WO2005029806A1 - Method for transmitting data and data transmitter - Google Patents

Method for transmitting data and data transmitter Download PDF

Info

Publication number
WO2005029806A1
WO2005029806A1 PCT/JP2004/002947 JP2004002947W WO2005029806A1 WO 2005029806 A1 WO2005029806 A1 WO 2005029806A1 JP 2004002947 W JP2004002947 W JP 2004002947W WO 2005029806 A1 WO2005029806 A1 WO 2005029806A1
Authority
WO
WIPO (PCT)
Prior art keywords
transmission
data
phase
packet
transmission phase
Prior art date
Application number
PCT/JP2004/002947
Other languages
French (fr)
Japanese (ja)
Inventor
Makoto Takizawa
Takuya Tojo
Original Assignee
Tokyo Denki University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tokyo Denki University filed Critical Tokyo Denki University
Priority to JP2005513992A priority Critical patent/JP4382752B2/en
Publication of WO2005029806A1 publication Critical patent/WO2005029806A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/0001Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0009Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the channel coding
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/12Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1812Hybrid protocols; Hybrid automatic repeat request [HARQ]

Definitions

  • the present invention relates to a data transmission method and a data transmission device.
  • TCP Internet Protocol
  • XII Xpress Transport Protocol
  • P Xpress Transport Protocol
  • TCP Transmission Control Protocol
  • XII Xpress Transport Protocol
  • Xpress Transport Protocol a communication algorithm called slow start to avoid network congestion. This slow start starts transmission at the base window size and receives data from the receiver.
  • the window size is increased at a fixed rate. If a received ACK does not arrive or a reception failure NACK arrives, (a) the transmission window size is returned to the base window size and finally Transmission resumes from the bucket following the bucket that received the ACK for successful reception, and the window size gradually increases in the same manner.
  • Patent Literature 3 adds a parity bucket having data that can be restored from the data of the remaining buckets even if one of them is lost for each of a fixed number of k ⁇ 1 buckets, and the k packet data is reduced to 1 packet.
  • a technology for transmitting one segment at a time is disclosed.
  • Patent Document 4 discloses a communication speed control technology that increases the communication speed while congestion does not occur in a packet network and reduces the communication speed if there is a possibility of congestion.
  • this communication speed control can be expected to substantially improve the communication speed
  • a predetermined number of packets is regarded as one segment, and if the transmission is successful, the communication speed is increased by one step.
  • One and one parity bucket are transmitted as one segment, and if successful, a predetermined number of data buckets are transmitted as one segment without a parity bucket. No control is disclosed for increasing the number of data buckets at the same communication speed to substantially increase the communication speed.
  • Patent Document 5 utilizes a plurality of transmission paths. For example, (N-1) data packets and one parity packet are used to divide N buckets into one segment. A technique for transmitting buckets of M segments in a matrix manner over M transmission paths is disclosed. As a result, even if one bucket is lost on one of the transmission lines, the packet can be restored on the receiving side, eliminating the need for retransmission and enabling high reliability and high-speed communication. However, since multiple transmission lines are used, equipment costs are increased, and the process of restoring a series of data on the receiving side is complicated.
  • Patent Document 1 Japanese Patent Application Publication No.
  • Patent Document 2 Japanese Patent Application Laid-Open No. 2003-7785556
  • Patent Literature 3 Japanese Patent Application Laid-Open No. 2-2199336
  • Patent Document 4 Japanese Patent Application Laid-Open No. 7-66683
  • Patent Document 5 Japanese Patent Application Laid-Open No. 2002-36811 Disclosure of the Invention
  • the present invention has been made in view of the above-mentioned conventional technical problems, and enables data recovery on the receiving side to minimize loss of data packets as much as possible, thereby achieving comprehensive data communication.
  • An object of the present invention is to provide a data transmission technology capable of improving speed throughput.
  • This data transmission method includes: (a) a fixed number N (N is an arbitrary natural number) to which a parity packet is added at a predetermined transmission rate at a predetermined transmission rate;
  • the packet group of (1) is defined as one segment, and a segment of a constant M (M is an arbitrary natural number) is included in each transmitted packet group in which the packets belonging to each segment consist of a series of M packets.
  • a data bucket group having the same transmission rate as that of the first transmission phase in (a) but not including parity packets and having the same total number of packets as the first transmission phase is transmitted.
  • This data transmission method comprises the steps of: (a) adding a parity packet at a predetermined transmission rate to a fixed number N of bucket groups on a single transmission path; A predetermined number M of segments, and N packets for each transmission packet group, so that one packet is included in each of the transmission packet groups consisting of a series of M buckets belonging to each segment, for a total of NXM (B) performing data transmission in the first transmission phase for distributing and transmitting the bucket of (a), if no data loss occurs in the transmission of the first transmission phase of (a), A second transmission phase in which a data packet group having the same transmission rate as one transmission phase, including no parity packet, and having the same total number of buckets as the first transmission phase is transmitted in a burst manner.
  • Still another feature of the present invention resides in a data transmission device, which includes a data holding unit that holds data to be transmitted, a network interface that sends and receives data to and from an information network through a single transmission path, A packetization processing unit configured to bucket data to be transmitted held in the data holding unit, a parity bucket generation unit configured to generate a parity packet for a fixed (N ⁇ 1) packet sequence, The transmission phase determining unit that determines the current transmission phase and the transmission speed based on the information on the unsuccessful transmission and the information on the previous transmission phase, and information on the transmission phase determined by the transmission phase determining unit.
  • the fixed number N of bucket groups to which the parity packet is A transmission packet group creation unit that generates or generates a data packet group that does not include the parity packet and has the same number as the total number of packets NXM in the predetermined number M of segments; and a transmission phase determination unit.
  • a data transmission processing unit that transmits the packet group generated by the transmission packet group creation unit to the information network via the network interface at the determined transmission speed, wherein the transmission phase determination unit comprises: (a) At a given transmission rate, a fixed number N of packet groups to which a parity bucket is added is divided into one segment, and a given number M of segments are transmitted.A packet belonging to each segment consists of a series of M buckets.
  • NXM buckets for each transmission bucket group are distributed and transmitted for N groups.
  • B If no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a). Transmission rate, not including parity packets, and Data transmission in the second transmission phase, in which data bucket groups having the same total number of buckets as one transmission phase are transmitted in a burst manner, is performed.
  • C Data loss also occurs in the transmission in the second transmission phase of (b) above. If not, the transmission rate is increased by a preset level to perform the data transmission in the first transmission phase, and (d) data loss occurs in the data transmission in the first transmission phase in (c). If not, the data transmission in the second transmission phase is performed.
  • the data in the first transmission phase in the above (c) is increased while the transmission speed is increased by a predetermined step unless data loss occurs.
  • the transmission and the data transmission of the second transmission phase (d) are repeated.
  • the data transmission is returned to the previous transmission phase one time before. To resume It is characterized in that the current transmission phase and transmission speed are determined based on the following rules.
  • Still another feature of the present invention resides in a data transmission device, which includes a data holding unit that holds data to be transmitted, a network interface that sends and receives data to and from an information network through a single transmission path.
  • a packetization processing unit configured to bucket data to be transmitted held in the data holding unit, and a parity bucket generation unit configured to generate a parity packet for a fixed (N ⁇ 1) packet sequence.
  • a transmission phase determining unit that determines the current transmission phase and transmission speed based on the information on the success or failure of the previous data transmission and the information on the previous transmission phase; and the transmission determined by the transmission phase determining unit.
  • a predetermined number M of segments Based on the phase information, a predetermined number M of segments, with a fixed number N of packet groups to which the parity packet is added as one segment Or a transmission packet group creation unit that does not include the parity packet and generates the same number of data bucket groups as the total number of packets NXM in the predetermined number M of segments, and the transmission packet At the transmission speed determined by the A data transmission processing unit for transmitting the packet group generated by the transmission bucket group creation unit to the information network via the network interface, wherein the transmission phase determination unit comprises: , A fixed number N (N is an arbitrary natural number) of packet groups to which a parity packet is added is defined as one segment, and a predetermined number M (M is an arbitrary natural number) of segments is formed.
  • FIG. 1 is a block diagram showing a functional configuration of a data transmission device according to one embodiment of the present invention.
  • FIG. 2 is a block diagram showing a functional configuration of a data receiving device corresponding to the data transmitting device of the embodiment.
  • FIG. 3 is an explanatory diagram of a bucketing process of application data in the embodiment.
  • FIG. 4 is an explanatory diagram of a distributed transmission process of a plurality of segments according to the above embodiment.
  • FIG. 5 is a flowchart of a data transmission process by the data transmission device of the above embodiment.
  • FIG. 6 is a sequence diagram of the first and first transmission phase data transmission processing by the data transmission device of the above embodiment.
  • FIG. 7 is a sequence diagram of the first and second transmission phase data transmission processing by the data transmission device of the above embodiment.
  • FIG. 8 is a sequence diagram of a second first transmission phase data transmission process in which the data transmission device according to the above embodiment has increased the speed by one step.
  • FIG. 9 is a sequence diagram of a second transmission phase data transmission process performed by the data transmission device of the above-described embodiment at a one-step speed reduction due to a data loss occurring in the second first transmission phase.
  • FIG. 10A is an explanatory diagram of the first transmission phase data transmission according to the embodiment
  • FIG. 10B is an explanatory diagram of the second transmission phase data transmission.
  • FIG. 11 shows data transmission by the data transmission device according to the second embodiment of the present invention. Flowchart of communication processing.
  • FIG. 12 is an explanatory diagram showing a data bucket string transmitted by the data transmission method according to the third embodiment of the present invention.
  • FIG. 13 is an explanatory diagram of a data bucket string and a parity bucket string transmitted by the data transmission method according to the third embodiment.
  • FIG. 14 is an explanatory diagram showing a first stage of data transmission according to the third embodiment.
  • FIG. 15 is an explanatory diagram showing a second stage of data transmission according to the third embodiment.
  • FIG. 16 is an explanatory diagram showing a third stage of data transmission according to the third embodiment.
  • FIG. 17 is an explanatory diagram showing a fourth stage of data transmission according to the third embodiment.
  • FIG. 18 is an explanatory diagram showing a fifth stage of the data transmission according to the third embodiment.
  • FIG. 19 is an explanatory diagram showing a sixth stage of data transmission according to the third embodiment.
  • FIG. 20 is an explanatory diagram showing a seventh stage of data transmission according to the third embodiment.
  • FIG. 21 is an explanatory diagram showing an eighth stage of data transmission according to the third embodiment.
  • FIG. 22 is a sequence diagram of data transmission in the first transmission phase at a speed of the base 5 O M bps in the data transmission method according to the first embodiment of the present invention.
  • FIG. 23 is a sequence diagram of data transmission in the first transmission phase at a speed of 60 Mbps in the data transmission method of the first embodiment.
  • FIG. 24 is a sequence diagram of data transmission in the first transmission phase at a speed of 70 Mbps in the data transmission method according to the first embodiment.
  • FIG. 25 is a graph of data transmission characteristics according to the first embodiment.
  • Figure 26 is a graph of data transmission characteristics according to Comparative Example 1.
  • FIG. 27 is a graph of a measurement result of a packet reception speed by the data transmission method according to the second embodiment of the present invention.
  • FIG. 28 is a graph of the measurement result of the bucket loss rate according to the second embodiment.
  • Fig. 29 is a graph of the results of measuring the bucket receiving speed by the data transmission method of Comparative Example 2.
  • FIG. 30 is a graph of a measurement result of a packet loss rate according to Comparative Example 2.
  • FIG. 1 shows a functional configuration of a data transmitting apparatus 10 according to a first embodiment of the present invention
  • FIG. 2 shows a functional configuration of a data receiving apparatus 20 that receives transmission data of the data transmitting apparatus 10. ing.
  • the data transmitting device 10 and the data receiving device 20 are both incorporated as software in a computer having a data packet communication function, and usually have both transmitting and receiving functions simultaneously. For convenience, it is shown separately for the transmitting side and the receiving side.
  • the functional configuration of the data transmission device 10 includes an HDD that stores large amounts of data, a data storage unit 11 that includes a RAM memory that temporarily stores data, and data to be transmitted from this data storage unit 11.
  • a bucketing processing unit 12 that takes out and performs a bucketing process, a parity adding process that generates a parity packet as necessary for a series of packet sequences of the bucketed data.
  • a hardware network interface (NW I / F) 17 including a LAN board, a LAN adapter, and a router and serving as an interface with the network 30 is provided.
  • the functional configuration of the data receiving device 20 includes a data receiving processing unit 21 for receiving data transmitted from the data transmitting device 10 through the network 30 and a memory 2 for temporarily storing received data.
  • a data receiving processing unit 21 for receiving data transmitted from the data transmitting device 10 through the network 30 and a memory 2 for temporarily storing received data.
