JP2002141931A - Router and route control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a router that conducts transfer of packets via networks at a low load by an IP header compression method and controls routing without confusing the packets with the same CID(Context ID) even outputted from nodes and to provide a route control method. SOLUTION: The router is provided with a packet type discrimination means 101 that discriminates a packet type of a regular header, a full header and a compression header of an IP(Internet Protocol) header of a received packets from header information, a sender MAC(Media Access Control) address extract means 112, a destination IP address extract means 106 that extracts a destination IP address from the regular header, an IP address/CID extract means 105 that extracts a destination IP address/CID from the full header, a CID extract means 103 that extracts a CID from the compression header, a storage means 104 that stores a record of the sender MAC address extracted by the means 112 and the destination IP address CID extracted by the means 105 and assigns a specific number to the record, and a CID rewrite means 113 that rewrites the CID of the full header and the compression header into a specific number.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a router device and a route control method used on the Internet, and more particularly to an IP (Internet Protocol) / TCP (Transmissio).
n Control Protocol) / UDP (User Datagram Protocol)
The present invention relates to a router device and a route control method for controlling a route of a packet in which a header portion of a network layer and a transport layer such as ol) is compressed.

[0002]

2. Description of the Related Art At a time when the utilization rate of the Internet is increasing, network traffic is steadily increasing, and at present, infrastructure development cannot keep up. In particular, telephone lines and ISDN lines used to connect homes and small business establishments have low transmission rates, and it is desired to reduce traffic. On the other hand,
In communication between two terminals on the Internet,
It often occurs that IP packets having similar contents of the IP header are continuously transmitted. In such a case, the IP packet transmits the same IP header many times, which is extremely wasteful. Therefore, a compression method of the IP header has been devised.
al Adapter).

As a method for compressing an IP header, an IEEE
(Internet Engineering Task Force) defined RF
C-1144 "Compressing TCP / IP
Headers ”and“ Remote Relay Device ”described in Japanese Patent Application Laid-Open No. 8-223222, etc. Recent ones include RFC-2507“ IP Header Com ”.
The compression method described in RFC-2507 described above uses IPv6 (Internet Protocol).
Version6) Basic header, IPv6 extension header, IPv4
Headers, TCP headers, UDP headers, and even encapsulated IPv6 / IPv4 headers are targets for compression. For convenience of explanation, the part to be compressed in the RFC-2507 will be referred to as an IP header, and the normal IP when the IP header compression is not performed.
The header will be called a "regular header". RF
The compression method described in C-2507 is generally performed by the following procedure.

FIG. 6 is a diagram showing a configuration of an IP header used in a compression method described in RFC-2507.
6A shows the configuration of a regular header, FIG. 6B shows the configuration of a full header, and FIG. 6C shows the configuration of a compressed header. 6, a number 500 indicates the entire IP header, a number 501 indicates a destination IP address included in the IP header 500, and a number 502 indicates a context (Context) ID (CID) included in the IP header 500. The CID 502 is a number indicating that the packet belongs to the same packet stream (a group of packets having similar IP header contents).

[0005] First, negotiation is performed between node A and node B that intend to use the IP header compression method. This negotiation is performed using a packet with a “regular header” shown in FIG.
When the negotiation is established and the IP header compression method becomes available between the node A and the node B, the node A first calls the IP header of the first packet of the packet stream to be transferred "full header". Send to node B as header. As shown in FIG. 6B, the “full header” is obtained by adding a context (Context) ID (CI) to the “regular header” shown in FIG.
D) 502 is added.

[0006] When the node B receives the "full header", it can create a context having exactly the same contents as the node A from the information contained therein. The context is basic data for compressing / decompressing an IP header in a packet stream. The compressed header to be transferred is created by taking the difference between the normal IP header and the context, and is returned to the normal IP header by combining with the context.

