KR100656317B1 - Label switching method using a ethernet frame and a labeled frame structure - Google Patents

Abstract

Disclosed is a label exchange method using an Ethernet frame and a labeled frame structure. When receiving a tunnel setup frame including a tunnel ID for identifying a path for transmitting an Ethernet frame and an address list of connected terminals, the address for frame forwarding is learned based on the tunnel ID of the frame, and the address list and the tunnel are received. Save the ID to set up the tunnel. Then, a labeled Ethernet frame is transmitted through the established tunnel. This eliminates wasted header lengths, improves compatibility with existing Ethernet devices, and improves quality assurance, scalability, and VPN support for Ethernet.

Tunnel ID, Local ID, Destination Address, Ethernet Frame

Description

Label switching method using a ethernet frame and a labeled frame structure}

1 is a view showing the structure of a conventional Ethernet frame,

2A illustrates the structure of a labeled Ethernet frame in accordance with the present invention;

2b illustrates an embodiment of a frame used for establishing an LSE tunnel according to the present invention;

3 is a diagram illustrating a transmission process of a labeled Ethernet frame according to the present invention in a linear network;

4A illustrates an embodiment of an LSE tunnel establishment process according to the present invention;

4B is a timing diagram illustrating a process of establishing a connection with a terminal by a terminal Ta in the network of FIG. 4A;

5A is a diagram illustrating a data transmission process in an LSE network according to the present invention;

FIG. 5B is a timing diagram illustrating a process of establishing a connection using an ARP extension protocol between Ta and Tc in an LSP tunnel established in the network of FIG. 5A;

6 is a diagram illustrating a flow for supporting mobility of a wireless LAN using an LSE according to the present invention;

7 is a flow chart illustrating the processing of a labeled Ethernet frame in accordance with the present invention;

8 is a flowchart illustrating a flow of an embodiment of a tunnel setting method according to the present invention;

9 is a flowchart illustrating a flow of an embodiment of a connection establishment method according to the present invention, and

 10A and 10B are flowcharts illustrating the flow of a data transmission method according to the present invention.

The present invention relates to an Ethernet frame switching method and a labeled frame structure, and more particularly, to a method for providing a label-switched switching using an Ethernet frame and a labeled Ethernet frame structure.

A conventional method of relaying Ethernet frames by an Ethernet switch node in an Ethernet communication network is as follows. When the Ethernet switch node receives the Ethernet frame format of the standard specification defined in IEEE 802.3 as shown in FIG. do. At this time, when the Ethernet switch does not know the port to which the received Ethernet frame is to be transmitted, the frame is broadcasted to all output ports of the switch device. However, this method increases the side effects such as unnecessary traffic spreads throughout the network when the Ethernet network grows, causing overload.

As one of the methods to solve this problem, VLAN Tagging method is introduced. This is intended to achieve functional improvements by inserting VLAN Tag fields containing additional information into the basic Ethernet header, such as grouping Ethernet devices into logical VLAN groups or providing differentiated transmission services according to priorities.

VLANs have become an important technology that will enable short-range Ethernet to become a campus-class enterprise network, resulting in improved Ethernet scalability, while wasteful bandwidth due to longer header fields and up to 4096 VLAN groups on a single Ethernet network. There is a problem such as a limitation that can only support.

The L2-VPN technology has been proposed to introduce the advantages of the IP MPLS technology to the Ethernet network while overcoming the above limitation of the number of VLAN groups. L2 VPN technology inserts an MPLS Label in front of a VLAN header to connect LSP connections between geographically separated VLAN groups. The L2-VPN technology further improves the scalability of Ethernet, which not only widens the range of Ethernet to the metrocore network, but also introduces some advantages of MPLS technology such as bandwidth control between VLANs.

However, since L2-VPN assigns MPLS Label depending on VLAN Tag information, the range of technology that can be implemented as MPLS technology is limited according to VLAN Tagging capability. For example, real-time application service traffic such as VoIP and bulk file transfer traffic should provide differentiated services in the lower transmission stage. However, L2-VPN cannot differentiate traffic flows according to higher applications. . In addition, bandwidth is wasted due to the increased header length.

The technical problem to be achieved by the present invention is to improve the problems such as the increased header length burden and dependency on VLAN tag of the conventional VLAN method and the L2-VPN method, and to use MPLS label switching without using a standard Ethernet header without modification. An object of the present invention is to provide a label switching method using an Ethernet frame.

