Media communication network bandwidth flow abnormity identification method and system based on big data

Abstract

The invention discloses a media communication network bandwidth flow abnormity identification method and system based on big data, and relates to the technical field of media communication.The method comprises the steps that firstly, network hardware, media server operation and terminal user experience indexes of all communication lines of a target media communication platform at the current moment are obtained, and a joint communication index evaluation value is obtained through analysis; secondly, identifying whether the network environment is abnormal or not according to the evaluation value; then, through the network environment abnormal data of each historical period, evaluating whether the platform needs to update and adapt to new equipment; and finally, if updating is needed, online live broadcast service, service quality and extension planning data of the current period are acquired, and a new equipment adaptation scheme is analyzed. According to the method, by means of a big data technology and integration of multi-dimensional indexes, effective identification and equipment adaptation analysis of bandwidth traffic abnormity of the media communication network are realized, and scientificity and accuracy of network management are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of media communication, and particularly to a method and system for identifying abnormal media communication network bandwidth traffic based on big data. Background Art

[0002] With the rise of big data technology, a vast amount of network operation data and service data have been collected and stored. However, how to effectively utilize these data to accurately identify abnormal bandwidth traffic remains an urgent problem to be solved. At the same time, in the media communication service scenario, if network bandwidth traffic anomalies cannot be identified in a timely and accurate manner, and network devices cannot be reasonably evaluated and updated, it will lead to a decline in network service quality, affect user experience, and even cause service interruption and economic losses. Therefore, there is a need for a method and system for identifying abnormal media communication network bandwidth traffic based on big data.

[0003] The prior art, such as an invention application patent with a publication number of CN106789323A, discloses a communication network management method and its device. The method includes: obtaining the operation data of a target device to be monitored; parsing the operation data according to the communication protocol corresponding to the target device to obtain the index parameters corresponding to the operation data; determining the operation state corresponding to the index parameters according to the preset corresponding relationship between the operation state and the index parameters; if the operation state corresponding to the index parameters is an abnormal state, performing abnormal processing on the target device in the abnormal state. In the embodiment of the present invention, through a general communication protocol, the operation data of all target devices within the jurisdiction of the communication network management device are obtained, so as to analyze the corresponding operation state of the target device, and then corresponding processing is performed, solving the problem that the existing communication network management technology cannot centrally monitor communication devices of multiple manufacturers, requires operation and maintenance personnel for manual management, increases the operation cost of the communication network, and reduces the management efficiency of the communication network.

[0004] In view of the above solution, the present application discovers that the above technology has at least the following technical problems: 1. The prior art lacks comprehensive consideration of multi-dimensional indicators and relies only on a single or a few indicators to judge abnormal network bandwidth traffic, which easily ignores the influence of other important factors, resulting in some abnormal situations not being discovered in time. It is impossible to comprehensively evaluate the operation status of the network environment and lacks analysis from multiple levels such as the operation of the media server and the experience of end users. It is difficult to detect the impact of potential faults inside the server on network bandwidth traffic; it will prevent network management from timely discovering network anomalies affecting service quality from the user's perspective.

[0005] 2. Existing Technologies When network anomalies occur, there is a lack of systematic analysis methods and processes, which may prevent quickly and accurately determining the root cause and scope of influence of the anomalies, resulting in slow or inappropriate response decisions. The full-flow mirroring cannot be started in a timely manner to copy the traffic of the abnormal communication line, and it is impossible to quickly switch to the disaster recovery center to transfer services, which may prolong the service interruption time and affect the user experience. There is a lack of effective traffic cleaning and optimization mechanisms, making it difficult to clean and optimize the network flow table through software-defined network technology, remove abnormal traffic rules, and restore normal network traffic transmission. At the same time, it is also impossible to limit the UDP port bandwidth in a timely manner to prevent network congestion caused by excessive UDP traffic, which may further deteriorate the network congestion situation.

[0006] 3. Existing Technologies It is difficult to accurately evaluate whether the target media communication platform needs to be updated and adapted to new devices, lacking a comprehensive analysis based on multi-factors such as historical network environment anomaly data, device design life, and rated capacity. If the device update is determined only based on the usage time of the device or simple performance indicators, it may lead to untimely device updates, affecting network performance and stability, or premature device updates, resulting in resource waste. It is impossible to reasonably formulate a new device adaptation plan according to the actual business needs and development plan of the platform, which may cause the new device to be incompatible with the existing system, or unable to fully utilize the performance advantages of the new device, and unable to meet the needs of future business expansion.

[0007] 4. Existing Technologies The business evaluation of the media communication platform is not comprehensive and accurate enough. It does not comprehensively consider various factors such as online live broadcast service data, service quality data, and business expansion planning data, making it difficult to accurately analyze the comprehensive business evaluation value of the platform. For example, only focusing on the number of concurrent live users while ignoring other online live broadcast service data such as the average daily on-demand request number and the proportion of popular videos will result in an incomplete evaluation of the platform's business load. There is a lack of a scientific coefficient analysis model, unable to accurately analyze the online live broadcast service coefficient, service quality coefficient, and business expansion planning coefficient, thus affecting the accurate judgment of the platform's business situation and the planning of future development. This may lead to problems such as unreasonable network resource allocation and decreased service quality during the business expansion process, unable to meet the growing needs of users and the requirements of business development. Summary of the Invention

[0008] Aiming at the above-mentioned existing technical deficiencies, the purpose of the present invention is to provide a method and system for identifying abnormal bandwidth traffic in a media communication network based on big data.