[0007] When the same context is created in both nodes by sending the "full header", the node A compresses the IP header using its own context and adds a "compressed header" packet. To the Node B. Upon receiving the packet with the “compressed header”, the node B performs decompression processing of the IP header based on the context owned by the node B.

The above-described IP header compression method uses a "compressed header" with a packet having a "full header" for updating the context and a "compressed header" in which the absolute length of the header is shortened by sending only the difference information from the context. By combining these packets and sending them, it is possible to reduce the number of bits of the IP header to be transferred as a whole.

[0009]

However, in such a conventional router device, the packet with the "compressed header" cannot be route-controlled because the destination IP address is not added to the "compressed header". For this reason, a packet with a “compressed header” cannot be exchanged between nodes existing in networks physically separated by a router.

[0010] Even if a method of implementing an IP header compression function in each router device and performing decompression / compression processing is employed to do this, there is a problem that the load on the router becomes extremely large. Was. Also, in the router device having the IP header compression function, when packets having the same CID are respectively output from a plurality of nodes in the same network, the packets are confused and decompressed. There was a problem.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and it is an object of the present invention to enable a packet in which an IP header is compressed by an IP header compression method to be route-controllable, and to transmit a packet whose IP header is compressed between a plurality of networks. Can be transferred with low load, and the same CID can be sent from multiple nodes.
It is an object of the present invention to provide a router device and a route control method that can control a route without confusing them even when a packet having a packet is output.

[0012]

The router device of the present invention comprises:
A router for connecting a plurality of IP networks, determining means for determining whether an IP header type of an input packet is a regular header, a full header, or a compressed header from header information of a data link layer or a network layer; , A source MAC for extracting a source MAC address from a data link layer header of the packet
Address extracting means; and a destination I from the regular header.
A destination IP address extracting means for extracting a P address; a destination IP address / CID extracting means for extracting a destination IP address and a CID from the full header; a CID extracting means for extracting a CID from the compressed header;
Holding means for holding a record including the source MAC address extracted by the AC address extraction means and the destination IP address and CID extracted by the destination IP address / CID extraction means, and assigning a unique number to the record; CID rewriting means for rewriting the CID of the compressed header to the unique number.

[0013] More preferably, there may be provided a path control means for transmitting a packet whose CID has been rewritten by the CID rewriting means based on the destination IP address. The router device of the present invention thus configured extracts a source MAC address, a destination IP address, and a CID from a packet with a full header of a packet stream to be transmitted, and attaches a unique number to each of them to form one record. It holds and rewrites the CID of the packet with the full header to the unique number, performs route control based on the destination IP address, and sends it out. Whether the packet with the compressed header transmitted thereafter belongs to the packet stream is determined by determining the source MAC address and C of the packet with the compressed header.
The ID is determined by referring to the record, the CID of the packet with the compressed header is rewritten to a unique number in the record, and route control is performed based on the destination IP address in the record and transmitted.

According to the routing control method of the present invention, a first terminal and a second terminal connected to an IP network are connected by a router, and a packet stream subjected to IP header compression is relayed through the router. In the route control method for delivering to a destination address, when transferring a packet stream subjected to IP header compression from the first terminal to the second terminal, the first terminal and the second terminal may include:
Using a packet with a regular header to perform negotiation to enable IP header compression; and, after the negotiation is established, the first
Transmitting the packet with the full header to the second terminal, and extracting the source MAC address, the destination IP address, and the CID from the transmitted packet with the full header. And a record including a source MAC address, a destination IP address, and a CID extracted from the packet with the full header, and an output packet CID as a unique number.
Holding a destination IP address table as one record by adding
Rewriting D to the unique number;
And performing a route control on the basis of the destination IP address and transmitting the rewritten packet.

