Adaptive video transmission method and apparatus, electronic device, and storage medium

Abstract

The present application relates to the technical field of communications, and discloses an adaptive video transmission method and apparatus, an electronic device, and a storage medium. The method comprises: acquiring a wireless channel environment parameter and a wireless scheduling parameter of a current period; inputting the wireless channel environment parameter and the wireless scheduling parameter of a preset target time sequence into a converged QoE prediction neural network model, and predicting a QoE measurement index for video transmission within a future preset duration; and on the basis of the predicted QoE measurement index, performing optimization adjustment on the wireless scheduling parameter of the next period to make the QoE measurement index of the next period reach a preset QoE standard for transmission.

Description

Video adaptive transmission method, device, electronic device and storage medium

[0001] Related applications

[0002] This application claims priority to Chinese patent application No. 202310759489.X filed on June 26, 2023, the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present application relates to the field of communication technology, and in particular to a method, device, electronic device, and storage medium for adaptive video transmission. Background Art

[0004] With the advancement of communication technologies and mobile devices, new video services (such as augmented reality (AR) and virtual reality (VR)) are becoming increasingly popular across various industries, including education, film, television, healthcare, and gaming, providing users with a richer experience. However, failing to provide a good user experience for these video services will lead to significant customer loss.

[0005] Therefore, optimizing user experience has become one of the primary goals of network operators (including equipment vendors, video service providers, and telecom operators). The ultimate goal of wireless video streaming services is to provide users with a satisfactory quality of experience. The International Telecommunication Union (ITU) defines Quality of Experience (QoE) as a metric that measures user subjective experience. It refers to the overall acceptability of an application or service as perceived by end users, and it directly reflects the user's subjective feelings about the service. QoE helps network operators understand the influencing parameters closely related to user satisfaction, ultimately improving user loyalty and maintaining an advantage in the fierce market competition. Therefore, configuring wireless scheduling parameters for wireless video streaming services based on QoE is not only a research focus in academia, but also a key factor for network operators to ensure a good quality of experience and retain and expand their user base.

[0006] However, current methods for optimizing user experience (QoE) rely primarily on users actively providing feedback on video playback status. This method consumes significant uplink resources and lacks real-time feedback, making it impossible to timely and accurately assess the QoE of wireless video streaming services and perform adaptive video transmission control to enhance users' subjective experience of video services.

[0007] Summary of the Invention

[0008] The main purpose of this application is to provide a method, device, electronic device and storage medium for adaptive video transmission, aiming to timely and accurately evaluate the QoE of wireless video streaming services and perform adaptive video transmission control.

[0009] To achieve the above objectives, the present application provides a method for adaptively transmitting a video, comprising:

[0010] Obtain wireless channel environment parameters and wireless scheduling parameters for the current period;

[0011] Inputting wireless channel environment parameters and wireless scheduling parameters of a preset target time series into a converged QoE prediction neural network model to predict the QoE metric for video transmission within a future preset duration, wherein the preset target time series includes a current period and a preset duration before the current period;

[0012] According to the predicted QoE metric, the wireless scheduling parameters for the next cycle are optimized and adjusted so that the QoE metric for the next cycle reaches the preset QoE standard for transmission.

[0013] In addition, to achieve the above-mentioned purpose, the present application also provides a video adaptive transmission device, including:

[0014] A data acquisition module is configured to obtain wireless channel environment parameters and wireless scheduling parameters of the current period;

[0015] A QoE prediction module is configured to input wireless channel environment parameters and wireless scheduling parameters of a preset target time series into a converged QoE prediction neural network model to predict a QoE metric for video transmission within a future preset duration, wherein the preset target time series includes a current period and a preset duration before the current period;

[0016] The wireless scheduling optimization module is configured to optimize and adjust the wireless scheduling parameters of the next cycle according to the predicted QoE metric index so that the QoE metric index of the next cycle reaches the preset QoE standard for transmission.