[0009] To solve the above technical problems, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a method for identifying abnormal bandwidth traffic in a media communication network based on big data, including: Step 1, obtaining the combined communication index evaluation value: Obtain the network hardware metric indicators, media server operation indicators, and terminal user experience indicators corresponding to each communication line in the target media communication platform at the current moment, and then analyze to obtain the combined communication index evaluation value corresponding to each communication line in the target media communication platform.

[0010] Step 2, identifying abnormal network environment: According to the combined communication index evaluation values corresponding to each communication line in the target media communication platform, further evaluate whether the network environment operation corresponding to each communication line in the target media communication platform is abnormal.

[0011] Step 3, evaluating historical abnormal network environment: Obtain the historical abnormal network environment data corresponding to each communication line in the target media communication platform in each historical period, and then evaluate whether the target media communication platform needs to be updated and adapted to new devices.

[0012] Step 4, analyzing the new device adaptation plan: If the target media communication platform needs to be updated and adapted to new devices, obtain the online live broadcast service data, service quality data, and service expansion plan data corresponding to the target media communication platform in the current period, and then analyze the new device adaptation plan corresponding to the target media communication platform.

[0013] In the second aspect, the present invention provides a system for identifying abnormal bandwidth traffic in a media communication network based on big data, including: A combined communication index evaluation value acquisition module: used to obtain the network hardware metric indicators, media server operation indicators, and terminal user experience indicators corresponding to each communication line in the target media communication platform at the current moment, and then analyze to obtain the combined communication index evaluation value corresponding to each communication line in the target media communication platform.

[0014] A network environment abnormal identification module: used to evaluate whether the network environment operation corresponding to each communication line in the target media communication platform is abnormal according to the combined communication index evaluation values corresponding to each communication line in the target media communication platform.

[0015] A historical network environment abnormal evaluation module: used to obtain the historical abnormal network environment data corresponding to each communication line in the target media communication platform in each historical period, and then evaluate whether the target media communication platform needs to be updated and adapted to new devices.

[0016] A new device adaptation plan analysis module: used to, if the target media communication platform needs to be updated and adapted to new devices, obtain the online live broadcast service data, service quality data, and service expansion plan data corresponding to the target media communication platform in the current period, and then analyze the new device adaptation plan corresponding to the target media communication platform.

[0017] The beneficial effects of the present invention are as follows: 1. In the embodiments of the present invention, by comprehensively considering network hardware metrics, media server operation metrics, and end-user experience metrics, the status of the media communication network can be comprehensively evaluated from multiple dimensions, accurately reflecting the actual operation of the network, and avoiding the one-sidedness of relying solely on a single metric for evaluation. By comparing the combined communication metric evaluation value with the set standard interval, it is possible to more accurately identify whether the network environment is operating abnormally, providing a strong basis for timely detecting and resolving network problems. When it is identified that the network environment is operating abnormally, a series of effective response decisions can be quickly taken, such as starting full-flow mirroring, switching to a disaster recovery center, triggering SDN flow table cleaning operations, restricting UDP port bandwidth, and sending alarm notifications, etc., to ensure network stability and service continuity and reduce the impact of abnormal situations on services.

[0018] 2. In the embodiments of the present invention, by analyzing historical network environment abnormal data and combining factors such as the designed lifespan and rated capacity of each device in the communication line, it is possible to reasonably evaluate whether the target media communication platform needs to be updated and adapted to new devices, providing a scientific basis for device updates and upgrades, and avoiding excessive or untimely device updates. At the same time, by analyzing the comprehensive service evaluation value based on the online live broadcast service data, service quality data, and business expansion planning data of the target media communication platform and comparing it with the comprehensive service evaluation value interval corresponding to each new device adaptation plan in the database, it is possible to select a suitable new device adaptation plan for the target media communication platform, improve the adaptability of the device to the service, and optimize network performance.

[0019] 3. In the embodiments of the present invention, by making full use of big data technology, it is possible to process and analyze a large amount of network data, mine the hidden information and rules in the data, provide strong support for network management and decision-making, improve the intelligent level and efficiency of network management, and provide intelligent decision-making support for network management. When evaluating whether the network environment is operating abnormally, it no longer relies on simple threshold judgments or empirical rules, but analyzes and predicts the combined communication metric evaluation value based on a big data model. When an anomaly is identified, the system can quickly initiate response decisions according to pre-set rules and best practices obtained from big data analysis, and can intelligently adjust the priority and execution order of the response strategy according to the severity and scope of the anomaly. When evaluating device update requirements and formulating new device adaptation plans, big data analysis can comprehensively consider various factors such as historical abnormal data, device performance metrics, and business development plans, providing scientific and reasonable decision-making suggestions for management personnel and avoiding blind decision-making and resource waste. Brief Description of the Drawings

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 This is the flowchart of the implementation steps of the method of the present invention.

[0022] Figure 2 This is the schematic diagram of the connection of system modules of the present invention. Detailed implementation manners

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present invention.