Another technical problem to be solved by the present invention is to improve the problems such as the increased header length burden and the dependency on the VLAN tag of the conventional VLAN method and the L2-VPN method, and the MPLS label without using the standard Ethernet header without modification. It is to provide a labeled frame structure that can be applied to the switching method.

Another technical problem to be solved by the present invention is to improve the problems such as the increased header length burden and the dependency on the VLAN tag of the conventional VLAN method and the L2-VPN method, and the MPLS label without using the standard Ethernet header without modification. The present invention provides a computer-readable recording medium that records a program for executing a label switching method using an Ethernet frame to enable a switching method.

In order to achieve the above technical problem, an embodiment of the tunnel configuration method according to the present invention, receiving a tunnel configuration frame including a tunnel ID for identifying the path for transmitting the Ethernet frame and the address list of the connected terminals; step; Learning an address for frame forwarding based on the tunnel ID of the frame; And storing the address list and the tunnel ID.

In order to achieve the above technical problem, an embodiment of the connection establishment method according to the present invention is to create a tunnel by broadcasting a tunnel setup frame including a first tunnel ID and an address list of connected terminals, and each of the Allocating a local ID for identifying terminals to each of the connected terminals; When receiving the connection establishment message through the generated tunnel, the destination address of the connection establishment message is exchanged with the address of the connection establishment target terminal and the source address is converted into the first tunnel ID and the first local ID of the connection establishment target terminal. A second step of forwarding the exchanged value after the exchange; And a third step of, when receiving a response to the connection establishment message, exchanging a destination address of an acknowledgment message including a second local ID of the terminal that has transmitted the connection establishment message to a predetermined tunnel ID and forwarding it. do.

In accordance with an aspect of the present invention, there is provided a method of data transmission, comprising: separating a destination address of an Ethernet frame received from a predetermined terminal into a tunnel ID and a local ID field; And exchanging a destination address of the Ethernet frame with the tunnel ID and exchanging a source address with the local ID.

In order to achieve the above technical problem, an embodiment of a computer-readable recording medium recording a labeled frame structure according to the present invention is provided in a space of an upper 3 byte of each of a destination address field and a source address field of an Ethernet frame. A TID field assigned to distinguish a given tunnel; An LID field allocated to spaces of lower 3 bytes of each of the destination address field and the source address field to distinguish predetermined terminals; And a type field indicating a case where the destination address field and the source address field are divided into the TID field and the LID field.

This eliminates wasted header lengths, improves compatibility with existing Ethernet devices, and improves quality assurance, scalability, and VPN support for Ethernet.

Devices related to LSE (Label Switched Ethernet) service according to the present invention are classified into the following four classes by function. Class-1 Equipment is a legacy device that supports Ethernet standards, including user terminals, Class-2 Equipment is an Ethernet switch device that supports LSE and label switching, and Class-3 Equipment is a heterogeneous network such as LSE network and IP network. Gateway device for interworking with. Class-4 Equipment is a circuit-based device that supports the LSE protocol.

Class-2 device performs frame transmission by label switching at Ethernet level and performs Ethernet switching and spanning tree establishment function for interworking with Class-1 device that does not support LSE protocol. Class-1 devices and Class-2 Ethernet devices form a unit network separated by a Class-3 device or an IP router. In the present invention, this is called a unit Class-2 network.

Class-3 device acts as Border Gateway between Class-2 network or between Class-2 network and external IP network. Class-3 device performs label switching by LSE protocol and IP forwarding by IP address lookup, and supports IP protocol when directly connected to IP network. Class-4 devices represent a family of circuit-based devices, such as ATM and OxC, which incorporate the LSE control protocol.

The present invention interprets the destination address of a received Ethernet frame as an MPLS label in configuring an Ethernet switching system, thereby determining an output port and an output label, exchanging labels, and transmitting the result according to a predefined service class and priority. The present invention relates to a method and an apparatus for enabling an Ethernet frame to be transmitted by an MPLS label switching method.

Hereinafter, with reference to the accompanying drawings will be described in detail a label switching method and apparatus using an Ethernet frame according to the present invention.

2A illustrates a frame structure for performing label switching according to the present invention.

Referring to FIG. 2A, the frame follows the Ethernet header format of the IEEE 802.3 standard, and the protocol protocol switching (MPLS) is performed by using a destination protocol 110 and a source address 120. Used as a label of). The destination address 110 and the source address 120 are 6 bytes each, and are divided into 3 bytes and used as TID (Tunnel ID) 205 and 215 and LID (Local ID) 210 and 220 fields. TID is used as a tunnel number, and LID is used for identification of terminals connected through a tunnel.

