JP2002271392A - Method of controlling voice quality every call in ip net - Google Patents

Abstract

PROBLEM TO BE SOLVED: To monitor the voice quality at every call from a remote place, measure it, without installing an external measuring instrument, suppress the cost and acquire end-to-end delay information. SOLUTION: The method comprises a step for detecting the quality deterioration in nearly real time, analyzing and indicating a section suspicious of quality deterioration, controlling to avoid the quality deterioration in an OSS, collating quality deterioration information taken from a plurality of GW with network topology information the OSS has, indicating the information to specify a quality deterioration suspicious section for enabling the cut off of a failure/congestion section, and making the control of detour, etc., for the specified section, etc., to early recover the service.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a telephone network and an IP (I
A plurality of GWs (Gateways) that join the Internet Protocol network and an OSS (Operation Su) that monitors and controls voice quality for each call of the IP network.
The present invention relates to a voice quality management method in a "port system".

[0002]

2. Description of the Related Art In a conventional IP network, information such as the number of transmitted / received packets and the number of lost packets for each physical port such as an IP router and a gateway is collected at regular intervals.
There was no concept of a call, and the quality of each call could not be monitored. In addition, it was not possible to collect end-end delay information. When an external measuring device is used to collect the delay information, the cost of the arrangement is increased. In addition, there is a method of generating a pseudo call and collecting delay information by analyzing the pseudo call.
This method has a drawback that a load is imposed on the network because traffic is generated on the IP network. Since a terminal for measurement is required, the cost increases.

[0003]

As described above, the conventional I
In the P network, voice quality cannot be managed for each call, en
There are problems that d-end delay information cannot be collected and that an external device is required, but costs increase.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned conventional problems, to be able to remotely monitor the call quality of each call, and to measure without placing an external measuring device. Another object of the present invention is to provide a voice quality management method for each call in an IP network that can collect end-end delay information by suppressing costs.

[0005]

In order to achieve the above object, a voice quality management method for each call in an IP network according to the present invention detects quality deterioration in a form close to real time and analyzes a section suspected of the quality deterioration. The OSS performs control to avoid quality degradation in the OSS. In addition, the quality degradation information obtained from a plurality of GWs is collated with the network topology information previously stored in the OSS, and displayed, whereby a section suspected of quality degradation can be specified, and a fault / congestion point can be separated. I do. By performing control such as detouring on the specified section, it is possible to recover the service early. Specifically, a threshold value is set in advance for each GW from the OSS, a determination is made by the GW that the threshold value has been exceeded, and only those exceeding the threshold value are notified to the OSS ( First Embodiment). G
W notifies the OSS of all information, and determines that the OSS has exceeded the threshold value (second embodiment).

[0006]

Embodiments of the present invention will be described below in detail with reference to the drawings. (First Embodiment) FIGS. 1 to 3 are explanatory views of the principle of operation of a voice quality management method for each call in an IP network according to a first embodiment of the present invention. FIGS. 1 to 3 show screens in the OSS at the upper and lower stages, respectively, so that a threshold can be set on the screen in the OSS. The symbol GW in the network topology is a gateway, and the other unmarked intersections are routers. In addition,
Generally, an IP address is assigned to each interface of the router.
One router has one IP address. 1 to 3 show the subnet mask 255.25.
There is a condition of 5.255.0, but the description is omitted. (1) First, a threshold value of voice quality management information is remotely set in advance from the OSS to a GW that collects voice quality management information for each call. For example, a setting is made to notify the OSS of a delay of 200 msec or more (FIG. 1 (1)). This inputs specific numerical values in a GUI. Actually, GW
Can be set as a) SNMP (Simple)
Network Management Proto
col) set method and b) config
(Setting) to telnet or FTP (File Tr
and setting the GW in the GW using the transfer protocol.

