JP4124065B2 - Data communication method and apparatus - Google Patents

Description

  The present invention relates to a data communication method and apparatus for performing data communication such as facsimile data via a packet network such as the Internet.

  As a real-time FAX communication system via an IP (Internet Protocol) network, ITU-U recommendation T.264 shown in Non-Patent Document 1 is used. There are 38. As an apparatus and method assuming an embodiment in accordance with such a standard, for example, a communication connection apparatus and a data output control method disclosed in Patent Document 1 can be cited.

One embodiment of the apparatus and method in the above document uses UDP (User Datagram Protocol) for packet transmission to perform communication in a packet network. Although the packet communication form using UDP is excellent in real time, packet loss may occur. In order to recover the data lost due to such packet loss to the receiving side, for example, a method is adopted in which an old packet sent before is added to a newly transmitted packet from the transmitting side. That is, the receiving side monitors the sequence number added to the packet, recognizes the packet loss from the discontinuity of the number, and picks up the old packet added to the new packet sent next. Data lost due to loss can be recovered. The number of old packets (that is, redundant packets) added to a newly sent packet is fixedly set so that, for example, the redundancy is set to 3 and three old packets up to three generations before are generated.
ITU-T recommendation 38 JP 2002-77225 A

  However, depending on such an apparatus or method, the optimal redundancy cannot be maintained in a communication environment in which the congestion state of the network changes with time, or the probability of occurrence of packet loss varies depending on the network route to the other party. . That is, depending on the fixed redundancy setting, there is a problem that the redundancy is not sufficient and data cannot be recovered, or the data is wastefully transferred because the redundancy is excessively high.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a data communication method and apparatus for realizing data communication in accordance with a redundancy that responds quickly to a network congestion state in order to solve the above problems.

Facsimile data communication method according to the present invention is a facsimile communication method for transmitting and receiving information data in a series of a plurality of packets of the form via the packet data network, received by the transmission previous transmission information data, the packet A traffic state estimation step for estimating a traffic state of the transmission information data from the data network to the other facsimile machine , and setting a packet redundancy of the transmission information data based on the traffic state, A transmission step of transmitting transmission information data to the counterpart facsimile machine according to the packet redundancy.

Facsimile communication apparatus according to the present invention is a facsimile communication apparatus for transmitting and receiving information data in a series of a plurality of packets of the form via the packet data network, received by the transmission previous transmission information data, the packet Traffic state estimation means for estimating a traffic state of the transmission information data reaching the counterpart facsimile device based on received information data from the data network, and setting packet redundancy of the transmission information data based on the traffic state, Transmitting means for transmitting transmission information data to the counterpart facsimile machine according to the packet redundancy.

  According to the data communication method and apparatus of the present invention, a configuration is provided in which the redundancy of packets transmitted and received in the packet network is changed according to the congestion state of the network. That is, an index representing the traffic state on the route to the other party via the network is estimated, and when the traffic state is high, transmission data is repeatedly transmitted. As a result, data communication having a redundancy that is immediately adapted to the congestion state of the network is realized. When transmitting / receiving FAX data, transmission / reception of a high-quality FAX image without data loss is achieved.

Embodiments of the present invention will be described in detail with reference to the accompanying drawings.
<First embodiment>
FIG. 1 shows the configuration of the Internet FAX connection device 20. The data communication apparatus according to the present invention, that is, the Internet FAX connection apparatus 20 is connected between the IP network 30 and the G3 FAX apparatus 10 and relays and transfers FAX data between the IP network 30 and the G3 FAX apparatus 10 in real time. . The communication method for relaying FAX data as image data in real time is ITU-T recommendation T.264. Follow 38 standards. The protocol related to the call is ITU-T recommendation H.264. It follows the H.323 standard. The G3 FAX apparatus 10 is an ITU-T recommendation T.30. The facsimile apparatus can be used on a normal telephone line in accordance with the 30 standard, and can be connected to the Internet FAX connection apparatus 20 by a normal circuit switching network.

