KR20070098169A - Method for transmission data for multicast based network, method for processing data and data transmission system - Google Patents

Abstract

A method for transmitting data for a multicast-based network, a method for processing the data, and a system for transmitting the data are provided to prevent unnecessary time delay by not requiring a retransmission process due to loss of data. A method for transmitting data for a multicast-based network includes the steps of: generating a data packet encoded with an FEC(Forwarding Equivalence Classes) method by a transmitter mode(S1); transmitting the encoded data packet(S2); grasping the loss of the data packet generated on each received path of relay nodes installed on a transmission path of the transmitted data packet(S4); restoring the loss when restoration is needed(S6); transmitting the restored data packet to a child node of the relay node(S8); and receiving the data packet transmitted by the relay node corresponding to a parent node of one of the relay nodes.

Description

Data transmission method, data processing method and data transmission system in a multicast-based network {Method for Transmission Data for Multicast Based Network, Method for Processing Data and Data Transmission System}

FIG. 1 is a block diagram showing a configuration of a multicast-based network to which a data transmission method according to the first embodiment of the present invention can be applied.

2 is a flowchart illustrating an operation flow of a data transmission method in a multicast-based network according to a first preferred embodiment of the present invention.

FIG. 3 is an exemplary view showing a path in which a transport packet transmitted from a transmitter node A in the overlay tree shown in FIG. 1 is transmitted to a receiver node C6 via a relay node B3 and a relay node B7.

4 is a configuration diagram showing a configuration of a multicast network according to a second embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

A: sender node

B1, B2, B3, B4, B5, B6, and B7: relay node

Recipient node: C1, C2, C3, C4, C5, C6, C7 and C8

The present invention relates to a data transmission method, a data processing method, and a data transmission system in a multicast-based network, and more particularly, each relay node existing between a sender node and a receiver node occurs in its own reception path. The present invention relates to a data transmission method in a multicast-based network for determining the loss of data packets and actively recovering the data packet, and a data processing method and a data transmission system related thereto.

In general, a multicast transmission method is a method of simultaneously transmitting the same data packet to two or more different receivers on a network, and is a concept that is opposed to a unicast transmission method of transmitting a data packet to only one specific receiver. It can be seen.

This multicast transmission method is a transmission method suitable for a network service requiring group communication. In the unicast transmission scheme, when data is transmitted to a plurality of receivers for the group communication, the same data packet must be repeatedly transmitted to each receiver, and the redundant transmission of the same data packet is a network efficiency. This is because of the degradation of. This problem becomes more prominent when the number of recipients increases.

On the other hand, as mentioned above, since the multicast transmission method transmits data packets to a plurality of receivers at the same time, it is possible to minimize waste of network resources due to redundant transmission of data.

Therefore, the multicast transmission method is a digital multimedia service that requires group communication, for example, IPTV (Internet Protocol Television), News Feed (News Feed), Internet video conferencing, online games, video on demand (VoD) It is mainly used in Video on Demand (DVB) and Digital Video Broadcasting (DVB). In particular, in an interactive service such as multi-party video conferencing, real-time transmission of voice and video data is required. Therefore, a multicast transmission method capable of simultaneously transmitting and receiving data is required.

In the Internet application of the multicast transmission method, the difference from the general unicast transmission method is in the transmission packet. Internet applications on the Transport Control Protocol / Internet Protocol (TCP / IP) send packets by indicating in the header of the transport packet the sender of the data indicates the recipient's Internet address. However, for multicast transmission, the packet is transmitted by indicating the group address of the receiver in the header of the packet instead of the recipient's address. For example, the group address for multicast transmission is a D-class IP address (224.0.0.0??239.255.255.255), which is a real host unlike the A, B, and C-class IP addresses that represent unique Internet hosts. The receiver receiving the multicast packet having the group address decides whether the packet belongs to the group of the packet or not.

On the other hand, RMTP (Reliable Multicast Transport Protocol) has been proposed as a protocol for realizing reliable multicast transmission. The RMTP attempts to provide reliability of data transmission using a response character (ACK: ACKnowledged, hereinafter abbreviated as ACK). That is, the sender transmits data to each receiver and then receives an ACK message from each receiver to confirm whether the transmission data is lost. If a data loss occurs, the sender retransmits the data.

