WO2005086436A1

WO2005086436A1 - Packet transfer method, packet transfer network system, and terminal device

Info

Publication number: WO2005086436A1
Application number: PCT/JP2004/002617
Authority: WO
Inventors: Hirotaka Saito; Tetsushi Matsuda; Hiroshi Hayakawa; Mitsuru Tsuchida
Original assignee: Mitsubishi Denki Kabushiki Kaisha
Priority date: 2004-03-03
Filing date: 2004-03-03
Publication date: 2005-09-15
Also published as: JP4252596B2; JPWO2005086436A1

Abstract

A redundancy degree adaptive processing section (26) decides a redundancy degree according to feedback information from a transfer destination and generates feedback information to be reported to a transmission source according to an original packet received and the number of discarded redundant packets. A syndrome processing section (22) performs syndrome calculation according to the original packet identified by a flow identification section (27). A restoration processing section (23) restores the original packet by decoding. A redundant packet generation section (24) generates a redundant packet by using the syndrome calculation value, the restored original packet, and the decided redundancy degree. A selector (28) selects and outputs the received original packet, the restored original packet, and the generated redundant packet.

Description

Packet transfer device, packet transfer network system, and terminal device, technical field

The present invention relates to a bucket transfer device for transferring a bucket, a bucket transfer network system, and a terminal device. More specifically, the present invention relates to a packet clear transfer device for transferring a packet using forward error correction, and a bucket transfer network system. And terminal devices.

Food n

Background art

The technology for realizing data communication of files and the like that require reliability can be broadly classified as follows: Automatic retransmission request method (retransmission request when data received has an error)

There are two methods: 'AR Q: Automatic Repeat reQuest) method and FEC (Forward Error Correction) method, which corrects errors when there is an error in data.

-A typical example of the ARQ method is TCP (Transmission Control Protocol). TCP secures its reliability by retransmitting the packet when the packet is discarded. In addition, TCP has functions such as window control, a slow start algorithm, and a fast retransmission algorithm to avoid congestion and suppress packet discarding.

On the other hand, the FEC scheme is a scheme in which an original packet consisting of original data to be transmitted and a redundant packet for error correction are transmitted together. For example, the sender generates 11—k (k <n, n is a natural number) redundant packets from k. (1 k, k is a natural number) original packets in the original packet, and Send packets. If the receiving side can receive k packets out of the transmitted n packets, it recovers the discarded bucket from the received k buckets. Therefore, there is no need to make a retransmission request as in the ARQ scheme.

In Patent Document 1, the receiving terminal measures the number of discarded packets, feeds back the measurement result to the transmitting terminal, and the transmitting terminal determines the number of redundant packets (redundancy in the FEC scheme) based on the number of discarded buckets fed back. ) Is determined, and a technology related to a data distribution control method adapted to the optimum redundancy is disclosed.

Patent Document 1

JP 2002-330118 A

However, in the conventional technology described in Patent Document 1, feedback is performed between the end and the end (between the transmitting terminal and the receiving terminal). Therefore, in a WAN (Wide Area Network) having a large network scale, In a situation where a packet is transmitted from a transmitting terminal to a receiving terminal via a plurality of data transfer devices with a long delay time, it is not possible to sufficiently cope with a rapidly changing network condition.

In networks with long distances and large bandwidths, such as LFN (Long Fat Network), if the quality of the network is not good, control is performed to avoid congestion and the bandwidth cannot be used up sufficiently (performance is poor). Not). In order to improve such problems, it is conceivable to apply the FEC method to the lower layer of TCP to lower the apparent loss rate.

However, in this case, despite the congestion occurring, the packets discarded by the FEC method are restored, so there is no packet discarding from the TCP perspective (congestion Is determined not to have occurred), but congestion control does not work. For this reason, there is a problem that the normal TCP without the FEC method is adversely affected.

Also, in order to detect congestion only when the degree of congestion is higher (congestion of Lenorée where buckets cannot be restored by FEC), by applying FEC to the lower layer of TCP, There was also a problem that the performance might deteriorate due to force. SUMMARY OF THE INVENTION The present invention has been made in view of the above, and has been made in consideration of the above, and has been made in consideration of the above, and has been achieved in a packet transfer device and a packet transfer network that realize highly reliable communication by optimizing the redundancy of the FEC method even in a large-scale network such as a WAN. The primary purpose is to obtain a system and a terminal device.

The second object is to provide a packet transfer device, a packet transfer network system, and a packet transfer device capable of performing congestion control without deteriorating network performance even when the FEC method is used in a lower layer of TCP. It is to obtain a terminal device. Disclosure of the invention

In the present invention, a redundant packet for forward error correction is generated based on an original packet and a redundant packet received from a transfer source via a network, and the generated redundant packet and the original packet are transmitted to a network. A bucket transfer device for transmitting to a transfer destination via a transmission destination, when performing decoding and encoding for the forward correction, determines redundancy based on feedback information from the transfer destination; And a forward error correction processing unit that generates a build-pack information to notify the transfer source based on the reception state of the original packet and the redundant packet received from the transfer source.

According to the present invention, when performing decoding and encoding for forward error correction, the redundancy is determined based on the feed pack information from the transfer destination, and the original bucket and the original bucket received from the transfer source are determined. Feed pack information to be notified to the transfer source is generated based on the reception state of the redundant packet. Brief Description of Drawings

FIG. 1 is a conceptual diagram showing an example of the configuration of a network to which the packet transfer device according to the first embodiment of the present invention is applied, and FIG. 2 is a conceptual diagram showing an example of the FEC node shown in FIG. FIG. 3 is a block diagram showing the configuration of the FEC processing unit. FIG. 3 is a flowchart for explaining the operation of the FEC processing unit of the FEC node according to the first embodiment of the present invention. FIG. 4 is a flowchart for explaining the operation of the FEC processing unit of the FEC node according to the first embodiment of the present invention, and FIG. 5 is a second embodiment of the present invention. FIG. 6 is a block diagram showing a configuration of an FEC processing unit included in the FEC node of FIG. 6. FIG. 6 is a block diagram showing a restoration processing unit and a redundancy processing unit based on the redundancy of a received encoded block and the required redundancy. FIG. 7 is a diagram showing a relationship between processing requests to an adjustment unit. FIG. 7 is a flowchart for explaining the operation of the FEC processing unit of the FEC node according to the second embodiment of the present invention. FIG. 9 is a flowchart for explaining the operation of the FEC processing unit of the FEC node according to the second embodiment of the present invention. FIG. 9 shows a redundant packet generation process of the FEC node according to the third embodiment of the present invention. FIG. 10 is a block diagram showing the configuration of the FIG. 11 is a flowchart for explaining the operation of the redundant packet generation processing unit of the FEC node according to the third embodiment of the present invention. FIG. 11 is a thin flowchart of the FEC node according to the fourth embodiment of the present invention. FIG. 12 is a block diagram showing the configuration of the syndrome processing unit. FIG. 12 is a flowchart for explaining the operation of the syndrome processing unit 22 of the FEC node according to the fourth embodiment of the present invention. FIG. 14 is a diagram for explaining the relationship between the measured time of the timer and the received packet. FIG. 15 is a block diagram showing the configuration of the FEC processing unit of the FEC node according to the fifth embodiment of the present invention, and FIG. 16 is a FEC node of the fifth embodiment of the present invention. FIG. 17 is a diagram showing the timing of a reception bucket and a transmission bucket in the FEC processing section of FIG. 17.FIG. 17 is a block diagram showing the configuration of the FEC processing section of the FEC node according to the sixth embodiment of the present invention. Fig. 18 is a flowchart for explaining the operation of the FEC processing unit of the FEC node according to the sixth embodiment of the present invention. Fig. 19 is a flowchart showing the operation of the FEC node according to the seventh embodiment of the present invention. FIG. 20 is a diagram showing the relationship between the received packet and the transmitted packet of FIG. 20. FIG. 20 is a diagram for explaining the operation of the FEC node according to the eighth embodiment of the present invention, and FIG. Form 9 shows the configuration of the FEC processing unit of the FEC node. FIG. 22 is a conceptual diagram showing the configuration of a network according to the tenth embodiment of the present invention. FIG. 23 is a FEC node 12 of the tenth embodiment according to the present invention. FIG. 24 is a block diagram showing the configuration of terminal a. FIG. 24 is a block diagram showing the configuration of a terminal according to Embodiment 11 of the present invention. FIG. 25 is a block diagram showing Embodiment 12 of the present invention. FIG. 26 is a block diagram showing a configuration of a terminal according to Embodiment 13 of the present invention. FIG. 27 is a block diagram showing a configuration of a terminal according to Embodiment 13 of the present invention. FIG. 4 is a block diagram illustrating a configuration of an FEC processing unit of the FEC node of FIG. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of a packet transfer device, a packet transfer network system, and a terminal device according to the present invention will be described in detail with reference to the drawings. The present invention is not limited by the embodiment.