The 6-byte destination address 110 can implement various types of label construction methods, such as using the entire label as a single label value or using a total of six labels for 1 byte each as multiple layered label structures. have. Length / type 225 indicates that the LSE (Label Switched Ethernet) protocol according to the present invention is used. The payload 230 contains the contents of the frame, and the FCS 235 is an abbreviation of Frame Check Sum.

LSE is a new frame transmission method proposed by the present invention that improves quality assurance, scalability, VPN support ability, etc. by introducing advantages of connected services such as MPLS to Ethernet. LSE protocol is interconnected by establishing Tunnel Merging Tree or Tunnel LSP (TLSP) Tree in Ethernet network. Terminal devices may establish connection and resource reservation by exchanging and receiving a signaling protocol that extends Address Resolution Protocol (ARP) through a merging tree or a signaling protocol similar to LDP (Label Distribution Protocol) of MPLS.

2B is a diagram illustrating an embodiment of a frame used to establish an LSE tunnel according to the present invention. Through the LSE tunnel, class-2 devices establish signaling protocol exchange and connection with each other.

Referring to FIG. 2B, the LSE Tunnel Establishment (LTE) frame includes an Ethernet Multicast / Broadcast Address field 255, a Tunnel ID field 260, a length / type field 275, a Root address field 280, The policy data field 285 and the access list field 290 are included.

Ethernet Multicast / Broadcast Address (255) is an address for broadcasting a frame for establishing a tunnel in a network, Tunnel ID 260 is an identifier for identifying a tunnel, and a Length / Type field 275 is a frame for LSE protocol. Root Address 280 indicates the address of the node that transmitted the tunnel setup frame, and each node includes one or more trees whose root node is the node. The policy data 285 includes tunnel characteristics such as bandwidth of a frame. The access list 290 contains IP addresses of terminals allowed to access through the established tunnel.

3 is a diagram illustrating a delivery process of a labeled Ethernet frame according to the present invention.

Specifically, FIG. 3 is transmitted while the destination address labeled is replaced in the process of transmitting an Ethernet frame from a terminal 330 connected to a general Ethernet network 300 to a terminal 340 connected to another Ethernet network 320. The process is showing.

In FIG. 3, the label switching network 310 is a network composed of switches to which the present invention is applied, and the switches 312 and 314 constituting the label switching network 310 all interpret the target address of the Ethernet frame as a label and replace the label. I have it. The switches 302, 304, 322, and 324 constituting other general Ethernet networks 300 and 320 are all devices that perform Ethernet frame forwarding by a method defined by the IEEE 802.3 standard.

In order to apply the label replacement method of the present invention, a label that can be mutually recognized by a signal protocol between a user terminal and devices supporting the method should be defined. Devices using this method, for example, using a signaling protocol similar to MPLS's Label Distribution Protocol (LDP) protocol, or a modification of the ARP (Address Resolution Protocol) used by IP terminals to know the Ethernet address. You can define the Ethernet address to use as a label by sending and receiving messages between them.

3 has negotiated the use of the Ethernet address b1 as a label between the start node 330 and the first intermediate node 312 by the above-described signaling protocol, and the first intermediate node 312 and the second intermediate node 314 are negotiated. B2 is negotiated between the nodes and b3 is negotiated between the second intermediate node 314 and the destination node 324. As such, the Ethernet addresses negotiated between the nodes to which the present invention is applied should be assigned a unique value so as not to overlap with the actual Ethernet addresses of other nodes on the intermediate path.

When the Ethernet address to be used as the label is determined by the above process, the switches to which the present invention is applied broadcast the arbitrary Ethernet frame having the label value as the source address to the Ethernet network so that general Ethernet switches can recognize the address value. broadcast).

For example, in FIG. 3, the intermediate node 312 uses b1 as a source address so that the nodes 302 and 304 of the first Ethernet network 300 recognize b1, which is a label value negotiated with the start node 330. Any frame having a broadcast address as a destination address is broadcasted to the first Ethernet network 300.

The nodes 302 and 304 of the first Ethernet network 300 then learn the source address and input port by the MAC learning function of the Ethernet switches. Therefore, when the start node 330 transmits an Ethernet frame whose destination address is b1 to the first Ethernet network 300, the nodes 302 and 304 of the first Ethernet network 300 transmit the frame to the first intermediate node 312. To pass.

