Method for adjusting media stream of new call service and new call network system

Abstract

The application discloses a media stream adjustment method of a new call service and a new call network system. The method comprises the following steps: acquiring a first transmission data set of a media stream corresponding to a first new call service; in the case that first target transmission data in the first transmission data set exceeds a preset threshold value, analyzing the first target transmission data by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media stream corresponding to the first new call service; determining a first processing strategy corresponding to the first abnormal type, and sending the first processing strategy to a control network element, wherein the control network element is used for executing the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service to be less than the threshold value. The application solves the technical problem that related media stream adjustment methods only rely on static threshold value judgment or user feedback for abnormality and processing, which leads to the inability to quickly and effectively optimize call quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, in particular to a media stream adjustment method of new call service and a new call network system. BACKGROUND

[0002] With the intelligentization and large-screenization of mobile terminals, the demands of real-time communication of users are no longer limited to the exchange of harsh and images between the two parties of the call, the interaction of touch, drag, and drag operations, and the common cooperation for the same thing. These more complex interaction requirements have gradually emerged, giving voice services new life and vitality. Therefore, 5G new call services have emerged.

[0003] Specifically, in the 5G new call network, the quality of the media stream is affected by various factors, including terminal movement, wireless bandwidth changes, and network packet loss. These factors can cause problems such as unclear, intermittent, video flickering, green screen, blue screen, mosaic, and slow file transfer speed during the call, which seriously affects the user experience. However, the existing media stream adjustment technology often relies on static threshold judgment or user feedback when detecting and handling exceptions, and lacks real-time adaptive ability and intelligent decision-making mechanism. When the network conditions change, the existing technology may not be able to quickly and effectively adjust the media stream parameters to optimize the call quality, especially in resource-limited or complex network environments.

[0004] To address the above problems, no effective solutions have been proposed so far. SUMMARY

[0005] The embodiments of the present application provide a media stream adjustment method of new call service and a new call network system to at least solve the technical problem that the related media stream adjustment method only relies on static threshold judgment or user feedback for exception and processing, resulting in the inability to quickly and effectively optimize the call quality.

[0006] According to an aspect of an embodiment of the present application, a media stream adjustment method of new call service is provided, comprising: obtaining a first transmission data set of a media stream corresponding to a first new call service, wherein the first transmission data set includes a plurality of first transmission data; in a case where a first target transmission data in the first transmission data set exceeds a preset threshold value, analyzing the first target transmission data using a pre-trained target exception classification model to obtain a first exception type of the media stream corresponding to the first new call service; determining a first processing strategy corresponding to the first exception type, and sending the first processing strategy to a control network element, wherein the control network element is used to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service not to exceed the threshold value.

[0007] Optionally, the training process of the target abnormality classification model comprises: obtaining a first training sample set and a first sample label set, wherein the first training sample set comprises a plurality of first training samples each composed of second target transmission data lower than a preset threshold value in a second transmission data set of a media stream corresponding to a historical new call service, the first sample label set comprises a second abnormality type corresponding to each first training sample as a first sample label, and the second target transmission data comprises at least one of bandwidth, packet loss rate, code rate, frame rate, resolution, jitter rate, and time delay, and the second abnormality type comprises at least one of black screen, blue screen, mosaic, screen flower, and voice discontinuity; constructing an initial abnormality classification model; and iteratively training the initial abnormality classification model by using the first training sample set and the first sample label set to obtain the target abnormality classification model.

[0008] Optionally, the iteratively training the initial abnormality classification model by using the first training sample set and the first sample label set to obtain the target abnormality classification model comprises: sequentially inputting batches of first training samples in the first training sample set into the initial abnormality classification model to obtain each predicted abnormality type output by the initial abnormality classification model; constructing a first target loss function according to the predicted abnormality type of each batch of first training samples and the corresponding first sample label; adjusting the model parameters of the initial abnormality classification model according to the first target loss function until the model converges to obtain the target abnormality classification model.

[0009] Optionally, the determining the first processing strategy corresponding to the first abnormality type comprises: analyzing the first abnormality type by using a pre-trained target strategy decision model to obtain the corresponding first processing strategy; and the first processing strategy comprises at least one of adjusting the code rate to a target code rate value, adjusting the session description protocol parameter to a target parameter value, and adjusting the bandwidth to a target bandwidth value.

[0010] Optionally, the training process of the target strategy decision model comprises: obtaining a second training sample set and a second sample label set, wherein the second training sample set comprises a plurality of training samples each composed of a second abnormality type of a media stream corresponding to a plurality of historical new call services, the second sample label set comprises a second processing strategy corresponding to each second training sample as a second sample label, and the second processing strategy is a processing strategy with the lowest network cost and the highest user evaluation; constructing an initial strategy decision model; and iteratively training the initial strategy decision model by using the second training sample set and the second sample label set to obtain the target strategy decision model.

[0011] Optionally, the initial strategy decision model is iteratively trained by using the second training sample set and the second sample label set to obtain a target strategy decision model, including: inputting batches of second training samples in the second training sample set into the initial strategy decision model in sequence to obtain respective predicted processing strategies output by the initial strategy decision model; constructing a second target loss function according to the respective predicted processing strategies of the batches of second training samples and the corresponding second processing strategies; adjusting model parameters of the initial strategy decision model according to the second target loss function until the model converges to obtain the target strategy decision model.

