Network slice resource adjustment method, device and computer-readable storage medium

Abstract

The present invention discloses a network slice resource adjustment method, device and computer-readable storage medium, and relates to the field of communications. The network slice resource adjustment method includes: identifying different users in historical network slice multimedia data and the moment when each user performs a preset behavior; marking the moment of user switching in the historical network slice multimedia data according to the identification result; using the marked historical network slice multimedia data to train a prediction model, so as to use the prediction model to process the real-time collected network slice multimedia data, obtain the channel resource requirements of the users of the network slice to be tested at future moments, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results. The embodiments of the present invention save wireless channel resources while reducing the probability of untimely or erroneous resource adjustments, thereby improving the user service quality as a whole.

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a network slice resource adjustment method, device and computer-readable storage medium. Background Art

[0002] Currently, network slicing is considered an important means of social communication for 5G and 6G, and an important technical guarantee for deterministic networks. When network slice users use cellular network access, there are differences and connections between different users in the uplink and downlink channel resources they occupy. For example, in teaching network slices, teachers will use more uplink channel resources to send video data such as PPT and voice data to other users; while network slice users who are listening to lectures use more downlink channels to receive data. In related technologies, when configuring network slices, the network controller will allocate the same uplink and downlink channel resources to each user in order to prevent SLA breaches. Summary of the invention

[0003] After analysis, the inventors found that, however, with the increase in wireless devices, spectrum resources are gradually scarce, and the method of allocating the same uplink and downlink channel resources to each user further causes a waste of channel resources.

[0004] A technical problem to be solved by an embodiment of the present invention is: how to reduce the waste of channel resources in a network slicing scenario.

[0005] According to a first aspect of some embodiments of the present invention, a network slice resource adjustment method is provided, comprising: identifying different users in historical network slice multimedia data and the moment when each user performs a preset behavior; marking the moment when the user switches in the historical network slice multimedia data according to the identification result; using the marked historical network slice multimedia data to train a prediction model, so as to use the prediction model to process the network slice multimedia data collected in real time, obtain the channel resource requirements of the users of the network slice to be tested at future moments, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

[0006] In some embodiments, the multimedia data includes voice data, and identifying different users in the historical network slice multimedia data and the moment when each user performs a preset behavior includes: performing timbre recognition on the voice data in the historical multimedia data to identify different users corresponding to different timbres and the moment when each user performs the preset behavior; based on the recognition result, marking the moment when the user switches in the historical network slice multimedia data includes: based on the recognition result, marking the moment when the timbre switches in the historical network slice multimedia data.

[0007] In some embodiments, at the moment of user switching in the historical network slice multimedia data, the user who speaks at the next moment is marked.

[0008] In some embodiments, training a prediction model using labeled historical network slice multimedia data includes: training a prediction model using labeled historical network slice multimedia data and corresponding historical network slice information, wherein the historical network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user microphone status.

[0009] In some embodiments, the prediction model is a Transformer.

[0010] In some embodiments, the network slice resource adjustment method also includes: using a prediction model to process the network slice multimedia data collected in real time to obtain the channel resource requirements of users of the network slice to be tested at future times; and adjusting the network slice resources of users of the network slice to be tested according to the prediction results.

[0011] In some embodiments, adjusting the network slice resources of the user of the network slice to be tested according to the prediction results includes: when it is predicted that the uplink channel resource demand of the user at the future time will increase, increasing the uplink channel resources of the user at the future time; when it is predicted that the downlink channel resource demand of the user at the future time will increase, increasing the downlink channel resources of the user at the future time; when it is predicted that the uplink channel resource demand of the user at the future time will decrease, reducing the uplink channel resources of the user at the future time; when it is predicted that the downlink channel resource demand of the user at the future time will decrease, reducing the downlink channel resources of the user at the future time.

[0012] In some embodiments, adjusting the network slice resources of the user of the network slice to be tested according to the prediction results includes: when it is predicted that the user will speak at a future moment, increasing the uplink channel resources of the user at the future moment; when it is predicted that the user will not speak at a future moment, reducing the uplink channel resources of the user at the future moment.

[0013] In some embodiments, using a prediction model to process network slice multimedia data collected in real time to obtain channel resource requirements of users of the network slice to be tested at future times includes: using a prediction model to process network slice multimedia data and network slice information collected in real time to obtain channel resource requirements of users of the network slice to be tested at future times, wherein the network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user's microphone status.