  • Parity check unit 23 that performs parity check on received data 23.
  • Data recovery processing unit 24 that attempts to recover data using a parity bucket added to the received data sequence when a data error occurs.
  • the communication status determination unit 25 which determines the success or failure of the data reception based on the status of the memory 22 and the success or failure of the data recovery of the data recovery processing unit 24, and the result of the data reception success / failure determination by the communication status determination unit 25, which is based on A data transmission processing unit 26 for transmitting ACK and NAK signals, and also includes, for example, a LAN board, a LAN adapter and a router as well as the transmission side, and an interface with the network 30. It is equipped with a hardware of Netw network interface (NW I / F) 2 7 which acts as a face.
  • NW I / F Netw network interface
  • the network 30 is a LAN, a public communication network, the Internet, or an information network that uses these in an integrated manner, and is capable of transmitting data packets.
  • the data bucket communication used in the present embodiment will be described.
  • the application data is divided into fixed-length packets pk1, k2, ..., and the exclusive-OR (XOR) of all packets for a fixed number m of packets pk1 to pkm is calculated. Calculation is performed, and the calculation result is used as a parity packet.
  • the predetermined number of data packets and one parity bucket constitute one segment SG.
  • a predetermined number of N-segment packets are transmitted in a distributed manner.
  • Fig. 4 shows an example of transmitting a 2-segment SG1, SG2 packet group, which forms one segment with three buckets + one parity packet.
  • the first packet (1) of segment SG1, the first packet (4) of segment SG2, the second packet (2) of segment SG1, and the second packet of segment SG2 (5),... Are transmitted in order, and finally the parity packets PT 1 and PT 2 of each segment are transmitted.
  • a parity packet is generated for each bucket group consisting of a fixed number m of data buckets, and the bucket group consisting of m + 1 packets including the parity bucket is transmitted as one segment, thereby transmitting the packet. Even if one of the bucket groups is lost, the receiving side can correctly recover the remaining bucket groups and the lost packet data from the knowledge packet that were normally received.
  • the data transmission device 10 and the data reception device 20 perform data transmission by the method shown in the flowchart of FIG. 5 by using the generation / addition of the parity bucket and the distributed packet transmission.
  • the data transmission device 10 if application data to be transmitted is generated, the data is divided into packets of a fixed length, in the following description, every 30,000 bytes, and parity is calculated for every 10 data packets in the first transmission stage. Then, a parity packet is generated, and a packetization process for forming one segment with 11 packets of 10 data packets + 1 parity packet is performed (steps S1, S3).
  • step S5 In the transmission start stage, it is assumed that the previous reception was successful (YE S in step S5), and that no parity packet was added in the previous transmission (0 in step 37), and the first transmission phase (first phase) is transmitted.
  • first transmission phase first phase
  • one parity bucket is added to 10 data packets to make 11 segments into 1 packet, and 20 segments are prepared as shown in FIG. 6 (step S9A).
  • the base speed is set to 55 Mb s, which is 1.1 times the base speed of 50 Mb ps (step S11), and distributed transmission of 20 segments is started in the first transmission phase (step S11).
  • step S11 the base speed is set to 55 Mb s, which is 1.1 times the base speed of 50 Mb ps (step S11), and distributed transmission of 20 segments is started in the first transmission phase (step S11).
  • the receiving side transmits a transmission speed gap to the transmitting side together with ACK due to the successful reception. It should be noted that if the transmitting side receives the ACK, it may decide to increase the speed by itself.
  • step S7 the parity bucket is initially transmitted. Since the packet is transmitted in the first transmission phase with the data added, the process branches to YES and prepares for transmission in the second transmission phase (second phase) (step S9B).
  • second phase second phase
  • all packets in one segment are transmitted as data packets at the same transmission speed as the previous one, thus substantially increasing the speed.
  • a data packet for 20 segments is transmitted in a burst manner (step S13; SK21). Suppose that the second transmission was successful without any data loss.
  • the base transmission rate was 50 Mbps for the data packet at the first real transmission rate shown in Fig. 6, but the second transmission was possible without the parity bucket. This means that the data could be transmitted at a transmission speed of 55 Mb ps, which is twice as high.
  • the speed is increased to 60.5 Mb ps, which is 1.1 times the initial transmission speed of 55 Mb ps, and the parity bucket is processed as in the first transmission.
  • 1 bucket with 1 segment added is transmitted in a distributed manner in the first transmission phase for 20 segments (SK 21).
  • the transmission speed is maintained at 60.5 Mb ps in the subsequent second transmission phase, as in the second transmission process.
  • a 1-segment 1-data packet is transmitted and a 20-segment data packet is transmitted in a burst manner. This allows real data transmission
  • the transmission speed can be increased to 60.5 Mbps.
  • the packet group consisting of 10 data packets + 1 parity packet and 1 packet constitutes 1 segment in the 1st transmission phase while increasing the transmission rate by one step unless data loss occurs.
  • the data is composed of 1 segment and 1 data bucket and transmitted at the same transmission rate in a burst manner. Repeat the operation to increase the data transmission speed.
  • the speed-up rate can be set, for example, by 1 OM bps or by 10%, but is not particularly limited.
  • the first transmission phase (that is, the transmission phase in which a packet group consisting of 10 data buckets + 1 parity packet and 1 packet consisting of 1 segment is distributed for 20 segments) is transmitted in the first phase. If data loss occurs, one step before, in this case, as shown in Fig. 9, the procedure returns to the same procedure as the second transmission procedure and data transmission is restarted in the second transmission phase (SK twenty two ) . In other words, in the flowchart of FIG. 5, the process branches to NO in step S5, branches to YES in step S15 because the parity was transmitted last time, and performs the parity adding process in preparation for the transmission in the second transmission phase.
  • Step S17A the communication speed is reduced by one step, and the data transmission in the second transmission phase is restarted (steps S19, S13). If data loss still occurs by the transmission operation shown in FIG. 9, if the data transmission is performed one more time before, that is, the first transmission is performed at the same transmission speed as the first transmission. The transmission in the transmission phase will be performed.
  • the process branches to NO in step S5 in FIG. 5, and also branches to NO in step S15.
  • the first transmission phase is prepared (step S17B), and data is transmitted in the first transmission phase with the parity bucket added at the same transmission speed as the previous time (step S13).
  • the bucket group is divided into one segment and distributed transmission of a predetermined number of segments is performed.
  • B As shown in Fig. 10 (B), if no data loss occurs, the second transmission phase (second phase) is performed.
  • a data packet group that does not include a parity packet and is the same as the total number of packets in the first transmission phase is transmitted in bursts.
  • C If no data loss occurs in the second transmission phase, the transmission speed The data transmission in the first transmission phase shown in Fig. 10 (A) and the data transmission in the second transmission phase shown in Fig.
  • the redundancy is increased by the addition of the parity data, and the data loss on the network can be recovered by the receiving side using the parity data from the received data, and complete data loss can be achieved. Therefore, if there is no data loss in the first transmission phase, the parity packet is not included in the second transmission phase, and the retransmission is not performed in the first transmission phase.
  • the parity packet was transmitted at the transmission speed that is likely to succeed this time because it was successful last time.
  • the number of data packets transmitted at the same transmission rate can be increased by replacing Only the actual transmission speed can be increased, and as a result of all these, the throughput of data transmission can be improved.
  • the second embodiment differs from the first embodiment in the method of reducing the communication speed when the transmission fails in the first transmission phase. If the transmission fails in the first transmission phase, the transmission speed is reduced by one step. Data transmission is resumed in the same first transmission phase after the floor has been lowered. Therefore, the functional configuration of the data transmission device is the same as that shown in FIGS. 1 and 2 as in the first embodiment. However, in the case of the present embodiment, if there is a transmission failure in the previous second transmission phase, the transmission phase determination unit 15 determines to transmit in the first transmission phase at the same transmission speed. If there is a transmission failure in one transmission phase, the transmission speed is reduced by one step and the same Decide to resume data transmission in the transmission phase.
  • FIG. 11 is a flowchart of a data transmission process performed by the data transmission device according to the present embodiment.
  • the process of substantially increasing the speed when the previous data transmission succeeds is the same as that of the first embodiment.
  • the processing when the previous data transmission failed and the previous transmission phase was the second transmission phase is the same as in the first embodiment.
  • the procedure branches to NO, and also branches to NO in step S15, prepares for the first transmission phase (step S17B), and adds the parity bucket at the same transmission rate as the previous transmission phase. (Step S13).
  • the process branches to YES in step S15, and is the same as the previous time. It prepares for data transmission in the first transmission phase (step S17C), and performs processing to reduce the transmission speed by one step from the previous time and restart data transmission (steps S19, S13).
  • redundancy is increased by adding parity data, and data loss on the network can be recovered by the receiving side using parity data from the received data.
  • the chances of complete data loss can be reduced, the chances of retransmission for data loss can be reduced accordingly, and if no data loss occurs in the first transmission phase, the parity packet is transmitted as the second transmission phase.
  • the system configuration of the data transmission device of the third embodiment is the same as that of the first embodiment and the second embodiment, but has a feature in its redundant data transmission method.
  • the transmission method shown in 1 is taken.
  • the data packets to be transmitted are data packets (1) to (6), and data packets (1), (3) and (5) are segments. It is assumed that the packet group belonging to SG1 and the data packets (2), (4), (6) are the packet group belonging to segment SG2.
  • the redundant packet transmission is performed as follows. .
  • the transmission starts with the packet group of (6) arranged in a line in the transmission order. Then, the packet (1) belonging to the segment SG1 to be transmitted first is transmitted, and at this time, the parity is recorded in the parity packet PT1 for the segment SG1.
  • the first data packet (2) belonging to the segment SG2 is transmitted, and at this time, the parity is recorded in the parity packet PT2 for the segment SG2.
  • the second data packet (3) belonging to the reproduction segment SG1 is transmitted, and at this time, the parity is recorded in the parity packet PT1 for the segment SG1.
  • Figure 17 shows As described above, the second data packet (4) belonging to the segment SG2 is transmitted, and the parity is recorded in the parity bucket PT2 for the segment SG2.
  • the same transmission procedure is repeated.
  • the last data packet (5) of the segment SG 1 is transmitted, its parity is recorded in the parity packet PT1, and then, as shown in FIG.
  • the data packet (6) is transmitted and its parity is recorded in the parity bucket PT2.
  • a parity packet PT1 for the segment SG1 and a parity packet PT2 for the segment SG2 are transmitted.
  • the results of evaluating the data transmission by the two-phase slow-start transmission method of the present invention and the data transmission by the conventional slow-start congestion avoidance control transmission method according to the present invention shown in Figs. 22 to 24 are shown in the graphs of Figs. 25 and 26. Indicated.
  • the data transmission amount is the same.In the embodiment of the present invention, the speed is increased by 10 Mbps when data transmission is successful, the speed is decreased by 10 Mbps when data loss occurs, and the control is performed in the second transmission phase. Is omitted.
  • the first phase parity at 50 Mbps 1 segment 2 including 1 packet is transmitted for 20 segments (SK2111). This base transmission rate is 4.8 msec / packet.
  • the transmission of SK2221 succeeds, the speed is further increased by 10 Mbps, and as shown in Fig. 24, the first phase, 70 Mbps, and 1 segment 21 packets including the knowledge packet are 20 segments. Minute (SK 2 3 1). The transmission speed at this time is 3.4 msec / packet.
  • the speed is reduced by 1 OMbs, and the transmission is performed in the first phase at a speed of 60 Mbps.
  • Example 1 The transmission characteristics and transmission results of the data transmission method of Example 1 and the conventional data transmission method (Comparative Example 1) were measured. According to this, as shown in the graph of Example 1 in FIG. 25 and the graph of Comparative Example 1 in FIG. 26, in the one-to-one evaluation using Gigabit Ethernet (registered trademark), It was demonstrated that the data transmission according to the first embodiment can communicate with a loss rate of 43.9% less than the slow start of the first embodiment. In the case of the present invention, the packet recovery rate was 41.2%.
  • the data transmission by the two-phase slow start transmission method according to the third embodiment of the present invention and the data transmission by the conventional slow start transmission method were evaluated.
  • two computers, PC1 and PC2 were connected by Gigabit Ethernet (registered trademark), and PC1 and PC2 were operated with NIST Net (trade name) software of a delay generation emulator.
  • the delay time required for data transmission from PC 1 to PC 2 was set to 100 msec. This results in a round-trip delay of 200 ms, which is equivalent to the delay of a long-distance network between Japan and the United States.
  • This NIST Net is "The Linux-based Network Emulation Tool NIST Net" and was obtained from http://snad.ncsl.nist.gov/itg/nistnet/.
  • the receiver requests the transmitter to increase the transmission speed by 0.8, otherwise it increases the speed by 1.2 times. Decided to request.
  • the redundancy is increased by the addition of parity data, and the data loss on the network can be recovered by the receiving side using the parity data from the received data, so that the opportunity for complete data loss can be reduced.
  • the chances of retransmission for data loss can be reduced, and if no data loss occurs in the first transmission phase, the parity packet is not included in the second transmission phase, and the total buckets in the first transmission phase
  • the parity packet is replaced with data packets at the transmission rate that is likely to succeed this time because it succeeded the previous time.
  • the number of data packets transmitted at the same transmission speed can be increased, and the actual transmission speed is increased accordingly. Can and this, as these overall results, the throughput of data transmission can be improved.

