KR101338485B1 - Quality of each service management Method and system in total IP network - Google Patents

Abstract

The present invention relates to a quality control method and system for each service type in a large IP network. The present invention extracts a session between each service server and an end user from a probe of a backbone network, randomly samples the extracted session packet until the service utilization rate is obtained, and extracts the packet of the sampled session by region based on distribution information of an IP address. And classify it by the access service network and classify it again by service type, and then make the IP address of the classified packet into the IP list of interest for each measurement area and transmit it to the corresponding area, and thus located in the access network of each measurement area. The probe receives a list of interested IPs of a corresponding region and measures the quality of a packet having an IP address of the received list of interested IPs.

QA, Probe, Session, IP Address

Description

Quality control method and system for service type in large IP network {Quality of each service management Method and system in total IP network}

1 is a block diagram of a quality control system according to an embodiment of the present invention.

2 is a flowchart illustrating a quality control method for each service type in a large IP network according to an exemplary embodiment of the present invention.

3 is a block diagram illustrating a structure of an access network probe according to an exemplary embodiment of the present invention.

4 is a block diagram showing the structure of a backbone probe according to an embodiment of the present invention.

5 is a flowchart illustrating an operation of a backbone network probe generating an IP list of interest according to an embodiment of the present invention.

6 is a block diagram showing the structure of a quality analysis server according to an embodiment of the present invention.

7 is a diagram illustrating an IP address management structure managed by an IP address manager of a quality analysis server according to an exemplary embodiment of the present invention.

8 is a diagram illustrating a general RTP protocol header structure.

9 is a diagram illustrating a general SR type RTCP protocol header structure.

FIG. 10 is a diagram illustrating a general RR type RTCP protocol header structure. FIG.

The present invention relates to service quality management in an IP (Internet Protocol) network, and more particularly, to a method and system for enabling quality control for various types of services accommodated by a large IP network.

At present, with the development of high-speed Internet, network operators have been expanding large networks every year for stable Internet service, and recently MPLS (Multi Protocol Label) in addition to general Internet network for high-quality Internet service to be accommodated in Broadband Convergence Network (BCN). It is building and operating a quality based premium network based on switching.

In addition, network operators operate various types of operation servers such as Domain Name System (DNS), Dynamic Host Configuration Protocol (DHCP), and Authentication Authorization Accounting (AAA) server in the server farm for stable internet service for subscribers. In addition, service servers of many operators that provide services in IDC (Internet Data Center) are operating.

In order to provide more than a certain level of service to customers, network operators always monitor network status by operating multiple NMS (network management system) and SMS (server management system) for stable network operation and server operation. However, it is difficult and difficult to manage and manage numerous network resource elements and servers without failure due to the frequent appearance of new services. Managing the quality of service in terms of network operation is an important and difficult task.

In particular, various types of networks such as general IP network (eg ADSL, VDSL, FTTH, metro game, metro office, wireless LAN, etc.), premium IP network, virtual private network (VPN) network, Wibro network, BCN network, etc. Large IP network operators, which maintain networks across the country and region, operate numerous routers and switches and the ports, lines, etc. that are responsible for connecting them, and management servers (DNS, DHCP, AAA servers) to support them. Etc.) and the application server that stores and delivers the contents of these services, it is very important to provide stable services to customers by operating these devices and servers without any obstacles.

However, this requires a system that supports the operator's work by detecting the quality degradation by monitoring the quality of each service network type and region, and each network is widely distributed in the region and the vast information (ie, information by network and region). Because the quality of the data must be measured, there are many difficulties in terms of cost and data processing. As a result, no efficient network management system or method for large IP networks has been proposed.

An object of the present invention is to provide a quality control method and system for each service type in a large IP network that enables quality measurement by region, network, and service type in a large IP network.

According to a feature for achieving the above technical problem, the present invention comprises a first step of setting the measurement area and the service utilization rate; Extracting a session between each service server and an end user in a probe of a backbone network; Randomly sampling the extracted session packets until a service utilization rate is reached; A fourth step of classifying packets of the sampled sessions by region based on distribution information of IP addresses, classifying them by access service network, and classifying them again by service type; A fifth step of making an IP address of the classified packet into a target IP list for each measurement area and transmitting the same to a corresponding area; And a sixth step of receiving a list of interested IPs of a corresponding region from a probe located in an access network of each measuring region, and measuring a quality of a packet having an IP address of the received interested IP list. It provides quality control method by service type in network.

