Service quality guarantee method and device, processing equipment, storage medium and program product

Abstract

The invention discloses a service guarantee method, processing equipment, a storage medium and a program product, and the method comprises the steps: obtaining the network index characteristics and park characteristics of a park, and enabling the park characteristics to comprise the business type of the park and the busy time period of the park; predicting a service quality category of the park based on the network index characteristics and the park characteristics through a service awareness model; a park guarantee strategy is determined based on the service quality category, and the park guarantee strategy is used for guaranteeing the service quality of the park; in this way, real-time end-to-end service quality perception for the park service is realized.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a service quality assurance method and apparatus, processing equipment, computer-readable storage medium, and computer program product. Background Art

[0002] 5G applications have been integrated into various industries. In enterprise (To Business, ToB) scenarios, video services in 5G private networks require low latency and large bandwidth. Therefore, a video quality assurance system is proposed to achieve service quality perception by optimizing video services and make adjustments when the service quality is poor. However, 5G private networks have different campus scenarios, and the services in each 5G private network campus scenario have different characteristics. Existing technologies are time-consuming and labor-intensive. Therefore, how to combine video service quality assessment with the real-time characteristics of the campus to provide end-to-end low-latency video quality assurance is a technical problem that needs to be solved in this field. Summary of the Invention

[0003] To solve the above technical problems, embodiments of the present invention provide a service quality assurance method and apparatus, a processing device, a computer-readable storage medium, and a computer program product.

[0004] In a first aspect, an embodiment of the present application provides a method for ensuring service quality, including:

[0005] Obtaining network indicator characteristics and park characteristics of the park, wherein the park characteristics include the business type of the park and the busy time period of the park;

[0006] Predicting the service quality category of the park based on the network indicator characteristics and the park characteristics through a service perception model;

[0007] A campus protection policy is determined based on the service quality category, wherein the campus protection policy is used for service quality protection of the campus.

[0008] In a second aspect, an embodiment of the present application provides a service quality assurance device, including:

[0009] An acquisition unit, configured to acquire network indicator characteristics and park characteristics of a park, wherein the park characteristics include a business type of the park and a busy time period of the park;

[0010] A processing unit, configured to predict the service quality category of the park based on the network indicator characteristics and the park characteristics through a service perception model;

[0011] The determining unit is configured to determine a park security policy based on the service quality category, wherein the park security policy is used for service quality assurance of the park.

[0012] In a third aspect, the processing device provided by the embodiment of the present application includes: a processor and a memory, the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute any one of the above-mentioned service quality assurance methods.

[0013] In a fourth aspect, the computer-readable storage medium provided in an embodiment of the present application is used to store a computer program, and the computer program enables a computer to execute any one of the above methods.

[0014] In a fifth aspect, the computer program product provided in the embodiments of the present application includes computer program instructions, which enable a computer to execute any one of the above methods.

[0015] The technical solution of the embodiments of this application achieves service quality assurance for the park by acquiring network and park characteristics, using a service perception model to predict the park's service quality category based on these characteristics, and determining a park assurance strategy based on the service quality category. This approach, on the one hand, enables end-to-end service quality perception of park services, acquiring service quality perception results in real time; on the other hand, it matches the park service quality perception results with corresponding assurance strategies, enabling automatic service adjustments and optimizing low-quality services. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flowchart of a service quality assurance method according to an embodiment of the present application;

[0017] Figure 2 This is a type distribution diagram of a service quality assurance method according to an embodiment of the present application;

[0018] Figure 3 This is a flow rate ratio diagram of a service quality assurance method according to an embodiment of the present application;

[0019] Figure 4 This is a campus monitoring diagram of a service quality assurance method according to an embodiment of the present application;

[0020] Figure 5 This is a scenario-based schematic diagram of a service quality assurance method according to an embodiment of the present application;

[0021] Figure 6 This is a model training flow chart of a service quality assurance method according to an embodiment of the present application;

[0022] Figure 7 This is a schematic diagram of an evaluation system for a service quality assurance method according to an embodiment of the present application;

[0023] Figure 8This is a schematic diagram of the structure of the service quality assurance device provided in an embodiment of the present application;