Abstract

A method for transmitting data and a data transmitter in which (a) distributed transmission of a specified number of segments, each consisting of a predetermined number of packet groups added with a parity packet as a first phase at each transmission rate, is performed, (b) distributed transmission of a specified number of segments, each consisting of a predetermined number of data packet groups not including a parity packet as a second phase, is performed if data loss does not occur, (c) data transmission of first phase and data transmission of second phase are repeated by increasing the data transmission rate by one stage if data loss does not occur even in the second phase, and (d) throughput of data communication is enhanced by performing an operation for returning to the immediately previous phase upon occurrence of data loss during any one data transmission interval and for retransmitting the data so that the data can be recovered on the receiving side thereby minimizing missing of data packets.

Description

明 細 書 データ送信方法及びデータ送信装置 技術分野  Description Data transmission method and data transmission device
本発明は、 データ送信方法及びデータ送信装置に関する。 背景技術  The present invention relates to a data transmission method and a data transmission device. Background art
特許文献 1、 特許文献 2に開示されているように、 従来、 情報ネットヮ ーク上でデータをパケット送信する際に利用するプロ トコル、 例えば T C P ( iransmi ssion Control Protocol 、 X 丄、 P ( Xpress Transport Protocol) では高度なウィンドウ制御技術を採用し、 ネットワークの輻輳 回避のためにスロースタートと呼ばれる通信アルゴリズムを利用している 。 このスロースター トは、 基底ウィンドウサイズで送信を開始し、 受信側 から受信成功の A C Kが届く と一定割合でウインドウサイズを大きく して ゆき、 受信 A C Kが届かなくなったり、 受信失敗の N A C Kが届いたりす れば、 (a ) 送信ウィンドウサイズを基底ウィンドウサイズに戻して最後 に受信成功の A C Kを受けたバケツ トの次のバケツトから送信再開し、 同 じょうにして徐々にウィンドウサイズを上げていく、 あるいは (b ) 送信 ウィンドウサイズを最後に受信成功の A C Kを受けときのウインドウサイ ズまで小さく し、 最後に受信成功の A C Kを受けたバケツ トの次のバケツ トから送信再開し、 同じようにして徐々にウインドウサイズを上げていく という通信アルゴリズムである。 このような従来のスロースタート方式の ウィンドウ制御を利用する通信アルゴリズムは、 送信側の望むデータのす ベてをネットワーク環境が許容する範囲内で比較的に速く、 かつ確実に受 信側に送信できる特長がある。 As disclosed in Patent Literature 1 and Patent Literature 2, protocols conventionally used for transmitting data packets over an information network, for example, TCP (iranssion Control Protocol, XII, P (Xpress Transport Protocol) employs advanced window control technology and uses a communication algorithm called slow start to avoid network congestion.This slow start starts transmission at the base window size and receives data from the receiver. When a successful ACK arrives, the window size is increased at a fixed rate. If a received ACK does not arrive or a reception failure NACK arrives, (a) the transmission window size is returned to the base window size and finally Transmission resumes from the bucket following the bucket that received the ACK for successful reception, and the window size gradually increases in the same manner. Or (b) reduce the transmission window size to the window size at the time of receiving the ACK of the last successful reception, restart transmission from the bucket next to the bucket of the last ACK of the successful reception, A communication algorithm that gradually increases the window size in the same way.The communication algorithm that uses the conventional slow-start window control allows the network environment to accept all the data desired by the sender. Within a range that is relatively fast and reliable There is a feature that can be transmitted to the receiving side.
ところが、 近年の高速化された情報ネットワークを利用して送信される データ量は膨大なものになってきているが、 受信側の P C (パーソナルコ ンピュータ) 内のソフ トウェアの受信データ処理速度がネットワークの高 速な通信速度に追従できず、 ネットワーク上ではなくて、 この受信側での 受信データ処理速度がポトルネックとなって多くのデータバケツトが紛失 してしまうとレヽぅ状況が発生するようになつている。  However, the amount of data transmitted using recent information networks has become enormous, but the processing speed of the data received by the software in the receiving PC (personal computer) has increased. If the high-speed communication speed of the receiver cannot follow, and the data processing speed of the receiving side becomes a bottleneck rather than on the network, a lot of data buckets will be lost, and a situation will occur. I'm familiar.
このようなネッ トワーク通信環境の下で、 従来のスロースタート方式の 通信アルゴリズムでデータバケツトを送信する場合、 ネッ トワーク上では なくて受信側の装置内のソフトウエアの処理速度の問題で紛失するバケツ トが多くなり、 パケット再送の頻度が大きくなるが、 スロースタート方式 の通信アルゴリ ズムでは、 受信側でのデータ紛失発生から送信側が上述し た (a ) あるいは (b ) の手順で紛失した最初のパケッ トから送信再開す るまでのラウンドトリップ時間が長くかかる。 そのため、 スロースタート 方式のウィンドウ制御技術を採用しても、 パケッ ト紛失の頻度が多く発生 するほどに一定量のデータ送信にかかる伝送時間が長くなる問題点があつ た。  When a data bucket is transmitted using the conventional slow start communication algorithm in such a network communication environment, the data bucket is lost not because of the network but because of the processing speed of the software in the receiving device. The number of buckets increases and the frequency of packet retransmission increases.However, in the slow-start communication algorithm, the data lost on the receiving side causes the transmitting side to lose the data in the procedure (a) or (b) above. It takes a long round-trip time from the start of the packet to the restart of transmission. Therefore, even if the window control technology of the slow start method is adopted, there is a problem that the transmission time for transmitting a certain amount of data becomes longer as the frequency of packet loss increases.
データ伝送の信頼性の保証する技術として従来からパリティを利用する ものが知られている。 特許文献 3には一定個数 k一 1のバケツト毎にその うちの 1つが消失しても残りのバケツトのデータから復元できるデータを 持たせたパリティバケツ トを付加し、 k個のパケッ トデータを 1セグメン トとして、 1セグメントずつ送信する技術が開示されている。 しかしなが ら、 この伝送方式では、 2つ以上の連続するパケットが消失することが多 い受信側コンピュータ内でのデータ消失事象一コンピュータ側のソフ トゥ アの処理速度が通信速度に追いつかず、 コンピュータ内で一塊のデータ が喪失してしまう事象一には対応できず、 データ再送信要求が頻発し、 実 質的な通信速度の向上が図れない問題点があった。 As a technique for guaranteeing the reliability of data transmission, a technique using parity has been known. Patent Literature 3 adds a parity bucket having data that can be restored from the data of the remaining buckets even if one of them is lost for each of a fixed number of k−1 buckets, and the k packet data is reduced to 1 packet. As a segment, a technology for transmitting one segment at a time is disclosed. However, in this transmission method, two or more consecutive packets are often lost, so the data loss event in the receiving computer, the processing speed of the software on the computer side cannot keep up with the communication speed, Mass of data in computer However, there was a problem that it was not possible to cope with the event that the data was lost, data retransmission requests frequently occurred, and it was not possible to improve the actual communication speed.
またバケツト伝送において、 パケッ ト網に輻輳が発生しない間は通信速 度を上昇させ、 輻輳の恐れがあれば通信速度を低減させる通信速度制御技 術が特許文献 4に開示されている。 この通信速度制御は実質的な通信速度 向上が期待できるが、 所定個数のパケットを 1セグメントとして、 その送 信が成功すれば 1段階通信速度を上げるようにしており、 最初には (所定 個数一 1個) のデータバケツ トと 1個のパリティバケツ トを 1セグメント として送信し、 それに成功すればパリティバケツ トを付けずに所定個数の データバケツ トを 1セグメントとして送信することはしておらず、 同じ通 信速度でもデータバケツ ト数を増加させることで実質的に通信速度を上げ る制御は開示していない。  Also, in packet transmission, Patent Document 4 discloses a communication speed control technology that increases the communication speed while congestion does not occur in a packet network and reduces the communication speed if there is a possibility of congestion. Although this communication speed control can be expected to substantially improve the communication speed, a predetermined number of packets is regarded as one segment, and if the transmission is successful, the communication speed is increased by one step. (One) and one parity bucket are transmitted as one segment, and if successful, a predetermined number of data buckets are transmitted as one segment without a parity bucket. No control is disclosed for increasing the number of data buckets at the same communication speed to substantially increase the communication speed.
さらにデータ転送の信頼性を高める技術として特許文献 5には複数の伝 送路を利用し、 例えば (N— 1 ) 個のデータパケットと 1個のパリティパ ケットとで N個のバケツ トを 1セグメントとし、 M本の伝送路で Mセグメ ント分のバケツトをマトリタス的に送信する技術が開示されている。 これ により、 いずれかの伝送路の 1本で 1個のバケツ トが消失しても受信側で 復元可能であり、 再送の必要がなくて通信の信頼性と高速性を可能にして いる。 しかしながら、 複数の伝送路を利用するものであるため、 設備コス トがかさむこと、 受信側での一連のデータの復元処理が複雑である問題点 カ あつた。  As a technique for further improving the reliability of data transfer, Patent Document 5 utilizes a plurality of transmission paths. For example, (N-1) data packets and one parity packet are used to divide N buckets into one segment. A technique for transmitting buckets of M segments in a matrix manner over M transmission paths is disclosed. As a result, even if one bucket is lost on one of the transmission lines, the packet can be restored on the receiving side, eliminating the need for retransmission and enabling high reliability and high-speed communication. However, since multiple transmission lines are used, equipment costs are increased, and the process of restoring a series of data on the receiving side is complicated.
特許文献 1 特開 2 0 0 3— 2 3 4 4 8号公報  Patent Document 1 Japanese Patent Application Publication No.
特許文献 2 特開 2 0 0 3— 7 8 5 5 6号公報  Patent Document 2 Japanese Patent Application Laid-Open No. 2003-7785556
特許文献 3 特開平 2— 2 1 9 3 3 6号公報  Patent Literature 3 Japanese Patent Application Laid-Open No. 2-2199336
特許文献 4 特開平 7— 6 6 8 3 7号公報 特許文献 5 :特開 200 2— 3 68 1 1号公報 発明の開示 Patent Document 4 Japanese Patent Application Laid-Open No. 7-66683 Patent Document 5: Japanese Patent Application Laid-Open No. 2002-36811 Disclosure of the Invention
本発明は、 上記のような従来の技術的課題に鑑みてなされたもので、 受 信側でのデータ復旧を可能にして可能な限りデータパケッ トの紛失を少な く し、 総合的なデータ通信速度スループットを向上できるデータ送信技術 を提供することを目的とする。  The present invention has been made in view of the above-mentioned conventional technical problems, and enables data recovery on the receiving side to minimize loss of data packets as much as possible, thereby achieving comprehensive data communication. An object of the present invention is to provide a data transmission technology capable of improving speed throughput.
本発明の 1つの特徴はデータ送信方法にあり、 このデータ送信方法は、 単一の伝送路上で、 (a) 所定の伝送速度で、 パリティパケットを付加し た一定個数 N (Nは任意の自然数) のパケッ ト群を 1セグメントにして所 定数 M (Mは任意の自然数) のセグメントを、 各セグメントに属するパケ ットが一連の M個のバケツ トで成る送信パケット群それぞれに 1個ずつ含 まれるようにして当該送信パケッ ト群ごとに N群分、 合計 NXMのバケツ トを分散送信する第 1送信フェーズのデータ送信を行うステップと、 (b ) 前記 (a) の第 1送信フェーズの送信でデータ損失が発生しなければ、 前記 (a) の第 1送信フェーズの伝送速度と同じ伝送速度で、 パリティパ ケットを含まず、 かつ前記第 1送信フェーズと同じ総パケット数のデータ バケツト群をパースト的に送信する第 2送信フェーズのデータ送信を行う ステップと、 (c) 前記 (b) の第 2送信フェーズの送信でもデータ損失 が発生しなければ伝送速度を予め設定されている 1段階アップして、 前記 第 1送信フェーズのデータ送信を行うステップと、 (d) 前記 (c) の第 1送信フェーズのデータ送信でデータ損失が発生しなければ、 前記第 2送 信フェーズのデータ送信を行うステップと、 (e) 以降、 データ損失が発 生しない限り伝送速度を所定の 1段階ずつアップしながらの前記 (c) の 第 1送信フェーズのデータ送信と、 前記 (d) の第 2送信フェーズのデー タ送信を繰り返すステップと、 ( f ) いずれかのデータ伝送期間にデータ 損失が発生すれば、 その 1回前のフェーズに戻してデータ送信を再開する ステップとを有する。 One feature of the present invention resides in a data transmission method. This data transmission method includes: (a) a fixed number N (N is an arbitrary natural number) to which a parity packet is added at a predetermined transmission rate at a predetermined transmission rate; The packet group of (1) is defined as one segment, and a segment of a constant M (M is an arbitrary natural number) is included in each transmitted packet group in which the packets belonging to each segment consist of a series of M packets. Performing a data transmission in a first transmission phase for distributing and transmitting a total of NXM buckets for N groups for each transmission packet group in a rare case; and (b) performing the first transmission phase in (a). If no data loss occurs during transmission, a data bucket group having the same transmission rate as that of the first transmission phase in (a) but not including parity packets and having the same total number of packets as the first transmission phase is transmitted. Pa And (c) increasing the transmission rate by one preset level if no data loss occurs even in the transmission of the second transmission phase of (b). Performing data transmission in the first transmission phase; and (d) performing data transmission in the second transmission phase if no data loss occurs in the data transmission in the first transmission phase in (c). ( E ) transmitting the data in the first transmission phase of (c) while increasing the transmission rate by a predetermined step unless a data loss occurs after ( e ); and transmitting the second transmission of (d) Phase Day (F) If data loss occurs during any data transmission period, return to the previous phase and restart data transmission.
本発明の他の特徴はデータ送信方法にあり、 このデータ送信方法は、 単 —の伝送路上で、 (a ) 所定の伝送速度で、 パリティパケットを付加した —定個数 Nのバケツ ト群を 1セグメントにして所定数 Mのセグメントを、 各セグメントに属するバケツトが一連の M個のバケツトで成る 信パケッ ト群それぞれに 1個ずつ含まれるようにして当該送信パケット群ごとに N 群分、 合計 N X Mのバケツ トを分散送信する第 1送信フェーズのデータ送 信を行うステップと、 (b ) 前記 (a ) の第 1送信フェーズの送信でデー タ損失が発生しなければ、 前記 (a ) の第 1送信フェーズの伝送速度と同 じ伝送速度で、 パリティパケットを含まず、 かつ、 前記第 1送信フェーズ と同じ総バケツ ト数のデータパケット群をバースト的に送信する第 2送信 フェーズのデータ送信を行うステップと、 (c ) 前記 (b ) の第 2送信フ ユーズの送信でもデータ損失が発生しなければ伝送速度を予め設定されて いる 1段階アップして、 前記第 1送信フェーズのデータ送信を行うステツ プと、 (d ) 前記 (c ) の第 1送信フェーズのデータ送信でデータ損失が 発生しなければ、 前記第 2送信フェーズのデータ送信を行うステップと、 ( e ) 以降、 データ損失が発生しない限り伝送速度を所定の 1段階ずっァ ップしながらの前記 (c ) の第 1送信フェーズのデータ送信と、 前記 (d ) の第 2送信フェーズのデータ送信を繰り返すステップと、 ( f ) いずれ かのデータ伝送期間にデータ損失が発生すれば、 それが第 2送信フェーズ のデータ送信であった場合には同じ伝送速度の第 1送信フェーズに戻して データ送信を再開するステップと、 (g ) いずれかのデータ伝送期間にデ ータ損失が発生すれば、 それが第 1送信フェーズのデータ送信であった場 合には伝送速度を所定の 1段階ダウンし、 かつ、 第 1送信フェーズでデー タ送信を再開するステップとを有する。 Another feature of the present invention resides in a data transmission method. This data transmission method comprises the steps of: (a) adding a parity packet at a predetermined transmission rate to a fixed number N of bucket groups on a single transmission path; A predetermined number M of segments, and N packets for each transmission packet group, so that one packet is included in each of the transmission packet groups consisting of a series of M buckets belonging to each segment, for a total of NXM (B) performing data transmission in the first transmission phase for distributing and transmitting the bucket of (a), if no data loss occurs in the transmission of the first transmission phase of (a), A second transmission phase in which a data packet group having the same transmission rate as one transmission phase, including no parity packet, and having the same total number of buckets as the first transmission phase is transmitted in a burst manner. Transmitting data; and (c) increasing the transmission speed by one preset level if no data loss occurs even in the transmission of the second transmission fuse in the above (b). (D) performing data transmission in the second transmission phase if no data loss occurs in the data transmission in the first transmission phase in (c); and (e) Repeating the data transmission in the first transmission phase of the above (c) and the data transmission in the second transmission phase of the above (d) while keeping the transmission speed in a predetermined one-step increment unless data loss occurs. (F) If data loss occurs during any of the data transmission periods, if it is data transmission in the second transmission phase, return to the first transmission phase at the same transmission rate and restart data transmission. Steps and, (g) if data loss occurs in any of the data transmission period, place it is a data transmission of the first transmission phase In this case, the transmission speed is reduced by one predetermined stage, and the data transmission is restarted in the first transmission phase.
