KR101109658B1 - Multicast routing method using network coding scheme in wireless mobile multihop network system - Google Patents

Abstract

The present invention relates to a multicast routing method in a wireless mobile multihop network system. Multicast routing method in a wireless mobile multi-hop network system according to an embodiment of the present invention comprises the steps of establishing a data transmission path between the source node and the destination node through the transmission and reception of control packets; Whenever a data packet generated at the source node is stored in a buffer of the source node, at least one data packet having the same generation number among the data packets stored in the buffer of the source node is encoded by a network coding scheme to encode an encoded packet. Generating a; And transmitting the encoded packet from the source node to the destination node through the data transmission path. According to the multicast routing method according to an embodiment of the present invention, the reception rate (reliability) of a data packet can be increased while reducing the overhead due to transmission and reception of control packets in a wireless mobile multihop network system.

Description

Multicast routing method using network coding technique in wireless mobile multi-hop network system {MULTICAST ROUTING METHOD USING NETWORK CODING SCHEME IN WIRELESS MOBILE MULTIHOP NETWORK SYSTEM}

The present invention relates to a multicast routing method, and more particularly to a multicast routing method in a wireless mobile multi-hop network.

In a wireless mobile multi-hop or mobile ad-hoc network (hereinafter referred to as MANET), nodes can communicate with each other without a wired infrastructure or a central control unit. Therefore, MANET is useful in environments where wired infrastructure is not built or difficult to access.

According to the On-Demand Multicast Routing Protocol (ODMRP) used in MANET, the multicast data transmission path allows the transmission and reception of control packets between nodes in a multicast group. Is set through. However, the path set in this way may be broken due to the mobility of nodes. Therefore, transmission and reception of control packets should be performed periodically to maintain the path.

However, in a network based on ODMRP, the transmission and reception of these periodic control packets incurs a lot of overhead throughout the network. Here, the time interval in which control packets occur is one of the factors affecting overhead and reliability. If the time interval over which control packets occur is too short, limited resources at a given channel bandwidth are wasted due to increased overhead. On the other hand, if the time interval over which control packets occur is too long, reliability is lost due to data packet loss caused by path disconnection. This means that the data packet reception rate is reduced.

It is an object of the present invention to provide a multicast routing method for recovering lost data packets during path recovery or reconfiguration in a wireless mobile multi-hop network system.

Multicast routing method in a wireless mobile multi-hop network system according to an embodiment of the present invention comprises the steps of establishing a data transmission path between the source node and the destination node through the transmission and reception of control packets; Whenever a data packet generated at the source node is stored in a buffer of the source node, at least one data packet having the same generation number among the data packets stored in the buffer of the source node is encoded by a network coding scheme to encode an encoded packet. Generating a; And transmitting the encoded packet from the source node to the destination node through the data transmission path.

In one embodiment, the multicast routing method includes at least one of the encoded packets stored in a buffer of the target node each time the encoded packet received at the target node has the same generation number. Determining whether to decode the encoded packet; And if it is determined that the at least one encoded packet can be decoded, the method may further include decoding the at least one encoded packet.

The setting of the data transmission path may include: periodically generating a subscription query packet among the control packets; And flooding the subscription query packet to all nodes belonging to a multicast group.

In an embodiment, the multicast routing method may be configured to generate an encoded packet having a generation size of a largest value among encoded packets generated during a time between a generation time of a previous subscription query packet and a generation time of a next subscription query packet. The method may further include adding to the packet.

In an embodiment, the generation size of the encoded packet is determined according to the number of data packets that are encoded to generate the encoded packet.

In an embodiment, in the step of periodically generating the subscription query packet, the generation period of the subscription query packet varies according to the reception rate of the data packet.

In an embodiment, in the step of periodically generating the subscription query packet, the generation period of the subscription query packet decreases as the reception rate of the data packet increases, and increases as the reception rate of the data packet decreases.

In an embodiment, the generation number of data packets and encoded packets generated during the time between the generation of the previous subscription query packet and the generation of the next subscription query packet is updated in response to the generation of the previous subscription query packet.