[0024] The embodiment of the present invention is as Figure 1 shown. The method for identifying abnormal bandwidth traffic in a media communication network based on big data includes: Step 1: Obtaining the combined communication index evaluation value: Obtain the network hardware metric indicators, media server operation indicators, and terminal user experience indicators corresponding to each communication line in the target media communication platform at the current moment, and then analyze to obtain the combined communication index evaluation value corresponding to each communication line in the target media communication platform.

[0025] In a specific embodiment, the process of analyzing to obtain the combined communication index evaluation value corresponding to each communication line in the target media communication platform is as follows: A1. According to the network hardware metric indicators, media server operation parameters, and terminal user experience indicators corresponding to each communication line in the target media communication platform, the network hardware metric indicators include the inbound traffic mutation rate, UDP traffic ratio change rate, and CPU temperature deviation degree, the media server operation indicators include the disk I / O deviation degree, HTTP error frequency, and abnormal connection ratio, and the terminal user experience indicators include the signal attenuation gradient, packet loss anomaly index, and user channel switching frequency. Then analyze to obtain the network hardware coefficient, media server operation coefficient, and terminal user experience coefficient corresponding to each communication line in the target media communication platform, and record them as Z q , X q and V q respectively, where q represents the number corresponding to each communication line, q = 1, 2......g, and g is any integer greater than 2.

[0026] It should be noted that the inbound traffic data of each communication line is collected in real time through the network traffic monitoring tool, and the traffic value per unit time is recorded. The difference in traffic changes in adjacent time intervals is calculated, and then divided by the initial traffic value to obtain the traffic change rate. The network traffic monitoring tool is also used to perform protocol analysis on the traffic data of each communication line to separate the UDP traffic. The proportion of UDP traffic in the total traffic is obtained, and the proportion values ​​at different times are recorded in time series. The proportion of UDP traffic at adjacent times or different time periods is compared to obtain the change value and change rate of the proportion. With the help of hardware monitoring sensors, the temperature data of the CPU of the network device is obtained in real time. The normal working temperature range of the device CPU is set, and the actually measured CPU temperature is compared with the average value or the middle value of the normal temperature range to obtain the temperature deviation value, which is then divided by the width of the normal temperature range (the difference between the maximum value and the minimum value) to obtain the CPU temperature deviation.

[0027] It should also be noted that the system's own disk I / O monitoring tool or third-party monitoring software is installed on the media server to monitor the disk read and write operations in real time, including indicators such as read and write rate, I / O request queue length, and response time. According to historical data or the disk I / O performance benchmark when the server is operating normally, calculate the deviation degree of the current disk I / O indicators from the benchmark value. By calculating the difference between the current disk read and write rate and the historical average read and write rate, and then dividing it by the historical average read and write rate, the disk I / O read and write rate deviation is obtained. The deviation of multiple I / O-related indicators is combined and added to calculate the overall disk I / O deviation. Use the server log analysis tool to parse the HTTP access log generated by the media server. Extract the error status code information recorded in the log, count the number of error status codes that appear per unit time, and divide it by the total number of HTTP requests in the same time period to obtain the HTTP error frequency. Use the server's network connection monitoring tool to obtain the server's network connection information in real time, count the number of abnormal connections, and divide it by the total number of connections to obtain the abnormal connection ratio.

[0028] Once again, it should be noted that the signal strength monitoring function module is integrated on the terminal device, and the API interface provided by the device operating system is used to obtain signal strength data. For example, the TelephonyManager class in the Android system is used to obtain the mobile network signal strength, and the Wi-FiManager class is used to obtain the Wi-Fi signal strength. The signal strength values are recorded at certain time intervals. Calculate the difference in signal strength between adjacent time points, and then divide it by the time interval to obtain the signal attenuation gradient. On the network path between the terminal device and the target media communication platform, by sending test data packets, record the number of sent data packets and the number of successfully received data packets. Calculate the packet loss rate, that is, (number of sent data packets - number of received data packets) / number of sent data packets. According to the historical packet loss rate data or industry standards, set the normal packet loss rate range, compare the current packet loss rate with the normal range, and obtain the packet loss anomaly index. Embed statistical code in the media player application. When the user performs a channel switching operation, trigger the counting function to record the number of channel switches by the user in a certain time period in real time. Take the number of channel switches as the user channel switching frequency index.

[0029] A2. Substitute the network hardware coefficient, media server operation coefficient, and terminal user experience coefficient corresponding to each communication line in the target media communication platform into the calculation formula: to obtain the joint communication index evaluation value φ corresponding to each communication line in the target media communication platform g .

[0030] It should be noted that erf is the error function, which is widely used in fields such as statistics, probability theory, physics, and engineering. In the calculation scenario of this joint communication index evaluation value, the main role of the erf function is to introduce non-linearity: it can perform non-linear transformation on the combined effects of the network hardware coefficient, media server operation coefficient, and terminal user experience coefficient, better simulate the complex interaction relationships of various factors in the network, and make the evaluation value calculation more in line with the actual network situation. It can map the relevant logarithmic operation results to a specific range, affect the evaluation value calculation, and highlight or weaken the contribution of certain factor combinations to the final evaluation value.