[0017] In addition, to achieve the above-mentioned purpose, the present application also provides an electronic device, which includes: a memory, a processor, and an adaptive transmission program for video stored on the memory and runnable on the processor. When the adaptive transmission program for video is executed by the processor, the adaptive transmission method for video as described above is implemented.

[0018] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and the computer-readable storage medium stores an adaptive transmission program for video. When the adaptive transmission program for video is executed by a processor, the adaptive transmission method for video as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0020] FIG1 is a flow chart of a first embodiment of a method for adaptively transmitting video according to the present application;

[0021] FIG2 is a schematic diagram of a training framework of a QoE prediction neural network model in an example of the present application;

[0022] FIG3 is a schematic diagram of the architecture of an adaptive video transmission method in an example of the present application;

[0023] FIG4 is a schematic diagram of a process framework of a method for adaptively transmitting video in an example of the present application;

[0024] FIG5 is a schematic diagram of functional modules of an adaptive video transmission device according to an embodiment of the present application;

[0025] FIG6 is a schematic diagram of the hardware structure of the electronic device involved in the embodiment of the present application.

[0026] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0027] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0028] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0029] It should be noted that all directional indications in the embodiments of the present application (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0030] In this application, unless otherwise specified or limited, the terms "connection" and "fixation" should be understood in a broad sense. For example, "fixation" can mean fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two elements or interaction between two elements, unless otherwise specified. For those skilled in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0031] In addition, the descriptions of "first", "second", etc. in this application are for descriptive purposes only and should not be understood as indicating or implying their relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but this must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0032] Currently, the main method for optimizing user experience (QoE) is to rely on users to actively provide feedback on video playback status. This method consumes a large amount of uplink resources and cannot provide real-time feedback. As a result, it is impossible to timely and accurately evaluate the QoE of wireless video streaming services and perform adaptive video transmission control to improve users' subjective experience of video services.

[0033] Based on this, an embodiment of the present application provides a method for adaptive transmission of a video. Referring to FIG1 , FIG1 is a flow chart of an embodiment of an adaptive transmission method for a video of the present application. In this embodiment, the adaptive transmission method for a video includes:

[0034] Step S10, obtaining wireless channel environment parameters and wireless scheduling parameters of the current period;

[0035] In this embodiment, the wireless channel environment parameter refers to a channel environment characteristic parameter of the wireless channel, and the wireless channel environment parameter may include channel quality indication CQI (Channel Quality Indication) information, signal to interference plus noise ratio SINR (Signal to Interference plus Noise Ratio, signal to interference plus noise ratio) information, reference signal received power RSRP (Reference Signal Receiving Power, reference signal received power) information, path loss PL (Path Loss, path loss) information, and base station transmit power information.

[0036] Exemplarily, the wireless scheduling parameter refers to a policy parameter for scheduling optimization of wireless channel environment parameters, and the wireless scheduling parameter may include modulation and coding strategy MCS (Modulation and Coding Scheme, modulation and coding strategy) information, and resource block RB (Resource Block, resource block) quantity information.

[0037] Step S20: Inputting the wireless channel environment parameters and wireless scheduling parameters of the preset target time series into the converged QoE prediction neural network model to predict the QoE metric of video transmission within a preset time period in the future;

[0038] The preset target time sequence includes the current cycle and a preset duration before the current cycle. In other words, the wireless channel environment parameters and wireless scheduling parameters of the preset target time sequence refer to the wireless channel environment parameters and wireless scheduling parameters of the current cycle, as well as the wireless channel environment parameters and wireless scheduling parameters of the preset duration before the current cycle. This preset duration can be equal to the duration of one cycle or the duration of multiple cycles, and this embodiment does not impose any specific limitation on this.