  The Internet FAX connecting apparatus 20 includes a FAX control unit 21, an IFP / CODEC unit 22, an UDPTL / PAD unit 23, and a LAN control unit 24. These are connected in series. The Internet FAX connecting apparatus 20 further includes a packet drop detection unit 27 and a congestion detection unit 28.

  The FAX control unit 21 has a transmission function for transmitting data received from the G3 FAX apparatus 10 to the IFP / CODEC unit 22 and a reception function for transmitting data transferred from the IFP / CODEC unit 22 to the G3 FAX apparatus 10.

  The IFP / CODEC unit 22 encodes the FAX data passed from the FAX control unit 21 (IFP packetization), and the reception function decodes the IFP packet passed from the UDPTL / PAD unit 23 into FAX data. Have Such encoding and decoding are performed based on the packetization information stored in the packetization information storage unit 25 connected to the IFP / CODEC unit 22.

  The UDPTL / PAD unit 23 is connected to the transmission buffer 26. If there is old data in the transmission buffer 26, the UDPTL / PAD unit 23 adds the IFP packet to the UDPTL packet, and the IFP / CODEC unit converts the necessary IFP packet from the UDPTL packet. 22 has a receiving function.

  The LAN control unit 24 has a transmission function of transmitting an instructed packet toward the IP network 30 and a reception function of passing a packet received from the IP network 30 to the UDPTL / PAD unit 23.

  The packet drop detection unit 27 is connected between the IFP / CODEC unit 22 and the UDPTL / PAD unit 23 and has a function of detecting a packet drop from the sequence number of the received IFP packet and outputting a packet drop notification. The congestion control unit 28 is connected between the packet drop detection unit 27 and the UDPTL / PAD unit 23, calculates the packet loss rate from the packet drop notification output from the packet drop detection unit 27, and determines the redundancy of the transmission UDPTL packet. It has a function of determining and notifying the UDPTL / PAD unit 23.

  FIG. 2 shows the structure of a UDPTL packet. The UDPTL packet is a packet based on UDPTL (Facsimile UDP Transport Layer Protocol), and performs communication via the IIP Internet 30 using UDP (User Datagram Protocol). The UDPTL packet includes a header part for control based on the UDPTL protocol and a plurality of IFP packets.

The IFP packet is a packet based on IFP (Internet Facsimile Protocol), and is data obtained by encoding and packetizing FAX data exchanged with the G3 FAX apparatus 20. The IFP packet is composed of a primary part packet and a secondary part packet. The primary part packet is new data to be transmitted, while the secondary part packet is a packet stored in the transmission buffer 26 as a recovery means when the packet is lost. Each packet is identified by a unique number (hereinafter referred to as a sequence number). In the example of this figure, it consists of a primary part packet with sequence number 5 as a new packet and secondary part packets with sequence numbers 4 and 3 as two secondary part packets. In this example, two redundant packets are stored with a redundancy of 2. Such packetization is performed based on the redundancy instructed from the congestion detection unit 28 in the UDPTL / PAD unit 23. The redundancy of the UDPTL packet is optimized to reliably perform real-time FAX communication by UDP. That is, when the packet loss rate of the IP network 30 is large, the redundancy is increased and the number of IFP packets added as the secondary part is increased. Conversely, when the packet loss rate of the IP network 30 is small, the redundancy of the UDPTL packet is lowered to reduce the number of IFP packets added as the secondary part.

  FIG. 3 shows a processing procedure at the time of reception by the Internet FAX connecting apparatus 20. Hereinafter, the processing procedure shown in this figure will be described with reference to the components shown in FIG.

  First, in response to a reception request from another FAX apparatus via the IP network 30, the Internet FAX connection apparatus 20 performs an IFP reception session start process (step S1). Next, the LAN control unit 24 receives a UDPTL packet from the IP network 30 (step S2). Then, the UDPTL / PAD unit 23 decomposes the UDPTL packet into IFP packets (step S3).