The first intermediate node 312 replaces the destination address of the received frame with the new intermediate address b2 by negotiating with the second intermediate node 314, and the second intermediate node 314 again transfers the destination address (the destination node). According to the negotiation with 340, it is replaced by b3 and transmitted through the second Ethernet network 320. In the second Ethernet network 320, a frame having a destination address of b3 may be delivered to the destination node 340 by broadcasting and MAC learning functions.

4A is a diagram illustrating an embodiment of an LSE tunnel establishment process according to the present invention.

In a network in which Ethernet and Class-2 devices are mixed, Class-2 devices 415, 420, and 425 establish Ethernet tunnels for exchanging signaling protocols and establishing connections.

Class-2 devices periodically broadcast the LSE Tunnel Establishment (LTE) Frame of FIG. 2B containing information on tunnel characteristics such as List and Resource information of a terminal directly connected to the network.

In a subnet, Class-2 devices have a 3-byte TID number space and a 3-byte LID number space that can be dynamically allocated and used. Class-2 device selects and assigns a unique number from the TID Number Pool for each tunnel it establishes. This number must be unique within the unit Class-2 network. The TID number is mapped to the SA (Source Address) field of the LTE frame in the form of xx: xx: xx: 00: 00: 00, which is a tunnel address to be used as a destination address of all Ethernet frames using a tunnel in the future.

When each class-2 switch broadcasts an LTE frame, the Ethernet switches present on the frame transmission path are xx: xx: xx: 00: 00 recorded in the SA field of the LTE frame while the LTE frame is propagated on the Ethernet network. Learn about: 00 type tunnel address. Therefore, if a node transmits a frame whose destination address (DA) is the SA of the LTE frame, the frame is delivered to a class-2 switch that broadcasts the LTE frame. The data transmission path learned from LTE frame broadcasting is called an 'Ethernet Tunnel'.

FIG. 4A illustrates an LTE frame in which three LSE devices S1 415, S2 420, and S3 425 are connected to two Ethernet devices E1 430 and E2 435. By broadcasting, the LSE Tunnel Merging Tree having S3 425 as a Root Node is established.

Ta 400, Tb 405, and Tc 410 represent User Terminals, respectively. G1 is a TID assigned to this tunnel by S3425, and as a result of tunnel establishment by broadcasting the LTE frame, all Ethernet frames whose destination address is G1 can be delivered to S3425.

The LTE frame includes policy data containing tunnel characteristics such as bandwidth and an access list containing IP addresses of terminals allowed to access through the established tunnel. Therefore, Class-2 devices (ie, S1 and S2) receiving the LTE frame store the tunnel information and the access list of the LTE frame.

If a Class-2 device has an IP packet to forward to another Class-2 device, the Access List is searched based on the destination address of the IP packet to select the most appropriate tunnel ID (TID), and the Ethernet frame whose target address is the TID. Encapsulate and transmit through selected tunnel. Tunnel Merging Tree is a unidirectional path that can transmit Frame only to Root Node, but in Class-2 network, all Class-2 devices establish one or more Tunnel Merging Tree which makes itself Root, so mutual IP You can send and receive packets. Class-2 devices establish resource reservation and label switched connection by sending and receiving signaling packet through tunnel in this way.

FIG. 4B is a timing diagram illustrating a process of establishing a connection with the Tc by the terminal Ta in the network of FIG. 4A.

Terminal devices supporting LSE require a connection establishment for IP packet delivery to a Class-2 device using an ARP-extended protocol. The extended ARP protocol includes traffic characteristics, bandwidth requirements and authentication information required by the terminal. In FIG. 4B, the frame type transmitted between nodes is expressed in the form of 'Frame_type [DA] [SA]'.

Referring to FIG. 4, S1 415, which is a Class-2 device that receives an ARP request from the terminal Ta 400, searches for stored Access Lists and selects a suitable tunnel for Tc 410. S1 (415) encapsulates information necessary to establish ARP Frame and other connection in LSE Frame with selected tunnel address [G1: 0] as DA and 0x01 as SA, and delivers it to Root Node S3 (425) through tunnel. do. Class-2 devices recognize and handle a frame whose SA value is 0x01 among the frames received through the tunnel as a control message between tunnel edge nodes.

S3 425 selects two LID numbers a1 and a2 from the LID Number Pool assigned to them and assigns them to Ta 400 and Tc 410, respectively. The LID to be allocated to the UE is always assigned by the root node, and the allocated LID must be unique within the unit Class-2 network.

