum number of reference signal received powers corresponding to the multiple reference signal received power groups and the multiple different error tolerance values, wherein each of the multiple reference signal received power groups includes at least one reference signal received power.

[0020] Optionally, before sending M to the second communication device, the method further includes:

[0021] Based on historical feedback information, determine the minimum number J of reference signal received power that needs to be included in the feedback information reported by the second communication device. The historical feedback information includes the minimum number J of reference signal received power that needs to be reported sent by the first communication device to the second communication device in the last time, where J is a positive integer.

[0022] Optionally, obtaining the set error tolerance value includes:

[0023] Receive the set error tolerance value sent by the second communication device;

[0024] or,

[0025] The error tolerance value preset by the first communication device is determined as the set error tolerance value.

[0026] Secondly, this application provides a device for adjusting the number of beam measurement reports, applied to a first communication device, the device comprising:

[0027] The receiving module is configured to receive feedback information reported by the second communication device at a first moment after the first communication device sends K beams to the second communication device. The feedback information includes N reference signal received powers among K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams, where K is a positive integer and N is a positive integer less than or equal to K.

[0028] The acquisition module is used to acquire a set error tolerance value, which represents the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation.

[0029] The determining module is used to determine, based on the N reference signal received power and the set error tolerance value, the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device, wherein M is a positive integer less than or equal to K;

[0030] A sending module is used to send the M to the second communication device.

[0031] Thirdly, this application also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in the first aspect above.

[0032] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect above.

[0033] Fifthly, this application also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the method described in the first aspect above.

[0034] This application provides a method, apparatus, and related equipment for adjusting the number of beam measurement reports, relating to the field of artificial intelligence technology, and applied to a first communication device. The method includes: after the first communication device sends K beams to a second communication device, receiving feedback information reported by the second communication device at a first moment, the feedback information including N reference signal received powers from K reference signal received powers obtained by the second communication device after receiving the K beams, where K is a positive integer and N is a positive integer less than or equal to K; obtaining a set error tolerance value, the set error tolerance value representing the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation; determining, based on the N reference signal received powers and the set error tolerance value, the minimum number M of reference signal received powers that need to be included in the feedback information reported by the second communication device, where M is a positive integer less than or equal to K; and sending M to the second communication device. The technical solution of this application, after the first communication device receives the feedback information from the second communication device based on N beams at the first moment, determines the minimum number M of reference signal received power that the second communication device needs to include in the feedback information according to the set error tolerance limit and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal received power in subsequent feedback, thereby reducing resource overhead. Attached Figure Description

[0035] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart illustrating a method for adjusting the number of beam measurement reports provided in an embodiment of this application;

[0037] Figure 2 A typical process for beam management of the AI ​​model provided in the embodiments of this application on the network side;

[0038] Figure 3 A schematic diagram of a beam measurement reporting quantity adjustment device provided in this application embodiment;

[0039] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0041] The terms "first," "second," etc., used in the embodiments of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices. Additionally, the use of "and / or" in this application indicates at least one of the connected objects, such as A and / or B and / or C, representing seven possibilities: including A alone, B alone, C alone, and the presence of both A and B, both B and C, both A and C, and the presence of A, B, and C.

[0042] See Figure 1 , Figure 1 This is a flowchart illustrating the method for adjusting the number of beam measurement reports provided in an embodiment of this application. Figure 1As shown, the method for adjusting the number of beam measurement reports applied to the first communication device may include the following steps:

[0043] Step 101: After the first communication device sends K beams to the second communication device, the feedback information reported by the second communication device at a first moment is received. The feedback information includes N reference signal received powers among the K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams. K is a positive integer and N is a positive integer less than or equal to K.

[0044] In this embodiment, as Figure 2 As shown, the first communication device is in Figure 2 The explanation is based on the network terminal (base station) of China and Israel. The second communication device is in Figure 2 The explanation will focus on the user terminal. Figure 2 This is a typical process for AI models to perform beam management on the network side. Taking spatial downlink beam prediction as an example, during the training phase, the network measures the beam, the UE feeds back RSRP, and the network-side model determines the optimal beam ID.

[0045] After training, the network sends only sparse beams for measurement. The UE feeds back the RSRP of a small number of scanned beams. The network performs beam inference based on the AI ​​model and sends the predicted Top-K beams to the UE for measurement. The UE then measures these beams and finally determines the best beam ID and reports it to the network.