[0039] Those skilled in the art will understand that the current cycle refers to the period during which the current wireless channel environment parameters and wireless scheduling parameters are obtained for optimization and scheduling. This means that "obtaining the wireless channel environment parameters and wireless scheduling parameters" should be performed periodically, and "inputting the wireless channel environment parameters and wireless scheduling parameters into the converged QoE prediction neural network model to predict the QoE metrics for video transmission within a preset future duration" should also be performed periodically. Before the next cycle of each cycle arrives, the QoE metrics for video transmission within a preset future duration need to be predicted. This allows for periodic iterative optimization of the wireless scheduling parameters based on the QoE metrics predicted for each cycle.

[0040] In this embodiment, the types of QoE metrics may include at least one of the number of freezes, freeze duration, number of switches, and switching resolution. It is known to those skilled in the art that the types of QoE metrics can be defined by analyzing the R17 protocol. For different video services, it is necessary to consider which QoE metrics to collect and how to deploy monitoring and real-time collection of these data in the system. Currently, existing terminals do not support accurate real-time collection of video playback status information (including QoE metrics), but the R17 protocol defines the collection of real-time playback status information of terminals.

[0041] In response to the problems of insufficient objective measurement and prediction of QoE for video services in existing networks, imperfect technical means for measuring and predicting QoE for video services, and unclear optimization of QoE based on scheduling means from the network side, the embodiments of the present application collect wireless channel environment parameters and wireless scheduling parameters from the base station side to optimize the scheduling strategy and adjust the wireless scheduling parameters, which can effectively ensure the user's QoE in video services, as shown in Figure 4. This embodiment deploys the QoE prediction neural network model on the base station side. Without the need for real-time feedback on video playback status from the user terminal (reducing uplink resource consumption), the base station only needs to collect wireless channel environment parameters and wireless scheduling parameters to predict the user's quality of experience in real time, making better video transmission scheduling decisions in advance and improving the user's quality of experience.

[0042] Exemplarily, the type of the QoE prediction neural network model may be an LSTM (Long Short-Term Memory) neural network model, a Transform neural network model, a Kmeans clustering neural network model, or a time series neural network model.

[0043] In this embodiment, the preset duration can be equal to the duration of one cycle, and of course, it can also be equal to the duration of multiple cycles. Those skilled in the art can set it according to actual conditions. This embodiment does not make specific limitations on this, so as to better (accurately and sensitively) predict the QoE measurement indicators for future video transmission.

[0044] In one embodiment, the types of QoE measurement indicators may include the number of freezes, freeze duration, number of switches, and switching resolution. By setting different weights for the number of freezes, freeze duration, number of switches, and switching resolution, the various types of QoE measurement indicators are weightedly combined and input into the QoE prediction neural network model.

[0045] After step S20, step S30 is executed to optimize and adjust the wireless scheduling parameters of the next cycle according to the predicted QoE metric index, so that the QoE metric index of the next cycle reaches the preset QoE standard for transmission.

[0046] In this embodiment, after obtaining the wireless channel environment parameters and wireless scheduling parameters for the current period, the wireless channel environment parameters and wireless scheduling parameters for a preset target time series are input into a QoE prediction neural network model to predict a QoE metric. This facilitates optimizing and adjusting the wireless scheduling parameters for the next period based on the predicted QoE metric, so that the QoE metric for the next period meets the preset QoE standard for transmission. This enables adaptive adjustment of video stream transmission control, effectively combating network fluctuations while improving video playback quality. In wireless transmission, the wireless network environment is the primary factor affecting QoE. To provide users with a better quality of experience, the QoE prediction neural network model can be used to efficiently and accurately establish a mapping relationship between the wireless channel environment parameters, wireless scheduling parameters, and the QoE metric. After the QoE metric is predicted, an optimization strategy is implemented to adjust the wireless scheduling parameters based on the predicted QoE metric to improve user QoE.

[0047] The preset QoE standard for optimal user experience quality during video playback can be defined based on the DASH (Dynamic Adaptive Streaming over HTTP) video transmission protocol and the user's intuitive experience of watching videos.