[0014] According to a second aspect of some embodiments of the present invention, a network slice resource adjustment method is provided, comprising: using a prediction model to process network slice multimedia data collected in real time to obtain channel resource requirements of users of the network slice to be tested at future moments, wherein the prediction model is trained using historical network slice multimedia data marked with moments of user switching, and the historical network slice multimedia data is marked by identifying different users in the historical network slice multimedia data and the moments when each user performs preset behaviors; adjusting network slice resources of users of the network slice to be tested according to the prediction results.

[0015] According to a third aspect of some embodiments of the present invention, there is provided a network slice resource adjustment device, comprising: an identification module, configured to identify different users in historical network slice multimedia data and the moment when each user performs a preset behavior; a marking module, configured to mark the moment of user switching in the historical network slice multimedia data according to the identification result; a training module, configured to use the marked historical network slice multimedia data to train a prediction model, so as to use the prediction model to process the network slice multimedia data collected in real time, obtain the channel resource requirements of the users of the network slice to be tested at future moments, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

[0016] According to a fourth aspect of some embodiments of the present invention, there is provided a network slice resource adjustment device, comprising: a prediction module, configured to use a prediction model to process network slice multimedia data collected in real time, and obtain channel resource requirements of users of the network slice to be tested at future moments, wherein the prediction model is trained using historical network slice multimedia data marked with moments of user switching, and the historical network slice multimedia data is marked by identifying different users in the historical network slice multimedia data and the moments when each user performs preset behaviors; an adjustment module, configured to adjust network slice resources of users of the network slice to be tested according to the prediction results.

[0017] According to a fifth aspect of some embodiments of the present invention, a network slice resource adjustment system is provided, comprising: the network slice resource adjustment device described in the second aspect; and the network slice resource adjustment device described in the third aspect.

[0018] According to a sixth aspect of some embodiments of the present invention, a network slice resource adjustment device is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute any one of the aforementioned network slice resource adjustment methods based on instructions stored in the memory.

[0019] According to a seventh aspect of some embodiments of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the program is executed by a processor, any one of the aforementioned network slice resource adjustment methods is implemented.

[0020] Some embodiments of the above invention have the following advantages or beneficial effects. By identifying historical network slice multimedia data to determine different users, data with user tags can be automatically generated. By using the labeled data to train the prediction model, the user's behavior at a future moment can be predicted, so that the user's demand for network slice resources can be estimated based on the predicted user behavior to adjust the network slice resources. The above embodiment automatically generates training data to train the prediction model, and dynamically adjusts the user's network slice resources according to the prediction results of the model. Therefore, the embodiments of the present invention save wireless channel resources while reducing the probability of untimely or erroneous resource adjustments, thereby improving the user service quality as a whole.

[0021] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] Figure 1 A flow chart of a network slice resource adjustment method according to some embodiments of the present invention is shown.

[0024] Figure 2 A flow chart of a network slice resource adjustment method according to some embodiments of the present invention is shown.

[0025] Figure 3 A structural schematic diagram of a network slice resource adjustment device according to some embodiments of the present invention is shown.

[0026] Figure 4 A structural schematic diagram of a network slice resource adjustment device according to some embodiments of the present invention is shown.

[0027] Figure 5 A structural schematic diagram of a network slice resource adjustment system according to some embodiments of the present invention is shown.

[0028] Figure 6A structural schematic diagram of a network slice resource adjustment device according to some further embodiments of the present invention is shown.

[0029] Figure 7 A structural schematic diagram of a network slice resource adjustment device according to some further embodiments of the present invention is shown. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0032] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0033] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0035] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0036] Figure 1 A schematic diagram of a flow chart of a network slice resource adjustment method according to some embodiments of the present invention is shown. Figure 1 As shown, the network slice resource adjustment method of this embodiment includes steps S102 to S110.

[0037] In step S102, different users in the historical network slice multimedia data and the moments when each user performs preset behaviors are identified.

[0038] Multimedia data includes, for example, voice data, video data, etc. Network slice multimedia data refers to multimedia data generated during the life cycle of a network slice. For example, for a network slice of a conference, the network slice multimedia data may be a video or voice recording of the conference. Historical network slice multimedia data refers to network slice multimedia data corresponding to a terminated network slice.

