A method and system for detecting the quality of APP video services

Abstract

The present invention provides a method and system for detecting the quality of APP video services, including: (1) obtaining the playback address of the video content to be monitored on the APP and pulling the stream for the playback address; (2) monitoring the quality of video services from three levels: the signal network layer, the bitstream layer, and the video and audio content layer, analyzing the stream metrics, and determining whether an alarm is generated; (3) calculating the video availability according to the program data and alarm data of the video content; (4) establishing a dynamic display graph of the video failure rate, and presenting the failure rate of the video content at different time periods in different color blocks at different time periods. The method and system for detecting the quality of APP video services provided by the present invention conduct comprehensive quality monitoring from the signal network layer, the bitstream layer, and the video and audio content layer, and dynamically display the monitoring results, improving the user experience.

Description

Technical Field

[0001] The present invention belongs to the technical field of video services, and particularly relates to a method and system for detecting the quality of APP video services. Background Art

[0002] With the rapid development of broadband networks, APP video services have developed rapidly and gradually become an important area for operators to develop value-added services. On the other hand, users' quality requirements for APP video services are also getting higher and higher.

[0003] The existing main implementation solutions for APP automated testing are to use a large number of real machines and a large number of testers to provide APP testing services, including in-depth compatibility testing, security vulnerability testing, in-depth traversal testing, and functional testing. In the prior art, scripts are commonly used for compatibility testing. The most commonly used script compatibility testing method is to first customize and develop scripts to cover the main pages or functions to be tested, and then adapt to mainstream test models. After adapting to different models, a large number of real machines are used to install the corresponding APP for actual testing. The main testing directions are the availability of functions, whether the UI interface is presented normally, the compatibility of various models, and the conditions of memory, traffic, power consumption, etc. The disadvantages of script compatibility testing are obvious: it requires dependence on hardware and a large investment in equipment and manpower; customized development is required for the testing of different APPs; it is biased towards APP performance and function testing, and there is no monitoring of the quality of APP video services.

[0004] Therefore, it is necessary to provide a method and system for detecting the quality of APP video services to detect the quality of APP video services. Summary of the Invention

[0005] Aiming at the defects existing in the prior art, the method and system for detecting the quality of APP video services provided by the present invention are free from the limitations of hardware or equipment, releasing human and material resources; automatically adapting to detect various APPs, reducing customized development work; monitoring the quality of APP video services, ensuring the security of the released content and the quality of business services, and continuously improving the quality of the media integration business and the user experience.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a method for detecting the quality of APP video services, the method comprising: (1) obtaining the playback address of the video content to be monitored on the APP, and pulling the stream for the playback address;

[0007] (2) monitoring the quality of the video service from three levels of the signal network layer, the bitstream layer, and the video and audio content layer, analyzing the stream metrics, and determining whether an alarm is generated;

[0008] (3) calculating the video availability according to the program data and alarm data of the video content;

[0009] (4) A dynamic display diagram of the video failure rate is established, which uses different color blocks to present the failure rate of the video content in different time periods.

[0010] Furthermore, the signal network layer related flow indicators include: maximum TCP connection duration, minimum TCP connection duration, average TCP connection duration, TCP retransmission rate and TCP repetition rate; the flow indicators of the code stream layer include screen freeze error, index file request failure, index file update timeout, index file format error, fragment request error; the flow indicators of the audio and video content layer include: black field failure, still frame, silent failure, volume.

[0011] Furthermore, the monitoring and analysis of flow indicators in step (2) and determining whether to generate an alarm specifically include: when the monitored value is greater than the threshold value of each flow indicator, generating an alarm and recording alarm data; the alarm data includes: alarm type and alarm duration.

[0012] Furthermore, the formula for calculating the video availability in step (3) is:

[0013]

[0014] Where: Avb is the video availability, m is the statistical period; n is the number of alarm types; μ1, μ2,…, μ n The weight of each alarm type; k1, k2, ..., k n The duration of each alarm type occurring for a certain video program.