[0007] (2) The OSS is provided with connection relation information (network topology information) of an IP address assigned to a communication device in advance. For example, each MG
(Gateway) and 10.10.1 for each node
IP address connection information of 0.1 to 10.10.10.10.1 is provided (FIG. 1 (2)). (3) When the threshold value set in the GW is exceeded, the OSS is notified of the quality information of the call. As the quality information, for example, route information (10.
10.20.1 → 10.10.90.1 → 10.10.
60.1 → 10.10.30.1) and the delay time is 23
If it is 0 msec or more (exceeding the threshold), the OS
S is notified (FIG. 2 (3)). In addition, as a specific method of setting the network topology information in the OSS,
In order to discover a node by using SNMP polling, first, the OSS itself is polled by SNMP (SNMP management command get) to know the default gateway. Next, SNMP polling is performed on the default gateway to obtain routing table information. SNMP polling is performed again on the obtained next hop router. This operation is sequentially repeated to create a topology. Further, the host information (IP address) is known from the ARP table. (4) The quality information notified from the plurality of GWs is compared with the connection relationship information of the IP address stored in the OSS, and the route in which the quality deterioration occurs is displayed. For example, in FIG. 2, route information (10.100.20.1 → 10.10.90.
(1 → 10.10.60.1 → 10.10.30.1), a plurality of pieces of quality information notified from the MG are compared with the OSS topology information, and the quality linking path is obtained. Is displayed (FIG. 2 (4)).

[0008] (5) A section where a plurality of pieces of route information having deteriorated quality are overlapped is displayed as a section suspected of quality deterioration. Here, in addition to the held route information, 10.10.10.1 → 10.10.40.
1 → 10.10.90.1 → 10.10.60.1 → 1
Since the route of 0.10.30.1 was also notified as the quality degradation, both the overlapping portions, 10.10.0.1 → 1
The route of 0.10.60.1 is displayed as the quality degradation suspected section (router A to router B) (FIG. 3 (5)). OSS
First, the node (router) address on the route of each call corresponding to the errored RTCP is written out.
Next, since the section where the addresses overlap is the section suspected of quality deterioration, this section is displayed. This can be automatically displayed programmatically. (6) Information likely to be related to the quality deterioration suspected section, such as failure information and construction information, is displayed. In FIG. 3, the overlapping router 10.10.10.0.1 indicates “failure information, router A, port 1: 1.10.90.1 Link Do.
wn ”is displayed (FIG. 3 (6)).

FIG. 4, FIG. 5, and FIG. 6 are explanatory diagrams of detour control, band control, and call control, respectively, illustrating the operation principle of the present invention. 4, 5 and 6 are shown in FIGS.
When the quality deterioration suspected section is displayed, the service is restored by performing the detour control, the band control, or the call control. 4 to 6 show the subnet mask 25.
There is a condition of 5.255.255.0, but the description is omitted. FIGS. 4 (7) and (8): Control is performed on the quality deterioration suspected section. In FIG. 4, three detour control routes are shown as indicated by broken lines. These are 10.10.20.
1 → 10.10.70.1 → 10.10.80.1 → 1
0.10.100.1 → first of 10.10.10.10.1
Detour route and 10.10.20.1 → 10.10.7
0.1 → 10.10.50.1 → 10.10.60.1
→ The second detour route of 10.10.30.1 and 10.1
0.10.1 → 10.10.40.1 → 10.10.5
0.1 → 10.10.60.1 → 10.10.30.1
This is the third detour route (FIGS. 4 (7) and (8)).

FIGS. 5 (7) and (8): Control is performed for the section suspected of quality deterioration. In FIG. 5, the quality degradation is displayed on three routes, and 10.10.
On the route from 0.1 to 10.10.60.1, the suspected quality degradation section is displayed (router A to router B). As a result, the bandwidth control (bandwidth increase) is performed for the quality deterioration suspected section.
Is performed. FIGS. 6 (7) and (8): Control is performed on the quality deterioration suspected section. In FIG. 6, the quality deterioration is displayed on three routes, and the three routes overlap 10.10.90.1 → 1
On the route of 0.10.60.1, the quality deterioration suspected section is displayed (router A to router B). As a result, call control is performed, and MG 10.10.2.
MG 10.1, which is the entrance of 0.1 and one other path
Incoming call control of 0.10.1, that is, call restriction is performed at this gateway by dividing time.