However, in this case, there is a high probability of ACK flooding due to frequent transmission of the ACK message, and there is a problem in that real time deteriorates due to retransmission of data.

Therefore, by improving the RMTP, an Automatic Repeat Request (ARQ: Automatic Repeat Request) based on an Error Correction Code (FEC) or Negative ACKnowledged (NACK) RMTP II is proposed as an optional option. The RMTP II is described in detail in IETF internet Draft (Brian Whetten, et. Al., 'The RMTP II Protocol, IETF Draft-Whetten-rmtp-ii-00.txt. April, 8, 1999.).

The NACK-based ARQ technique is a technique for automatically requesting retransmission of data by notifying the sender that a transmission error has occurred through a NACK message when there is a loss in the received data. However, since the NACK-based ARQ scheme requires a retransmission operation due to data loss, a time delay due to retransmission occurs. Therefore, there is a problem that the data transmission rate is lowered, which is not suitable for real time transmission, which is one of the important characteristics of multicast transmission.

In addition, the FEC technique is a technique for transmitting the data is encoded in the FEC block by overlapping the data so that the receiver can recover the loss in the case of data loss. That is, when the first sender transmits a packet including a data packet and a duplicate packet, the receiver who finally receives the data determines whether the packet is lost and recovers the loss. Therefore, the FEC technique is a forward error recovery technique that can actively cope with the loss on the receiver side.

However, the FEC scheme is suitable for real time transmission, but requires a network bandwidth more than necessary due to transmission of duplicate data.

Accordingly, in order to overcome the problem of excessive bandwidth occupancy of the FEC scheme, a dynamic FEC scheme and the like have been proposed, and efforts have been made to reduce the amount of redundant data. However, even if the dynamic FEC technique is applied to a multicast network, the amount of bandwidth change is not so large in terms of the entire network, and there is a problem that recovery is not possible at all for loss of packets exceeding the amount of overlapping data.

Also, given that the location of packet loss and error in the network is usually concentrated in a particular router, applying the same error recovery rate across the entire path from the original sender to the final receiver, given limited network resources It may be unnecessary waste.

SUMMARY OF THE INVENTION The present invention has been made to solve this problem. In the multicast-based network, each communication node performs loss check of the FEC block and recovers the loss when the loss occurs, thereby improving the real-time and service quality of the data transmission. It is a first object of the present invention to provide a data transmission method in a multicast-based network that can be satisfied at the same time.

Another object of the present invention is to provide a data processing method that can be applied to a relay node which relays data transmission between a sender node and a receiver node.

In addition, a third object of the present invention is to provide a multicast data transmission system capable of performing loss determination and recovery of a data packet using a relay node provided in a data transmission path between a sender node and a receiver node.

A data transmission method in a multicast network according to the present invention for achieving the first object of the present invention comprises the steps of: generating and transmitting an FEC encoded data packet by a sender node; The relay nodes hierarchically provided in the transmission path of the transmitted data packet grasp the degree of loss of the data packet generated in their own reception path, and if the recovery is necessary, recover the loss and then return to the child node. Transmitting; And a receiver node receiving a data packet transmitted by a relay node corresponding to its parent node among the relay nodes. In this case, the data packet transmitted by the sender node may consist of h duplicate packets and k FEC data packets generated by the FEC encoding.

The relay node determines whether the loss of the data packet received by the relay node is greater than or equal to the acceptance criterion, and if the determination result is greater than or equal to the acceptance criterion, the relay node performs FEC decoding on the received data packet to recover the loss, and then performs FEC encoding to recover the loss. If the determination result is less than the acceptance criteria, the received data packet is transmitted to the child node.

In addition, the relay node may feed back the loss information of the data packet received to the parent node to which the data packet is transmitted. In this case, the parent node performs FEC encoding so that the combination ratio of the FEC data packet and the duplicate packet of the data packet is changed in consideration of the feedback data packet loss information, and then transmits it to the relay node. Here, the combination ratio of the FEC data packet and the duplicate packet may be determined such that the error recovery rate is greater than the packet loss rate.