  At this time, the order of the sequence numbers is checked by the UDPTL / PAD unit 23 and the packet drop detection unit 27 of the Internet FAX connection apparatus 20 (step S4). That is, the UDPTL / PAD unit 23 notifies the packet drop detection unit 27 of the sequence number of the primary part of the received UDPTL packet. Next, the packet drop detection unit 27 determines whether or not a packet drop has occurred (step S5). For example, if (reception sequence number) = (previous sequence number + 1), the packets are continuously received normally, so the congestion control unit 28 is notified of the normal reception state. . On the other hand, if (reception sequence number) = (sequence number received last time + 2 or more), it is recognized that one packet has dropped. In response to this recognition, the congestion control unit 28 is notified of the discontinuous reception state. Next, the congestion controller 28 records a packet drop notification (step S8).

  On the other hand, when the packets are continuously received normally, the Internet FAX connecting apparatus 20 decodes the IFP packet into FAX data by the IFP / CODEC unit 22 (step S6). Next, the decoded FAX data is transferred to the G3 FAX apparatus via the FAX control unit 21 (step S7). Next, it is determined whether or not the session is ended (step S9). If not completed, the process returns to step S2 to repeat the reception process in order to receive the next UDPTL packet.

  On the other hand, if the IFP session has ended, the Internet FAX connecting apparatus 20 performs an IFP session end process (step S10). Next, the congestion control unit 28 calculates the packet loss rate from the packet drop notification record and determines the redundancy (step S11). That is, the congestion control unit 28 calculates the packet loss rate in reception and instructs the UDPTLPAD unit 23 to set transmission redundancy. The UDPTL / PAD unit 23 creates a UDPTL packet by changing the number of IFP packets in the secondary unit in accordance with the designation of the redundancy in the subsequent FAX data transmission. The packet loss rate is calculated by the number of dropped packets / (number of dropped packets + number of received packets), and the redundancy that is the number of retransmissions of the same transmission packet is determined according to this rate. For example, when the packet loss rate is 10% to 20%, one retransmission is performed with redundancy 1. The relationship between the packet loss rate and the redundancy can be arbitrarily set in advance. The data acquisition time for calculating the loss rate is not limited to one session, and may be short or long. Therefore, the determination of redundancy need not be determined for each session, but may be determined at a shorter frequency, such as every predetermined number of received packets, or may be determined by measurements over more sessions. good. The determined redundancy specifies the number of retransmissions of the same packet, and the value can be set from 0 to a plurality of times.

  FIG. 4 shows a processing procedure at the time of transmission by the Internet FAX connecting apparatus 20. Hereinafter, the processing procedure shown in this figure will be described with reference to the components shown in FIG.

  First, in response to a transmission request from the G3 FAX apparatus 10, the Internet FAX connection apparatus 20 performs an IFP transmission session start process (Step S11). Next, the FAX control unit 21 receives the transfer of FAX data from the G3 FAX apparatus (Step S12). Next, the IFP / CODEC unit 22 encodes the FAX data into an IFP packet (step S13). At this time, the IFP packet is stored in the transmission buffer 26 by the UDPTL / PAD unit 23 in preparation for addition to the next new packet (step S14).

  Next, the Internet FAX connecting apparatus 20 converts the latest IFP packet into a UDPTL packet using the latest IFP packet as a primary part and the old packet stored in the transmission buffer 26 as a secondary part in the UDPTL / PAD part 23 (step S15). At this time, the number of old packets packed in the secondary part depends on the redundancy specified by the congestion control part 28.

  Next, the Internet FAX connecting apparatus 20 transmits the UDPTL packet to the other party by the LAN control unit 24 (step S16). Then, it is determined whether or not the session is ended. If it is ended, a session end process is performed (step S17). On the other hand, if not finished, the process returns to step S12 to repeat the process for processing the next FAX data.

As is clear from the above description of the flowchart, as shown in the reception process of FIG. 3, by calculating the packet loss rate from continuously received UDPTL packets, it indicates the congestion state reaching the other party in the IP network. As a result, the packet loss rate is estimated. As shown in the transmission process of FIG. 4, by dynamically changing the redundancy of the UDPTL packet transmitted by the own device based on this packet loss rate, real-time FAX communication with reliable and optimum redundancy can be performed. This can be done by UDP / IP.
<Second embodiment>
FIG. 5 shows the configuration of the Internet FAX communication apparatus 20 in the second embodiment. Since the configuration has a common part with the first embodiment, only a different part will be described. Furthermore, in the second embodiment, an example is shown in which an Internet FAX communication device 20 ′ having the same configuration is connected to face through the IP network 30.

