JP2007282004A - Voice monitoring and recording system, and method therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To monitor and store a voice at an arbitrary monitor point based on minimum facilities by guiding a voice packet to a voice monitoring device while using functions packaged in a router and a switch. <P>SOLUTION: A session control server 33 sets session information to a BGP protocol based on IP address information and port information of a transmitter extracted from a session management database 32 and sets a destination BGP NextHop address to an address of a voice monitoring device 34, so that the voice packet is guided to the voice monitoring device 34. The voice monitoring device 34 copies and stores the voice packet corresponding to a transmission source IP address, a destination IP address, a transmission source port number and a destination port number designated from the session control server 33 and all the packets are returned to an IP network, thereby extracting only the voice packet to be monitored or recorded. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technique for monitoring or recording a voice packet call flowing on a large-scale network represented by the Internet, for example.

  The voice monitoring function typified by a circuit-switched call has been realized via a monitor trunk provided in the exchange. In the case of circuit switching, since the signaling protocol and voice data for establishing a call are realized through the same exchange, it is easy to specify the voice data for monitoring. On the other hand, in an IP phone represented by VoIP, a signaling protocol for establishing a call is controlled via a SIP proxy server, and voice packets are directly exchanged between terminals. It was difficult to monitor the packets.

  Under the circumstances described above, Patent Document 1 proposes a method of monitoring voice packets by arranging agents for transferring voice packets at various locations. Moreover, in patent document 2, the voice packet can be monitored by guiding the voice packet to the voice conference device to make a multi-party call.

However, the above method requires a special device such as an agent or a voice conference device, and when the IP network becomes large, the number of installed devices and capacity are also required. It was. In addition, when using an audio conference device, it becomes difficult to monitor audio for a call during communication.
In addition, as the prior art documents related to the present invention, there are Patent Documents 1 and 2 described above, but as the techniques used in the present invention, there are Non-Patent Documents 1 to 3.

JP 2004-040538 A (VoIP voice monitoring method and VoIP voice monitoring system) JP 2005-033311 A (voice call recording method) "Dissemination of flow specification rules", [online], [searched February 14, 2006], Internet <http://www.ietf.org/internet-drafts/draft-marques-idr-flow-spec-03 .txt> "IPFIX Protocol Specification", [online], [February 14, 2006 search], <http://www.ietf.org/internet-drafts/draft-ietf-ipfix-protocol-19.txt> "IP Multicast Applications: Challenges and Solutions", [online], [Search February 14, 2006], <http://www.ietf.org/rfc/rfc3170.txt>

Currently, the following methods are utilized when monitoring and recording voice / image calls in an IP network environment. In this case, the audio / image data is transferred by the RTP protocol (Real-time Transport Protocol). The quality status between the end hosts is transmitted and received by the RTCP protocol. Here, these are collectively referred to as voice packets.
・ It is possible to monitor and record audio by arranging a number of audio monitoring devices that manage audio packets in advance on the network through which audio packets pass, and restoring the audio packets to audio on this audio monitoring device. (See FIG. 1).
・ A large number of agents that manage voice packets are placed in advance on the network through which voice packets pass. When this agent manages call information during a call and monitors specific call information, In response to the instruction, the agent transfers the voice packet to the voice monitoring device, and enables voice monitoring and recording on the voice monitoring device (see FIG. 2).
・ Voice monitoring / recording is possible by drawing the voice packet to be monitored into the audio conference device and making it a third party call.

  In the above method, it is necessary to arrange a voice monitor device and an agent / voice conference device that receive voice packets on the network and manage call information in advance. Currently, the network through which voice packets pass is becoming larger year by year, and an environment in which various packets other than VoIP packets are mixed. Under such circumstances, it is difficult to place voice monitoring devices and agents that manage voice packets at various locations in the network. When voice monitoring devices and agents are arranged at various locations on the edge of the network, the number of installations becomes enormous, and when voice monitoring devices and agents are arranged in the network backbone, enormous processing capacity is required.

  In addition, the method of drawing a voice packet into the voice conference apparatus can be realized by arranging only the voice conference apparatus on the IP network. However, it is problematic that voice cannot be monitored for a call during communication. .

  As described above, it has been difficult to monitor voice packets exchanged directly between terminals on an IP network. Further, when realizing a voice monitor, it is necessary to mount a special device, which is very inefficient in terms of equipment cost. Also, it has been desired to implement a voice monitor for calls during communication.