Class-2 device acts as proxy of peer terminal to terminal directly connected through Physical Link, and uses 6 bytes of [TID: LID] combining TID of tunnel and LID assigned to terminal as proxy address. That is, in FIG. 4B, S3425 has [G1: a2] combining the TID (= G1) and the LID (= a2) given to the Tc 410, and the proxy of Ta (400) which is a peer terminal of the Tc (410). Used as an address. Therefore, S3425 transmits an ARP request with [G1: a2] as SA to Tc 410, and in response, Tc 410 returns an ARP reply frame with [G1: a2] as DA.

S3 425 forwards the processing result (ie, LID number, etc.) for Ta A400 request to S1 415, which does not necessarily have to match the ARP transmission path. S1 415 combines the TID (= G1) and the LID (= a1) assigned by S3 425 for Ta (400) and uses A [G1: a1] as the proxy address to ARP to Ta (400). By transmitting the reply, information transfer for label switching between the two terminals Ta 400 and Tc 410 is completed.

4B illustrates a timing diagram in which Ta 400 and Tc 410 exchange Unicast data using a tunnel.

Terminal devices communicate with the counterpart terminal using proxy addresses in the form of [TID: LID] given by Class-2 devices without having to be aware of the existence of the tunnel.

First, Ta 400 transmits an Ethernet Data Frame having DA of [G1: a1] to S1 415. S1 415 separates the DA of the received frame and replaces [G1: 00] with DA and [00: a1] with SA. The address-substituted Frame is delivered to S3425 through the tunnel via E1430, an Ethernet node. At this time, the intermediate nodes such as E1 430 and S3 425 learn [00: a1] which is the Source MAC Address of the frame. S3 425 retrieves the SA (= 00: a1) of the received frame from its local table and retrieves the MAC address and LID (= a2) of the Tc 410 that will deliver the frame. S3 425 replaces the MAC address of Tc 410 with DA and replaces [G1: a2], which combines TID and LID, with a proxy source address, and transmits the Tc 410 to Tc 410.

The Tc 410 transfers a reply frame to the S3425 using DA of [G1: a2]. When the S3 425 forwards the frame downward to the S1 415 along the tunnel path, the S3 425 uses the learned address as the destination address of the frame. S1 415 delivers the frame to E1 430, and the intermediate nodes such as E1 430 have previously learned in the process of data frame delivery in the direction of S1 415-> S3 425. The data frame is transmitted to the S1 415 according to the MAC information. S1 (415) retrieves the DA (= 00: a1) of the received Frame from its local table, replaces the MAC address of Ta (400) with DA, and replaces the proxy source address (= G1: a1) with SA. Deliver to Ta (400).

If the tunnel passes through Legacy Ethernet switches in the data transfer process as described above, the transmission quality of the tunnel traffic may be affected by the traffic introduced from the path outside the tunnel. In this case, the VLAN tag can be used to ensure that tunnel traffic is always processed at high priority.

5A is a diagram illustrating a data transmission process in an LSE network according to the present invention.

The label switching method using the Ethernet tunnel enables the introduction of a connected quality assurance service using ARP in a mixed network of Class-1 and Class-2 devices. However, this method has a limitation in scalability because the TID and LID numbers used by Class-2 devices must be unique within the unit network. If all switch devices in the network support LSE, the tunnel can be replaced by LSP Tunnel, which switches Label at each node, thus eliminating the above limitation of Tunnel Number.

The network of FIG. 5A is Merging to Root Node S3 525 when E1 430, E2 435, which is a Class-1 device of FIG. 4A, is replaced with S4 530, S5 535, which is a Class-2 device. LSP Tunnels gather together

In FIG. 5A, when S1 515 transmits a frame using a label g2 negotiated with an upper node S4 530, S4 530 substitutes g1 and transfers the frame to g3 to S3 525. Therefore, the label used in the tunnel only needs to be unique on the link with the neighbor node, so a sufficient number of LSP tunnels can be established even in a large Class-2 network with a 3-byte TID Number Space. The LID number assigned by the root node (= S3) also needs to be unique within the LSP (Label Switched Path). Such LSP Merging Tree can be established by MPLS protocol or Flood Routing Protocol.

FIG. 5B is a timing diagram illustrating a process of establishing a connection using an ARP extension protocol between Ta and Tc in an LSP tunnel established in the network of FIG. 5A.

Label switching in an ALL LSE network (that is, a network in which all network nodes support LSE) has the following characteristics.