本発明のさらに他の特徴はデータ送信装置にあり、 このデータ送信装置 は、 送信すべきデータを保持するデータ保持部と、 単一の伝送路を通じて 情報ネットワークに対してデータ授受するネットワークインタフェースと 、 前記データ保持部に保持されている送信すべきデータをバケツ ト化する パケット化処理部と、 一定個 (N— 1 ) のパケット列に対してパリティパ ケットを生成するパリティバケツト生成部と、 前回のデータ送信の成功 Z 不成功の情報と前回の送信フェーズの情報とに基づき、 今回の送信フヱー ズと送信速度を決定する送信フェーズ決定部と、 前記送信フェーズ決定部 の決定した送信フェーズの情報に基づき、 前記パリティパケットを付加し た一定個数 Nのバケツト群を 1セグメントにして所定数 Mのセグメントを 生成し、 又は、 前記パリティパケッ トを含まず、 かつ前記所定数 Mのセグ メント内の総パケット数 N X Mと同じ個数のデータパケット群を生成する 送信パケット群作成部と、 前記送信フェーズ決定部の決定した送信速度で 、 前記送信パケッ ト群作成部の生成したパケット群を前記ネットワークィ ンタフヱースを介して前記情報ネットワークに送信するデータ送信処理部 とを備え、 前記送信フェーズ決定部は、 (a ) 所定の伝送速度において、 パリティバケツトを付加した一定個数 Nのパケット群を 1セグメントにし て所定数 Mのセグメントを、 各セグメントに属するパケッ トが一連の M個 のバケツ トで成る送信パケット群それぞれに 1個ずつ含まれるようにして 当該送信バケツト群ごとに N群分、 合計 N X Mのバケツ トを分散送信する 第 1送信フェーズのデータ送信を行い、 (b ) 前記 (a ) の第 1送信フエ ーズの送信でデータ損失が発生しなければ、 前記 (a ) の第 1送信フエ一 ズの伝送速度と同じ伝送速度で、 パリティパケッ トを含まず、 かつ前記第 1送信フェーズと同じ総バケツト数のデータバケツト群をバースト的に送 信する第 2送信フェーズのデータ送信を行い、 (c ) 前記 (b ) の第 2送 信フェーズの送信でもデータ損失が発生しなければ伝送速度を予め設定さ れている 1段階アップして、 前記第 1送信フェーズのデータ送信を行い、 ( d ) 前記 (c ) の第 1送信フェーズのデータ送信でデータ損失が発生し なければ、 前記第 2送信フェーズのデータ送信を行い、 (e ) 以降、 デー タ損失が発生しない限り伝送速度を所定の 1段階ずつアップしながらの前 記 (c ) の第 1送信フェーズのデータ送信と、 前記 (d ) の第 2送信フエ ーズのデータ送信を繰り返し、 ( f ) いずれかのデータ伝送期間にデータ 損失が発生すれば、 その 1回前の送信フェーズに戻してデータ送信を再開 する、 というルールに基づいて今回の送信フェーズと送信速度を決定する ことを特徴とするものである。 Still another feature of the present invention resides in a data transmission device, which includes a data holding unit that holds data to be transmitted, a network interface that sends and receives data to and from an information network through a single transmission path, A packetization processing unit configured to bucket data to be transmitted held in the data holding unit, a parity bucket generation unit configured to generate a parity packet for a fixed (N−1) packet sequence, The transmission phase determining unit that determines the current transmission phase and the transmission speed based on the information on the unsuccessful transmission and the information on the previous transmission phase, and information on the transmission phase determined by the transmission phase determining unit. Based on the above, the fixed number N of bucket groups to which the parity packet is A transmission packet group creation unit that generates or generates a data packet group that does not include the parity packet and has the same number as the total number of packets NXM in the predetermined number M of segments; and a transmission phase determination unit. A data transmission processing unit that transmits the packet group generated by the transmission packet group creation unit to the information network via the network interface at the determined transmission speed, wherein the transmission phase determination unit comprises: (a) At a given transmission rate, a fixed number N of packet groups to which a parity bucket is added is divided into one segment, and a given number M of segments are transmitted.A packet belonging to each segment consists of a series of M buckets. NXM buckets for each transmission bucket group are distributed and transmitted for N groups. (B) If no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a). Transmission rate, not including parity packets, and Data transmission in the second transmission phase, in which data bucket groups having the same total number of buckets as one transmission phase are transmitted in a burst manner, is performed. (C) Data loss also occurs in the transmission in the second transmission phase of (b) above. If not, the transmission rate is increased by a preset level to perform the data transmission in the first transmission phase, and (d) data loss occurs in the data transmission in the first transmission phase in (c). If not, the data transmission in the second transmission phase is performed. After (e), the data in the first transmission phase in the above (c) is increased while the transmission speed is increased by a predetermined step unless data loss occurs. The transmission and the data transmission of the second transmission phase (d) are repeated. (F) If data loss occurs during any of the data transmission periods, the data transmission is returned to the previous transmission phase one time before. To resume It is characterized in that the current transmission phase and transmission speed are determined based on the following rules.
本発明のさらに他の特徴はデータ送信装置にあり、 このデータ送信装置 は、 送信すべきデータを保持するデータ保持部と、 単一の伝送路を通じて 情報ネッ トワークに対してデータ授受するネットワークインタフェースと 、 前記データ保持部に保持されている送信すべきデータをバケツ ト化する パケット化処理部と、 一定個 (N— 1 ) のパケット列に対してパリティパ ケッ トを生成するパリティバケツト生成部と、 前回のデータ送信の成功ノ 不成功の情報と前回の送信フェーズの情報とに基づき、 今回の送信フエ一 ズと送信速度を決定する送信フェーズ決定部と、 前記送信フェーズ決定部 の決定した送信フェーズの情報に基づき、 前記パリティパケットを付加し た一定個数 Nのパケット群を 1セグメントにして所定数 Mのセグメントを 生成し、 又は、 前記パリティパケッ トを含まず、 かつ前記所定数 Mのセグ メント内の総パケット数 N X Mと同じ個数のデータバケツト群を生成する 送信パケット群作成部と、 前記送信フ ーズ決定部の決定した送信速度で 、 前記送信バケツト群作成部の生成したパケッ ト群を前記ネットワークィ ンタフエースを介して前記情報ネットワークに送信するデータ送信処理部 とを備え、 前記送信フェーズ決定部は、 (a ) 所定の伝送速度で、 パリテ ィパケッ トを付加した一定個数 N ( Nは任意の自然数) のパケット群を 1 セグメントにして所定数 M (Mは任意の自然数) のセグメントを、 各セグ メントに属するバケツトが一連の M個のバケツ トで成る送信パケット群そ れぞれに 1個ずつ含まれるようにして当該送信バケツト群ごとに N群分、 合計 N X Mのバケツトを分散送信する第 1送信フェーズのデータ送信を行 い、 ( b ) 前記 (a ) の第 1送信フェーズの送信でデータ損失が発生しな ければ、 前記 (a ) の第 1送信フェーズの伝送速度と同じ伝送速度で、 パ リティバケツ トを含まず、 かつ前記第 1送信フェーズと同じ総パケッ ト数 のデータバケツ ト群をパースト的に送信する第 2送信フェーズのデータ送 信を行い、 (c ) 前記 (b ) の第 2送信フェーズの送信でもデータ損失が 発生しなければ伝送速度を予め設定されている 1段階アップして、 前記第 1送信フェーズのデータ送信を行い、 (d ) 前記 (c ) の第 1送信フエ一 ズのデータ送信でデータ損失が発生しなければ、 前記第 2送信フェーズの データ送信を行い、 (e ) 以降、 データ損失が発生しない限り伝送速度を 所定の 1段階ずつアップしながらの前記 (c ) の第 1送信フェーズのデー タ送信と、 前記 (d ) の第 2送信フェーズのデータ送信を繰り返し、 ( f ) いずれかのデータ伝送期間にデータ損失が発生すれば、 それが第 2送信 フェーズのデータ送信であった場合には同じ伝送速度の第 1送信フェーズ に戻してデータ送信を再開し、 (g ) いずれかのデータ伝送期間にデータ 損失が発生すれば、 それが第 1送信フェーズのデータ送信であった場合に は伝送速度を所定の 1段階ダウンし、 かつ、 第 1送信フェーズでデータ送 信を再開する、 というルールに基づいて今回の送信フェーズと送信速度を 決定することを特徴とするものである。 図面の簡単な説明 Still another feature of the present invention resides in a data transmission device, which includes a data holding unit that holds data to be transmitted, a network interface that sends and receives data to and from an information network through a single transmission path. A packetization processing unit configured to bucket data to be transmitted held in the data holding unit, and a parity bucket generation unit configured to generate a parity packet for a fixed (N−1) packet sequence. A transmission phase determining unit that determines the current transmission phase and transmission speed based on the information on the success or failure of the previous data transmission and the information on the previous transmission phase; and the transmission determined by the transmission phase determining unit. Based on the phase information, a predetermined number M of segments, with a fixed number N of packet groups to which the parity packet is added as one segment Or a transmission packet group creation unit that does not include the parity packet and generates the same number of data bucket groups as the total number of packets NXM in the predetermined number M of segments, and the transmission packet At the transmission speed determined by the A data transmission processing unit for transmitting the packet group generated by the transmission bucket group creation unit to the information network via the network interface, wherein the transmission phase determination unit comprises: , A fixed number N (N is an arbitrary natural number) of packet groups to which a parity packet is added is defined as one segment, and a predetermined number M (M is an arbitrary natural number) of segments is formed. Data packets in the first transmission phase for distributed transmission of NXM buckets for N groups for each transmission packet group so that one packet is included in each transmission packet group consisting of (b) If no data loss occurs in the transmission of the first transmission phase of (a), the parity transmission is performed at the same transmission rate as that of the first transmission phase of (a). (C) performing data transmission in a second transmission phase for transmitting a data packet group having the same total number of packets as in the first transmission phase in a burst manner, including no packets, and (c) performing the second transmission in (b). If data loss does not occur even in the transmission of the transmission phase, the transmission rate is increased by one step set in advance, and the data transmission of the first transmission phase is performed. (D) The first transmission phase of (c) is performed. If no data loss occurs during the data transmission of the delay, the data transmission in the second transmission phase is performed. (E) Thereafter, as long as the data loss does not occur, the above (c) while increasing the transmission speed by a predetermined step at a time. ) Of the first transmission phase and the data transmission of the second transmission phase (d) are repeated. (F) If data loss occurs during any data transmission period, this is the second transmission phase. data from If it is a transmission, return to the first transmission phase at the same transmission rate and resume data transmission. (G) If data loss occurs during any of the data transmission periods, this is the data transmission in the first transmission phase. If this is the case, the transmission speed and the transmission speed are reduced based on the rule that the transmission speed is reduced by one predetermined level and data transmission is restarted in the first transmission phase. It is characterized in that it is determined. Brief Description of Drawings
図 1は、 本発明の 1つの実施の形態のデータ送信装置の機能構成を示す ブロック図。  FIG. 1 is a block diagram showing a functional configuration of a data transmission device according to one embodiment of the present invention.
図 2は、 上記実施の形態のデータ送信装置に対応するデータ受信装置の 機能構成を示すプロック図。  FIG. 2 is a block diagram showing a functional configuration of a data receiving device corresponding to the data transmitting device of the embodiment.
図 3は、 上記実施の形態におけるアプリケーションデータのバケツ ト化 処理の説明図。  FIG. 3 is an explanatory diagram of a bucketing process of application data in the embodiment.
図 4は、 上記実施の形態による複数セグメントの分散送信処理の説明図 図 5は、 上記実施の形態のデータ送信装置によるデータ送信処理のフロ ーテヤー卜。  FIG. 4 is an explanatory diagram of a distributed transmission process of a plurality of segments according to the above embodiment. FIG. 5 is a flowchart of a data transmission process by the data transmission device of the above embodiment.
図 6は、 上記実施の形態のデータ送信装置による初回、 第 1送信フエ一 ズデータ送信処理のシーケンス図。  FIG. 6 is a sequence diagram of the first and first transmission phase data transmission processing by the data transmission device of the above embodiment.
図 7は、 上記実施の形態のデータ送信装置による初回、 第 2送信フエ一 ズデータ送信処理のシーケンス図。  FIG. 7 is a sequence diagram of the first and second transmission phase data transmission processing by the data transmission device of the above embodiment.
図 8は、 上記実施の形態のデータ送信装置による 1段階速度アップした 2回目の第 1送信フェーズデータ送信処理のシーケンス図。  FIG. 8 is a sequence diagram of a second first transmission phase data transmission process in which the data transmission device according to the above embodiment has increased the speed by one step.
図 9は、 上記実施の形態のデータ送信装置による 2回目の第 1送信フエ ーズでデータ損失が発生したために 1段階速度ダウンして行う第 2送信フ ヱーズのデータ送信処理のシーケンス図。  FIG. 9 is a sequence diagram of a second transmission phase data transmission process performed by the data transmission device of the above-described embodiment at a one-step speed reduction due to a data loss occurring in the second first transmission phase.
図 1 0は、 (A ) は上記実施の形態による第 1送信フェーズデータ送信 の説明図、 (B ) は第 2送信フェーズデータ送信の説明図。  10A is an explanatory diagram of the first transmission phase data transmission according to the embodiment, and FIG. 10B is an explanatory diagram of the second transmission phase data transmission.
図 1 1は、 本発明の第 2の実施の形態のデータ送信装置によるデータ送 信処理のフローチヤ一ト。 FIG. 11 shows data transmission by the data transmission device according to the second embodiment of the present invention. Flowchart of communication processing.
図 1 2は、 本発明の第 3の実施の形態のデータ伝送方法によって送信す るデータバケツト列を示す説明図。  FIG. 12 is an explanatory diagram showing a data bucket string transmitted by the data transmission method according to the third embodiment of the present invention.
図 1 3は、 上記第 3の実施の形態によるデータ伝送方法によって送信す るデータバケツ ト列及びパリティバケツト列の説明図。  FIG. 13 is an explanatory diagram of a data bucket string and a parity bucket string transmitted by the data transmission method according to the third embodiment.
図 1 4は、 上記第 3の実施の形態によるデータ伝送の第 1段階を示す説 明図。  FIG. 14 is an explanatory diagram showing a first stage of data transmission according to the third embodiment.