[0039] Users have various behaviors in network slicing multimedia data, such as speaking. The moment when each user performs a preset behavior is, for example, the speaking moment. When a user speaks, he or she may also show visual materials to other users, so that the user mainly sends data to the network side, and other users mainly receive data from the network side.

[0040] In step S104, based on the recognition result, the moment of user switching in the historical network slice multimedia data is marked.

[0041] In some embodiments, after identifying different users, the duration of preset actions performed by different users and the moments in the network slice lifecycle (i.e., the length of time the network slice provides communication services) when different users perform preset actions are recorded. For example, the duration of each speech of users with different timbres and the moments of speech in the network slice lifecycle are recorded.

[0042] In some embodiments, at the moment of user switching in the historical network slice multimedia data, the user who speaks at the next moment is marked. Thus, it can be determined which users will occupy more uplink channel resources and which users will occupy more downlink channel resources at the next moment. This method automatically marks the historical network slice multimedia data, thereby saving the workload of manual marking and improving the efficiency of generating training data.

[0043] In step S106, the prediction model is trained using the labeled historical network slice multimedia data.

[0044] In some embodiments, the prediction model is a time series prediction model, such as a Transformer.

[0045] For example, historical network slice multimedia data is input into the prediction model to predict which user will have a preset behavior in the future. The parameters of the prediction model are adjusted based on the gap between the predicted results and the pre-marked users.

[0046] In some embodiments, a prediction model is trained using marked historical network slice multimedia data and corresponding historical network slice information, wherein the historical network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user microphone status. For example, for a network slice corresponding to a certain video conference, the number of users is the number of users accessing the conference, and the life cycle is the duration of the conference. The network slice type indicates the type of service provided by the network slice, such as conference, remote classroom, live broadcast, etc. The user microphone status includes, for example, microphone on or microphone off.

[0047] In step S108, the prediction model is used to process the network slice multimedia data collected in real time to obtain the channel resource requirements of users of the network slice to be tested at future times.

[0048] For example, the real-time collected multimedia data of the network slice is input into the prediction model, which can predict the user who is most likely to speak at a future moment.

[0049] In some embodiments, a prediction model is used to process network slice multimedia data and network slice information collected in real time to obtain channel resource requirements of users of the network slice to be tested at future times, wherein the network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user's microphone status.

[0050] The number of users of the corresponding network slice or the duration of the communication service corresponding to the corresponding network slice can be obtained in advance, for example, filled in by the user when applying for the service corresponding to the network slice, or estimated by the server based on the information filled in by multiple users. For example, the user who applied for the video conference estimated in advance that the number of participants is 50 and the meeting time is 1:00-3:00 p.m. on a certain day of a certain month of a certain year, then the number of users can be filled in as 50, and the background server can calculate the meeting duration as 2 hours as the life cycle of the network slice; or, the user who applied for the video conference published the conference link in advance, and 50 users clicked on the link and applied to join the meeting, then the server can pre-estimate the number of users as 50. Of course, the number of users can also be determined based on the number of users at the beginning of the communication service, or adjusted in real time as the number of users changes during the communication service.

[0051] In addition, network slicing information can also include information such as whether the user turns on the camera, whether to send files, and when to access.

[0052] In step S110, the network slice resources of the user of the network slice to be tested are adjusted according to the prediction result, so that the network slice resources of the user match the user's behavior.

[0053] In some embodiments, when it is predicted that the user's demand for uplink channel resources at a future time will increase, the user's uplink channel resources at the future time are increased; when it is predicted that the user's demand for downlink channel resources at a future time will increase, the user's downlink channel resources at the future time are increased; when it is predicted that the user's demand for uplink channel resources at a future time will decrease, the user's uplink channel resources at the future time are reduced; when it is predicted that the user's demand for downlink channel resources at a future time will decrease, the user's downlink channel resources at the future time are reduced.

[0054] The increase or decrease range of the channel resources is, for example, a preset fixed value, or is determined according to the service type, or is determined according to the statistical value of historical data.

[0055] For example, if the prediction result is that a certain user will only receive data at the next moment, the number of channel resources for sending data of the user will be reduced; if the prediction result is that the user will only send data at the next moment, the number of channel resources for receiving data of the user will be reduced.