[0024] Figure 9 This is a schematic structural diagram of a processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0026] With the large-scale construction and operation of 5G networks, 5G applications in vertical industries are gradually expanding, and 5G applications have been integrated into various industries. A 5G private network (Private 5G Network) is a local area network (LAN) built based on 5G technology, featuring unified connectivity, optimized services, and secure communication within a specific area. A 5G private network can provide secure, reliable, and customized network services for individual (ToC) users. However, compared to ToC scenarios, in ToB scenarios, private networks exist across multiple campuses, each with different service characteristics and varying service quality requirements. For example, high-definition video services in 5G private networks require low latency and high bandwidth. However, different video services exist in different campuses, such as HD video, live streaming, on-demand video, and surveillance video. Video services in different campuses have different characteristics, so private networks require a video quality assurance system that can monitor the end-to-end video service quality in real time and make timely adjustments when quality deteriorates to ensure low latency for video services. The prior art discloses a cross-layer resource scheduling method for multi-user real-time video streams in 5G scenarios. First, during the user access adjustment period, based on network state-related parameters perceived by the 5G base station, a demand model is used to calculate the minimum bandwidth required for each user to achieve the user-perceived quality of service (QoE) requirement. The user flow congestion factor for each user is then constructed based on the minimum bandwidth. Second, the priority of each user is determined based on the minimum bandwidth and the user flow congestion factor. The minimum bandwidth is allocated to each user in descending order of priority until the remaining bandwidth does not meet the user access requirement. Finally, the bandwidth of each user is continuously adjusted based on the user's congestion status to minimize the number of users in the congested state. However, the prior art does not take into account the existence of multiple campuses in 5G private networks, each with different service characteristics, and is therefore time-consuming and labor-intensive. Furthermore, it does not combine video service quality assessment with the real-time characteristics of the campuses, resulting in little interaction between quality assessment and video quality assurance. To this end, considering that 5G private networks have different campus scenarios and that services vary significantly between campuses, 5G private networks need to provide customized solutions to meet different needs. Therefore, the technical solutions of the embodiments of this application are proposed.

[0027] To facilitate understanding of the technical solutions of the embodiments of the present application, the technical solutions of the present application are described in detail below through specific embodiments. The above related technologies can be combined arbitrarily with the technical solutions of the embodiments of the present application as optional solutions, and all of them fall within the scope of protection of the embodiments of the present application. The embodiments of the present application include at least part of the following contents.

[0028] Figure 1This is a flow chart of a service quality assurance method provided by an embodiment of the present application, such as Figure 1 As shown, the service quality assurance method includes:

[0029] Step 101: Obtain network indicator characteristics and park characteristics of the park, wherein the park characteristics include the business type of the park and the busy time period of the park.

[0030] Here, due to the different distribution ratios, business volumes, and busy time periods of video services in different parks, the network requirements of park video services also vary greatly. Therefore, network indicator characteristics and park characteristics of different parks are obtained separately.

[0031] In some embodiments, the network indicator characteristics of different campuses are obtained by real-time parsing of network indicator data of the Transmission Control Protocol (TCP) at the transport layer and the Real Time Streaming Protocol (RTSP) at the application layer. For example, the network indicator characteristics of different campuses can be obtained by real-time parsing of network indicator data of the TCP protocol at the transport layer and the RTSP protocol at the application layer through the N6 pointer port, with minute-by-minute accuracy.

[0032] In some implementations, the park feature can be the park's service type. Since video services have relatively high network requirements, and different video service types have different requirements for communication indicators such as latency, jitter, and packet loss rate, the video service type is refined according to the park scenario requirements. Here, refer to Figure 2 , Figure 2 This is a type distribution diagram of a service quality assurance method provided by the embodiment of this application. Figure 2 As shown, the distribution of video service types in different parks is different. In some implementations, based on research and analysis, video service types can be divided into conference type, i.e., high-definition video type services, live type, i.e., live video type services, on-demand type, i.e., on-demand video type services, and monitoring type, i.e., monitoring video type services. Among them, high-definition video type services can be understood as high-definition conference videos. Due to the popularity of new office models, video conferencing applications are increasing, so there is a demand for low latency for high-definition video type services. Live video services can be understood as uplink live broadcast services, and there are requirements for high interactivity and low latency for uplink live broadcast services. On-demand video services can be understood as multimedia on-demand and video download services. Such services have a high proportion of video services and are relatively insensitive to latency. Monitoring video services can be understood as high-definition video monitoring, which is used for centralized video monitoring in the park and requires real-time management.

[0033] In some embodiments, the park feature may be the busy time period of the park. Figure 3, Figure 3 This is a flow chart of a service quality assurance method provided by an embodiment of the present application. Figure 3 As shown in Figure 1, Park A and Park B are two different parks. The daily video traffic curves of each park are different. Therefore, it can be understood that due to the different production modes of different parks, the busyness of video services at different times is also different. Therefore, it is necessary to dynamically perceive the busy time period of the park video service. Here, the busy time period T of the park can be calculated based on the business traffic data using formulas (1) and (2). b .

[0034]

[0035]

[0036] Among them, U is the average video traffic in the past seven days; d represents the current day, d-1 represents yesterday; h is the 30-minute granularity time period mapped every day, and the constant is represented by H; u t,h It is expressed as the video transmission volume in time period h on day t, and the constant is represented by T; u Tb For time T b The amount of video data transmitted during a given period; α is the busy ratio coefficient, typically set to 1.5. A busy period is defined as a period when the video traffic exceeds a certain percentage α of the average video traffic. The average video traffic is the average of the video traffic over the past seven days. If the campus traffic is relatively evenly distributed throughout the day, there is no busy period on that day.

[0037] Step 102: Predict the service quality category of the park based on the network indicator characteristics and park characteristics through the service perception model.

[0038] In some implementations, the service perception model is used to evaluate the video service quality of different campuses in real time. Specifically, the service quality category of the campus is predicted based on network indicator characteristics and campus characteristics.

[0039] In some embodiments, the options for service quality categories include: a first quality category, a second quality category, and a third quality category, wherein the service quality corresponding to the first quality category is higher than the service quality corresponding to the second quality category, and the service quality corresponding to the second quality category is higher than the service quality corresponding to the third quality category.

[0040] Here, this application divides service quality categories into three quality categories, which are used by the service perception model to indicate video service quality results. The first quality category can be high-quality services, that is, the service perception model evaluates service quality in real time, and its network performance has a relatively ample safety margin; the second quality category can be good services, that is, the service perception model evaluates service quality in real time as between high and poor quality, and its network performance is normal and can meet the needs of normal service operation, but the safety margin is small; the third quality category can be poor quality services, that is, the service perception model evaluates service quality in real time, and its network performance cannot meet the needs of normal service operation.

[0041] Step 103: Determine a park assurance policy based on the service quality category, wherein the park assurance policy is used to ensure service quality of the park.

[0042] Here, the service perception model predicts three service quality categories based on the park's network metrics and park characteristics. Based on these three service quality categories, the park's assurance policies are determined, implementing automated adjustments to ensure service quality. Once the service perception model predicts the three service quality categories, the park system displays the predictions on the system's large screen. Specifically, the park system converts complex network metric data into easily understandable indicators for park users, reflecting the operational status of the industry network's video services.

[0043] In some implementations, in step 103, determining a campus security strategy based on the service quality category includes: if the service quality category is the third quality category, determining an encoding strategy, a compression strategy, and a storage strategy for the campus's service data.

[0044] Here, if the service perception model predicts in real time that the video service quality category is the third quality category, that is, the video service quality is poor quality service, then according to the service type of the park: high-definition video service, live video service, on-demand video service and monitoring video service, the encoding strategy, compression strategy and storage strategy of the park video service are determined respectively.

[0045] In some embodiments, the encoding strategy includes: if the business type of the park is a high-definition video business, a live video business, or a video-on-demand business, and the park is in a busy period, then the encoding rate of the park's business data is lowered by one or more levels.

[0046] Here, if the service perception model predicts in real time that the video service quality is poor quality, the park's service type is high-definition video service, live video service, or on-demand video service, and the park is in a busy period, the service experience can be improved by adjusting the bit rate. Specifically, when the park's video conferencing or uplink live broadcast service is predicted by the service perception model to be the third quality category, i.e., poor quality service, the priority of the video service that is not sensitive to delay requirements can be lowered, and the service with a high priority level can be processed first. The video bit rate is divided into {0, 1, 2, 3} according to the clarity, which respectively represent ordinary L normal , SD L good , HD L HD , Ultra HD L UHD The higher the level, the higher the network requirements. When the park is in a non-busy period, that is, when the park business is idle, the default playback bit rate of the park video is L HD If you need high definition, you can manually adjust the playback bit rate to L UHD When the zone is in a busy period, that is, when the park business is busy, the default playback bit rate of the park video is L good If necessary, you can manually increase the bit rate.