According to another feature for achieving the above technical problem, the present invention is installed in the backbone network to extract the session between each service server and the end user, and randomly sampled until the service utilization rate for the packet of the extracted session, At least one backbone network probe for classifying packets of sessions by region, access service network, and service by using distribution information of IP addresses, and then generating an IP list of interest for each measurement target region; One or more access network probes installed in an access network and extracting packets between the end user and the backbone network and performing quality measurements on packets corresponding to the IP list of interest provided by the backbone network probe among the extracted packets; And a quality analysis server connected to the backbone network probe and the access network probe and generating quality data for each region, access service network, and service for the quality measurement result measured by the access network probe. Provides a quality management system for each service type.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification, when a part is said to "include" a certain component, it means that it can further include other components, except to exclude other components unless otherwise stated. Also, the terms " part, "" module," " module, "and " block" refer to units that process at least one function or operation, Lt; / RTI >

Now with reference to the drawings will be described in detail with respect to the quality management method and system for each service type in a large IP network according to an embodiment of the present invention.

1 is a block diagram of a quality control system according to an embodiment of the present invention. As shown in FIG. 1, the quality management system according to the present invention includes at least one backbone network probe 200 installed in a backbone network, a plurality of access network probes 100 installed in an access network, and an access network probe 100. And a quality analysis server 300 connected to the backbone network probe 200.

The access network probe 100 is installed at an output (or input) of an arbitrary router located in a measurement area to be measured for quality among routers respectively connected to the access service network of each region. In FIG. 1, access network probes 100a and 100b are installed in measurement area A and measurement area B. The backbone network probe 200 is installed at an output (or input) of a router connected to IDC in which various service servers 20 are collected.

The access network probe 100 has a function of collecting packets coming and going between the end user 10 and the service server 20 and performing quality measurement on the collected packets.

The access service network to which the access network probe 100 is connected is one of a general IP network, a premium IP network, a VPN network, a WiBro network, a BcN network, and the like. Therefore, packets collected by the access network probe 100 have quality characteristics according to the type of the access service network. The service server 20 in the backbone network (that is, IDC) that the end user 10 accesses through the access service network depends on the type of service requested by the end user 10.

Therefore, the packet collected by the access network probe 100 may be a packet for an application service (a service such as a web or an e-mail that a subscriber accesses a server to receive traffic) or a management service (DNS, DHCP, etc.) to facilitate an application service. It may be a packet for a network operator prepared and operated as a backend for use, and may be a packet differentiated by service class (eg, gold, silver, bronze, best effort, etc.). have. As a result, the access network probe 100 installed in a specific region may basically collect packets classified by region, access service network type, and service type.

The packets collected by the access network probe 100 installed in each measurement area are indiscriminate. Therefore, if the quality measurement for all packets collected by the access network probe 100, it is necessary to process a large amount of data as in the conventional problem.

The backbone network probe 200 solves the problem of processing massive amounts of data in the access network probe 100.

That is, the backbone network probe 200 generates session information by collecting packets between each service server and a subscriber installed in the backbone network, and processes the session information based on IP address distribution information (that is, an IP address management scheme). A service object to be measured for quality, i.e., an IP list of interest, is selected, and the selected IP list of interest is provided to the access network probe 100. Accordingly, the access network probe 100 measures quality only for packets corresponding to the IP list of interest provided by the backbone network probe 200 (eg, RTT (Round Trip Time), delay, jitter, loss, etc.). ) And provides the quality measurement result to the quality analysis server 300.

 In general, IP addresses are allocated to a specific address band by KRNIC (Korea Network Information Center), registered and used, and assigned and managed according to the network design policy of the operator and the distribution policy for subscribers. In this case, the operator can efficiently manage and classify IP bands consistently according to organization information, facility information, and service policy. As a result, the current IP address has been allocated to the region and service consistently, so as to show the region, service type, access service network, etc. as shown in FIG.

On the other hand, the IP list of interest is classified by region and service, and the IP address of representative packets using a specific service for a specific region and the access service network and service type corresponding to each IP address can be known. Information such as source IP, destination IP source port, destination port, destination field and time information, and the like. The IP list of interest will be described in more detail with reference to FIGS. 4 and 5.