In the determining of whether the decoding is possible, the number of at least one encoded packet (hereinafter, at least one decoded encoded packet) having the same generation number among the encoded packets stored in the buffer of the target node may be determined. If at least one decoding object encoded packet has a value equal to or greater than a largest value among generations, it is determined that the at least one decoding object encoded packet can be decoded.

In an embodiment, the multicast routing method may further include recovering lost data from data packets obtained by decoding the at least one encoded packet.

In an embodiment, the establishing of the data transmission path may include performing a local path recovery.

The transmitting of the encoded packet from the source node to the destination node may include determining whether the encoded packet is normally transmitted through a passive acknowledgment response.

In an embodiment, in the generating the encoded packet, a random linear coding scheme is applied as the network coding scheme.

According to the multicast routing method according to an embodiment of the present invention, the reception rate (reliability) of a data packet can be improved while reducing overhead due to transmission and reception of control packets in a wireless mobile multi-hop network system.

1 is a diagram for exemplarily describing a multicast routing method in an ODMRP-based network system.
2A through 2C are diagrams for describing a multicast routing method using local path recovery according to an embodiment of the present invention.
3 is a flowchart illustrating a multicast routing method of adjusting a rerouting period according to a reception rate of a data packet.
4 is a diagram illustrating a multicast routing method using a passive ACK.
5 is a diagram illustrating data packets encoded by a network coding technique.
6 is a flowchart illustrating an encoding method by a network coding scheme according to an embodiment of the present invention.
7 is a flowchart illustrating a method of recovering a lost data packet through decoding by a network coding scheme according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, so that those skilled in the art can easily carry out the technical idea of the present invention.

The multicast routing method according to an embodiment of the present invention is applied to a mobile ad-hoc network (hereinafter referred to as MANET). The network system according to an embodiment of the present invention is based on an on-demand multicast routing protocol (hereinafter, referred to as ODMRP).

1 is a diagram for exemplarily describing a multicast routing method in an ODMRP-based network system. Referring to FIG. 1, a multicast group of a MANET according to an embodiment of the present invention is illustrated by way of example. The multicast group includes a source node S, a destination node D, and first to sixth forwarding nodes F1 to F6.

For simplicity, it is assumed that a multicast group according to an embodiment of the present invention includes one source node S and one destination node D. However, this is exemplary and it is apparent to those skilled in the art that a multicast group may include a plurality of source nodes and a plurality of destination nodes.

In order to establish a path for data packet transmission between the source node S and the destination node D, a process of configuring a mesh by selecting forwarding nodes through periodic transmission and reception of control packets is required. Here, the set of forwarding nodes elected to mediate the data packet between the source node S and the destination node D is called a forwarding group.

The data packet transmission path between the source node S and the destination node D is established through an inquiry step and a response step. First, the querying step is described. The source node S floods a control packet (hereinafter referred to as a subscription query packet) including a join query message and identification information. In this case, flooding means transmitting a packet to all neighboring nodes. The first forwarding node F1, which is a source node S and a one-hop neighbor node, receives the flooded subscription query packet from the source node S. The first forwarding node F1 stores the identification information of the source node S in the routing table and adds its identification information to the subscription query packet. The first forwarding node F1 then broadcasts the subscription query packet.

The second to fourth nodes F2 to F4 receive the subscription query packet broadcast from the first forwarding node F1. Each of the second to fourth forwarding nodes F2 to F4 stores the identification information of the source node S and the first forwarding node F1 in the routing table, and adds its identification information to the subscription query packet. . Thereafter, each of the second to fourth forwarding nodes F2 to F4 broadcasts a subscription query packet. In this way the subscription query packet is flooded to all other nodes in the multicast group.

Next, the response step is described. After receiving the subscription query packet, the destination node D transmits a control packet (hereinafter, referred to as a subscription reply packet) including a join reply message to the source node S in response thereto. At this time, the first and second forwarding nodes F1 and F2 mediate the subscription response message with reference to the identification information stored in the respective routing tables. The first and second forwarding nodes F1 and F2 which mediate the subscription query packet and the subscription response packet between the source node S and the destination node D are elected as members of the forwarding group.