On the other hand, the data processing method according to the present invention for achieving the above-described second object of the present invention, the data processing method of the relay node provided at least one or more hierarchically to relay the data packet between the sender node and the receiver node. A method comprising: receiving an FEC encoded data packet from one of the sender node and a higher layer relay node; Determining whether a loss of the received data packet is above a tolerance criterion; And recovering the loss and transmitting the loss to any one of the receiver node and the lower layer relay node when the loss is higher than the acceptance criteria, and if the loss of the data packet does not occur, the received data packet. Transmitting to either a receiver node or the lower layer relay node.

The transmitting may include recovering the lost part by FEC decoding the received data packet when the determination result is equal to or greater than an acceptance criterion; And FEC encoding the recovered data into an FEC data packet having a predetermined combination ratio and a duplicate packet. In this case, the combination ratio of the FEC data packet and the duplicate packet may be determined to have an error recovery rate greater than the packet loss rate.

On the other hand, the multicast data transmission system according to the present invention for achieving the above-described third object of the present invention, the sender node for transmitting the FEC encoded data packet; At least one relay node for relaying the transmitted data packet; And a receiver node receiving a data packet transmitted from the relay node forms a hierarchical structure, and the relay node determines a loss degree of the FEC encoded data packet received from its parent node and recovers it when necessary. It recovers the loss and transmits it to its own child node, and if the recovery is not necessary, transmits the received FEC encoded data packet to the child node.

The relay node compares the loss degree of the received FEC encoded data packet with an allowance criterion, and if the comparison result is equal to or greater than the allowance criterion, FEC decodes the FEC encoded data packet to recover the loss, and then performs FEC encoding. .

The relay node may be an information terminal, in which case the FEC encoding and FEC decoding may be performed at an application layer of the information terminal. In addition, the relay node may be a router. In this case, the FEC encoding and the FEC decoding may be performed by an FEC layer provided in the router.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. In addition, in describing the preferred embodiment of the present invention illustrated in the drawings, specific technical terms are used for clarity of content. However, it is to be understood that the invention is not limited to these specific selected terms, and that each specific term includes all technical synonyms that operate in a similar manner to achieve a similar purpose.

<Example 1>

1 is a block diagram showing a configuration of a multicast-based network to which a data transmission method according to a first embodiment of the present invention can be applied, and shows a configuration of an overlay multicast tree formed by a session for transmitting data. Giving.

First, prior to referring to FIG. 1, an overlay multicast to which the data transmission method according to the first exemplary embodiment of the present invention can be easily applied will be described.

Overlay multicast is to enable multicast transmission using the current unicast-based Internet network, and refers to a multicast transmission scheme at a higher layer such as an application layer. For example, in overlay multicast, a plurality of communication nodes existing in the current unicast-based Internet network are formed into a logical tree structure divided into senders, relays, and receivers, and then the sender sends packets to the receivers through relays of the relays. To transmit the data.

Using this overlay multicast transmission technology, a multicast application service can be provided in software using a unicast network even without a separate multicast network.

Referring to FIG. 1, an overlay tree for connecting a sender node A and a plurality of receiver nodes C1 to C8 is configured on a network for data transmission through multicast. At this time, between the sender node A and the receiver nodes C1 to C8, relay nodes B1 to B7 are hierarchically formed for data relay.

In this case, the nodes A, B1 to B7, and C1 to C8 may refer to information terminals operating in an upper application layer. In other words, the overlay tree refers to a tree that logically connects information terminals. This overlay tree corresponds to the routing tree in the multicast network. In the multicast network, the relay node is a router, whereas in the overlay multicast network, each node is an information terminal.

The relay nodes B1 to B7 are nodes that may have their own child nodes, for example, other relay nodes or receiver nodes C1 to C8, and may be servers separately installed in a network, or may be software in communication terminals. It may be implemented in the form. The relay nodes B1 to B7 receive data from the sender node A or higher relay nodes and relay data to child nodes.

2 is a flowchart illustrating an operation flow of a data transmission method in a multicast-based network according to a first preferred embodiment of the present invention.