The quality analysis server 300 classifies the quality measurement result by region, service, and access service network type based on the IP address, and generates statistical data by region, service, and access service network type by day, week, and month. And the quality analysis server 300 provides a UI (User Interface) that the network administrator can check the generated statistical data.

Hereinafter, a quality management method for each service type in a large IP network according to an embodiment of the present invention will be described with reference to FIG. 2. 2 is a flowchart illustrating a quality control method for each service type in a large IP network according to an embodiment of the present invention, and is a flowchart according to a conceptual description to help understanding of the present invention.

As shown in FIG. 2, the backbone network probe 200 collects session information for accessing each service server (S110), classifies session information by region, access service network, and service type based on an IP address and randomly. By using the sampler, the utilization rate measured by the packets sampled using the random sampler for the packet of each service session is set to the set utilization rate K%. Here, using the random sampler to make the packet for each service at the set utilization rate is the same as sampling the circuit to be measured among the entire circuits because the entire circuit cannot be measured.

The backbone network probe 200 classifies the session information for each service classified and sampled by S120 by region based on IP address information to create an IP list of interest (S130), and then creates the IP list of interest in the corresponding region. Provided to the access network probe 100 located (S140).

When the access network probe 100 of a specific region receives the corresponding IP list of interest, the IP network is detected from the collected packets, and the packets included in the IP list of interest are extracted (S150 and S160).

Then, the access network probe 100 performs quality measurement on the extracted packet (S170) and provides the quality measurement result to the quality analysis server 300 (S180).

Hereinafter, the present invention will be described in more detail with reference to FIGS. 3 to 10.

First, the access network probe will be described with reference to FIG. 3. 3 is a block diagram illustrating a structure of an access network probe according to an exemplary embodiment of the present invention.

As shown in FIG. 3, the access network probe 100 includes a reception buffer 110, a session manager 120, a quality computation processor 130, an interworking manager 140, a primary statistics processor 150, and an IP address. And a table 160 and a config manager 170.

The reception buffer 100 temporarily stores packets collected from a network information center (NIC) on a physical line (that is, a gigabit line), and the session manager 120 stores the source IP, the destination IP, and the source port of the collected packets. Session information is generated based on the destination port, protocol ID, and so on. At this time, the session manager 120 discards the packet that is not the IP address of the subscriber or server of the own station.

The quality calculation processor 130 filters the packets for the session generated by the session manager 120 based on the IP list of interest received from the backbone network probe 200, and extracts only packets corresponding to the IP list of interest. The extracted packets are subjected to quality measurements such as delay, jitter, loss, and RTT.

Here, the filtering based on the IP list of interest is based on the source IP, destination IP, source port, destination port, destination field, and time information of the packet for the session generated by the session manager 120. Extract only packets that match.

The quality calculation processing unit 130 extracts only packets corresponding to the IP list of interest from the collected packets, and performs quality measurement on the extracted packets. This is performed by classifying the quality measurement for the streaming service and the streaming service (VoD, AoD, Voip, IPTV, etc.).

Quality measurements for general services use the packets collected in the usual way to measure the availability and RTT, delay, jitter, and loss that occur between the access network and the backbone network. In this case, the quality operation processor 130 measures delay variation, that is, jitter, on the upstream packet based on the source IP.

As an example of quality measurement for a general service, the quality operation processor 130 records sender information, system time, and packet number to create a measurement packet at regular intervals and transmits the measurement packet to the multicast service server. Then, in response, the service server receives the measurement packet, compares the number of the previous packet with the number of the current packet, obtains a loss, and retransmits it to the quality processor 130. The quality processor 130 receives the retransmitted packet and provides a current time. RTT, jitter, delay and loss are compared by comparing the time stamp value with.

Quality measurement for the streaming service in the quality calculation processing unit 130 is delivered to the end by using a Real Time Transport Protocol (RTP), RTCP (RTP control Protocol) packets generated from the subscriber among the packets collected by the conventional method For delay, jitter, loss, and voip, measure call completion rate.

Here, a method of measuring quality using RTP and RTCP will be briefly described with reference to FIGS. 8 to 9.

FIG. 8 is a diagram illustrating a general RTP protocol header structure, FIG. 9 is a diagram illustrating a general SR type RTCP protocol header structure, and FIG. 10 is a diagram of a general RR type RTCP protocol header structure.