図 1 5は、 上記第 3の実施の形態によるデータ伝送の第 2段階を示す説 明図。  FIG. 15 is an explanatory diagram showing a second stage of data transmission according to the third embodiment.
図 1 6は、 上記第 3の実施の形態によるデータ伝送の第 3段階を示す説 明図。  FIG. 16 is an explanatory diagram showing a third stage of data transmission according to the third embodiment.
図 1 7は、 上記第 3の実施の形態によるデータ伝送の第 4段階を示す説 明図。  FIG. 17 is an explanatory diagram showing a fourth stage of data transmission according to the third embodiment.
図 1 8は、 上記第 3の実施の形態によるデータ伝送の第 5段階を示す説 明図。  FIG. 18 is an explanatory diagram showing a fifth stage of the data transmission according to the third embodiment.
図 1 9は、 上記第 3の実施の形態によるデータ伝送の第 6段階を示す説 明図。  FIG. 19 is an explanatory diagram showing a sixth stage of data transmission according to the third embodiment.
図 2 0は、 上記第 3の実施の形態によるデータ伝送の第 7段階を示す説 明図。  FIG. 20 is an explanatory diagram showing a seventh stage of data transmission according to the third embodiment.
図 2 1は、 上記第 3の実施の形態によるデータ伝送の第 8段階を示す説 明図。  FIG. 21 is an explanatory diagram showing an eighth stage of data transmission according to the third embodiment.
図 2 2は、 本発明の実施例 1のデータ送信方法の基底 5 O M b p sの速 度、 第 1送信フェーズでのデータ送信のシーケンス図。  FIG. 22 is a sequence diagram of data transmission in the first transmission phase at a speed of the base 5 O M bps in the data transmission method according to the first embodiment of the present invention.
図 2 3は、 上記実施例 1のデータ送信方法の 6 0 M b p sの速度、 第 1 送信フェーズでのデータ送信のシーケンス図。 図 2 4は、 上記実施例 1のデータ送信方法の 7 0 M b p sの速度、 第 1 送信フェーズでのデータ送信のシーケンス図。 FIG. 23 is a sequence diagram of data transmission in the first transmission phase at a speed of 60 Mbps in the data transmission method of the first embodiment. FIG. 24 is a sequence diagram of data transmission in the first transmission phase at a speed of 70 Mbps in the data transmission method according to the first embodiment.
図 2 5は、 上記実施例 1によるデータ送信特性のグラフ。  FIG. 25 is a graph of data transmission characteristics according to the first embodiment.
図 2 6は、 比較例 1によるデータ送信特性のグラフ。  Figure 26 is a graph of data transmission characteristics according to Comparative Example 1.
図 2 7は、 本発明の実施例 2のデータ伝送方法によるパケット受信速度 の測定結果のグラフ。  FIG. 27 is a graph of a measurement result of a packet reception speed by the data transmission method according to the second embodiment of the present invention.
図 2 8は、 上記実施例 2によるバケツト損失率の測定結果のグラフ。 図 2 9は、 比較例 2のデータ伝送方法によるバケツト受信速度の測定結 果のグラフ。  FIG. 28 is a graph of the measurement result of the bucket loss rate according to the second embodiment. Fig. 29 is a graph of the results of measuring the bucket receiving speed by the data transmission method of Comparative Example 2.
図 3 0は、 比較例 2によるパケット損失率の測定結果のグラフ。 発明を実施するための最良の形態  FIG. 30 is a graph of a measurement result of a packet loss rate according to Comparative Example 2. BEST MODE FOR CARRYING OUT THE INVENTION
以下、 本発明の実施の形態を図に基づいて詳説する。 図 1は本発明の第 1の実施の形態のデータ送信装置 1 0の機能構成を示し、 図 2はこのデー タ送信装置 1 0の送信データを受信するデータ受信装置 2 0の機能構成を 示している。  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a functional configuration of a data transmitting apparatus 10 according to a first embodiment of the present invention, and FIG. 2 shows a functional configuration of a data receiving apparatus 20 that receives transmission data of the data transmitting apparatus 10. ing.
データ送信装置 1 0、 データ受信装置 2 0は共にデータパケッ ト通信機 能を備えたコンピュータにソフトウエアとして組み込まれるものであり、 また通常は送受信両機能を同時に備えるものであるが、 ここでは説明の便 宜上、 送信側と受信側に分けて示してある。  The data transmitting device 10 and the data receiving device 20 are both incorporated as software in a computer having a data packet communication function, and usually have both transmitting and receiving functions simultaneously. For convenience, it is shown separately for the transmitting side and the receiving side.
データ送信装置 1 0の機能構成としては、 大容量データを保存する H D D、 一時的にデータを保存する R AMメモリを含めたデータ保存部 1 1、 このデータ保存部 1 1から送信すべきデータを取り出してバケツ ト化処理 をするバケツト化処理部 1 2、 バケツト化されたデータの一連のパケット 列に対して、 必要に応じてパリティパケットを生成するパリティ付加処理 部 1 3、 データ伝送状況に応じて所定のルールに基づいて送信速度と、 第 1送ィ¾フェーズ (first phase) 又 fま第 2送 lフェーズ (second phase) の送信フ ーズを決定する送信フェーズ決定部 1 4、 送信フェーズ決定部 14の指示する送信速度、 送信フェーズでバケツト化されたデータを送信 するデータ送信処理部 1 5、 外部からのデータを受信するデータ受信処理 部 1 6、 例えば LANボード、 L ANアダプタとルータを含み、 ネットヮ ーク 30とのインタフェースとして働くハードウェアのネッ トワークイン タフエース (NW I /F) 1 7を備えている。 The functional configuration of the data transmission device 10 includes an HDD that stores large amounts of data, a data storage unit 11 that includes a RAM memory that temporarily stores data, and data to be transmitted from this data storage unit 11. A bucketing processing unit 12 that takes out and performs a bucketing process, a parity adding process that generates a parity packet as necessary for a series of packet sequences of the bucketed data. Part 13.Determine the transmission rate and the transmission phase of the first transmission phase (first phase) or the second transmission phase (second phase) based on a predetermined rule according to the data transmission situation The transmission phase deciding unit 14, the transmission speed indicated by the transmission phase deciding unit 14, the data transmission processing unit 15 for transmitting the data bucketed in the transmission phase, the data reception processing unit 16 for receiving external data, For example, a hardware network interface (NW I / F) 17 including a LAN board, a LAN adapter, and a router and serving as an interface with the network 30 is provided.
他方、 データ受信装置 20の機能構成としては、 データ送信装置 1 0か らネッ トワーク 30を通じて送られてくるデータの受信処理をするデータ 受信処理部 2 1、 受信データを一時的に保存するメモリ 2 2、 受信データ のパリティチェックを実施するパリティチェック部 23、 データエラーが 発生している場合に受信データ列に付加されているパリティバケツトを利 用してデータ復旧を試みるデータ復旧処理部 24、 メモリ 22の状態、 デ ータ復旧処理部 24のデータ復旧の成否に基づきデータ受信の成否を判定 する通信状態判定部 25、 通信状態判定部 2 5のデータ受信の成否判定結 果をデータ受信の ACK, NAK信号を送出するデータ送信処理部 2 6、 そして、 送信側と同様に例えば LANボード、 LANアダプタとルータを 含み、 ネットワーク 30とのインタフェースとして働くハードウエアのネ ットワークインタフェース (NW I /F) 2 7を備えている。  On the other hand, the functional configuration of the data receiving device 20 includes a data receiving processing unit 21 for receiving data transmitted from the data transmitting device 10 through the network 30 and a memory 2 for temporarily storing received data. 2.Parity check unit 23 that performs parity check on received data 23.Data recovery processing unit 24 that attempts to recover data using a parity bucket added to the received data sequence when a data error occurs. The communication status determination unit 25, which determines the success or failure of the data reception based on the status of the memory 22 and the success or failure of the data recovery of the data recovery processing unit 24, and the result of the data reception success / failure determination by the communication status determination unit 25, which is based on A data transmission processing unit 26 for transmitting ACK and NAK signals, and also includes, for example, a LAN board, a LAN adapter and a router as well as the transmission side, and an interface with the network 30. It is equipped with a hardware of Netw network interface (NW I / F) 2 7 which acts as a face.
ネッ トワーク 30は LAN、 公衆通信回線網、 インターネット、 あるい はこれらを統合的に利用する情報ネットワークであり、 データパケッ トの 伝送が可能なものである。  The network 30 is a LAN, a public communication network, the Internet, or an information network that uses these in an integrated manner, and is capable of transmitting data packets.
次に、 上記構成のデータ送信装置 1 0、 データ受信装置 20間のデータ 通信処理について説明する。 まず本実施の形態で利用するデータバケツト通信について説明する。 図 3に示すようにアプリケーションデータを一定長のパケット p k 1, k 2, …に分割し、 一定個数 mのパケッ ト p k 1 ~ p k mに対してすべての パケッ トの排他的論理和 (X O R ) を計算し、 その計算結果をパリティパ ケッ ト parity packetとする。 そしてこの所定個数のデータパケッ トと 1 個のパリティバケツトで 1セグメント S Gとする。 Next, a data communication process between the data transmitting device 10 and the data receiving device 20 having the above configuration will be described. First, the data bucket communication used in the present embodiment will be described. As shown in Fig. 3, the application data is divided into fixed-length packets pk1, k2, ..., and the exclusive-OR (XOR) of all packets for a fixed number m of packets pk1 to pkm is calculated. Calculation is performed, and the calculation result is used as a parity packet. The predetermined number of data packets and one parity bucket constitute one segment SG.
第 1送信フェーズ (first phase) のデータ送信に際しては、 所定の N セグメントのパケッ ト群を分散送信する。 図 4は 3バケツト + 1パリティ パケッ トで 1セグメントをなす 2セグメント S G 1, S G 2のパケット群 を送信する場合について例示している。 この場合、 セグメント S G 1の最 初のパケット (1 ) 、 セグメント S G 2の最初のパケッ ト (4 ) 、 セグメ ント S G 1の 2番目のパケット (2 ) 、 セグメント S G 2の 2番目のパケ ッ ト (5 ) 、 …と順に送信し、 最後に各セグメントのパリティパケット P T 1 , P T 2を送信することを示している。  In data transmission in the first transmission phase (first phase), a predetermined number of N-segment packets are transmitted in a distributed manner. Fig. 4 shows an example of transmitting a 2-segment SG1, SG2 packet group, which forms one segment with three buckets + one parity packet. In this case, the first packet (1) of segment SG1, the first packet (4) of segment SG2, the second packet (2) of segment SG1, and the second packet of segment SG2 (5),... Are transmitted in order, and finally the parity packets PT 1 and PT 2 of each segment are transmitted.
このようにデータバケツ トの一定個数 mで成るバケツ ト群ごとにパリテ ィパケットを生成し、 そのパリティバケツトを含めた m + 1のパケットで 成るバケツト群を 1セグメントして送信することにより、 そのバケツト群 のいずれか 1つバケツトが喪失したとしても、 受信側では正常に受信され た残りのバケツト群とノ リティパケットから喪失したパケッ トのデータを 正しく復元することができる。  In this manner, a parity packet is generated for each bucket group consisting of a fixed number m of data buckets, and the bucket group consisting of m + 1 packets including the parity bucket is transmitted as one segment, thereby transmitting the packet. Even if one of the bucket groups is lost, the receiving side can correctly recover the remaining bucket groups and the lost packet data from the knowledge packet that were normally received.
また、 通常、 データ通信ではバースト的にバケツト喪失が発生する事象 が多いが、 N ( = m + 1 ) 個で 1セグメントをなすパケット群を Mセグメ ント分ずつ分散送信すれば、 複数のセグメントそれぞれに属するバケツト 群がそのパースト的な期間内に送信される可能性が高く、 受信側ではセグ メントごとに上述したパリティパケットを利用したデータ復旧ができ、 回 復できるパケッ トが多くなり、 その分、 パケット喪失を少なくでき、 確実 なデータ伝送が可能であるメリットがある。 Normally, bucket loss often occurs in bursts in data communication.However, if N (= m + 1) packets forming one segment are distributed and transmitted by M segments, then multiple segments are lost. There is a high possibility that buckets belonging to the packet group will be transmitted within that bursty period, and the receiving side can recover data using the parity packet described above for each segment, and The number of packets that can be recovered is increased, so that there is an advantage that packet loss can be reduced and reliable data transmission is possible.
データ送信装置 1 0、 データ受信装置 20間では、 このようなパリティ バケツトの生成 ·付加とバケツト分散送信を利用し、 図 5のフローチヤ一 トに示す方法でデータ伝送を実行する。  The data transmission device 10 and the data reception device 20 perform data transmission by the method shown in the flowchart of FIG. 5 by using the generation / addition of the parity bucket and the distributed packet transmission.
<最初の送信手順 >  <First transmission procedure>
データ送信装置 1 0では、 送信すべきアプリケーションデータが生成さ れれば、 一定長、 以下の説明では 30000パイ トごとのパケットに分割 し、 最初の送信段階では 1 0データパケットごとにパリティを計算し、 パ リティパケッ トを生成し、 1 0データパケッ ト + 1パリティパケッ トの 1 1パケッ トで 1セグメントを構成するパケット化処理を行う (ステップ S 1, S 3) 。  In the data transmission device 10, if application data to be transmitted is generated, the data is divided into packets of a fixed length, in the following description, every 30,000 bytes, and parity is calculated for every 10 data packets in the first transmission stage. Then, a parity packet is generated, and a packetization process for forming one segment with 11 packets of 10 data packets + 1 parity packet is performed (steps S1, S3).
送信開始段階では、 前回受信成功とみなし (ステップ S 5で YE S) 、 また前回送信ではパリティパケットを付加しなかったものとし (ステップ 3 7で 0) 、 第 1送信フェーズ (first phase) の送信準備として、 1 0データパケットに対して 1パリティバケツトを付加し、 1 1バケツ ト 1 セグメントにし、 これを図 6に示すように、 20セグメント用意する (ス テツプ S 9 A) 。  In the transmission start stage, it is assumed that the previous reception was successful (YE S in step S5), and that no parity packet was added in the previous transmission (0 in step 37), and the first transmission phase (first phase) is transmitted. As a preparation, one parity bucket is added to 10 data packets to make 11 segments into 1 packet, and 20 segments are prepared as shown in FIG. 6 (step S9A).
続いて、 最初の送信 S K 1 1では基底速度 50Mb p sの 1. 1倍の 5 5 Mb sに設定し (ステップ S 1 1 ) 、 第 1送信フェーズで 20セグメ ントの分散送信を開始する (ステップ S 1 3) 。  Subsequently, in the first transmission SK11, the base speed is set to 55 Mb s, which is 1.1 times the base speed of 50 Mb ps (step S11), and distributed transmission of 20 segments is started in the first transmission phase (step S11). S1 3).
この最初の送信でデータ損失がなくて送信成功であったとする。 ここで 、 受信側はこの受信成功によつて送信速度ァップを送信側に A C Kと共に 送信する。 なお、 送信側が ACKを受信すれば自身で速度アップを判断し てもよレヽ。 <第 2回目の送信処理 > It is assumed that the first transmission has been successfully transmitted without data loss. Here, the receiving side transmits a transmission speed gap to the transmitting side together with ACK due to the successful reception. It should be noted that if the transmitting side receives the ACK, it may decide to increase the speed by itself. <Second transmission process>
送信側では、 受信成功を判断すれば、 第 2回目の送信処理に入り、 初回 の送信に対して受信成功したのでステップ S 5で YE Sに分岐し、 ステツ プ S 7では初回にはパリティバケツ トを付加して第 1送信フェーズで送信 したので YE Sに分岐し、 第 2送信フェーズ (second phase) の送信準備 をする (ステップ S 9 B) 。 つまり、 図 7に示すように、 第 2回目の送信 では、 パリティパケットを付加せずに、 1セグメント 1 1パケット全部を データパケットにして前回と同じ送信速度で送信し、 実質速度アップを図 る。 そして、 20セグメント分のデータパケットをパース ト的に送信する (ステップ S 1 3 ; SK2 1) 。 この第 2回目の送信でもデータ損失がな くて送信成功であったとする。  On the transmitting side, if it is determined that the reception was successful, the second transmission process is started. Since the reception was successful for the first transmission, the process branches to YES in step S5. In step S7, the parity bucket is initially transmitted. Since the packet is transmitted in the first transmission phase with the data added, the process branches to YES and prepares for transmission in the second transmission phase (second phase) (step S9B). In other words, as shown in Fig. 7, in the second transmission, without adding a parity packet, all packets in one segment are transmitted as data packets at the same transmission speed as the previous one, thus substantially increasing the speed. . Then, a data packet for 20 segments is transmitted in a burst manner (step S13; SK21). Suppose that the second transmission was successful without any data loss.