As such, when the data packet transmission path is established through the inquiry step and the response step, the source node S transmits the data packets to the destination node D. At this time, the first and second forwarding nodes F1 and F2 mediate data packets between the source node S and the destination node D. FIG.

However, the data packet transmission path may be broken due to the mobility of the nodes. Therefore, transmission and reception of control packets should be performed periodically to maintain the path. That is, periodic rerouting is necessary. In this case, the rerouting period affects the reception rate of the data packet and the overhead due to transmission and reception of control packets. As the rerouting period gets shorter, the overhead increases but the loss of data packets decreases. On the other hand, as the rerouting period is longer, the loss of data packets increases, but the overhead decreases.

The network system according to an embodiment of the present invention is enhanced to support a local path recovery function, a function of adjusting a path reset period according to a data packet reception rate, a passive acknowledgment (hereinafter, referred to as a passive ACK), and the like. It can be based on ODMRP (Enhanced-ODMRP, hereinafter referred to as E-ODMRP). The multicast routing method by local path recovery is described below with reference to FIGS. 2A to 2C. A multicast routing method for adjusting the rerouting period according to the reception rate of the data packet is described with reference to FIG. 3 below. A multicast routing method using a passive ACK is described with reference to FIG. 4 below.

2A through 2C are diagrams for describing a multicast routing method using local path recovery according to an embodiment of the present invention. For the sake of brevity, it is assumed that the destination node D leaves and rejoins the multicast group shown in FIG. In the following, the matters already described with reference to FIG. 1 are omitted.

Referring to FIG. 2A, the destination node D floods a control packet (hereinafter referred to as a join request packet) including a join request message to rejoin the multicast group. The destination node D and the second and fifth forwarding nodes F2 and F5, which are one-hop neighbor nodes, receive the flooded subscription request packet from the destination node D.

Referring to FIG. 2B, the second and fifth forwarding nodes F2 and F5 that receive the flooded subscription request packet operate as temporary forwarding nodes. That is, the second and fifth forwarding nodes F2 and F5 intermediate data packets temporarily until one of the second and fifth forwarding nodes F2 and F5 is selected as a member of the forwarding group.

The destination node D selects a temporary forwarding node constituting the shortest path as a member of the forwarding group. To this end, a control packet (hereinafter, referred to as a path selection packet) including the path selection message and identification information of the temporary forwarding node constituting the shortest path is flooded. Here, the temporary forwarding node constituting the shortest path is the second forwarding node F2. That is, the identification information included in the path selection packet indicates the second forwarding node F2. The second and fifth forwarding nodes F2 and F5 receive the flooded path selection packets from the destination node D.

Referring to FIG. 2C, the second and fifth forwarding nodes F2 and F5 confirm identification information included in the path selection packet. Since the identification information included in the path selection packet indicates the second forwarding node F2, the second forwarding node F2 is selected as a member of the forwarding group and continues to relay data packets. On the other hand, the fifth forwarding node F5 no longer transmits data packets to the destination node D.

As described above, according to the multicast routing method using the local path recovery, control packets are transmitted and received between the destination node D and the 1-hop neighbor node. Thus, for rerouting, overhead can be reduced compared to the transmission and reception of control packets between the source node S and all other nodes in the multicast group.

3 is a flowchart illustrating a multicast routing method of adjusting a rerouting period according to a reception rate of a data packet. Here, the rerouting period means a time interval in which the control packet (subscription query packet) is flooded from the source node (S).

Referring to FIG. 3, in a network system according to an exemplary embodiment of the present disclosure, a reception rate of a data packet is monitored (step S110). At this time, the reception rate of the high data packet means that the mobility of the nodes is low, so that the data packet loss is low. In this case, it is desirable to increase the rerouting period to reduce the overhead. On the other hand, the low data packet reception rate means that the mobility of nodes is high, resulting in high data packet loss. In this case, it is desirable to reduce the rerouting period to increase the reliability.

Then, the reception rate of the data packet is compared with the reference value (step S120). If the reception rate of the data packet is equal to the reference value, then the reception rate of the data packet is continuously monitored. On the other hand, if the reception rate of the data packet is different from the reference value, the rerouting period is adjusted (step S130). That is, the time interval at which the subscription query packet is flooded from the source node S is adjusted. For example, if the reception rate of the data packet is greater than the reference value, the rerouting period is increased to reduce overhead. On the other hand, if the reception rate of the data packet is smaller than the reference value, the rerouting period is reduced to increase the reliability.