In addition, FIG. 3 illustrates a path in which the transport packet transmitted from the sender node A in the overlay tree shown in FIG. 1 is transmitted to the receiver node C6 via the relay node B3 and the relay node B7.

2 to 3, first, the sender node A performs FEC encoding on original data with a predetermined error recovery rate through an application layer to generate n FEC blocks composed of k FEC data packets and h duplicate packets. Subsequently (step: S1), the data packet is converted into a data packet form that can be transmitted through a lower layer and transmitted to the relay node B3 (step: S2).

In this case, the data packet may be expressed as (n, k). Thus, the FEC encoding may be referred to as (n, k) encoding. Where k is the number of FEC data packets and n is the total number of transport packets. That is, n is the number of FEC data packets plus the number of duplicate packets. N is therefore k + h.

For example, if there are three FEC data packets and one duplicate packet, the transmission data packet may be expressed as (4, 3).

The FEC data packet is a FEC-encoded block of original data divided into equal sizes. In addition, the duplicate packet refers to a parity packet generated so that it can be recovered even if some of the k FEC data packets are lost. Information contained in the k FEC data packets is implicitly inserted into the FEC packet by a predetermined operation.

b

c" 같은 연산값을 가질 수 있다. 따라서, FEC 데이터 패킷의 일부가 손실되더라도 중복 패킷을 통하여 그 손실 부분을 복구할 수 있게 된다.For example, if the data packet to be transmitted is (4, 3), it can be seen that three FEC data packets are included. In this case, if the three FEC data packets are "a", "b", and "c", the duplicate packets are included. &Quot; a " includes all information of the three FEC data packets.

b

c ". Thus, even if a part of the FEC data packet is lost, the lost part can be recovered through the duplicate packet.

Meanwhile, the error recovery rate E may be expressed as in Equation 1 below.

E = (h / n) × 100

That is, if the transmission data packet is (4, 3), the error recovery rate is 25% since h is 1 and n is 4. In this case, packet loss within 25% is recoverable. Therefore, if you want to increase the error recovery rate, you can increase the number of heavy packets.

On the other hand, the relay node B3 receives the data packet transmitted by the sender node A, which is its parent node (step: S3), and then determines whether the loss level of the data packet is greater than or equal to the acceptance criteria (step: S4), As a result of the determination, if the loss is less than the acceptance criteria, the received data packet is transmitted to the relay node B7 which is its child node (step S8). On the other hand, if the determination result loss is above the acceptance criteria, the relay node B3 FEC decoding the received data packet (step: S5), recovers the lost packet (step: S6), and then FEC encodes again (step: S7), and transmits to the relay node B7 which is its child node (step S8).

For example, if the data packet transmitted from the sender node A is (4, 3) as shown in FIG. 3, but the data packet received at the relay node B3 is (4, 2), one FEC data packet is lost. As a result, the lost data packet can be recovered by using the duplicate packet. Therefore, the data packet transmitted by the relay node B3 may be represented by (4, 3) again.

On the other hand, when data transmitted by the relay node B3, which is the parent node, is transmitted to the relay node B7, the same process as the above-described relay node B3 is performed. That is, a data packet is received (step: S3), and it is determined whether the loss degree of the received data packet is greater than or equal to the acceptance criterion (step: S4). It sends to the receiver node C6 which is its child node (step S8). On the other hand, if the determination result loss is above the acceptance criteria, the received data packet is FEC decoded (step: S5), the lost packet is recovered (step: S6), and then FEC encoded (step: S7), the child The data is transmitted to the receiver node C6, which is a node (step S8).

For example, as shown in FIG. 3, if the data packet transmitted from the relay node B3 is (4, 3), but the data packet received at the relay node B7 is (3, 3), the duplicate packet is lost. have. Therefore, lost duplicate packets can be recovered using the FEC data packet information. The data packet sent to receiver node C6 will thus be (4, 3) as when sent from the original sender node A.

Subsequently, receiver node C6 receives a data packet transmitted from relay node B7. At this time, the receiver node C6 may also obtain the data by identifying the loss information of the data packet and recovering the loss in the FEC decoding as in the relay node B3 or B7 described above. Referring to FIG. 3, since no packet loss occurs in the path transmitted from the relay node B7 to the receiver node C6, the receiver node C6 may not need to perform a separate loss recovery.