Further, in the path control method according to the present invention, after transmitting the packet with the full header, the first terminal compresses and compresses an IP header of a next packet to be transmitted using a context held by the first terminal. Creating a header, transmitting the compressed header-attached packet to the second terminal, the router device determines a source MAC address and a CID in the compressed header from the transmitted compressed header-attached packet. Extracting
The source MAC address and the CID in the compressed header and the source MAC address and C in the destination IP address table
Detecting a record with a matching ID and reading the value of the destination IP address and the value of the output packet CID of the record that both match; rewriting the CID of the packet with the compressed header with the value of the output packet CID; Performing a route control based on the value of the destination IP address and transmitting a packet with a compressed header. Also, the second terminal executes a step of creating a context from the contents of the transmitted packet with the full header and a step of decompressing the transmitted packet with the compressed header based on the context held by the second terminal. May be used.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a block diagram illustrating a configuration of a router device according to an embodiment of the present invention. In FIG. 1, a router device includes a physical layer input I / F (interface) 111, a data link layer input I / F 100, a packet type determination unit 101, a branch unit 102, a CID extraction unit 103, a holding unit 104, a destination IP address / CID extraction means 10
5. Destination IP address extracting means 106, selecting means 10
7, route control means 108, data link layer output I / F1
09, physical layer output I / F 110, source MAC (Media
Access Control (Medium Access Control) It is composed of an address extracting means 112 and a CID rewriting means 113.

The router device is a device for interconnecting a plurality of physical networks. The router device shown in FIG. 1 shows only a configuration for performing a unidirectional process of outputting an input from a certain network to another network. In a router device actually used,
By using these in a time-division manner or by using a plurality of them at the same time, a bidirectional route control process and a route control process between a plurality of networks are performed. Hereinafter, the operation of each block of the router device configured as described above will be described in detail.

First, the types of IP headers used in the compression method described in RFC-2507 will be described. These include “regular headers” as described in FIG.
There are three types, “full header” and “compressed header”. The “regular header” is a normal uncompressed IP header and includes a destination IP address 501 as shown in FIG. The "full header" is the IP that carries the context for compressing and decompressing the "compressed header".
This is a header and includes a destination IP address 501 and a CID 502 as shown in FIG. The “compressed header” is an IP header that has been subjected to compression processing and is configured from difference information from a context, and includes a CID 502 as shown in FIG. CID 502
Is a number indicating which packet stream it belongs to.

FIG. 2 is a diagram for explaining a procedure in which the router processes a packet. As shown in FIG. 2A, the packet input to the router device is a packet type A shown in FIG.
01, an IP header 202, and data 203. The physical layer header 200 is equivalent to a preamble (preamble: a special bit string for delimiting) in the case of Ethernet (registered trademark), and is used for control such as distinguishing the head of a packet in the physical layer. .

First, the physical layer input I / F 111 corresponds to FIG.
When the packet shown in (a) is input, the physical layer header 2
00 and the packet type B shown in FIG.
, The data link layer header 201, the IP header 2
02, and a data link layer input I / F 100, a packet type determination unit 101, and a source MAC address extraction unit 11
Transfer to 2.

The data link layer input I / F 100 deletes the data link layer header 201 from the input packet, and comprises an IP header 202 and data 203 as in a packet type C shown in FIG. The packet is created and sent to the branching unit 102 and the CID rewriting unit 113.

The packet type determining means 101 determines whether the IP header in the packet is one of a “regular header”, a “full header”, and a “compressed header” from the data link layer header 201 and the IP header 202 of the input packet. It is determined whether there is. The judgment result is obtained by the branching means 102,
The information is supplied to the selection unit 107 and the CID rewriting unit 113, respectively.

The source MAC address extracting means 112
The source MAC address (address at the data link level of the node that sent the packet) is extracted from the data link layer header 201 of the input packet, and transferred to the holding unit 104. When the packet output from the data link layer I / F 100 is input, the branching unit 102 switches the output destination of the packet according to the determination result of the packet type determining unit 101. That is, the branching means 10
2 is that the input packet is transferred to the destination IP address extracting means 106 when the result of the determination by the packet type determining means 101 is “regular header”, and when the result is “full header”, the destination IP address is / CID extraction means 105, and if it is “compressed header”, it is forwarded to CID extraction means 103.