As shown in FIG. 8, a header of a packet according to a general RTP protocol includes a time stamp, payload, SSRC, and CSRC information. The delay is identified using a time stamp, and lost through a payload. Can be identified. And, as shown in Figure 9, the header of the packet according to the SR type RTCP protocol has a reception report block and a report block, by analyzing the reception report block can determine the loss and jitter value, LSR (last SR ) And delay since last SR (DSLR) to determine delay and RTT.

In addition, as shown in FIG. 10, the packet header according to the RR type RTCP protocol has a report block, and the delay and the RTT can be determined using the last SR (LSR) and delay since last SR (DSLR) of the report block. have.

Here, the quality calculation processor 130 may measure the quality by capturing only the RTCP packets without capturing the RTP packets in order to reduce the load on the system.

The interworking management unit 140 stores the IP list of interest received from the backbone network probe 200 and time information for each item of the IP list for each service, performs data transmission / reception communication with the backbone network probe 200, and performs quality calculation. The quality measurement result measured by the processor 130 is stored, the quality measurement result is transmitted to the quality analysis server 300, and the IP analysis table 300 receives the IP table corresponding to the access network probe.

The primary statistics processing unit 150 does not transmit the quality measurement results every time to the quality analysis server 300 as it is, and makes primary statistics in synchronization with the statistics processing cycle stored in the config management unit 170, the quality analysis server 300 To pass on. The IP address table 160 stores the IP table in which information about the entire IP address band managed by the corresponding node from the interlocking management unit 140 is recorded. In this case, the IP table is used as data that can know whether or not the IP address of the subscriber or service server of the host that the reception buffer 110 performs.

The config manager 170 receives and manages Top N, measurement cycle management information, and statistical processing cycle information from the interworking management unit of the quality analysis server 300. Here, the measurement cycle is a cycle of the quality measurement operation in the interlocking management unit 140.

Hereinafter, a backbone network probe according to an exemplary embodiment of the present invention will be described with reference to FIG. 4. 4 is a block diagram showing the structure of a backbone probe according to an embodiment of the present invention.

As shown in FIG. 4, the backbone network probe 200 may include a reception buffer 210, a session manager 220, a response manager 230, an interworking manager 240, an analysis target processor 250, and an IP address table ( 260 and a config manager 270.

The reception buffer 210 temporarily stores packets collected at a network information center (NIC) on a physical line (ie, a gigabit line), and temporarily stores a portion of a payload of each packet.

The session manager 220 generates session information based on the source IP, the destination IP, the source port, the destination port, and the protocol ID of the collected packet.

The response manager 230 analyzes the payload of the reception buffer 210 from the service session information generated by the session manager 200 and processes the session failure list.

The analysis target processor 250 generates an IP list of interest by using the remaining session information except the session failed session among the session information generated by the session manager 220. In this case, the analysis target processor 250 samples the packet information of the session stored as a result of the processing of the session manager 220 using a random sampler, and generates an IP list of interest based on the sampled packets.

The analysis target processing unit 250 uses the IP address information managed by the IP address management unit (see FIG. 6) of the quality analysis server 300 to generate an IP list of interest, and belongs to the organization through the IP address of each session. The access network probe 100 located in the information (that is, the region) and the institution information is identified. If the access network probe identified through this confirmation is an access network probe of a quality measurement area set according to the policy, a list of IPs of interest to be provided to the access network probe of the quality measurement area is collected. The analysis target processor 250 uses the IP address table for all IP addresses provided by the IP address management unit (see FIG. 9) of the quality analysis server 300 to collect the collected IP list of the access service network type and application service type. It classifies by management service type and processes IP list of interest by using logical port (source port and destination port) and packet size. Here, the list of interest IPs processed by the analysis object processing unit 250 is provided to the access network probe 100 located in the corresponding area according to the classified area information.

The interworking manager 240 transmits the IP list of interest processed by the analysis target processor 250 for each access network probe, and transmits the IP list and the session failure IP list that may cause the overload to the quality analysis server 300. To pass.

The IP address table 260 receives and stores the IP table in which information about the entire IP address band managed by the corresponding node is recorded from the quality analysis server 300.

The config manager 270 receives and manages the TOP N, various thresholds, measurement cycle management information, and the like, which should be set in the analysis target processor 250, from the quality analysis server 300.

Herein, a process of generating an IP list of interest in the backbone network probe will be described with reference to FIG. 5. 5 is a flowchart illustrating an operation of a backbone network probe generating an IP list of interest according to an embodiment of the present invention.