  The congestion control unit 28 of the Internet FAX communication device 20 is connected to the LAN control unit 24, and notifies the partner device, that is, the Internet FAX communication device 20 ′, that the packet drop has occurred via the IP network 30. Is further received from the Internet FAX communication apparatus 20 ′ via the IP network 30.

  The operation of the congestion control unit 28 will be described. Upon receiving a packet drop notification from the packet drop detection unit 27, the congestion control unit 28 creates a message notifying that a packet loss has occurred and passes it to the LAN control unit 24. . The FAX control unit 24 transmits this message to the partner apparatus, that is, the Internet FAX communication apparatus 20 '. A method such as a port number and a packet format between devices used at this time can be realized by a normal method.

  When receiving the packet generation message, the Internet FAX communication device 20 ′, which is the counterpart device, calculates the packet loss rate, changes the redundancy of the UDPTL packet currently being transmitted, that is, changes the number of IFP packets in the secondary part next time. From this, a UDPTL packet is created and transmitted with this redundancy.

  The same is executed in the Internet FAX communication apparatus 20. That is, the congestion control unit 28 of the Internet FAX communication device 20 calculates the packet loss rate based on a message notifying the packet drop from the partner device (Internet FAX communication device 20 ′ in the example of this figure). The redundancy of the transmission UDPTL packet, that is, the number of IFP packets in the secondary part is determined.

As described above, in the second embodiment, in addition to the effects of the first embodiment, it is possible to perform control with higher immediacy. That is, when a packet drop occurs, the partner apparatus is notified immediately, and the redundancy of the received UDPTL packet is quickly changed. Also, itself can immediately change its redundancy in response to a packet drop notification from the partner device.
<Third embodiment>
FIG. 6 shows the configuration of the Internet FAX communication apparatus 20 in the third embodiment. Since the configuration has a common part with the first embodiment, only a different part will be described.

  A congestion control unit 28 of the Internet FAX communication apparatus 20 is connected between the LAN control unit 24 and the UDPTL / PAD unit 23. The congestion control unit 28 requests the LAN control unit 24 to measure a congestion state using ICMP (Internet Control Message Protocol) Echo, estimates the congestion state based on the result, and determines the redundancy of the UDPTL packet as UDPTL. / PAD unit 23 has a function to notify. The ICMP Ech command corresponds to a ping command in MS-DOS (registered trademark).

  The operation of the congestion control unit 28 will be described. 38 During the communication, the congestion control unit 28 transmits ICMP Echo to the partner apparatus and measures the time until the response. From this response time, an instruction is issued to the UDPTL / PAD unit 23 to estimate the packet drop of the IP network 30 and change the redundancy of the UDPTL packet. The UDPTL / PAD unit 23 changes the number of IFP packets in the secondary unit in accordance with the instructed redundancy to make a UDPTL packet.

  As a response time measurement method, ICMP Echo transmission / reception processing may be performed a plurality of times, and the average response time value may be used, or time measurement may be performed periodically or regularly during communication.

  As described above, in the third embodiment, the state of the IP network can be estimated by measuring the response time of a general TCP / IP protocol, and the redundancy of the UDPTL packet transmitted by the device itself is dynamically changed. By doing so, real-time FAX communication with reliable and optimum redundancy can be performed by UDP / IP. Even if the counterpart device is not of the same model, congestion control can be performed using this embodiment as long as the device implements a general UDP / IP protocol.

  As a modified example, the congestion control unit 28 of the Internet FAX connecting apparatus 20 can recognize the number of hops to the partner apparatus by using ICMPTraceroot, and the redundancy of the UDPTL packet from the number of hops, that is, the IFP packet of the secondary part It may be configured to have a function of changing the number. ICMPTraceroot corresponds to Tracert in the MS-DOS (registered trademark) command.