これによつて、 基底送信速度が図 6に示した初回の実質送信速度がデー タパケッ トについては 50Mb p sであったものが、 第 2回目にはパリテ ィバケツ トなしで送信できたために 1. 1倍の 5 5Mb p sの送信速度で データ送信できたことになる。  As a result, the base transmission rate was 50 Mbps for the data packet at the first real transmission rate shown in Fig. 6, but the second transmission was possible without the parity bucket. This means that the data could be transmitted at a transmission speed of 55 Mb ps, which is twice as high.
く第 3回目の送信処理 >  3rd transmission process>
第 3回目の送信処理では、 図 8に示すように、 初回の送信速度 5 5 Mb p sに対してその 1. 1倍である 60. 5Mb p sに速度アップして、 初 回と同様にパリティバケツ トを付加した 1セグメント 1 1バケツ トを 20 セグメント分、 第 1送信フェーズで分散送信する (SK 2 1) 。  In the third transmission process, as shown in Fig. 8, the speed is increased to 60.5 Mb ps, which is 1.1 times the initial transmission speed of 55 Mb ps, and the parity bucket is processed as in the first transmission. 1 bucket with 1 segment added is transmitted in a distributed manner in the first transmission phase for 20 segments (SK 21).
この速度ァップした第 1送信フェーズのデータ送信でもデータ損失が発 生しない場合、 続く第 2送信フェーズでは、 第 2回目の送信処理の場合と 同様に、 送信速度は 6 0. 5 Mb p sに維持しながら、 パリティパケット を付加せずに、 1セグメント 1 1データパケットにして 20セグメント分 のデータパケッ トをパース ト的に送信する。 これによつて、 実質データ伝 送速度も 6 0 . 5 M b p sに高速化できることになる。 If no data loss occurs in the data transmission in the first transmission phase with this speed-up, the transmission speed is maintained at 60.5 Mb ps in the subsequent second transmission phase, as in the second transmission process. While adding a parity packet, a 1-segment 1-data packet is transmitted and a 20-segment data packet is transmitted in a burst manner. This allows real data transmission The transmission speed can be increased to 60.5 Mbps.
以下、 同様にして、 データ損失が発生しない限り、 伝送速度を 1段階ず っァップさせつつ、 第 1送信フェーズでは 1 0データパケット + 1パリテ ィパケッ トの 1 1パケットで 1セグメントを構成するバケツト群を所定セ グメント分だけ分散送信し、 その第 1送信フェーズの送信に成功すれば、 続く第 2送信フェーズでは 1セグメント 1 1データバケツトの構成にして 同じ送信速度でパースト的に送信し、 実質データ伝送速度も高める操作を 繰り返す。 なお、 速度アップ率については、 例えば 1 O M b p sずつ、 あ るいは 1 0 %ずっといつた設定が可能であるが、 特に限定されるものでは ない。  In the same manner, the packet group consisting of 10 data packets + 1 parity packet and 1 packet constitutes 1 segment in the 1st transmission phase while increasing the transmission rate by one step unless data loss occurs. Is transmitted in a predetermined segment, and if the transmission in the first transmission phase succeeds, in the subsequent second transmission phase, the data is composed of 1 segment and 1 data bucket and transmitted at the same transmission rate in a burst manner. Repeat the operation to increase the data transmission speed. The speed-up rate can be set, for example, by 1 OM bps or by 10%, but is not particularly limited.
図 8に示すように、 ここでは第 3回目のデータ伝送でバケツト損失が発 生したとする。  As shown in FIG. 8, it is assumed here that a bucket loss has occurred in the third data transmission.
<第 4回目の送信処理 >  <4th transmission process>
第 3回目に第 1送信フェーズ (つまり、 1 0データバケツト + 1パリテ ィパケットの 1 1パケットで 1セグメントとなるパケット群を 2 0セグメ ント分だけ分散送信する送信フェーズ: first phase) の送信でデータ損 失が発生した場合、 1段階前、 この場合には図 9に示すように第 2回目の 送信手順と同様の手順に戻って第 2送信フェーズでデータ送信を再開する ことになる (S K 2 2 ) 。 つまり、 図 5のフローチャートにおいて、 ステ ップ S 5では N Oに分岐し、 ステップ S 1 5では前回パリティ付加の送信 をしたので Y E Sに分岐し、 第 2送信フェーズの送信準備としてパリティ 付加処理をせず (ステップ S 1 7 A ) 、 通信速度を 1段階ダウンさせて第 2送信フェーズのデータ送信を再開する (ステップ S 1 9, S 1 3 ) 。 なお、 この図 9に示した送信操作によってもなおデータ損失が発生すれ ば、 さらに 1回前の操作手順、 つまり、 第 1回目と同様の送信速度で第 1 送信フェーズの送信を実施することになる。 In the third transmission, the first transmission phase (that is, the transmission phase in which a packet group consisting of 10 data buckets + 1 parity packet and 1 packet consisting of 1 segment is distributed for 20 segments) is transmitted in the first phase. If data loss occurs, one step before, in this case, as shown in Fig. 9, the procedure returns to the same procedure as the second transmission procedure and data transmission is restarted in the second transmission phase (SK twenty two ) . In other words, in the flowchart of FIG. 5, the process branches to NO in step S5, branches to YES in step S15 because the parity was transmitted last time, and performs the parity adding process in preparation for the transmission in the second transmission phase. (Step S17A), the communication speed is reduced by one step, and the data transmission in the second transmission phase is restarted (steps S19, S13). If data loss still occurs by the transmission operation shown in FIG. 9, if the data transmission is performed one more time before, that is, the first transmission is performed at the same transmission speed as the first transmission. The transmission in the transmission phase will be performed.
また、 前回の送信フェーズが第 2送信フェーズであって、 その第 2送信 フェーズにおいて受信失敗であった場合、 図 5のステップ S 5で NOに分 岐し、 またステップ S 1 5でも NOに分岐し、 第 1送信フェーズの準備を し (ステップ S 1 7 B) 、 前回と同じ送信速度でパリティバケツトを付加 した第 1送信フェーズでデータ送信する (ステップ S 1 3) 。  If the previous transmission phase was the second transmission phase and reception failed in the second transmission phase, the process branches to NO in step S5 in FIG. 5, and also branches to NO in step S15. Then, the first transmission phase is prepared (step S17B), and data is transmitted in the first transmission phase with the parity bucket added at the same transmission speed as the previous time (step S13).
このようにして、 本実施の形態のデータ通信方法では、 (a) 図 1 0 ( A) に示すように、 各伝送速度において第 1送信フェーズ (first phase ) としてパリティバケツトを付加した一定個数のバケツト群を 1セグメン トにして所定数のセグメントの分散送信を実施し、 (b) 図 10 (B) に 示すように、 データ損失が発生しなければ、 第 2送信フェーズ (second phase) としてパリティパケッ トを含まず、 かつ第 1送信フェーズの総パ ケット数と同じ個数のデータパケッ ト群をバース ト的に送信し、 (c) 第 2送信フェーズでもデータ損失が発生しなければ伝送速度を 1段階アップ して、 図 1 0 (A) に示す第 1送信フェーズのデータ伝送、 図 1 0 (B) に示す第 2送信フェーズのデータ伝送を繰り返し、 (d) いずれかのデー タ伝送期間にデータ損失が発生すればその 1回前の送信フェーズに戻して データ送信を再開する、 という操作を行う。 この (d) のデータ送信にお いては、 例えば、 図 1 0 (B) の第 2送信フェーズでデータ損失が発生す れば送信速度は維持したまま、 同図 (A) の第 1送信フェーズでデータ送 信を再開し、 図 1 0 (A) の第 1送信フェーズでデータ損失が発生すれば 送信速度を 1段階落として同図 (B) の第 2送信フェーズで送信再開する のである。  Thus, in the data communication method of the present embodiment, (a) As shown in FIG. 10 (A), a fixed number of packets each having a parity bucket added as a first transmission phase (first phase) at each transmission rate. The bucket group is divided into one segment and distributed transmission of a predetermined number of segments is performed. (B) As shown in Fig. 10 (B), if no data loss occurs, the second transmission phase (second phase) is performed. A data packet group that does not include a parity packet and is the same as the total number of packets in the first transmission phase is transmitted in bursts. (C) If no data loss occurs in the second transmission phase, the transmission speed The data transmission in the first transmission phase shown in Fig. 10 (A) and the data transmission in the second transmission phase shown in Fig. 10 (B) are repeated, and (d) one of the data transmissions If data loss occurs during the period Resume sending data back to the transmission phase of the previous one, perform the operation that. In the data transmission of (d), for example, if data loss occurs in the second transmission phase of FIG. 10 (B), the transmission speed is maintained and the first transmission phase of FIG. 10 (A) is maintained. Then, if data loss occurs in the first transmission phase of Fig. 10 (A), the transmission speed is reduced by one step and transmission is restarted in the second transmission phase of Fig. 10 (B).
また、 上記の実施の形態では例えば、 マルチメディアコンテンツのス ト リーム送信を想定し、 データ損失が発生しても該当データ群を再送するの ではなく、 次々に新たなデータ群からデータ送信を再開する方式にしたが 、 データ損失が発生した場合、 損失したデータ群のデータ送信を再開する 方式にすることもできる。 Also, in the above embodiment, for example, it is assumed that multimedia content is transmitted in a stream, and even if data loss occurs, the corresponding data group is retransmitted. Instead, the data transmission is restarted from a new data group one after another. However, if data loss occurs, the data transmission of the lost data group may be restarted.
これによつて第 1の実施の形態によれば、 パリティデータを付加する分 だけ冗長性が増し、 ネットワーク上でのデータ消失を受信側が受信データ からパリティデータを用いて復旧できて完全なデータ損失の機会を少なく でき、 その分データ損失に対する再送の機会を減少させることができ、 ま た第 1送信フェーズでデータ損失が発生しなければ、 第 2送信フェーズと してパリティバケツ トを含まず、 かつ第 1送信フェーズの総バケツト数と 同じ個数のデータパケッ ト群を同じ送信速度でパースト的に送信すること によって、 前回成功したので今回も成功の可能性の高い送信速度において パリティパケットをデータバケツトに置き換えて送信することにより、 同 じ送信速度でのデータパケットの送信個数を増やすことができ、 それだけ 実質送信速度を上げることができ、 これらの総合的な結果として、 データ 伝送のスループットを改善することができる。  As a result, according to the first embodiment, the redundancy is increased by the addition of the parity data, and the data loss on the network can be recovered by the receiving side using the parity data from the received data, and complete data loss can be achieved. Therefore, if there is no data loss in the first transmission phase, the parity packet is not included in the second transmission phase, and the retransmission is not performed in the first transmission phase. In addition, by transmitting the same number of data packets as the total number of buckets in the first transmission phase in a burst manner at the same transmission speed, the parity packet was transmitted at the transmission speed that is likely to succeed this time because it was successful last time. The number of data packets transmitted at the same transmission rate can be increased by replacing Only the actual transmission speed can be increased, and as a result of all these, the throughput of data transmission can be improved.
次に、 本発明の第 2の実施の形態のデータ送信装置及びデータ送信方法 について、 図 1、 図 2、 図 1 1を用いて説明する。 第 2の実施の形態は、 第 1の実施の形態に対して第 1送信フェーズでの送信失敗の時の通信速度 低減手法が異なり、 第 1送信フェーズで送信失敗すれば、 送信速度を 1段 階ダウンさせて同じ第 1送信フェーズでデータ送信を再開することを特徴 とする。 したがって、 データ送信装置の機能構成は第 1の実施の形態と同 様に図 1、 図 2に示すものと共通である。 ただし、 本実施の形態の場合、 送信フ ーズ決定部 1 5は、 前回の第 2送信フェーズで送信失敗があれば 同じ送信速度で第 1送信フェーズで送信する決定をするが、 前回の第 1送 信フェーズで送信失敗があれば、 送信速度を 1段階ダウンさせて同じ第 1 送信フェーズでデータ送信を再開する決定をする。 Next, a data transmission device and a data transmission method according to a second embodiment of the present invention will be described using FIG. 1, FIG. 2, and FIG. The second embodiment differs from the first embodiment in the method of reducing the communication speed when the transmission fails in the first transmission phase. If the transmission fails in the first transmission phase, the transmission speed is reduced by one step. Data transmission is resumed in the same first transmission phase after the floor has been lowered. Therefore, the functional configuration of the data transmission device is the same as that shown in FIGS. 1 and 2 as in the first embodiment. However, in the case of the present embodiment, if there is a transmission failure in the previous second transmission phase, the transmission phase determination unit 15 determines to transmit in the first transmission phase at the same transmission speed. If there is a transmission failure in one transmission phase, the transmission speed is reduced by one step and the same Decide to resume data transmission in the transmission phase.
本実施の形態のデータ送信装置によるデータ送信処理のフローチヤ一ト は図 1 1のようになる。 この第 2の実施の形態によるデータ送信処理では 、 前回のデータ送信が成功した場合の実質速度アップの処理は第 1の実施 の形態と同様である。 また、 前回のデータ送信が失敗した場合で、 その前 回の送信フェーズが第 2送信フェーズであった場合の処理についても第 1 の実施の形態と同様であり、 図 1 1のステップ S 5で N Oに分岐し、 また ステップ S 1 5でも N Oに分岐し、 第 1送信フェーズの準備をし (ステツ プ S 1 7 B ) 、 前回と同じ送信速度でパリティバケツトを付加した第 1送 信フェーズでデータ送信する (ステップ S 1 3 ) 。  FIG. 11 is a flowchart of a data transmission process performed by the data transmission device according to the present embodiment. In the data transmission process according to the second embodiment, the process of substantially increasing the speed when the previous data transmission succeeds is the same as that of the first embodiment. Also, the processing when the previous data transmission failed and the previous transmission phase was the second transmission phase is the same as in the first embodiment. The procedure branches to NO, and also branches to NO in step S15, prepares for the first transmission phase (step S17B), and adds the parity bucket at the same transmission rate as the previous transmission phase. (Step S13).
そして本実施の形態の特徴として、 前回のデータ送信が失敗した場合で 、 その前回の送信フェーズが第 1送信フェーズであった場合には、 ステツ プ S 1 5で Y E Sに分岐し、 前回と同じ第 1送信フェーズでのデータ送信 準備をし (ステップ S 1 7 C ) 、 前回よりも送信速度を 1段階ダウンさせ てデータ送信を再開する処理をする (ステップ S 1 9, S 1 3 ) 。  As a feature of the present embodiment, when the previous data transmission has failed and the previous transmission phase was the first transmission phase, the process branches to YES in step S15, and is the same as the previous time. It prepares for data transmission in the first transmission phase (step S17C), and performs processing to reduce the transmission speed by one step from the previous time and restart data transmission (steps S19, S13).
この第 2の実施の形態によっても、 第 1の実施の形態と同様に、 パリテ ィデータを付加する分だけ冗長性が増し、 ネットワーク上でのデータ消失 を受信側が受信データからパリティデータを用いて復旧できて完全なデー タ損失の機会を少なくでき、 その分データ損失に対する再送の機会を減少 させることができ、 また第 1送信フェーズでデータ損失が発生しなければ 、 第 2送信フェーズとしてパリティパケットを含まず、 かつ第 1送信フエ 一ズの総バケツ ト数と同じ個数のデータバケツ ト群を同じ送信速度でバー スト的に送信することによって、 前回成功したので今回も成功の可能性の 高い送信速度においてパリティパケッ トをデータバケツトに置き換えて送 信することにより、 同じ送信速度でのデータパケッ トの送信個数を増やす ことができ、 それだけ実質送信速度を上げることができ、 これらの総合的 な結果として、 データ伝送のスループットを改善することができる。 According to the second embodiment, as in the first embodiment, redundancy is increased by adding parity data, and data loss on the network can be recovered by the receiving side using parity data from the received data. As a result, the chances of complete data loss can be reduced, the chances of retransmission for data loss can be reduced accordingly, and if no data loss occurs in the first transmission phase, the parity packet is transmitted as the second transmission phase. By transmitting the same number of data bucket groups as the total number of buckets in the first transmission phase, in bursts at the same transmission speed, the transmission was successful the previous time, so this transmission is likely to succeed this time The number of data packets transmitted at the same transmission rate can be increased by replacing parity packets with data packets at the same transmission speed. To Therefore, the transmission rate can be substantially increased, and the overall result can improve the throughput of data transmission.