First, all LSE Labels are bidirectional. That is, the TID assigned by one of the two neighboring nodes is commonly used for uplink traffic and downlink traffic through the tunnel on the link. In addition, the LID assigned by the root node is also commonly used up and down. Therefore, as shown in FIG. 5B, the data frames transmitted by S1 or S3 have the same DA and SA values regardless of the up and down directions on the same link.

Second, the DA field of the LSE frame consists of 3 bytes of [TID: LID] combinations, and the TID value is swapped through the intermediate nodes of the tunnel. The SA field always contains the address of the leaf node regardless of the up / down transmission.

Third, Class-2 devices learn the LID value of the DA field when transferring the data frame toward the root of the LSP. The learned LID value is an index that guides the path at the intersection of the tree when the root node forwards the frame down to the leaf node.

Fourth, Class-2 devices all have multicast capability.

In the ALL LSE network, a process of establishing an interconnection between terminals by using an extended ARP protocol is similar to that of FIG. 4B. In FIG. 5B, S1 515 selects a suitable tunnel (TID = g2) that can reach Tc 510 for an ARP request of Ta 500, and ARP for an Ethernet frame with [g2: 01] as DA. Encapsulate the packet and deliver it to S3525. LID 0x01 is a pre-reserved number for the exchange of control messages between tunnel edge nodes. S3 525, which is a root node of the LSP tree, allocates LID numbers a1 and a2 for Ta (500) and Tc (510). As a result of the ARP procedure, Ta (500) assigns [g2: a1], which combines the TID (= g2) of the tunnel selected by S1 515 and the LID (= a1) given by S3 525, to the proxy address of Tc (510). Tc 510 uses [g1: a2] as the proxy address of Ta.

In FIG. 5B, while the Ta 500 transmits the frame toward the Tc 510, the intermediate node S4 530 learns the input link with the LID (= a1) of the DA field. The learned LID and the link are used to select which branch of the tree to transmit the frame forwarded downward along the LSP tree when the Tc 510 returns the frame to the Ta 500. S4 (530) retrieves the LID (= a1) of the DA field of the frame received from S3 (525) and transfers it to S1 (515) as previously learned, thereby exchanging data between Ta (500) and Tc (510) The path is complete.

In summary, Class-2 devices can transmit the data frame upstream from the leaf node of the LSP tree toward the root and learn the LID to deliver the frame transmitted by the root node to the leaf node. If two or more leaf nodes transmit a frame with the same LID value to the root, the class-2 device located at the merging point of both frames learns both paths.

For example, assume that in the network of FIG. 5A, Ta 500 and Tb 505 transmit a frame having DA of [g2: m1] and [g3: m1] toward Tc 510, respectively. The S4 530 located at the point where the uplink paths of the two frames meet each other learns the links in the directions S1 515 and S2 520 together with the LID (= m1) of the frame. Thereafter, when the Tc 510 transmits a frame having [g1: m1] as DA downward, S4 530 duplicates the frame and transmits the frame to S1 515 and S2 520, respectively. In this way, LSE provides multicast data transmission using LSP tunnel.

6 is a diagram illustrating a flow for supporting mobility of a wireless LAN using an LSE according to the present invention.

In the All LSE network, the mobile terminal provides a function of maintaining seamless connectivity while moving between leaf nodes of the LSP tree. 6 illustrates a process of reconfiguring the connection when the mobile terminal Ta 600 moves from the position of S1 610 to the position of S2 610 while communicating with the Tc 640 through an established connection. .

Referring to FIG. 6, while Ta 600 is communicating with Tc 640 via S1 610, [g2: a1] is used as a proxy address of Tc. When S1 610, which is a leaf node of the LSP tree, relays a frame of Ta 620 to S3 630, the SA of the frame is swapped to its own address.

If the Ta 600 moves to S2 610, the S2 side Local Label of the same LSP Tree used in S1 610 is requested through a subscription process to S2 610. When S2 610 informs the TID (= g3) of the LSP tunnel, Ta (600) sets [g3: a1], which combines the new TID and the previously used LID, as the new proxy address of the Tc 640 and S2 (610). To). S2 610 records its address in the SA field of the frame and transmits it upward through the LSP tunnel. In this process, S4 620 simultaneously learns S2 610 as a downstream link together with S1 610 which is the previous position of Ta 600. S3 630 recognizes that the SA address of the frame has been changed and learns that the position of Ta 600 has changed. However, the Tc 640 may continuously transmit data using [g1: a2] as the proxy address of the Ta 600 regardless of the change in the position of the Ta 600.