[0056] In some embodiments, the adjusted uplink channel resources or downlink channel resources may be restored to default values ​​at preset intervals.

[0057] By identifying historical network slice multimedia data to determine different users, data with user tags can be automatically generated. By using the labeled data to train the prediction model, the user's behavior at a future moment can be predicted, so as to estimate the user's demand for network slice resources based on the predicted user behavior, so as to adjust the network slice resources. The above embodiment automatically generates training data to train the prediction model, and dynamically adjusts the user's network slice resources according to the prediction results of the model. Therefore, the embodiment of the present invention saves wireless channel resources while reducing the probability of untimely or erroneous resource adjustment, thereby improving the user service quality as a whole.

[0058] Moreover, the embodiment of the present invention is implemented based on the existing channel resource scheduling process, without the need to improve the terminal, and has good backward compatibility and deployment feasibility. The embodiment of the present invention is an enhancement of the existing protocol, without introducing a new protocol process, which is convenient and concise, and has low implementation difficulty.

[0059] In some embodiments, different users can be identified by voice recognition to mark the historical network slice multimedia data. Figure 2 An embodiment of the network slice resource adjustment method of the present invention is described.

[0060] Figure 2 A schematic diagram of a flow chart of a network slice resource adjustment method according to some embodiments of the present invention is shown. Figure 2 As shown, the network slice resource adjustment method of this embodiment includes steps S202 to S212.

[0061] In step S202, timbre recognition is performed on the voice data in the historical multimedia data to identify different users corresponding to different timbre and the time when each user performs a preset behavior.

[0062] Voice recognition can also be called voiceprint recognition or speaker recognition. It is an unsupervised recognition method that saves operations such as labeling and training of the training data of the recognition model itself, thus saving human resources and computing resources.

[0063] In step S204, the moment of voice switching in the historical network slice multimedia data is marked according to the recognition result. For example, at the moment of user switching in the historical network slice multimedia data, the user who speaks at the next moment is marked.

[0064] In step S206, the prediction model is trained using the labeled historical network slice multimedia data.

[0065] In step S208, the prediction model is used to process the network slice multimedia data collected in real time to predict whether the user of the network slice to be tested will speak at a future moment.

[0066] In step S210, when it is predicted that the user will speak at a future time, the uplink channel resources of the user at the future time are increased, so that the user can send voice data more smoothly.

[0067] As required, the downlink channel resources of the user at a future time may also be reduced to make the overall resource occupancy of the user relatively balanced.

[0068] In step S212, when it is predicted that the user will not speak in the future, the uplink channel resources of the user in the future are reduced, thereby preventing the user from occupying too much uplink resources and causing a waste of uplink resources.

[0069] As required, the downlink channel resources of the user at a future time may also be increased to make the overall resource occupancy of the user relatively balanced.

[0070] The above embodiment uses an unsupervised timbre recognition method to mark historical network slice multimedia data, so as to use the marked data to train the prediction model, which can improve the marking efficiency of the training data, thereby improving the training efficiency of the model and saving computing resources.

[0071] Reference below Figure 3 and Figure 4 An embodiment of the network slice resource adjustment device of the present invention is described.

[0072] Figure 3 A schematic diagram of the structure of a network slice resource adjustment device according to some embodiments of the present invention is shown. Figure 3 As shown, the network slice resource adjustment device 300 of this embodiment includes: an identification module 3100, configured to identify different users in historical network slice multimedia data and the moment when each user performs a preset behavior; a marking module 3200, configured to mark the moment of user switching in the historical network slice multimedia data according to the identification result; a training module 3300, configured to use the marked historical network slice multimedia data to train a prediction model, so as to use the prediction model to process the network slice multimedia data collected in real time, obtain the channel resource requirements of the users of the network slice to be tested at future moments, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

[0073] In some embodiments, the multimedia data includes voice data, and the recognition module 3100 is further configured to perform timbre recognition on the voice data in the historical multimedia data to identify different users corresponding to different timbres; the marking module 3200 is further configured to mark the moment of timbre switching in the historical network slice multimedia data according to the recognition results.

[0074] In some embodiments, the marking module 3200 is further configured to mark the user who speaks at the next moment at the moment of user switching in the historical network slice multimedia data.

[0075] In some embodiments, the training module 3300 is further configured to train a prediction model using labeled historical network slice multimedia data and corresponding historical network slice information, wherein the historical network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user microphone status.

[0076] In some embodiments, the prediction model is a Transformer.

[0077] In addition, the network slice resource adjustment device 300 may also include a storage module for storing network slice multimedia data (including network slice information such as the number of network slice users and network slice lifetime), marked multimedia data, and prediction model information (including data such as trained parameter configuration).

[0078] Figure 4 The structure diagram of the network slice resource adjustment device according to some other embodiments of the present invention is shown. Figure 4As shown, the network slice resource adjustment device 400 of this embodiment includes: a prediction module 4100, configured to use a prediction model to process the network slice multimedia data collected in real time, and obtain the channel resource requirements of the users of the network slice to be tested at future moments, wherein the prediction model is trained using historical network slice multimedia data marked with the moment of user switching, and the historical network slice multimedia data is marked by identifying different users in the historical network slice multimedia data and the moment when each user performs a preset behavior; an adjustment module 4200, configured to adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

[0079] In some embodiments, the adjustment module 4200 is further configured to, when it is predicted that the user's uplink channel resource demand at the future time will increase, increase the user's uplink channel resources at the future time; when it is predicted that the user's downlink channel resource demand at the future time will increase, increase the user's downlink channel resources at the future time; when it is predicted that the user's uplink channel resource demand at the future time will decrease, reduce the user's uplink channel resources at the future time; when it is predicted that the user's downlink channel resource demand at the future time will decrease, reduce the user's downlink channel resources at the future time.

[0080] In some embodiments, the adjustment module 4200 is further configured to increase the uplink channel resources of the user at a future time when it is predicted that the user will speak at a future time; and to reduce the downlink channel resources of the user at a future time when it is predicted that the user will not speak at a future time.

[0081] In some embodiments, the prediction module 4100 is further configured to use a prediction model to process the network slice multimedia data and network slice information collected in real time to obtain the channel resource requirements of users of the network slice to be tested at future times, wherein the network slice information includes the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or at least one of the user microphone status.

[0082] Figure 5 A schematic diagram of the structure of a network slice resource adjustment system according to some embodiments of the present invention is shown. Figure 5 As shown, the network slice resource adjustment system 50 of this embodiment includes: a network slice resource adjustment device 300 and a network slice resource adjustment device 400.

[0083] Figure 6 A schematic diagram of the structure of a network slice resource adjustment device according to some other embodiments of the present invention is shown. Figure 6As shown, the network slice resource adjustment device 60 of this embodiment includes: a memory 610 and a processor 620 coupled to the memory 610, and the processor 620 is configured to execute the network slice resource adjustment method in any one of the aforementioned embodiments based on the instructions stored in the memory 610.

[0084] The memory 610 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.

[0085] Figure 7 A schematic diagram of the structure of a network slice resource adjustment device according to some further embodiments of the present invention is shown. Figure 7 As shown, the network slice resource adjustment device 70 of this embodiment includes: a memory 710 and a processor 720, and may also include an input / output interface 730, a network interface 740, a storage interface 750, etc. These interfaces 730, 740, 750 and the memory 710 and the processor 720 may be connected, for example, via a bus 760. Among them, the input / output interface 730 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 740 provides a connection interface for various networked devices. The storage interface 750 provides a connection interface for external storage devices such as SD cards and USB flash drives.

[0086] An embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, any one of the aforementioned network slice resource adjustment methods is implemented.

[0087] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A network slice resource adjustment method, comprising: Identifying different users in historical network slice multimedia data and the time when each user performs a preset behavior, including: performing timbre recognition on voice data in the historical multimedia data to identify different users corresponding to different timbres and the time when each user performs the preset behavior; Marking the moment of user switching in the historical network slice multimedia data according to the recognition result, including: marking the moment of timbre switching in the historical network slice multimedia data according to the recognition result; The prediction model is trained using the marked historical network slice multimedia data, so that the prediction model can be used to process the network slice multimedia data collected in real time, obtain the channel resource requirements of the users of the network slice to be tested at future times, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

2. The network slice resource adjustment method according to claim 1, wherein: At the moment of user switching in the historical network slice multimedia data, mark the user who speaks at the next moment.