Embodiment 1

FIG. 1 Embodiment 1 of the present invention will be described with reference to FIG. FIG. 1 is a conceptual diagram showing an example of a network configuration to which a bucket transfer device according to a first embodiment of the present invention is applied. The network to which the packet transfer device according to the first embodiment of the present invention is applied includes terminals 11 and 13 and a plurality of (three in this case) FEC decoding / encoding (packet restoration / redundant bucket generation). ) That perform bucket transfer (hereinafter referred to as FEC nodes) 12a, 12b, and 12c, and an intermediate network '14a and 14b. '.

'First, a case where terminal 11 (hereinafter, referred to as transmitting terminal 11) transmits a packet to terminal 13 (hereinafter, referred to as receiving terminal 13) will be described as an example. An outline of a network to which the FEC node is applied will be described.

The transmitting terminal 11 1 forwards the packet: ^, determines the redundancy based on the feedback information 16 a from the FEC node 12 a, generates a redundant bucket, and transmits the generated redundant packet together with the original packet. I do. FEC node 1 2 a, 1 2 b and 12 c are the original buckets discarded by the FEC processing to restore the discarded original bucket, and feed pack information from the downstream (in the case of the FEC node 12 a, feedback information 16 b from the FEC node 12 ′ b, In the case of the FEC node 12b, the feed pack information is 16c from the FEC node 12c.In the case of the FEC node 12c, the redundancy is determined based on the feed pack information 16d from the receiving terminal 13). A bucket is generated, and the generated redundant bucket is transmitted together with the original bucket. In other words, the FEC nodes 12a to l'2c force restore the discarded original packets, and generate a redundant bucket with the redundancy determined based on the downstream feedpack information 16b to 16d. The original packet and the redundant packet are transmitted, and the packet is finally transferred to the receiving terminal 13. The receiving terminal 13 also restores the discarded original bucket by FEC processing. Here, an example is shown in which the transmitting terminal 11 and the receiving terminal 13 also perform FEC processing.The transmitting terminal 11 and the receiving terminal 13 perform -only normal transfer, and the FEC node 12a and the FEC node 12 FEC processing may be performed only during c. , '

Next, the FEC nodes 12a to 12c of the first embodiment shown in FIG. 1 will be described. The FEC nodes 12a to l2c all have the same function. The function of the FEC node will be described with reference to a block diagram showing the configuration of the FEC node 12a shown in FIG. Note that the FEC node 12a includes a plurality of processing units that realize general functions related to packet transmission and reception. In FIG. 2, the FEC node 12a performs FEC decoding / encoding according to the present invention. Only the components of the FEC processing unit 21 relating to the forward error correction processing (hereinafter referred to as FEC processing) to be performed are described.

The FEC processing unit 21 is input from a syndrome processing unit 22 that performs syndrome calculation of FEC, a restoration processing unit 23 that restores a discarded packet, a syndrome processing unit 22, a restoration processing unit 23, and a redundancy adaptation processing unit 26. Bucket generation processing unit 24 that generates a new redundant bucket based on the information The redundancy at the time of transmission is determined based on the syndrome memory 25 that accumulates the intermediate results of the processing, the downstream FEC node 12 b power, and the feed pack information 16 b, and the syndrome processing unit 22 and the redundant bucket generation processing unit 2 And the information such as the packet discard rate from the syndrome processing unit 22 is transmitted to the upstream node (^^ of the FEC node 12a indicates that the transmitting terminal 11 is the upstream node and the FEC node 1 2b In the case of, the FEC node 12a is the upstream node, and in the case of the FEC node 12c, the FEC node 12b is the upstream node.) Selects one packet from the three types of packets, the original packet received, the restored original packet, and the redundant packet, and transmits the selected packet sequentially. Selector 28. ,

Next, the operation of the FEC processing unit 21 of the FEC node according to the first embodiment of the present invention will be described using the FEC node 12a as an example. Upon receiving the feed pack information 16b from the downstream FEC node 12b, the redundancy adaptation processing unit 26 determines the redundancy based on the feed pack information .16 (step S100). The redundancy adaptation processing unit 26 notifies the determined redundancy to the syndrome processing unit 22 and the redundant packet generation processing unit 24.

Upon receiving the packet, the flow identifying unit 27 determines whether the received packet is a FEC flow packet or not (step S101). If the received packet is not a packet of the FEC flow, the received packet is transmitted via the selector 28 without performing the FEC-related processing.

If the received packet is a packet of the FEC flow, the first identification unit 27 accesses the syndrome memory 25, and the accessed syndrome memory 25 transmits the held intermediate result of the syndrome calculation to the syndrome processing unit. Output to 22 (step S102).

Syndrome processing ¾522 calculates the syndrome from the result of the syndrome calculation from the syndrome memory 25 and the received packet, and synthesizes the calculation result. It is stored in the drome memory 25 (steps S103, S104).

The syndrome processing unit 22 stores the sequence number of the received bucket for each flow (step S105).

The final packet determines whether the received packet is a FEC flow packet or not, and if the received packet is an FEC flow packet, the syndrome processing unit 22 calculates the syndrome by the final bucket. Repeat until received (steps S101 to S106). '

When the syndrome calculation is completed after receiving the last packet, the syndrome processing unit 22 extracts the missing sequence number from the stored packet sequence number, that is, the sequence number of the discarded packet (step S107). The syndrome processing unit 22 outputs the syndrome calculation value and the sequence number of the discarded bucket to the restoration processing unit 23 and the redundant bucket generation processing unit 24, calculates the packet loss rate, and calculates the redundancy adaptive processing unit. Output to 26.

The restoration processing unit 23 restores the discarded original bucket based on the syndrome calculation value and the discarded bucket sequence number (step S108). The restoration processing unit 23 outputs the restored original packet to the redundant packet generation processing unit 24 and the selector 28.

The redundant packet generation processing unit 24 generates a redundant packet based on the syndrome calculation value, the discarded sequence number, and the restored original packet (Step S109). The redundant bucket generation processing unit 24 outputs the generated redundant packet to the selector 28.

The selector 28 determines whether the input packet is a received original packet (an original packet that has not been discarded), an original packet restored by the restoration processing unit 23, or a redundant packet generation processing unit. 24, identifies whether the packet is a redundant packet generated, and transmits the received original nano packet, the restored original bucket, and the redundant packet in that order.

When the discarded packet loss rate is input, the redundancy adaptation processing unit 26 The feedback information 16a is generated based on the packet loss rate input from the roaming processing unit 22, and the generated feedback information 16a is transmitted to the downstream (in this case, the transmitting terminal 11) (step S110).

Next, a specific example of packet restoration and change of redundancy will be described with reference to FIG. In the fourth section, four original packets 0 # 1 to 0 # 4 and two redundant packets R # 1 and R # 2 are to be input to the FEC node 12a. 3 is discarded, and three original packets O # 1, O # 2, 0 # 4 and two redundant packets R # 1, R # 2 are input. That is, the £ 0 node 12 & receives the coded block (the received packet of the same flow, in this case, five buckets) in which the original packet 0 # 3 of the sequence number 3 is discarded.

The original packets 0 # 1, 0 # 2, 0 # 4 of the sequence numbers 1, 2, 4 are fed from the desired port via the selector 28 immediately after the syndrome calculation is performed by the syndrome processing unit 22. Sent to node 12b.

After receiving the redundant packet R # 2, which is the maximum packet, the original bucket O # 3 of the sequence number 3 restored by the restoration processing unit 23 receives the FEC node from the desired port through the selector 28. Sent to 12b. Subsequently, new packets R # 1 to R # 3 generated by the redundant packet generation processing unit 24 are transmitted from the desired port to the FEC node 12b via the force selector 28. In addition, the order in which the order of the original packets is changed. Finally, the original packet is rearranged at the same time as the original packet is restored, so that the order inversion of the bucket in the middle may be corrected.

As described above, in the first embodiment, the transmitting terminal 11 and the receiving terminal 13 do not notify the feedback information, but the FEC nodes 12 a to l disposed between the transmitting terminal 11 and the receiving terminal 13. 2c notifies the upstream node of the feed pack information of the own node to the upstream node, determines redundancy based on the feed pack information notified from the node downstream of the own node, and executes FEC nodes 12a to 12c it Since each discarded original bucket is restored, bucket transfer with optimal redundancy and high reliability can be performed even in networks with long delay times such as WAN.

