CN108271189A - A kind of quality of service monitoring method and device - Google Patents

Abstract

The invention discloses a service quality monitoring method, which includes: obtaining the service characteristics and signaling characteristics of the current voice service; based on the service characteristics of the current voice service, performing a matching search in the service characteristic library of the abnormal voice service; when the When the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, a matching search is performed in the signaling feature database of the abnormal voice service; when the When the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, the service corresponding to the signaling feature of the abnormal voice service that matches the signaling feature of the current voice service A quality defect is used as a service quality defect of the current voice service. At the same time, the invention also discloses a service quality monitoring device.

Description

A service quality monitoring method and device

technical field

The invention relates to mobile communication technology, in particular to a service quality monitoring method and device.

Background technique

Voice over Long Term Evolution (VoLTE, Voice over Long Term Evolution) is a voice service solution based on Long Term Evolution (LTE, Long Term Evolution,) network data domain, which can provide users with voice services and video service. However, due to the complex network structure and long service process of the VoLTE network, the VoLTE service quality is not only affected by common abnormal problems such as dropped calls and unconnected calls, but also needs to pay special attention to the service experience of users in high-definition calls, such as during service use. Whether there are single-pass, short-term no sound, intermittent, stuttering, word swallowing and other problems that seriously affect user perception. Therefore, through the evaluation of service quality and analysis of abnormal problems, hidden dangers of network problems can be quickly discovered and located, from passively waiting for users to report service quality problems to proactively analyzing service deficiencies, and promoting network optimization, planning and construction, and marketing from network and Business-centric to user-centric, thereby promoting customer satisfaction and loyalty.

Service quality can be subdivided into network-level and user-level. Among them, the network is the carrier of services, and network-level service quality is a concept of macro balance, which refers to the average situation of network-carried services in a certain area in a certain period of time; Or, user-level service quality is a microscopic end-to-end concept, which refers to the service performance and user experience felt from the user's perspective. Therefore, the evaluation and analysis of user-level service quality can promote the transformation of network optimization and marketing from network-oriented to user-centered, so as to better measure the overall satisfaction of users with operators.

Commonly used service quality analysis methods mainly include subjective questionnaire survey, equipment performance statistics, and drive test. Subjective questionnaire survey and drive test belong to user-level service quality assessment, and equipment performance statistics belong to network-level service quality assessment. Among them, subjective questionnaire survey is to collect user feedback on service perception and service quality through design questions; device performance statistics is to regularly collect internal performance data of each device in the network through the network management system, and output the device after statistical analysis of the performance data The performance indicators can reflect the quality of service carried by the device in a macroscopic way; the drive test method is to conduct service dial-up test in the corresponding test scenario, track and capture the standard protocol messages and user plane content of the corresponding interface, and analyze the service based on these information. Quality is assessed.

However, the existing service quality analysis methods have certain limitations. The user-level service quality analysis cannot realize the long-term service quality tracking analysis and rapid problem location for the entire network users, while the network-level service quality analysis focuses more on the performance of network equipment. , Key Performance Indicator (KPI, Key Performance Indicator), etc., ignoring the user end-to-end experience difference between the network bearer quality and the actual use of the service. With the development of networks and services, especially for VoLTE networks, the difference between the KPI indicators that can be directly measured and the real user service quality experience is getting bigger and bigger, and different users in the same time and environment or the same user There are also differences in service quality at different times and environments, resulting in an increasing discrepancy between the sampling evaluation results and the service quality of users on the entire network.

The specific shortcomings of various existing service quality analysis methods are as follows:

The final result of the subjective questionnaire survey has a great relationship with the number and distribution of sampling samples, and is affected by factors such as local economic conditions, cultural environment, population size, etc., and cannot fully reflect the service quality of all users on the entire network; at the same time, the operation process is complex, The cost is high; it can only passively wait for users to report service quality problems and check them, but cannot actively discover hidden dangers and locate the specific reasons for service quality impacts.

Equipment performance statistics are generally based on "network", "network element", "link" and other units. It is an average summary of the network element and the performance of the area within a period of time. It cannot directly correspond to the user and cannot reflect the performance of a certain user. A certain service usage situation, such as whether the voice and image are clear and smooth; various service performance indicators for evaluating network quality are generated through device logs, network management reports, and signaling data, which cannot reflect the impact of the terminal on the user's service quality. A true measure of experience.

Although the drive test method can simulate the user's business use process, due to the limitations of the test scope, test tools, test times, and operation methods, the information obtained is a certain amount of sampling values, which has certain limitations and cannot fully reflect the overall situation. At the same time, this method requires special personnel and special vehicles to test, and the cost is relatively high, so it is not suitable for long-term implementation.

Therefore, in the situation of rapid development of VoLTE services, it is urgent to establish a user-level service quality analysis method.