次に、 本発明の第 3の実施の形態について、 図 1 2〜図 2 1を用いて説 明する。 第 3の実施の形態のデータ送信装置のシステム構成は第 1の実施 の形態、 第 2の実施の形態と同様であるが、 その冗長化データ送信方法に 特徴があり、 図 1 2〜図 2 1に示す送信方法をとる。 いま、 図 1 2、 図 1 3に示すように送信すべきデータパケットがデータパケッ ト (1 ) 〜デー タパケット (6) であり、 データパケット (1 ) , (3) , (5) がセグ メント S G 1に属するパケット群、 データパケッ ト ( 2) , (4) , (6 ) がセグメント S G 2に属するパケッ ト群であるとする。 そして、 これら セグメント S G 1, S G 2それぞれに対して冗長化のためにパリティパケ ット P T 1 とパリティパケッ ト P T 2を付加して送信する場合、 次のよう にして冗長化したバケツ ト送信を行う。  Next, a third embodiment of the present invention will be described with reference to FIGS. The system configuration of the data transmission device of the third embodiment is the same as that of the first embodiment and the second embodiment, but has a feature in its redundant data transmission method. The transmission method shown in 1 is taken. Now, as shown in FIGS. 12 and 13, the data packets to be transmitted are data packets (1) to (6), and data packets (1), (3) and (5) are segments. It is assumed that the packet group belonging to SG1 and the data packets (2), (4), (6) are the packet group belonging to segment SG2. In addition, when a parity packet PT1 and a parity packet PT2 are added to each of the segments SG1 and SG2 for redundancy to be transmitted, the redundant packet transmission is performed as follows. .
図 1 4に示すように、 複数のセグメント S G 1, S G 2それぞれに属す る送信すべきデータパケット (1 ) , (3) , (5) のパケット群とデー タパケット (2) , (4) , (6) のパケット群を送信順に一列に並べた 状態で送信を開始する。 そして、 最初に送信するセグメント S G 1に属す るパケット (1 ) を送信し、 この際にセグメント S G 1用のパリティパケ ット P T 1に対してパリティを記録する。  As shown in Fig. 14, the data packets (1), (3), and (5) belonging to each of the segments SG1 and SG2 and the data packets (2), (4), The transmission starts with the packet group of (6) arranged in a line in the transmission order. Then, the packet (1) belonging to the segment SG1 to be transmitted first is transmitted, and at this time, the parity is recorded in the parity packet PT1 for the segment SG1.
次に図 1 5に示すように、 セグメント S G 2に属する最初のデータパケ ット (2) を送信し、 この際にセグメント S G 2用のパリティパケット P T 2に対してパリティを記録する。  Next, as shown in FIG. 15, the first data packet (2) belonging to the segment SG2 is transmitted, and at this time, the parity is recorded in the parity packet PT2 for the segment SG2.
続いて、 図 1 6に示すように再ぴセグメント S G 1に属する 2番目のデ ータパケット (3) を送信し、 その際にもセグメント S G 1用のパリティ パケッ ト P T 1に対してパリティを記録する。 これに続いて、 図 1 7に示 すように、 セグメント S G 2に属する 2番目のデータパケッ ト (4 ) を送 信し、 セグメント S G 2用のパリティバケツト P T 2にパリティを記録す る。 Subsequently, as shown in FIG. 16, the second data packet (3) belonging to the reproduction segment SG1 is transmitted, and at this time, the parity is recorded in the parity packet PT1 for the segment SG1. . Following this, Figure 17 shows As described above, the second data packet (4) belonging to the segment SG2 is transmitted, and the parity is recorded in the parity bucket PT2 for the segment SG2.
以下、 同様の送信手順を操り返す。 そして図 1 8に示すようにセグメン ト S G 1の最後のデータパケット (5 ) を送信し、 そのパリティをパリテ ィパケット P T 1に記録し、 続いて図 1 9に示すようにセグメント S G 2 の最後のデータパケッ ト (6 ) を送信し、 そのパリティをパリティバケツ ト P T 2に記録する。 そして、 最後に図 2 0、 図 2 1に示すようにセグメ ント S G 1用のパリティパケット P T 1 とセグメント S G 2用のパリティ パケット P T 2を送信する。  Hereinafter, the same transmission procedure is repeated. Then, as shown in FIG. 18, the last data packet (5) of the segment SG 1 is transmitted, its parity is recorded in the parity packet PT1, and then, as shown in FIG. The data packet (6) is transmitted and its parity is recorded in the parity bucket PT2. Finally, as shown in FIGS. 20 and 21, a parity packet PT1 for the segment SG1 and a parity packet PT2 for the segment SG2 are transmitted.
このよ うにして、 同時に送信すべき複数のセグメントそれぞれに属する データバケツトを順に送信し、 その各データバケツトの送信の際に各セグ メント用のパリティパケッ トにパリティを記録してゆき、 最後に複数のセ グメントそれぞれのパリティパケッ ト群を送信することにより、 あらかじ めセグメント毎のパリティパケットを作成する処理をしなくて済み、 冗長 化したデータ伝送の効率を向上させることができる。 実施例 1  In this way, data buckets belonging to each of a plurality of segments to be transmitted at the same time are sequentially transmitted, and at the time of transmission of each data bucket, the parity is recorded in the parity packet for each segment. Finally, by transmitting a parity packet group of each of a plurality of segments, it is not necessary to generate a parity packet for each segment in advance, and the efficiency of redundant data transmission can be improved. Example 1
図 2 2〜図 2 4による本発明の 2フェーズスロースタート伝送方式によ るデータ送信と従来のスロースタート輻輳回避制御伝送方式によるデータ 送信を評価した結果を図 2 5、 図 2 6のグラフに示した。 データ伝送量は 同じであり、 本発明の実施例ではデータ送信成功で 1 0 M b p sずつ速度 を上昇させ、 データ損失発生で 1 0 M b p sずつ速度を下げる制御をし、 また第 2送信フェーズでのデータ伝送は省略した。  The results of evaluating the data transmission by the two-phase slow-start transmission method of the present invention and the data transmission by the conventional slow-start congestion avoidance control transmission method according to the present invention shown in Figs. 22 to 24 are shown in the graphs of Figs. 25 and 26. Indicated. The data transmission amount is the same.In the embodiment of the present invention, the speed is increased by 10 Mbps when data transmission is successful, the speed is decreased by 10 Mbps when data loss occurs, and the control is performed in the second transmission phase. Is omitted.
すなわち、 最初に図 2 2に示すように、 first phase、 50Mbpsでパリテ ィパケットを含めた 1セグメント 2 1パケットを 20セグメント分送信す る ( S K 2 1 1 ) 。 この基底送信速度は 4. 8 m s e c /p a c k e tで ある。 That is, first, as shown in Fig. 22, the first phase, parity at 50 Mbps 1 segment 2 including 1 packet is transmitted for 20 segments (SK2111). This base transmission rate is 4.8 msec / packet.
この S K 2 1 1の送信が成功すれば、 1 0Mb p sの速度増加を行い、 図 2 3に示すように first phase、 60Mbpsでパリティパケッ トを含めた 1 セグメント 2 1パケットを 2 0セグメント分送信する (S K 2 2 1 ) 。 こ のときの送 1言速度は 4. 0 m s e c / p a c k e tでめ o。  If the transmission of this SK211 succeeds, the speed is increased by 10 Mbps, and as shown in Fig. 23, 1 segment 21 packets including the parity packet at 60 Mbps in the first phase are transmitted for 20 segments. Yes (SK2 2 1). The transmission speed at this time is 4.0 msec / packet.
続いて S K 2 2 1の送信も成功すれば、 さらに 1 0Mb p sの速度増加 を行い、 図 24に示すように first phase、 70Mbpsでノ リティパケットを 含めた 1セグメント 2 1パケッ トを 2 0セグメント分送信する (S K 2 3 1 ) 。 このときの送信速度は 3. 4m s e c/p a c k e tである。  Then, if the transmission of SK2221 succeeds, the speed is further increased by 10 Mbps, and as shown in Fig. 24, the first phase, 70 Mbps, and 1 segment 21 packets including the knowledge packet are 20 segments. Minute (SK 2 3 1). The transmission speed at this time is 3.4 msec / packet.
この S K 2 3 1の送信でパケット損失が発生すれば、 次の送信では、 1 OMb sだけ速度を低下させ、 6 0 Mb p sの速度で first phaseで送 信する。  If a packet loss occurs in the transmission of SK231, in the next transmission, the speed is reduced by 1 OMbs, and the transmission is performed in the first phase at a speed of 60 Mbps.
このような実施例 1のデータ送信方法と従来のデータ送信方法 (比較例 1 ) での送信特性、 送信結果を測定した。 これによれば、 図 2 5の実施例 1のグラフ及び図 2 6の比較例 1のグラフに示すように、 ギガビッ ト ·ィ ーサネッ ト (登録商標) を用いた 1対 1の評価では、 比較例 1のスロース タートよりも実施例 1によるデータ送信の方が 4 3. 9 %少ない損失率で 通信できることが実証された。 また、 本発明の場合、 パケッ ト復旧率は 4 1. 2 %であった。  The transmission characteristics and transmission results of the data transmission method of Example 1 and the conventional data transmission method (Comparative Example 1) were measured. According to this, as shown in the graph of Example 1 in FIG. 25 and the graph of Comparative Example 1 in FIG. 26, in the one-to-one evaluation using Gigabit Ethernet (registered trademark), It was demonstrated that the data transmission according to the first embodiment can communicate with a loss rate of 43.9% less than the slow start of the first embodiment. In the case of the present invention, the packet recovery rate was 41.2%.
実施例 2 Example 2
本発明の第 3の実施の形態による 2フェーズスロースタート伝送方式に よるデータ送信と従来のスロースタート伝送方式によるデータ送信を評価 した。 実験環境としては、 P C 1, P C 2の 2台のコンピュータ間をギガビッ トイーサネット (登録商標) で接続し、 P C 1 と P C 2に NIST Net (商品 名) という遅延発生エミュレータのソフトウェアを動作させ、 P C 1から P C 2へのデータ送信にかかる遅延時間を 1 0 0 m s e cにした。 これで 往復遅延時間は 20 0m s e cになり、 この遅延時間は日本一ァメ リ力間 の長距離ネッ トワークの遅延時間に相当する。 この NIST Netは 「The Linux-based Network Emulation Tool NIST Net 」 で あ り 、 http://snad. ncsl. nist. gov/itg/nistnet/にて入手したものである。 The data transmission by the two-phase slow start transmission method according to the third embodiment of the present invention and the data transmission by the conventional slow start transmission method were evaluated. As an experimental environment, two computers, PC1 and PC2, were connected by Gigabit Ethernet (registered trademark), and PC1 and PC2 were operated with NIST Net (trade name) software of a delay generation emulator. The delay time required for data transmission from PC 1 to PC 2 was set to 100 msec. This results in a round-trip delay of 200 ms, which is equivalent to the delay of a long-distance network between Japan and the United States. This NIST Net is "The Linux-based Network Emulation Tool NIST Net" and was obtained from http://snad.ncsl.nist.gov/itg/nistnet/.
この条件で、 P C 2から P C 1へデータを送信し、 従来のスロースター トと本発明による 2フェーズスロースタート (T P S S) の 2つのアルゴ リズムによる送信速度とバケツト損失率の差を評価した。  Under this condition, data was transmitted from PC2 to PC1, and the difference between the transmission speed and the bucket loss rate due to the two algorithms of the conventional slow start and the two-phase slow start (TPSS) according to the present invention was evaluated.
送信速度の変化方法としては、 パケット損失率が 5%を越えた場合、 受 信側が送信側に送信速度を 0. 8倍するように要求し、 そうでない場合は 速度を 1. 2倍にするように要求することにした。  To change the transmission speed, if the packet loss rate exceeds 5%, the receiver requests the transmitter to increase the transmission speed by 0.8, otherwise it increases the speed by 1.2 times. Decided to request.
そして、 T P S Sでは、  And in T P S S,
( 1 ) 第 1送信フェーズの送信でパケット損失率が 5 %を越えていなけ れば第 2送信フェーズで冗長化せずに送信し、 第 2送信フェーズでも損失 率が 5 %を越えなければ送信速度が 1. 2倍になるように冗長化し、 再ぴ 第 1送信フェーズを開始する、  (1) If the packet loss rate does not exceed 5% in the transmission of the first transmission phase, transmit without redundancy in the second transmission phase, and transmit if the loss rate does not exceed 5% in the second transmission phase. Redundant so that the speed becomes 1.2 times, and start the first transmission phase.
(2) 第 1送信フェーズ、 第 2送信フェーズで損失率が 5 %を越えた場 合には、 1つ前の第 2送信フェーズに戻る、  (2) If the loss rate exceeds 5% in the first transmission phase and the second transmission phase, return to the previous second transmission phase,
という条件に設定した。 Was set to the condition.
実験条件は、 本実施例 2は次の表 1のものであった。  The experimental conditions in this Example 2 were as shown in Table 1 below.
表 1.本実施例の伝送条件 送信方式 TPSS Table 1 Transmission conditions of this embodiment Transmission method TPSS
分散送信するセグメント数 20セグメン卜  Number of segments for distributed transmission 20 segments
1セグメントのデータパケット数 5パケット  Number of data packets per segment 5 packets
送信時間 1 50sec  Transmission time 1 50sec
片道遅延時間 100msec 他方、 比較例 2の実験条件は、 次の表 2のものであった <  One-way delay time 100 ms On the other hand, the experimental conditions of Comparative Example 2 were as shown in Table 2 below.
表 2.比較例 2の伝送条件  Table 2 Transmission conditions for Comparative Example 2
Figure imgf000026_0001
Figure imgf000026_0001
これらの伝送条件で伝送試験を行った評価結果は、 次の表 3、 表 4 示し、 また図 2 7〜図 3 0に示すものであった。  The evaluation results of the transmission test performed under these transmission conditions are shown in Tables 3 and 4 below and shown in Figs. 27 to 30.
表 3.実施例 2の評価結果  Table 3 Evaluation results of Example 2
受信パケット数 267257パケット  Number of received packets 267 257 packets
損失パケット数 1 5181パケット  Number of lost packets 1 5181 packets
(損失率) (5.37%) 表 4.比較例 2の評価結果  (Loss rate) (5.37%) Table 4. Evaluation results of Comparative Example 2
受信パケット数 275646パケット  Number of received packets 275646 packets
損失パケット数 77292パケット  Number of lost packets 77292 packets
(損失率) (21 .88%) この実験結果から、 本実施例 2によるデータ伝送では実効伝送速度が向 上できることが確かめられた。 産業上の利用の可能性 (Loss rate) (21.88%) From this experimental result, it was confirmed that the data transmission according to the second embodiment can improve the effective transmission speed. Industrial potential
本発明によれば、 パリティデータを付加する分だけ冗長性が増し、 ネッ トワーク上でのデータ消失を受信側が受信データからパリティデータを用 いて復旧できて完全なデータ損失の機会を少なくでき、 その分データ損失 に対する再送の機会を減少させることができ、 また第 1送信フェーズでデ ータ損失が発生しなければ、 第 2送信フェーズとしてパリティバケツトを 含まず、 かつ第 1送信フェーズの総バケツト数と同じ個数のデータバケツ ト群を同じ送信速度でパースト的に送信することによって、 前回成功した ので今回も成功の可能性の高い送信速度においてパリティバケツトをデ一 タパケッ トに置き換えて送信することにより、 同じ送信速度でのデータパ ケッ トの送信個数を増やすことができ、 それだけ実質送信速度を上げるこ とができ、 これらの総合的な結果として、 データ伝送のスループットを改 善することができる。  According to the present invention, the redundancy is increased by the addition of parity data, and the data loss on the network can be recovered by the receiving side using the parity data from the received data, so that the opportunity for complete data loss can be reduced. The chances of retransmission for data loss can be reduced, and if no data loss occurs in the first transmission phase, the parity packet is not included in the second transmission phase, and the total buckets in the first transmission phase By transmitting the same number of data buckets in bursts at the same transmission rate, the parity packet is replaced with data packets at the transmission rate that is likely to succeed this time because it succeeded the previous time. As a result, the number of data packets transmitted at the same transmission speed can be increased, and the actual transmission speed is increased accordingly. Can and this, as these overall results, the throughput of data transmission can be improved.