[0031] In a specific embodiment, the analysis to obtain the network hardware coefficient, media server operation coefficient, and terminal user experience coefficient corresponding to each communication line in the target media communication platform is as follows: B1. Denote the inbound traffic mutation rate, UDP traffic ratio change rate, and CPU temperature deviation degree corresponding to each communication line in the target media communication platform as ω q , ξ q and ψ q , and perform normalization processing, and substitute them into the calculation formula:

[0032] to obtain the network hardware coefficient Z corresponding to each communication line in the target media communication platform g , where k represents the natural constant;

[0033] B2. Denote the disk I / O deviation degree, HTTP error frequency, and abnormal connection ratio corresponding to each communication line in the target media communication platform as σ q , and τ q respectively, and perform normalization processing, then substitute them into the calculation formula: to obtain the media server operation coefficient X corresponding to each communication line in the target media communication platform g .

[0034] It should be noted that arctan is the inverse tangent function, also called the inverse tangent operator. In the scenario of the formula for calculating the media server operation coefficient, the inverse tangent function plays a non-linear transformation role: it performs a non-linear transformation on the comprehensive value of the disk I / O deviation degree, HTTP error frequency, and abnormal connection ratio, simulating the complex non-linear relationship between various indicators during the operation of the media server, making the calculated operation coefficient more in line with the actual situation. It maps relevant values to a certain range, avoiding the situation where the calculation results are too large or too small to be processed, and making the finally obtained media server operation coefficient within a reasonable and controllable range.

[0035] B3. Denote the signal attenuation gradient, packet loss anomaly index, and user channel switching frequency corresponding to each communication line in the target media communication platform as θ q and respectively, and perform normalization processing, then substitute them into the calculation formula: to obtain the terminal user experience coefficient V corresponding to each communication line in the target media communication platform g , where is the standard user channel switching frequency corresponding to the set communication line.

[0036] It should be noted that sgn is the sign function. In the scenario of the formula for calculating the terminal user experience coefficient, sgn is used to adjust the calculation result according to the positive or negative situation of the difference between the user channel switching frequency and the standard user channel switching frequency. It can reflect the deviation direction of the actual channel switching frequency relative to the standard value, and thus affect the calculation of the terminal user experience coefficient, making this coefficient more accurately reflect the relevant situation of user experience.

[0037] Step 2. Identification of abnormal network environment: According to the joint communication index evaluation values corresponding to each communication line in the target media communication platform, further evaluate whether the network environment operation corresponding to each communication line in the target media communication platform is abnormal.

[0038] In a specific embodiment, it is determined whether the network environment corresponding to each communication line in the target media communication platform is operating abnormally. The specific evaluation process is as follows: Compare the combined communication index evaluation value corresponding to each communication line in the target media communication platform with the set combined communication index evaluation value range corresponding to the standard communication line. If the combined communication index evaluation value corresponding to a certain communication line in the target media communication platform is within the set combined communication index evaluation value range corresponding to the standard communication line, it is evaluated that the network environment corresponding to this communication line in the target media communication platform is operating normally. If the combined communication index evaluation value corresponding to a certain communication line in the target media communication platform is not within the set combined communication index evaluation value range corresponding to the standard communication line, it is evaluated that the network environment corresponding to this communication line in the target media communication platform is operating abnormally;

[0039] If it is evaluated that the network environment corresponding to a certain communication line in the target media communication platform is operating abnormally, an analysis of the response decision for this communication line in the target media communication platform is performed.

[0040] In a specific embodiment, the analysis of the response decision for this communication line in the target media communication platform is as follows: If the network environment corresponding to a certain communication line in the target media communication platform is operating abnormally, full-flow mirroring is immediately started to mirror and copy all the traffic of this communication line in the target media communication platform. At the same time, it is quickly switched to the disaster recovery center to transfer the service of this communication line in the target media communication platform to the standby network environment. Subsequently, an SDN flow table cleaning operation is triggered to clean and optimize the network flow table through software-defined network technology, remove abnormal traffic rules, and restore normal network traffic transmission. At the same time, the UDP port bandwidth is restricted to prevent network congestion caused by excessive UDP traffic, and an alarm notification is sent to relevant network management personnel or the operation and maintenance team to remind them to pay attention to the problem of this communication line in the target media communication platform.

[0041] Step 3: Evaluation of historical network environment anomalies: Obtain the historical network environment anomaly data corresponding to each communication line in the target media communication platform for each historical period, and then evaluate whether the target media communication platform needs to be updated and adapted to new devices.

[0042] In a specific embodiment, the evaluation of whether the target media communication platform needs to be updated and adapted to new devices is as follows: C1. Obtain the number and duration of each type of anomaly corresponding to each communication line in the target media communication platform for the historical period, and record them as μ qw and η qw, where q represents the number corresponding to each communication line, q = 1, 2......g, g is any integer greater than 2, and g is also the sum of all communication lines. w represents the number corresponding to each type of exception, w = 1, 2......h, h is any integer greater than 2. At the same time, obtain the design life and rated capacity of each device in each communication line in the target media communication platform, and record them as π qy and ρ qy , where y represents the number corresponding to each device, y = 1, 2......n, n is any integer greater than 2, and n is also the sum of all devices. Substitute into the calculation formula: to obtain the updated requirement evaluation value corresponding to the target media communication platform where ζ1 and ζ2 are the weight factors corresponding to the number of times of each type of exception and the duration of each type of exception in the set communication line respectively, υ1 and υ2 are the weight factors corresponding to the design life and rated capacity of each device in the set communication line respectively, δ q and Δδ q are the current cycle exception frequency and the historical cycle average exception frequency corresponding to each communication line in the historical cycle target media communication platform respectively.