As shown in FIG. 5, the backbone network probe 200 has a measurement area to be measured for quality and a service utilization rate which is a selection criterion for packets to be measured.

The session manager 220 of the backbone network probe 200 generates session information by using the collected packets (S220), and the analysis target processor 250 uses a random sampler to process packets for each service of the session using K%. The packet is extracted to be (S230).

The analysis target processor 250 selects a source IP address of the extracted packet as a target IP that may be a list of target IPs, identifies a destination IP in the packet of the target IP, and identifies a service server connected to the session by an end user. Therefore, target IP packet is classified by service type.

By the way, although the packet of each target IP (henceforth "target packet") was classified into the same service with the IP information of the service server (the destination IP in the case of an uplink), the traffic which has a different characteristic actually exists. For example, using port 80 on the web, downloading and playing web surfing and movie files is different.

In order to reflect these characteristics, the analysis target processor 250 classifies packets classified as the same service by logical ports, and classifies packets classified by service and logical ports by packet size (S240).

For example, among the packets using the same logical port, there are packets having different types of use. To filter out the packets, the packets are classified according to the packet size distribution. For example, IP packet size is classified into 1 ~ 100 bytes, 100 ~ 300 bytes, 300 ~ 500 bytes, 500 ~ 1000 bytes, 1000 bytes or more.

In the process of S240, most of the service classification and the logical port is divided into 1: N, the analysis target processing unit 250 finds a packet not applicable to this and excludes it from the target packet (S250).

Next, the analysis target processor 250 classifies the target packets classified by service, logical port, and packet size according to region information, and divides packets by service, logic, and packet size corresponding to the classified region. Create an IP list of interests using the IPs, and the interworking manager 240 grasps the access network probe 100 corresponding to each region (S260).

Then, the interlocking management unit 240 provides the region of interest IP list for each region generated by the analysis object processing unit 250 to the access network probe 100 of the corresponding region (S270).

On the other hand, when the backbone network probe 100 occupies more than a certain threshold value of the source IP, which is a form of abnormal use or the possibility of the impact of worms / viruses exist, the backbone probe 100 is notified to the quality analysis server 300 to generate an alarm. The analysis of these packets is useful information for the management of the service server, so it can be monitored later by leaving a log.

Hereinafter, a quality analysis server 300 according to an embodiment of the present invention will be described with reference to FIG. 6. 6 is a block diagram showing the structure of a quality analysis server according to an embodiment of the present invention.

As shown in FIG. 6, the quality analysis server 300 includes an IP address manager 310, a raw data collector 320, a backup manager 330, an interworking manager 340, an alarm processor 350, and a statistics processor. And a config manager 370.

The IP address management unit 310 manages the IP address, and reprocesses the IP address information that is used by all the NMSs and network operators and stores them in the DB 380. Here, the structure of the IP address reprocessed by the IP address management unit 310 is shown in FIG. As shown in FIG. 7, the IP address information managed by the IP address management unit 310 may include a substantial IP address, an access service network used by the IP address, a service class, an application service type, an affiliated organization (an organization that provides IP), Affiliation department, group number, access service name, subscriber accepting router name, IP of subscriber accepting router, and the like. Therefore, when looking at the IP address structure shown in FIG.

The raw data collection unit 320 periodically communicates with each access network probe to receive a result of quality measurement and store the received data in the DB 380. The backup manager 330 is responsible for managing the storage cycle of raw data, backup of statistical data, reports, logs, and the like.

The interworking manager 340 is in charge of interworking with the plurality of access network probes 100 and the backbone network probe 200 that perform quality measurements.

The alarm processor 350 registers and manages the conditions registered by the operator, and generates an alarm to notify the operator when an event corresponding to these conditions occurs. An example of an alarm condition is the condition "If source ip is not 211. *, 220. *, 61. *", which is an alarm in the event of a packet containing an IP that has been assigned and not assigned to the subscriber. An alarm condition that notifies the operator.

In addition, an alarm condition when an abnormally large number of packets are generated for a session registered to a service server is “The packet num of port num (100, 200, 300)> 10000 in a AA min”. In addition, the condition for abnormally large packet inflow to the port used by the known worm / virus based on the source IP is "If service code (a) & service grade (a) & network delay (over aaa ms) & quality. Lowering holding time (AA or more) ".