As described above, according to the multicast routing method of adjusting the rerouting period according to the reception rate of the data packet, the rerouting period is maintained at an appropriate level to reduce the overhead due to transmission and reception of control packets.

4 is a diagram illustrating a multicast routing method using a passive ACK. Referring to FIG. 4, the source node S transmits a data packet to the first forwarding node F1, which is a one-hop neighbor node. The first forwarding node F1 broadcasts the data packet transmitted from the source node S. The source node S may receive the broadcast data packet from the first forwarding node F1. At this time, the data packet transmitted from the first forwarding node F1, which is a lower node, to the source node S, which is an upper node, is treated as a passive ACK.

Since each node belonging to the multicast group stores the identification information of the upper node and the lower node in the routing table, when the source node S receives the passive ACK, whether the data packet is normally transmitted by referring to the routing table. You can check. Therefore, even if the first forwarding node F1 does not transmit a separate ACK to the source node S after receiving the data packet from the source node S, the source node S normally receives the data packet through the passive ACK. You can check whether it was sent.

If the source node S does not receive the passive ACK several times from the first forwarding node F1, the source node S determines that the first forwarding node F1 does not belong to the forwarding group and removes the data packet. 1 does not transmit to the forwarding node F1. The same applies to the relationship with the first and second forwarding nodes F1 and F2.

As described above, according to the multicast routing method using the passive ACK, it is possible to reduce the overhead due to the data packet is transmitted to the unnecessary forwarding node.

An embodiment of the present invention provides a multicast routing method for recovering a lost data packet by applying a network coding scheme in an E-ODMRP-based network system. When the network coding scheme is applied to the E-ODMRP-based network system, the target node may recover the lost data packet by decoding the encoded packets stored in the buffer without retransmitting the lost data packet. This is explained in more detail with reference to FIGS. 5 to 7 below.

5 is a diagram illustrating data packets encoded by a network coding technique. Referring to FIG. 5, first to N th data packets P1 to PN and first to N th encoded packets EP1 to EPN are illustrated.

The first to Nth data packets P1 to PN include original data. The first to Nth encoded packets EP1 to EPN include encoding vectors and encoded data. Here, the encoding vector means a set of values used for encoding and decoding by a network coding technique.

As an embodiment of the present invention, a random linear coding scheme may be applied to the network coding scheme. According to the random linear coding scheme, each element of the encoding vector is randomly selected from values defined in a finite field (also referred to as a Galois field).

All operations used in network coding techniques are defined and performed in finite fields. Thus, a finite field algorithm is used. Addition and subtraction operations in finite field operations consist of exclusive ORs only. On the other hand, for the multiplication operation, an algorithm or a table reference algorithm in which shift and addition operations are repeated is used.

Encoding and decoding by a network coding scheme according to an embodiment of the present invention are performed in consideration of generation number and size. The generation number is updated every time a subscription query packet occurs at the source node. That is, data packets and coded packets generated during the time (path resetting period) from the time of generation of the previous subscription query packet to the time of generation of the next subscription query packet have the same generation number. For the sake of brevity, it is assumed that the first to N th data packets P1 to PN occur sequentially at the source node. In addition, it is assumed that the first to Nth data packets P1 to PN and the first to Nth encoded packets EP1 to EPN belong to the same generation. Therefore, the first to Nth data packets P1 to PN and the first to Nth encoded packets EP1 to EPN have the same generation number.

Data packets originating at the source node are once stored in a buffer. Whenever a data packet occurs, all data packets belonging to the same generation among the data packets stored in the buffer are encoded. For example, when the first data packet P1 occurs, the first data packet P1 is stored in a buffer and encoded. At this time, the first encoded packet EP1 is obtained as a result of encoding the first data packet P1. Thereafter, when the second data packet P2 occurs, the second data packet P2 is stored in the buffer, and the first and second data packets P1 and P2 belonging to the same generation are encoded. At this time, the second encoded packet EP2 is obtained as a result of encoding the first and second data packets P1 and P2. When the N-th data packet PN continues in this manner, the N-th data packet PN is stored in a buffer, and the first to N-th data packets P1 to PN belonging to the same generation (having the same generation number). ) Is encoded. At this time, the Nth encoded packet EPN is obtained as a result of encoding the first to Nth data packets P1 to PN.