[0015] Furthermore, the step (3) further includes: when the occurrence periods of different alarm types overlap, the overlapping time is calculated using the weight value of the alarm type with a higher weight.

[0016] Furthermore, in the dynamic display graph of the video failure rate, the horizontal axis is the time period, where the unit time period is the statistical period m; the vertical axis is the program name and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1-availability.

[0017] Furthermore, the step (4) includes: selecting five color blocks to display five levels of failure rate, and the five levels of failure rate are divided into the following standards:

[0018] Level 1: min: 0, max: 0.3; Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min: 50, max: 100.

[0019] To achieve the above objectives, the technical solution adopted by the present invention is: a system for detecting the quality of APP video services, the system includes: a probe subsystem, a monitoring subsystem, and a management and analysis platform; the probe subsystem includes a video content acquisition module and a streaming module; the monitoring subsystem includes a monitoring module and a judgment module; the management and analysis platform includes a data processing module and a display module;

[0020] The video content acquisition module is used to obtain the playback address of the video content to be monitored on the APP;

[0021] The streaming module is used to pull the stream for the playback address;

[0022] The monitoring module is used to monitor the video service quality from three levels: the signal network layer, the bitstream layer, and the video and audio content layer;

[0023] The judgment module is used to analyze the stream metrics according to the monitoring results reported by the monitoring module and judge whether an alarm is generated;

[0024] The data processing module is used to calculate the video availability according to the program data and alarm data reported by the probe subsystem and the monitoring subsystem;

[0025] The display module is used to establish a dynamic display chart of the video failure rate, and present the failure rate of the video content at different time periods with different color blocks by time period.

[0026] Further, the stream metrics analyzed in the monitoring module include:

[0027] The stream metrics related to the signal network layer include the maximum TCP connection duration, the minimum TCP connection duration, the average TCP connection duration, the TCP retransmission rate, and the TCP duplication rate; the stream metrics of the bitstream layer include freeze frame errors, index file request failures, index file update timeouts, index file format errors, and shard request errors; the stream metrics of the video and audio content layer include: black field failures, still frames, silent failures, and volume.

[0028] Further, when the monitoring results reported by the monitoring module are greater than the threshold values of each stream metric, the judgment module determines that an alarm is generated and records the alarm data; the alarm data includes: alarm type and alarm duration.

[0029] Further, the formula for calculating the video availability by the data processing module is:

[0030]

[0031] In the formula: Avb is the video availability, m is the statistical period; n is the number of alarm types; μ1, μ2, …, μ n is the weight of each alarm type; k1, k2, …, kn The duration of occurrence of each alarm type for a certain video program.

[0032] Furthermore, when the data processing module calculates the video availability, when the occurrence periods of different alarm types overlap, the overlapping time is calculated using the weight value of the alarm type with a higher weight.

[0033] Furthermore, in the dynamic display graph of the video failure rate, the horizontal axis represents the time period, and the unit time period is the statistical period m; the vertical axis represents the program names and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1 - availability.

[0034] Furthermore, the legend of the dynamic display graph of the video failure rate selects five color blocks to display 5 levels of the failure rate. The 5 - level division standard of the failure rate is as follows:

[0035] Level 1: min: 0, max: 0.3; Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min: 50, max: 100.

[0036] The effect of the present invention is that, to ensure the broadcast safety and broadcast quality of the video content of the media integration APP, for the video programs provided by the media integration APP, comprehensive quality monitoring is carried out from the signal network layer, the bitstream layer, and the video - audio content layer, and the monitoring results are dynamically displayed to improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic flowchart of a method for measuring the video service quality of an APP;

[0038] Figure 2 It is a schematic diagram of the dynamic display of the video failure rate;

[0039] Figure 3 It is a schematic structural diagram of a system for measuring the video service quality of an APP. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the technical problems solved, the technical solutions adopted, and the technical effects achieved by the present invention clearer, the technical solutions of the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0041] Refer to Figure 1 ,Figure 1 It is a schematic flowchart of a method for measuring the quality of APP video services.

[0042] Step 101: Obtain the playback address of the video content to be monitored on the APP, and pull the stream for the said playback address.