FIGS. 7 and 8 are diagrams showing examples of information and an acquisition method which can be acquired by the GW in an application (the specification and drawings of Japanese Patent Application No. 2000-246424) previously proposed by the present applicant, and quality information. It is. The present invention utilizes this technology. In the present invention, a quality management packet (RTCP) used during communication between communication devices,
By using traceroute, quality information is generated. The quality information includes the time of occurrence of the call, the STM line number, the destination IP address of the IP side, the packet loss rate,
There are fluctuation time, delay, and IP address of a passing IP host. A threshold value is remotely set by the OSS in advance for a plurality of GWs collecting quality management information. Then, when the threshold value is exceeded, the GW notifies the OSS of the quality information of the call.

The OSS is provided with a connection relationship of an IP address assigned to a communication device in advance. The quality information possessed by the GW includes between the GW and the terminal, or between the GW and the G.
Since there is information on the passing host IP address through which the call whose quality has deteriorated between W has passed, the IP address and the I
It compares with the P address information and displays the result of the comparison. If one section is deteriorated, a plurality of calls using the section are deteriorated.
S is notified. When the route information of the plurality of GWs whose quality has deteriorated is displayed, there is a portion where the route information overlaps. The overlapping portion is a section in which the quality is highly likely to have deteriorated, and from this, the section suspected of quality deterioration can be specified. In addition, failure information, construction information, and the like that are likely to be related to the quality deterioration suspected section are displayed. The maintenance person performs control (detour, band control, call control) on the quality deterioration suspected section based on the information, and restores the service.
According to the control content determined in advance, it is possible to automatically shift to the control without the intervention of a maintenance person.

FIG. 7 shows information that can be acquired by the GW described in the specification of Japanese Patent Application No. 2000-246424 and an acquisition method (paragraph numbers in parentheses). The packet loss rate from the GW (communication system) to the communication terminal is calculated from the number of lost packets from the GW to the communication terminal included in the RR packet and the number of transmission packets from the GW to the communication terminal calculated by the transmission packet number counter unit. It is calculated using the formula (number of lost packets / number of transmitted packets). The packet loss rate in the GW direction from the communication terminal is calculated as follows: the communication terminal included in the SR packet → the number of transmission packets in the GW direction and the communication terminal notified from the voice packet quality measurement unit →
It is calculated from the number of lost packets in the GW direction using a calculation formula (number of lost packets / number of transmitted packets).

As the fluctuation time of the packet from the GW to the communication terminal, the fluctuation time of the voice packet from the GW to the communication terminal included in the RR packet is used. The fluctuation time of the packet in the GW direction from the communication terminal uses the fluctuation time of the communication terminal → voice packet in the GW direction notified from the voice packet quality measurement unit. The delay time of the packet between the GW and the communication terminal is Ping
It is calculated from the transmission start time of the packet and the reception time of the Ping packet returned in response to the packet by a calculation formula (reception time-transmission start time). The packet route information between the GW and the communication terminal is based on the received T
The pass IP host (router) route information included in the raceRoute packet is used.

Next, FIG. 8 shows a structure of Japanese Patent Application No. 2000-2464.
An example of the quality information described in the specification of No. 24 is shown. Call identification information (call occurrence time, STM line number,
IP communication destination IP address), IP communication destination → packet loss and fluctuation time of communication system, communication system → IP
The packet loss and fluctuation time, delay time, and IP address of the passing IP host at the side communication destination are shown.