The generation size is determined by the number of data packets to be encoded. For example, the generation size of the first encoded packet EP1 obtained by encoding the first data packet P1 is one. The generation size of the second encoded packet EP2 obtained by encoding the first and second data packets P1 and P2 is two. The generation size of the Nth encoded packet EPN obtained by encoding the first to Nth data packets P1 to PN is N. As such, encoded packets having the same generation number may have different generation sizes. Here, the last N-th coded packet (EPN) generated among the coded packets having the same generation number has the largest generation size. On the other hand, the size of the encoding vector (the number of elements of the encoding vector) corresponds to this generation size.

In an E-ODMRP-based network system according to an embodiment of the present invention, as described with reference to FIG. 3, a rerouting period (time interval at which the subscription query packet is flooded) is adjusted according to the reception rate of the data packet. Thus, the maximum size of the generation for encoding varies with the reception rate of the data packet. For example, the longer the rerouting period, the greater the number of data packets that occur during one period, so the maximum size of the generation will also increase. On the other hand, a shorter rerouting period will reduce the number of data packets that occur during one period, thus reducing the maximum size of the generation.

6 is a flowchart illustrating an encoding method by a network coding scheme according to an embodiment of the present invention. For the sake of brevity, an encoding method for first to Nth data packets occurring at a source node during any rerouting period is described.

Referring to Fig. 6, first, a first data packet is generated (step S210). Then, the first data packet is stored in the buffer (step S220). Data packets having the same generation number among the data packets stored in the buffer are encoded (step S230). Here, the first coded packet is obtained as a result of encoding the first data packet.

Next, it is checked whether a subscription query packet is generated (step S240). If no subscription query packet occurs, the first coded packet is broadcasted (step S250). Thereafter, steps S210 to S240 are repeated. On the other hand, if a subscription query packet occurs, an encoded packet obtained by encoding data packets having the same generation number among the data packets stored in the buffer is added to the subscription query packet (step S260). This subscription query packet is then broadcasted (step S270). Thereafter, the generation number is updated (step S280).

Assume that the subscription query packet does not occur until the Nth data packet occurs. Therefore, steps S210 to S250 are repeated until the Nth data packet occurs.

When the second data packet occurs (step S210), the second data packet is stored in the buffer (step S220). Then, a second encoded packet is obtained as a result of encoding the first and second data packets stored in the buffer (step S230). Thereafter, the second coded packet is broadcasted (step S250). When the Nth data packet continues to be generated in this manner (step S210), the Nth data packet is stored in the buffer (step S220). Then, the first to Nth data packets stored in the buffer are encoded (step S230).

Since the subscription query packet is generated next to the Nth data packet, the Nth encoded packet obtained as the encoding result of the first to Nth data packets is added to the subscription query packet (step S260). This subscription query packet is then broadcasted (step S270). Here, the Nth coded packet broadcasted through the subscription query packet has a generation size of the largest value among the coded packets belonging to the same generation. Thereafter, the generation number is updated (step S280).

The encoding method as described above may also be applied to the forwarding node. However, in the forwarding node, instead of generating data packets, data packets or encoded packets will be received. Instead of confirming whether a subscription query packet is generated, whether or not a subscription query packet is received will be confirmed.

7 is a flowchart illustrating a method of recovering a lost data packet through decoding by a network coding scheme according to an embodiment of the present invention. For the sake of brevity, a decoding method for encoded packets received at a destination node during any rerouting period is described.

Referring to Fig. 7, first, an encoded packet is received (step S310). In this case, the encoded packet may be delivered through a subscription query packet. The received encoded packet is stored in a buffer (step S320). In step S330, it is possible to decode the encoded packets having the same generation number among the encoded packets stored in the buffer. At this time, if the number of encoded packets having the same generation number is equal to or greater than the largest value among the generation sizes they have, it is determined that the encoded packets can be decoded. This means that as the number of encoded packets belonging to the same generation increases, the decoding probability of encoded packets increases.