On the other hand, the error recovery rate of the transmitted data packet may be varied according to the degree of loss of the data packet grasped by each relay node, thereby constituting the data packet variably. In fact, in a multicast-based network, the loss of data packets is mostly concentrated in a specific path, so if the loss of data packets is much higher at one relay node, the error recovery rate is higher at the previous relay node. You can also construct a data packet.

To this end, the loss information of the data packet identified in each relay node may be fed back to the relay node of the parent node. For example, in the example shown in FIG. 3, if loss of data packets is concentrated in the path transmitted from the relay node B3 to the relay node B7, the relay node B7 may feed back the lost information of the data packet to the relay node B3. . This may be selectively performed only at the relay node larger than the relay node having a different degree of loss, or may be performed at all relay nodes for stability.

Based on this feedback information, each relay node can adjust the number of data packets and duplicate packets during FEC encoding. That is, the number of duplicate packets is increased while leaving the number of FEC data packets intact. For example, the relay node B3 can increase the error recovery rate based on the feedback information from the relay node B7. That is, if the received data packet is (4, 3), it can be configured as (5, 3) by adding one more duplicate packets to increase the error recovery rate. In this case, the error recovery rate can be improved from 25% to 40%.

The packet loss rate is taken into account when combining the number of packets. In this case, the packet loss rate indicates the ratio of packets lost in the received data packet, and the error recovery rate should be higher than the packet loss rate. This is because, if the packet loss rate is higher than the error recovery rate, recovery of the packet loss may be impossible.

As described above, according to the present invention, since the loss degree of the data packet is actively recovered for each relay node existing between the sender node and the receiver node, the stability of the data packet transmitted to the receiver node can be ensured. In addition, since there is no need for a separate retransmission request like the NACK-based ARQ function, unnecessary time delays can be prevented, thereby improving real-time.

On the other hand, in the conventional FEC scheme, since the relay node performs only simple data packet transmission, and only the receiver node performs the loss recovery of the data packet, the loss of packets generated in each transmission path is accumulated, and thus the loss cannot be recovered. Possible cases occurred. In addition, in order to prevent this, a method of increasing the number of duplicate packets may be used. However, in this case, the capacity of the data packet to be transmitted increases, which may burden the network.

At this time, in the above-described embodiment of the present invention, if the FEC decoding / encoding for loss recovery is performed for each relay node, it may be considered that the load of the transmitting node may be increased. In most cases, loss recovery is required only on the node, as it is mostly focused on the receive path. Therefore, other relay nodes only need to determine the loss of packets, so they do not have a significant effect from the viewpoint of the entire network.

In the above, the first embodiment of the present invention has described a data transmission method in a multicast-based network having excellent reliability and real-time data packet transmission. However, as mentioned above, the multicast-based network in the first embodiment refers to overlay multicast performed at the application layer. However, the technical idea of the present invention can be easily performed even in a multicast environment constructed in hardware. Therefore, the following second embodiment will be described a data transmission method based on a multicast network.

<Example 2>

4 is a configuration diagram showing a configuration of a multicast network according to a second embodiment of the present invention.

Referring to FIG. 4, a logical tree connecting a sender node D and a plurality of receiver nodes F1 to F3 is configured on a network for data transmission through multicast. At this time, routers E1 to E4 and M1 to M3 are provided between the transmitter node D and the receiver nodes F1 to F3 as relay nodes for data relay.

In this case, the routers E1 to E4 and M1 to M3 are nodes capable of having their own child nodes, for example, other routers or receiver nodes F1 to F3. Among the routers E1 to E4 and M1 to M3, the router E1 to E4 and the routers E1 to E4 that do not have multicast functions may be classified into MRouters M1 to M3. The data packet is transmitted through tunneling between the emrouters M1 to M3.

Each router (E1 ~ E4, M1 ~ M3) grasps the loss of the data packet received to it, and if the loss occurs above the allowable standard, it performs FEC decoding to recover the lost packet and then performs FEC encoding again. After executing, transmit to the router of the child node.