The destination IP address extracting means 106 extracts the destination I address from the packet having the "regular header".
The P address 501 (see FIG. 6A) is extracted, and the destination IP address is transferred to the selection unit 107. At this time, the next node to be transferred, such as a routing header, which is one of the IPv6 extension headers, is included in the IP header.
If the information to be specified is specified in preference to the destination address included in the v6 basic header or the IPv4 header, this IP address is extracted.

Destination IP address / CID extraction means 105
Is the destination I from the input packet with "full header".
P address 501 and CID 502 (see FIG. 6B)
Are extracted and transferred to the holding unit 104, and the destination IP address 501 is transferred to the selection unit 107. The destination IP address 501 is extracted in the same manner as in the processing in the destination IP address extraction means 106.

The CID extracting means 103 extracts the CID 502 (FIG. 6) from the input packet with the "compressed header".
(See (c)), and stores the CID 502 in the holding unit 1
04. The operation of the holding unit 104 is roughly divided into two cases: a case where a packet with a “full header” is input to the router device and a case where a packet with a “compressed header” is input. Hereinafter, description will be made in order.

(1) When a packet with a “full header” is input to the present router device A packet with a “full header” is input to the present router device, and the destination IP address / CID extracting means 105 extracts the packet from the packet. When the extracted destination IP address and CID are transferred, the holding unit 104 stores the destination IP address as shown in FIG. 3 based on the input source MAC address, input packet CID, and destination IP address.
Create and maintain an address table.

FIG. 3 is a view showing a destination IP address table which the router device has for controlling the route of the "compressed header". As shown in FIG. 3, the holding unit 104
A destination IP address table is created and held in which a record including four items of the AC address, the input packet CID, the destination IP address, and the output packet CID is defined as one record. The following items are held in the four items of the destination IP address table.

Source MAC address term: "full header"
The source MAC address described in the data link layer header of the attached packet (in FIG. 3, SA1, SA2, SA3,
…) Input packet CID: C described in “full header”
ID (s, t, u,... In FIG. 3) Destination IP address: Destination IP written in “full header”
Address (DA1, DA2, DA3,... In FIG. 3) Output packet CID: a unique number (unique number) arbitrarily added to the record by the holding unit 104 (FIG.
Then, w, x, y, ...)

The destination IP address and CID extracted from the packet with the “full header” are stored in the holding unit 104.
Thus, the data is written in units of records in the destination IP address table of FIG. At this time, the destination IP address and the CI
If a record that matches D already exists, it is written over the corresponding record, and if not, a new record is created. The source MAC address field of the corresponding record includes the source MAC address transferred from the source MAC address extraction unit 112, and the output packet CID field includes a unique MAC address not assigned to another record. The number is assigned and written by the holding unit 104. The value of the CID term in the output packet is
Is forwarded to

(2) When a packet with a “compressed header” is input to the present router device When a packet with a “compressed header” is input to the present router device and the CID is transferred from the CID extracting means 103 , Holding means 104, the transferred source MAC
Based on the address and the CID, the destination IP address table shown in FIG. 3 is searched and the output packet CID is read. That is, the holding unit 104 is configured to transmit the source MAC address
The source MAC address transferred from the destination IP address table and the CID transferred from the CID extraction means 103 are the source MAC address term and the input packet C of the destination IP address table.
A record matching the ID term is detected, and the value of the destination IP address term in the matched record is transferred to the selecting means 107, and the value of the output packet CID term is transferred to the CID rewriting means 113.

Returning to FIG. 1, the CID rewriting means 113
Is the CID of the packet from which the data link layer header transferred from the data link layer input I / F 100 is deleted.
In the output packet C transferred from the holding unit 104.
Rewrite with the value of the ID term. At this time, depending on whether the IP header in the packet is “regular header”, “full header”, or “compressed header”, there is a presence / absence of CID, a difference in header configuration, and the like.
The CID is rewritten while switching the processing according to the determination result of 01.