If it is determined that the encoded packets cannot be decoded, steps S310 and S320 are repeated. On the other hand, if it is confirmed that the encoded packets can be decoded, the encoded packets having the same generation number among the encoded packets stored in the buffer are decoded (step S340). When the encoded packets are decoded, data packets originating at the source node are obtained. Thus, data packets lost during path recovery or resetting are recovered from the data packets obtained as a result of decoding the encoded packets (step S350).

In an E-ODMRP-based network system, subscription query packets are duplicated because subscription query packets are flooded throughout the network. However, as described above, when applying a network coding scheme according to an embodiment of the present invention, an encoded packet added to a subscription query packet includes encoded components for all data packets belonging to the corresponding generation. Therefore, if the subscription query packet is received in duplicate, the number of encoded packets having the same generation number increases, so that the decoding possibility increases. This means that the probability of data packets lost during path recovery or reestablishment being increased is high.

As described above, when the network coding scheme is applied to the E-ODMRP-based network system, the destination node decodes the encoded packets stored in the buffer without retransmission of the lost data packets, thereby losing the lost data packets. Can be restored. As a result, in an E-ODMRP-based network system, reliability may be improved without increasing overhead.

It will be apparent to those skilled in the art that the structure of the present invention can be variously modified or changed without departing from the scope or spirit of the present invention. In view of the foregoing, it is intended that the present invention cover the modifications and variations of this invention provided they fall within the scope of the following claims and equivalents.

Claims (13)

In a multicast routing method in a wireless mobile multihop network system:
Establishing a data transmission path between a source node and a destination node through transmission and reception of control packets;
Whenever a data packet generated at the source node is stored in a buffer of the source node, at least one data packet having the same generation number among the data packets stored in the buffer of the source node is encoded by a network coding scheme to encode an encoded packet. Generating a; And
And transmitting the encoded packet from the source node to the destination node through the data transmission path. The method of claim 1,
Whenever the encoded packet received at the target node is stored in the buffer of the target node, determining whether to decode at least one encoded packet having the same generation number among the encoded packets stored in the buffer of the target node. ; And
If it is determined that the at least one encoded packet can be decoded, further comprising decoding the at least one encoded packet. The method of claim 1,
The setting of the data transmission path may include: periodically generating a subscription query packet among the control packets; And
Flooding the subscription query packet to all nodes belonging to a multicast group. The method of claim 3, wherein
And adding to the next subscription query packet an encoded packet having a generation size of the largest value among the encoded packets generated during the time between the generation of the previous subscription query packet and the generation of the next subscription query packet. Routing method. The method of claim 4, wherein
The generation size of the encoded packet is determined according to the number of data packets that are encoded to generate the encoded packet. The method of claim 3, wherein
In the step of periodically generating the subscription query packet, the generation period of the subscription query packet is variable according to the reception rate of the data packet. The method of claim 5, wherein
In the step of generating the subscription query packet periodically, the generation period of the subscription query packet decreases as the reception rate of the data packet increases, and increases as the reception rate of the data packet decreases. The method of claim 2,
The generation number of data packets and encoded packets generated during the time between the generation of the previous subscription query packet and the generation of the next subscription query packet is updated in response to the generation of the previous subscription query packet. The method of claim 2,
In the determining of the decodability, the number of at least one encoded packet (hereinafter, at least one decoded encoded packet) having the same generation number among the encoded packets stored in the buffer of the target node is the at least one decoded object. And determining that the at least one encoded packet to be decoded can be decoded if it is equal to or greater than the largest value of the generation sizes of the encoded packet. The method of claim 2,
Recovering lost data from data packets obtained by decoding the at least one encoded packet. The method of claim 1,
Establishing the data transmission path comprises performing local path recovery. The method of claim 1,
The step of transmitting the encoded packet from the source node to the destination node includes determining whether the encoded packet is normally transmitted through a passive acknowledgment response. The method of claim 1,
In the generating the coded packet, a random linear coding scheme is applied as the network coding scheme.

Images