[0046] In this embodiment, step 101 is the process of the first communication device sending a full-beam scan to the second communication device. Specifically, the first communication device sends K beams to the second communication device, the second communication device measures the K beams to obtain the received power of K N reference signals, selects the received power of the N reference signals, and then sends feedback information to the first communication device. The feedback information includes the received power of the N reference signals out of the K received power obtained after measuring the K beams.

[0047] Specifically, reference signal received power is an important indicator used to evaluate signal quality in wireless communication systems (especially LTE and 5G networks). Reference signal received power represents the power value of the reference signal received by the receiver at a specific frequency, usually expressed in decibels and milliwatts (dBm).

[0048] Step 102: Obtain the set error tolerance value, which represents the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed on the first communication device during operation.

[0049] In this embodiment, the set error tolerance limit is a predefined value. Specifically, the set error tolerance limit represents the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed in the first communication device during operation. That is, it represents the minimum requirement of the machine learning model (AI model) deployed in the first communication device for the amount of received power of the input reference signal. It should be noted that the set error tolerance limit can be adaptively adjusted according to the actual situation of the AI ​​model, and is not specifically limited in this embodiment.

[0050] Step 103: Based on the N reference signal received power and the set error tolerance value, determine the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device, where M is a positive integer less than or equal to K.

[0051] In this embodiment, the first communication device determines the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device by means of looking up a table, calculation or inputting into a machine learning model, based on the N reference information received power sent by the second communication device and the set error tolerance value. When the second communication device measures the N beams next time, it can only feed back the received power of M reference signals, thereby achieving the purpose of saving resources.

[0052] It should be noted that in this embodiment, M is a positive integer less than or equal to K. That is, in some cases, M can be equal to K. In this case, it indicates that the second communication device needs to feed back all the reference signal receiving power corresponding to all K beams to the first communication device.

[0053] Step 104: Send M to the second communication device.

[0054] In this embodiment, after determining the specific value of M, M is sent to the second communication device. During the subsequent feedback process from the second communication device to the first communication device, only M reference signal received power values ​​need to be fed back, thus achieving dynamic adjustment of the RSRP M value that the UE needs to report. This method can flexibly determine the number of RSRPs to be reported based on the inference requirements of the network-side model and the characteristics of the actual measurement data, effectively avoiding the resource waste or insufficient information problems that may be caused by a fixed reporting number in the prior art.

[0055] The technical solution of this application, after the first communication device receives the feedback information from the second communication device based on N beams at the first moment, determines the minimum number M of reference signal received power that the second communication device needs to include in the feedback information according to the set error tolerance limit and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal received power in subsequent feedback, thereby reducing resource overhead.

[0056] Optionally, after the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at a first moment, the method further includes:

[0057] Obtain a training dataset, which includes X training samples. Each training sample includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the best beam. The best beam is the beam with the strongest reference signal receiving power between the second communication device and the first communication device. X is a positive integer.

[0058] The training dataset is augmented to obtain Y training samples, where Y is a positive integer greater than X;

[0059] The initial machine learning model deployed on the first communication device is trained based on the Y training samples to obtain the machine learning model.

[0060] In this embodiment, before receiving feedback information reported by the second communication device, a training dataset needs to be prepared to train the machine learning model deployed on the first communication device. Specifically, the training dataset includes X training samples, each of which includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the optimal beam, which is the beam with the best transmission effect between the second communication device and the first communication device.

[0061] Specifically, the initial training set is denoted as ,in Indicates the first Initial training data, . Indicates the first Input of initial training data, Indicates the first The labels corresponding to the initial training data, among which The number of beams in a set of beams transmitted by the first communication device. This refers to the number of beams in all downlink transmit beams of the first communication device. More specifically, the... In the initial training data, Indicates the UE measurement of the first RSRP of 1 beam, of which And there are . This represents the first downlink transmit beam of the first communication device. The probability that a beam is the optimal beam. .

[0062] In this embodiment, after obtaining the training dataset, the training dataset is augmented to generate Y training samples. It should be noted that Y is a positive integer greater than X. Specifically, during the model training phase, the first communication device first performs an augmentation operation on the initial training set, that is, augments each initial training data... Expand into The training data, where the first... The input of each training data point is overwritten from the end. One training coverage value, With the corresponding labels unchanged, we obtain the augmented training set:

[0063] ;

[0064] in, For the first Initial training data The augmented training subset obtained by expansion For example, when the training coverage value is 0, the first... Initial training data It is expanded to the following augmented training subset:

[0065] ;

[0066] in:

[0067] .