The selecting means 107 determines the destination IP address extracting means 106, the destination IP address / CID extracting means 105,
Destination IP output from any of ID extraction means 103
The address is selected and transferred to the route control means 108.

Based on the destination IP address transferred from the selection means 107, the route control means 108 detects the IP address of the next node to be transferred next from the routing control table held inside like a normal router. Then, the IP address is transferred to the data link layer output I / F 109.

The data link layer output I / F 109 receives A from the IP address transferred from the route control means 108.
M using RP (Address Resolution Protocol)
An AC address is obtained, a data link layer header is added to the packet transferred from the CID rewriting means 113, and the packet is transferred to the physical layer I / F 110. Physical layer output I /
F110 adds a physical layer header to the packet transferred from the data link layer output I / F 109 and outputs the packet.

The present router operates as described above when a packet is input.
It is possible to control the route of a packet with either a “full header” or a “compressed header”, and to process packet streams of the same CID transferred from different nodes at the same time. Next, a processing procedure when the router device is actually connected to a network will be described in detail with reference to FIGS.

FIG. 4 is a connection diagram when the present router device is connected to the network A and the network B. FIG. 4 shows a state in which the router 300 as the present router device is connected to two physical networks, the network A 305 and the network B 306. Is shown. In FIG. 4, terminal C301 and terminal E303 are connected to network A305, and terminal D302 and terminal F304 are connected to network B306.

FIG. 5 shows an example of the operation performed by the router 300 shown in FIG.
FIG. 9 is a control sequence diagram in which terminals (terminal C301 and terminal D302) transmit and receive packets. Now, network A3
It is assumed that a packet stream subjected to IP header compression is transferred from terminal C301 belonging to network 05 to terminal D302 belonging to network B306.

First, the terminal C301 and the terminal D302
Using the packet with the “regular header”, negotiation is performed to enable the use of IP header compression in subsequent packet transmission (number 400). The “regular header” is a normal IP header when the IP header is not compressed as shown in FIG.

When the negotiation is established, first, the terminal C301 sends a packet with a “full header” as the first packet of the packet stream to the terminal D302 through the router 300 (number 401). The “full header” includes a CID 50 as shown in FIG.
2 are included. At this time, the terminal C301 creates and holds a context as basic data for compressing / decompressing the IP header in the packet stream from the contents of the “full header” (number 409). This context is held in association with the CID described in the “full header”.

When the router 300 receives the packet with the "full header", the source MAC address (which becomes the MAC address of the terminal C301) described in the data link layer header 201 (FIG. 2) of the packet and The destination IP address and CID in the “full header” are extracted, and a destination IP address table shown in FIG. 3 is created (number 4).
02). In the record of the destination IP address table, there are an input packet CID term and an output packet CID term, but the extracted CID is included in the input packet CID term, and the output packet CID is an arbitrary one that is not assigned to another record. Number is stored. And the router 30
0 rewrites the CID of the packet with the "full header" with the output packet CID (No. 411), performs a route control process using the destination IP address extracted from the "full header", and transfers the packet with the "full header" to the terminal D.
Transfer to 302 (No. 403). In the CID rewriting process, for example, when a packet having a different source MAC address and the same CID is transferred from the terminal C301 and the terminal E303 to the router 300, the packet output from the router 300 to the network B306 is Since both have the MAC address of the router 300 and cannot be correctly processed by the network B306, this is performed to prevent this.