[0047] In some embodiments, if the campus is in a busy period, when the service perception model determines that the minute-by-minute video service quality type is the third quality category (poor quality service) for N consecutive times, that is, the video quality is in a poor state for N minutes and no more network resources can be coordinated, the system automatically reduces the bit rate of the campus service data by one or more levels. Here, the value of N is generally 2 to 5, and this application does not specifically limit the value of N.

[0048] In some implementations, the compression strategy includes: if the business type of the park is a surveillance video business, locally compressing the business data of the park.

[0049] Here, if the service perception model predicts in real time that the video service quality is poor, and the park's service type is a surveillance video service, the service experience can be improved by locally compressing the park's service data. Figure 4 , Figure 4 This is a campus monitoring diagram of a service quality assurance method provided by an embodiment of the present application. Figure 4 As shown in the figure, to ensure safe production, the park's global monitoring generates a large amount of surveillance video every day. Uploading this surveillance video to the cloud network consumes a large amount of network resources. Therefore, by compressing the video service data locally and processing it before uploading it to the cloud network, the pressure on the park network environment is alleviated and the quality of park services is guaranteed.

[0050] In some implementations, edge computing is first performed, and the intelligent video analysis service is deployed on the local server to intelligently identify abnormal events and compress the surveillance video. Without affecting the image quality, video compression can reduce the memory size by about 80%, while ensuring the quality of the captured clips when abnormal events occur. When the abnormal monitoring time T is identified, the abnormal event is detected. abn , real-time transmission of alarm segments, generating alarm segments {T abn +T interval , T abn -T interval}, where T interval is a constant, such as T interval The value can be 10s. Key alarm clips are uploaded to the core network, enabling remote applications to monitor campus conditions in real time. If a remote user requests to view the video, compressed video is uploaded in real time. During busy campus hours, upload to the core network can be delayed until the peak has passed.

[0051] In some implementations, the storage strategy includes: if the business type of the park is a video-on-demand business, locally caching the business data of the park.

[0052] If the service perception model predicts in real time that the video service quality is poor and the campus service type is on-demand video, the service experience can be improved by locally caching the campus service data. Specifically, on-demand video services include campus on-demand services and video download services. Campus on-demand services are less sensitive to latency, while video download services have looser latency requirements. These two types of video services have lower priority and experience high concentrations of traffic, such as concentrated online learning and training courses and popular campus services. Therefore, on-demand services can be optimized through edge caching to ensure service quality. First, the campus's service popularity is calculated based on the number of on-demand video service visits over the past seven days using historical campus data. This calculates the campus's popular video resources. Highly accessed video resources are then pre-cached to local devices or servers, reducing video service latency and alleviating pressure on campus network resources.

[0053] In some implementations, if the service quality category is the third quality category, a service alarm log is generated, and the service alarm log includes network indicator characteristics and the campus characteristics.

[0054] If the service perception model predicts the video service quality as Category III in real time, the video service type corresponding to the service data is considered poor quality. A video service alarm record corresponding to the current time is generated, and a service alarm log is generated based on the campus system's alarm policy. The campus system can automatically adjust measures based on the service alarm log to ensure service quality.

[0055] As can be seen from the above, the service quality assurance method provided by the embodiments of this application achieves service quality assurance for the park by obtaining network indicator characteristics and park characteristics of the park, predicting the park's service quality category based on the network indicator characteristics and park characteristics through a service perception model, and determining the park assurance strategy based on the service quality category. In this way, on the one hand, end-to-end service quality perception of the park's services is achieved, and service quality perception results are obtained in real time; on the other hand, the corresponding assurance strategy is matched to the park's service quality perception results, achieving automatic service adjustment and optimization of poor-quality services.

[0056] Figure 5 This is a scenario diagram of a service quality assurance method provided by an embodiment of the present application, such as Figure 5 As shown, based on campuses, training data for service quality assessment and assurance is maintained based on different 5G private network scenarios, campus characteristics, and real-time network indicator characteristics. The service perception model is regularly and automatically iterated to achieve real-time, end-to-end video service quality perception. If the service quality category is predicted to be the third quality category (poor quality), a system alarm is pushed, and the campus assurance policy is automatically adjusted and optimized to optimize the video service in the third quality category. This optimizes the video service status under existing network resources and updates the gene library. If the service quality category is predicted to be the first or second quality category, a system alarm log is generated and displayed on the platform screen.

[0057] Figure 6 This is a model training flow chart of a service quality assurance method provided by an embodiment of the present application, such as Figure 6 As shown, the service perception model is a trained service perception model, and the specific training steps include:

[0058] Step 601: Build a gene library, which includes historical park characteristics, historical network indicator characteristics, and historical service quality category labels.