  In the present invention, the voice packet is guided to the voice monitoring device by using a function implemented in a network device such as a router or a switch, and the voice at an arbitrary monitoring point is monitored based on the minimum equipment. It is an object of the present invention to provide a system and a method that enable to perform and accumulate.

Of the inventions disclosed in this specification, the outline of typical ones will be briefly described as follows.
On the IP network through voice packets, a minimum necessary voice monitor device is arranged, and voice packets flowing on the network are guided to the voice monitor device, thereby enabling voice monitoring and recording. In order to guide the voice packet to the voice monitor device, the SIP proxy server uses session management information obtained by receiving the SIP message.

  By manipulating route information in the network, guiding voice packets to the voice monitoring device, replicating the voice packets, returning the voice packets to the original network, and using the router function to monitor the packets There is a method of mirroring (duplicating) or sampling a packet on the router side and guiding it to a voice monitoring device.

  In order to perform the route operation, this is realized by rewriting the BGP NextHop address of the packet to the address of the voice monitoring apparatus and injecting the route using the BGP protocol.

  When mirroring (duplicating) a voice packet on the router side and transferring it to the voice monitoring device, the IP address of the router through which the voice packet passes is specified, and a mirroring instruction (duplication instruction) for the packet is sent to this router To achieve this.

  When a voice packet is sampled on the router side and transferred to the voice monitoring device, the IP address of the router through which the voice packet passes is specified and sent to the voice monitoring device using the packet sampling protocol. The voice monitoring apparatus achieves this by restoring the original IP packet from the received packet sampling protocol packet.

  According to the present invention, it is possible to monitor the sound without incurring a great expense. For example, it is possible to monitor the voice of a malicious user or perform a voice quality evaluation and a busy determination by an inquiry from the user. In addition, it is possible to provide a voice recording function that has conventionally been provided in a portable terminal on the network side. On the terminal, since the storage capacity for storing the recorded voice data is not sufficient, the recording time is limited. However, by realizing it on the network side, this limitation can be eliminated.

  Embodiments of the present invention will be described below with reference to the drawings.

This embodiment uses the BGP protocol to direct voice packets to the voice monitoring device.
FIG. 3 shows a configuration diagram of the voice monitoring recording system of the present embodiment. In FIG. 3, 31 is a SIP (Session Initiation Protocol) proxy server, 32 is a session management database, 33 is a session control server, and 34 is a voice monitor device. Reference numeral 35 denotes an IP network. Terminal # 1, terminal # 3, terminal # 4, and terminal # 6 are connected to IP network 35 via router # 1, router # 3, router # 4, and router # 6, respectively. The SIP proxy server 31, the session management database 32, the session control server 33, and the voice monitor device 34 are connected to the IP network 35 via the router # 5. In this embodiment and the embodiments described later, the IP network 35 may be another network, the session control server 33 may be a control device, and the routers # 1 to # 6 may be network devices. In this embodiment, the routers # 1 to # 4 and # 6 are edge routers installed at the edge of the IP network 35. Generally, however, the routers inside the IP network 35 are routers through which voice packets pass. It may be an installed router. In the present invention, the audio packet includes a packet of video data accompanied by audio. The audio monitor device also includes a monitor device for video data packets accompanied by audio.

  The system of this embodiment includes a SIP proxy server 31 that relays a signaling protocol such as SIP, and a session management database 32 that stores session management information extracted from SDP (Session Description Protocol) information in the SIP. And a session control server 33 for extracting information to be monitored from the session management database 32 and instructing packet control to a specific router, and a voice monitor device 34 having a voice monitor and recording function. The SIP proxy server 31 has a function of relaying an outgoing call signal (INVITE) from a terminal and subsequent signaling, and manages session information established based on SDP information described in INVITE, ACK, and 200 OK. It can be stored in the database 32. The SIP proxy server 31 deletes the session information from the session management database 32 when the call processing is terminated by SIP signaling (BYE). The session information stored in the session management database 32 stores the source and destination IP addresses and port numbers via voice packets, and the codec type.

  The session control server 33 collects BGP (Border Gateway Protocol) information in the IP network in advance, and specifies the router through which the voice packet passes when the destination IP address is specified from the BGP NextHop address information. Is possible. At this time, the session control server 33 can acquire this information by establishing peering with a route reflector (not shown) in the IP network using the BGP protocol.

  FIG. 4 shows an outline of the operation of the SIP proxy server 31 when the terminal # 3 and the terminal # 6 make a call. A call is made from the terminal # 3, and the SIP proxy server 31 receives the INVITE, 200OK, and ACK signals, and writes them in the session management database 32 based on the SDP information (S41). Terminal # 3 and terminal # 6 can exchange voice packets by exchanging session information (S42).

  FIG. 5 shows an outline of the operation when monitoring sound. Based on the session information included in the session management database 32, the session control server 33 is instructed to monitor voice (S51). Upon receiving the instruction, the session control server 33 collects the session information to be monitored from the session management database 32 and takes it over to the voice monitor device 34 (S52), and uses the NextHop of the BGP protocol as the address of the voice monitor device 34. The session is instructed to pass through the voice monitor device 34 (S53). If the router on the IP network implements BGP Flow Spec, session information (source IP address, destination IP address, source port number, destination port number) is specified, and only the session is It is possible to go through the monitor device 34. When a router on the IP network does not implement BGP Flow Spec, session information (destination IP address) is specified, and only packets for the terminal can pass through the voice monitor device 34. .

  The voice monitoring device 34 enables voice recording by duplicating and storing the session information packet instructed by the session control server 33, and enables voice monitoring by restoring the voice to the voice. The packet received by the voice monitor device 34 is transferred to the destination side edge router # 6 by connecting the voice monitor device 34 and the incoming router # 6 through the VPN tunnel 36. The received destination edge router # 6 transfers the voice packet to the receiving terminal # 6 (S54).

  This embodiment uses the mirroring function of the router to guide the voice packet to the voice monitor device. The normal mirroring function refers to a function of copying all packets flowing into the monitor port (monitoring IF) and transmitting them through the mirror port (IF that sends out the mirror packet). Selective mirroring that adds specific conditions (source IP address, destination IP address, source port number, destination port number, etc.) to this mirroring function, and sends only those that match the conditions from the mirror port Called a function.

  In this embodiment, a selective mirroring function implemented in the router is used. FIG. 6 shows this outline. FIG. 6A is a diagram for explaining a normal mirroring function. When Packets #A to #D flow into the router, the router sends out Packets #A to #D to the destination port, and all Duplicate the packet and send it to the mirror port. On the other hand, FIG. 6B is a diagram for explaining the selective mirroring function, and the router duplicates only the packet (Packet # B in the figure) that matches the condition and sends it to the mirror port. .

  FIG. 7 shows a system configuration diagram of this embodiment. The voice monitor device 34 is directly connected to each edge router by the VPN 37 or is connected to the wide area LAN 38.

  The outline of the operation of the SIP proxy server 31 when the terminal # 3 and the terminal # 6 talk is also the same as in FIG. FIG. 8 shows an outline of the operation when monitoring sound. Based on the session information included in the session management database 32, the session control server 33 is instructed to monitor voice (S81). Receiving the instruction, the session control server 33 collects the session information to be monitored from the session management database 32 and takes it over to the voice monitor device 34 (S82).

  The session control server 33 specifies the router through which the voice packet passes based on the BGP route information. The router also specifies the IF (interface) that receives the voice packet based on the BGP route information.

  The session control server 33 instructs the specified router (router # 6 in the case of FIG. 8) to perform selective mirroring (S83). The received IF is used as a monitor port, and mirrored packets are selected as mirrored packets based on the session information (source IP address, destination IP address, source port number, destination port number).

  The mirrored voice packet is guided from the mirror port to the voice monitoring device 34 via the wide area LAN 38 or the VPN 37 (S84). The voice monitoring device 34 enables voice recording by duplicating and saving the session information packet instructed by the session control server 33, and enables voice monitoring by restoring the voice to the voice (S85).

  In this embodiment, in order to guide the voice packet to the voice monitor device, a method of sampling the packet and sampling the packet is used. As packet sampling methods, sFlow and IPFIX are implemented in the router. The normal packet sampling function specifies the router IF, extracts packets at a predetermined sampling rate for all packets that flow into the IF, and uses sFlow and IPFIX for the upper header portion in the packet. Distribution is performed according to a predetermined packet format. It is possible to designate in the router as to how many bytes are to be distributed from which protocol part. (This is called MAXIMUM_HEADER_SIZE.)