[0068] After obtaining Y training datasets, the initial machine learning model deployed on the first communication device is trained using the Y training samples to obtain the machine learning model. Specifically, the neural network model of the first communication device utilizes the augmented training set. During training, the neural network model uses a loss function based on probability distribution metrics, such as cross-entropy, i.e.:

[0069] ;

[0070] in This represents the first downlink transmit beam of the first communication device. The probability that the nth beam is the optimal beam (when the nth beam is the optimal beam) When each beam is the optimal beam ,otherwise ), .use Represents the output of the neural network model, where This represents the first downlink transmit beam of the predicted first communication device. The probability that a beam is the optimal beam. .

[0071] In this embodiment, by expanding the diversity and coverage of training data, the generalization ability and inference accuracy of the neural network model are improved. This approach helps the model maintain stable performance in complex and ever-changing communication environments.

[0072] Optionally, after sending the M to the second communication device, the method further includes:

[0073] The received power of M reference signals sent by the second communication device at a second time, where the second time is a time after the first time;

[0074] The received power of the M reference signals is input into the machine learning model to obtain the probability value of each beam in all beams transmitted between the first communication device and the second communication device. The probability value is used to represent the probability that each beam in all beams transmitted between the first communication device and the second communication device is the optimal beam.

[0075] In this embodiment, the first communication device sets the minimum number of RSRPs that the UE needs to report. After instructing the UE, the UE receives the M reference signal reception power sent by the second communication device at the second time, that is, the UE reports the second time to the first communication device. (Time) The first set of RSRP measurement results of the first communication device transmitting a group of beams One measurement result, namely ,in .

[0076] Therefore, the received power of M reference signals is input into the machine learning model for identification, that is, the first communication device receives the measurement data reported from the UE side. Then, the measurement data reported by the UE side Perform interpolation to insert at the end of the vector. Given a set of interpolated measurements (interpolated values ​​can be the median, mean, minimum, or 0 of the measurements), inference data is obtained. For example, when the measurement interpolation value is 0, we get:

[0077] ;

[0078] Inference data As input to the neural network model, the output of the neural network model is obtained, which is the probability that each of the downlink transmit beams on the first communication device side is the optimal beam.

[0079] In this embodiment, by adaptively adjusting the number of reports, the network side can obtain only the RSRP measurement results that are most valuable to model inference, thereby minimizing the amount of data transmission and reducing system resource consumption while ensuring model performance.

[0080] In this embodiment, the minimum number of valid RSRP measurements is selected. This method helps improve the inference accuracy of the network-side model with limited information, thereby better supporting beam management decisions.

[0081] Optionally, determining the minimum number M of reference signal received power to be included in the feedback information reported by the second communication device based on the N reference signal received power and the set error tolerance value includes:

[0082] Based on the N reference signal received power and the set error tolerance value, a query is performed in the target lookup table to obtain the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device;

[0083] The target lookup table includes multiple reference signal received power groups, multiple different error tolerance values, and the minimum number of reference signal received powers corresponding to the multiple reference signal received power groups and the multiple different error tolerance values, wherein each of the multiple reference signal received power groups includes at least one reference signal received power.

[0084] In this embodiment, the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device is determined by looking up a table.

[0085] Specifically, a lookup table is pre-stored in the first communication device. The lookup table includes the input of each training data in the initial training set, different error tolerance limits, and the corresponding data input and the minimum number of RSRPs that the UE needs to report under the error tolerance limit. For example, as shown in Table 1, the row name is the input of each training data in the initial training set, the column name is the different error tolerance limits, and the corresponding data is the minimum number of RSRPs that the UE needs to report.

[0086] Table 1. Query table stored in the first communication device

[0087] In this table, the parameter value in the i-th row and j-th column represents the minimum number of reference signal received powers corresponding to the i-th reference signal received power and the j-th error tolerance limit. For example, the first row and first column represents the minimum number of reference signal received powers when the error tolerance limit is set to η1. .

[0088] Therefore, the first communication device, based on the received... RSRP measurement results at time and the set error tolerance Select the corresponding data from the query table as the minimum number of RSRPs that the UE needs to report. .