S3 630 records the address of S2 610 which is a newly moved position of Ta 600 in the SA field when transmitting the frame of Tc 640 downward. The frame of the Tc 640 is downlinked and replicated in S4 620 and simultaneously transmitted to S1 610 and S2 610. S1 (610) knows that the SA field of the received frame is not its address, and stops relaying data through the Wireless Link. Therefore, wireless data is transmitted to Ta 600 only through the moved position of S2 610. Data replicated in S4 620 and delivered to S1 610 is stopped when the MAC Address Refresh Time of S4 620 expires and is deleted.

Next, let's take a look at interconnecting LSE networks through border class-2 devices.

When the size of Class-2 network grows, Class-2 network can be separated based on IP subnet boundary. In this case, each Class-2 network establishes independent LSP Trees, and the Border Class-2 device located at the boundary of the two networks interconnects the LSP Trees of both Class-2 networks.

Border Class-2 device summarizes the information of UEs collected from Local Tree and propagates it to the other network, and the other network uses this information to establish the Exterior LSP Tree to reach the external network. If the connection between Class-2 networks becomes complicated, looping paths may be established. In this case, network should be connected by Class-3 device which performs IP Routing function.

Class-3 device performs interworking with IP network as well as connection function between Class-2 network. Class-3 devices use the extended IP Routing Protocol to connect LSP trees without looping between Class-2 networks. Class-3 device performs frame forwarding by LSP switching between Class-2 networks or packet forwarding by IP address lookup.

Class-4 devices, on the other hand, form a core transport network by incorporating LLS control protocols similar to MPLS to connection-based devices such as ATM and OxC.

7 is a flow chart illustrating the processing of a labeled Ethernet frame in accordance with the present invention.

Specifically, FIG. 7 is a flowchart illustrating a method of processing a label when the label stacking method using multiple labels is applied in the label substitution method of MPLS.

Referring to FIG. 7, when a node receives a labeled Ethernet frame according to the present invention (S700), the node extracts a destination address of the Ethernet frame (S705). Then, the label of the destination address is searched (S710), and one of three predefined actions is taken according to the label value.

First, when the processing form of the label is a PoP label (S720), the highest label among the plurality of labels recorded in the Ethernet destination address is removed (S725). If the label is a switch label (S720), simply replace the upper label with a new label (S730). If the label is a Push label (S720), a new label is added as the top label (S735). In this way, a label switched path (LSP) that is hierarchically included may be implemented. In this case, when the network to which the present invention is applied is hierarchically configured, there is an advantage that a large network can be constructed.

After processing the label according to the shape of each label, the frame is output with reference to the highest label value (S745). If the label is not defined, the frame is discarded (S740).

8 is a flowchart illustrating the flow of an embodiment of a tunnel setting method according to the present invention.

Referring to FIG. 8, a tunnel setup frame including a tunnel ID for identifying a path for transmitting an Ethernet frame and an address list of connected terminals is received (S800). The tunnel setup frame has been described in detail with reference to FIG. 2.

The address for frame forwarding is learned based on the tunnel ID of the frame (S810). Nodes supporting the LSE according to the present invention recognize the tunnel ID as the source address of the tunnel setup frame and perform a MAC learning function based on the tunnel ID recognized as the source address to create a forwarding table. Therefore, if the frame received by each node has the tunnel ID as the destination address, the frame is transmitted to the node which generated and broadcast the tunnel ID.

The address list and the tunnel ID are stored (S820). IP addresses of terminals directly connected to the node broadcasting the tunnel ID are included in the address list. When the frame is received, the address list is searched based on the frame's destination address to read out the corresponding tunnel ID, and the tunnel ID is transmitted as the frame's destination address. Then, the frame addressed to the tunnel ID is delivered to the node which generated the tunnel ID and broadcasted it.

9 is a flowchart illustrating a flow of an embodiment of a connection establishment method according to the present invention.

Referring to FIG. 9, a tunnel is created by broadcasting a tunnel setup frame including a first tunnel ID and an address list of connected terminals, and a local ID for identifying each of the terminals is assigned to each of the connected terminals. (S900).

When the connection establishment message is received through the generated tunnel, the destination address of the connection establishment message is exchanged with the address of the connection establishment target, and the source address is exchanged with a value consisting of the first tunnel ID and the first local ID of the connection establishment target terminal. After forwarding (S910).