Claims

請 求 の 範 囲 1. 単一の伝送路上で、 Scope of Claim 1. On a single transmission line,
( a ) 所定の伝送速度で、 パリティパケッ トを付加した一定個数 N (Nは任意の自然数) のパケット群を 1セグメントにして所定数 M (Mは 任意の自然数) のセグメントを、 各セグメントに属するパケッ トが一連の M個のバケツトで成る送信バケツ ト群それぞれに 1個ずつ含まれるように して当該送信バケツト群ごとに N群分、 合計 NX Mのパケットを分散送信 する第 1送信フェーズのデータ送信を行うステップと、  (a) At a given transmission rate, a fixed number N (N is an arbitrary natural number) packet group to which a parity packet is added is divided into one segment, and a predetermined number M (M is an arbitrary natural number) of segments is added to each segment. A first transmission phase in which packets belonging to N groups, a total of NXM packets, are distributed and transmitted for each of the transmission bucket groups so that one packet belongs to each transmission bucket group consisting of a series of M buckets. Transmitting the data of
(b ) 前記 ( a) の第 1送信フェーズの送信でデータ損失が発生し なければ、 前記 (a) の第 1送信フェーズの伝送速度と同じ伝送速度で、 パリティバケツトを含まず、 かつ前記第 1送信フェーズと同じ総バケツト 数のデータパケット群をパースト的に送信する第 2送信フェーズのデータ 送信を行うステップと、  (b) If no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a), and no parity bucket is included; and Performing data transmission in a second transmission phase for transmitting a data packet group having the same total number of buckets as the first transmission phase in a burst manner;
( c ) 前記 (b ) の第 2送信フェーズの送信でもデータ損失が発生 しなければ伝送速度を予め設定されている 1段階アップして、 前記第 1送 信フェーズのデータ送信を行うステップと、  (c) performing a data transmission in the first transmission phase by increasing the transmission speed by one preset stage if no data loss occurs even in the transmission in the second transmission phase in (b);
(d) 前記 ( c ) の第 1送信フェーズのデータ送信でデータ損失が 発生しなければ、 前記第 2送信フェーズのデータ送信を行うステップと、 (e ) 以降、 データ損失が発生しない限り伝送速度を所定の 1段階 ずつアップしながらの前記 ( c) の第 1送信フェーズのデータ送信と、 前 記 (d) の第 2送信フェーズのデータ送信を繰り返すステップと、  (d) performing a data transmission in the second transmission phase if no data loss occurs in the data transmission in the first transmission phase in (c); and Repeating the data transmission in the first transmission phase of (c) and the data transmission of the second transmission phase in (d) while increasing the
( f ) いずれかのデータ伝送期間にデータ損失が発生すれば、 その 1回前のフェーズに戻してデータ送信を再開するステップとを有するデー タ送信方法。 (f) If data loss occurs during any of the data transmission periods, returning to the previous phase and restarting data transmission.
2 . 単一の伝送路上で、 2. On a single transmission line,
( a ) 所定の伝送速度で、 パリティパケッ トを付加した一定個数 N ( Nは任意の自然数) のパケット群を 1セグメントにして所定数 M (Mは 任意の自然数) のセグメントを、 各セグメントに属するパケッ トが一連の M個のパケッ 卜で成る送信バケツ ト群それぞれに 1個ずつ含まれるように して当該送信バケツト群ごとに N群分、 合計 N X Mのバケツトを分散送信 する第 1送信フェーズのデータ送信を行うステップと、  (a) At a given transmission rate, a fixed number N (N is an arbitrary natural number) packet group to which a parity packet is added is divided into one segment, and a predetermined number M (M is an arbitrary natural number) of segments is added to each segment. The first transmission phase in which packets belonging to N groups are distributed for each of the transmission bucket groups so that one packet belongs to each transmission bucket group consisting of a series of M packets, and a total of NXM buckets are transmitted. Transmitting the data of
( b ) 前記 (a ) の第 1送信フェーズの送信でデータ損失が発生し なければ、 前記 (a ) の第 1送信フェーズの伝送速度と同じ伝送速度で、 パリティパケッ トを含まず、 かつ、 前記第 1送信フェーズと同じ総バケツ ト数のデータバケツト群をパースト的に送信する第 2送信フェーズのデー タ送信を行うステップと、  (b) If no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a), and no parity packet is included, and Performing a data transmission in a second transmission phase for transmitting a data bucket group having the same total number of buckets as the first transmission phase in a burst manner;
( c ) 前記 (b ) の第 2送信フェーズの送信でもデータ損失が発生 しなければ伝送速度を予め設定されている 1段階アップして、 前記第 1送 信フェーズのデータ送信を行うステップと、  (c) performing a data transmission in the first transmission phase by increasing the transmission speed by one preset stage if no data loss occurs even in the transmission in the second transmission phase in (b);
( d ) 前記 ( c ) の第 1送信フェーズのデータ送信でデータ損失が 発生しなければ、 前記第 2送信フェーズのデータ送信を行うステップと、 (d) performing data transmission in the second transmission phase if no data loss occurs in the data transmission in the first transmission phase in (c);
( e ) 以降、 データ損失が発生しない限り伝送速度を所定の 1段階 ずつアップしながらの前記 (c ) の第 1送信フェーズのデータ送信と、 前 記.(d ) の第 2送信フェーズのデータ送信を繰り返すステップと、 (e) Thereafter, as long as no data loss occurs, the data transmission in the first transmission phase of (c) while increasing the transmission speed by a predetermined step, and the data of the second transmission phase in (d) above Repeating the transmission;
( f ) いずれかのデータ伝送期間にデータ損失が発生すれば、 それ が第 2送信フェーズのデータ送信であった場合には同じ伝送速度の第 1送 信フェーズに戻してデータ送信を再開するステップと、  (f) If data loss occurs during any of the data transmission periods, if it is data transmission in the second transmission phase, return to the first transmission phase at the same transmission speed and restart data transmission When,
( g ) いずれかのデータ伝送期間にデータ損失が発生すれば、 それ が第 1送信フェーズのデータ送信であった場合には伝送速度を所定の 1段 階ダウンし、 かつ、 第 1送信フェーズでデータ送信を再開するステップと を有するデータ送信方法。 (g) If data loss occurs during any of the data transmission periods, and if it is data transmission in the first transmission phase, the transmission speed is increased by one predetermined stage. And d. Resuming data transmission in the first transmission phase.
3 . 送信すべきデータを保持するデータ保持部と、  3. A data holding unit for holding data to be transmitted,
単一の伝送路を通じて情報ネッ トワークに対してデータ授受するネ ッ トワークインタフェースと、  A network interface for exchanging data with the information network through a single transmission line,
前記データ保持部に保持されている送信すべきデータをパケッ ト化 するバケツト化処理部と、  A packetization processing unit that packetizes data to be transmitted held in the data holding unit;
一定個 (N— 1 ) のパケッ ト列に対してパリティパケッ トを生成す るパリティパケット生成部と、  A parity packet generator for generating a parity packet for a fixed (N-1) packet sequence;
前回のデータ送信の成功ノ不成功の情報と前回の送信フェーズの情 報とに基づき、 今回の送信フェーズと送信速度を決定する送信フェーズ決 定部と、  A transmission phase determining unit that determines the current transmission phase and the transmission speed based on the information on the success or failure of the previous data transmission and the information on the previous transmission phase;
前記送信フェーズ決定部の決定した送信フェーズの情報に基づき、 前記パリティバケツトを付加した一定個数 Nのバケツト群を 1セグメント にして所定数 Mのセグメントを生成し、 又は、 前記パリティパケットを含 まず、 かつ前記所定数 Mのセグメント内の総バケツト数 N X Mと同じ個数 のデータバケツ ト群を生成する送信バケツト群作成部と、  Based on the information on the transmission phase determined by the transmission phase determining unit, a predetermined number M of segments is generated by dividing the fixed number N of bucket groups to which the parity bucket is added into one segment, or including the parity packet. And a transmission bucket group creation unit for generating a data bucket group of the same number as the total number of buckets NXM in the predetermined number M of segments,
前記送信フェーズ決定部の決定した送信速度で、 前記送信パケッ ト 群作成部の生成したバケツト群を前記ネットワークインタフェースを介し て前記情報ネッ トワークに送信するデータ送信処理部とを備え、  A data transmission processing unit that transmits the packet group generated by the transmission packet group generation unit to the information network via the network interface at the transmission speed determined by the transmission phase determination unit;
前記送信フェーズ決定部は、  The transmission phase determination unit,
( a ) 所定の伝送速度で、 パリティパケッ トを付加した一定個 数 N ( Nは任意の自然数) のパケッ ト群を 1セグメントにして所定数 M ( Mは任意の自然数) のセグメントを、 各セグメントに属するパケットがー 連の M個のパケットで成る送信バケツ ト群それぞれに 1個ずつ含まれるよ うにして当該送信パケット群ごとに N群分、 合計 N X Mのバケツ トを分散 送信する第 1送信フェーズのデータ送信を行い、 (a) At a given transmission rate, a packet group of a fixed number N (N is an arbitrary natural number) to which a parity packet is added is made into one segment, and a segment of a predetermined number M (M is an arbitrary natural number) is divided into A packet belonging to a segment is included in each of a series of M packets of transmission packets. In this way, data transmission in the first transmission phase for distributing and transmitting NXM buckets for N groups for each transmission packet group is performed.
( b ) 前記 ( a ) の第 1送信フェーズの送信でデータ損失が発 生しなければ、 前記 (a ) の第 1送信フェーズの伝送速度と同じ伝送速度 で、 パリティパケットを含まず、 かつ前記第 1送信フェーズと同じ総パケ ット数のデータパケット群をパースト的に送信する第 2送信フェーズのデ ータ送信を行い、  (b) if no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a), no parity packet is included, and Perform data transmission in the second transmission phase for transmitting data packets of the same total number of packets as in the first transmission phase in a burst manner,
( c ) 前記 (b ) の第 2送信フェーズの送信でもデータ損失が 発生しなければ伝送速度を予め設定されている 1段階アップして、 前記第 1送信フェーズのデータ送信を行い、  (c) If data loss does not occur even in the transmission of the second transmission phase of (b), the transmission rate is increased by one preset level, and the data transmission of the first transmission phase is performed.
( d ) 前記 ( c ) の第 1送信フェーズのデータ送信でデータ損 失が発生しなければ、 前記第 2送信フェーズのデータ送信を行い、  (d) if no data loss occurs in the data transmission in the first transmission phase of (c), perform the data transmission in the second transmission phase;
( e ) 以降、 データ損失が発生しない限り伝送速度を所定の 1 段階ずつアップしながらの前記 (c ) の第 1送信フェーズのデータ送信と 、 前記 (d ) の第 2送信フェーズのデータ送信を繰り返し、  (e) Thereafter, the data transmission in the first transmission phase of (c) and the data transmission of the second transmission phase in (d) while increasing the transmission speed by a predetermined step at a time unless data loss occurs, Repeatedly,
( f ) いずれかのデータ伝送期間にデータ損失が発生すれば、 その 1回前の送信フェーズに戻してデータ送信を再開する、  (f) If data loss occurs during any data transmission period, return to the previous transmission phase and restart data transmission;
というルールに基づいて今回の送信フェーズと送信速度を決定することを 特徴とするデータ送信装置。 A data transmission device that determines the current transmission phase and transmission speed based on the following rule.
4 . 送信すべきデータを保持するデータ保持部と、  4. A data holding unit for holding data to be transmitted,
単一の伝送路を通じて情報ネッ トワークに対してデータ授受するネ ッ トワークインタフェースと、  A network interface for exchanging data with the information network through a single transmission line,
前記データ保持部に保持されている送信すべきデータをバケツ ト化 するパケット化処理部と、  A packetization processing unit configured to bucket data to be transmitted held in the data holding unit;
—定個 (N— 1 ) のパケッ ト列に対してパリティパケッ トを生成す るパリティバケツト生成部と、 —Generate parity packets for a fixed number (N-1) of packet sequences A parity bucket generator,
前回のデータ送信の成功/不成功の情報と前回の送信フェーズの情 報とに基づき、 今回の送信フェーズと送信速度を決定する送信フェーズ決 定部と、  A transmission phase determining unit that determines the current transmission phase and transmission speed based on the information on the success / failure of the previous data transmission and the information on the previous transmission phase;
前記送信フェーズ決定部の決定した送信フェーズの情報に基づき、 前記パリティパケッ トを付加した一定個数 Nのパケッ ト群を 1セグメント にして所定数 Mのセグメントを生成し、 又は、 前記パリティパケッ トを含 まず、 かつ前記所定数 Mのセグメント内の総バケツト数 N X Mと同じ個数 のデータパケッ ト群を生成する送信パケット群作成部と、  Based on the information on the transmission phase determined by the transmission phase determination unit, a predetermined number M of segments is generated by dividing the fixed number N of packet groups to which the parity packet is added into one segment, or the parity packet is generated. First, a transmission packet group creation unit that generates a data packet group of the same number as the total number of buckets NXM in the predetermined number M of segments,
前記送信フェーズ決定部の決定した送信速度で、 前記送信パケッ ト 群作成部の生成したバケツト群を前記ネットワークインタフェースを介し て前記情報ネッ トワークに送信するデータ送信処理部とを備え、  A data transmission processing unit that transmits the packet group generated by the transmission packet group generation unit to the information network via the network interface at the transmission speed determined by the transmission phase determination unit;
前記送信フエ一ズ決定部は、  The transmission phase determining unit,
( a ) 所定の伝送速度で、 パリティパケッ トを付加した一定個 数 N ( Nは任意の自然数) のパケッ ト群を 1セグメントにして所定数 M ( Mは任意の自然数) のセグメントを、 各セグメントに属するパケッ トがー 連の M個のバケツ トで成る送信パケッ ト群それぞれに 1個ずつ含まれるよ うにして当該送信バケツト群ごとに N群分、 合計 N X Mのバケツ トを分散 送信する第 1送信フエ一ズのデ一タ送信を行い、  (a) At a given transmission rate, a packet group of a fixed number N (N is an arbitrary natural number) to which a parity packet is added is made into one segment, and a segment of a predetermined number M (M is an arbitrary natural number) is divided into A total of NXM buckets are distributed and transmitted so that one packet belonging to the segment is included in each of the M transmission packets, and one N packet for each transmission packet group. Performs data transmission of the first transmission phase,
( b ) 前記 (a ) の第 1送信フェーズの送信でデータ損失が発 生しなければ、 前記 (a ) の第 1送信フェーズの伝送速度と同じ伝送速度 で、 パリティパケッ トを含まず、 かつ前記第 1送信フェーズと同じ総パケ ット数のデータバケツト群をパースト的に送信する第 2送信フェーズのデ ータ送信を行い、  (b) if no data loss occurs in the transmission of the first transmission phase of (a), the transmission rate is the same as the transmission rate of the first transmission phase of (a), and no parity packet is included; and Performing data transmission in a second transmission phase for transmitting a data packet group having the same total number of packets as the first transmission phase in a burst manner,
( c ) 前記 (b ) の第 2送信フェーズの送信でもデータ損失が 発生しなければ伝送速度を予め設定されている 1段階アップして、 前記第 1送信フェーズのデータ送信を行い、 (c) Even in the transmission of the second transmission phase of (b), data loss occurs. If it does not occur, increase the transmission speed by one preset level, perform data transmission in the first transmission phase,
( d ) 前記 ( c ) の第 1送信フェーズのデータ送信でデータ損 失が発生しなければ、 前記第 2送信フェーズのデータ送信を行い、  (d) if no data loss occurs in the data transmission in the first transmission phase of (c), perform the data transmission in the second transmission phase;
( e ) 以降、 データ損失が発生しない限り伝送速度を所定の 1 段階ずつアップしながらの前記 (c ) の第 1送信フェーズのデータ送信と 、 前記 (d ) の第 2送信フェーズのデータ送信を繰り返し、  (e) Thereafter, the data transmission in the first transmission phase of (c) and the data transmission of the second transmission phase in (d) while increasing the transmission speed by a predetermined step at a time unless data loss occurs, Repeatedly,
( f ) いずれかのデータ伝送期間にデータ損失が発生すれば、 それが第 2送信フェーズのデータ送信であった場合には同じ伝送速度の第 1送信フェーズに戻してデータ送信を再開し、  (f) If data loss occurs during any of the data transmission periods, if it is data transmission in the second transmission phase, return to the first transmission phase at the same transmission rate and resume data transmission,
( g ) いずれかのデータ伝送期間にデータ損失が発生すれば、 それが第 1送信フェーズのデータ送信であった場合には伝送速度を所定の 1段階ダウンし、 かつ、 第 1送信フェーズでデータ送信を再開する、 というルールに基づいて今回の送信フェーズと送信速度を決定することを 特徴とするデータ送信装置。  (g) If data loss occurs during any of the data transmission periods, if it is data transmission in the first transmission phase, the transmission speed is reduced by one predetermined level, and data is lost in the first transmission phase. A data transmission device characterized in that the current transmission phase and transmission speed are determined based on a rule that transmission is restarted.
PCT/JP2004/002947 2003-09-19 2004-03-08 Method for transmitting data and data transmitter WO2005029806A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2005513992A JP4382752B2 (en) 2003-09-19 2004-03-08 Data transmission method and data transmission apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2003-328132 2003-09-19
JP2003328132 2003-09-19

Publications (1)

Publication Number Publication Date
WO2005029806A1 true WO2005029806A1 (en) 2005-03-31

Family

ID=34372892

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2004/002947 WO2005029806A1 (en) 2003-09-19 2004-03-08 Method for transmitting data and data transmitter

Country Status (2)

Country Link
JP (1) JP4382752B2 (en)
WO (1) WO2005029806A1 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009118244A (en) * 2007-11-07 2009-05-28 Internatl Business Mach Corp <Ibm> Technology for transmitting data whose regeneration unit is variable
JP2009159368A (en) * 2007-12-27 2009-07-16 Mitsubishi Electric Corp Data transmission apparatus and transmission band estimation method
JP2009206608A (en) * 2008-02-26 2009-09-10 Nec Engineering Ltd Communication device
JP2012517130A (en) * 2009-02-02 2012-07-26 スカイプ・リミテッド Data transmission method in communication system
JP2013544475A (en) * 2010-12-03 2013-12-12 インスティテュート フューア ランドファンクテクニック ゲーエムベーハー Method and system for controlling data packet transmission over lossy protocols

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016220187A (en) * 2015-05-20 2016-12-22 富士電機株式会社 Communication system and communication device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03274937A (en) * 1990-03-26 1991-12-05 Nec Corp Transmission error controller
JPH08223624A (en) * 1995-02-15 1996-08-30 Nec Corp Radio selective call receiver and radio data transmitting system
JP2003078560A (en) * 2001-09-05 2003-03-14 Oki Electric Ind Co Ltd Flow control system in transport layer protocol

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03274937A (en) * 1990-03-26 1991-12-05 Nec Corp Transmission error controller
JPH08223624A (en) * 1995-02-15 1996-08-30 Nec Corp Radio selective call receiver and radio data transmitting system
JP2003078560A (en) * 2001-09-05 2003-03-14 Oki Electric Ind Co Ltd Flow control system in transport layer protocol

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009118244A (en) * 2007-11-07 2009-05-28 Internatl Business Mach Corp <Ibm> Technology for transmitting data whose regeneration unit is variable
JP2009159368A (en) * 2007-12-27 2009-07-16 Mitsubishi Electric Corp Data transmission apparatus and transmission band estimation method
JP2009206608A (en) * 2008-02-26 2009-09-10 Nec Engineering Ltd Communication device
JP2012517130A (en) * 2009-02-02 2012-07-26 スカイプ・リミテッド Data transmission method in communication system
JP2013544475A (en) * 2010-12-03 2013-12-12 インスティテュート フューア ランドファンクテクニック ゲーエムベーハー Method and system for controlling data packet transmission over lossy protocols

Also Published As

Publication number Publication date
JPWO2005029806A1 (en) 2006-11-30
JP4382752B2 (en) 2009-12-16

Similar Documents

Publication Publication Date Title
Karzand et al. Design of FEC for low delay in 5G
US7502860B1 (en) Method and apparatus for client-side flow control in a transport protocol
JP4587053B2 (en) Communication apparatus, communication system, packet loss detection method, and packet loss detection program
US6445717B1 (en) System for recovering lost information in a data stream
Akyildiz et al. TCP-Peach+: Enhancement of TCP-Peach for satellite IP networks
Lin et al. TCP fast recovery strategies: Analysis and improvements
US8320248B2 (en) Method and system for communicating data in xDSL using data retransmission
US20020054570A1 (en) Data communication system, data communication method, and recording medium with data communication program recorded thereon
KR20020079796A (en) Wireless network system and method
EP2613497B1 (en) Method of transporting data in a sub-segmented manner
WO1996017462A2 (en) A reliable connectionless network protocol
JP4924285B2 (en) Communication apparatus, communication system, transfer efficiency improvement method, and transfer efficiency improvement program
CN113765626B (en) Data transmission method and device of mobile communication system
Ming-Chit et al. Improving TCP performance over asymmetric networks
CN109560897A (en) A kind of TCP repeating method and device
WO2005029806A1 (en) Method for transmitting data and data transmitter
US7100078B1 (en) Method and apparatus for restoration of lost blocks in a multicast data transmission
Peisa et al. Analytical model for TCP file transfers over UMTS
KR102601348B1 (en) Data transmission method, transmission device, data reception method and receiving device
US6981194B1 (en) Method and apparatus for encoding error correction data
JP5046197B2 (en) Communication apparatus, communication system, packet loss detection method, and packet loss detection program
West et al. TCP enhancements for heterogeneous networks
US7764624B2 (en) Methods and devices for preventing protocol stalling
Dhara et al. A performance enhanced transmission control protocol for inter-planetary networks
Epple et al. An Error Protection Protocol for user-transparent bridging of Fast Ethernet data transmission over the optical fading channel in an Aeronautical environment

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BW BY BZ CA CH CN CO CR CU CZ DK DM DZ EC EE EG ES FI GB GD GE GM HR HU ID IL IN IS JP KE KG KP KZ LC LK LR LS LT LU LV MA MD MK MN MW MX MZ NA NI NO NZ PG PH PL PT RO RU SC SD SE SG SK SY TJ TM TN TR TT TZ UA UG US UZ VN YU ZA ZM

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): BW GH GM KE LS MW MZ SD SL TZ UG ZM ZW AM AZ BY KG MD RU TJ TM AT BE BG CH CY DE DK EE ES FI FR GB GR HU IE IT MC NL PL PT RO SE SI SK TR BF CF CG CI CM GA GN GQ GW ML MR SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2005513992

Country of ref document: JP

122 Ep: pct application non-entry in european phase