[0043] After obtaining the playback addresses of the video content to be monitored on the APP through a crawler, the monitoring probe pulls the streams for these playback addresses to detect whether the stream pulling is successful. The acquisition of the playback addresses of the video content is not the focus of this application and can be achieved through existing technologies, so it will not be elaborated here.

[0044] Step 102: Monitor the video service quality from three levels: the signal network layer, the bitstream layer, and the video and audio content layer, analyze the stream metrics, and determine whether an alarm is generated.

[0045] The stream metrics related to the signal network layer include: the maximum TCP connection duration, the minimum TCP connection duration, the average TCP connection duration, the TCP retransmission rate, and the TCP duplication rate.

[0046] Specifically, the maximum, minimum, and average TCP connection durations are used to measure the end-to-end network latency; the TCP retransmission rate and the TCP duplication rate are used for the quantitative measurement of the network packet loss rate. When there are problems with the video quality, these metrics can be used to initially determine whether the problem lies in the network layer. Acquisition method: Obtain the basic parameters of the TCP connection initiated by the terminal to the WEB or CDN server.

[0047] The stream metrics of the bitstream layer include: freeze frame error, index file request failure, index file update timeout, index file format error, and shard request error. When there are problems with the video quality, these metrics can be used to initially determine whether there are problems with video playback.

[0048] The stream metrics of the video and audio content layer include: black field fault, still frame, silent fault, and volume. These metrics can intuitively reflect the quality of the user experience.

[0049] Monitoring and analyzing the stream metrics and determining whether an alarm is generated specifically include: when the monitored value is greater than the threshold of each stream metric, an alarm is generated and the alarm data is recorded; the alarm data includes: the alarm type and the alarm duration.

[0050] It should be noted that the threshold values are different under different application scenarios. In a specific embodiment, the alarm criteria for each flow index are as follows: TCP connection timeout: the time for establishing TCP three-way handshake. The default value of the threshold configuration item time is 200 ms, which means an alarm is generated if the time for establishing TCP three-way handshake exceeds this value. TCP retransmission rate exceeding the standard error: the ratio of the number of retransmitted packets within 5 seconds to the total number of packets within 5 seconds. The default threshold value is 5%, which means an alarm is generated if the retransmission rate within 5 seconds exceeds this value. It can be used to measure whether there is packet loss in the upstream link of the monitoring point. The higher the alarm frequency, the more serious the packet loss. TCP duplication rate exceeding the standard error: the ratio of the number of duplicate packets within 5 seconds to the total number of packets within 5 seconds. The default threshold value is 5%, which means an alarm is generated if the duplication rate within 5 seconds exceeds this value. It can be used to measure whether there is packet loss in the downstream link of the monitoring point. The higher the alarm frequency, the more serious the packet loss.

[0051] Freeze error: an alarm is generated when the screen freezes during playback; the main reasons are abnormal TS shard download or inability to download, update, or format the index file; Index file request failure: an alarm is generated when the m3u8 index file fails to download and the server has no response or returns an error message; Index file update timeout: an alarm is generated when the text content in the m3u8 index file has not been updated for more than the described total ts duration, resulting in no new valid ts shards available for download; Index file format error: an alarm is generated when there is a problem with the text format of the m3u8 index file, such as inability to parse the shard duration or obtain shard information; Shard request error: an alarm is generated when the TS shard request has no response or returns 4xx, 5xx errors.

[0052] Black screen fault: the duration is 2 s, which means an alarm is generated if the black screen time exceeds this value. Still frame fault: the duration is 10 s, which means an alarm is generated if the still frame phenomenon time exceeds this value. Mute fault: the level threshold is -48 dB, and an alarm is generated after 30 S. Too low volume: the level threshold is -38 dB, and an alarm is generated after 30 s. Too high volume: the level threshold is -9 dB, and an alarm is generated after 8 s.

[0053] Step 103: Calculate the video availability based on the program data and alarm data of the video content.