Second Embodiment FIGS. 9 to 12 are diagrams showing a specific example of a voice quality management method according to a second embodiment of the present invention. As described above, in the second embodiment, the GW notifies the OSS of all information, and determines that the OSS has exceeded the threshold. FIG. 9 is an explanatory diagram of a VoIP service for each QoS. In this method, for a GW accommodating a high-quality VoIP service, a threshold is set strictly for information such as delay and loss, and for a GW accommodating a low-quality VoIP service. The threshold is set loosely for information such as delay, loss, and the like. This makes it possible to monitor at one place whether or not the calls accommodated in each MG satisfy the service standards. Then, the maintenance person can find the quality deteriorated portion at an early stage. In FIG. 9, a GW 6b of a low-quality VoIP service is provided at the boundary between the PSTN (Public Switched Telephone Network) 2b and the IP-NW 3, and a GW 6a of the high-quality VoIP service is provided at the boundary between the PSTN 2a and the IP-NW 3. Is connected to OSS4. O
In SS, the call accommodated in each GW is SLA
It is possible to monitor whether the criteria are satisfied.

FIG. 10 is an explanatory diagram of a monitoring method by polling. Each bold arrow is a miracle of polling. By polling from the OSS (operation system) 4, information such as the number of transmitted / received packets and the number of lost packets for each physical port such as the GW (gateway) 6a and IP routers (MIBa to i) is collected at regular intervals. That is, information (such as the number of passing packets) stored in the MIB of each node is obtained from the OSS by SNMP. Specifically, the OSS issues an SNMP management command “get” to the management target node, and the node that has received the SNMP management command “get” the target information from the MIB to the OSS.
Return with et-Response. By performing this at regular intervals, information is recorded, collected, and statistically performed by the OSS.

FIG. 11 is an explanatory diagram of a method of arranging probes and capturing packets. In this method, probes a to l are arranged on a network, and a flowing packet is captured by a manager (OSS) 4 and analyzed. In other words, a device (probe) that exclusively collects packet information outside the node instead of the MIB inside the node (probe)
Put. Information is acquired from the OSS by SNMP in the same manner as described above for the MIB of the device that collects information exclusively. The difference from FIG. 10 is that more dedicated information can be acquired by using a dedicated device for collecting information, and that no load is placed on the node by using an external device. The dotted arrow in the figure is an abnormal time notification signal from each probe arranged in the network. 3 is a network, and 6a is a GW (gateway).

FIG. 12 is an explanatory diagram of a method of throwing a test packet. In this method, measuring devices a to
e, a test packet (RTP, etc.) (shown by a thick arrow) flows, and the packets are measured for delay, jitter, etc. by measuring devices a to e, and collated with a threshold value. And
The matching result is notified to the manager (OSS) 4 (dotted arrow is a notification signal). Note that only the result exceeding the threshold may be notified to the manager 4. That is, a device for transmitting a test packet and a device for receiving and analyzing the test packet are installed. For example, by analyzing the packet from the measuring device a in FIG. 12 with the measuring device b, the delay of the packet (for example,
When a delay of 400 ms is found, an SNMP trap notification is sent to the OSS (manager). The OSS receives the trap and can confirm that an abnormality has occurred in the measuring device b.

[0020]

As described above, according to the present invention,
It is possible to remotely monitor the call quality of each call,
In this case, measurement can be performed without placing an external measurement device, so that cost can be reduced. In addition, since the quality deterioration information acquired from a plurality of GWs is compared with the network topology information previously provided in the OSS, and the result is displayed, the quality deterioration suspected section is specified.
The failure / congestion point can be separated. Furthermore, by performing control such as detour for the specified section,
Service can be restored early.

[Brief description of the drawings]

FIG. 1 is an operation principle diagram (1) of a voice quality management method for each call in an IP network according to a first embodiment of the present invention.

FIG. 2 is an operation principle diagram (2) of the voice quality management method for each call in the IP network.

FIG. 3 is an operation principle diagram (3) of the voice quality management method for each call in the IP network.

FIG. 4 is an operation principle diagram of a voice quality management method (detour control) for each call in the same IP network.