  In this embodiment, a selective sampling function is used. This selective sampling function adds a specific condition (source IP address, destination IP address, source port number, destination port number, etc.) and samples and extracts only those that match the condition. FIG. 9 shows this outline. FIG. 9A is a diagram for explaining a normal packet sampling function. When Packets #A to #D flow into a router, the router sends Packets #A to #D to a destination port and performs sampling. Packets are sampled at a rate of 1 / N, and the sampled packets are sent to a port different from the destination port in sFlow or IPFIX data format. On the other hand, FIG. 9B is a diagram for explaining the selective packet sampling function, in which the router samples only packets that meet the conditions (Packet # B in the figure) at the sampling rate 1 / N, The sampled packet is sent to a port different from the destination port in the sFlow or IPFIX data format.

  The system configuration diagram is the same as FIG. The outline of the operation of the SIP proxy server when the terminal # 3 and the terminal # 6 make a call is the same as in FIG. FIG. 10 shows an outline of the operation when monitoring sound. Based on the session information included in the session management database 32, the session control server 33 is instructed to monitor voice (S101). Upon receiving the instruction, the session control server 33 collects the session information to be monitored from the session management database 32 and takes over to the voice monitor device 34 (S102).

  The session control server 33 specifies the router through which the voice packet passes based on the BGP route information. The router also identifies the IF that receives the voice packet based on the BGP route information. The session control server instructs selective packet sampling to the identified router (router # 6 in FIG. 10) (S103). Designate an IF to receive the packet and perform selective packet sampling on the IF. The packet condition to be selected is session information to be monitored (source IP address, destination IP address, source port number, destination port number), and sampling rate = 1/1 is specified. Since the RTP data length differs depending on the Codec, MAXIMUM_HEADER_SIZE is designated according to the Codec condition. For the type extracted by sampling, the HEADER type that can be extracted from the IP header or the EtherFrame header is selected.

  The extracted packet is shaped into the data format of sFlow and IPFIX, and then delivered to the voice monitor device 34 as an sFlow and IPFIX datagram (S104). In the router, the voice monitor device 34 is designated in advance as a collector for collecting sFlow and IPFX information.

  From the sFlow and IPFIX packets received by the voice monitor device 34, the RTP packet portion of the original packet portion is extracted. These RTP packets are rearranged based on the time interval according to the Codec, and are internally processed as if the RTP packets were received. As a result, the voice packet is restored to voice, enabling voice monitoring and recording (S105).

  The SIP proxy server, the session management database, the session control server, and the voice monitoring device in the first to third embodiments include means for performing each operation, and the means can be configured by a computer and a program. Moreover, you may comprise a part or all of the program with a hardware.

  Generally speaking, the voice monitoring recording system according to the first to third embodiments is a voice monitoring system for monitoring or recording a voice packet call flowing on the network, and monitoring or recording a voice packet call flowing on the network. A control device that instructs the network device so that a packet corresponding to the voice packet to be monitored or recorded is guided to the voice monitoring device, and a packet transmitted from the network device according to the instruction of the control device, A voice monitor device for monitoring or recording voice. Then, when the control device exchanges session information regarding the voice packet by SIP between the caller and the callee to establish a session, the source IP address of the voice packet stored in the session management database by the SIP proxy server Based on the destination IP address, the source port number, and the destination port number, the network device is instructed to guide the IP packet to the voice monitoring device.

  In the voice monitoring recording system of the first embodiment, the control device collects route information based on the BGP protocol, and a voice packet to be monitored or recorded based on the IP address information of the caller extracted from the session management database. Identify the router through. Also, the control device sets the session information in the BGP protocol based on the caller's IP address information and port information extracted from the session management database, and sets the destination BGP NextHop address as the address of the voice monitoring device. Thus, the voice packet is guided to the voice monitoring device, and the voice monitoring device sends the voice packet corresponding to the transmission source IP address, the destination IP address, the transmission source port number, and the destination port number designated by the control device. Is copied and stored, and all packets are returned to the IP network, thereby extracting only the voice packet to be monitored.

  In the voice monitoring system recording system according to the second embodiment, the control device transmits a voice packet included in the session based on the source IP address, destination IP address information, source port information, and destination port number extracted from the session management database. By specifying the router through which the packet is routed and instructing that router to mirror only the extracted packets related to the source IP address, destination IP address, source port number, and destination port number. Guide to monitor device.