The terminal D302, which has received the packet with the "full header", creates and holds a context from the contents of the "full header" (number 404). The above is the sequence when the packet with the “full header” is transmitted from the terminal C301 to the terminal D302 through the router 300. Next, a description will be given of a sequence when a packet with a “compressed header” is transmitted from the terminal C301 to the terminal D302. After transmitting the packet with the “full header”, the terminal C301 compresses the IP header of the next packet to be transmitted using the context held by the terminal C301 to create a “compressed header” (No. 410), and The attached packet is sent to the terminal D302 (number 405). The "compressed header" contains the CID. In the compression process, the “compression header” may update the context itself, but does not affect the operation of the router 300 as described later.

When the router 300 receives the packet with the “compressed header”, it extracts the source MAC address described in the data link layer header of the packet and the CID in the “compressed header”, and extracts the “full header” described above. Referring to the destination IP address table shown in FIG. 3 created at the time of receiving the packet with "", the source MAC address is the destination IP address.
Source MAC address entries in the address table (SA1, SA2,
SA3,...) And a record whose CID matches the input packet CID (s, t, u,...) Of the destination IP address table.
Address Item (eg, DA1) Value and Output Packet CI
The value of the D item (for example, w) is read (number 406). Here, even if the “compressed header” updates the context as described above, in the present invention, the destination IP address is updated in the “full header”.
No special processing is required.

The value of the CID term of the output packet in the corresponding record of the destination IP address table is used as the "compression header".
The CID of the attached packet is rewritten (number 412), the route control process is performed from the value of the destination IP address item in the corresponding record of the destination IP address table, and the packet with the “compressed header” is transferred to the terminal D302 ( Number 40
7).

The terminal D302 that has received the packet with the “compressed header” decompresses the “compressed header” based on the context held by the terminal D302 (number 408). When the context of the terminal C301 is updated by the “compressed header”, the context of the terminal D302 is similarly updated. Thereafter, the same processing is performed as long as the transmission of the packet with the “full header” and the transmission of the packet with the “compression header” are repeated.

As described above, the router device according to the present embodiment determines the type of the IP header of an input packet from the header information of the data link layer or the network layer to any of a regular header, a full header, and a compressed header. Packet type determining means 101, a source MAC address extracting means 112 for extracting a source MAC address from a data link layer header of the packet, and a destination IP address extracting means 106 for extracting a destination IP address from a regular header. , Destination IP address / CID extraction means 105 for extracting the destination IP address and CID from the full header, and CID for extracting the CID from the compressed header
Extracting means 103 and source MAC address extracting means 11
2 holds a record including the source MAC address and the destination IP address / CID extracted by the destination IP address / CID extraction means 105 and assigns a unique number to the record; A CID rewriting means 113 for rewriting the CID to a unique number, from the packet with the "full header" of the transmitted packet stream to the source M
An AC address, a destination IP address, and a CID are extracted, a unique number is assigned to each of them, and the record is held as one record. The CID of the packet with the “full header” is rewritten to a unique number, and the route is determined based on the destination IP address. This is controlled and transmitted. Also, it is determined whether the packet with the “compressed header” transmitted thereafter belongs to the packet stream by the source M of the packet with the “compressed header”.
The record is discriminated from the AC address and the CID by referring to the record, the CID of the packet with the “compressed header” is rewritten to a unique number in the record, and route control is performed based on the destination IP address in the record. Is transmitted, so that a packet stream transmitted over a plurality of networks can be transmitted over an IP such as RFC-2507.
When IP header compression by the header compression method is applied, the route can be controlled without decompressing the compressed IP header, so that decompression / recompression is not required in the router, and the load on the router can be reduced. It becomes possible.

Even if packets having the same CID are output from different nodes in the same network, the combination of the MAC address and CID of the source node and the destination IP address are managed in association with each other, and packets with different CIDs are managed. "Compressed header"
CID confusion does not occur in the attached packet, and the route control becomes possible.

[0048]

As described above in detail, according to the present invention, when IP header compression by the IP header compression method is applied, path control can be performed without decompressing the compressed IP header. This eliminates the need for the router to perform decompression / recompression, thereby reducing the load on the router and transferring the packet with the compressed IP header through a plurality of networks with low load. Further, even if packets having the same CID are output from a plurality of nodes, path control can be performed without confusing them.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a router device according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a procedure in which a router device of the present embodiment processes a packet.