FIG. 5 is an operation principle diagram of a voice quality management method (bandwidth control) for each call in the IP network.

FIG. 6 is an operation principle diagram of a voice quality management method (call control) for each call in the IP network.

FIG. 7 is an explanatory diagram of information that can be acquired by a GW and an acquisition method described in the specification of the prior application used in the present invention.

FIG. 8 is a diagram showing an example of quality information shown in the specification of the prior application.

FIG. 9 is a diagram showing Vo for each QoS according to the second embodiment of the present invention;
It is an explanatory view of an IP service.

FIG. 10 is an explanatory diagram of a monitoring method using polling.

FIG. 11 is an explanatory diagram of a method of arranging probes and capturing packets.

FIG. 12 is an explanatory diagram of a method of throwing a test packet.

[Explanation of symbols]

OSS: Operation support system, GW: Gateway, 1a, 1b, 1c: Communication terminal, 2a, 2b ...
PSTN (Public Switched Telephone Network), 3 ... IP-NW (Internet Protocol Network), 4 ... OSS, 6
a: Gateway for high-quality VoIP service, 6b: Gateway for low-quality VoIP service, probes a to l
... probes, measuring devices a to e ... measuring devices that measure delays and the like of test packets and compare them with threshold values.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Takehisa Ichijo 2-3-1 Otemachi, Chiyoda-ku, Tokyo Within Nippon Telegraph and Telephone Corporation (72) Inventor Ken Fujiwara 2-3-3, Otemachi, Chiyoda-ku, Tokyo No. 1 Inside Nippon Telegraph and Telephone Corporation (72) Masato Ogawa 2-3-1 Otemachi, Chiyoda-ku, Tokyo Nippon Telegraph and Telephone Corporation (72) Takao Matsuda 2-chome Otemachi, Chiyoda-ku, Tokyo No. 3 1 Nippon Telegraph and Telephone Corporation (72) Minoru Mizusawa 3-19-2 Nishi Shinjuku, Shinjuku-ku, Tokyo East Japan Nippon Telegraph and Telephone Corporation (72) Inventor Osamu Matsuura Konan, Minato-ku, Tokyo 1-9-1-1 NTT COMWARE CORPORATION (72) Inventor Yoshiki Tsuzuki 1-9-1 Konan, Minato-ku, Tokyo Co., Ltd. in the F-term (reference) 5K030 HB01 HD03 JA10 MA04 MB04 MB20 MC07 MC09 5K033 DB20 EA02 EA07

Claims (2)

[Claims] 1. A threshold value of voice quality management information is remotely set in advance by an operation support system for a plurality of gateways that collect voice quality management information for each call, and communication equipment is previously set in the operation support system. In the case where the connection relationship information of the IP address assigned to is stored, and when the threshold value set in the gateway is exceeded, the operation support system is notified of the corresponding call quality information. The quality information notified from the plurality of gateways is compared with the connection information of the IP address provided in the operation support system, and the route in which the quality is deteriorated is displayed. Sections with overlapping route information are suspected of quality degradation. As a time interval, this is displayed, and information that is likely to be related to the suspected quality deterioration section such as failure information and construction information is displayed. By the displayed information, a detour, band control, or call to the suspected quality deterioration section is performed. A voice quality management method for each call in an IP network, which performs control to restore service. 2. Collecting all quality information held in a plurality of gateways for collecting voice quality management information for each call from an operation support system, and providing the operation support system with network topology information in advance; A threshold value for quality information is set in advance in the operation support system, and the operation support system extracts information exceeding the threshold value from all quality information collected from a plurality of gateways, By comparing the stored network topology information with the extracted value, the route in which quality deterioration has occurred is displayed, and a section where the displayed plurality of pieces of route information overlap is defined as a section suspected of quality deterioration, and Display, trouble information, construction information, etc. A call in an IP network characterized by displaying information likely to be related to a section, and performing a detour, a band control, and a call control on a section suspected of quality degradation based on the displayed information to restore service. Voice quality management method for each.