In this case, in the first embodiment described above, the FEC processing related functions are performed through the application layer of the information terminal which is the relay node. However, in the second embodiment, the relay nodes are the routers E1 to E4 and M1 to M3. For this reason, each router E1 to E4 and M1 to M3 should be equipped with an FEC apparatus for performing FEC related functions. This may mean that the FEC layer is provided separately.

The FEC layer divides the data transmitted from the upper layer into k blocks having the same size, FEC-encodes the k data blocks, generates k FEC data packets and n duplicate packets, and transmits the same to the lower layer. Performs the function and vice versa. Therefore, each of the routers E1 to E4 and M1 to M3 may perform loss detection and loss recovery of data packets through the FEC layer.

As described above, according to the present invention, since each relay node existing between the sender node and the receiver node can actively recover by determining the degree of loss of the data packet occurring in its reception path, the data packet transmission stability Can be secured. In addition, since a retransmission process is not required due to the loss of data packets, unnecessary time delays can be prevented, which is an advantage in terms of real time.

Claims (13)

A sender node generating and transmitting an FEC encoded data packet; The relay nodes hierarchically provided in the transmission path of the transmitted data packet grasp the degree of loss of the data packet generated in their own reception path, and if the recovery is necessary, recover the loss and then return to the child node. Transmitting; And And a receiver node receiving a data packet transmitted by a relay node corresponding to its parent node among the relay nodes. 2. The method of claim 1, wherein the data packet transmitted by the sender node includes h duplicate packets and k FEC data packets generated by the FEC encoding. The method of claim 1, wherein the relay node determines whether the loss of the data packet received by the relay node is greater than or equal to the acceptance criterion, and if the determination result is greater than or equal to the acceptance criterion, the relay node recovers the loss by FEC decoding the received data packet. And transmitting the received data packet to the own child node if FEC encoding is performed and transmitted to the own child node. The method of claim 1, wherein the relay node feeds back lost information of the data packet received to the parent node to the parent node that transmitted the data packet. 5. The method of claim 4, wherein the parent node performs FEC encoding so that a combination ratio of the FEC data packet and the duplicate packet of the data packet is changed in consideration of the feedback data packet loss information, and then transmits the FEC encoding to the relay node. A data transmission method in a multicast-based network. 6. The method of claim 5, wherein the combination ratio of the FEC data packet and the duplicate packet determines that an error recovery rate is greater than a packet loss rate. A data processing method of a relay node provided with at least one layer hierarchically for relaying data packets between a sender node and a receiver node, Receiving an FEC encoded data packet from either the sender node or a higher layer relay node; Determining whether a loss of the received data packet is above an acceptance criterion; And If the result of the determination results in a loss exceeding an acceptable criterion, the loss is recovered and transmitted to either the receiver node or a lower layer relay node. If the result of the determination does not cause a loss of the data packet, the receiver receives the received data packet. Transmitting to any one of a node and the lower layer relay node. The method according to claim 7, wherein in the transmission step, the determination result is greater than or equal to the acceptance criteria. Recovering the lost portion by FEC decoding the received data packet; And FEC encoding said recovered data into a FEC data packet having a predetermined combination ratio and a duplicate packet. 9. The data processing method according to claim 8, wherein the combination ratio of the FEC data packet and the duplicate packet is determined to have an error recovery rate greater than a packet loss rate. A sender node for transmitting FEC encoded data packets; At least one relay node for relaying the transmitted data packet; And The receiver node receiving the data packet transmitted from the relay node has a hierarchical structure, The relay node detects the loss level of the FEC encoded data packet received from its parent node, recovers the loss if necessary, and transmits the loss to its child node. And transmitting the FEC encoded data packet to the child node. The method of claim 10, wherein the relay node compares the loss of the received FEC-encoded data packet with an acceptance criterion, and if the comparison result is greater than or equal to the acceptance criterion, recovers the loss by FEC decoding the FEC-encoded data packet. And FEC encoding again. 12. The system of claim 11, wherein the relay node is an information terminal, and the FEC encoding and FEC decoding are performed at an application layer of the information terminal. 12. The system of claim 11, wherein the relay node is a router, and the FEC encoding and FEC decoding are performed by an FEC layer provided in the router.

Images