FIG. 3 is a diagram showing a destination IP address table that the router device of the present embodiment has for controlling the path of a “compressed header”.

FIG. 4 is a connection diagram when the router device of the present embodiment is connected to a network A and a network B;

FIG. 5 is a control sequence diagram in which two terminals transmit and receive packets via the router device of the present embodiment.

FIG. 6 is a diagram showing a configuration of a regular header, a full header, and a compression header used in the compression method described in RFC-2507.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 data link layer input I / F 101 packet type determination means 102 branching means 103 CID extraction means 104 holding means 105 destination IP address / CID extraction means 106 destination IP address extraction means 107 selection means 108 route control means 109 data link layer output I / F 110 Physical layer output I / F 111 Physical layer input I / F 112 Source MAC address extracting means 113 CID rewriting means 200 Physical layer header 201 Data link layer header 202 IP header 203 Data 300 Router 301 Terminal A 302 Terminal B 303 Terminal E 304 Terminal F 305 Network A 306 Network B 200 Physical layer header 500 IP header 501 Destination IP address 502 CID

Claims (5)

[Claims] In a router device for connecting a plurality of IP (Internet Protocol) networks, the type of an IP header of an input packet is determined based on header information of a data link layer or a network layer from a regular header, a full header, and a compressed header. And a transmission source MAC from a data link layer header of the packet.
(Media Access Control) Source M to extract address
AC address extracting means; destination IP address extracting means for extracting a destination IP address from the regular header; destination IP address and CID (Contex) from the full header.
t ID), a destination IP address / CID extraction unit for extracting CID from the compressed header, and a source M extracted by the source MAC address extraction unit.
Holding means for holding a record including an AC address and the destination IP address and CID extracted by the destination IP address / CID extraction means, and assigning a unique number to the record; and storing the CID of the full header and the compressed header in the unique And a CID rewriting means for rewriting to a unique number. 2. The router device according to claim 1, further comprising a route control unit that sends out a packet whose CID has been rewritten by the CID rewriting unit based on the destination IP address. 3. A route control for connecting a first terminal and a second terminal connected to an IP network by a router device, and relaying a packet stream subjected to IP header compression to the destination address by relaying the router device. In the method, when transferring a packet stream subjected to IP header compression from the first terminal to the second terminal, the first terminal and the second terminal use a packet with a regular header. Performing a negotiation to enable IP header compression; and after the negotiation is established, the first terminal transmits a packet with a full header to the second terminal, and the router In the device, the transmission source M
Extracting an AC address, a destination IP address, and a CID; and a source MA extracted from the packet with the full header.
Holding a destination IP address table as one record by adding an output packet CID, which is a unique number, to a record including a C address, a destination IP address, and a CID; and setting the CID of the packet with the full header to the unique And retransmitting the packet whose CID has been rewritten by performing path control based on the destination IP address. 4. After transmitting the packet with the full header, the first terminal compresses an IP header of a next packet to be transmitted using a context held by the first terminal to create a compressed header; Transmitting a packet with a compressed header to the second terminal; extracting, in the router device, a source MAC address and a CID in a compressed header from the transmitted packet with a compressed header; Source MAC address and CID in the compressed header and source MAC address and C of the destination IP address table
A step of detecting a record having a matching ID and reading a value of a destination IP address and a value of an output packet CID of the record that both match; a step of rewriting a CID of a packet with a compressed header with the value of the output packet CID; And performing a route control based on the value of the destination IP address to send out a packet with a compressed header. 5. The second terminal creates a context from the contents of the transmitted packet with the full header, and decompresses the transmitted packet with the compressed header based on the context held by the second terminal. The route control method according to claim 3, wherein the following is performed.