[0012] Optionally, the control network element includes: an MF network element, a VoLTE application server, a policy and charging rules function network element / packet control function network element, and the sending of the first processing strategy to the control network element includes: in the case where the first processing strategy is to adjust a code rate to a target code rate value, sending a code rate adjustment request message carrying the target code rate value to the MF network element, where the MF network element is configured to send a real-time transport control protocol request message carrying a temporary maximum code rate request field as the target code rate value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust a code rate of a media stream corresponding to the first new call service; in the case where the first processing strategy is to adjust a session description protocol parameter to a target parameter value, sending a session description protocol parameter adjustment request message carrying the target parameter value to the VoLTE application server, where the VoLTE application server is configured to send a Re-invite request message carrying the target parameter value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust a session description protocol parameter of a media stream corresponding to the first new call service; and in the case where the first processing strategy is to adjust a bandwidth to a target bandwidth value, sending a bandwidth adjustment request message carrying the target bandwidth value to the policy and charging rules function network element / packet control function network element, where the policy and charging rules function network element / packet control function network element is configured to send a bandwidth adjustment instruction carrying the target bandwidth value to a core network element to instruct the core network element to adjust a bandwidth of a media stream corresponding to the first new call service.

[0013] According to another aspect of the embodiments of the present application, a new call network system is also provided, and the new call network system at least includes a media stream adjustment network element and a control network element. The media stream adjustment network element is configured to obtain a first transmission data set of a media stream corresponding to a first new call service, wherein the first transmission data set includes a plurality of first transmission data. When a first target transmission data in the first transmission data set exceeds a preset threshold value, the first target transmission data is analyzed by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media stream corresponding to the first new call service. A first processing strategy corresponding to the first abnormal type is determined and sent to the control network element. The control network element is configured to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service to be less than the threshold value.

[0014] Optionally, the new call network system further includes a calling user equipment, a called user equipment and a core network element. The control network element includes an MF network element, a VoLTE application server and a policy and charging rules function network element / packet control function network element. When the first processing strategy is to adjust a code rate to a target code rate value, the media stream adjustment network element is configured to send a code rate adjustment request message carrying the target code rate value to the MF network element. When the first processing strategy is to adjust a session description protocol parameter to a target parameter value, the media stream adjustment network element is configured to send a session description protocol parameter adjustment request message carrying the target parameter value to the VoLTE application server. When the first processing strategy is to adjust a bandwidth to a target bandwidth value, the media stream adjustment network element is configured to send a bandwidth adjustment request message carrying the target bandwidth value to the policy and charging rules function network element / packet control function network element. The MF network element is configured to send a real-time transport control protocol request message carrying a temporary maximum code rate request field as the target code rate value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust the code rate of the media stream corresponding to the first new call service. The VoLTE application server is configured to send a Re-invite request message carrying the target parameter value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust the session description protocol parameter of the media stream corresponding to the first new call service. The policy and charging rules function network element / packet control function network element is configured to send a bandwidth adjustment instruction carrying the target bandwidth value to the core network element to instruct the core network element to adjust the bandwidth of the media stream corresponding to the first new call service.

[0015] According to another aspect of the embodiments of the present application, a network device is also provided, and the network device includes a memory and a processor. The memory stores a computer program, and the processor is configured to execute the media stream adjustment method of the new call service by using the computer program.

[0016] In the embodiment of the present application, the first transmission data set of the media stream corresponding to the first new call service is obtained, wherein the first transmission data set includes a plurality of first transmission data; in the case that the first target transmission data in the first transmission data set exceeds the preset threshold value, the pre-trained target abnormal classification model is used to analyze the first target transmission data, and the first abnormal type of the media stream corresponding to the first new call service is obtained; the first processing strategy corresponding to the first abnormal type is determined, and the first processing strategy is sent to the control network element, wherein the control network element is used to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service not to exceed the threshold value. In the whole adjustment process, the media stream quality is intelligently analyzed by the large model, the possible call problems are predicted, and the most suitable adjustment strategy is automatically selected. This not only improves the accuracy of adjustment, but also reduces the dependence on manual intervention. Compared with the traditional method depending on rules or threshold, the embodiment of the present application can better adapt to the complex and changeable network environment and user demand. Further, the technical problem that the related media stream adjustment method only depends on static threshold judgment or user feedback for exception and processing, resulting in the inability to quickly and effectively optimize the call quality is solved. BRIEF DESCRIPTION OF DRAWINGS

[0017] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 is a structural schematic diagram of an optional new call network system according to an embodiment of the present application;

[0019] Figure 2 is a flowchart of an optional adaptive adjustment of a media stream according to an embodiment of the present application;

[0020] Figure 3 is another optional flowchart of an adaptive adjustment of a media stream according to an embodiment of the present application;

[0021] Figure 4 is another optional flowchart of an adaptive adjustment of a media stream according to an embodiment of the present application;

[0022] Figure 5 is a flowchart of an optional media stream adjustment method of a new call service according to an embodiment of the present application;

[0023] Figure 6 is a structural schematic diagram of an optional network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application, so that those skilled in the art can better understand the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in other than the order illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a list of steps or units need not be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to such processes, methods, products or devices.

