KR20000061000A - Method for intelligent observation of exchange network service - Google Patents

Abstract

PURPOSE: A method for checking an intelligent type service of a switching network is provided for judging whether there is an error in a service based on a route-based characteristic of a switch station in accordance with a result of a comparison and analysis with a performance checking traffic data. CONSTITUTION: A performance data which occurs based on the current state of a switch station is gathered at one hour period and is accumulated for a certain time, for example, 8 weeks. An intermediate value and average value are computed based on the accumulated performance data(S300). A knowledge based data base is periodically updated based on an event occurrence circumference such as a date-based/time-based holiday and maintenance occurrence circumference and route shape information change data of the switch station.

Description

Method for intelligent observation of exchange network service

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile communication system, and in particular, after accumulating and accumulating performance data on a call processing occurring in an exchange network of an integrated network management system for a certain period of time, the accumulated data is constructed and then a knowledge base is established. On the basis of this, the present invention relates to an intelligent service monitoring method for an exchange network that is suitable for judging whether there is an abnormality of a service considering the characteristics of each exchange route based on a result of comparing and analyzing the performance monitoring traffic data collected in real time for each route.

In general, Integrated Network Management System (INMS) provides real-time fault and performance monitoring to maintain communication service, provides statistical data for network quality analysis and management, ticket generation and management by country / consignee, and wire / wireless network design. The main function is to provide statistical data for.

1 is a block diagram showing the basic configuration of the integrated network management system according to the prior art.

Referring to Fig. 1, the exchange offices 50a to 50n of each region composed of a mobile switching center (MSC), a home location register (HLR), a base station management apparatus (BSM), and other network equipment, and the exchange stations of each region. A management center (OMC) 40a to 40n that manages 50a to 50n, and a network management system (NMC) to receive failure information data according to the failure information data through the optical fiber distributed data interface 20 when a failure occurs at the switching center. 10).

The integrated network management system collects data on fault information generated in the exchange or measured data of call processing performance of the exchange every 5 minutes to determine whether there is an abnormality of the exchange based on an absolute reference value.

At this time, the operating maintenance center (OMC) 40a to 40n of each region outputs the failure information data as a system console message or manages the operation so that the failure information data according to the switching center can be managed by the region itself. The network management system (NMS) 10 is notified via the network dedicated line 30.

Then, the network management system 10 notified of the failure information data through the leased line 30 collects the failure information data of each region and recognizes that a failure occurs in the switching center.

In addition, the network management system 10 collects the performance data of the 5-minute cycle for the exchange call processing, and whether there is an abnormality in the service of the exchange through the call traffic rate and the completion rate, which are absolute values of the exchange stations for each route according to the analysis result. Monitor in real time.

The integrated network management system monitors the failure occurring in the communication network in real time, and then informs the user where the failure occurred is a communication line, a communication device, or a computer.

Meanwhile, the integrated network management system additionally includes a network operation function such as dynamically changing a network configuration or measuring traffic volume as well as the above-described monitoring function.

But. Such a conventional integrated network management system may not detect 100% of the service status of a switching center for each route even with collected failure information data such as an unmonitorable state due to a power failure or a power failure.

That is, in the conventional network management system, if the failure information data is not notified to the network management system from the exchange station or if there is no number of attempts for the 5-minute monitoring performance data, the conventional network management system checks whether there is an error in the call service of the exchange station caused by an external environmental factor other than its own network. It is impossible to judge quickly and accurately.

In addition, the conventional network management system does not manage the call number, traffic rate, completion rate and traffic, which are various call patterns for each route, and thus cannot quickly determine whether or not there is a service abnormality of the switching center.

In other words, the performance data monitoring function of the integrated network management system judges the service status of the switching center as the completion rate of call success of the absolute reference value, and it is difficult to fully monitor the current status of the switching center for each route.

Therefore, alarming about absolute communication rate and completion rate for a short time period in case of very low number of attempts or traffic causes confusion in actual service monitoring, resulting in deterioration of overall network service quality and consequent customer dissatisfaction. There is this.

The object of the present invention was devised in view of the above-mentioned problems of the prior art, and after the knowledge-based database of the types of calls generated at the exchanges was compared, the service of the exchanges was compared with the actual call situation of each exchange route. The purpose of the present invention is to provide an intelligent service monitoring method for an exchange network that can quickly and accurately determine the presence of an abnormality.