[0059] Here, the gene bank can be understood as a database. The park's gene bank is composed of historical park characteristics, historical network indicator characteristics, and historical business quality category labels.

[0060] In some embodiments, historical park characteristics include historical business types and historical busy time periods; constructing a gene library includes: detecting historical business in the park; classifying historical business to obtain historical business types; determining historical busy time periods based on historical business traffic; and constructing a gene library based on historical business types and historical busy time periods.

[0061] Here, the service perception model detects the park's historical services by acquiring historical video services and classifying them according to communication indicators and park scenario requirements. The resulting historical service types include HD video services, live video services, on-demand video services, and surveillance video services. HD video services can be understood as HD conferencing videos. With the rise of new office models, video conferencing applications are increasingly common, requiring low latency for HD video services. Live video services can be understood as uplink live streaming services, requiring high interactivity and low latency. On-demand video services can be understood as multimedia on-demand and video download services, which have a high video traffic ratio and are relatively insensitive to latency. Surveillance video services can be understood as HD video surveillance, which is used for centralized park video monitoring and requires real-time management.

[0062] In some embodiments, based on the traffic of historical services, a historical busy time period is determined, including: if within a first time unit, the traffic of historical services is greater than or equal to a certain proportion of the average video traffic, then it is determined that the first time unit belongs to a historical busy time period; wherein the average video traffic is the average of the traffic of historical services in N second time units before the first time unit, and N is a positive integer.

[0063] Here, the busy time period T of the park can be calculated based on the traffic data of historical services with a 30-minute granularity by using the following formulas (3) and (4): b .

[0064]

[0065]

[0066] Among them, U is the average value of the historical service traffic of N second time units before the first time unit, that is, it can be the average video traffic of the past seven days, and N is a positive integer; d represents the first time unit, which can be understood as the current day, and d-1 represents yesterday; h is the 30-minute granularity time period mapped every day, and the constant is represented by H; u t,h represents the video transmission volume in time period h on day t, and the constant is represented by T; u Tb For time T b The peak time period is determined by the video traffic volume at that time; α is the busy ratio coefficient, generally set to 1.5. If, within the first time unit, the video traffic exceeds a certain percentage α of the average video traffic, the park is considered to be in a busy period. The average video traffic is the average of the video traffic over the past seven days. If the park's daily traffic is relatively even, that is, within the first time unit, the historical traffic volume is less than a certain percentage of the average video traffic, then the park is not in a busy period that day.

[0067] In some embodiments, the park system builds a gene library based on historical service types and historical busy time periods. Specifically, by classifying historical service types according to communication indicators, the service classification results are mapped into digital features, for example, conference, live broadcast, on-demand, and monitoring are mapped into digital features [1, 2, 3, 4], and the classification results are converted into digital features and written into the gene library. b To determine whether the minute intensity data is within the busy period, the result is written into the gene library. For example, if the park has a busy period, it can be represented as 1; if the park does not have a busy period, it can be represented as 0, and the mapping result is written into the gene library.

[0068] In some embodiments, constructing a gene library includes: collecting historical network indicators of a park; calculating the importance of the historical network indicators, and screening out historical network indicators whose importance is greater than or equal to a threshold from the collected historical network indicators; and constructing a gene library based on the historical network indicators whose importance is greater than or equal to the threshold.

[0069] Here, the historical network indicators for different campuses can be minute-by-minute TCP and RTSP data, analyzed through the N6 pointer port. The model's accuracy and computational speed are improved by calculating the importance of these historical network indicators. This involves simplifying the features of the collected historical network indicator data and selecting those with importance greater than or equal to a threshold.

[0070] In some implementations, the collected historical network indicator data may be feature simplified using a random forest algorithm. Specifically, the simplified historical network indicator data may be obtained using the following formula (5).

[0071] X net ={f impt (x i )>0} (5)

[0072] Among them, f impt Represents the importance calculation function; X={x0,x1,....x n} represents historical network metric data.

[0073] In some implementations, historical service quality category labels can be obtained through rule labeling, collecting active feedback from the current network and users, abnormal data and alarm data, etc. Specifically, obtaining through rule labeling is to collect historical data of the park based on the service requirements of latency, jitter, etc. of different parks, and use video service rules to label the network data into categories. Obtaining through collecting active feedback from the current network and users is to collect feedback from the current network and users through the user feedback module, correct the category labels of the data, and enrich the diversity of data labeling. Obtaining through abnormal data and alarm data is to improve the video service quality data through abnormal data and alarm data of the park service, where abnormal data includes data such as session failure, no response, high latency, and high jitter.