[0026] In order to better understand the embodiments of the present application, first, the part of the nouns or terms appearing in the description of the embodiments of the present application are translated and explained as follows:

[0027] Real-time Control Protocol (RTCP): provides out of band control for RTP (Real-time Protocol) media streams. RTCP does not transport data in the present application, but it can work with RTP to package and send multimedia data. RTCP transmits control data periodically between participants in a streaming multimedia session.

[0028] TMMBR (Temporal Max Media Bitrate Request): used to request the sender to transmit data stream according to the specified maximum bit rate, usually used to ensure the smoothness of communication in the case of network jitter.

[0029] Embodiment 1

[0030] According to the embodiments of the present application, a new call network system is provided, and the structure of the new call network system 10 is as follows Figure 1As shown, the new call network system 10 comprises a media flow adjustment network element 11 and a control network element, wherein: the media flow adjustment network element 11 is configured to detect and analyze transmission data of a media flow in real time, and when the transmission data does not meet the expected requirements, a corresponding processing strategy can be formulated according to the transmission data; and the control network element is configured to receive the processing strategy of the media flow adjustment network element 11 to ensure the continuity and high quality of the call.

[0031] Specifically, the media flow adjustment network element 11 and the control network element can interact according to the following process to optimize and adjust the media flow corresponding to the new call service, comprising:

[0032] Step S1, the media flow adjustment network element 11 can obtain a first transmission data set of the media flow corresponding to the first new call service, wherein the first transmission data set comprises a plurality of first transmission data;

[0033] Step S2, when the first target transmission data in the first transmission data set exceeds the preset threshold value, the media flow adjustment network element 11 can analyze the first target transmission data by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media flow corresponding to the first new call service;

[0034] Step S3, the media flow adjustment network element 11 determines a first processing strategy corresponding to the first abnormal type and sends the first processing strategy to the control network element.

[0035] Step S4, the control network element can execute the first processing strategy to adjust the first target transmission data of the media flow corresponding to the first new call service to not exceed the threshold value.

[0036] In addition, the new call network system 10 further comprises: a calling user equipment 13, a called user equipment 14, a core network element 15, an SBC (Session Border Controller, Session Border Controller) network element / P-CSCF (Proxy Call Session Control Function, Proxy Call Session Control Function) 16, an I-CSCF (Interrogating Call Session Control Function, Interrogating Call Session Control Function) / S-CSCF (Serving Call Session Control Function, Serving Call Session Control Function) network element 17, a data channel application server 18, and a data channel signaling function (Data Channel Signaling Function, DCSF) network element 19, wherein:

[0037] The calling user equipment 13 is a terminal device initiating a new call service, which can be a 5G-enabled smartphone, tablet computer or other mobile device, etc. It is mainly responsible for establishing a call connection with the called user equipment 14, sending audio, video or DC data, and receiving media streams from the called user equipment 14;

[0038] The called user equipment 14 is a terminal device accepting and responding to a new call request, which can also be a 5G-enabled smartphone, tablet computer or other mobile device, etc. It is mainly responsible for displaying media streams from the calling user equipment 13 while sending its own media streams;

[0039] The core network element 15 plays a control and routing role in the new call network system, and the core network element 15 includes but is not limited to: 4G EPC (Evolved Packet Core) or 5G Core.

[0040] In addition, since different control network elements in the new call network system have different functions, the functional specificity of the control network elements determines that they can effectively execute the policies related to their functions. Specifically, the control network elements can be divided into: MF (Media Function) network element 121, VoLTE application server 122, policy and charging rules function network element PCRF (Policy and Charging Rules Function) / packet control function network element PCF (Policy Control Function) 123, wherein the MF network element 121 is mainly responsible for the management of media resources and the forwarding of traffic, and is therefore closely related to the code rate adjustment policy; the VoLTE application server 122 provides voice and video call services, and therefore plays a key role in resolution and frame rate adjustment policies; the PCRF network element / PCF network element 123 is used to control network resources and policy execution, and is suitable for dynamic bandwidth adjustment. Therefore, the first processing policy is different, and the control network elements and the interaction process of the media stream adjustment network element 11 are also different, specifically:

[0041] When the first processing policy is to adjust the code rate to a target code rate value, the media stream adjustment network element 11, the MF network element 121, the calling user equipment 13 and the called user equipment 14 can interact according to the process shown in Figure 2 to adaptively adjust the media stream corresponding to the new call service, including:

[0042] Step 1: The media stream adjustment network element 11 sends a code rate adjustment request message carrying a target code rate value to the MF network element;

[0043] Second step: the MF network element 121 sends a real-time transport control protocol request message carrying a temporary maximum media bitrate request (TMMBR) field with a target bitrate value to the calling user equipment 13 and the called user equipment 14, to request the calling user equipment 13 and the called user equipment 14 to adjust the bitrate of the media stream corresponding to the first new call service;

[0044] Third step: the calling user equipment 13 and the called user equipment 14 respectively send corresponding real-time transport control protocol response messages to the MF network element 121, to feed back the bitrate adjustment result.

[0045] Therefore, after the bitrate is adjusted through the above interaction process, the experience of the ongoing new call service can be greatly improved, and the screen flashing situation can be reduced.

[0046] In the case where the first processing strategy is to adjust the session description protocol parameter to a target parameter value, the media stream adjustment network element 11, the VoLTE application server 122, the calling user equipment 13 and the called user equipment 14 can interact according to the process as shown in Figure 3 to adaptively adjust the media stream corresponding to the new call service, including:

[0047] First step: the media stream adjustment network element 11 sends a session description protocol parameter adjustment request message carrying a target parameter value to the VoLTE application server, wherein the target parameter value includes a resolution parameter and / or a frame number;

[0048] Second step: the VoLTE application server 122 sends a Re-invite request message carrying a target parameter value to the calling user equipment 13 and the called user equipment 14, to request the calling user equipment 13 and the called user equipment 14 to adjust the session description protocol parameter of the media stream corresponding to the first new call service;

[0049] Third step: the calling user equipment 13 and the called user equipment 14 respectively send corresponding Re-invite response messages to the VoLTE application server 122, to feed back the session description protocol parameter adjustment result.

[0050] In the case where the first processing strategy is to adjust the bandwidth to a target bandwidth value, the media stream adjustment network element 11, the PCRF network element / PCF network element 123 and the core network element 15 can interact according to the process as shown in Figure 4 to adaptively adjust the media stream corresponding to the new call service, including:

[0051] First step: the media stream adjustment network element 11 sends a bandwidth adjustment request message carrying a target bandwidth value to the PCRF network element / PCF network element 123, wherein the target bandwidth value can be an audio bandwidth, a video bandwidth or a data channel bandwidth;

[0052] Second step: the PCRF network element / PCF network element 123 sends a bandwidth adjustment instruction carrying the target bandwidth value to the core network element 15, to instruct the core network element 15 to adjust the bandwidth of the media stream corresponding to the first new call service;

[0053] Third step: the core network element 15 sends a corresponding bandwidth adjustment result to the PCRF network element / PCF network element 123.

[0054] When the architecture of the new call network system is a 4G EPC (Evolved Packet Core) network, if the media stream adjustment network element 11 detects that the bandwidth requirement of an audio / video call or a data channel (DC) call exceeds the current allocation, or the network condition leads to insufficient bandwidth, serious packet loss and other problems, it will interact with the PCRF network element to request bandwidth adjustment. The PCRF will formulate and implement corresponding bandwidth adjustment strategies based on user subscription information, QoS (Quality of Service) requirements, network resource conditions and other factors, such as dynamically increasing the bandwidth allocation of audio, video or data channel. When the architecture of the new call network system is a 5G Core network, the PCF network element will replace the PCRF network element in the 4G network and serve as a control plane to manage policies and QoS rules. That is, when the media stream adjustment network element detects bandwidth-related problems in the 5G network, it will interact with the PCF network element. The PCF network element also dynamically adjusts network bandwidth allocation based on user policies, service level agreements and real-time network resource conditions to meet the media stream transmission requirements of new call services.

[0055] It should be noted that the media stream adjustment network element 11 uses different interfaces when sending adjustment request messages to the MF network element 121, the VoLTE application server 122 and the PCRF network element / PCF network element 123, respectively.

[0056] Through the cooperation of these modules and network elements in the new call network system, a complete control and optimization system is formed, so that the transmission parameters of the media stream corresponding to the new call service can be intelligently adjusted according to real-time network conditions and user requirements, ensuring stable and smooth call experience even in poor network conditions.

[0057] Embodiment 2

[0058] According to the embodiments of the present application, a media stream adjustment method for a new call service is also provided. It should be noted that the steps shown in the flowchart can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0059] Figure 5 FIG. 1 is a flowchart illustrating a media stream adjustment method for a new call service according to an embodiment of the present application. As shown in FIG. 1, the method comprises the following steps S502-S506, wherein: Figure 5

[0060] In step S502, first transmission data of a media stream corresponding to a first new call service is obtained.

[0061] In the technical solution provided in step S502, the first transmission data set is a set of real-time transmission quality information of the media stream such as audio and video in the ongoing call, which includes a plurality of first transmission data such as packet loss rate, delay, jitter, bandwidth usage, code rate, frame rate, resolution, and codec type, etc. These data can evaluate the media transmission state of the corresponding first new call service.

[0062] In step S504, when any first transmission data in the first transmission data set exceeds a preset threshold value, the first target transmission data is analyzed using a preset target abnormality classification model to obtain a first abnormality type of the media stream corresponding to the first new call service.

[0063] In the technical solution provided in step S504, the media stream adjustment network element compares each first transmission data (such as packet loss rate, delay, jitter, bandwidth usage, code rate stability, etc.) in the first transmission data set with a preset threshold value (i.e., a benchmark for evaluating whether the media stream quality reaches a normal level or whether there is a potential problem) to determine which first transmission data exceeds the threshold value. Specifically:

[0064] Packet loss rate refers to the ratio of data packets that fail to reach the destination to the total number of sent data packets due to network congestion, transmission errors, or device failures, etc. Therefore, when the packet loss rate exceeds the preset ratio threshold, it may affect the continuity of the media stream, causing video lag or audio discontinuity.

[0065] Delay refers to the time from sending to receiving a data packet. Therefore, when the delay of the media stream exceeds the preset delay threshold, it may affect the real-time nature of the call, making the conversation parties feel a significant delay.

[0066] ​Network jitter refers to the instability of the transmission time of data packets. Therefore, when the jitter value of the media stream exceeds the preset jitter threshold, it may affect the smoothness of the media stream playback, causing the video to play unsmoothly or the audio to be intermittent.

[0067] Network bandwidth refers to the maximum rate at which a network communication link can transmit data, usually expressed in bits per second (bps). Therefore, when the network bandwidth usage reaches or exceeds the preset threshold, it may indicate that the current available bandwidth is insufficient to support high-quality media stream transmission;

[0068] Therefore, once it is known that the first target transmission data exceeds the preset threshold, the media stream adjustment network element can analyze the first target transmission data using the pre-trained target anomaly classification model to obtain the first abnormal type of the media stream corresponding to the first new call service, such as screen flashing, green screen, blue screen, mosaic, or intermittent audio call, etc.

[0069] Step S506, determining a first processing strategy corresponding to the first abnormal type, and sending the first processing strategy to the control network element.

[0070] In the technical solution provided in the above step S506, the media stream adjustment network element can determine the corresponding first processing strategy (i.e. optimization strategy) according to the first abnormal type obtained in the above steps, wherein the first processing strategy is used to adjust the media parameters such as code rate, frame rate, resolution and bandwidth, so that the first target transmission data does not exceed the preset threshold. And the determined first processing strategy is fed back to the control network element, and the control network element executes the first processing strategy.

[0071] Based on the scheme defined in the above steps S502 to S506, the media stream adjustment network element monitors and analyzes the transmission quality of the media stream corresponding to the new call service in real time, discovers and predicts potential audio and video quality problems such as unclear, intermittent, screen flashing, green screen, blue screen, mosaic, etc. in time; then, according to the identified audio and video quality problems, a corresponding processing strategy is formulated, and the transmission parameters of the media stream are adjusted in time, which realizes the guarantee of real-time audio, video and data channel service experience in the case of network quality decline, thereby effectively improving the user call experience.

[0072] The steps of the media stream adjustment method for new call services will be described below in conjunction with a specific implementation process.

[0073] As an optional implementation, the training process of the above target anomaly classification model includes the following steps:

[0074] Step S61, obtaining a first training sample set and a first sample label set.

[0075] The first training sample set includes a plurality of first training samples each composed of second target transmission data below a preset threshold value in a second transmission data set of a media stream corresponding to a historical new call service, the first sample label set includes a second abnormal type corresponding to each first training sample as a first sample label, and the second target transmission data includes but is not limited to bandwidth, packet loss rate, code rate, frame rate, resolution, jitter rate, and time delay, and the second abnormal type includes but is not limited to black screen, blue screen, mosaic, screen, and voice discontinuity.

[0076] In step S62, an initial abnormal classification model is constructed. The model architecture of the initial abnormal classification model can be a convolutional neural network (CNN), a recurrent neural network (RNN), a Transformer architecture, etc. The selection of the model architecture can be determined according to the complexity of the problem, the type of data, the real-time requirement, and the availability of computing resources, and the present application does not make specific limitations.

[0077] In step S63, the initial abnormal classification model is iteratively trained using the first training sample set and the first sample label set to obtain a target abnormal classification model.

[0078] Specifically, in the technical solution provided in step S63, the method can include:

[0079] In step S631, a batch of first training samples in the first training sample set are sequentially input into the initial abnormal classification model to obtain each predicted abnormal type output by the initial abnormal classification model.

[0080] In step S632, a first target loss function is constructed according to the predicted abnormal type of each first training sample in the batch and the corresponding first sample label, and the first target loss function can be a cross-entropy loss.

[0081] In step S633, the model parameters of the initial abnormal classification model are adjusted according to the first target loss function until the model converges to obtain a target abnormal classification model.

[0082] As an optional implementation, in the technical solution provided in step S506, the media stream adjustment network element can determine the first processing strategy according to the following method, including: analyzing the first abnormal type by using the pre-trained target strategy decision model to obtain the corresponding first processing strategy. The first processing strategy includes but is not limited to adjusting the code rate to a target code rate value, adjusting the session description protocol parameter to a target parameter value, adjusting the bandwidth to a target bandwidth value, etc.

[0083] Optionally, the training process of the target policy decision model comprises the following steps S71-S73, wherein:

[0084] Step S71, obtaining a second training sample set and a second sample label set. The second training sample set comprises a plurality of training samples composed of a second abnormal type of media stream corresponding to a plurality of historical new call services, and the second sample label set comprises a second processing policy corresponding to each second training sample as a second sample label.

[0085] Since network resources (such as bandwidth, processing capacity, etc.) are limited, especially in high-density user areas or network congestion situations, effective resource allocation is crucial for network stability and efficiency. Therefore, the second processing policy must be the one with the lowest network cost. At the same time, for any communication service, ensuring user experience is always the top priority. Strategies with high user ratings mean that these strategies can provide higher quality services in actual applications, such as smoother video calls, clearer voice quality, or faster data transmission speed. Network cost and user experience are two interrelated factors. Excessive pursuit of resource saving may sacrifice user experience, while providing high experience services may lead to resource waste. Therefore, when training the model, a balance between the two needs to be found to optimize the overall performance. Therefore, the second processing policy needs to be the one with the lowest network cost and the highest user rating, so that the model can meet user experience while minimizing the consumption of network resources, achieving the dual goals of economy and service quality.

[0086] Step S72, constructing an initial policy decision model. The model architecture of the initial policy decision model can be a decision tree or random forest, support vector machine (SVM), neural network, etc. In the new call scenario, a model that can quickly respond to network changes, process real-time data, and has high accuracy may be needed. Therefore, deep learning models, especially those that can handle time series data (such as recurrent neural networks RNN, long short-term memory LSTM networks) and large-scale data sets (such as deep neural networks DNN, Transformers), may be the preferred architecture. At the same time, considering the real-time and lightweight of the model, reinforcement learning or ensemble learning may also be an effective method to build a policy decision model, which is not limited in this application.

[0087] Step S73, iteratively training the initial policy decision model using the second training sample set and the second sample label set to obtain the target policy decision model.

[0088] Specifically, in the technical solution provided in the above step S73, the method can include:

[0089] Step S731, the second training sample in the second training sample set is input into the initial policy decision model in batches, and each predicted processing strategy output by the initial policy decision model is obtained.

[0090] Step S732, a second target loss function is constructed according to the predicted processing strategy of each batch of second training samples and the corresponding second processing strategy. The second target loss function can be a policy gradient loss, a Q-Learning loss, an Actor-Critic loss, etc.

[0091] Step S733, the model parameters of the initial policy decision model are adjusted according to the second target loss function until the model converges, and a target policy decision model is obtained.

[0092] Further, since different control network elements in the new call network system bear different functions, the functional specificity of the control network elements determines that they can effectively execute the policies related to their functions. Specifically, the control network elements can be divided into: MF (Media Function) network elements, VoLTE application servers, policy and charging rule function network elements PCRF (Policy and Charging Rules Function) / packet control function network elements PCF (Policy Control Function), wherein the MF network element is mainly responsible for the management of media resources and the forwarding of traffic, and is therefore closely related to the code rate adjustment policy; the VoLTE application server provides voice and video call services, and therefore plays a key role in resolution and frame rate adjustment policies; the PCRF network element / PCF network element is used to control network resources and policy execution, and is suitable for dynamic bandwidth adjustment. Therefore, the first processing strategy is different, and the control network elements and the interaction process of the media stream adjustment network element are also different. Specifically:

[0093] In the case where the first processing strategy is to adjust the code rate to a target code rate value, the media stream adjustment network element can send a code rate adjustment request message carrying the target code rate value to the MF network element, wherein the MF network element is used to send a real-time transport control protocol request message carrying a temporary maximum code rate request field with the target code rate value to the calling user equipment and the called user equipment, and request the calling user equipment and the called user equipment to adjust the code rate of the media stream corresponding to the first new call service;

[0094] In a case where the first processing strategy is to adjust the session description protocol parameter to a target parameter value, the media stream adjustment network element can send a session description protocol parameter adjustment request message carrying the target parameter value to a VoLTE application server, where the VoLTE application server is configured to send a Re-invite request message carrying the target parameter value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust the session description protocol parameter of the media stream corresponding to the first new call service.

[0095] In a case where the first processing strategy is to adjust the bandwidth to a target bandwidth value, the media stream adjustment network element can send a bandwidth adjustment request message carrying the target bandwidth value to a policy and charging rules function network element / packet control function network element, where the policy and charging rules function network element / packet control function network element is configured to send a bandwidth adjustment instruction carrying the target bandwidth value to a core network network element to instruct the core network network element to adjust the bandwidth of the media stream corresponding to the first new call service.

[0096] In the media stream adjustment method for a new call service, the quality of the media stream is intelligently analyzed by a large model, possible call problems are predicted, and the most suitable adjustment strategy is automatically selected. This not only improves the accuracy of adjustment, but also reduces the dependence on manual intervention. Compared with traditional methods that rely on rules or thresholds, the embodiments of the present application can better adapt to complex and variable network environments and user needs. Further, the technical problem that related media stream adjustment methods only rely on static threshold judgment or user feedback for exception and processing, resulting in the inability to quickly and effectively optimize call quality, is solved.

[0097] Embodiment 3

[0098] According to the embodiments of the present application, a computer program product is also provided, which includes a computer program. When the computer program is executed by a processor, the media stream adjustment method for a new call service in the embodiment 2 is implemented.

[0099] According to the embodiments of the present application, a non-volatile storage medium is also provided, which includes a stored computer program. The device where the non-volatile storage medium is located executes the media stream adjustment method for a new call service in the embodiment 2 by running the computer program.

[0100] According to the embodiments of the present application, a processor is also provided, which is used to run a computer program. When the computer program is run, the media stream adjustment method for a new call service in the embodiment 2 is executed.

[0101] According to the embodiment of the present application, a network device is also provided, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the media stream adjustment method of the new call service in the embodiment 2 by using the computer program.

[0102] Specifically, the computer program performs the following steps when running: obtaining a first transmission data set of a media stream corresponding to a first new call service, wherein the first transmission data set comprises a plurality of first transmission data; in a case where a first target transmission data in the first transmission data set exceeds a preset threshold value, analyzing the first target transmission data by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media stream corresponding to the first new call service; determining a first processing strategy corresponding to the first abnormal type, and sending the first processing strategy to a control network element, wherein the control network element is configured to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service to be within the threshold value.

[0103] As an optional implementation, the network device can exist in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 6 A hardware structure block diagram of a network device for implementing the media stream adjustment method of the new call service is shown. As shown in the figure, Figure 6 The network device 60 can include one or more (in the figure, 602a, 602b, …, 602n are used to show) processors 602 (the processor 602 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 604 for storing data, and a transmission device 606 for communication function. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that, Figure 6 The structure shown in the figure is only schematic, and it does not limit the structure of the network device described above. For example, the network device 60 can also include more or fewer components than those shown in the figure, or have a different configuration from that shown in the figure. Figure 6 For example, the network device 60 can also include more or fewer components than those shown in the figure, or have a different configuration from that shown in the figure. Figure 6 For example, the network device 60 can also include more or fewer components than those shown in the figure, or have a different configuration from that shown in the figure.

[0104] It should be noted that the one or more processors 602 and / or other data processing circuitry described above can be referred to herein generically as "data processing circuitry." The data processing circuitry can be embodied in whole or in part as software, hardware, firmware, or any combination thereof. Furthermore, the data processing circuitry can be a single standalone processing module, or it can be incorporated in whole or in part within any one of the other elements of network device 60. As referred to herein, the data processing circuitry functions as a processor to control, for example, selection of variable resistance terminal paths in connection with an interface.

[0105] Memory 604 can be used to store software programs and modules for application software, such as program instructions / data storage for the media stream adjustment method for new call service described in embodiments of the present application. Processor 602 can execute various functions and data processing, i.e., implement the vulnerability detection method for application programs described above, by running software programs and modules stored in memory 604. Memory 604 can include high-speed random access memory and can also include nonvolatile memory, such as one or more magnetic data storage devices, flash memory, or other nonvolatile solid state memory. In some examples, memory 604 can further include memory that is remotely located with respect to processor 602, which can be connected to network device 60 via a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communications network, and combinations thereof.

[0106] Transmitting device 606 is configured to receive or send data via a network. Examples of the network include, but are not limited to, a wireless network provided by a communications provider of network device 60. In one example, transmitting device 606 includes a network interface controller (NIC) that can be connected to other network devices via a base station to communicate with the Internet. In one example, transmitting device 606 can be a radio frequency (RF) module configured to communicate with the Internet via wireless means.

[0107] Display can be, for example, a touch screen type liquid crystal display (LCD) that can enable a user to interact with a user interface of network device 60.

[0108] The above-mentioned embodiment numbers are merely for description, and do not represent the advantages or disadvantages of the embodiments.

[0109] In the above-described embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0110] In several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented by other ways. Among them, the above-described device embodiments are only schematic, for example, the division of units can be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0111] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e. they can be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0112] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0113] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the technical solutions or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0114] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be considered as the protection scope of the present application.

Claims

1. A method for adjusting media stream of a new call service, characterized in that, The method comprises the following steps: acquiring a first transmission data set of a media stream corresponding to a first new call service, wherein the first transmission data set comprises a plurality of first transmission data; in a case where first target transmission data in the first transmission data set exceeds a preset threshold value, analyzing the first target transmission data by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media stream corresponding to the first new call service; determining a first processing strategy corresponding to the first abnormal type, and sending the first processing strategy to a control network element, wherein the first processing strategy is obtained by analyzing the first abnormal type by using a pre-trained target strategy decision model; and the first processing strategy comprises at least one of the following: adjusting a code rate to a target code rate value, adjusting a session description protocol parameter to a target parameter value, and adjusting a bandwidth to a target bandwidth value; and the control network element is configured to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service to be below the threshold value.

2. The method of claim 1, wherein, The training process of the target anomaly classification model comprises the following steps: acquiring a first training sample set and a first sample label set, wherein the first training sample set comprises a plurality of first training samples composed of second target transmission data exceeding a preset threshold value in a second transmission data set of a media stream corresponding to a historical new call service, the first sample label set comprises a second abnormal type corresponding to each first training sample as a first sample label, and the second target transmission data comprises at least one of the following: bandwidth, packet loss rate, code rate, frame rate, resolution, jitter rate, and time delay; and the second abnormal type comprises at least one of the following: black screen, blue screen, mosaic, screen flower, and voice discontinuity; constructing an initial anomaly classification model; iteratively training the initial anomaly classification model by using the first training sample set and the first sample label set to obtain the target anomaly classification model.

3. The method of claim 2, wherein, The iteratively training the initial anomaly classification model by using the first training sample set and the first sample label set to obtain the target anomaly classification model comprises the following steps: inputting a batch of first training samples in the first training sample set into the initial anomaly classification model in sequence to obtain each predicted abnormal type output by the initial anomaly classification model; constructing a first target loss function according to the predicted abnormal type of each first training sample in the batch and the corresponding first sample label; adjusting model parameters of the initial anomaly classification model according to the first target loss function until the model converges to obtain the target anomaly classification model.

4. The method of claim 1, wherein, The training process of the target strategy decision model comprises the following steps: acquiring a second training sample set and a second sample label set, wherein the second training sample set comprises a plurality of training samples composed of a second abnormal type of a media stream corresponding to a plurality of historical new call services, and the second sample label set comprises a second processing strategy corresponding to each second training sample as a second sample label, and the second processing strategy is a processing strategy with the lowest network cost and the highest user evaluation; constructing an initial strategy decision model; The initial strategy decision model is iteratively trained by using the second training sample set and the second sample label set, to obtain the target strategy decision model.

5. The method of claim 4, wherein, The initial strategy decision model is iteratively trained by using the second training sample set and the second sample label set, to obtain the target strategy decision model, including: The batch of second training samples in the second training sample set are sequentially input into the initial strategy decision model, to obtain each predicted processing strategy output by the initial strategy decision model; A second target loss function is constructed according to the predicted processing strategy of each second training sample in the batch and the corresponding second processing strategy; The model parameters of the initial strategy decision model are adjusted according to the second target loss function, until the model converges, to obtain the target strategy decision model.

6. The method of claim 1, wherein, The control network element includes a media function (MF) network element, a voice over long term evolution (VoLTE) application server, a policy and charging rules function (PCRF) network element / packet control function (PCF) network element, wherein the first processing strategy is sent to the control network element, including: In a case where the first processing strategy is to adjust a code rate to a target code rate value, a code rate adjustment request message carrying the target code rate value is sent to the MF network element, wherein the MF network element is configured to send a real-time transport control protocol (RTCP) request message carrying a temporary maximum code rate request field as the target code rate value to a calling user equipment (UE) and a called UE, to request the calling UE and the called UE to adjust a code rate of a media stream corresponding to the first new call service; In a case where the first processing strategy is to adjust a session description protocol (SDP) parameter to a target parameter value, an SDP parameter adjustment request message carrying the target parameter value is sent to the VoLTE application server, wherein the VoLTE application server is configured to send a Re-invite request message carrying the target parameter value to the calling UE and the called UE, to request the calling UE and the called UE to adjust an SDP parameter of a media stream corresponding to the first new call service; In a case where the first processing strategy is to adjust a bandwidth to a target bandwidth value, a bandwidth adjustment request message carrying the target bandwidth value is sent to the PCRF network element / the PCF network element, wherein the PCRF network element / the PCF network element is configured to send a bandwidth adjustment instruction carrying the target bandwidth value to a core network element, to instruct the core network element to adjust a bandwidth of a media stream corresponding to the first new call service.

7. A new call network system, characterized by, The new call network system at least includes a media stream adjustment network element and a control network element, wherein The media stream adjustment network element is configured to: acquire a first transmission data set of a media stream corresponding to a first new call service, wherein the first transmission data set includes a plurality of first transmission data; in a case where first target transmission data in the first transmission data set exceeds a preset threshold value, analyze the first target transmission data by using a pre-trained target anomaly classification model to obtain a first abnormal type of the media stream corresponding to the first new call service; determine a first processing strategy corresponding to the first abnormal type, and send the first processing strategy to the control network element, wherein the first abnormal type is analyzed by using a pre-trained target strategy decision model to obtain the corresponding first processing strategy; and the first processing strategy includes at least one of the following: adjusting a code rate to a target code rate value, adjusting a session description protocol parameter to a target parameter value, and adjusting a bandwidth to a target bandwidth value. The control network element is configured to execute the first processing strategy to adjust the first target transmission data of the media stream corresponding to the first new call service to be below the threshold value.

8. The new call network system according to claim 7, wherein, The new call network system further includes a calling user equipment, a called user equipment, and a core network element, and the control network element includes an MF network element, a VoLTE application server, and a policy and charging rules function network element / packet control function network element. The media stream adjustment network element is further configured to, in a case where the first processing strategy is to adjust the code rate to the target code rate value, send a code rate adjustment request message carrying the target code rate value to the MF network element; in a case where the first processing strategy is to adjust the session description protocol parameter to the target parameter value, send a session description protocol parameter adjustment request message carrying the target parameter value to the VoLTE application server; and in a case where the first processing strategy is to adjust the bandwidth to the target bandwidth value, send a bandwidth adjustment request message carrying the target bandwidth value to the policy and charging rules function network element / packet control function network element. The MF network element is configured to send a real-time transport control protocol request message carrying a temporary maximum code rate request field with the target code rate value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust the code rate of the media stream corresponding to the first new call service. The VoLTE application server is configured to send a Re-invite request message carrying the target parameter value to the calling user equipment and the called user equipment to request the calling user equipment and the called user equipment to adjust the session description protocol parameter of the media stream corresponding to the first new call service. The policy and charging rules function network element / packet control function network element is configured to send a bandwidth adjustment instruction carrying the target bandwidth value to the core network element to instruct the core network element to adjust the bandwidth of the media stream corresponding to the first new call service.

9. A network device, comprising: The method comprises the following steps: A memory and a processor, wherein the memory has stored therein a computer program, and the processor is configured to execute the method of adjusting media stream of new call service according to any one of claims 1 to 6 by the computer program.

Images