The statistical processor 360 generates statistical data for each classification of the raw data stored in the DB 380. In this case, the statistical data is divided by access service network, service, and region, and is generated daily, weekly, and monthly.

The config manager 370 stores and distributes the quality measurement cycle and the threshold of the entire system.

The DB 280 stores management information, statistical information, logs, and the like, and the web server 390 allows the operator to access the browser to view the results, manage access rights in the system administrator and user modes, and graph Generate tabular reports and download them to Excel.

The embodiments of the present invention are not limited to the above-described apparatuses and / or methods, but may be implemented through a program for realizing functions corresponding to the configuration of the embodiment of the present invention, a recording medium on which the program is recorded And such an embodiment can be easily implemented by those skilled in the art from the description of the embodiments described above.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

According to an embodiment of the present invention, the present invention analyzes the quality in detail according to the nature of the traffic circulated on each line, and can identify and identify the quality appearing by region using the IP address management module, management services and applications The service can be monitored periodically for each access service network to manage network quality and service quality of the entire large IP network.

In addition, the present invention can be monitored at a glance by the service network by service network for the national network at the same time by event and urbanization that the quality degradation occurs.

Claims (12)

A first step of setting a measurement area and a service usage rate; Extracting a session between each service server and an end user in a probe of a backbone network; Randomly sampling the extracted session packets until a service utilization rate is reached; A fourth step of classifying packets of the sampled sessions by region based on distribution information of IP addresses, classifying them by access service network, and classifying them again by service type; A fifth step of making an IP address of the classified packet into a target IP list for each measurement area and transmitting the same to a corresponding area; And A probe located in an access network of each measurement region receives a list of interested IPs of a corresponding region, collects packets transmitted between the end user and the backbone network, and collects IP addresses of the interested IP lists among the collected packets. And a sixth step of measuring the quality of the received packet, Quality control method by service type in large IP network. The method of claim 1, And a seventh step of receiving the quality measurement result measured in the sixth step and generating statistical data for each classified information-region, access service network, and service type-. . 3. The method of claim 2, Through the statistical data, the quality management method for each service type in a large IP network further comprises an eighth step of generating an alarm by detecting that an abnormal packet has occurred. 4. The method according to any one of claims 1 to 3, After the fourth step, And a ninth step of dividing the packet classified in the fourth step by logical port and classifying the packet again by packet size. 5. The method of claim 4, The third step, Quality management method by service type in large IP network that collects packets and generates session information with collected packets. 5. The method of claim 4, If the packet collected in the sixth step is a packet of a general service, quality measurement is performed using the packet of the general service, and if the packet collected in the sixth step is a packet for a streaming service, at least RTP or PTCP is performed. Quality control method by service type in large IP network that performs quality measurement by using The method according to claim 6, After the third step, In the backbone network probe, the quality management method for each service type in a large IP network further includes a tenth step of notifying an IP list that may cause an overload or an IP list that may cause an overload by measuring the quality of the extracted session. It is installed in the backbone network and extracts sessions between each service server and end user, randomly samples the extracted session packets until the service utilization rate is reached, and accesses the sampled session packets by region and access using IP address distribution information. At least one backbone network probe for classifying by service network and service, and generating an IP list of interest for each measurement target region; One or more access network probes installed in an access network to collect packets transmitted between the end user and the backbone network, and perform quality measurement on packets corresponding to the IP list of interest provided by the backbone network probe among the collected packets; And A service in a large IP network connected to the backbone network probe and the access network probe and including a quality analysis server generating statistical data for each region, access service network, and service for the quality measurement result measured by the access network probe. Quality management system by type. 9. The method of claim 8, The quality analysis server, Quality management system for each service type in a large IP network that generates an alarm when an abnormal packet is detected through the statistical data. 9. The method of claim 8, The backbone network probe, Quality management system for each service type in a large IP network that divides packets classified by service into logical ports and classifies them again by packet size to generate an IP list of interest. The method of claim 10, The access network probe, When the packet collected by the access network probe is a packet of a general service, quality measurement is performed using the packet of the general service, and when the packet collected by the access network probe is a packet for a streaming service, at least RTP or Quality management system for each service type in large IP network that performs quality measurement using PTCP. 12. The method of claim 11, The backbone network probe measures the quality of the extracted session and informs the quality analysis server of a list of failed IPs or an IP list that may cause an overload to the quality analysis server.

Images