[0054] The formula for calculating video availability is:

[0055]

[0056] In the formula: Avb is the video availability, m is the statistical period; n is the number of alarm types; μ1, μ2, …, μ n is the weight of each alarm type; k1, k2, …, k n is the duration of each alarm type occurring in a certain video program.

[0057] It should also be noted that the common statistical period m is 5 minutes. However, it has different values in different situations, and the present invention does not limit this.

[0058] In a specific embodiment, within one minute, the warning weights and occurrence durations of various warnings of a certain program are as follows, and the occurrence times of each warning do not overlap with each other:

[0059] Freeze warning: weight: 0.1, occurrence duration: 20s;

[0060] Audio loss: weight: 0.2, occurrence duration: 15s;

[0061] Video loss: weight: 0.3, occurrence duration: 10s; [[ID=…]]

[0062] Stream interruption error: weight: 0.4, occurrence duration: 5s.

[0063] At this time, the statistical period m is 60 seconds. Then, the availability is:

[0064] Avb = 1 - (20 * 0.1 * 4 + 15 * 0.2 * 4 + 10 * 0.3 * 4 + 5 * 0.4 * 4) * 100% / 60 = 66.7%

[0065] It should also be noted that when the occurrence periods of different warning types overlap, the weight value of the warning type with a higher weight is used to calculate the overlapping time.

[0066] In another specific embodiment, among the above-mentioned warnings, the freeze warning and the audio loss warning occur simultaneously for 10 seconds. Then, the availability of this program within this minute is:

[0067] Avb = 1 - (10 * 0.1 * 4 + 15 * 0.2 * 4 + 10 * 0.3 * 4 + 5 * 0.4 * 4) * 100% / 60 = 80%

[0068] Step 104: Establish a dynamic display graph of the video failure rate, and present the failure rates of the video content in different time periods with different color blocks in different time periods.

[0069] Refer to Figure 2 , Figure 2 which is a schematic diagram of the dynamic display of the video failure rate. In the dynamic display graph of the video failure rate, the horizontal axis is the time period, and the unit time period is the statistical period m; the vertical axis is the program name and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1 - availability.

[0070] It should be noted that the purpose of establishing a dynamic display chart of video failure rate is to more intuitively see whether there are alarms for a certain video stream within a short period of time. Video availability is to reflect the overall quality of a stream over a long period of time. If video availability is used to represent it, a too large value over a long time will make people ignore the importance of alarms.

[0071] Five color blocks are selected to display five levels of failure rate. The classification criteria for the five levels of failure rate are as follows:

[0072] Level 1: min: 0, max: 0.3; Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min: 50, max: 100.

[0073] In a specific embodiment, the green color block represents that the failure rate is at Level 1, the light green color block represents that the failure rate is at Level 2; the yellow color block represents that the failure rate is at Level 3; the orange color block represents that the failure rate is at Level 4; the red color block represents that the failure rate is at Level 5.

[0074] By vertically comparing the dynamic display chart of video failure rate, it can be found which video content has low service quality. By horizontally comparing, it can be found in which event segments the video service quality is low, and the reasons for the low video service quality can be investigated targeted.

[0075] Refer to Figure 3 , Figure 3 For the structural schematic diagram of a system for detecting the service quality of APP video. The present application provides a system for detecting the service quality of APP video 100. The system 100 includes: a probe subsystem 101, a monitoring subsystem 102, and a management and analysis platform 103; the probe subsystem 101 includes a video content acquisition module 1011 and a stream pulling module 1012; the monitoring subsystem 102 includes a monitoring module 1021 and a judgment module 1022; the management and analysis platform 103 includes a data processing module 1031 and a display module 1032.

[0076] The video content acquisition module 1011 is used to obtain the playback address of the video content to be monitored on the APP. In a specific embodiment, a crawler obtains the playback address of the video content to be monitored on the APP.

[0077] The stream pulling module 1012 is used to pull the stream for the playback address.

[0078] The monitoring module 1021 is used to monitor the video service quality from three levels: the signal network layer, the bitstream layer, and the video and audio content layer.

[0079] The flow metrics related to the signal network layer include: the maximum duration of a TCP connection, the minimum duration of a TCP connection, the average duration of a TCP connection, the TCP retransmission rate, and the TCP duplication rate.