According to an aspect of the present invention for achieving the above object, the intelligent service monitoring method for the switching network accumulates a database of performance data generated according to the operating state of the switching center for a predetermined period of time, and by the root of the switching center Setting a monitoring class and a threshold, comparing performance data generated according to the current operating state of the switching center and the database, and according to the comparison result, presenting the current state of the switching center according to the set threshold. An alarm is generated to notify the operator.

Preferably, the database of the performance data may include classifying performance data generated during a certain time period of the switching center by day, time zone, and route, and classifying the classified performance data according to various event conditions of the switching center. Accumulating and calculating average data excluding abnormal performance data from the accumulated performance data to make a database.

In addition, accumulation of the performance data is made according to the holidays, maintenance occurrence situation and the event occurrence state of the route shape information change data for each day / time zone of the switching center, the threshold is based on the number of attempts collected in real time by the location of the switching center Will change accordingly.

In addition, the current performance data of the switching center is compared with the modified knowledge-based analysis data by sampling the databased performance data at regular intervals.

In addition, the alarm is generated by the operator terminal according to the alarm of the class according to the current state of the switching center, and if the alarm is repeatedly generated for a predetermined time to raise the alarm level.

In addition, the alarm is generated by being divided into an overload alarm, a traffic rate / completion rate alarm, an uncollected alarm based on the threshold set by the operator, a buzzer sound or as a result of comparing the monitoring data and knowledge-based analysis data with each other Outputs a graph according to

In addition, after the alarm is generated, the information according to the alarm is stored in the history management table, and the stored information is re-database to adjust the alarm level and threshold by itself.

1 is a block diagram showing the basic configuration of the integrated network management system according to the prior art.

2 is a functional block diagram of an integrated network management system according to the present invention.

3 is a flowchart illustrating an intelligent service monitoring method for an intelligent switching network according to the present invention;

4 is a graph illustrating an embodiment of sampling knowledge-based data in an intelligent service monitoring method for a switching network according to the present invention;

5 is a diagram illustrating an embodiment of displaying alarm information on an operator terminal in an intelligent service monitoring method for a switching network according to the present invention;

* Description of the symbols for the main parts of the drawings *

200: shared memory

210: Knowledge Base

220: exchange office

230: operator terminal

Hereinafter, a configuration and an operation according to an exemplary embodiment of the present invention will be described with reference to the accompanying drawings.

2 is a functional block diagram of an integrated network management system according to the present invention.

Referring to FIG. 2, a knowledge base database 200 which is constructed by accumulating performance data of the switching center 300 for a predetermined period of time, a shared memory 100 for storing data for network management, and an operator terminal 400. It consists of.

The knowledge base 200 calculates the median value and the mean value from the performance data collected periodically and collected for a certain period of time at the switching center 300, and then the holidays, maintenance by day and time zone of the switching center in the calculated median value. It reflects the event occurrence situation such as occurrence situation, route shape information change data, etc. and automatically becomes database based on this.

In this case, the median value and the average value are calculated except for data having a large number of attempted calls, which are performance data in abnormal cases, or data having a particularly low traffic rate.

The shared memory 100 stores the knowledge base analysis data 110, the event information management data 120, the monitoring data 130, and the alarm information data 140.

In this case, the knowledge-based analysis data 110 is data obtained by sampling the performance data of the knowledge-based database 200 at regular intervals in order to compare the monitoring data 130 which is periodically generated, and the event information management data ( 120 is data for registering the event occurrence history and the exchange shape information change for each route of the switching center, and the monitoring data 130 is performance data collected in real time at the switching center 300 every 5 minutes.

The comparative analysis function of the integrated network management system compares and analyzes the knowledge base analysis data 110, the event information management data 120, and the monitoring data 130 stored in the shared memory 100 to the threshold set by the operator. According to the grade.

At this time, according to the result of the comparative analysis, alarm information data 140 for generating alarm information is generated and stored in the shared memory 100.

Then, the alarm notification function according to the alarm information data 140 notifies the operator of the alarm information generation result.

Meanwhile, the history management function compares the monitoring data and the knowledge-based analysis data of the switching center 300 collected at regular intervals when the system operator requests a history matter for a specific switching center, and compares the operator terminal 400 with the comparison result. To the operator through the graph.

3 is a flowchart illustrating an intelligent service monitoring method for an intelligent switching network according to the present invention.