[0043] It should be noted that by deploying a log recording tool in the network devices, media servers and related systems of the target media communication platform, various exception events are recorded in real time. The log will contain information such as the timestamp of the exception occurrence and the exception type identifier. When obtaining historical cycle data, according to the pre-set exception type classification standard, the log data is screened and statistically analyzed, classified by communication line number and exception type number, and the number of occurrences of various exceptions in each communication line is counted. Similarly, based on the log record, the start time is recorded when the exception is detected, and the end time is recorded when the exception ends. For continuously occurring exceptions, its duration can be calculated through the timestamp. In the historical cycle, the duration of each exception event is statistically analyzed by communication line and exception type respectively, and accumulated to obtain the duration corresponding to each type of exception in each communication line.

[0044] It should also be noted that the design life of the equipment is usually provided by the equipment manufacturer in the product documentation. When deploying the equipment on the target media communication platform, relevant documentation of the equipment should be collected and sorted out, and an equipment information database should be established. Information such as the number, model, and design life of each equipment should be clearly recorded in the database. Subsequently, when evaluating the update requirements, the design life data of the corresponding equipment in each communication line can be directly retrieved from the database. For network equipment, such as the port bandwidth capacity of routers, the backplane bandwidth of switches, and media servers, such as storage capacity and processing capacity, their rated capacities are also given by the manufacturer. These can be obtained through materials such as the equipment specification sheets and technical manuals. Similarly, these information are entered into the equipment information database to facilitate accurately obtaining the rated capacity data of the corresponding equipment in each communication line when calculating the update requirement evaluation value.

[0045] Once again, it should be noted that according to the number of times of each type of anomaly corresponding to each communication line obtained previously in the current period, combined with the time span of the current period. Divide the total number of occurrences of various anomalies in each communication line by the duration of the current period to obtain the anomaly frequency corresponding to each communication line in the current period.

[0046] Historical period average anomaly frequency: Collect the data on the number of times of each type of anomaly corresponding to each communication line in multiple historical periods. Calculate the anomaly frequency for each historical period separately, and then perform an arithmetic average of the anomaly frequencies of these historical periods to obtain the historical period average anomaly frequency.

[0047] C2. Compare the update requirement evaluation value corresponding to the target media communication platform with the update requirement evaluation value corresponding to the set standard media communication platform. If the update requirement evaluation value corresponding to the target media communication platform is greater than the update requirement evaluation value corresponding to the set standard media communication platform, it is evaluated that the target media communication platform needs to be updated to adapt to new equipment. If the update requirement evaluation value corresponding to the target media communication platform is less than or equal to the update requirement evaluation value corresponding to the set standard media communication platform, it is evaluated that the target media communication platform does not need to be updated to adapt to new equipment.

[0048] Step Four: Analysis of the new equipment adaptation plan: If the target media communication platform needs to be updated to adapt to new equipment, obtain the online live service data, service quality data, and service expansion planning data corresponding to the target media communication platform in the current period, and then analyze the new equipment adaptation plan corresponding to the target media communication platform.

[0049] In a specific embodiment, the analysis of the new device adaptation solution corresponding to the target media communication platform is as follows: Analyze the comprehensive service evaluation value corresponding to the target media communication platform, and compare the comprehensive service evaluation value corresponding to the target media communication platform with the comprehensive service evaluation value ranges corresponding to each new device adaptation solution in the database. If the comprehensive service evaluation value corresponding to the target media communication platform is within the comprehensive service evaluation value range corresponding to a certain new device adaptation solution in the database, then use the new device adaptation solution in the database as the new device adaptation solution corresponding to the target media communication platform.

[0050] It should be noted that the database is used to store the comprehensive service evaluation value ranges corresponding to each new device adaptation solution.

[0051] In a specific embodiment, the analysis of the comprehensive service evaluation value corresponding to the target media communication platform is as follows: Analyze the online live broadcast service coefficient, service quality coefficient, and service expansion planning coefficient corresponding to the target media communication platform, and denote them as S, G, and H respectively, and substitute them into the calculation formula: to obtain the comprehensive service evaluation value χ corresponding to the target media communication platform, where S′, G′, and H′ are the standard online live broadcast service coefficient, standard service quality coefficient, and standard service expansion planning coefficient corresponding to the set media communication platform respectively, and Θ1, Θ2, and Θ3 are the weight factors corresponding to the online live broadcast service coefficient, service quality coefficient, and service expansion planning coefficient of the set media communication platform respectively.

[0052] It should be noted that Θ1, Θ2, and Θ3 are all greater than 0 and less than 1.