[0048] The QoE prediction neural network model is pre-trained through a learning and training process, that is, a converged QoE prediction neural network model, which can accurately predict the QoE measurement indicators for the next cycle based on the wireless channel environment parameters and wireless scheduling parameters of the preset target time series.

[0049] The application environment of the embodiment of the present application can be a video transmission adaptive optimization scenario based on QoE prediction for NR (New Radio), LTE (Long Term Evolution), FDD (Frequency-division Duplex), or TDD (Time-division Duplex). This embodiment can collect wireless channel status (wireless channel environment parameters) and scheduling information (i.e., wireless scheduling parameters) on the base station side, use the QoE prediction neural network model to predict user QoE in real time, execute optimization strategies based on the QoE prediction value, adjust the wireless scheduling parameters, and thus improve user QoE.

[0050] With reference to Figures 1 and 3, the technical solution of the embodiment of the present application is to obtain the wireless channel environment parameters and wireless scheduling parameters of the current period, and input the wireless channel environment parameters and wireless scheduling parameters of a preset target time series (including the current period and a preset time period before the current period) into a converged QoE prediction neural network model to predict the QoE metric indicators of video transmission within a future preset time period (which may include at least one of the number of freezes, freeze time, number of resolution switching, and resolution after switching). Then, based on the predicted QoE metric indicators, the wireless scheduling parameters of the next period are optimized and adjusted so that the QoE metric indicators of the next period meet the preset QoE standard for transmission, thereby realizing the QoE metric indicators predicted by the QoE prediction neural network model, executing the QoE wireless scheduling strategy, adjusting the wireless scheduling parameters, and improving the user QoE of the video transmission service.

[0051] Compared to the current method of optimizing user QoE by actively providing user feedback on video playback status, the embodiment of the present application does not require a large amount of uplink resources and can provide real-time feedback, thereby enabling the embodiment of the present application to timely and accurately evaluate the QoE of wireless video streaming services, and optimize and adjust the current wireless scheduling parameters based on the evaluated QoE (i.e., the predicted QoE metric), thereby achieving adaptive video transmission control and improving the user's subjective experience of the video service.

[0052] In a possible implementation manner, before the step of optimizing and adjusting the radio scheduling parameters for the next period, the method further includes:

[0053] Step A10, determining whether the predicted QoE metric meets the preset QoE standard;

[0054] Exemplarily, the QoE metric includes the number of freezes, freeze duration, the number of resolution switches, and the resolution value after switching. Step A10, determining whether the predicted QoE metric meets the preset QoE standard, includes:

[0055] Step B10: If it is determined that the number of freezes reaches a first preset number threshold, the freeze duration reaches a preset duration threshold, the number of switches reaches a second preset number threshold, or the resolution value after switching is less than a preset resolution threshold, determining that the predicted QoE metric does not meet the preset QoE standard;

[0056] Step B20, when it is determined that the number of freezes does not reach the first preset number threshold, the freeze duration does not reach the preset duration threshold, the number of switches does not reach the second preset number threshold, and the resolution value after switching is not less than the preset resolution threshold, it is determined that the predicted QoE measurement index meets the preset QoE standard.

[0057] In this embodiment, the first preset number threshold, the preset duration threshold, the second preset number threshold and the preset resolution threshold can be set by those skilled in the art according to actual conditions. This embodiment does not make specific limitations, so as to more accurately detect whether the predicted QoE measurement indicator meets the preset QoE standard.

[0058] After step A10, step A20 is executed. If it is determined that the predicted QoE metric does not meet the preset QoE standard, the step of optimizing and adjusting the wireless scheduling parameters for the next cycle is executed.

[0059] In this embodiment, the preset QoE standard is not specifically limited in this embodiment, and is based on providing the best user QoE as much as possible on the basis of reducing the network channel environment requirements.

[0060] This embodiment performs subsequent steps such as optimizing and adjusting the wireless scheduling parameters for the next cycle only when it is determined that the predicted QoE metric does not meet the preset QoE standard. This improves user QoE while minimizing the operating load of wireless scheduling optimization, and enables smooth video playback even when the network signal is unstable.