When the FEC function implemented in the terminal is simply applied to an intermediate node, the packet of the encoded block (for example, the original packets 0 # 1, 0 # 2 , 0 # 4, R #l, R # 2) must be accumulated, so a large amount of memory is required for packet recovery processing S, and it is difficult to handle many flows due to memory limitations. there were. However, according to the first embodiment of the present invention, since only the number of redundant packets (intermediate result of the syndrome calculation) is stored in the syndrome memory, the discarded original bucket is restored with a small amount of memory. It is possible to handle more flows with the same memory capacity than when simply applying the FEC function implemented in the terminal. ,

Embodiment 2.

Embodiment 2 of the present invention will be described with reference to FIGS. 5 to 8. FIG. In Embodiment 1, the encoding itself (redundant packet) is changed to match the redundancy, but in Embodiment 2, the encoding (redundant packet) is the same, and the redundancy is reduced by discarding or restoring the bucket. Is to change. '

The network to which the packet transfer according to the second embodiment of the present invention is applied is the same as the network to which the bucket transfer device according to the first embodiment shown in FIG. 1 is applied, and the description thereof is omitted here. .

FIG. 5 is a block diagram showing a configuration of an FEC processing unit 21a provided with FEC nodes 12a to 12c according to the second embodiment of the present invention. The FEC processing unit 21a according to the second embodiment of the present invention shown in FIG. 5 replaces the redundancy adaptation processing unit 26 of the FEC processing unit 21 of the first embodiment shown in FIG. It has a redundancy adaptation processing unit 26 a and a redundancy adjustment unit 29 instead of the redundancy bucket generation processing unit 24. The components having the same functions as the FEC processor 21 of the first embodiment shown in FIG. The same reference numerals are given and duplicate description will be omitted.

The redundancy adaptation processing unit 26a determines the redundancy at the time of transmission based on the feedback information from the downstream, notifies the determined redundancy to the syndrome processing unit 22 and, at the same time, the syndrome processing unit 22 In addition to the function of transmitting information such as the bucket discard rate from the upstream as feedback information, the redundancy of the received coded block (the number of packets discarded from the redundancy of the Value) and the redundancy requested from downstream, determine whether to output a processing request to the restoration processing unit 23 and whether to output a processing request to the redundancy adjustment unit 29. I do.

FIG. 6 shows the relationship between the processing requests to the restoration processing unit 23 and the redundancy adjustment unit 29 based on the received redundancy of the encoded block and the required redundancy. As shown in Fig. 6, when the redundancy R i of the code block that can be received is larger than the redundancy Ro required from the downstream, that is, the received data is smaller than the number of packets requested downstream. If the number of buckets is large, the redundancy is excessive, and there is no need to restore the discarded original packet. Further, the bucket needs to be discarded according to the required redundancy Ro. Therefore, the redundancy adaptation processing unit 26 a does not output the processing request to the restoration processing unit 23, but outputs the number of packets to be transmitted to the redundancy adjustment unit 29. It outputs a processing request to discard the bucket so that it becomes Ro.

If the redundancy R i of the received encoded block and the required redundancy Ro are equal, that is, if the number of buckets requested downstream and the number of received buckets are equal, the original There is no need to restore the bucket or discard the bucket, and the received packet can be transmitted as it is. Therefore, the redundancy adaptation processing unit 26 a does not output a processing request to the restoration processing unit 23 and the redundancy adjustment unit 29.

If the redundancy R i of the received coded block is smaller than the redundancy R o requested from the downstream, that is, if the number of received packets is smaller than the number of packets requested downstream, the redundancy is used. Original discarded due to lack of degree Need to restore the ket. Therefore, the redundancy adaptation processing unit 26a outputs a processing request to the restoration processing unit 23 to restore the discarded original packet so that the number of packets to be transmitted becomes the required redundancy Ro.

Next, the operation of the FEC processing unit 21a of the FEC node according to the second embodiment of the present invention will be described with reference to the flowcharts of FIGS. 7 and 8, taking the FEC node 12a as an example. '

Upon receiving the feed pack information 16b from the downstream FEC node 12b, the redundancy adaptation processing unit 26a determines the redundancy based on the feed pack information 16 (step S200 in FIG. 7). The redundancy adaptation processing unit 26a notifies the syndrome processing unit 22 of the determined redundancy.

'Upon receiving the packet, the flow identifying unit 27 determines whether the received packet is a FEC flow packet or not (Step S201 in FIG. 7). If the received bucket is not a bucket of the FEC flow, the received bucket is transmitted via the selector 28 without performing the processing related to the FEC.

If the received packet is a packet of the FEC flow, the flow identification unit 27 accesses the syndrome memory 25, and the accessed syndrome memory 25 sends the held intermediate result of the syndrome calculation to the syndrome processing unit 22. Output (Step S202 in FIG. 7).

The syndrome processing unit 22 calculates the syndrome from the result of the syndrome calculation from the syndrome memory 25 and the received packet, and stores the calculation result in the syndrome memory 25 (steps S203 and S204 in FIG. 7). The syndrome processing unit 22 stores the sequence number of the received bucket for each flow (Step S205 in FIG. 7).

The flow identification unit 27 determines whether the received packet is a FEC flow bucket.If the received packet is an FEC flow packet, the syndrome processing unit 22 calculates the syndrome and the final bucket is received. (Steps S201 to S206 in Fig. 7). When the syndrome calculation is completed after receiving the final packet, the syndrome processing unit 22 extracts the sequence number missing from the sequence number of the stored packet, that is, the sequence number of the discarded bucket (FIG. 7). Step S207). The syndrome processing unit 22 outputs the syndrome calculation value and the sequence number of the discarded packet to the restoration processing unit 23, calculates the packet discard rate, and outputs the packet discard rate to the redundancy adaptation processing unit 26a. . .

The redundancy adaptation processing unit 26a compares the redundancy R i of the received coded block, which is obtained by subtracting the number of discarded buckets from the redundancy of the coding, with the redundancy o requested from the downstream. I do. If the redundancy R i of the received encoded block is larger than the redundancy R o required from the downstream ^ (step S 208, Y es in FIG. 8), the redundancy adaptation processing unit 26 a outputs a processing request including the number of packets to be discarded so as to match the number of buckets to be transmitted with the requested redundancy Ro to the redundancy adjusting unit 29. The redundancy adjuster 29 discards the number of packets included in the processing request (step S209).

The redundancy R for which the received coding block redundancy R i is required. (Step S210, Yes in FIG. 8), the 'redundancy adaptation processor 26a restores the number of buckets to be transmitted to match the required redundancy Ro. A processing request including the number of original buckets to be processed is output to the restoration processing unit 23. The restoration processing unit 23 restores the number of original packets included in the processing request based on the syndrome calculation value and the sequence number of the discarded packet (step S211 in FIG. 8).

When the number of discarded packets is input, the redundancy adaptation processing unit 26a generates feed pack information 16a based on the bucket discard rate input from the syndrome processing unit 22 and generates The transmitted feed pack information 16a is transmitted downstream (in this case, the transmission terminal 11) (step S2 12 in FIG. 8).

Thus, in the second embodiment, instead of generating a new redundant bucket, based on the received coding block redundancy R i and the required redundancy Ro, Since the original packets that have been discarded are restored or the packets are discarded and the number of packets to be sent is adjusted, only the minimum necessary restoration of the original packets is performed, and the delay time of WAN etc. is reduced. Even in networks with long data traffic, highly reliable bucket transfer with optimal redundancy can be performed.

Embodiment 3.

Embodiment 3 and Embodiment 3 of the present invention will be described with reference to FIGS. 9 and 10. FIG. In the third embodiment, a redundant packet generation processing unit of an FEC node that generates a redundant bucket first will be described.

FIG. 9 is a block diagram showing the configuration of the redundant bucket generation processing section 24 shown in FIG. The redundant packet generation processing unit 24 includes an encoder 241, which generates a redundant bucket based on the redundancy at the time of transmission obtained by the redundancy adaptation processing unit 26 based on the feed pack information from downstream. A timer 242 for measuring the reception interval of the original nano bucket and notifying the encoder of the end of the encoding when the reception interval becomes equal to or greater than a preset threshold.

Next, referring to the flowchart in Fig. 10.10, based on the feedback information from the downstream, n (k <n, where n is a natural number) from .k (1 k, k is a natural number) original packets The operation of the redundant packet generation processing means according to the third embodiment will be described by taking as an example a case where one k redundant packets are generated and n packets are generated.