[0080] Specifically, the maximum, minimum, and average durations of a TCP connection are used to measure the end-to-end network latency; the TCP retransmission rate and the TCP duplication rate are used for the quantitative measurement of the network packet loss rate. When there are problems with the video quality, these metrics can be used to initially determine whether the problem lies at the network level. Acquisition method: Obtain the basic parameters of the TCP connection initiated by the terminal to the WEB or CDN server.

[0081] The flow metrics of the bitstream layer include freeze frame errors, index file request failures, index file update timeouts, index file format errors, and shard request errors. When there are problems with the video quality, these metrics can be used to initially determine whether there are problems with video playback.

[0082] The flow metrics of the video and audio content layer include: black field faults, still frames, mute faults, and volume. These metrics can intuitively reflect the quality of the user experience.

[0083] The judgment module 1022 is used to analyze the flow metrics based on the monitoring results reported by the monitoring module and determine whether an alarm is generated.

[0084] When the monitoring results reported by the monitoring module 1021 are greater than the threshold values of each flow metric, the judgment module 1022 determines that an alarm is generated and records the alarm data; the alarm data includes: alarm type and alarm duration.

[0085] Monitoring and analyzing the flow metrics to determine whether an alarm is generated specifically includes: when the monitored value is greater than the threshold value of each flow metric, an alarm is generated and the alarm data is recorded; the alarm data includes: alarm type and alarm duration.

[0086] It should be noted that the threshold values are different in different application scenarios. In a specific embodiment, the alarm criteria for each flow metric are as follows: TCP connection timeout: The time for establishing a TCP three-way handshake. The default value of the threshold configuration item time is 200 ms, which means that an alarm is generated if the TCP three-way handshake establishment time exceeds this value. TCP retransmission rate exceeding the standard error: The ratio of the number of retransmitted packets within 5 seconds to the total number of packets within 5 seconds. The default threshold value is 5%, which means that an alarm is generated if the retransmission rate within 5 seconds exceeds this value. It can be used to measure whether there is packet loss in the upstream link of the monitoring point, and the higher the alarm frequency, the more serious the packet loss. TCP duplication rate exceeding the standard error: The ratio of the number of duplicate packets within 5 seconds to the total number of packets within 5 seconds. The default threshold value is 5%, which means that an alarm is generated if the duplication rate within 5 seconds exceeds this value. It can be used to measure whether there is packet loss in the downstream link of the monitoring point, and the higher the alarm frequency, the more serious the packet loss.

[0087] Freeze error: Alarm when the video freezes during playback; The main reasons are abnormal TS shard download or failure to download, update, or incorrect format of the index file; Index file request failure: Alarm when the m3u8 index file fails to download and the server has no response or returns an error message; Index file update timeout: Alarm when the text content in the m3u8 index file exceeds the described total duration of the ts and has not been updated, resulting in no new valid ts shards available for download; Index file format error: Alarm when there is a problem with the text format of the m3u8 index file, such as unable to parse the shard duration or obtain shard information; Shard request error: Alarm when the TS shard request has no response or returns a 4xx or 5xx error.

[0088] Black screen fault: Duration is 2s, meaning an alarm is triggered when the black screen time exceeds this value. Still frame fault: Duration is 10s, meaning an alarm is triggered when the still frame phenomenon time exceeds this value. Silent fault: Level threshold is -48dB, and an alarm is generated after 30S. Too low volume: Level threshold is -38dB, and an alarm is generated after 30s. Too high volume: Level threshold is -9dB, and an alarm is generated after 8s.

[0089] The data processing module 1031 is used to calculate the video availability according to the program data and alarm data reported by the probe subsystem 101 and the monitoring subsystem 102.

[0090] The formula for the data processing module 1031 to calculate the video availability is:

[0091]

[0092] In the formula: Avb is the video availability, m is the statistical period; n is the number of alarm types; μ1, μ2, …, μ n is the weight of each alarm type; k1, k2, …, k n is the duration of each alarm type occurrence for a certain video program.