[0074] Step 602: Construct training data based on historical campus characteristics, historical network indicator characteristics, and historical service quality category labels.

[0075] The training data generated here includes three parts: historical campus characteristics, namely historical service types and historical busy time periods; historical network indicator characteristics with importance greater than or equal to a threshold selected from the collected historical network indicators; and historical service quality category labels obtained through rule labeling, collection of live network and user active feedback, abnormal data, and alarm data to construct training data.

[0076] Step 603: Train the service perception model based on the training data.

[0077] In some embodiments, training the business perception model based on the training data includes: training the business perception model based on the training data according to a preset period; or training the business perception model based on the training data when obtaining an update operation of the business perception model.

[0078] Here, the business perception model is trained by using end-to-end training data. The business perception model can be a neural network model. Of course, the business perception model can also be other multi-classification models, which is not limited in this application. When the gene library is maintained and / or video data is recently added, the business perception model is automatically updated and iterated every two weeks or every month according to the preset cycle, or when the business adjustment is relatively large, it can be manually updated with one click. Specifically, the business perception model can be expressed by the following formula (6):

[0079] Y=F θ (X) (6)

[0080] Where θ represents a parameter vector. The training data can be expressed as X = {historical campus characteristics, historical network parameter indicator characteristics}. The service quality category Y is output, and the historical service quality category label corresponding to the training data X is denoted as Y'. X is input into the service perception model, and the service quality category Y is output by the service perception model. The loss value between the service quality category Y and the service quality category label Y' is calculated, and the parameters of the service perception model are updated based on this loss value. The training data X is input into the updated service perception model, and the service perception model outputs Y. The loss value between Y and Y' is again calculated, and the parameters of the service perception model are updated based on this loss value. When the number of iterations reaches the preset number or the loss value obtained in a certain iteration is less than or equal to the preset value, the iteration is stopped, and the service perception model training is completed.

[0081] From the above, it can be seen that the embodiment of the present application provides a service quality assurance method, which constructs a gene library, constructs training data based on the historical park characteristics, historical network indicator characteristics and historical service quality category labels in the gene library, and trains the service perception model based on the training data. By using a method of evaluating the quality of video services using a multi-classification algorithm based on the park scenario of the 5G private network, real-time evaluation and monitoring of the quality of video services in the ToB scenario can be achieved, thereby achieving automated correction and flexible error correction, and dynamically perceiving the needs of park users. In this way, on the one hand, end-to-end service quality perception of park services is achieved, and service quality perception results are obtained in real time; on the other hand, the corresponding assurance strategy is matched to the park service quality perception results, so as to automatically adjust the service and optimize the poor-quality service.

[0082] Figure 7 This is a schematic diagram of a service quality assurance method evaluation system provided by the application embodiment. Figure 7 As shown, by taking the park as a unit and analyzing the characteristics of the park video service categories according to different park scenarios of the 5G private network, the park characteristics are generated by combining the business volume distribution and scenario-based needs. Specifically, the park video services are finely classified and the busyness of the park at different times is dynamically perceived. The park video service category characteristics and busy time periods are generated based on the park's video service distribution ratio, business volume and busy time periods. The park gene library records historical park characteristics and combines the historical network indicator characteristics with minute strength and historical service quality category labels parsed by the N6 pointer to maintain training data for video service evaluation. Based on the gene library and the video data added in the preset period, the park system regularly and automatically iterates the quality perception model, thereby realizing real-time end-to-end video service quality perception, realizing automatic correction and flexible error correction, and dynamically perceiving the needs of park users.

[0083] Figure 8 This is a schematic diagram of the structure of a service quality assurance device provided in an embodiment of the present application. Figure 8As shown, the service quality assurance device 800 includes:

[0084] The acquisition unit 801 is configured to acquire network indicator characteristics and park characteristics of a park, wherein the park characteristics include the business type of the park and the busy time period of the park.

[0085] The processing unit 802 is configured to predict the service quality category of the campus based on network indicator characteristics and campus characteristics through a service perception model.

[0086] The determining unit 803 is configured to determine a campus assurance policy based on the service quality category, wherein the campus assurance policy is used to ensure the service quality of the campus.