When receiving the original packet, encoder 2 4 1 performs a coding process for generating redundant packets (Step S 3 0 0, S 3 0 1) 0 coder 2 4 1 Fugoi匕After performing the processing, the original packet is output to the selector 28 (step S302).

Each time the encoder 2411 outputs the original packet to the selector 28, the encoder 2411 counts the number of output original packets, and outputs the k-th packet after the output original packet starts encoding. It is determined whether the packet is an original packet (step S303). The output original packet is not the kth original packet In this case, the encoder 241 resets the timer 242 (step S305), restarts the measurement of the reception interval of the timer 242, and performs encoding by the next received original packet. After performing the processing, the operation of outputting the original packet is repeated until the k-th bucket is output (steps S300 to S304). If the output original packet is the k-th original packet since the start of encoding, the encoder 2241 stops the timer 242 and then outputs the first original packet packet. A redundant packet is generated based on the result of the encoding process using the k-th original packet (step S305), and the generated redundant packet is output to the selector 28.

On the other hand, the original packet was not received, and a timeout was notified from the timer 242 ^^, that is, the original packet was received even if the measurement interval of the timer 242 exceeded the threshold. If not (step S306), the encoder 2441 generates a redundant bucket based on the result of the encoding process using the original bucket received before the notification of the time-out (step S306). S 305), and outputs the generated redundant packet to the selector 28.

As described above, in the third embodiment, when a redundant packet is generated first, the redundant packet generation processing unit 24 is provided with the timer 242 to measure the reception interval of the original bucket, and When time elapses, a redundant bucket is generated based on the result of the encoding process performed by the original buckets received up to that time. For example, the amount of information to be transmitted (the number of original packets) is determined. Even if it does not, a redundant bucket can be generated.

Although Embodiment 3 shows a method of transmitting n−k redundant packets even when the number of received original packets is less than k, n−k packets are transmitted according to the redundancy required by the feed pack information. It is also possible to discard some redundant packets from k and output them.

Embodiment 4.

Embodiment 4 of the present invention will be described with reference to FIGS. 11 to 13. FIG. Real In the third embodiment, the redundant packet generation processing unit 24 of the FEC node that performs encoding is described first. However, in the fourth embodiment, the intermediate FEC node, that is, the FEC node that receives the redundant bucket The following describes the case where the bucket including the last bucket is discarded at the node.

In the FEC node according to the fourth embodiment of the present invention, the syndrome processing unit 22 according to the first embodiment shown in FIG. 2 is configured as shown in FIG. The syndrome processing unit 22 of the FEC node according to the fourth embodiment of the present invention measures the syndrome reception unit 221, which calculates the syndrome from the received packet, and the bucket reception interval, and sets the measured reception interval in advance. A timer 222 for notifying the syndrome calculation unit 221 of the end of the syndrome calculation when the calculated threshold value is exceeded.

The operation of the syndrome processing unit 22 of the FEC node according to the fourth embodiment of the present invention will be described with reference to the flowchart in FIG. 12 and FIG. Upon receiving the packet, the syndrome calculation unit 221 resets the timer 222 and restarts the measurement of the reception interval of the timer 222 (steps S4'00, S401).

) o '

The syndrome is calculated from the intermediate result of the syndrome calculation stored in the syndrome memory 25 and the received packet, and the calculation result is stored in the syndrome memory 25 (steps S402 and S403).

The syndrome calculation unit 221 stores the sequence number of the received bucket for each flow (step S404).

Upon receiving the bucket, the syndrome calculation unit 221 resets the timer 222 and repeats the operation of calculating the syndrome until the last packet is received (steps S400 to S405).

When the last packet is received, the syndrome calculation unit 221 stops the timer 222 and searches for a sequence number missing from the stored sequence number, that is, a sequence number of a discarded packet, and searches for the discarded bucket. No one A cans number is extracted (step S406).

On the other hand, when a packet is not received and a notification of a timeout is sent from the timer 222, that is, a packet is not received even if the reception interval measured by the timer 222 exceeds the threshold value. (Step S407), the sequence number missing from the stored sequence number, that is, the sequence number of the discarded packet is searched, and the sequence number of the discarded bucket is extracted (Step S406). . The restoration processing unit 23 restores the original bucket of the extracted sequence number. The redundant packet generation processing unit 24 generates a redundant packet based on the received original packet and the restored original bucket. .

The relationship between the measurement time of the timer 222 and the reception bucket will be described with reference to FIG. 13. As shown in Fig. 13, four original buckets O # 1 to 0 # 4 and three redundant buckets R # 1 to R # 3 were transmitted from the transmitting terminal 11 1 as coding blocks. Assume that the original packet O # 4 and the redundant packets R # 2 and R # 3 are discarded on the way. That is, suppose that redundant packet R # 3, which is the last packet, has been discarded.

The timer 222 is reset when the original packet O # 3 is received, and restarts the measurement of the reception interval. The original bucket 0 # 4 has been discarded.Next, since the redundant packet R # 1 is received, the timer 222 is reset when the redundant packet R # 1 is received, and the measurement of the reception interval is performed. Restart. However, since both redundant packets R # 2 and R # 3 to be received after redundant packet R # 1 have been discarded, no packet is received after redundant packet R # 1. Therefore, the reception interval of the timer 222 exceeds the threshold, and a notification of the timeout is output to the syndrome calculation unit 221. As a result, even when the final packet R # 3 cannot be received, the syndrome calculation unit 22i can end the calculation of the syndrome at the reception interval.

As described above, in the fourth embodiment, the syndrome processing unit 22 is provided with the timer 22 to measure the packet reception interval, and the measured packet reception interval is equal to or smaller than the threshold. If the above time elapses, it is determined that the following packet has been discarded, and the packet is calculated based on the syndrome calculated by the received bucket, and the packet is restored and the redundant packet is generated. Even if the packet is discarded, 'the original discarded packet can be restored and re-encoded. '' Embodiment 5.

Embodiment 5 of the present invention will be described with reference to FIGS. FIG. 14 shows the timing of a reception packet and a transmission packet in the FEC processing unit 21 of the FEC node according to the first embodiment shown in FIG. In FIG. 14, reception is performed by encoding (10, 6) (original packets: 6 from 0 # 1 to 0 # 6, redundant packets: 4 from R # 1 to R # 4), Orientation packet # 4 and redundant packet R # 2: R # 3 is input to the FEC node in a discarded state, and is re-encoded and transmitted at (9, 6). The timing is shown.

As described above, the FEC processing unit 21 according to the first embodiment receives and transmits a bucket while performing syndrome calculation, receives all buckets that have not been discarded, and then discards the original packet. Has been restored. That is, as shown in FIG. 14, from the time when the original bucket 〇 # 1 which is the first bucket is received until the time when the redundant packet R # 4 which is the last packet is received, the syndrome processing unit 22 Each time the original packets 0 # 1 to 0 # 3, 0 # 5, 0 # 6 and the redundant packets R # l, R # 4 are received, the syndrome calculation is performed. After the syndrome calculation of the redundant packet R # 4 is completed, the original packet O # 4 is restored after restoring the original packet O # 4 discarded by the restoration processing unit 23, and then returning to the redundant packet. The generation processing unit 24 generates redundant packets R # 1 'to R # 3. As a result, there was a problem that it took time from transmitting the last received original packet 0 # 6 to transmitting the restored original packet 0 # 4.

In the fifth embodiment, in order to improve such a problem, When the required number of buckets to receive the null packet are received, the restoration of the discarded original bucket starts, and the original bucket received last is transmitted, and then the restored original bucket is transmitted. To reduce the time it takes '

FIG. 15 is a block diagram showing a configuration of an FEC processing unit 21b included in the FEC nodes 12a to 12c according to the fifth embodiment of the present invention. The FEC processing unit 21b of the fifth embodiment according to the present invention includes a function of calculating the number of buckets received for each coding block (flow) by the FEC processing unit 21 of the first embodiment shown in FIG. , A reception packet counter 30 is added. Components having the same functions as those of the FEC processing unit 21 of the first embodiment shown in FIG. 2 are denoted by the same reference numerals, and redundant description will be omitted.