  The operation of the congestion control unit 28 will be described. During 38 communications, ICMPTraceroot is transmitted to the partner device, and the number of hops that is the number of routers to the partner device is measured. The UDPTL / PAD unit 23 is instructed to estimate the packet drop of the IP network 30 from this hop count and change the redundancy of the UDPTL packet. The UDPTL / PAD unit 23 executes UDPTL packetization by changing the number of IFP packets in the secondary unit according to the instructed redundancy.

  In the modification of the third embodiment described above, the state of the IP network can be estimated by using the normal UDP / IP protocol. Thus, by dynamically changing the redundancy of the UDPTL packet transmitted by the own apparatus, real-time FAX communication with reliable and optimum redundancy can be performed by UDP / IP. Even if the counterpart device is not of the same model, congestion control can be performed using the present embodiment as long as the device implements a general TCP (Transmission Control Protocol) / IP protocol.

  In the above embodiments, the packet method for real-time FAX communication using the Internet FAX communication apparatus has been described. However, the present invention is not limited to the Internet FAX communication apparatus but can be applied to implementation by software. In FIG. 1, the G3 FAX and the Internet FAX communication apparatus are described as separate apparatuses connected by a circuit switching network. However, the data communication apparatus according to the present invention may be an apparatus having both the G3 FAX function and the Internet FAX communication function. good. Further, the data communication method and apparatus according to the present invention are described in Recommendation T.30. Although the method and apparatus using the redundancy method in the 38 UDP communication has been described, the present invention is also applicable to the case where the FEC processing amount is optimized in the communication device implemented in the FEC (Forward Error Correction) method.

It is a block diagram which shows the structure of the internet FAX connection apparatus in 1st Example of this invention. It is a figure which shows the structure of the UDPTL packet shown by FIG. It is a flowchart which shows the process sequence at the time of reception of an internet FAX connection apparatus. It is a flowchart which shows the process sequence at the time of transmission of an internet FAX connection apparatus. It is a block diagram which shows the structure of the internet FAX connection apparatus in 2nd Example of this invention. It is a block diagram which shows the structure of the internet FAX connection apparatus in 3rd Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 G3 FAX apparatus 20, 20 'Internet FAX communication apparatus 21 FAX control part 22 IFP / CODEC part 23 UDPTL / PAD part 24 LAN control part 25 Packetization information storage part 26 Transmission buffer 27 Packet drop detection part 28 Congestion control part 30 IP network

Claims (6)

A facsimile data communication method for transmitting and receiving information data in the form of a series of a plurality of packets via a packet data network,
A traffic state estimation step for estimating a traffic state of the transmission information data that has been received before transmission of the transmission information data, based on the reception information data from the packet data network, reaching the counterpart facsimile device;
A transmission step of setting packet redundancy of the transmission information data based on the traffic state, and performing transmission of the transmission information data to the counterpart facsimile device according to the packet redundancy;
A facsimile data communication method comprising: The traffic state estimation step detects a loss of the packet based on a discontinuity of a sequence number given to each of a plurality of packets of reception information data received before transmission of the transmission information data, 2. The facsimile data communication method according to claim 1, wherein a packet loss rate is estimated as an index indicating the traffic state. In the traffic state estimation step, a test packet is transmitted to the packet data network, a response packet is received as the received information data, and at least one of the response time and the number of hops indicates the current traffic state. 2. The facsimile data communication method according to claim 1, wherein the facsimile data communication method is estimated as an index. A facsimile data communication device that transmits and receives information data in the form of a series of packets via a packet data network,
Traffic state estimation means for estimating the traffic state of the transmission information data reaching the counterpart facsimile device based on the reception information data from the packet data network received before transmission of the transmission information data;
Transmission means for setting packet redundancy of the transmission information data based on the traffic state, and transmitting the transmission information data to the counterpart facsimile machine according to the packet redundancy;
A facsimile data communication apparatus comprising: 5. The facsimile data communication apparatus according to claim 4, further comprising: a facsimile apparatus that inputs or reproduces facsimile data; and a network connection apparatus that connects the facsimile data to the packet data network as the information data. 6. The network connection device according to claim 5, further comprising: means for encoding the facsimile data into a packet suitable for the packet data network; and means for decoding the packet into the facsimile data. Facsimile data communication device.

Images