Contents of the invention

In view of this, the embodiment of the present invention expects to provide a service quality monitoring method and device, which can realize accurate evaluation and analysis of user-level service quality.

In order to achieve the above object, technical solution of the present invention is achieved in that way:

An embodiment of the present invention provides a service quality monitoring method, the method comprising:

Obtain the service characteristics and signaling characteristics of the current voice service;

Based on the service feature of the current voice service, a matching search is performed in the service feature database of the abnormal voice service;

When the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, performing a matching search in the signaling feature database of the abnormal voice service based on the signaling feature of the current voice service;

When the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature library, the signaling feature of the abnormal voice service that matches the signaling feature of the current voice service corresponds to The service quality defect of is used as the service quality defect of the current voice service.

In the above solution, the acquisition of the service characteristics and signaling characteristics of the current voice service includes:

Collecting service data packets and signaling messages of the current voice service, and analyzing the service data packets and the signaling messages respectively, obtaining the service data and signaling of the current voice service, and obtaining the service data and signaling messages from the service data Filter out the service characteristics of the current voice service, and filter out the signaling characteristics of the current voice service from the signaling.

In the above solution, the method further includes:

Collecting service data packets of at least one voice service in the communication network, and analyzing the collected service data packets of the at least one voice service, obtaining service data of the at least one voice service, and obtaining the service data from the at least one voice service Selecting the service characteristics of the at least one voice service from the service data;

Based on the service feature of the at least one voice service, perform a matching search in the service feature database of the abnormal voice service;

When the service feature of the at least one voice service matches the service feature of the abnormal voice service in the service feature database, storing the service feature of the at least one voice service in the service feature database;

Perform cluster analysis on the business features in the business feature database, generate analysis results, and update the business feature database according to the analysis results.

In the above scheme, the method also includes:

Locating the management subject of the service quality defect of the current voice service, and outputting alarm information for the current voice service to the management subject.

In the above solution, the management subject for locating the service quality defect of the current voice service includes:

Determining and discovering the abnormal network element in the signaling message of the current voice service, and obtaining the management subject of the service quality defect of the current voice service according to the management domain to which the network element belongs and the management subject of each management domain.

An embodiment of the present invention provides a service quality monitoring device, which includes: a first processing unit, a second processing unit, a third processing unit, and a fourth processing unit; wherein,

The first processing unit is configured to obtain service characteristics and signaling characteristics of the current voice service;

The second processing unit is configured to perform a matching search in the service feature library of the abnormal voice service based on the service feature of the current voice service;

The third processing unit is configured to, when the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, in the abnormal voice service Matching lookup in the signaling signature database;

The fourth processing unit is configured to, when the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, match the signaling feature of the current voice service The quality of service defect corresponding to the signaling feature of the abnormal voice service is used as the quality of service defect of the current voice service.

In the above solution, the first processing unit is specifically used for:

Collecting service data packets and signaling messages of the current voice service, and analyzing the service data packets and the signaling messages respectively, obtaining the service data and signaling of the current voice service, and obtaining the service data and signaling messages from the service data Filter out the service characteristics of the current voice service, and filter out the signaling characteristics of the current voice service from the signaling.

In the above scheme,

The first processing unit is further configured to collect service data packets of at least one voice service in the communication network, analyze the collected service data packets of the at least one voice service, and obtain service data packets of the at least one voice service data, and filter out the service characteristics of the at least one voice service from the service data of the at least one voice service;

The second processing unit is further configured to perform a matching search in the service feature library of the abnormal voice service based on the service feature of the at least one voice service;

The third processing unit is further configured to store the service feature of the at least one voice service in the A business feature database; performing cluster analysis on the business features in the business feature database, generating analysis results, and updating the business feature database according to the analysis results.

In the solution above, the device further includes: a fifth processing unit, configured to locate the management subject of the service quality defect of the current voice service, and output alarm information for the current voice service to the management subject.

In the above solution, the fifth processing unit is specifically used for:

Determining and discovering the abnormal network element in the signaling message of the current voice service, and obtaining the management subject of the service quality defect of the current voice service according to the management domain to which the network element belongs and the management subject of each management domain.

The service quality monitoring method and device provided by the embodiments of the present invention acquire the service characteristics and signaling characteristics of the current voice service; based on the service characteristics of the current voice service, perform a matching search in the service feature library of the abnormal voice service; when the When the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, a matching search is performed in the signaling feature database of the abnormal voice service; when the When the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, the service corresponding to the signaling feature of the abnormal voice service that matches the signaling feature of the current voice service Quality defects, as the service quality defects of the current voice service; it can be seen that the embodiment of the present invention is based on the service characteristic analysis of the voice service, combined with the signaling characteristics of the voice service, and establishes a user-level service quality evaluation analysis model, thereby quickly judging Quality defects and hidden dangers in the use of voice services, solve the problem that traditional service quality analysis methods cannot find and locate abnormal end-to-end service experience of a single user, and realize accurate evaluation and analysis of user-level service quality.

In addition, it expands the way to discover service quality problems, solves the problem that equipment alarms and performance indicators in the network cannot detect the degradation of single-user service quality, and improves the early warning capability; it can determine the degree of impact of service quality defects on user perception, and realizes End-to-end business quality assessment and analysis, helping maintenance personnel achieve the goal of focusing on and prioritizing problems that users cannot tolerate; adopting active operation and maintenance methods, intelligently and accurately locating and demarcating problems, improving the efficiency of the entire process of problem discovery, location, and processing, and improving user Business experience, satisfaction and online loyalty, and through real-time alarms and early warnings, find and locate potential problems before user complaints; provide timely and effective information for monitoring, maintenance, optimization, emergency response, customer service and other departments, Promote the transformation of various tasks from network-centric to user-centric, link network support and customer service more closely together, and promote integrated management to achieve excellent operations.

Description of drawings

FIG. 1 is a schematic diagram of an implementation process of a service quality monitoring method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a specific implementation process of a service quality monitoring method according to an embodiment of the present invention;

Fig. 3 is a schematic diagram of the composition and structure of a service quality monitoring device according to an embodiment of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 is a schematic diagram of the implementation process of the service quality monitoring method of the embodiment of the present invention, the method comprising:

Step 101: Obtain the service characteristics and signaling characteristics of the current voice service;

Specifically, collect the service data packet and signaling message of the current voice service, and analyze the service data packet and the signaling message respectively, obtain the service data and signaling of the current voice service, and obtain the service data and signaling message from the The service characteristics of the current voice service are selected from the service data of the current voice service, and the signaling characteristics of the current voice service are selected from the signaling of the current voice service.

Here, the service data packets and signaling messages of the current voice service may be collected from an interface such as LTE S1 on the network side; the current voice service may be such services as making a call and receiving a call.

Step 102: Based on the service feature of the current voice service, perform a matching search in the service feature library of the abnormal voice service;

Specifically, according to the service feature of the current voice service, search for the service feature database of the abnormal voice service, and judge whether there is a service feature of the abnormal voice service matching the service feature of the current voice service in the service feature database, That is, it is judged whether the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature library, if it matches, it indicates that the current voice service is an abnormal voice service, and step 103 is performed; if not matched, then Indicate that the current voice service is not an abnormal voice service, and end the analysis.

Here, the service feature library of the abnormal voice service includes the service feature of the abnormal voice service; the abnormal voice service can be ultra-short call, over-frequency call, early release and other services; the service feature of the abnormal voice service refers to Some specific features usually appear in abnormal voice services, for example, for an ultra-short call, the service feature of the ultra-short call may be that the duration of the call is within a certain time threshold.

Step 103: When the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, perform match lookup;

Specifically, according to the matching search result in step 102, when it is determined that the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, according to the service feature of the current voice service acquired in step 101, Signaling feature, searching the signaling feature library of the abnormal voice service, judging whether there is a signaling feature of the abnormal voice service matching the signaling feature of the current voice service in the signaling feature library, that is, judging the current Whether the signaling feature of the voice service matches the signaling feature of the abnormal voice service in the signaling feature database, if they match, execute step 103; if not, end the analysis.

Here, the signaling feature library of the abnormal voice service contains the signaling features of the abnormal voice service; the signaling feature of the abnormal voice service refers to some signaling features that usually appear in the abnormal voice service, such as for The ultra-short call, the signaling feature of the ultra-short call may be that the packet loss rate is greater than a certain threshold value, the MOS value is lower than a certain threshold value, and the like.

Step 104: When the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, send the signaling of the abnormal voice service that matches the signaling feature of the current voice service The quality of service defect corresponding to the feature is used as the quality of service defect of the current voice service.

Specifically, according to the matching search result in step 103, when it is determined that the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, the signaling feature of the current voice service will be matched with the signaling feature of the current voice service. The quality of service defect corresponding to the signaling feature of the abnormal voice service whose feature matches is used as the quality of service defect of the current voice service.

Here, the quality of service defect corresponding to the signaling feature of the abnormal voice service can be stored in the signaling feature library, so that the defect corresponding to the signaling feature of the abnormal voice service can be quickly obtained according to the signaling feature of the abnormal voice service. Service quality defects; each abnormal voice service may have multiple service quality defects due to different signaling characteristics, that is, for any abnormal voice service, when the signaling characteristics of the abnormal voice service are different, the abnormal voice service The QoS defects corresponding to signaling features are also different.

Here, when the signaling feature of the current voice service matches the signaling feature of any abnormal voice service in the signaling feature library, it indicates that the signaling feature of the current voice service is consistent with the signaling feature Match the signaling features of abnormal voice services in the library.

Further, after step 104, the method may also include:

Locating the management subject of the service quality defect of the current voice service, and outputting alarm information for the current voice service to the management subject.

Specifically, determine and discover the abnormal network element in the signaling message of the current voice service, and obtain the service quality defect information of the current voice service according to the management domain to which the network element belongs and the management subject of each management domain A management subject, and output warning information for the current voice service to the management subject, so as to inform the management subject of information such as the type and cause of service quality defects of the current voice service.

Here, since the signaling interaction of the voice service is performed serially, failure of any step in the signaling will lead to the failure of the subsequent signaling interaction. Therefore, the current voice can be found for the first time according to the signaling message of the voice If there is an abnormal network element in the signaling message of the service, then determine the management domain of the network element and the corresponding management subject.

Further, before step 101, the method may also include:

Collecting service data packets of at least one voice service in the communication network, and analyzing the collected service data packets of the at least one voice service, obtaining service data of the at least one voice service, and obtaining the service data from the at least one voice service Selecting the service characteristics of the at least one voice service from the service data;

Based on the service feature of the at least one voice service, perform a matching search in the service feature database of the abnormal voice service;

When the service feature of the at least one voice service matches the service feature of the abnormal voice service in the service feature database, storing the service feature of the at least one voice service in the service feature database;

Perform cluster analysis on the business features in the business feature database, generate analysis results, and update the business feature database according to the analysis results.

Specifically, the service data packets of one or more voice services in the communication network are collected in advance, and after analyzing the service data packets of the one or more voice services, the information contained in the service data packets of the one or more voice services is obtained. Corresponding service data, and then screen out the service characteristics of the one or more voice services from the service data; According to the service characteristics of the one or more voice services, search the service feature database of the abnormal voice service, and judge the Whether the service characteristics of the one or more current voice services match the service characteristics of the abnormal voice service in the service characteristic database; When the service characteristics are matched, the service characteristics of the one or more voice services are stored in the service characteristic library; then, cluster analysis is performed on the service characteristics corresponding to the same abnormal voice service in the service characteristic library, etc. Self-learning, to generate new service features of the same abnormal voice service, and implement update processing of the service features of the abnormal voice service in the service feature database.

Here, the service features of the abnormal voice service in the service feature library can be set in advance, such as for any abnormal voice service, a service feature can be set respectively first, and then the service feature of the set abnormal voice service can be updated in subsequent processing; Alternatively, after step 104 is performed, the service feature of the current voice service may be stored in the service feature database, and the service feature database may be updated.

Here, it should be noted that, the signaling feature of the abnormal voice service in the signaling feature database may also be acquired according to the manner of acquiring the service feature of the abnormal voice service in the service feature database.

The service quality monitoring method provided in this embodiment can change from passively waiting for users to report service quality problems to proactively discovering and solving service quality problems, from focusing on equipment performance to considering end-to-end user experience, and from network element-centered optimization The model has changed to a user-centered optimization model to improve user satisfaction with service quality. Only when users feel that service quality meets or exceeds their expectations can they improve service satisfaction and online loyalty, and they are more willing to try New business, thus laying a solid foundation for the continuous development of operators.

The following is a specific example to further illustrate the embodiment of the present invention. FIG. 2 is a schematic diagram of a specific implementation flow of the service quality monitoring method of the embodiment of the present invention. The method includes:

Step 201: analyzing current user behavior;

Here, the user behavior refers to a specific operation behavior of the user in the process of using the voice service, such as making a call.

Specifically, the service data packet of the current user in the process of using the voice service is obtained from the network side; according to the service data packet, the service characteristics of the current user in the process of using the voice service are obtained; Business characteristics in the course of business use and business characteristics of user abnormal behavior in the established user behavior model, analyze whether the business characteristics of current user behavior match the business characteristics of user abnormal behavior in the user behavior model, and generate analysis results .

Here, before step 201, the user behavior model can be established according to the historical user behavior data; the user behavior model includes abnormal user behaviors such as ultra-short calls, over-frequency calls, early releases, etc. business characteristics; when the business characteristics of the current user behavior meet the business characteristics of any abnormal user behavior, the current user behavior belongs to the abnormal user behavior; the business characteristics of each abnormal user behavior can be Based on the historical data of one or more users.

For the ultra-short call behavior, a kind of ultra-short call threshold algorithm is proposed in the present embodiment to obtain the service characteristics of the ultra-short call behavior, and the specific process is as follows:

The first step is to analyze the user's ultra-short call behavior; after the voice call is connected, if the voice call is a one-way call, since the user often says: "Hello, do you hear it?" Judging the speech rate of 180-200 words per minute, the call time of the ultra-short call behavior is about 3-10 seconds;

The second step, historical data analysis; use the call duration and MOS value in the historical data of one or more users to perform cluster analysis, further narrow the call time range of ultra-short call behavior, and find out how many seconds after the call duration, MOS value When the percentage growth rate is less than 2.8, the inflection point is the threshold value, so as to obtain the service characteristics of ultra-short call behavior, that is, the call duration;

The third step is the live network verification; according to the single-way complaints of one or more users, analyze the bill duration, and verify the service characteristics of the ultra-short call behavior.

For the overfrequency call behavior, a kind of overfrequency call threshold value algorithm is proposed in this embodiment to obtain the service characteristics of the overfrequency call behavior, and the specific process is as follows:

The first step is to analyze the user's overfrequency call behavior; if there is a single call during the call, the user will often look at the phone screen to confirm after the call is hung up, and then find the last call record in the call record, and then choose to re-initiate the voice call, the interval between overclocking calls is about 8-12 seconds;

The second step is historical data analysis; use the overfrequency call interval and packet loss rate value in the historical data to perform cluster analysis, further narrow the range of overfrequency call interval, and find out how many seconds after the call interval, the packet loss rate is greater than If the proportion increase of more than 3% starts to decrease, this inflection point is the threshold value, so as to obtain the service characteristics of overfrequency call behavior, that is, the call interval time;

The third step is the live network verification; according to the single-way complaints of one or more users, analyze the call interval between two bills, and verify the service characteristics of the overfrequency call behavior.

In this embodiment, the user behavior model includes service characteristics corresponding to three abnormal user behaviors of ultra-short call, over-frequency call, and early release as an example; wherein, the ultra-short call refers to the Hang up within 4 seconds to 9 seconds after the end; the overfrequency call is that the user initiates a call again within 8 seconds after the calling or called last call ends, satisfying that the call object is consistent with the last call; The explanation is that the user hangs up before the call is not connected.

Step 202: judging whether the current user behavior is an abnormal user behavior, if so, execute step 203, otherwise execute step 206;

Specifically, according to the analysis result in step 201, it is judged whether the current user behavior is an abnormal user behavior in the user behavior model, that is, it is judged whether the current user behavior is an ultra-short call, an over-frequency call, or an early release. Either, if the current user behavior is any abnormal user behavior in the user behavior model, perform step 203; if the current user behavior is not any abnormal user behavior in the user behavior model , then step 206 is executed.

Here, when the current user behavior is any abnormal user behavior in the user behavior model, it can be preliminarily determined that the current user may have a quality problem that seriously affects perception during the use of the voice service; for example, when If the call duration in the current user behavior is 6 seconds, it can be judged that the current user behavior is an ultra-short call behavior, and there may be a problem of poor voice perception.

Step 203: analyzing the signaling of the current user behavior;

Specifically, collect signaling messages from interfaces related to the current user behavior on the network side, such as collecting signaling messages of the LTES1 interface, and after parsing the signaling messages, obtain the signaling of the current user behavior, and from The signaling features of the current user behavior are filtered out from the signaling of the current user behavior.

Step 204: Judging whether the signaling feature of the current user behavior satisfies the abnormal condition, if so, execute step 205, otherwise execute step 206;

Specifically, according to the established signaling model, analyze the signaling of the current user at each stage of the service use process, such as statistics of key indicators such as MOS value, delay, jitter, packet loss, and frame rate; determine the current Whether the signaling feature of the user behavior satisfies the signaling feature of the user's abnormal behavior in the established signaling model, if yes, go to step 205; if not, go to step 206.

Here, the signaling features of the abnormal user behavior contained in the signaling model include: the MOS value is lower than the first threshold value, and the packet loss rate is greater than the second threshold value; The value changes from high to low; the time interval jitter exceeds the third threshold; the time of no data packet transmission on the bearer link exceeds the fourth threshold; the threshold of eSRVCC switching times exceeds the threshold; RRC connection re-establishment; release; the video code rate is lower than the fifth threshold value; the video and voice data packets are not synchronized, etc.

Here, you can focus on the abnormal indicators that exceed the threshold value; the specific value of the threshold value can be defined according to the actual business situation.

Step 205: problem location and delimitation;

Specifically, according to the signaling characteristics of the current user behavior and the set quality of service problem corresponding to the abnormal user behavior, as shown in Table 1, determine the quality of service problem corresponding to the current user behavior; A signaling message to determine the management subject of the service quality defect of the current voice service.

Here, through the correlation analysis between the current user behavior and signaling messages, the service quality problems of the current user behavior are classified, and the specific problem types are determined, such as single communication, no sound, intermittent, word swallowing, noise, etc.

Here, in general, the signaling of the service is performed serially, and any failure of any step in the signaling will cause the subsequent signaling interaction to fail. Therefore, if the service fails, it is necessary to find the earliest failure point, that is, the first disconnection network element.

The first disconnection network element may find that the signaling message itself is abnormal or cannot process the signaling normally due to various reasons such as insufficient resources and signaling timeout. The first disconnection message sent by the first disconnection network element carries the The failure code will generally indicate the exact root cause of the failure.

Since VoLTE involves three domains: Packet Core Network (EPC, Evolved Packet Core), Internet Protocol Multimedia Subsystem (IMS, IP Multimedia Subsystem), Circuit Switched Domain/Circuit Switched Domain (CS, Circuit Switched Domain), there are many network elements and signaling It is complex, and its particularity requires signaling analysis to gradually narrow down the scope of analysis from large to small. Therefore, when defining and locating the problem, for a single service that is judged to have a quality problem after analyzing user behavior and signaling messages, further analyze the signaling process of this service and pay attention to the first disconnection: first analyze and issue the first disconnection The network element of the message belongs to the domain, and delimits the maintenance department of the hidden problem; re-analyzes the network element that sends the first disconnection message in the domain, divides the responsible subject of the hidden problem, and automatically obtains the performance indicators and alarm information within the time period attached to the network element; Finally, analyze the failure code of the first disconnection message, correlate performance indicators, equipment alarms and other information, and finally locate the root cause of hidden problems, so that maintenance personnel can quickly formulate corresponding solutions.

Step 206: Result output.

Specifically, according to the problem delineation and location results, a user-level service quality warning message is generated in real time, and the warning message includes information such as quality problem type, problem location result, failure code message, and reason, and is pushed to the monitoring Maintenance personnel.

Here, when the current user behavior is not an abnormal user behavior, or the current user behavior is an abnormal user behavior and the signaling characteristics of the current user behavior do not meet the abnormal conditions, the judgment result can be pushed to the monitoring and maintenance personnel to Make monitoring and maintenance personnel more aware of network conditions.

The service quality monitoring method provided in this embodiment can also generate early warning information in time before customer complaints for large-scale user service quality degradation, which is convenient for the maintenance department to carry out troubleshooting, the marketing department to actively care for customers, and the customer service department to provide targeted To appease customers by formulating a customer service explanation caliber and complaint interception channels; at the same time, from the perspective of user signaling, the service quality monitoring method provided in this embodiment provides a new way for the monitoring work to quickly discover hidden problems, and solves the problem of simple equipment alarms and The performance indicators cannot detect and locate hidden dangers of user service quality; provide automatic demarcation and location methods for hidden troubles for maintenance work, greatly improve the efficiency of problem location, and reduce the impact on user business perception; provide new solutions for optimization work, and promote Optimize the transformation from network-oriented to user-oriented; provide a more refined way of carrying out security work, refine the granularity of security, and realize real-time security of single-user, single-service awareness for VIP customers; provide more reliable information sources for customer service work, Complaint explanations and customer care can be carried out in a targeted manner.

In order to implement the above method, the embodiment of the present invention also provides a service quality monitoring device. FIG. 3 is a schematic diagram of the composition and structure of the service quality monitoring device according to the embodiment of the present invention. The device includes: a first processing unit 11 and a second processing unit 12 , the third processing unit 13 and the fourth processing unit 14; wherein,

The first processing unit 11 is configured to obtain service characteristics and signaling characteristics of the current voice service;

The second processing unit 12 is configured to perform a matching search in the service feature database of the abnormal voice service based on the service feature of the current voice service;

The third processing unit 13 is configured to, when the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, in the abnormal voice service The matching search is performed in the signaling signature database;

The fourth processing unit 14 is configured to match the signaling feature of the current voice service when the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database The quality of service defect corresponding to the signaling feature of the abnormal voice service is used as the quality of service defect of the current voice service.

The first processing unit 11 is specifically used for:

Collect the service data packets and signaling messages of the current voice service, and analyze the service data packets and the signaling messages respectively, obtain the service data and signaling of the current voice service, and obtain the service data and signaling messages from the current voice service The service characteristics of the current voice service are selected from the service data of the current voice service, and the signaling characteristics of the current voice service are selected from the signaling of the current voice service.

Here, the service data packets and signaling messages of the current voice service may be collected from an interface such as LTE S1 on the network side; the voice service may be services such as making calls and receiving calls.

The second processing unit 12 is specifically used for:

According to the service feature of the current voice service, search the service feature database of the abnormal voice service, and judge whether there is a service feature of the abnormal voice service matching the service feature of the current voice service in the service feature database, that is, determine the Whether the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, if it matches, it indicates that the current voice service is an abnormal voice service; if it does not match, it indicates that the current voice service is not Abnormal voice service, end analysis.

Here, the service feature library of the abnormal voice service includes the service feature of the abnormal voice service; the abnormal voice service can be ultra-short call, over-frequency call, early release and other services; the service feature of the abnormal voice service refers to Some specific features usually appear in abnormal voice services, for example, for an ultra-short call, the service feature of the ultra-short call may be the duration of the call.

The third processing unit 13 is specifically used for:

According to the matching search result of the second processing unit 12, when it is determined that the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, according to the information obtained by the first processing unit 11 Describe the signaling feature of the current voice service, search for the signaling feature library of the abnormal voice service, and judge whether there is a signaling feature of the abnormal voice service matching the signaling feature of the current voice service in the signaling feature library, That is, it is judged whether the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature library, and if they match, then inform the fourth processing unit 14 to continue processing; if they do not match, then end analyze.

Here, the signaling feature library of the abnormal voice service contains the signaling features of the abnormal voice service; the signaling feature of the abnormal voice service refers to some signaling features that usually appear in the abnormal voice service, such as for The ultra-short call, the signaling characteristics of the ultra-short call may be that the packet loss rate is greater than the threshold value, and the MOS value is low.

The fourth processing unit 14 is specifically used for:

According to the matching search result of the third processing unit 13, when it is determined that the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, it will be matched with the signaling feature of the current voice service. The quality of service defect corresponding to the signaling feature of the abnormal voice service that matches the feature is used as the quality of service defect of the current voice service.

Here, the quality of service defect corresponding to the signaling feature of the abnormal voice service can be stored in the signaling feature library, so that the defect corresponding to the signaling feature of the abnormal voice service can be quickly obtained according to the signaling feature of the abnormal voice service. Service quality defects; each abnormal voice service may have multiple service quality defects due to different signaling characteristics, that is, for any abnormal voice service, when the signaling characteristics of the abnormal voice service are different, the abnormal voice service The QoS defects corresponding to signaling features are also different.

Here, when the signaling feature of the current voice service matches the signaling feature of any abnormal voice service in the signaling feature library, it indicates that the signaling feature of the current voice service is consistent with the signaling feature Match the signaling features of abnormal voice services in the library.

Further, the device further includes: a fifth processing unit 15, configured to locate the management subject of the service quality defect of the current voice service, and output alarm information for the current voice service to the management subject.

Specifically, the fifth processing unit 15 determines and finds the network element whose signaling message of the current voice service is abnormal, and obtains the current The management subject of the service quality defect of the voice service, and output the alarm information for the current voice service to the management subject, so as to inform the management subject of the type and cause of the service quality defect of the current voice service.

Here, since the signaling interaction of the voice service is performed serially, any failure of any step will lead to the failure of the subsequent signaling interaction. Therefore, the first discovery of the current voice service information can be obtained according to the voice service signaling message. Find the network element whose message is abnormal, and then determine the management domain of the network element and the corresponding management subject.

further,

The first processing unit 11 is further configured to collect service data packets of at least one voice service in the communication network, analyze the collected service data packets of the at least one voice service, and obtain the service data packets of the at least one voice service. Service data, and filter out the service characteristics of the at least one voice service from the service data of the at least one voice service;

The second processing unit 12 is further configured to perform a matching search in the service feature library of the abnormal voice service based on the service feature of the at least one voice service;

The third processing unit 13 is further configured to store the service feature of the at least one voice service in the at least one voice service when the service feature of the at least one voice service matches the service feature of the abnormal voice service in the service feature database. The business feature database; performing cluster analysis on the business features in the business feature database, generating analysis results, and updating the business feature database according to the analysis results.

Specifically, the first processing unit 11 collects service data packets of one or more voice services in the communication network in advance, and after analyzing the service data packets of the one or more voice services, obtains the one or more The service data corresponding to the service data packet of the voice service, and then filter out the service characteristics of the one or more voice services from the service data; the second processing unit 12 is based on the one or more voice services. Service characteristics, searching the service characteristic library of the abnormal voice service, and judging whether the service characteristics of the one or more current voice services match the service characteristics of the abnormal voice service in the service characteristic library; the third processing unit 13 When the service features of the one or more voice services match the service features of the abnormal voice service in the service feature database, store the service features of the one or more voice services in the service feature database; then, the The third processing unit 13 performs self-study such as cluster analysis on the service features corresponding to the same abnormal voice service in the service feature database, so as to generate new service features of the same abnormal voice service, and realize the analysis of the service features Update processing of service characteristics of abnormal voice services in the library.

Here, the service features of the abnormal voice service in the service feature library can be set in advance, such as for any abnormal voice service, a service feature can be set respectively first, and then the service feature of the set abnormal voice service can be updated in subsequent processing; The third processing unit 13 may also store the service feature of the current voice service in the service feature database after the fourth processing unit 14 determines the quality of service defect of the current voice service, and update the service feature database. Described business feature database.

Here, it should be noted that, the signaling feature of the abnormal voice service in the signaling feature database may also be acquired according to the manner of acquiring the service feature of the abnormal voice service in the service feature database.

The service quality monitoring device provided in this embodiment can change from passively waiting for users to report service quality problems to proactively discovering and solving service quality problems, from focusing on equipment performance to considering end-to-end user experience, and from network element-centered optimization The model has changed to a user-centered optimization model to improve user satisfaction with service quality. Only when users feel that service quality meets or exceeds their expectations can they improve service satisfaction and online loyalty, and they are more willing to try New business, thus laying a solid foundation for the continuous development of operators.

In practical applications, the first processing unit 11, the second processing unit 12, the third processing unit 13, the fourth processing unit 14, and the fifth processing unit 15 all can be controlled by a central processing unit (CPU) or a microprocessor located at the terminal. It can be implemented by a digital signal processor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA).

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements and improvements made within the spirit and scope of the present invention are included in the protection scope of the present invention.

Claims (10)

1. A method for monitoring service quality, characterized in that the method comprises: Obtain the service characteristics and signaling characteristics of the current voice service; Based on the service feature of the current voice service, a matching search is performed in the service feature database of the abnormal voice service; When the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, performing a matching search in the signaling feature database of the abnormal voice service based on the signaling feature of the current voice service; When the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature library, the signaling feature of the abnormal voice service that matches the signaling feature of the current voice service corresponds to The service quality defect of is used as the service quality defect of the current voice service. 2. The method according to claim 1, wherein said acquiring the service characteristics and signaling characteristics of the current voice service comprises: Collecting service data packets and signaling messages of the current voice service, and analyzing the service data packets and the signaling messages respectively, obtaining the service data and signaling of the current voice service, and obtaining the service data and signaling messages from the service data Filter out the service characteristics of the current voice service, and filter out the signaling characteristics of the current voice service from the signaling. 3. The method according to claim 1, wherein, based on the service feature of the current voice service, before performing a matching search in the service feature library of the abnormal voice service, the method also includes: Collecting service data packets of at least one voice service in the communication network, and analyzing the collected service data packets of the at least one voice service, obtaining service data of the at least one voice service, and obtaining the service data from the at least one voice service Selecting the service characteristics of the at least one voice service from the service data; Based on the service feature of the at least one voice service, perform a matching search in the service feature database of the abnormal voice service; When the service feature of the at least one voice service matches the service feature of the abnormal voice service in the service feature database, storing the service feature of the at least one voice service in the service feature database; Perform cluster analysis on the business features in the business feature database, generate analysis results, and update the business feature database according to the analysis results. 4. The method according to claim 2, characterized in that the method further comprises: Locating the management subject of the service quality defect of the current voice service, and outputting alarm information for the current voice service to the management subject. 5. The method according to claim 4, characterized in that, the management subject for positioning the quality of service defect of the current voice service comprises: Determining and discovering the abnormal network element in the signaling message of the current voice service, and obtaining the management subject of the service quality defect of the current voice service according to the management domain to which the network element belongs and the management subject of each management domain. 6. A service quality monitoring device, characterized in that the device comprises: a first processing unit, a second processing unit, a third processing unit, and a fourth processing unit; wherein, The first processing unit is configured to obtain service characteristics and signaling characteristics of the current voice service; The second processing unit is configured to perform a matching search in the service feature library of the abnormal voice service based on the service feature of the current voice service; The third processing unit is configured to, when the service feature of the current voice service matches the service feature of the abnormal voice service in the service feature database, based on the signaling feature of the current voice service, in the abnormal voice service Matching lookup in the signaling signature database; The fourth processing unit is configured to, when the signaling feature of the current voice service matches the signaling feature of the abnormal voice service in the signaling feature database, match the signaling feature of the current voice service The quality of service defect corresponding to the signaling feature of the abnormal voice service is used as the quality of service defect of the current voice service. 7. The device according to claim 6, wherein the first processing unit is specifically configured to: Collecting service data packets and signaling messages of the current voice service, and analyzing the service data packets and the signaling messages respectively, obtaining the service data and signaling of the current voice service, and obtaining the service data and signaling messages from the service data Filter out the service characteristics of the current voice service, and filter out the signaling characteristics of the current voice service from the signaling. 8. The device of claim 6, wherein: The first processing unit is further configured to collect service data packets of at least one voice service in the communication network, analyze the collected service data packets of the at least one voice service, and obtain service data packets of the at least one voice service data, and filter out the service characteristics of the at least one voice service from the service data of the at least one voice service; The second processing unit is further configured to perform a matching search in the service feature library of the abnormal voice service based on the service feature of the at least one voice service; The third processing unit is further configured to store the service feature of the at least one voice service in the A business feature database; performing cluster analysis on the business features in the business feature database, generating analysis results, and updating the business feature database according to the analysis results. 9. The device according to claim 7, further comprising: a fifth processing unit, configured to locate the management subject of the quality of service defect of the current voice service, and output to the management subject the Alarm information of the current voice service. 10. The device according to claim 9, wherein the fifth processing unit is specifically configured to: Determining and discovering the abnormal network element in the signaling message of the current voice service, and obtaining the management subject of the service quality defect of the current voice service according to the management domain to which the network element belongs and the management subject of each management domain.