  In the voice monitoring system according to the third embodiment, the control device passes the voice packet included in the session based on the source IP address, destination IP address information, source port information, and destination port number extracted from the session management database. A router to be identified is specified, and only the packets related to the extracted source IP address, destination IP address information, source port information, and destination port number are transmitted to the voice monitoring apparatus using the packet sampling protocol. The voice monitoring apparatus restores the original IP packet from the received packet sampling protocol packet.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Of course.

System configuration diagram of the prior art in which a large number of voice monitoring devices are arranged Prior art system configuration diagram with many agents System configuration diagram of Embodiment 1 The figure which shows the operation | movement outline | summary of a SIP proxy server The figure which shows the operation | movement outline | summary at the time of monitoring the sound in Example 1. The figure for demonstrating the mirroring function used in Example 2 System configuration diagram of embodiment 2 The figure which shows the operation | movement outline | summary at the time of monitoring the sound in Example 2. The figure for demonstrating the packet sampling function used in Example 3 The figure which shows the operation | movement outline | summary at the time of monitoring the sound in Example 3.

Explanation of symbols

31 ... SIP proxy server, 32 ... Session management database, 33 ... Session control server, 34 ... Voice monitoring device, 35 ... IP network, 36 ... VPN tunnel, 37 ... VPN, 38 ... Wide area LAN

Claims (7)

A voice monitoring recording system for monitoring or recording a voice packet call flowing over a network,
A control device for instructing the network device so that a packet corresponding to the voice packet to be monitored or recorded is guided to the voice monitoring device when monitoring or recording a call of the voice packet flowing on the network;
A voice monitoring device that receives a packet transmitted from a network device according to an instruction of the control device, and monitors or records a voice;
A voice monitoring recording system comprising: The voice monitoring recording system according to claim 1,
The control device, when exchanging session information regarding a voice packet by SIP between a caller and a callee for establishing a session, a source IP address of a voice packet stored in a session management database by a SIP proxy server, A voice monitoring recording system characterized by instructing a network device to direct an IP packet to a voice monitoring device based on a destination IP address, a transmission source port number, and a destination port number. The voice monitoring recording system according to claim 2,
The control device collects route information based on the BGP protocol, and specifies a router through which a voice packet to be monitored or recorded passes based on the IP address information of the caller extracted from the session management database. Features voice monitoring recording system. The voice monitoring recording system according to claim 3,
The control device sets the session information to the BGP protocol based on the caller's IP address information and port information extracted from the session management database, and sets the destination BGP NextHop address as the address of the voice monitoring device. By setting, the voice packet is guided to the voice monitoring device,
The voice monitoring device duplicates and stores voice packets corresponding to the source IP address, destination IP address, source port number, destination port number designated by the control device, and returns all packets to the IP network. A voice monitoring recording system characterized by extracting only voice packets to be monitored or recorded. The voice monitoring recording system according to claim 2,
The control device identifies a router through which a voice packet included in the session passes based on the source IP address, destination IP address information, source port information, destination port number extracted from the session management database, and Instructing the router to mirror only the packet related to the extracted source IP address, destination IP address, source port number, and destination port number, thereby guiding the voice packet to the voice monitoring device. Voice monitoring recording system. The voice monitoring recording system according to claim 2,
The control device identifies a router through which a voice packet included in the session passes based on the source IP address, destination IP address information, source port information, destination port number extracted from the session management database, and Instructs the router to send only the packets related to the extracted source IP address, destination IP address information, source port information, and destination port number to the voice monitoring device using the packet sampling protocol (sFlow, IPFIX). And
The voice monitoring recording system, wherein the voice monitoring device restores an original IP packet from a packet according to a received packet sampling protocol. A voice monitoring recording method in a voice monitoring recording system for monitoring or recording a call of a voice packet flowing over a network,
The voice monitoring recording system includes a control device and a voice monitoring device,
Instructing the network device to direct a packet corresponding to the voice packet to be monitored or recorded to the voice monitoring device when the control device monitors or records a call of the voice packet flowing on the network;
The voice monitoring device receives a packet transmitted from a network device in accordance with an instruction from the control device, and performs voice monitoring or recording;
A voice monitoring recording method comprising:

Images