[0061] Based on the first embodiment of the present application, a second embodiment of the method for adaptively transmitting video of the present application is proposed. In this embodiment, the method further includes:

[0062] Step C10: Using the wireless channel environment parameters and wireless scheduling parameters sampled in a preset time series as learning samples, and using the QoE metric associated with video transmission based on the preset time series as a sample label corresponding to the learning sample;

[0063] Step C20, associating the learning sample with the sample label to obtain a sample set;

[0064] Step C30: Perform supervised learning on the loss function of the QoE prediction neural network model using the sample set to obtain a converged QoE prediction neural network model.

[0065] Exemplarily, the step of performing supervised learning on the loss function of the QoE prediction neural network model using the sample set to obtain a converged QoE prediction neural network model includes:

[0066] Step D10, dividing the sample set into a training sample set and a validation sample set;

[0067] Step D20: Train the loss function of the QoE prediction neural network model using the training sample set, and verify the prediction accuracy of the QoE prediction neural network model using the verification sample set until the prediction accuracy reaches a preset ratio threshold, then stop training and obtain a converged QoE prediction neural network model.

[0068] This embodiment uses wireless channel environment parameters and wireless scheduling parameters sampled in a preset time series as learning samples, and uses the QoE metric associated with video transmission based on the preset time series as the sample label corresponding to the learning sample. The learning sample and the sample label are then associated to obtain a sample set. The sample set is then used to perform supervised learning on the loss function of the QoE prediction neural network model, thereby accurately obtaining a converged QoE prediction neural network model. This facilitates the subsequent efficient and accurate establishment of a mapping relationship between the wireless channel environment parameters, wireless scheduling parameters, and QoE metric based on the converged QoE prediction neural network model, thereby predicting the QoE metric for a preset future duration. Based on the predicted QoE metric, an optimization strategy is executed to adjust the wireless scheduling parameters to improve QoE. This optimization process is not completed offline once, but rather is repeatedly optimized online.

[0069] To help understand the technical concept or technical principle of the embodiments of the present application, a specific embodiment of a method for adaptively transmitting video is listed, including the following main process steps:

[0070] Step 1: Data Collection

[0071] During the video service playback process on the client, wireless channel environment parameters and wireless scheduling parameters are collected. Wireless channel environment parameters and wireless scheduling parameters can be directly obtained on the base station side. The specific description is as follows:

[0072] Wireless channel environment parameters primarily describe the state of the wireless channel during video transmission. These parameters include, but are not limited to, CQI information, RSRP information, PL information, SINR information, and base station transmit power information. Wireless scheduling parameters primarily describe the scheduling decisions made by the base station based on the wireless channel environment for optimal data transmission. Wireless scheduling parameters include, but are not limited to, MCS information and the number of allocated RBs.

[0073] QoE metrics primarily describe the user's intuitive experience while watching a video. User QoE metrics vary for different video services. However, the vast majority of current video services are based on the DASH video transmission protocol, which adaptively switches between different video resolutions based on network conditions and transmission resources. Therefore, this embodiment defines common QoE metrics for DASH-based video services, taking into account both the user's intuitive experience (stuttering) and the DASH transmission protocol (switching). These metrics include the number of stutters, stutter duration, number of switches, and switching resolution, as defined below: Number of stutters refers to the number of stutters counted by the client during video playback. Stutter duration refers to the duration of stutters counted by the client during video playback. Number of switches refers to the number of video resolution switches counted by the client during video playback. Switching resolution refers to the resolution before and after the switch, as counted by the client during video playback. Previously, these metrics could not be accurately obtained in real time. However, with the release of the R17 (TS26.247) protocol, the protocol defined a terminal quality reporting feedback mechanism, which allows these metrics to be obtained through the Play List in the quality report.

[0074] Step 2: QoE prediction

[0075] The QoE prediction module consists of two parts: time series neural network model training and QoE prediction.