[0087] In some embodiments, the options for service quality categories include: a first quality category, a second quality category, and a third quality category, wherein the service quality corresponding to the first quality category is higher than the service quality corresponding to the second quality category, and the service quality corresponding to the second quality category is higher than the service quality corresponding to the third quality category.

[0088] In some implementations, the determining unit 803 is further configured to determine an encoding strategy, a compression strategy, and a storage strategy for the business data of the campus if the service quality category is the third quality category.

[0089] In some embodiments, the determination unit 803 is further used to lower the encoding rate of the park's business data by one or more levels if the park's business type is a high-definition video business, a live video business, or a video-on-demand business, and the park is in a busy period.

[0090] In some implementations, the determining unit 803 is further configured to locally compress the business data of the park if the business type of the park is a surveillance video business.

[0091] In some implementations, the determining unit 803 is further configured to locally cache the business data of the park if the business type of the park is a video-on-demand business.

[0092] In some implementations, the determining unit 803 is further configured to generate a service alarm log if the service quality category is the third quality category, where the service alarm log includes network indicator characteristics and campus characteristics.

[0093] In some embodiments, the processing unit 803 is also used to construct a gene library, which includes historical park characteristics, historical network indicator characteristics and historical service quality category labels; construct training data based on the historical park characteristics, historical network indicator characteristics and historical service quality category labels; and train the service perception model based on the training data.

[0094] In some embodiments, the processing unit 803 is further used to detect historical business of the park; classify historical business to obtain historical business types; determine historical busy time periods based on the traffic of historical business; and construct a gene library based on historical business types and historical busy time periods.

[0095] In some embodiments, the processing unit 803 is further used to determine that the first time unit belongs to a historical busy time period if the traffic of historical services in the first time unit is greater than or equal to a certain proportion of the average video traffic; wherein the average video traffic is the average of the traffic of historical services in N second time units before the first time unit, and N is a positive integer.

[0096] In some embodiments, the processing unit 803 is also used to collect historical network indicators of the park; calculate the importance of the historical network indicators, and screen out historical network indicators with an importance greater than or equal to a threshold from the collected historical network indicators; and construct the gene library based on the historical network indicators with an importance greater than or equal to the threshold.

[0097] In some implementations, the processing unit 803 is further configured to train the service awareness model based on the training data according to a preset period; or, when an update operation of the service awareness model is obtained, train the service awareness model based on the training data.

[0098] Those skilled in the art should understand that Figure 8 The functions implemented by each unit in the service quality assurance device shown can be understood by referring to the relevant description of the aforementioned method. Figure 8 The functions of the various units in the service assurance device shown can be implemented by a program running on a processor, or by a specific logic circuit.

[0099] Figure 9 900 is a schematic structural diagram of a processing device provided in an embodiment of the present application. The processing device may be a terminal device or a network device. Figure 9 The processing device 90 shown includes a processor 910, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0100] Alternatively, as Figure 9 As shown, the processing device 900 may further include a memory 920. The processor 910 may call and execute a computer program from the memory 920 to implement the method in the embodiment of the present application.

[0101] The memory 920 may be a separate device independent of the processor 910 , or may be integrated into the processor 910 .

[0102] Alternatively, as Figure 9 As shown, the processing device 900 may further include a transceiver 930 , and the processor 910 may control the transceiver 930 to communicate with other devices. Specifically, the transceiver 930 may send information or data to other devices, or receive information or data sent by other devices.

[0103] The transceiver 930 may include a transmitter and a receiver. The transceiver 930 may further include an antenna, and the number of antennas may be one or more.

[0104] Optionally, the processing device 900 may specifically be a network device in an embodiment of the present application, and the processing device 900 may implement the corresponding processes implemented by the network device in each method in the embodiment of the present application. For the sake of brevity, they are not described here in detail.

[0105] Optionally, the processing device 900 may specifically be a mobile terminal / terminal device in an embodiment of the present application, and the processing device 900 may implement the corresponding processes implemented by the mobile terminal / terminal device in each method in the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0106] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly implemented as a hardware decoding processor, or can be implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0107] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0108] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0109] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0110] Optionally, the computer-readable storage medium can be applied to the network device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0111] Optionally, the computer-readable storage medium can be applied to the mobile terminal / terminal device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the mobile terminal / terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0112] An embodiment of the present application also provides a computer program product, including computer program instructions.