Next, the operation of the FEC processing unit 21b according to the fifth embodiment of the present invention will be described with reference to FIG. When the syndrome calculation is completed after receiving the final packet, the sequence number of the discarded packet is extracted from the stored sequence number. The sequence number of the discarded bucket and the syndrome calculation value are output to the restoration processing section 23 and the redundant bucket generation processing section 24 (see steps S106 and S107 in FIG. 3). On the other hand, the FEC processor 21b counts the number of buckets received by the reception packet 1 and the counter 30 for each encoding block, and calculates the number of buckets that can recover the original packet whose count value has been discarded. When the value becomes the value, the start of the restoration processing is notified to the syndrome processing unit 22 and the notified syndrome processing unit 22 is discarded from the sequence number stored immediately before receiving the notification. It extracts the sequence number of the extracted packet and outputs the extracted sequence number of the discarded packet and the syndrome calculation value to the restoration processing unit 23 and the redundant bucket generation processing unit 24. This operation is the same as the FEC process shown in the flowchart of FIG. 3, and thus detailed description is omitted here.

In Fig. 16, the first received bucket, the original packet 0 # 1, is received. Upon receipt, the reception bucket counter 30 starts counting the number of received packets, and the syndrome processing unit 22 starts syndrome calculation. .

Packets are received in the order of original packets 0 # 2, 0 # 3, 0 # 5, 0 # 6, and the syndrome processing section 22 sends the original packets 0 # 2, 0 # 3, 0 # 5, 0 # 6 Calculate the syndrome upon receiving.

When the redundant packet R # 1 is received, the count value of the received packet counter 30 becomes “6”, and when the minimum number of buckets for recovering the discarded original packet is reached, the received packet counter 30 becomes: The syndrome processing unit 22 is notified of the start of the restoration processing. Upon receiving the notification of the start of the restoration processing, the syndrome processing unit 22 extracts the sequence and the number of the discarded packet from the sequence number stored immediately before, and calculates the sequence number and the syndrome of the extracted discarded packet. The values are output to the restoration processing section 23 and the redundant bucket generation processing section 24. As a result, as shown in FIG. 16, after receiving the redundant packet R # 1, the restoration processor 23 restores the discarded original packet (in this case, the original packet 0 # 4). To start.

As described above, in the fifth embodiment, when the reception packet counter 30 counts the number of reception buckets in the flow and receives the minimum number of packets that can recover the discarded original packet, the syndrome processing unit 2 2 requests the start of the restoration process and starts the restoration process, so the received original packet is transmitted and then restored compared to ^ which starts the restoration process after receiving the final packet. The time required for transmitting the original packet can be reduced.

When the received packet counter 30 counts the number of received buckets of the flow, and receives the minimum number of buckets that can restore the discarded original packet, it requests the syndrome processing unit 22 to start the restoration processing. Since the restoration process is started, the time for the encoding block (flow) to use the syndrome memory 25 can be shortened, and the syndrome memory 25 can be used efficiently. However, the restoration process of the original bucket is basically to solve a system of linear equations.Since the system of linear equations has an order power 0 (n ³ ) for the number n of unknowns, it greatly depends on the number of unknowns. The amount of calculation differs. Therefore, the restoration time may be longer depending on the relationship between the packet transmission interval and the processing capability of the FEC processing means. For such ^, if a new packet is received during the calculation, the contents of the bucket should be reflected, and the number of unknowns should be reduced so that the time does not increase.

Embodiment 6

Embodiment 6 of the present invention will be described with reference to FIG. 17 and FIG. In the sixth embodiment, when the necessary data is already stored in the syndrome memory and it is not possible to execute the original bucket recovery processing or the redundant bucket generation processing, etc., no extra processing is performed. It reduces the processing load on FEC nodes.

FIG. 17 is a block diagram showing a configuration of the FEC processing unit 21c of the FEC node according to the sixth embodiment of the present invention. The FEC node 21c of the sixth embodiment of the present invention includes a flow identification unit 27a instead of the flow identification unit 27 of the FEC processing unit 21 of the first embodiment shown in FIG. A selector 28 a is provided instead of the selector 28. Components having the same functions as those of the FEC processing unit 21 of the first embodiment shown in FIG. 2 are denoted by the same reference numerals, and redundant description will be omitted.

The flow identification unit 27 a monitors the use state of the syndrome memory 25 in addition to the flow identification unit 27 of the first embodiment, and stores the original state in the restoration processing unit 23 in the syndrome memory 25. It is possible to determine whether there is an unused area that can be used for bucket restoration processing, redundant packet generation processing in the redundant packet generation processing unit 24, etc., and to execute restoration processing and redundant bucket generation processing. It also has the ability to judge whether it is strong or not. The flow identification unit 27 holds, for each flow, whether or not the restoration processing or the redundant bucket processing has been performed, as processing determination information.

The selector 28a is based on the processing determination information stored in the flow identification unit 27. And select and output the received original packet, restored original packet, or redundant bucket. '

Next, the operation of the FEC processing unit 21c according to the sixth embodiment will be described with reference to the flowchart in FIG. If there is a usable area in the syndrome memory 25, the operations for performing the syndrome calculation processing, the restoration processing, and the redundant packet generation processing are the same as the FEC processing of the flowchart shown in FIG. Since this is the operation, the description is omitted here.

The flow identification unit 27a monitors the usage status of the thin-drome memory 25, and receives the process determination information indicating whether or not the restoration process and the redundant packet generation process have been performed. Hold each time.

When transmitting a packet, the selector 28a determines, based on the processing identification information, that there is no free space in the roam memory (insufficient unused area) and no restoration processing or redundant bucket generation processing is performed. The power flow, which is the packet of the flow 否 IJ is cut off (step S500). The packet to be transmitted is a bucket of a flow for which the restoration processing or the redundant bucket generation processing has not been executed: ^, the selector 28a transmits the received packet as it is (step S501).

If the packet to be transmitted is a flow packet on which the restoration processing and the redundant bucket generation processing have been executed, the selector 28a determines whether the received bucket is an original packet “ ^{: i} ” packet (step S500). 2). If the received packet is an original bucket, the selector 28a transmits the received bucket as it is (step S501).

When there is no received bucket, or when the received bucket is not the original bucket, the selector 28a determines whether or not there is a packet that can be transmitted to the restoration processing unit 23 (the restored original bucket). (Step S503). When there is a packet that can be transmitted to the restoration processing unit 23, the selector 28a fibrate the bucket that can be transmitted by the restoration processing unit 23 (step S504). When there is no packet that can be transmitted to the restoration processing unit 23, the selector 28a The transmittable packet (redundant packet) of the packet generation processing unit 24 is transmitted (step S505).

As described above, in the sixth embodiment, since the unused area of the syndrome memory 25 is small, the packet of the flow for which the restoration processing or the redundant bucket generation processing cannot be executed is received from the upstream FEC node. Since the packet is sent to the downstream FEC node as it is, it is not possible to restore the original packet discarded by the own node! / Even in this case, restoration processing can be performed in the downstream FEC node. In other words, since the original nano buckets that have been dispersed and discarded throughout the network can be restored, many flows can be handled.

Embodiment 7.

Embodiment 7 of the present invention will be described with reference to FIG. First, the operation of the FEC node according to the seventh embodiment of the present invention will be described. FIG. 19 is a diagram showing a relationship between a reception bucket and a transmission bucket of the FEC node 12a according to the seventh embodiment of the present invention. In FIG. 19, reception is performed by coding (6, 5) (original packet is 個 # 1 to 0 # 5, and redundant bucket is R # 1). Two packets 0 # 2 and 0 # 4 are input to FEC node 12a in a discarded state. In this case, since more packets than the number of redundant buckets have been discarded, the £ 0 node 12 & cannot recover the discarded original packets 0 # 2 and 0 # 4. When the FEC node 12a determines that the received packet has received an irrecoverable number of discards (this is due to receiving the original packet 0 # 5 and the original packet 0 # 2, 0 # 4 When it is determined that this packet has been discarded), the syndrome processing is stopped, and only the original packets # 1, 0 # 3, and 0 # 5 are transferred, and the redundant packet R # l is discarded. ·

If the FEC node 12a has the FEC processing unit 21 shown in FIG. 2, the number of discarded packets is counted by the flow identification unit 27, and the count value is calculated. If the number of discardable buckets determined from the redundancy of the own node is exceeded, a processing stop notification is output to the syndrome processing unit 22 and only the original bucket of the received packets is transmitted to A function to output a process change notification for discarding the bucket to the selector 28 is provided.The syndrome processing unit 22 stops the syndrome process when receiving the process stop notification, and the selector 28 receives the process change notification. It is sufficient to transmit only the original packet and discard the redundant bucket. If the FEC node 12a has the FEC processing unit 21a shown in FIG. 5, the number of discarded packets is counted by the port identification unit 27, and the If the event value exceeds the number of discardable packets determined from the redundancy of the own node, a processing stop notification is output to the syndrome processing unit 22 and the redundancy adjustment unit 29, and the received packet A function to transmit only the original packet and output a processing change notification for discarding the redundant bucket to the selector 28 is provided. The syndrome processing unit 22 stops the syndrome processing upon receiving the processing stop notification, and When the processing change notification is received, only the original packet is transmitted and the redundant packet is discarded. When the processing stop notification is received, the redundancy adjusting unit 29 stops the processing and outputs the packet from the selector 28 as it is. It suffices to be.