[0093] It should also be noted that the commonly used statistical period m is 5 minutes. However, it has different values in different situations, and the present invention does not limit this.

[0094] In a specific embodiment, within one minute, the weights and occurrence durations of various alarms for a certain program are as follows, and the occurrence times of each alarm do not overlap with each other:

[0095] Freeze alarm: Weight: 0.1, Occurrence duration: 20s;

[0096] Audio loss: Weight: 0.2, Occurrence duration: 15s;

[0097] Video loss: Weight: 0.3, Occurrence duration: 10s;

[0098] Flow interruption error: Weight: 0.4, Occurrence duration: 5s.

[0099] At this time, the statistical period m is 60 seconds. Then, the availability is:

[0100] Avb = 1 - (20 * 0.1 * 4 + 15 * 0.2 * 4 + 10 * 0.3 * 4 + 5 * 0.4 * 4) * 100% / 60 = 66.7%

[0101] It should also be noted that when calculating the video availability by the data processing module 1031, when the occurrence periods of different alarm types overlap, the weight value of the alarm type with a higher weight is used to calculate the overlapping time.

[0102] In another specific embodiment, in the above alarms, the screen freeze alarm and the audio loss alarm occur simultaneously for 10 seconds. Then, the availability of this program within this minute is:

[0103] Avb = 1 - (10 * 0.1 * 4 + 15 * 0.2 * 4 + 10 * 0.3 * 4 + 5 * 0.4 * 4) * 100% / 60 = 80%

[0104] The display module 1032 is used to establish a dynamic display graph of the video failure rate, and present the failure rates of the video content in different time periods with different color blocks by time period.

[0105] Refer to Figure 2 , Figure 2 is a schematic diagram of the dynamic display of the video failure rate. In the dynamic display graph of the video failure rate, the horizontal axis is the time period, and the unit time period is the statistical period m; the vertical axis is the program name and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1 - availability.

[0106] Five color blocks are selected to display 5 levels of the failure rate. The 5 - level division standard of the failure rate is as follows:

[0107] Level 1: min: 0, max: 0.3; Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min: 50, max: 100.

[0108] In a specific embodiment, the green color block indicates that the failure rate is at Level 1, the light - green color block indicates that the failure rate is at Level 2; the yellow color block indicates that the failure rate is at Level 3; the orange color block indicates that the failure rate is at Level 4; the red color block indicates that the failure rate is at Level 5.

[0109] By making a vertical comparison of the dynamic display chart of video failure rates, it can be found that the service quality of those video contents is low. By making a horizontal comparison, it can be found in which event segments the video service quality is low, and the reasons for the low video service quality can be investigated targeted.

[0110] Different from the prior art, a method and system for detecting the service quality of APP videos provided by the present invention comprehensively monitor the quality of video programs provided by the media integration APP from the signal network layer, the bitstream layer, and the video and audio content layer to ensure the broadcast safety and broadcast quality of the video content of the media integration APP, and dynamically display the monitoring results, which is convenient for users to detect whether the APP video service is normal and whether the user experience of watching videos is good.

[0111] Those skilled in the art should understand that the method and system of the present invention are not limited to the embodiments described in the specific embodiments. The above specific description is only for explaining the purpose of the present invention and is not used to limit the present invention. Other embodiments obtained by those skilled in the art according to the technical solution of the present invention also belong to the scope of technical innovation of the present invention. The protection scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for testing the quality of APP video service, characterized in that: The method comprises the following steps: (1) Obtain the playback address of the video content to be monitored on the APP and pull the stream of the playback address; (2) Monitor video service quality at three levels: signal network layer, bitstream layer, and audio / video content layer, analyze stream indicators, and determine whether to generate an alarm; (3) calculating video availability based on program data and alarm data of the video content; (4) establishing a dynamic display graph of video failure rate, using different color blocks to present the failure rate of the video content in different time periods; The calculation formula for calculating the video availability in step (3) is: Where: Avb is the video availability; m is the statistical period; n is the number of alarm types; μ1,μ2,…,μ n The weight of each alarm type; k1, k2, ..., k n The duration of each alarm type for a certain video program: The step (3) further includes: when the occurrence periods of different alarm types overlap, calculating the overlapping time using the weight value of the alarm type with a higher weight.

2. The method according to claim 1, characterized in that The signal network layer related flow indicators include: maximum TCP connection duration, minimum TCP connection duration, average TCP connection duration, TCP retransmission rate and TCP repetition rate; the flow indicators of the code stream layer include screen freeze error, index file request failure, index file update timeout, index file format error, and fragment request error; the flow indicators of the audio and video content layer include: black field failure, static frame, silent failure, and volume.

3. The method according to claim 2, characterized in that Monitoring and analyzing flow indicators in step (2) to determine whether an alarm is generated specifically includes: When the monitored value is greater than the threshold value of each flow indicator, an alarm is generated and the alarm data is recorded; the alarm data includes: alarm type and alarm duration.

4. The method according to claim 1, wherein In the dynamic display diagram of the video failure rate, the horizontal axis is the time period, and the unit time period is the statistical period m; the vertical axis is the program name and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1-availability.

5. The method according to claim 1, characterized in that The step (4) includes: selecting five color blocks to display five levels of failure rate, and the five levels of failure rate are divided into the following standards: Level 1: min: 0, max: 0.3; Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min: 50, max:

100.

6. A system for dialing and testing the quality of APP video service, characterized in that: The system includes: a probe subsystem, a monitoring subsystem, and a management and analysis platform; the probe subsystem includes a video content acquisition module and a stream pulling module; the monitoring subsystem includes a monitoring module and a judgment module; the management and analysis platform includes a data processing module and a display module; Video content acquisition module, used to obtain the playback address of the video content to be monitored on the APP; A stream pulling module, used for pulling the stream of the playback address; The monitoring module is used to monitor the quality of video services at three levels: signal network layer, bit stream layer, and audio and video content layer; A judgment module, configured to analyze flow indicators based on the monitoring results reported by the monitoring module and determine whether an alarm is generated; The data processing module is used to calculate the video availability based on the program data and alarm data reported by the probe subsystem and the monitoring subsystem. The calculation formula for the video availability is: Where: Avb is the video availability, m is the statistical period; n is the number of alarm types; μ1,μ2,…,μ n The weight of each alarm type; k1, k2, ..., k n The duration of each alarm type occurring in a certain video program. When the occurrence periods of different alarm types overlap, the weight value of the alarm type with the higher weight is used to calculate the overlapping time. The display module is used to establish a dynamic display diagram of the video failure rate, and present the failure rate of the video content in different time periods by using different color blocks.

7. The system according to claim 6, characterized in that The flow indicators analyzed in the monitoring module include: The signal network layer's related flow indicators include the maximum TCP connection duration, the minimum TCP connection duration, the average TCP connection duration, the TCP retransmission rate, and the TCP repetition rate; the code stream layer's flow indicators include screen freeze errors, index file request failures, index file update timeouts, index file format errors, and fragment request errors; the audio and video content layer's flow indicators include: black field failures, still frames, silent failures, and volume.

8. The system according to claim 6, wherein: When the monitoring result reported by the monitoring module is greater than the threshold value of each flow indicator, the judgment module determines to generate an alarm and records alarm data; the alarm data includes: alarm type and alarm duration.

9. The system according to claim 6, wherein: In the dynamic display diagram of the video failure rate, the horizontal axis is the time period, and the unit time period is the statistical period m; the vertical axis is the program name and availability of each video content; a single color block represents the failure rate of the video program corresponding to the vertical axis in the time period corresponding to the horizontal axis, and the failure rate = 1-availability.

10. The system according to claim 6, wherein: The legend of the dynamic display graph of the video failure rate uses five color blocks to display the five levels of failure rate. The classification standards for the five levels of failure rate are as follows: Level 1: min: 0, max: 0.3； Level 2: min: 0.3, max: 2; Level 3: min: 2, max: 30; Level 4: min: 30, max: 50; Level 5: min:50,max:100.

Images