[0113] Optionally, the computer program product can be applied to the network device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0114] Optionally, the computer program product can be applied to the mobile terminal / terminal device in the embodiments of the present application, and the computer program instructions enable the computer to execute the corresponding processes implemented by the mobile terminal / terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0115] The embodiment of the present application also provides a computer program.

[0116] Optionally, the computer program can be applied to the network device in the embodiments of the present application. When the computer program runs on a computer, the computer executes the corresponding processes implemented by the network device in the various methods of the embodiments of the present application. For the sake of brevity, they are not described here.

[0117] Optionally, the computer program can be applied to the mobile terminal / terminal device in the embodiments of the present application. When the computer program runs on the computer, the computer executes the corresponding processes implemented by the mobile terminal / terminal device in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0118] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0119] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0123] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling 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 method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0124] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A service quality assurance method, characterized in that: The method comprises: Obtaining network indicator characteristics and park characteristics of the park, wherein the park characteristics include the business type of the park and the busy time period of the park; Predicting the service quality category of the park based on the network indicator characteristics and the park characteristics through a service perception model; A campus protection policy is determined based on the service quality category, wherein the campus protection policy is used for service quality protection of the campus.

2. The method according to claim 1, characterized in that The options for the service quality category include: a first quality category, a second quality category, and a third quality category, wherein the service quality corresponding to the first quality category is higher than the service quality corresponding to the second quality category, and the service quality corresponding to the second quality category is higher than the service quality corresponding to the third quality category; The determining of the campus protection strategy based on the service quality category includes: If the service quality category is the third quality category, the encoding strategy, compression strategy and storage strategy of the service data of the park are determined.

3. The method according to claim 2, characterized in that The encoding strategy includes: if the service type of the park is a high-definition video service, a live video service, or a video-on-demand service, and the park is in a busy time period, lowering the encoding rate of the service data of the park by one or more levels; The compression strategy includes: if the business type of the park is a surveillance video business, locally compressing the business data of the park; The storage strategy includes: if the business type of the park is a video-on-demand business, locally caching the business data of the park.

4. The method according to claim 2, characterized in that The method further comprises: If the service quality category is the third quality category, a service alarm log is generated, and the service alarm log includes the network indicator characteristics and the campus characteristics.

5. The method according to any one of claims 1 to 4, characterized in that The service perception model is a trained service perception model; the method further includes: Constructing a gene library, wherein the gene library includes historical park characteristics, historical network indicator characteristics, and historical service quality category labels; Constructing training data based on the historical park characteristics, the historical network indicator characteristics, and the historical service quality category labels; The service awareness model is trained based on the training data.

6. The method according to claim 5, characterized in that The training of the service perception model based on the training data includes: Training the service perception model based on the training data according to a preset period; or When obtaining the update operation of the service awareness model, the service awareness model is trained based on the training data.

7. The method according to claim 5, characterized in that The historical park characteristics include historical business types and historical busy time periods; and the gene bank construction includes: Detecting historical business of the park; Classifying the historical business to obtain the historical business type; Determining the historical busy time period based on the traffic of the historical services; The gene library is constructed based on the historical business types and the historical busy time periods.

8. The method according to claim 7, characterized in that The determining the historical busy time period based on the historical service traffic includes: If, within a first time unit, the traffic volume of the historical business is greater than or equal to a certain proportion of the average video traffic volume, it is determined that the first time unit belongs to the historical busy time period; wherein, the average video traffic volume is the average value of the traffic volume of the historical business in the N second time units before the first time unit, and N is a positive integer.

9. The method according to claim 5, characterized in that The construction of the gene library comprises: Collecting historical network metrics for the park; Calculating the importance of the historical network indicators, and selecting historical network indicators whose importance is greater than or equal to a threshold from the collected historical network indicators; The gene library is constructed based on the historical network indicators whose importance is greater than or equal to a threshold.

10. A service quality assurance device, characterized in that: The device comprises: An acquisition unit, configured to acquire network indicator characteristics and park characteristics of a park, wherein the park characteristics include a business type of the park and a busy time period of the park; A processing unit, configured to predict the service quality category of the park based on the network indicator characteristics and the park characteristics through a service perception model; The determining unit is configured to determine a park security policy based on the service quality category, wherein the park security policy is used for service quality assurance of the park.

11. A processing device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that Used to store a computer program, wherein the computer program causes a computer to execute the method according to any one of claims 1 to 9.

13. A computer program product, characterized in that The method comprises computer program instructions for causing a computer to execute the method according to any one of claims 1 to 9.

Images