[0053] It should also be noted that by collecting a large amount of online live broadcast service data (such as the number of concurrent live users, the average daily on-demand request number, the proportion of popular videos, etc.), service quality data (the number of data transmission path devices, the expected data transmission volume, the bandwidth fluctuation range required for the service, etc.), and service expansion planning data (the number of new live channels, the annual growth rate of the user scale, the expected number of concurrent users for each channel, etc.) of the same type and similar-scale platforms, the industry average value is obtained through statistical analysis as the standard online live broadcast service coefficient, standard service quality coefficient, and standard service expansion planning coefficient corresponding to the media communication platform. Through a combination of expert evaluation, multiple rounds of simulation tests, and data analysis, according to the correlation between the comprehensive service evaluation value and the actual service effect of the platform under different weight combinations, optimization and adjustment are carried out to ensure that the setting of the weight factors is scientific and reasonable, and the weight factors corresponding to the online live broadcast service coefficient, service quality coefficient, and service expansion planning coefficient of the set media communication platform are set by experts.

[0054] In a specific embodiment, the online live broadcast service coefficient, service quality coefficient, and service expansion planning coefficient corresponding to the target media communication platform are analyzed as follows: D1. Obtain the online live broadcast service data, service quality data, and service expansion planning data corresponding to the target media communication platform in the current period. The online live broadcast service data includes the number of concurrent live users, the average daily on-demand request number, and the proportion of popular videos. The service quality data includes the number of devices on the data transmission path, the expected data transmission volume, and the fluctuation range of the bandwidth required for the service. The service expansion planning data includes the number of new live channels, the annual growth rate of the user scale, and the expected number of concurrent users for each channel. Then, normalize the online live broadcast service data, service quality data, and service expansion planning data corresponding to the target media communication platform.

[0055] D2. Take the number of concurrent live users, the average daily on-demand request number, and the proportion of popular videos corresponding to the target media communication platform in the current period as input items and import them into the online live broadcast service coefficient analysis model. After the operation and analysis of the online live broadcast service coefficient analysis model, finally output the online live broadcast service coefficient S corresponding to the target media communication platform.

[0056] It should be noted that the analysis process of the online live broadcast service coefficient corresponding to the target media communication platform is as follows: Denote the number of concurrent live users, the average daily on-demand request number, and the proportion of popular videos corresponding to the target media communication platform in the current period as Ω, Ξ, and Π respectively, and substitute them into the analysis formula to obtain the online live broadcast service coefficient S corresponding to the target media communication platform.

[0057] D3. Take the number of devices on the data transmission path, the expected data transmission volume, and the fluctuation range of the bandwidth required for the service corresponding to the target media communication platform in the current period as input items and import them into the service quality coefficient analysis model. After the operation and analysis of the service quality coefficient analysis model, finally output the service quality coefficient G corresponding to the target media communication platform.

[0058] It should be noted that the service quality coefficient corresponding to the target media communication platform is analyzed according to the above analysis process of the online live broadcast service coefficient corresponding to the target media communication platform.

[0059] D4. Take the number of new live channels, the annual growth rate of the user scale, and the expected number of concurrent users for each channel corresponding to the target media communication platform in the current period as input items and import them into the service expansion planning coefficient analysis model. After the operation and analysis of the service expansion planning coefficient analysis model, finally output the service expansion planning coefficient H corresponding to the target media communication platform.

[0060] It should be noted that the business expansion planning coefficient corresponding to the target media communication platform is obtained by analyzing according to the analysis process of the online live broadcast service coefficient corresponding to the above-mentioned target media communication platform.

[0061] As shown in the embodiments of the present invention Figure 2 The media communication network bandwidth traffic anomaly recognition system based on big data includes: a joint communication index evaluation value acquisition module: used to obtain the network hardware metric indicators, media server operation indicators, and terminal user experience indicators corresponding to each communication line in the target media communication platform at the current moment, and then analyze to obtain the joint communication index evaluation value corresponding to each communication line in the target media communication platform.

[0062] A network environment anomaly recognition module: used to evaluate whether the network environment operation corresponding to each communication line in the target media communication platform is abnormal according to the joint communication index evaluation value corresponding to each communication line in the target media communication platform.

[0063] A historical network environment anomaly evaluation module: used to obtain the historical network environment anomaly data corresponding to each communication line in the target media communication platform in each historical period, and then evaluate whether the target media communication platform needs to be updated and adapted to new devices.

[0064] A new device adaptation plan analysis module: used to obtain the online live broadcast service data, service quality data, and business expansion planning data corresponding to the target media communication platform in the current period if the target media communication platform needs to be updated and adapted to new devices, and then analyze the new device adaptation plan corresponding to the target media communication platform.

[0065] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for identifying abnormal bandwidth flow in a media communication network based on big data, characterized in that: include: Step 1: Acquisition of joint communication index evaluation values: Acquiring network hardware metrics, media server operation metrics, and terminal user experience metrics corresponding to each communication line in the target media communication platform at the current moment, and then analyzing and obtaining joint communication index evaluation values ​​corresponding to each communication line in the target media communication platform; Step 2: Identification of network environment anomalies: Based on the joint communication index evaluation value corresponding to each communication line in the target media communication platform, it is evaluated whether the network environment corresponding to each communication line in the target media communication platform is operating abnormally; Step 3: Evaluation of historical network environment anomalies: Obtain historical network environment anomaly data corresponding to each communication line in the target media communication platform in each historical period, and then evaluate whether the target media communication platform needs to be updated and adapted to new equipment; Step 4: Analysis of new equipment adaptation solutions: If the target media communication platform needs to be updated to adapt to new equipment, obtain the online live broadcast service data, service quality data and business expansion planning data corresponding to the target media communication platform in the current period, and then analyze the new equipment adaptation solution corresponding to the target media communication platform.

2. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 1, characterized in that: The analysis obtains the joint communication index evaluation value corresponding to each communication line in the target media communication platform. The specific analysis process is as follows: A1. According to the network hardware metrics, media server operating parameters and terminal user experience indicators corresponding to each communication line in the target media communication platform, the network hardware metrics include the inbound traffic mutation rate, UDP traffic share change rate and CPU temperature deviation, the media server operating indicators include disk I / O deviation, HTTP error frequency and abnormal connection ratio, and the terminal user experience indicators include signal attenuation gradient, packet loss anomaly index and user channel switching frequency. Then, the network hardware coefficient, media server operating coefficient and terminal user experience coefficient corresponding to each communication line in the target media communication platform are analyzed and recorded as Z respectively. q , X q and V q , where q represents the number corresponding to each communication line, q=1,2...g, g is any integer greater than 2; A2. Substitute the network hardware coefficient, media server operation coefficient and terminal user experience coefficient corresponding to each communication line in the target media communication platform into the calculation formula: The joint communication index evaluation value φ corresponding to each communication line in the target media communication platform is obtained g .

3. The method for identifying abnormal bandwidth flow in a media communication network based on big data as claimed in claim 2, characterized in that: The analysis obtains the network hardware coefficient, media server operation coefficient and terminal user experience coefficient corresponding to each communication line in the target media communication platform. The specific analysis process is as follows: B1. The inbound traffic mutation rate, UDP traffic ratio change rate and CPU temperature deviation corresponding to each communication line in the target media communication platform are recorded as ω q , q and ψ q , and normalize it and substitute it into the calculation formula: The network hardware coefficient Z corresponding to each communication line in the target media communication platform is obtained g , where k represents a natural constant; B2. The disk I / O deviation, HTTP error frequency and abnormal connection ratio corresponding to each communication line in the target media communication platform are respectively denoted as σ q , and τ q , and normalize it and substitute it into the calculation formula: The media server operation coefficient X corresponding to each communication line in the target media communication platform is obtained g ; B3. The signal attenuation gradient, packet loss anomaly index and user channel switching frequency corresponding to each communication line in the target media communication platform are recorded as θ q and And perform normalization and substitute into the calculation formula: The terminal user experience coefficient V corresponding to each communication line in the target media communication platform is obtained g ,in, The standard user channel switching frequency corresponding to the set communication line.

4. The method for identifying abnormal bandwidth flow in a media communication network based on big data as claimed in claim 3, characterized in that: The specific evaluation process of whether the network environment corresponding to each communication line in the target media communication platform is abnormal is as follows: Compare the joint communication index evaluation value corresponding to each communication line in the target media communication platform with the set joint communication index evaluation value interval corresponding to the standard communication line; if the joint communication index evaluation value corresponding to a communication line in the target media communication platform is within the set joint communication index evaluation value interval corresponding to the standard communication line, then it is assessed that the network environment corresponding to the communication line in the target media communication platform is operating normally; if the joint communication index evaluation value corresponding to a communication line in the target media communication platform is not within the set joint communication index evaluation value interval corresponding to the standard communication line, then it is assessed that the network environment corresponding to the communication line in the target media communication platform is operating abnormally; If the network environment corresponding to a communication line in the target media communication platform is evaluated to be operating abnormally, an analysis of the response decision of the communication line in the target media communication platform is performed.

5. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 4, characterized in that: The specific analysis process of the response decision analysis of the communication line in the target media communication platform is as follows: If the network environment corresponding to a communication line in the target media communication platform operates abnormally, full traffic mirroring is immediately started to mirror all traffic of the communication line in the target media communication platform. At the same time, the system quickly switches to the disaster recovery center to transfer the business of the communication line in the target media communication platform to the backup network environment. Subsequently, the SDN flow table cleaning operation is triggered to clean and optimize the network flow table through software-defined network technology, remove abnormal traffic rules, and restore normal traffic transmission of the network. At the same time, the UDP port bandwidth is limited to prevent network congestion caused by excessive UDP traffic, and an alarm notification is sent to the relevant network management personnel or operation and maintenance team to remind them to pay attention to the problem of the communication line in the target media communication platform.

6. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 5, characterized in that: The specific evaluation process of evaluating whether the target media communication platform needs to be updated to adapt to new devices is as follows: C1. Obtain the number and duration of each type of anomaly in each communication line in the target media communication platform in the historical period, and record them as μ qw and η qw , where q represents the number corresponding to each communication line, q = 1, 2...g, g is any integer greater than 2, g is also the sum of each communication line, w represents the number corresponding to each type of abnormality, w = 1, 2...h, h is any integer greater than 2, and at the same time, the design life and rated capacity of each device in each communication line in the target media communication platform are obtained and recorded as π respectively qy and ρ qy , where y represents the number corresponding to each device, y = 1, 2...n, n is any integer greater than 2, and n is also the sum of all devices. Substitute it into the calculation formula: In the above example, we obtain the update demand evaluation value corresponding to the target media communication platform. Among them, ζ1 and ζ2 are the weight factors corresponding to the number of abnormalities of each type in the communication line and the weight factors corresponding to the duration, υ1 and υ2 are the weight factors corresponding to the design life and rated capacity of each device in the communication line, respectively. q and Δδ q They are respectively the current cycle abnormal frequency and the historical cycle average abnormal frequency corresponding to each communication line in the target media communication platform of the historical cycle; C2. Compare the update requirement evaluation value corresponding to the target media communication platform with the update requirement evaluation value corresponding to the set standard media communication platform. If the update requirement evaluation value corresponding to the target media communication platform is greater than the update requirement evaluation value corresponding to the set standard media communication platform, it is assessed that the target media communication platform needs to be updated to adapt to new equipment. If the update requirement evaluation value corresponding to the target media communication platform is less than or equal to the update requirement evaluation value corresponding to the set standard media communication platform, it is assessed that the target media communication platform does not need to be updated to adapt to new equipment.

7. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 6, characterized in that: The new device adaptation solution corresponding to the target media communication platform is analyzed, and the specific analysis process is as follows: Analyze the comprehensive business evaluation value corresponding to the target media communication platform, and compare the comprehensive business evaluation value corresponding to the target media communication platform with the comprehensive business evaluation value intervals corresponding to each new equipment adaptation scheme in the database; if the comprehensive business evaluation value corresponding to the target media communication platform is within the comprehensive business evaluation value interval corresponding to a new equipment adaptation scheme in the database, then the new equipment adaptation scheme in the database will be used as the new equipment adaptation scheme corresponding to the target media communication platform.

8. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 7, characterized in that: The comprehensive service evaluation value corresponding to the target media communication platform is analyzed, and the specific analysis process is as follows: Analyze the online live broadcast business coefficient, business service quality coefficient and business expansion planning coefficient corresponding to the target media communication platform, and record them as S, G and H, and substitute them into the calculation formula: The comprehensive business evaluation value χ corresponding to the target media communication platform is obtained, where S′, G′, and H′ are the standard online live broadcast business coefficient, standard business service quality coefficient, and standard business expansion planning coefficient corresponding to the set media communication platform, respectively; Θ1, Θ2, and Θ3 are the weight factors corresponding to the online live broadcast business coefficient, the weight factors corresponding to the business service quality coefficient, and the weight factors corresponding to the business expansion planning coefficient of the set media communication platform, respectively.

9. The method for identifying abnormal bandwidth flow in a media communication network based on big data according to claim 8, characterized in that: The online live broadcast business coefficient, business service quality coefficient and business expansion planning coefficient corresponding to the target media communication platform are analyzed, and the specific analysis process is as follows: D1. Obtain the online live broadcast service data, service quality data and business expansion planning data corresponding to the target media communication platform in the current cycle. The online live broadcast service data includes the number of concurrent live broadcast users, the average number of on-demand requests per day and the proportion of popular videos. The service quality data includes the number of data transmission path devices, the expected data transmission volume and the fluctuation range of the bandwidth required for the service. The business expansion planning data includes the number of newly added live broadcast channels, the annual growth rate of user scale and the expected number of concurrent users of each channel. The online live broadcast service data, service quality data and business expansion planning data corresponding to the target media communication platform are normalized; D2. The number of concurrent live broadcast users, the average number of daily on-demand requests, and the proportion of popular videos corresponding to the target media communication platform in the current period are taken as input items and imported into the online live broadcast business coefficient analysis model. After calculation and analysis by the online live broadcast business coefficient analysis model, the online live broadcast business coefficient S corresponding to the target media communication platform is finally output; D3, taking the number of data transmission path devices, expected data transmission volume and bandwidth fluctuation range required by the target media communication platform in the current period as input items, and importing them into the business service quality coefficient analysis model, after calculation and analysis by the business service quality coefficient analysis model, finally outputting the business service quality coefficient G corresponding to the target media communication platform; D4. The number of newly added live broadcast channels, the annual growth rate of user scale and the expected number of concurrent users of each channel corresponding to the target media communication platform in the current period are taken as input items and imported into the business expansion planning coefficient analysis model. After calculation and analysis by the business expansion planning coefficient analysis model, the business expansion planning coefficient H corresponding to the target media communication platform is finally output.

10. A media communication network bandwidth flow anomaly identification system based on big data that executes the media communication network bandwidth flow anomaly identification method based on big data as described in any one of claims 1 to 9, characterized in that: include: The joint communication index evaluation value acquisition module is used to obtain the network hardware measurement indicators, media server operation indicators and terminal user experience indicators corresponding to each communication line in the target media communication platform at the current moment, and then analyze and obtain the joint communication index evaluation value corresponding to each communication line in the target media communication platform; Network environment anomaly identification module: used to evaluate whether the network environment corresponding to each communication line in the target media communication platform is abnormal according to the joint communication index evaluation value corresponding to each communication line in the target media communication platform; Historical network environment anomaly assessment module: used to obtain historical network environment anomaly data corresponding to each communication line in the target media communication platform in each historical period, and then assess whether the target media communication platform needs to be updated and adapted to new equipment; New equipment adaptation solution analysis module: used to obtain the online live broadcast business data, business service quality data and business expansion planning data corresponding to the target media communication platform in the current period if the target media communication platform needs to be updated to adapt to new equipment, and then analyze the new equipment adaptation solution corresponding to the target media communication platform.

Images