When receiving a response to the connection establishment message, the destination address of the response message including the second local ID of the terminal that transmitted the connection establishment message is exchanged with a predetermined tunnel ID and forwarded (S920).

10A and 10B are flowcharts illustrating the flow of a data transmission method according to the present invention.

Referring to FIG. 10A, a destination address of an Ethernet frame received from a predetermined terminal is separated into a tunnel ID and a local ID field (S1000). The destination address of the Ethernet frame is exchanged with the tunnel ID, and the source address is exchanged with the local ID and transmitted (S1010).

Referring to FIG. 10B, a destination address field of an Ethernet frame received from a predetermined terminal is separated into a tunnel ID and a local ID field (S1020). The destination address of the Ethernet frame is exchanged with a value consisting of a tunnel ID and a local ID of the terminal, and the source address is exchanged by the node itself receiving the Ethernet frame and transmitted (S1030).

The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

LSE (Label Switched Ethernet) introduces the advantages of connected services such as MPLS to Ethernet, improving quality assurance, scalability and VPN support in Ethernet.

In addition, the conventional MPLS technology for Ethernet or IP requires an additional information field such as a Shim header for label display in a frame or packet format, whereas the LSE according to the present invention is a destination address and a source address of an Ethernet frame. Interpretation and exchange of label information eliminates waste of header length and improves compatibility with existing Ethernet devices.

In addition, devices supporting the LSE protocol are interconnected by establishing a Tunnel Merging Tree or TLSP (Tunnel LSP) Tree in the Ethernet network, thereby reducing signaling overhead in providing Ethernet-connected services as well as scalability of the LSE protocol. Makes it good LSE provides multicast and wireless LAN mobility support.

Claims (11)

delete Receiving a tunnel setup frame including a tunnel ID for identifying a path for transmitting an Ethernet frame and an address list of connected terminals; Learning an address for frame forwarding based on the tunnel ID of the frame; And Storing the address list and the tunnel ID; In the learning step, after establishing 6 bytes of 3 byte tunnel IDs and 3 byte values as the source address of the tunnel setting frame, MAC learning is performed, and tunnel setting in label switching using the Ethernet frame is performed. Way. The method of claim 2, Receiving a frame, retrieving the address list based on a destination address of the frame, reading the tunnel ID, and transmitting the tunnel ID as a destination address of the frame; Tunnel Establishment Method in Label Switching A first step of generating a tunnel by broadcasting a tunnel setup frame including a first tunnel ID and an address list of connected terminals, and allocating a local ID for identifying each of the terminals to each of the connected terminals; When receiving the connection establishment message through the generated tunnel, the destination address of the connection establishment message is exchanged with the address of the connection establishment target terminal and the source address is converted into the first tunnel ID and the first local ID of the connection establishment target terminal. A second step of forwarding the exchanged value after the exchange; And Receiving a response to the connection establishment message, a third step of exchanging a destination address of an acknowledgment message including a second local ID of the terminal that has transmitted the connection establishment message to a predetermined tunnel ID and forwarding the message; Connection setting method in label switching using an Ethernet frame, characterized in that. The method of claim 4, wherein And the third step includes forwarding a tunnel ID having a node directly connected to a terminal that has transmitted the connection establishment message as a root node, as a destination address of the response message; and label switching using an Ethernet frame. How to establish a connection in. The method of claim 5, The node connected to the terminal that has sent the connection establishment message exchanges the destination address of the response message with the address of the terminal that has transmitted the establishment message and replaces the source address with a value consisting of the first tunnel ID and the second local ID. Exchanging and forwarding; the method of establishing a connection in label switching using an Ethernet frame further comprising. Separating a destination address of an Ethernet frame received from a predetermined terminal into a tunnel ID and a local ID field; And Exchanging a destination address of the Ethernet frame with the tunnel ID and exchanging a source address with the local ID and transmitting the local address. The method of claim 7, wherein Receiving an Ethernet frame labeled with a destination address and a source address, exchanging a destination address of the labeled Ethernet frame with a MAC address of the terminal and exchanging a source address with a destination address of the labeled Ethernet frame. A data transmission method in label switching using an Ethernet frame further comprising. delete A TID field allocated to spaces of upper 3 bytes of each of the destination address field and the source address field of the Ethernet frame to distinguish a predetermined tunnel; An LID field allocated to spaces of lower 3 bytes of each of the destination address field and the source address field to distinguish predetermined terminals; And And a type field representing a case in which the destination address field and the source address field are divided into the TID field and the LID field. 2. delete

Images