[0076] The training process of the temporal neural network model is as follows:

[0077] Step 2.1: Based on the data collected in step 1, a time series is constructed from two dimensions: wireless channel environment parameters and wireless scheduling parameters. The QoE metrics corresponding to the time series are used as labels to construct a data set (training set and test set).

[0078] Step 2.2 Time series neural network training: Use the training set constructed in step 2.1 to train the time series neural network model, and use the test set to test the model. The training framework is shown in Figure 2.

[0079] In step 2.3, the trained model is deployed on the base station side. The base station collects time series data (wireless channel environment parameters and wireless scheduling parameters) over the past period of time in real time as the input of the time series neural network model. The model output is the user's QoE measurement indicator for the future period of time.

[0080] Step 3: QoE wireless scheduling optimization

[0081] Based on the QoE metrics predicted in step 2 (number of freezes, freeze duration, number of handoffs, and handoff resolution), an optimization strategy is implemented to adjust wireless scheduling parameters and improve the QoE of the video transmission service.

[0082] Referring to Figure 3, this embodiment establishes a mapping relationship between wireless channel environment parameters, wireless scheduling parameters and QoE measurement indicators by training a neural network, uses the collected wireless channel environment parameters and wireless scheduling parameters as inputs to the neural network, and uses the QoE measurement indicators as labels to perform time-series neural network training.

[0083] Referring to Figure 4, this embodiment embeds the trained neural network model into the resource scheduling system of the base station. By inputting time series data (wireless channel environment parameters and wireless scheduling parameters) over a period of time into the neural network model, the user QoE measurement indicators for a period of time in the future are predicted. Then, based on the predicted QoE measurement indicators, the wireless resource scheduling parameters are optimized to improve the user's QoE.

[0084] The above disclosure is only a preferred embodiment of the present application, and it is certainly not intended to limit the scope of protection of the present application. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present application are still within the scope covered by the present application.

[0085] That is to say, the above specific embodiments are only used to help understand the technical concept of the embodiments of this application, and do not constitute a limitation of this application. More simple transformations based on the technical concept should all be within the scope of protection of this application.

[0086] In addition, an embodiment of the present application further proposes an adaptive transmission device for video. Referring to FIG. 5 , FIG. 5 is a schematic diagram of functional modules of an adaptive transmission device for video according to an embodiment of the present application.

[0087] In this embodiment, the video adaptive transmission device includes:

[0088] The data acquisition module 10 is configured to obtain wireless channel environment parameters and wireless scheduling parameters of the current period;

[0089] The QoE prediction module 20 is configured to input wireless channel environment parameters and wireless scheduling parameters of a preset target time series into a converged QoE prediction neural network model to predict a QoE metric for video transmission within a future preset duration, wherein the preset target time series includes a current period and a preset duration before the current period;

[0090] The wireless scheduling optimization module 30 is configured to optimize and adjust the wireless scheduling parameters of the next cycle according to the predicted QoE metric index, so that the QoE metric index of the next cycle reaches the preset QoE standard for transmission.

[0091] In some embodiments, the wireless scheduling optimization module 30 is further configured to:

[0092] Determine whether the predicted QoE metrics meet the preset QoE standards;

[0093] If it is determined that the predicted QoE metric does not meet the preset QoE standard, the step of optimizing and adjusting the wireless scheduling parameters for the next period is performed.

[0094] In some embodiments, the QoE metric includes the number of freezes, freeze duration, the number of resolution switches, and the resolution value after switching. The QoE prediction module 20 is further configured as follows:

[0095] When it is determined that the number of freezes reaches a first preset number threshold, the freeze duration reaches a preset duration threshold, the number of switches reaches a second preset number threshold, or the resolution value after switching is less than a preset resolution threshold, determining that the predicted QoE metric does not meet the preset QoE standard;

[0096] When it is determined that the number of freezes does not reach the first preset number threshold, the freeze duration does not reach the preset duration threshold, the number of switches does not reach the second preset number threshold, and the resolution value after switching is not less than the preset resolution threshold, it is determined that the predicted QoE measurement index meets the preset QoE standard.

[0097] In some embodiments, the video adaptive transmission apparatus further includes a training module (not shown) configured to:

[0098] Using wireless channel environment parameters and wireless scheduling parameters sampled in a preset time series as learning samples, and using QoE metrics associated with video transmission based on the preset time series as sample labels corresponding to the learning samples;

[0099] Associating the learning sample with the sample label to obtain a sample set;

[0100] Through the sample set, supervised learning is performed on the loss function of the QoE prediction neural network model to obtain a converged QoE prediction neural network model.

[0101] In some embodiments, the training module is further configured to:

[0102] Dividing the sample set into a training sample set and a validation sample set;

[0103] The loss function of the QoE prediction neural network model is trained using the training sample set, and the prediction accuracy of the QoE prediction neural network model is verified using the verification sample set, until the prediction accuracy reaches a preset ratio threshold and the training is stopped to obtain a converged QoE prediction neural network model.

[0104] In some embodiments, the wireless scheduling parameters include modulation and coding strategy (MCS) information and resource block (RB) quantity information.

[0105] In some embodiments, the wireless channel environment parameters include channel quality indication CQI information, signal to interference plus noise ratio SINR information, reference signal received power RSRP information, path loss PL information, and base station transmit power information.

[0106] The adaptive video transmission device provided in this embodiment and the adaptive video transmission method provided in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the embodiments of the above-mentioned adaptive video transmission method. This embodiment has the same beneficial effects as the embodiments of the adaptive video transmission method, and will not be repeated here.

[0107] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0108] In addition, an embodiment of the present application also provides an electronic device, which may be, for example, an edge router, or a broadband remote access server (BRAS), a broadband network gateway (Broadband Network Gateway), a serving GPRS support node (SGSN), a gateway GPRS support node (Gateway GPRS Support Node, GGSN), a mobility management entity (MME), or a serving gateway (S-GW), etc.

[0109] 6, FIG6 is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. As shown in FIG6, the electronic device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0110] Those skilled in the art will appreciate that the structure shown in FIG6 does not limit the electronic device and may include more or fewer components than shown, or combine certain components, or arrange the components differently. As shown in FIG6 , the memory 1005 as a storage medium may include an operating system, a data storage module, a network communication module, a user interface module, and a video adaptive transmission program.

[0111] In the electronic device shown in Figure 6, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in this embodiment can be set in the communication device, and the communication device calls the adaptive transmission program of the video stored in the memory 1005 through the processor 1001, and executes the adaptive transmission method applied to the video provided in any of the above embodiments.

[0112] The terminal proposed in this embodiment and the adaptive transmission method applied to video proposed in the above embodiment belong to the same inventive concept. Technical details not fully described in this embodiment can be referred to any of the above embodiments, and this embodiment has the same beneficial effects as executing the adaptive transmission method for video.

[0113] In addition, an embodiment of the present application also proposes a storage medium, which is a computer storage medium. The computer-readable storage medium can be a non-volatile computer-readable storage medium. The computer-readable storage medium stores an adaptive transmission program for video. When the adaptive transmission program for video is executed by a processor, the adaptive transmission method for video of the present application as described above is implemented.

[0114] The various embodiments of the electronic device and computer-readable storage medium of the present application can refer to the various embodiments of the adaptive transmission method of the video of the present application, and will not be repeated here.

[0115] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0116] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0117] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course, by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling an electronic device to execute the methods described in each embodiment of the present application.

[0118] The above are merely optional embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for adaptive transmission of video, wherein: The method comprises: Obtaining wireless channel environment parameters and wireless scheduling parameters of the current period; Inputting the wireless channel environment parameters and wireless scheduling parameters of the preset target time series into the converged QoE prediction neural network model, and predicting the QoE metric index of video transmission within a future preset time period, wherein the preset target time series includes the current period and the preset time period before the current period; According to the predicted QoE metric, the wireless scheduling parameters of the next cycle are optimized and adjusted so that the QoE metric of the next cycle reaches the preset QoE standard for transmission.

2. The method for adaptively transmitting a video according to claim 1, wherein: Before the step of optimizing and adjusting the wireless scheduling parameters of the next period, the method further includes: Determine whether the predicted QoE metric meets the preset QoE standard; When it is determined that the predicted QoE metric index does not reach the preset QoE standard, the step of optimizing and adjusting the wireless scheduling parameters of the next period is performed.

3. The method for adaptively transmitting a video according to claim 2, wherein: The QoE metric includes the number of freezes, the freeze duration, the number of resolution switchings, and the resolution value after switching. The step of determining whether the predicted QoE metric reaches the preset QoE standard includes: When it is determined that the number of freezes reaches a first preset number threshold, the freeze duration reaches a preset duration threshold, the number of switches reaches a second preset number threshold, or the resolution value after switching is less than a preset resolution threshold, it is determined that the predicted QoE metric indicator does not meet the preset QoE standard; When it is determined that the number of freezes does not reach the first preset number threshold, the freeze duration does not reach the preset duration threshold, the number of switches does not reach the second preset number threshold, and the resolution value after switching is not less than the preset resolution threshold, it is determined that the predicted QoE measurement indicator meets the preset QoE standard.

4. The method for adaptively transmitting a video according to claim 1, wherein: The method further comprises: Using the wireless channel environment parameters and wireless scheduling parameters sampled in a preset time series as learning samples, and using the QoE metric associated with video transmission based on the preset time series as sample labels corresponding to the learning samples; Associating the learning sample with the sample label to obtain a sample set; Through the sample set, supervised learning is performed on the loss function of the QoE prediction neural network model to obtain a converged QoE prediction neural network model.

5. The method for adaptively transmitting a video according to claim 4, wherein: The step of performing supervised learning on the loss function of the QoE prediction neural network model through the sample set to obtain a converged QoE prediction neural network model comprises: Dividing the sample set into a training sample set and a validation sample set; The loss function of the QoE prediction neural network model is trained through the training sample set, and the prediction accuracy of the QoE prediction neural network model is verified through the verification sample set, until the prediction accuracy reaches a preset ratio threshold and the training is stopped to obtain a converged QoE prediction neural network model.

6. The method for adaptively transmitting a video according to any one of claims 1 to 5, wherein: The wireless scheduling parameters include modulation and coding strategy MCS information and resource block RB quantity information.

7. The method for adaptively transmitting a video according to any one of claims 1 to 5, wherein: The wireless channel environment parameters include channel quality indication CQI information, signal to interference plus noise ratio SINR information, reference signal received power RSRP information, path loss PL information, and base station transmission power information.

8. A video adaptive transmission device, wherein: The device comprises: A data acquisition module, configured to obtain wireless channel environment parameters and wireless scheduling parameters of the current period; The QoE prediction module is configured to input the wireless channel environment parameters and wireless scheduling parameters of a preset target time series into a converged QoE prediction neural network model to predict the QoE metric index of video transmission within a future preset time period, wherein the preset target time series includes the current period, and the preset duration before the current cycle; The wireless scheduling optimization module is configured to optimize and adjust the wireless scheduling parameters of the next cycle according to the predicted QoE metric index, so that the QoE metric index of the next cycle reaches the preset QoE standard for transmission.

9. An electronic device, wherein: The electronic device comprises: a memory, a processor, and an adaptive transmission program for video stored in the memory and executable on the processor, wherein the adaptive transmission program for video, when executed by the processor, implements the adaptive transmission method for video as described in any one of claims 1 to 7.

10. A storage medium, wherein: The storage medium is a computer-readable storage medium, on which a video adaptive transmission program is stored. When the video adaptive transmission program is executed by a processor, the video adaptive transmission method according to any one of claims 1 to 7 is implemented.

Images