[0089] When looking up a table, various query methods can be included. For example, selecting one or more values ​​from the query table, such as selecting a single value. Another example is selecting data, such as searching for data in the query table shown in Table 1 that matches the values ​​reported by the UE. RSRP measurement results at time The row name with the smallest mean square error (MSE) ,in Then based on column name Find the corresponding data as the minimum amount of RSRP that the UE needs to report. .

[0090] This embodiment establishes a lookup table based on prediction error and error tolerance. This proposal ensures that, under certain prediction error conditions, the minimum number of valid RSRP measurement results is selected. This method helps improve the inference accuracy of the network-side model with limited information, thereby better supporting beam management decisions.

[0091] Optionally, after the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at a first moment, the method further includes:

[0092] Based on historical feedback information, determine the minimum number J of reference signal received power that needs to be included in the feedback information reported by the second communication device. The historical feedback information includes the minimum number J of reference signal received power that needs to be reported sent by the first communication device to the second communication device in the last time, where J is a positive integer.

[0093] In this embodiment, when determining the minimum amount J of the reference signal received power that needs to be included in the feedback information, the minimum amount J of the reference signal received power can be directly determined through historical feedback information.

[0094] Specifically, if the UE has not previously received the minimum number of RSRPs that the UE needs to report from the first communication device, the UE reports... All RSRP measurement results at any given time, i.e. ,in .

[0095] The minimum number of RSRPs that the UE needs to report if it previously received an instruction from the first communication device. UE reporting The preceding time in the RSRP measurement results One measurement result, namely ,in .

[0096] In this embodiment, the temporal correlation of the wireless channel is utilized to predict and guide the setting of the M value at the current moment based on the RSRP measurement results reported by the UE at the previous moment. This intelligent reporting strategy based on historical data can further improve the rationality of the reporting quantity, reduce frequent adjustments to the reporting quantity due to environmental changes, and improve the stability and robustness of the system.

[0097] Optionally, obtaining the set error tolerance value includes:

[0098] Receive the set error tolerance value sent by the second communication device;

[0099] or,

[0100] The error tolerance value preset by the first communication device is determined as the set error tolerance value.

[0101] In this embodiment, the error tolerance value can be sent from the second communication device to the first communication device, that is, the first communication device receives the error tolerance value from the UE. For example, it can be represented by 1 bit, where "0" corresponds to a low fault tolerance limit. "1" corresponds to the high error tolerance limit. The error tolerance value can also be preset by the first communication device if the second communication device does not send it to the first communication device. In actual use, the error tolerance value can be determined according to the actual situation, and no specific limitation is made in this embodiment.

[0102] The technical solution of this application, after the first communication device receives the feedback information from the second communication device based on N beams at the first moment, determines the minimum number M of reference signal received power that the second communication device needs to include in the feedback information according to the set error tolerance limit and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal received power in subsequent feedback, thereby reducing resource overhead.

[0103] See Figure 3 , Figure 3 This is a structural diagram of the beam measurement reporting quantity adjustment device provided in the embodiments of this application. (See diagram below.) Figure 3 As shown, the beam measurement reporting quantity adjustment device 300 includes:

[0104] The receiving module 310 is configured to receive feedback information reported by the second communication device at a first moment after the first communication device sends K beams to the second communication device. The feedback information includes N reference signal received powers among the K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams. K is a positive integer and N is a positive integer less than or equal to K.

[0105] The acquisition module 320 is used to acquire a set error tolerance limit value, wherein the set error tolerance limit value is used to represent the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation.

[0106] The determining module 330 is used to determine, based on the N reference signal received power and the set error tolerance value, the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device, wherein M is a positive integer less than or equal to K;

[0107] The sending module 340 is used to send the M to the second communication device.

[0108] Optional, also includes:

[0109] The sample acquisition module is used to acquire a training dataset, which includes X training samples. Each training sample includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the best beam. The best beam is the beam with the strongest reference signal receiving power between the second communication device and the first communication device. X is a positive integer.

[0110] An augmentation module is used to perform augmentation operations on the training dataset to obtain Y training samples, where Y is a positive integer greater than X;

[0111] The training module is used to train the initial machine learning model deployed on the first communication device based on the Y training samples to obtain the machine learning model.

[0112] Optional,

[0113] The receiving module 310 is also configured to receive the received power of M reference signals sent by the second communication device at a second time, wherein the second time is a time after the first time.

[0114] The input module is used to input the received power of the M reference signals into the machine learning model to obtain the probability value corresponding to each beam in all beams transmitted between the first communication device and the second communication device. The probability value is used to represent the probability that each beam in all beams transmitted between the first communication device and the second communication device is the optimal beam.

[0115] Optionally, the determining module 330 includes:

[0116] The query submodule is used to query the target query table based on the N reference signal received power and the set error tolerance value to obtain the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device;

[0117] The target lookup table includes multiple reference signal received power groups, multiple different error tolerance values, and the minimum number of reference signal received powers corresponding to the multiple reference signal received power groups and the multiple different error tolerance values, wherein each of the multiple reference signal received power groups includes at least one reference signal received power.

[0118] Optional, also includes:

[0119] The historical information determination module is used to determine, based on historical feedback information, the minimum number J of reference signal received power that needs to be included in the feedback information reported by the second communication device. The historical feedback information includes the minimum number J of reference signal received power that needs to be reported sent by the first communication device to the second communication device in the last instance, where J is a positive integer.

[0120] Optionally, the acquisition module 320 includes:

[0121] A receiving submodule is used to receive the set error tolerance value sent by the second communication device;

[0122] or,

[0123] The second determining submodule is used to determine the preset error tolerance value of the first communication device as the set error tolerance value.

[0124] The technical solution of this application, after the first communication device receives the feedback information from the second communication device based on N beams at the first moment, determines the minimum number M of reference signal received power that the second communication device needs to include in the feedback information according to the set error tolerance limit and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal received power in subsequent feedback, thereby reducing resource overhead.

[0125] This application also provides an electronic device. Please refer to [link to relevant documentation]. Figure 4 The electronic device may include a processor 401, a memory 402, and a program 4021 stored in the memory 402 and executable on the processor 401.

[0126] When program 4021 is executed by processor 401, it can achieve the following: Figure 1 Any step in the corresponding method embodiment:

[0127] After the first communication device sends K beams to the second communication device, it receives feedback information reported by the second communication device at a first moment. The feedback information includes N reference signal received powers from K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams. K is a positive integer and N is a positive integer less than or equal to K.

[0128] Obtain a set error tolerance value, wherein the set error tolerance value is used to represent the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation;

[0129] Based on the N reference signal received power and the set error tolerance value, determine the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device, where M is a positive integer less than or equal to K;

[0130] The M is sent to the second communication device.

[0131] Optionally, after the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at a first moment, the method further includes:

[0132] Obtain a training dataset, which includes X training samples. Each training sample includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the best beam. The best beam is the beam with the strongest reference signal receiving power between the second communication device and the first communication device. X is a positive integer.

[0133] The training dataset is augmented to obtain Y training samples, where Y is a positive integer greater than X;

[0134] The initial machine learning model deployed on the first communication device is trained based on the Y training samples to obtain the machine learning model.

[0135] Optionally, after sending the M to the second communication device, the method further includes:

[0136] The received power of M reference signals sent by the second communication device at a second time, where the second time is a time after the first time;

[0137] The received power of the M reference signals is input into the machine learning model to obtain the probability value of each beam in all beams transmitted between the first communication device and the second communication device. The probability value is used to represent the probability that each beam in all beams transmitted between the first communication device and the second communication device is the optimal beam.

[0138] Optionally, determining the minimum number M of reference signal received power to be included in the feedback information reported by the second communication device based on the N reference signal received power and the set error tolerance value includes:

[0139] Based on the N reference signal received power and the set error tolerance value, a query is performed in the target lookup table to obtain the minimum number M of reference signal received power that needs to be included in the feedback information reported by the second communication device;

[0140] The target lookup table includes multiple reference signal received power groups, multiple different error tolerance values, and the minimum number of reference signal received powers corresponding to the multiple reference signal received power groups and the multiple different error tolerance values, wherein each of the multiple reference signal received power groups includes at least one reference signal received power.

[0141] Optionally, before sending M to the second communication device, the method further includes:

[0142] Based on historical feedback information, determine the minimum number J of reference signal received power that needs to be included in the feedback information reported by the second communication device. The historical feedback information includes the minimum number J of reference signal received power that needs to be reported sent by the first communication device to the second communication device in the last time, where J is a positive integer.

[0143] Optionally, obtaining the set error tolerance value includes:

[0144] Receive the set error tolerance value sent by the second communication device;

[0145] or,

[0146] The error tolerance value preset by the first communication device is determined as the set error tolerance value.