If the FEC node 12a has the FEC processing unit 21c shown in FIG. 17, the number of packets discarded by the flow identification unit 27a is counted, and the If the value exceeds the number of discardable packets determined from the redundancy of the own node, a processing stop notification is output to the syndrome processing unit 22 and only the original packets out of the received packets are transmitted. A function that outputs a processing change notification for discarding redundant packets to the selector 28a.The syndrome processing unit 22 stops the syndrome processing upon receiving the processing stop notification, and the selector 28a transmits the processing change notification. Upon receipt, only the original bucket should be transmitted and the redundant bucket should be discarded.

As described above, in the seventh embodiment, the processing in each FEC node is performed by not performing extra processing for the flow in which the number of unrecoverable packets has been discarded. It can be reduced and can handle many flows.

In order to prevent the downstream FEC node from performing extra processing on the flows that have been irretrievably discarded, the original packets that have been determined to have irreparable discards should be May be set. In this case, the flow identification unit 27 identifies the flag of the received original bucket, and discards the unrecoverable number of packets for the original bucket determined to have an unrecoverable number of discards. May be performed in the same manner as when there is an error. This eliminates the need to count the number of discarded packets, further reduces processing, and can handle many flows. '

Embodiment 8.

Embodiment 8 of the present invention will be described with reference to FIG. The eighth embodiment of the present invention supports a class-based discard control such as Differentiated Services (Diffse erV).

To support discard control according to the class such as Diffsserv, the FEC node 12a stores packets in a queue and transmits the packets. As shown in FIG. 20 (a), the queue 40 stores two original packets 0 # 1 and 0 # 2. In the queue 40, a low-priority class discard threshold value X1 and a high-priority class discard threshold value X2 are set for discard control according to the class. Whether the queue 40 queues newly input packets according to the low priority class discard threshold X1 and the high priority class discard threshold X2, and the amount of accumulated packets and the class of the input packet. Decide what to discard. In Fig. 20 (a), the number of buckets stored in the queue 40 is two, and does not exceed the low-priority class discard threshold X1 (because congestion has not occurred). Original packets 0 # 1 and 0 # 2 remain as they are.

In FIG. 20 (b), three original packets 0 # 3, 0 # 5, 0 # 6 are further accumulated, and five original packets O # 1 to 0 # 3, 0 # 5 are stored in the queue 40. , 0 # 6 are stored, and the number of packets stored in the queue 40 is low priority. It exceeds the degree class discard threshold _x1 , but does not exceed the high priority class discard threshold _X2 . In other words, clothing in the low priority class has occurred. In this case, only packets of the lower priority class are dropped. Therefore, if the accumulated original packets O # 1 to # 5 are in the low-priority class, the original packets 0 # 1 to 0 # 5 are discarded, and if they are in the high-priority class, they are not discarded.

In FIG. 20 (c), two redundant buckets R # 1 and R # 2 are further accumulated, and the queue 40 has five original packets 0 # 1 to 0 # 5 and two A total of 7 packets including the redundant packets R # l and R # 2 are accumulated, and the number of packets accumulated in the queue 40 exceeds the high priority class discard threshold X2. That is, congestion occurs in the high priority class. In this case, drop all packets.

Thus, bucket 1, transport network supports discard control according to class such as Diffserv ^^, original bucket is mapped to high priority class, and redundant packets are mapped to low priority class Then, the discard of the original 'packet can be suppressed.

The selector 28 of the FEC processor 21 shown in FIG. 2, the selector 28 of the FEC processor 21 a shown in FIG. 5, the selector 28 of the FEC processor 21 b shown in FIG. 15 When selecting a packet to be transmitted, the selector 28a shown in Fig. 17 maps a high-priority class to the received original packet and the restored original packet, and a constant priority to the redundant packet. It should have a function to map the degree class.

Thus, in the eighth embodiment, the packet forwarding network supports discard control according to the class: ^, mapping the original bucket to the high priority class and mapping the redundant packet to the low priority As a result, the discard of original packets can be suppressed, the number of restoration processes can be reduced, the load of FEC processing can be reduced, and many flows can be handled. Embodiment 9 Embodiment 9 of the present invention will be described with reference to FIG. FIG. 21 is a block diagram showing a configuration of an FEC processing unit 21c included in the FEC node 12a according to the ninth embodiment of the present invention. The FEC processing unit 21c of the FEC node 12a according to the ninth embodiment of the present invention transmits a monitor packet from the upstream to the FEC processing unit 21 included in the FEC node 12a of the first embodiment shown in FIG. A monitor bucket receiving unit 31 for receiving and a monitor bucket generating unit 32 for generating a monitor bucket to be transmitted downstream at regular intervals are added.

In the first embodiment, the packet discard rate and the like are measured from received packets (normal user packets). In the ninth embodiment, the measurement of the packet discard rate and the like are performed using the monitor packet.

The monitor bucket generation unit 32 of the upstream FEC node 12a generates a monitor bucket and transmits it to the downstream FEC node 12b at regular time intervals. The monitor packet receiving unit 31 of the FEC node 12b calculates the number of discarded packets (discard rate) from the sequence number added to the received monitor packet. The monitor bucket receiving unit 32 of the FEC node 12b notifies the calculated bucket discard rate to the redundancy adaptation processing unit 26, and the redundancy adaptation processing unit 26 performs a field pack as in the first embodiment. It generates information and sends the generated feedback information to the upstream FEC node 12a. '

As described above, in the ninth embodiment, a monitor packet is transmitted at regular intervals in order to generate the buried pack information, so that the state of the network can be known even when the user bucket is not transferred. Thus, even when the user bucket is intermittently transferred, the bucket can be transferred with an appropriate redundancy.

Embodiment 10.

Embodiment 10 of the present invention will be described with reference to FIG. 22 and FIG. In the tenth embodiment, as shown in FIG. 22, when the FEC node 12a copies a packet to the FEC nodes 12b and 12c and performs multicast transfer. Will be described.

FIG. 23 is a block diagram showing a configuration of an FEC node l ′ 2 a according to the tenth embodiment of the present invention. The FEC node 12a according to the tenth embodiment of the present invention includes, in addition to the FEC processing unit 21 according to the first embodiment shown in FIG. 2, a redundant packet in accordance with a plurality (three in this case) of redundancy. And a queue 34a-34c for storing a bucket for each destination. Here, it is assumed that the queue 34a corresponds to the FEC node 12b, and the queue 34b corresponds to the FEC node 12c.

In Fig. 22, FEC node Γ2b requests 40% redundancy as feedback information 16b from FEC node 12a, and FEC node 12c feeds back to FEC node 12a. Information 16c requires 50% redundancy. .

An encoding block of three original buckets O # 1 to 0 # 3 and one redundant packet R # 1 is input to the FEC node 12a. In the FEC node 12a, the FEC processing unit 21 of the FEC node 12c, which requires a high degree of redundancy, re-encodes based on the feed pack information 16c to generate a new redundant packet. '. In this case, the number of the original buckets is 3, and the redundancy requested from the FEC node 12c by the feedback information 16c is 50%, so that the FEC processing unit 21 of the FEC node 12 , And three redundant packets R # 1 to R # 3. The FEC processing unit 21 includes a redundancy adjustment unit 33a located upstream of the queue 34a corresponding to the FEC node 12b, and a redundancy adjustment unit 33b located upstream of the queue 34b corresponding to the FEC node 12c. Then, original packets 0 # 1 to 0 # 3 and redundant packets R # l, to R # 3 'are output.

Since the redundancy requested from the FEC node 12b is 40%, the redundancy adjusting unit 33a discards one redundant packet (here, redundant packet R # 2 ') and replaces the original packet 0 # 1 to 0 # 3 and redundant packets R # 1, R # 3 are output to queue 34a. The required redundancy from FEC node 12c is 50% Therefore, the redundancy adjusting unit 34b outputs the original buckets 0 # 1 to 0 # 3 and all the redundant packets R # 1 'to R # 3 generated by the FEC processing unit 21 to the queue 34b. That is, the FEC node 12a generates a redundant packet with the highest required redundancy among a plurality of pieces of feedback information, and then discards the redundant packet according to the redundancy for each destination.