3. The network slice resource adjustment method according to claim 1, wherein: The training of the prediction model using the labeled historical network slice multimedia data includes: A prediction model is trained using marked historical network slice multimedia data and corresponding historical network slice information, wherein the historical network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user microphone status.

4. The network slice resource adjustment method according to claim 1 or 3, wherein: The prediction model is a Transformer.

5. The network slice resource adjustment method according to claim 1, further comprising: The prediction model is used to process the network slice multimedia data collected in real time to obtain the channel resource requirements of users of the network slice to be tested at a future time; Adjust the network slice resources of the user of the network slice to be tested according to the prediction results.

6. The network slice resource adjustment method according to claim 5, wherein: The adjusting the network slice resources of the user of the network slice to be tested according to the prediction result includes: In case that it is predicted that the uplink channel resource demand of the user at a future time will increase, increasing the uplink channel resource of the user at the future time; In case that it is predicted that the downlink channel resource demand of the user at a future time will increase, increasing the downlink channel resource of the user at the future time; When it is predicted that the uplink channel resource demand of the user at a future time is reduced, reducing the uplink channel resource of the user at the future time; When it is predicted that the downlink channel resource demand of the user at a future time is reduced, the downlink channel resource of the user at the future time is reduced.

7. The network slice resource adjustment method according to claim 6, wherein: The adjusting the network slice resources of the user of the network slice to be tested according to the prediction result includes: In case that it is predicted that the user will speak at a future time, increasing the uplink channel resources of the user at the future time; When it is predicted that the user will not speak at a future time, uplink channel resources of the user at the future time are reduced.

8. The network slice resource adjustment method according to claim 5, wherein: The using the prediction model to process the network slice multimedia data collected in real time to obtain the channel resource requirements of the users of the network slice to be tested at a future moment includes: The prediction model is used to process the network slice multimedia data and network slice information collected in real time to obtain the channel resource requirements of users of the network slice to be tested at future times, wherein the network slice information includes at least one of the number of users of the corresponding network slice, the duration of the communication service corresponding to the corresponding network slice, the network slice type, or the user microphone status.

9. A network slice resource adjustment method, comprising: The network slice multimedia data collected in real time is processed by using a prediction model to obtain the channel resource requirements of users of the network slice to be tested at a future moment, wherein the prediction model is trained by using historical network slice multimedia data marked with a moment of user switching, and the historical network slice multimedia data is marked by identifying the timbre of voice data in the historical network slice multimedia data to identify different users corresponding to different timbres and the moment when each user performs a preset behavior, and the moment of user switching is the moment of timbre switching; Adjust the network slice resources of the user of the network slice to be tested according to the prediction results.

10. A network slice resource adjustment device, comprising: The identification module is configured to identify different users in the historical network slice multimedia data and the time when each user performs a preset behavior, including: performing timbre recognition on the voice data in the historical multimedia data to identify different users corresponding to different timbres and the time when each user performs the preset behavior; The marking module is configured to mark the moment of user switching in the historical network slice multimedia data according to the recognition result, including: marking the moment of timbre switching in the historical network slice multimedia data according to the recognition result; The training module is configured to use the marked historical network slice multimedia data to train the prediction model, so as to use the prediction model to process the network slice multimedia data collected in real time, obtain the channel resource requirements of the users of the network slice to be tested at future times, and adjust the network slice resources of the users of the network slice to be tested according to the prediction results.

11. A network slice resource adjustment device, comprising: A prediction module is configured to process the network slice multimedia data collected in real time by using a prediction model to obtain the channel resource demand of the user of the network slice to be tested at a future moment, wherein the prediction model is trained by using historical network slice multimedia data marked with a moment of user switching, the historical network slice multimedia data is marked by identifying the timbre of voice data in the historical network slice multimedia data to identify different users corresponding to different timbres and the moment when each user performs a preset behavior, and the moment of user switching is the moment of timbre switching; The adjustment module is configured to adjust the network slice resources of the user of the network slice to be tested according to the prediction results.

12. A network slice resource adjustment system, comprising: The network slice resource adjustment device as described in claim 10; as well as The network slice resource adjustment device as described in claim 11.

13. A network slice resource adjustment device, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the network slice resource adjustment method as described in any one of claims 1-9 based on instructions stored in the memory.

14. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the network slice resource adjustment method described in any one of claims 1 to 9.

Images