[0147] The technical solution of this application, after the first communication device receives the feedback information from the second communication device based on N beams at the first moment, determines the minimum number M of reference signal received power that the second communication device needs to include in the feedback information according to the set error tolerance limit and the feedback information, and then sends M to the second communication device. The second communication device only needs to feed back M reference signal received power in subsequent feedback, thereby reducing resource overhead.

[0148] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described embodiment for adjusting the number of beam measurement reports, achieving the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0149] This application also provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described method embodiment for adjusting the number of beam measurement reports, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0150] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0151] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a second communication device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0152] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for adjusting the number of beam measurement reports, characterized in that, Applied to a first communication device, the method includes: After the first communication device sends K beams to the second communication device, it receives feedback information reported by the second communication device at a first moment. The feedback information includes N reference signal received powers from K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams. K is a positive integer and N is a positive integer less than or equal to K. Obtain a set error tolerance value, wherein the set error tolerance value is used to represent the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation; Based on the N reference signal received powers and the set error tolerance values, determine the minimum number M of reference signal received powers that need to be included in the feedback information reported by the second communication device, where M is a positive integer less than or equal to K. This determination includes: querying a target lookup table based on the N reference signal received powers and the set error tolerance values ​​to obtain the minimum number M of reference signal received powers that need to be included in the feedback information reported by the second communication device; wherein the target lookup table includes multiple reference signal received power groups, multiple different set error tolerance values, and the minimum number of reference signal received powers corresponding to each of the multiple reference signal received power groups and the multiple different set error tolerance values, and each of the multiple reference signal received power groups includes at least one reference signal received power. The M is sent to the second communication device.

2. The method according to claim 1, characterized in that, After the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at the first moment, the method further includes: Obtain a training dataset, which includes X training samples. Each training sample includes a set of beam data and a label. The label is used to indicate the probability that the corresponding beam data is the best beam. The best beam is the beam with the strongest reference signal receiving power between the second communication device and the first communication device. X is a positive integer. The training dataset is augmented to obtain Y training samples, where Y is a positive integer greater than X; The initial machine learning model deployed on the first communication device is trained based on the Y training samples to obtain the machine learning model.

3. The method according to claim 2, characterized in that, After sending M to the second communication device, the method further includes: The received power of M reference signals sent by the second communication device at a second time, where the second time is a time after the first time; The received power of the M reference signals is input into the machine learning model to obtain the probability value of each beam in all beams transmitted between the first communication device and the second communication device. The probability value is used to represent the probability that each beam in all beams transmitted between the first communication device and the second communication device is the optimal beam.

4. The method according to claim 1, characterized in that, After the first communication device sends K beams to the second communication device and before receiving the feedback information reported by the second communication device at the first moment, the method further includes: Based on historical feedback information, determine the minimum number J of reference signal received power that needs to be included in the feedback information reported by the second communication device. The historical feedback information includes the minimum number J of reference signal received power that needs to be reported sent by the first communication device to the second communication device in the last time, where J is a positive integer.

5. The method according to any one of claims 1-4, characterized in that, The process of obtaining the set error tolerance value includes: Receive the set error tolerance value sent by the second communication device; or, The error tolerance value preset by the first communication device is determined as the set error tolerance value.

6. A device for adjusting the number of beam measurement reports, applied to a first communication device, characterized in that, The device includes: The receiving module is configured to receive feedback information reported by the second communication device at a first moment after the first communication device sends K beams to the second communication device. The feedback information includes N reference signal received powers among K reference signal received powers obtained by the second communication device after receiving the K beams and measuring the K beams, where K is a positive integer and N is a positive integer less than or equal to K. The acquisition module is used to acquire a set error tolerance value, which represents the upper limit of the absolute value of the difference between the prediction result and the actual result of the machine learning model deployed by the first communication device during operation. A determining module is configured to determine, based on the N reference signal received powers and the set error tolerance values, the minimum number M of reference signal received powers that need to be included in the feedback information reported by the second communication device, wherein M is a positive integer less than or equal to K. The determining module includes a query submodule, configured to query a target query table based on the N reference signal received powers and the set error tolerance values ​​to obtain the minimum number M of reference signal received powers that need to be included in the feedback information reported by the second communication device; wherein the target query table includes multiple reference signal received power groups, multiple different set error tolerance values, and the minimum number of reference signal received powers corresponding to the multiple reference signal received power groups and the multiple different set error tolerance values, wherein each of the multiple reference signal received power groups includes at least one reference signal received power. A sending module is used to send the M to the second communication device.

7. An electronic device, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 5.

Images