As described above, according to the tenth embodiment, in the multicast communication, when the request of the redundancy is different, instead of generating the redundant packet for each destination, the redundant packet is generated in accordance with the higher required redundancy. Redundant packets are discarded in accordance with the redundancy of each destination and individual redundancy is adjusted, so that FEC processing can be reduced and many flows can be handled.

Embodiment 11.

Embodiment 11 of the present invention will be described with reference to FIG. FIG. 24 is a block diagram showing components related to the TCP transmission function of terminal 11 on the transmitting side and components related to the TCP reception function of terminal 13 on the receiving side according to the eleventh embodiment of the present invention. . Here, for the sake of simplicity, only the components related to the transmission function are shown in the terminal 11 on the transmission side, and only the components related to the reception function are shown in the terminal 13 on the reception side. Each of the terminals 11 and 13 has a transmission function and a reception function.

The terminal 11 includes a TCP transmission processing unit 111 that performs processing related to TGP transmission, an FEC encoding unit 112 that performs FEC encoding on data from the TCP transmission processing unit 111 to generate a redundant packet, A layer 13 processing unit 113 for performing layer 13 transmission processing is provided.

The terminal 13 includes a layer 1-3 processing unit 133 for performing layer 1-3 reception processing, an FEC decoding unit 132 for performing recovery processing of discarded packets, and a TCP for performing processing related to TCP reception. Receive processing unit 131 and TC based on bucket discard status

A window size changing unit 134 for rewriting the window size of the TCP ACK transmitted from the P reception processing unit 131 is provided. Next, the operation of the eleventh embodiment of the present invention will be described. The transmission-side TCP transmission processing section 111 performs transmission processing on TCP on transmission data, and outputs the transmission-processed data to the FEC encoding section 112. FEC encoding section 112 performs FEC encoding on the data subjected to the transmission processing to generate a redundant packet. The layer 13 processing unit 113 transmits the original packet and the redundant packet as user data to the terminal 13 via the layer 13 processing unit 113. The FEC decoding unit 132 of the receiving terminal 13 restores the discarded original packet based on the original packet and the redundant bucket received via the layer 13 processing unit 133. The FEC decoding unit 132 outputs the received original packet and the restored original packet to the TCP reception processing unit 131, and outputs the packet discard status (information such as the discard number and discard rate) to the TCP connection. The result is stored every time, and the result is notified to the window size changing unit 134.

The window size changing unit 134 rewrites the window size of the TCP ACK transmitted from the TCP reception processing unit 131 based on the notified bucket discarding state, and Send to

In terminal 11, TCP transmission processing section 111 changes the transmission band based on the window size of ACK received via layer 13 processing section 113. For example, when there is no discard, the window size changing unit 134 transmits the ACK from the TCP reception processing unit 131 as it is, and reduces the ACK window size every time discard increases. Thereby, the transmission band from terminal 11 on the transmission side can be reduced.

Thus, in the eleventh embodiment, even when FEC is applied to the lower layer of TCP and the original packet is restored by FEC, the window size of the ACK is reduced according to the number of discarded packets. However, even with low quality LFN, performance can be extracted and congestion control can be performed appropriately, and highly reliable packet transfer can be realized.

Embodiment 12. Embodiment 12 of the present invention will be described with reference to FIG. FIG. 25 is a block diagram showing a configuration of terminal 11 and terminal 13 according to Embodiment 12 of the present invention. In the terminal 11 of the embodiment 12 of the present invention, a redundancy adaptation processing unit 114 is added to the terminal 11 of the embodiment 11 shown in FIG. A feed pack information transmission processing unit 135 is connected to the terminal 13 of the embodiment 11 shown in FIG. 24. '

That is, in addition to the function of the embodiment 11, the number of discarded packets and the discard rate detected by the FEC decoding unit 132 of the terminal 13 on the receiving side are represented by the feedback information transmission processing unit 1. 3 5 feeds back as feed pack information to the terminal 11 on the transmitting side via the layer 1-3 processing unit 13 3, and the redundancy adaptation processing unit 1 14 of the terminal 11 is based on this feed pack information. Determine redundancy.

Thus, in this Embodiment 12, even when FEC is applied to the lower layer of TCP and the original packet is restored by FEC, the window size of ACK is reduced according to the number of discarded packets, Redundancy is determined based on feedpack information such as the number of dropped packets and the loss rate, so that even low-quality LFNs can extract performance and perform congestion control appropriately. The occurrence of congestion can be suppressed, and highly reliable transfer of packets can be realized.

Embodiment 13

Embodiment 13 of the present invention will be described with reference to FIG. In Embodiment 12, the ACK of the TCP and the feed pack information are individually transmitted. In Embodiment 13, however, the ACK of the TCP and the feedback information are transmitted together.

FIG. 26 is a block diagram showing a configuration of terminal 11 and terminal 13 according to Embodiment 13 of the present invention. In the terminal 11 of the embodiment 13 of the present invention, a feed pack information extracting unit 1 15 is added to the terminal 11 of the embodiment 11 shown in FIG. Is the terminal 13 window of the embodiment 11 shown in FIG. An ACK rewriting unit 135 is provided instead of the window size changing unit 134.

ACK rewrite unit 135 The special packet generated by adding feedback information such as the number of discarded packets and the discard rate notified from the decoding unit 132 is generated.

The feedback information extracting unit 115 extracts the feedback information from the special packet including the ACK and the FEC feedback information. The feed pack information extraction unit 115 determines the redundancy based on the extracted feed pack information, notifies the FEC encoding unit 112 of the determined redundancy, and converts the special bucket into a normal ACK format. And outputs it to the TCP transmission processing unit 111.

Thus, in the thirteenth embodiment, the ACK rewriting unit 135 transmits the ACK and the feed pack information as one special packet, and outputs the feedback information from the special packet. Extraction and format conversion to normal ACK eliminates the need to send buckets to notify feedpack information, thus reducing the amount of traffic on the network.

Embodiment 14.

Embodiment 14 of the present invention will be described with reference to FIG. Embodiments 11 to 13 have described processing between terminals in TCP. In Embodiment 14, as shown in FIG. 1, a case will be described in which FEC nodes 12a to 12c are arranged between terminals 11 and 13, and the redundancy is changed for each section. . As shown in FIG. 1, when the FEC—Kl 2 a to 12 c is placed between the transmission terminal 11 and the reception terminal 13, the reception terminal 13 determines the number of discarded buckets and the number of discarded buckets in the entire network. The rate cannot be determined. Therefore, the window size of ACK cannot be adjusted to an appropriate value. Therefore, an information area for recording the discard number or the discard rate is provided in the transferred packet, and the receiving terminal 13 is notified of the discard number and the discard rate of the packets of the entire network. ,

FIG. 27 is a block diagram of the FEC node 12a according to the fourteenth embodiment of the present invention. FIG. 9 is a block diagram illustrating a configuration of an FEC processing unit 21d. The FEC processing unit 21d according to the fourteenth embodiment of the present invention includes a discard number updating unit 35 added to the FEC processing unit 21 according to the first embodiment shown in FIG. '' The discard number update unit 35 adds the discard number of the packet notified from the syndrome processing unit 22 to the discard number of the received bucket information area, and calculates the discard number of the bucket discarded up to the own node in the information area. To be added.

As shown in FIG. 1, when a packet is transmitted from the transmitting terminal 11 to the receiving terminal 13, no packet is discarded at the transmitting terminal 11, so that a packet having an information area of "0" is a FEC node. Entered in 12a. If the syndrome processing unit 22 of the FEC node 12a detects discard of two packets, the discard number update unit 35 of the FEC node 12a sends the packet with the information area of "2" to the FEC node 12b. , Send. When the syndrome processing unit 22 of the FEC node 12 b detects discard of one bucket, the discard number updating unit 35 of the FEC node 12 b discards the received packet in the information area “2” of the received packet by its own node. Is added, and the bucket with the information area set to "3" is sent to the FEC node 12c. If the syndrome processing unit 22 of the FEC node 12c does not detect the bucket discard, the discard number update unit 35 of the FEC node 12c does not change the information area (adds "0" to the information area to change the information area). ) The packet whose information area is “3” is transmitted to the receiving terminal 13. The FEC decoding unit 132 having no receiving end 13 extracts the number of discarded packets in the information area and rewrites the ACK shown in FIG. 26 or the window size changing unit 134 shown in FIG. 24 or FIG. Notify part 135.