Referring to Figure 3, after collecting the performance data generated in accordance with the current state of the exchange in a period of time and accumulates for a certain period (for example, eight weeks). The median value and the average value are calculated from the accumulated performance data (S300).

The knowledge base is automatically built periodically by reflecting the median and mean values of events such as holidays, maintenance occurrences, and route shape information change data for each day / time zone of the switching center (S310).

Subsequently, the operator sets a monitoring level and a threshold for each route of the switching center (S320).

At this time, the threshold setting of the monitoring data alarm judgment is the width of the threshold value according to the number of attempts by location (that is, by exchange, IN / OUT / BOTH CLASS, route) based on the knowledge base.

In other words, the threshold is relatively changed according to the characteristics of the exchanges located in each region.

Then, the knowledge-based database stores the knowledge base data of the monitoring class and the threshold value such as call characteristics by day of the exchange, time-specific call characteristics, and traffic rate / completion rate for generated calls.

Subsequently, the knowledge-based database samples the stored knowledge-based data at predetermined intervals (for example, 5 minute periods) and converts the knowledge-based data into knowledge-based analysis data that is 5 minutes of data (S330).

This is to compare with the monitoring data actually collected every 5 minutes.

At this time, the sampling method proportionally subdivides the event of the knowledge-based data (that is, the number of attempts, the number of traffic, the number of completions) to follow the time trend in a predetermined time interval.

The following describes the formula for calculating the average number of events of sampling data.

MinInterval is the real-time monitoring data reporting period, Nmonhour means the hour during reporting time of real-time monitoring data, and nMonmin means the minute of reporting time.

NmaxDispHour is the time representation of the maximum monitoring interval.

Maxhour means 24, the maximum time of day.

Sampling_count = 60 / MinInterval, which means the number of subdivision sections of sampling data per hour.

Then, Nmin can be calculated | required as following (Equation 1).

Nmin = (nMonhour% nMaxDisphour) * sampling_count + nMonmin / MinInterval

Then, the sampling data of the knowledge base data is calculated by the following sampling method.

If ((nmin% sampling_count <6) {

Index = nmin% (sampling_count) + 6

Ae (nmin) =

H (nMonhour) / sampling_count + index * ((H ((nMonhour + 1 + MAXHOUR)% MAXHOUR) -H (nMonhour)) / sampling_count / sampling_count)

} Else {

index = nmin% (sampoing_count)-6

Ae (nmin) =

H (nMonhour + 1) / sampling_count + index * ((H ((nMonhour + 2 + MAXHOUR)% MAXHOUR) -H (nMonhour)) / sampling_count / sampling_count)

}

Then, the knowledge-based analysis data and the monitoring data is compared and analyzed with each other (S340), and determines whether there is a service abnormality of the switching center (S350).

This compares and analyzes Me (Nmin), which is the number of monitoring events of monitoring data calculated by the sampling method, and Ae (Nmin), which is sampling data of knowledge-based analysis data.

As a result of the determination, if an abnormality occurs in the service, the alarm information for the switching center having the abnormality is displayed through the operator terminal (S360).

That is, the performance monitoring data collected periodically is compared / analyzed with the knowledge base to generate traffic call congestion alarms for attempted calls or alarms for call rate / completion rate.

At this time, the alarm is generated by being divided into an overload alarm, a traffic rate / completion rate alarm and an uncollected alarm for the attempted number.

The trial lake alarm is when the threshold is out of the threshold compared with the trial number collected in real time based on the knowledge base analysis data. The traffic rate / completion rate alarm is based on the critical value set by the operator. It is divided into "Minor" part and "Normal" part, and when alarm occurs continuously, it is generated by increasing the alarm grade by weighting.

Therefore, the sensitivity of alarm occurrence varies according to the number of attempts.

At this time, in order to improve the accuracy of the alarm judgment, the operator's operator's terminal is compared to the operator's terminal by comparing the monitoring data and the knowledge analysis data collected for a certain period of time (5 minutes interval) before the alarm occurs at the operator's request. In order to make a quick judgment, the graph shows the comparison result such as traffic rate, completion rate, traffic and alarm status.

Since the operator terminal is continuously connected to the performance server system, an alarm is generated through a real-time monitoring screen (GUI screen).

When an alarm occurs, alarm information (ie alarm type, alarm class, location, occurrence time, end time) is stored in the history management table so that service status can be analyzed even after alarm termination.

On the other hand, real-time monitoring data is backed up for a certain time to provide a simulation function to virtually reproduce the service failure situation.

4 is a graph illustrating an embodiment of sampling knowledge-based data in an intelligent service monitoring method for an exchange network according to the present invention.

Referring to FIG. 4, the horizontal axis represents a time axis and accumulates knowledge-based data through monitoring of an exchange office for three hours, and the vertical axis represents a call effective number and represents an average number of events of the knowledge-based database.

Then, if the knowledge-based data is sampled for one hour, a sampling period 400 (for example, 12) of a predetermined period is made, and thus the number of events per hour 410 of the knowledge-based database is calculated.

Therefore, the knowledge-based data sampled every 5 minutes according to the above-described sampling method draws a curve as shown in the figure.

5 is a diagram illustrating an embodiment of displaying alarm information on an operator terminal in an intelligent service monitoring method for a switching network according to the present invention.

Referring to Figure 5, the operator sets the monitoring level and the threshold to generate an alarm by the class, and if a problem occurs repeatedly for a certain time to raise the alarm level.

That is, when the number of attempts is less than or equal to the threshold, an alarm is generated only when the number of attempted time thresholds for accumulating data of the previous time slot is reached.

At this time, the alarm is a "Critical" part 540, "Major" part 530, "Minor" part 520, "Normal" part 500, "Attempt" part 510 and " missed "portion 550.

These alarms are generated by exchanges, by INCOMING / OUTGOING / BOTHWAY CLASS, and by route.

The items displayed in the graph are location, route name, number of attempts, traffic rate, completion rate, traffic, average usage time, overload rate, number of available trunks, number of installed trunks, alarm type, alarm duration, and number of alarm occurrences. And so on.

The present invention as described above has the following effects.

First, the service monitoring function is reinforced by complementing the monitoring function of the switching center of the integrated network management system and by detecting and recovering service failures caused by factors outside the local network in the shortest time.

Second, by providing an auxiliary tool for determining the failure of the route of the service switching center, the efficiency of network monitoring can be improved, and the service quality is improved by minimizing the service down time due to system failure, and the profitability is increased by increasing the volume of calls.

Third, it is possible to quickly determine the situation of the root failure of the switching center and the failure of the switching center to provide information to the large power to maintain a smooth communication service.

Claims (9)

Accumulating the performance data generated according to the operation state of the switching center for a predetermined period of time and making a database; Setting a monitoring level and a threshold for each route of the switching center; Comparing the database with performance data generated according to a current operating state of the switching center; And informing the operator by generating an alarm according to the set threshold value according to the comparison result. The method of claim 1, wherein the step of making the database, Classifying performance data generated during a predetermined time period of the switching center by day, time zone, and route; Accumulating the classified performance data for a period of time according to various event conditions of the switching center; And calculating average data excluding the abnormal performance data from the accumulated performance data to make a database. The method of claim 2, wherein the accumulating the performance data comprises: Intelligent service monitoring method for a switching network, characterized in that according to the day of the week / time zone of the switching center according to the holidays, maintenance occurrence situation and event occurrence state of the route shape information change data. The intelligent service monitoring method according to claim 1, wherein the threshold is changed according to the number of attempts collected in real time for each location of the switching center. The method of claim 1, wherein the comparing step, Intelligent service monitoring method for a switching network, characterized in that the database-based performance data is sampled at regular intervals and transformed into knowledge-based analysis data, and the modified knowledge-based analysis data and performance data generated at the current switching center are compared with each other. . The method of claim 1, wherein generating the alarm comprises: Intelligent service monitoring method for a switching network, characterized in that the alarm for each class according to the current state of the switching center is generated through an operator terminal, and if the alarm is repeatedly generated for a predetermined time, the alarm level is raised. The intelligent service monitoring method of claim 6, wherein the alarm is generated by being divided into an overload alarm, a traffic rate / completion rate alarm, and an uncollected alarm based on a threshold set by the operator. The intelligent service monitoring method of claim 6, wherein the alarm outputs a graph based on a result of a buzzing sound or a comparison between the monitoring data and the knowledge-based analysis data. The method of claim 6, wherein after the alarm occurs, The information according to the alarm is stored in a history management table, intelligent service monitoring method for the switching network, characterized in that to automatically control the alarm level and threshold by re-database the stored information.