As described above, in the fourteenth embodiment, the information control area for recording the number of discarded buckets or the discarded rate is provided in the packet, and the FEC nodes arranged for each section are detected in each bucket. Is added to the information area, so that the receiving terminal can grasp the number of buckets discarded or the discarded rate of the entire network, and adjust the TCP ACK window size according to the network be able to. In this Embodiment 14, an example has been described in which the FEC processing unit 21 of Embodiment 1 shown in FIG. 2 is provided with the discard number updating unit 35, but this is shown in FIG. Even if the FEC processing unit 21a, the FEC processing unit 21b shown in Fig. 15 or the FEC processing unit 21c shown in Fig. 17 has the discard number update unit 35, the FEC processing unit Needless to say, the same effect as when the discard number updating unit 35 is provided in 21 can be obtained. Industrial applicability

As described above, the packet transfer device according to the present invention is useful for a large-scale network, and is particularly suitable for a bucket transfer device in a network such as a WAN having a long delay time.

Claims

1. Generate a redundant packet for forward error correction based on the original bucket and the redundant bucket received from the transfer source via the network, and transfer the generated redundant bucket and the original bucket via the network. In the packet transfer device that transmits first,

When performing the decoding for the forward error correction, the blue from the transfer destination is used.

Determine the redundancy degree based on the feedback information and receive from the transfer source

Three

A forward error correction processing unit that generates feedback information to be notified to the transfer source based on the reception state of the original bucket and the redundant bucket 5

A packet transfer device comprising: Enclosure

2. The forward error correction processing unit comprises:

Redundancy is determined based on the feed pack information from the transfer destination, and redundancy forming the build pack information to be notified to the transfer source based on the number of discarded original packets and redundant buckets received from the transfer source. An adaptive processing unit, a flow identification unit for identifying a flow of the original bucket and the redundant bucket received from the transfer source,

A syndrome processing unit that performs a syndrome calculation based on the original packet identified by the flow identification unit;

A restoration processing unit for restoring the original packet discarded by the decoding unit; a syndrome calculation value calculated by the syndrome processing unit; an original bucket restored by the restoration processing unit; and the redundancy adaptation processing unit. A redundant packet generation processing unit that generates the redundant packet by re-encoding based on the redundancy determined in the above manner;

An original bucket identified by the flow identification unit, an original packet restored by the restoration processing unit, and a redundant packet generation processing unit A selector for selecting and outputting the generated redundant packet,

2. The bucket transfer device according to claim 1, comprising:

3. The forward error correction processing section includes:

(5) A method of restoring a discarded original nano packet for adjusting the redundancy based on the feed pack information from the transfer destination, determining whether to discard the received redundant packet, and A redundancy adaptation processing unit for generating feedpack information to the transfer source based on the number of discarded original packets and redundant buckets received from the transfer source;

0 a flow identification unit for identifying the flow of the original packet and the redundant packet received from the transfer source,

(I) a syndrome processing unit that performs a syndrome calculation based on the original packet identified by the contact identification unit;

When the redundancy adaptation processing unit decides to restore the original bucket, 15 is a restoration processing unit for restoring the original packet discarded by the decoding, and an original bucket identified by the flow identification unit. A selector for selecting and outputting the original bucket restored by the restoration processing unit, a redundant bucket, and the original bucket;

If the redundancy adaptation processing unit decides to discard the redundant packet, the redundancy adjustment unit discards the redundant bucket output from the selector 0; and

2. The packet transfer device according to claim 1, comprising:

4. The redundant bucket generation processing unit includes:

A timer for measuring a bucket receiving interval from a transfer source,

25 an encoding unit that generates a redundant bucket from an original bucket from a source by a code based on the redundancy determined by the redundancy adaptive processing unit; The reception interval power of the original bucket from the transfer source measured by the timer is greater than a predetermined threshold value, and if it exceeds a predetermined threshold, a redundant bucket is generated from the result of the encoder at that time. 3. The bucket transfer device according to claim 2, wherein

5. The syndrome processing unit is

A timer that measures the interval between receipts of packets from the transfer source,

With

When the reception interval of the original bucket from the transfer source measured by the timer exceeds a predetermined threshold value, the restoration processing unit determines whether the original bucket has received the original bucket based on the syndrome calculation value at that time. 3. The bucket transfer device according to claim 2, wherein restoration is performed.

6. The forward error correction processing unit comprises:

The received packets are counted for each flow of the packet received from the transfer source, and when the count value becomes a number that enables the restoration processing unit to restore the discarded original packet, the restoration processing unit starts processing. Counter to notify,

Further comprising

3. The bucket transfer device according to claim 2, wherein the restoration processing unit starts restoring the discarded original packet upon receiving the processing start notification.

7. The forward error correction processing unit comprises:

A syndrome memory for holding an intermediate result of a syndrome calculation value calculated by the syndrome processing unit;

Further comprising

The flow identification unit, When the usage status of the syndrome memory is monitored and it is detected that there is no available area in the syndrome memory, it is determined that the original packet and the redundant bucket from the transfer source are output as they are,

The selector is

3. The bucket transfer device according to claim 2, wherein the packet transfer device outputs the original packet and the redundant packet from the transfer source.

8. The flow identification unit comprises:

The number of original buckets and redundant packets discarded based on the original bucket and the redundant bucket from the transfer source is counted, and the counted value is used to determine the number of discardable packets determined from the redundancy of the own device. If it exceeds, the syndrome calculation of the syndrome processing unit is stopped, and the

'I decided to output only the original bucket 1,

The selector is

3. The bucket transfer device according to claim 2, wherein only the original packet from the transfer source is output.

9. The selector is

A high-priority class is assigned to an original bucket from the transmission source and an original packet restored by the restoration processing unit, and a low-priority class is assigned to a redundant packet generated by the redundant packet generation processing unit. 3. The packet transfer device according to claim 2, wherein the packet transfer device is characterized in that:

10. The forward error correction processing unit comprises:

A monitor packet generator for generating a monitor bucket for monitoring the discarded state of the bucket in the network, and transmitting the generated monitor bucket to the transfer destination; And a monitor bucket receiving means for receiving a monitor bucket transmitted from the transmission source and obtaining a discarded state of a bucket of the network between the transmission source and the own device. The bucket transfer device described in paragraph 2.

1 1. The forward error correction processing unit comprises:

A redundancy adjustment unit configured to discard a redundant bucket generated by the redundant bucket generation processing unit for each transfer destination in the multicast and adjust the redundancy bucket to the redundancy for each transfer destination;

The redundancy adaptation processing unit includes:

The feedback information transmitted for each of the transfer destinations has the highest redundancy, and it is determined that a redundant bucket is generated in accordance with the feedback information.

3. The packet transfer device according to claim 2, wherein the redundant bucket generation processing unit generates a redundant packet based on the redundancy determined by the redundancy adaptation processing unit.

12. The original bucket has an information area for recording the number of discard rates of the bucket,

The forward error correction processing unit,

A discard number update unit that detects the number of discarded packets based on the original bucket or the redundant bucket received from the transfer source and adds the detected number of discarded buckets to the information area;

3. The bucket transfer device according to claim 2, further comprising:

13. A packet transfer network system comprising a plurality of the packet transfer devices according to claim 1, wherein a forward error correction process is performed between the packet transfer devices based on feedback information.

1 4. In a terminal device that is connected via a network and communicates using TCP in the transport layer,

A forward error correction encoding unit that performs encoding for forward error correction on a bucket to be transmitted, generates and transmits an original bucket and a redundant bucket,

A forward error correction decoding unit that decodes the received original packet and the redundant packet to recover the discarded original bucket and detects the number of discarded buckets;

A terminal device, comprising: a window size changing unit that changes the ACK window size based on the number of discarded packets detected by the forward error correction decoding unit and transmits the result.

15. A feedpack information transmission processing unit that generates feedback information based on the number of discarded packets detected by the forward error correction decoding unit and transmits the generated feedpack information,

A redundancy adaptation processing unit that determines the redundancy of the forward error correction coding unit based on the received feedpack information;

15. The terminal device according to claim 14, further comprising:

16. Based on the number of discarded buckets detected by the forward error correction coding unit, generate a special packet including the generated feedpack information in the ACK packet and transmit the generated special packet. ACK rewriting unit

The feedback information is extracted from the received special bucket, and the redundancy of the forward error correction coding unit is determined based on the extracted feedback information. Information extraction unit and '

15. The terminal device according to claim 14, further comprising: