JPH08130558A - Digital exchange - Google Patents

Abstract

PURPOSE: To distribute group address data to a terminal which is not directly stored in an exchange constituting a public communication network. CONSTITUTION: When a variable length message transmitted from any of terminals is divided into cells and transferred to an interstation server device, a DA analyzing part 11 detects whether a top cell includes a group address(GA) or not, and when the cell includes the GA, information (RGA) corresponding to the GA is written in the top cell. The top cell is temporarily written in a buffer part 14 by a write control part 13. The top cell written in the buffer part 14 is transferred to a transfer control part 17 by a read control part 15 and written in a computer buffer 16. The control part 17 reads out an individual address corresponding to the RGA from a RAM 23 storing a member IA or the like, reads out the top cell from the buffer 16 in accordance with the read and copied value and the read individual address is written in each copied top cell.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital exchange capable of realizing connectionless communication, and more particularly to a digital exchange having a broadcast distribution function in connectionless communication.

[0002]

2. Description of the Related Art In recent years, with the introduction of a large number of computers in companies and the like, LANs have been developed to enable effective use of hardware resources and data between these computers.
The construction of (Local Area Network) has been developed. In this LAN, a connectionless communication protocol is generally used as a communication protocol between terminals.
This connectionless protocol is an LA protocol for digital communication data (packets, etc.) to which the destination address of the destination terminal is attached without performing a procedure for call setup in advance.
The terminal which has sent this digital communication data to M and recognizes whether or not this digital communication data is addressed to itself based on this destination address, and further transfers this digital communication data when it is not addressed to itself. It is a thing.

When such a LAN further develops, WAN (Wide Area Network) like a public communication network is developed.
LANs constructed at remote locations are connected via rk). For example, LANs constructed in each office of a company are connected to each other by ISDN (Integrat).
ed Service Digital Network
k) and other digital public communication networks are used to connect to each other and operate as if they were a single LAN. To make such a LAN-to-LAN connection practical,
It is necessary not to be aware that a public communication network exists between LANs. That is, even when transmitting data to a terminal under the control of another LAN connected by the public communication network, communication must be possible without taking steps such as connecting a line in the public communication network. , The communication speed is remarkably slower than the communication within the own LAN.

Therefore, connectionless high-speed data communication services are also provided in public communication networks.
What is this connectionless high-speed data communication service?
When digital communication data such as a packet, frame, or cell with an incoming address is transmitted to a public communication network without performing a call setup procedure in advance, the transfer of the digital communication data in the switching network Recognize the destination (relay switch, called terminal) based on the called address,
This is a message transfer method for transfer. Therefore, LAN
From the side, it can be treated as if a dedicated line for connecting the LANs to the public communication network is set.

On the other hand, in the LAN, it is possible to deliver one piece of communication data to a plurality of receiving terminals. In this case, TCP / IP (Transmission Co
ntrol Protocol / Internet
Using a broadcast distribution function such as Protocol), digital communication data to which a group address such as an IP address (one address corresponding to the addresses of multiple terminals) is attached is taken in by each terminal, or in a bit map format, etc. Digital communication data is distributed by distributing the digital communication data to each terminal.

[0006]

The delivery of such digital communication data (digital communication data with a group address, hereinafter referred to as "group address data") to a plurality of terminals is performed by a plurality of LANs mutually. It is more useful if it can be realized on the public communication network to be connected. In this way, even if the terminals indicated by the group address exist over a plurality of LANs, it is possible to copy the communication data on the public communication network and distribute the communication data to each LAN. .

However, the connection between the LAN and the public communication network is made by a gateway device such as a router. That is, it is only this gateway device that is directly accommodated in the public communication network, and the individual terminals that make up the LAN are not directly accommodated in the public communication network. Then, each exchange forming the public communication network can recognize only the individual address of each terminal directly accommodated in the public communication network. Therefore, the group address data cannot be delivered to a terminal that is not directly accommodated in each exchange forming the public communication network.

[0008]

An object of the present invention is to provide an exchange capable of delivering group address data to a terminal which is not directly accommodated in each exchange constituting a public communication network.

[0009]

[Means for Solving the Problems]

<Indispensable constituent requirements of the present invention> In order to solve the above problems, the present invention provides a receiving unit (101) for receiving the digital communication data in an exchange (100) which constitutes a communication network using the digital communication data, A transfer information detection unit (102) for detecting transfer information to be distributed to a terminal not directly accommodated in the communication network from the digital data received by the reception unit (101), and the transfer information. An address storage unit (104) that stores address information of the corresponding terminal in advance, and the transfer information detection unit (10).
When the transfer information is detected by 2), the transfer control unit (which reads the address information of the terminal corresponding to the transfer information from the address storage unit (104) and writes the address information in the digital communication data ( 10
3) is provided (corresponding to claim 1).

The significance of the configuration used in this exchange is as follows. [Digital Communication Data] The digital communication data here is block-shaped digital communication data having at least an incoming address. This is because this incoming address enables connectionless communication.

Therefore, this digital communication data may be a packet in packet switching, a frame in a frame relay, or an ATM cell in a cell relay.

The transmission source of this digital communication data may be a terminal directly accommodated in the communication network, or a terminal connected to a LAN (regional information communication network) connected to the communication network. Or a terminal connected to another communication network having a different service area connected to the communication network. [Communication Network] The communication network may be a packet switching network, a frame relay network, or a cell relay network. Moreover, in this case, the switching network need not be the one that transfers the same type of data as the type of digital communication data described above. This is because such a difference in the type of data can be easily overcome by converting the data format in the receiving unit.

[Exchange] The exchange is an exchange suitable for the above-mentioned communication network. Further, this exchange is not limited to an exchange that directly operates a terminal (or a LAN connected to the terminal), but is also applicable to an exchange for transferring digital communication data from such an exchange. [Reception Unit] The reception unit may have a function of simply receiving digital communication data and transferring the digital communication data to the transfer information detecting unit, or dividing the received digital communication data into a plurality of segments. The first segment of the plurality of segments may have a function of storing the transfer information (corresponding to claim 5). For example, when a packet which is a variable length message is received, it is sent to an ATM.
It may be divided into cells.

In the latter case, the receiving section may attach the same identification information to each of the plurality of segments based on the same digital communication data (corresponding to claim 7). [Terminal not directly accommodated in communication network] This terminal is, for example, a terminal connected to a LAN (area information communication network) connected to the communication network (corresponding to claim 1). [Transfer Information] If only the address information corresponding to one piece of transfer information is stored in the address storage unit, the transfer information may be any information that can simply identify whether or not it is delivered to the above-mentioned terminal. In this case, for example, a 1-bit flag may be used. However, when address information corresponding to two or more pieces of transfer information is stored in the address storage unit, it is desirable that the transfer information can be identified from each other.

The transfer information may be written in a position where the incoming address is written in the digital communication data (corresponding to claim 11). This is because the digital communication data in which such transfer information is written does not require an incoming address having a specific terminal as an incoming terminal. [Transfer Information Detection Unit] When the transfer information detection unit detects the transfer information, the transfer information detection unit may notify the transfer control unit of the transfer information as it is, or the information indicating that the transfer information has been detected is converted into digital communication data. It may be written and transmitted to the transfer control unit.

When the receiving section divides the digital communication data into a plurality of segments, the transfer information detecting section reads the transfer information from the first segment. [Address Storage Unit] The address storage unit may store address information of a plurality of terminals corresponding to one piece of transfer information. Also, a plurality of such combinations may be stored. [Transfer Control Unit] When there is a plurality of address information corresponding to the transfer information detected by the transfer information detecting unit, the transfer control unit copies the digital communication data according to the number of the address information. The address data can be written one by one in each of the copied digital communication data (corresponding to claim 3).

To execute this copy, for example,
A copy buffer for storing the digital communication data received by the receiving unit may be further provided. With this configuration, when the transfer control unit copies the digital communication data, the digital communication data can be read from the copy buffer (corresponding to claim 4).

Further, the transfer control unit, when the transfer information detecting unit detects the transfer information from the first segment,
The address information of the terminal corresponding to the transfer information may be read from the address storage unit and the address information may be written in the first segment (corresponding to claim 6).

Further, when there is a plurality of address information corresponding to the transfer information detected by the transfer information detecting unit, the transfer control unit copies the head segment according to the number of this address information, and copies it. It is possible to write the address information one by one to each of the head segments and copy the other segments having the same identification information as the head segment by the same number as the number of copies of the head segment ( (Corresponding to claim 8).

Further, the transfer control unit is configured to have a counter for measuring the number of copied segments and a copy number table for storing the measured value of this counter, and this counter is provided when the copying of the first segment is completed. It can be configured to store the measured value of in the copy number table, and when copying other segments, copy until the measured value of this counter becomes the same as the value stored in the copy number table. Yes (corresponding to claim 9). In this case, if an error occurs while the transfer control unit is copying the segment, the measurement value of the counter up to that point may be stored in the copy number table (claim 10). Corresponding to). This is because it is possible to distribute at least the number of successfully copied digital communication data.

Further, when the transfer information detecting unit has an error in the segment, the transfer information detecting unit is configured to display an error in the segment, and the transfer control unit
When the segment is marked with the error, it is possible to set the minimum value of the counter (that is, "0") in the copy number table (corresponding to claim 12). By doing so, it is possible to prevent the subsequent copying of the segment based on the same digital communication data as this segment. This prevents omission of digital communication data.

Further, when the transfer control unit stores a value equal to or greater than the minimum value (that is, "0") of the counter in the copy number table at the time of receiving the head segment, it sets a dummy segment. The address information may be written in the leading segment after outputting the number of differences between the value of the copy number table and the minimum value of the counter (corresponding to claim 13). By doing so, the value of the copy number table can be returned to the minimum value without causing an error in the first segment. <Method of the Present Invention> A method of transferring digital communication data according to the present invention is a method of transferring the digital communication data in an exchange forming a communication network using the digital communication data, wherein the digital communication data is received and the communication is performed. Transfer information to the effect that it is distributed to a terminal that is not directly accommodated in the network is detected from the digital communication data received by the receiving unit, and is stored in advance as address information of the terminal corresponding to this transfer information. Address information is written in the digital communication data, and the digital communication data is output according to the address information (corresponding to claim 14).

[0023]

The digital communication data transmitted from any terminal to the communication network is received by the receiving unit (10) of the exchange (100).
1) is received. This digital communication data is
It is analyzed by the transfer information detection unit (102). That is,
If the digital communication data includes transfer information indicating that the transfer information is to be distributed to a terminal that is not directly accommodated in the communication network, the transfer information detection unit (102) detects this. The fact that this transfer information is detected is notified to the transfer control unit (103). Upon receiving this notification, the transfer control unit (103)
A combination of the transfer information stored in advance in the address storage unit and the corresponding address information of the terminal is read, and the address information corresponding to the notified transfer information is read. The transfer control unit (103) writes the read address information in the digital communication data. Therefore, according to the present invention, since the exchange (100) stores the incoming address for transferring the digital communication data to the terminal not directly accommodated in the communication network, the group address is also stored in such terminal. Data can be delivered (operation of claim 1).

Further, when there is a plurality of read address information, the transfer control unit (103) copies the digital communication data by the number corresponding to the address information, and adds the address information to each of the copied digital communication data. If it is configured to write one by one, it is possible to deliver the group address data to a plurality of terminals (operation of claim 3).

In this case, if the copy is executed by reading the digital communication data from the copy buffer, the process for the copied digital communication data can be performed serially (operation of claim 4). .

Further, if the digital communication data received by the receiving unit is divided into segments, the format of the digital communication data transmitted from the terminal is converted,
Higher-speed data transfer can be realized (operation of claim 5).

In this case, if the same identification information is attached to the segment based on the same digital communication data, the process for the first segment can be accompanied by the process for the other segment (claim 7).
Action). For example, when a plurality of head segments are copied, if the same number of segments having the same identification information as the head segment are copied, a plurality of pieces of digital communication data can be completely restored later. 8 action). In this case, if the counter and the copy number table are prepared to manage the copy number of the segment, the copy number will not be mistaken (acts to claim 9). Also, in this case, if an error occurs during copying of a segment, the measured value of the counter up to that point is stored in the copy number table and the copy number of the subsequent segment is managed. It is possible to prevent the output of useless segments that cannot be reproduced (operation of claim 10).

[0028]

An embodiment of the present invention will be described with reference to the drawings. In this embodiment, the digital exchange according to the present invention is applied to a B-ISDN network. <Structure of B-ISDN Network> First, FIG. 2 shows a schematic diagram of a B-ISDN network as a communication network to which an embodiment of the present invention is applied. This B-ISDN network transfers digital communication data to ATM (Asynchronous Transfer).
r Mode) is a public communication network that is segmented into a cell format according to r Mode) and is mixed with cells for other communication and transferred in the transmission path. In FIG. 2, B-ISDN
The network has a configuration in which n ATM exchanges (first ATM exchange 1a, second ATM exchange 1b, ..., Nth ATM exchange) are connected to each other. The connection between these ATM exchanges 1 is performed using the conversion unit 2b as an interface.

Further, the first ATM exchange 1a is connected to a LAN 9 as an individual regional information communication network with the conversion unit 2a as an interface. Similarly, other A
TM switch (second ATM switch 1b, ..., nATM
A LAN 9 is also connected to each of the exchanges 1n).

On the other hand, in each LAN 9, a gateway device (not shown) such as a router is directly connected to the ATM exchange 1, and this gateway device and the other terminals 10A to 10F.
And are connected to each other by a token ring or the like.
That is, these terminals 10A to 10F correspond to terminals that are not directly accommodated in the communication network.

Although not shown in FIG. 2, an independent terminal having an individual address is also directly connected to each exchange 1. <Cell Configuration> Next, FIG. 3 shows a cell format which is a segment of digital communication data transmitted and received between the exchanges 1. The format shown in FIG. 3 is CCI.
Since the format complies with the TT recommendation, in principle, the cell format outside the exchange 1 complies with this cell format. As is apparent from FIG. 3, each cell is composed of a 5-byte cell header part and a 48-byte payload part. The cell header part is a field indicating which communication the cell belongs to. The payload section is information that stores user information.

The first 4 bits of the first byte of the cell header portion are a general control flow portion (GFC). The latter 4 bits of the first byte and the first 4 bits of the second byte of the cell header portion are virtual path identifiers (VPI), and the second 4 bits of the second byte to the first 4 bits of the fourth byte are virtual channel identifiers (VCI). ). The virtual path identifier (VPI) and the virtual channel identifier (VCI) are routing bits that specify the transfer path of the cell, that is, the specific channel (VC) in the specific virtual path (VP). The 3rd bit from the 5th bit of the 4th byte of the cell header part is a payload type identifier (PTI). The payload type identifier (PTI) indicates the type of contents of the payload part. This type is user information, OAM information, resource information, or the like. The last bit of the fourth octet of the cell header part is a cell loss priority indication (CLP). The cell loss priority indicator (CLP) indicates whether the priority is low (“1”) or high (“0”). The fifth byte of the cell header section is a header error control section (HEC).

<Structure of ATM Switch> Next, the internal structure of the ATM switch 1 will be described in detail with reference to FIG. As shown in FIG. 4, each ATM exchange 1 has a plurality of A
A plurality of conversion units 2 each connected to a TM subscriber line φ1 or a plurality of ATM relay lines φ2, and an ATM cell connected to all the conversion units 2 via a highway (HW) φ3. The switch (SW) 3 and the intra-office server device 7 connected to the ATM cell switch (SW) 3
A central control unit (CC) 4 connected to all conversion units 2 and ATM cell switches 3 via a system bus φ4; and a central control unit (CC) 4 via an internal bus φ5. And a ROM (Read Only Memory) 5 and a RAM (Random Access Memory) 6 connected to each other.

[Central Control Unit (CC)] The central control unit (CC) 4 controls the entire exchange 1 in accordance with the control program stored in the ROM 5 and the data stored in the RAM 6. Specifically, the central control unit (CC) 4
Issues a conversion instruction of the data format to the conversion unit 2. Further, the central control unit (CC) 4 reads the VPI / VCI of the cell flowing into the ATM cell switch 3 and gives an instruction to switch the transfer path in the ATM cell switch 3.

[ATM Cell Switch (SW)] The ATM cell switch (SW) 1 has a network-like signal path and
It is composed of a plurality of cell multiplexing circuits that integrate these signal paths, buffers, cell distribution circuits, circuits such as selectors. Then, by controlling each circuit according to an instruction from the central control (CC) 4, each received cell is output via an appropriate transfer path.

[Conversion Unit] The conversion unit 2 as a reception unit time-division-multiplexes cells received from a plurality of ATM subscriber lines φ1 or ATM relay lines φ2 and places them on the highway (HW) φ3, and ) It has a function of dividing the cell received from φ3 and transmitting the divided signal to an appropriate signal line of the plurality of ATM subscriber lines φ1 or ATM relay line φ2. Here, the ATM subscriber line φ1 is a line connecting the exchange 1 and the subscriber terminal (including the gateway device of the LAN 9). The ATM trunk line φ2 is a line that connects the exchanges 1 to each other. In FIG. 3, the ATM switch 1 is provided with m conversion units 2a connected to the ATM subscriber line φ1 and n conversion units 2b connected to the ATM relay line φ1.

Each converter 2 also has a function of converting the data format. Specifically, ATM
The conversion unit 2a connected to the subscriber line φ1 exclusively uses the variable length message (packet) as digital communication data received from the subscriber terminal (including the gateway device of the LAN 9) in the ATM switch 1 as a segment. ATM cell switch (SW) 3 after converting to cells of the cell format (hereinafter referred to as "dedicated format")
Transfer to. Further, the conversion unit 2a converts the message of the dedicated format received from the ATM cell switch (SW) 3 into a variable-length message (packet) and sends it to the subscriber terminal (including the gateway device of the LAN 9).

On the other hand, the conversion unit 2b connected to the ATM trunk line φ2 receives the CCI received from another ATM exchange 1.
An ATM cell compliant with the TT recommendation is converted into a cell of a dedicated format and transferred to the ATM cell switch (SW) 3. Further, the conversion unit 2b uses the ATM cell switch (S
W) CCIT the cell of the dedicated format received from 3.
The cell format is converted to a cell format compliant with the T recommendation and sent to the transfer destination ATM switch 1.

Here, the conversion from the variable length message (packet) to the cell of the dedicated format, which is executed in the conversion unit 2a, will be described with reference to FIG. As shown in FIG. 5, a message (packet) received by the conversion unit 2a from a subscriber terminal (including the terminal 10 connected to the LAN 9).
Has a variable length of n bytes to m bytes. This message (packet) may constitute the completed information by itself, but may be a part of the completed information. The information contained in this packet is well known. That is, at least the destination address (DE)
And the source address (SA).

When such a variable length message (packet) is transmitted, the converting unit 2a divides it into 48 bytes, adds an ATM header of 5 bytes, and converts it into an ATM cell. However, the format of this ATM cell is a dedicated cell format (dedicated format) in the exchange 1, as described above. Therefore, it is slightly different from the cell format used for transmission and reception between the exchanges 1 shown in FIG. The dedicated format will be described below with reference to FIGS. 29 and 34. In addition,
The highway φ3 connecting the conversion unit 2 and the ATM cell switch (SW) 3 is a 21 parallel signal line. Then, one of them (MCLK) is allocated for clock transfer, one (XCFP) is allocated to the cell frame signal, one (XCEN) is allocated to the cell enable signal, and two (CEF1, CEF2). ) Is assigned to the error status display signal, and the remaining 16 lines (CD00 to
CD15) is assigned for cell data.

When converting a variable length message (packet) into an ATM cell, the converting unit 2a stores the beginning of the variable length message (packet) (the part in which the header part of the packet is stored) in the payload. The cell is the head cell (BOM), and the cell in which the portion after the head of the variable length message (packet) is stored in the payload is the intermediate cell (C
OM), and the cell storing the last part of the variable length message is the final cell (EOM). The conversion unit 2a writes the type (IST) of these cells in the ATM header of each cell.

The cell structure converted into this dedicated format is as shown in FIGS. 29 and 34. In each figure, from the beginning of cell data transmitted through 16 parallel signal lines to 03WORD (1WORD = 16 bits) is a cell header, and from 04WORD to 26WORD.
Up to is the payload.

Among the cell headers of the top cell (BOM) shown in FIG. 29, SST is the intra-device segment type. DM is a DA match signal. Also, GA
Is a group address signal for writing whether the incoming address (DA) includes a group address (GA = 1) or does not include it (GA = 0). IF is a line type signal. In addition, Link No. Is a physical link number, and an appropriate initial value is set by the conversion unit 2. PL Len. Is the payload length. TCB
Is a test cell bit indicating whether it is a test cell (TCB = 1) or a normal cell (TCB = 0). SN
Is a sequential number given to each MID. The MID is a message identifier serving as identification information, and the same value is set in all cells forming the same variable length message.

The variable length message (packet) header is stored in the payload of the head cell (BOM). That is, ISSIP DA is the destination address of the variable length message (packet), and SA is the source address of the variable length message. Also, ISSI
BEtag is identification information of BOM and EOM that are CS-PDUs. ISSIBAsize is a value indicating the message length. In addition, SVTYPE is for each Subic
This is a signal indicating the type of e Entity. Also, He
ader Extension is SIP Head
er Extension This is a field that guarantees consistency with four. IITYPE is a Packet
It is the display information of the input path of. In addition, ES is Souc
e User is information indicating whether or not the IC has been selected in the L3-PDU. In HCIND, Packet is S
This is information for detecting an infinite loop between S. In addition, the Carrier is a destination address (D
This is information indicating the relay network when A) is another network. Also, the INNET ID is
This is the identification information of the Network to which the Packet has been transferred. In addition, INICI TPSET is a Pack
This is the identification information of the ICI TPS to which et has been transferred. Further, If is a line type signal. Also, CRC
Is a code for error detection and correction.

The contents of the data portion of the variable length message (packet) are stored in the payload of the intermediate cell (COM) or the final cell (EOM) shown in FIG. When the cells of the dedicated format are transmitted from the ATM cell switch 3, the conversion unit 2a extracts the contents of the payload of these cells and restores them into a variable length message. On the other hand, the conversion unit 2b also receives CCITT from the other exchange 1.
When a cell having a format conforming to the above is received, the format of the received cell is converted into the dedicated format as shown in FIG. 29 or 34.

When any cell is transmitted to the ATM cell switch 3, each of the conversion sections 2 sends routing information so that the cell is transferred in the ATM cell switch 3 toward the in-office server device 7. Set. [In-office server device, inter-office server device] In-office server device 7
When the received cell is data having a group address (GA) as transfer information, the inter-office server apparatus 8 duplicates the received cell and assigns an individual address corresponding to the group address to each duplicated cell. It has a function to attach. The group address is an address collectively assigned to a plurality of terminals, and is an address represented by the upper 5 bits of the incoming address (DA). This group address (GA) is registered in advance in either the intra-station server device 7 or the inter-station server device 8 of any of the exchanges 1 as corresponding to the individual address as the address information of the plurality of terminals. Then, the server devices 7 and 8 copy and transfer only to the cell having the same group address as the group address registered in itself.

The roles of these server devices 7 and 8 are the subscriber terminals that accommodate the copied cells in their own exchange 1 (including the terminal 10 connected to the LAN 9 accommodated in their own exchange 1). It is divided depending on whether it is directly delivered to the switch or transferred to another exchange 1. That is, in the intra-office server 7, group addresses corresponding to individual addresses of a plurality of subscriber terminals (including the terminal 10 connected to the LAN) accommodated in the own exchange 1 are registered. On the other hand, in the inter-office server 8, group addresses corresponding to individual addresses of a plurality of subscriber terminals (including the terminal 10 connected to the LAN) accommodated in the other exchange 1 are registered.

The individual address corresponding to the group address registered in the inter-office server device 8 of a certain exchange includes the group address registered in the intra-office server device 7 of another exchange 1. There are also cases. Similarly, in the case where the group address registered in the inter-office server device 8 of the same exchange 1 is included in the individual address corresponding to the group address registered in the intra-office server device 7 of a certain exchange. is there.

Further, as described above, the cell is the conversion unit 2
When transmitted from the ATM cell switch 3 to the ATM cell switch 3, these cells are first input to the in-office server device 7. Then, this cell (and the copied cell when copying is performed in the in-office server device 7) is sent to the ATM cell switch 3. Since these cells are provided with routing information having the content of being transferred to the interoffice server device 8, all of them are transferred to the interoffice server device 8. The inter-office server device 8 transfers the transferred cell (and the copied cell when copying is performed in the intra-office server device 7).
Is sent to the ATM cell switch 3. Since routing information according to each individual address is added to these cells, conversion to connect to the called terminal (another switch connected to the called terminal or another switch corresponding to the group address) indicated by each individual address Forwarded to part 2.

The only difference between the two server devices 7 and 8 in structure is that the group addresses registered therein are different, and the destinations of the cells are also different. The other structures are exactly the same. Therefore, the configuration will be described below by taking the interoffice server device 7 as an example.

As shown in FIG. 6, in the inter-station server device 7, there are n connectionless (CL) connections connected in parallel.
The server 8a is built in. In each CL server 8a,
Different destination addresses (DA) are registered. That is, each CL server 8a is divided into roles so as to process only the cell having the incoming address (DA) registered in each CL server 8a. And each CL server 8
a is the incoming address (DA) of the input cell at the entrance
Is checked, and only cells that match the registered destination address (DA) are passed.

In FIG. 6, processed cells are output from each CL server 8a via any physical link. This physical link shows the mouth when the cell actually exits the transmission path. A bundle of these physical links is a transfer route.

Each CL server 8a has an internal structure as shown in FIG. That is, the DA analysis unit 11, the carrier (Carrier) detection unit 12, the write control unit 13, the buffer unit 14, the read control unit 15, and the transfer control unit 17 are interconnected in order from the cell entrance. Further, the DA analysis unit 11 includes a DA / GA storage RAM.
25 are connected. A carrier valid / invalid storage RAM 18 is connected to the carrier detection unit 12. A write ok information storage RAM 19 is connected to the write control unit 13. A copy buffer 16 is connected to the read control unit 15 and the transfer control unit 17. Further, the transfer control unit 17 includes a GA copy number storage RAM 21, a GA member start address storage RAM 22, a GA member IA etc. storage RAM 23,
And a RAM 24 for storing Link no etc. are connected. The function of each component will be described below.

The CL is stored in the DA / GA storage RAM 25.
An incoming address (DA) of a cell scheduled to be processed by the server 8a and a group address (GA) as transfer information are registered. Also, this DA ・ G
Each group address (GA) registered in the A storage RAM 25 has a member start table (FIG. 25) for reading an entry corresponding to each group address.
The information (RGA) for referencing (reference) is also associated.

DA analysis section 11 as transfer information detection section
Refers to the DA / GA storage RAM 25 to determine whether the input cell is valid / invalid, whether the destination address (DA) of this cell is registered in advance, and the destination address (DA).
It is checked whether the upper 5 bits in the group are registered in advance as a group address (GA).

In the carrier valid / invalid storage RAM 18,
The carrier table shown in FIG. 21 and the TP shown in FIG.
The S table name is stored. This carrier table is a table in which carrier information (Carrier) information stored in the payload of the first cell (BOM) is listed in advance. This carrier information (Carrier)
This is information required when transferring the cell copied in the CL server 8a to another network. However, this carrier table contains carrier information (Carrier) that can be supported by the CL server 8a.
Only registered. As shown in FIG. 21, each carrier table entry is assigned an address value. This address value is used when searching the TPS table.

This TPS table includes the address value described in the carrier table and the transfer route number (TPS N
o. ), And support information (S) indicating whether or not the transfer route is supported.

The carrier detection unit 12 uses the top cell (BO
The content of the carrier field of M) is extracted, the carrier table in the carrier valid / invalid storage RAM 18 is searched, and it is detected whether the same carrier is registered in advance. Then, the carrier detection unit 12
When there is a corresponding carrier, reads out the address value corresponding to the record on the carrier table in which this carrier is described. Then, the TPS table is searched based on the read address value, the transfer route number (TPS No.) necessary for the group address transfer control (the transfer control of the cell based on the group address) is read, and this is written in the cell. . In addition, the carrier detection unit,
RGA corresponding to the cell group address (GA)
The data is read from the DA / GA storage RAM 25 and written in the cell.

In the write-ok information storage RAM 19, FIG.
The write ok table shown in FIG. 3 is stored. This write ok table has 6 bits of VCI and 1 of the input cell.
Whether the buffer memory write succeeds (=
1) This is a table for writing whether or not (= 0). In the following description, the record (OK0) in which the information about the normal cell is written is referred to as the write ok table # 0, and the record (OK1) in which the information about the test cell is written is referred to as the write ok table # 1.

The write control unit 13 includes a buffer unit 14
You will be informed if the is full or free. Then, according to the state of the buffer unit 14, the DA analysis unit 11
Control is performed to write the cell received from the buffer unit 14 into the buffer unit 14. The write control unit records whether or not the BOM has been written successfully in the write ok table in the write ok information storage RAM 19, and the COM and EOM.
With respect to, the writing to the buffer unit 14 is performed only when it is recorded in the write ok table that the writing of the BOM is successful.

The buffer section 14 is a buffer memory. The buffer unit 14 notifies the write control unit 13 whether or not the buffer is full, and notifies the read control unit 15 whether or not the buffer is empty.

The read control unit 15 reads out cells from the buffer unit 14 and sends them to the transfer control unit 17 and writes them to the copy buffer 16 unless the buffer unit 14 notifies that the buffer is empty. Take control.

The copy buffer 16 is a buffer having a capacity for writing one cell. Therefore, the copy buffer 16 updates the cell written by overwriting each time the cell is written from the read control unit 15. Since the copy buffer 16 holds the same cell until overwritten, the transfer control unit 1
It is possible to output the same cell to the transfer control unit 17 a plurality of times in response to a request from 7. This allows the cells to be copied.

The RAM 21 for storing the GA copy number is shown in FIG.
The copy number table shown in FIG. This copy number table is a table for writing the number of times of copying for a normal cell or a test cell, using the 6-bit VCI of the input cell and the 10-bit MID as an address.

RAM for storing GA member start address
22 stores the member start table shown in FIG. This member start table is a table in which the carrier detection unit 12 uses the 10-bit RGA written in the cell at the time of writing to the buffer unit 13 as an address and associates the start address of the member IA / TPS table (see FIG. 26). is there.

The member member IA / TP shown in FIG. 26 is stored in the RAM 23 for storing the GA member IA etc. as an address storage unit.
The S table is stored. This member IA / TPS
The table is a start address read from the member start table or a pointer (pointer) described later.
Using the value as an address, an individual address (mem.I.
A), member ID (mem.ID), LT bit (L
T), transfer route number (TPS No.), pointer (p
This is a table that stores a combination of C. Here, the individual address (mem.
IA) is the address of each individual terminal. The member ID is a code for identifying each individual terminal.
The LT bit indicates whether the member IA is in the service area of the network (LT = 0, hereinafter, this case is referred to as “inside LATA”) or not (LT = 1, below, in this case, outside “LATA”). ""). The TPS number is the number of the transfer route to each terminal indicated by the member IA. The pointer is a field indicating the start address of the above-mentioned data for other individual terminals forming the group. That is, the data storage position of the terminal to be read next time is shown. This pointer is regularly set so that data is read in order from a terminal having a small member ID to a terminal having a large member ID. The C bit is a bit indicating whether the BOM copy is continued (C = 1) or completed (C = 0). Therefore, the value of the pointer is not set in the data of the individual terminal set to C = 0.

Link no. The storage RAM 24, etc.,
The link number table shown in FIG. 27 and the ICI / ISSI valid / invalid table shown in FIG. 28 are stored. This link number table uses the 7-bit member ID read from the member IA / TPS table and the 5-bit TPS number (or the 5-bit TPS number written in the cell) as an address and the physical link of the corresponding physical link. Link number (Link No.) and whether or not the physical link is supported (S = 1) (S =)
It is a table that stores the information of 0). Also, ICI
In the / ISSI valid / invalid table, a 3-bit link number read from the link number table is used as an address, and a physical link corresponding to the link number is physically set (A = 1, that is, an intercarrier interface, and It is a table showing whether or not the inter-switching system interface is available) (A = 0, inter-carrier interface, and inter-switching system interface not available).

When the cell received from the read control unit 15 is a BOM, the transfer control unit 17 determines from the RGA written in this cell whether or not this cell has a group address. If it has a group address, RGA from the member start table
The start address corresponding to is read and the member IA / TPS table is referred to based on this start address. Then, the same cell is read from the copy buffer 16 until the C bit becomes 1, and each read cell is given an individual address (mem.IA). Further, the transfer control unit 17 reads the member ID and TPS number (only in LATA) from the member IA / TPS table, and the member ID and TPS number (when outside LATA, the TPS number attached to the cell). ), The link number table is referred to and the corresponding link number is read. The transfer control unit 17 attaches the read link number to each cell. Further, the transfer control unit 17 refers to the ICI / ISSI valid / invalid table based on the read link number to check whether or not the link is physically set, and only when it is set, the cell is set. Is output to the outside.

The transfer control section 17 has a built-in copy counter 17a as a counter. The copy counter 17a counts the number of times the same cell is read from the buffer unit 14 or the copy buffer 16. The count value of the copy counter 17a is added to each of the read cells, and is written in the copy number table when the copy of the BOM (reading from the copy buffer 16) is completed.

When the cell received from the read control unit 15 is COM or EOM, the transfer control unit 17 determines the VCI and MID of the cell (these are all cells (BOM, COM, EOM) based on the same variable length message). Is common to) and reads the copy number table with the key. And
The transfer control unit 17 reads the cell from the copy buffer 16 and outputs the cell to the outside until the number of copies recorded for the BOM corresponding to the cell is reached. <Contents of processing in connectionless server> Next,
The contents of the processing executed in the connectionless server 8a will be described based on the flowchart. In the connectionless server 8a, the process from receiving a cell from the outside to writing the cell in the buffer unit 14 and the process from reading the cell from the buffer unit 14 to outputting the cell to the rear part are asynchronous in parallel. To be executed. Therefore, in the following description, each process will be described. [Processing from Reception of Cell to Writing to Buffer Unit] First, in order to write a cell to the buffer unit D
The processing executed by the A analysis unit 11 and the carrier detection unit 12 will be described based on the flowcharts of FIGS. 8 to 10.

In the processing of FIG. 8 which is the main routine of this processing, the ATM cell (BOM, COM,
It starts by inputting EOM, etc.).
Then, in the first step S0001, it is checked whether the input ATM cell is valid or invalid. If it is invalid (for example, if the cell length is out of the standard, etc.), it waits until a valid ATM cell is input.

When it is determined in step S0001 that the ATM cell is valid, it is determined in subsequent step S0002 whether the incoming address (DA) of the cell is valid or invalid. That is, if the cell is a BOM, as shown in FIG.
Incoming address (DA) of variable length message (packet)
Is included. In step S0002, the case where the incoming address (DA) is registered in the DA / GA storage RAM 25 of the CL server 8a in advance is treated as "valid". Therefore, if the incoming address (DA) is not registered in advance for a cell other than the BOM or for the BOM, it is determined to be “invalid”.

In step S0002, the incoming address (D
If A) is valid, step S000
At 3, the permission signal is output. This permission signal is a group address (GA) within the incoming address (DA).
Is a signal indicating whether or not the code indicating is GA.

When the processing in step S0003 is completed, and in step S0002, the destination address (DA)
In any case where is determined to be invalid, carrier detection processing is performed in subsequent step S0004.

FIG. 9 is a subroutine of the carrier detection process executed in step S0004 of FIG.
In the first step S0101 after entering this subroutine, it is checked by the segment type field (IST) included in the cell header of the input cell whether or not the cell is a BOM. If it is not the BOM, this subroutine is finished as it is, and the process returns to the process of FIG. Therefore, in that case, the carrier detection process is not performed.

On the other hand, if it is determined in step S0101 that it is a BOM, in step S0102.
It is checked whether the upper 5 bits of the incoming address (DA) included in the cell match the group address (GA) registered in the DA / GA storage RAM 25 in advance. If both match, the destination address of the cell (D
Group address transfer instruction by A) and CL server 8a
It is determined that the state of the GA registered in the group indicates the group address transfer control (which means the broadcast distribution control according to the group address. The same applies to the following), and the process is performed in step S0103. Proceed to. On the other hand, when the two do not match, it is determined that the individual address transfer control (which is the normal distribution control according to the individual address. The same applies hereinafter) is instructed, and the process is performed in step S01.
Proceed to 08.

When the group address transfer control is selected, GA = 1 is set in the group address field (GA) included in the cell header in step S0103. In the following step S0104, it is determined that DA / G corresponds to the group address (GA) of the cell.
The RGA value registered in the A table 25 is set as the RGA field in the cell header part of the cell (see FIG. 30).

In the next step S0105, it is checked whether the carrier is valid. That is, the carrier information (Carrier) included in the payload of the cell is read and it is checked whether or not this is registered in the carrier table of the carrier valid / invalid storage RAM 18. Then, the same carrier information (Carri
er) is registered, it is determined that the carrier is valid, and the process proceeds to step S0106.

In step S0106, the address of the entry in which the carrier information (Carrier) is written on the carrier table is read. Then, based on this address, the carrier valid / invalid storage RAM
The TPS table 18 is searched, and the corresponding transfer route number (TPS no.) Is read. Then, the read transfer route number (TPS no.) Is set as an in-cell TPS number field in the cell header portion of the cell (see FIG. 30). Furthermore, in the cell header of the cell,
TPS support information (SS = 1) indicating that TPS is supported is set (see FIG. 30). Upon completion of step S0106, the subroutine of FIG. 9 is terminated and the process returns to the main routine of FIG. As a result of executing the above processing, the content of the cell (BOM) is as shown in FIG.
It is rewritten as shown in 0.

On the other hand, if it is determined in step S0105 that the same carrier information (Carrier) is not registered, it is determined that the carrier is invalid, and the process proceeds to step S0107. In step S0107, TPS support information (S = 0) indicating that TPS is not supported is set in the cell header of the cell (see FIG. 30). Upon completion of step S0105, the subroutine of FIG. 9 is terminated and the process returns to the main routine of FIG.

In step S0108, which is executed when it is determined in step S0102 that individual address transfer control is instructed, the state of the group address field (GA) included in the cell header of the cell is checked. .

This group address field (GA)
Is set to GA = 1 for the first time in step S0103, so G is set at step S0109.
It is unlikely that A = 1 is set. for that reason,
If the group address field (GA) is GA = 1, an error flag (EFGB = 0) is set in step S0109. As is apparent from FIG. 30, this error flag (EFGB) indicates that the signal line WEF1
Set on. Then, in step S0110, the group address field (GA) of the cell.
Is set to disable (GA = 0). Upon completion of step S0105, the subroutine of FIG. 9 is terminated and the process returns to the main routine of FIG. On the other hand, when the group address field (GA) is GA = 0 in step S0108, this subroutine is finished as it is, and the process is returned to the main routine of FIG. Therefore, also in these cases, the carrier detection process is not performed.

When the process returns from the subroutine of FIG. 9, the process proceeds to step S0005 in the main routine of FIG. In step S0005, a buffer write control process subroutine shown in FIG. 10 is executed.

In the first step S0201 after entering the subroutine of FIG. 10, it is checked whether or not the relevant cell is a test cell. That is, whether the state of the test cell bit (TCB) included in the cell header of the cell indicates a test cell (TCB = 1) or a normal cell (TCB).
= 0) is checked.

If the test cell is indicated, in step S0203, the write ok used for the subsequent processing is set.
When the write ok table # 1 is selected as the table and indicates a normal cell, in step S0202, the write ok table is used as the write ok table used in the subsequent processing.
Select table # 2.

In the next step S0204, the capacity of the buffer section 14 is monitored. Then, when it is determined that the buffer unit 14 is full and is in the unwritable state, the process proceeds to step S0208. In this step S0208, the corresponding entry (6 bit VCI and 10 bit MID of the cell) of the write ok table is set so that the subsequent cells (cells based on the same variable length message) of the same MID combination cannot be written. The unwritable information (= 0) is set in the entry corresponding to the address obtained by combining and. Then, in subsequent step S0210, the cell is discarded and a discard signal is output.

On the other hand, if it is determined in step S0204 that there is a vacancy of one cell or more in the buffer section 14, it is checked in step S0205 whether the cell is the head cell (BOM). And B
If it is OM, in step S0206,
The write permission information (= 1) is set in the corresponding entry of the write ok table (the entry corresponding to the address that combines the 6-bit VCI of the cell and the 10-bit MID). Then, in the following step S0207,
The cell is written in the buffer unit 14.

If it is determined in step S0205 that the cell is other than the BOM, the state of the write ok table is monitored in step S0209. That is, the state of the entry corresponding to the combined address of the 6-bit VCI and the 10-bit MID of the cell is checked. When the unwritable information (= 0) is set in this entry, the BOM having the same MID as the cell is still in the buffer unit 14.
It is determined that it has not been written in step S0210.
At, the cell is discarded and a discard signal is output. On the other hand, when the write permission information (= 1) is set in this entry, it is determined that the BOM having the same MID as the cell is already written in the buffer unit 14, and in step S0207, The cell is also written in the buffer unit 14.

As described above, the discard signal is output (step S0210) or the buffer is written (step S0).
When (207) is completed, the subroutine of FIG. 10 is terminated and the process returns to the main routine of FIG.

In the main routine of FIG. 8 where the processing is returned,
To wait for the next ATM cell, the process proceeds to step S0001.
Return to. [Processing from Reading Cell from Buffer Unit 14 to Outputting to Outside] Next, the cell is buffer unit 14
Read-out control unit 1 for reading out from and outputting to the outside
5 and the processing executed by the transfer control unit 17 will be described based on the flowcharts of FIGS. 11 to 20.

11 and 12, which are the main routines of this processing, start at system startup.
Then, in the first step S0301, the subroutine of buffer read control shown in FIG. 13 is started.

In the first step S0401 after entering the subroutine of FIG. 13, it is checked whether or not the buffer unit 14 is empty. If it is empty, that is, if the cell to be read does not exist in the buffer section 14, it is determined that the cell cannot be read and the cell is read (step S0).
This subroutine is ended as it is without performing 402). In that case, step S0310 is executed in the main routine to which the process is returned. That is, the transfer cell control unit 17 transfers the cell as an invalid cell (a signal including only the cell frame signal), and the cell is read again (steps S0309 and S030).
1).

On the other hand, if there is a cell in the buffer unit 14 in step S0401 of FIG. 13 and the cell is in a readable state, one cell is read from the buffer unit 14 in step S0402. The reading order of the cells is first-in first-out. This read cell (BO
The state M) is shown in FIG.

When the data is read from the buffer unit 14, the cell is transferred to the transfer cell control unit 17 and is also written in the copy buffer 16 (step S0403). In the following description, unless otherwise specified that the data is read from the copy buffer 16, the relevant cell means a cell directly transferred from the read control unit 15.

When the process returns to the main routine after the reading from the buffer unit 14 (step S0402), step S0302 is executed. In step S0302, the copy number counter 17a for registering the cumulative value of the copy number is initialized. In the next step S0303, the setting processing subroutine shown in FIG. 14 is executed. In the first step S0501 after entering this subroutine, the value of the copy number counter 17a is checked. If the value of the copy number counter 17a is "0", it means that the copy is not yet made, and therefore the process proceeds directly to step S0503. On the other hand, when the value of the copy number counter 17a is 1 or more, copying is in progress. Therefore, in step S0502, masking processing is performed to prevent error cells, and the process proceeds to step S0503.

In step S0503, it is checked whether the cell is a test cell based on the test cell bit (TCB) in the cell header section. If the cell is a test cell, the copy number table # 1 is selected as the copy number table used in the subsequent processing in step S0505. On the other hand, if the cell is a normal cell, in step S0504, the copy number table # 0 is selected as the copy number table used in the subsequent processing. When the above processing is completed, the processing is returned to the main routine of FIG.

When the process returns from the subroutine of FIG. 14,
In the main routine, step S0304 is executed.
In this step S0304, the segment type (ST) included in the cell header part of the cell is the head cell (BO
M) is checked. If the cell other than the head cell (BOM) is indicated, the process is performed in step S03.
Proceed to 17.

On the other hand, if it is determined in step S0304 that the cell is the head cell (BOM), the value in the copy number table is checked in step S0305. That is, 1 written in the cell header of the cell
The value written in the corresponding entry on the copy number table is read using the 0-bit MID value and the 6-bit VCI value as addresses. Then, if the read value is not "0", the process proceeds to step S0313. Unless copying is completed for the cell (BOM), the value of the copy number table remains “0” which is the initial value.

On the other hand, when the value read from the copy number table in step S0305 is "0", processing is performed in the next step S0306 to determine whether it is an error cell. That is, the state of the signal transmitted from the signal line REF1 or REF2 is checked to detect whether any error bit indicates an error (= 0). If it is determined that the cell is an error cell, the process proceeds to step S0321.

On the other hand, if it is determined in step S0306 that the cell is not an error cell, then in the next step S0307, the group address field (GA) of the cell concerned.
Check the status of. If it is determined that the address is not a group address (GA = 0), the process advances to step S0323.

On the other hand, if it is determined in step S0307 that the address is a group address (GA = 1), the process proceeds to step S0308. Step S030 above
The processing contents for each case identified by the checks from 4 to step S0307 will be described below. << First Cell (BOM), Copy Number Table = 0, No Error, GA = 1 >> In this case, step S03
07 to step S0308, and FIG.
The BOM copy processing subroutine shown in is executed. In the first step S0601 after entering this subroutine,
The next process is selected depending on whether the copy (beginning) of the cell (BOM) has been started or is being copied. If this subroutine is first entered for the cell, it means that copying has not yet started, and the process advances to step S0602.

In step S0602, it is checked whether or not there is an address in the member start table (FIG. 25). That is, it is checked whether or not the member start address value is described in the entry on the member start table corresponding to the RGA value stored in the cell header portion of the cell. Then, if there is no member start address value, the process proceeds to step S0615.

On the other hand, if it is determined in step S0602 that the member start address value is described, then in step S0603, the member start address value read from the member start table is used as the address, and the member IA / TPS table (see FIG. Read 26). That is, the individual address (memIA), member ID, LT value, transfer route number (TPS), C bit, and pointer value are read from the entry corresponding to this address value. After that, the process advances to step S0604.

On the other hand, if the subroutine is entered for the cell for the second time or later, it is determined in step S0601 that copying is in progress, and the process advances to step S0611. In this step S0611,
At the time of processing the subroutine in the immediately preceding round, step S
0603 or the pointer value read in step S0611 as an address, the member IA / TPS table (see FIG. 2).
Read 6). That is, from the entry corresponding to this pointer value, the individual address (memIA), the member ID,
The LT value, transfer route number (TPS), C bit, and pointer value are read.

In the following step S0612, the member ID (new me) read in step S0611 of this time.
m. ID) is the previous step S0603 or step S
Member ID read in 0611 (current mem.ID)
Check if greater than. And new me
m. ID is the current mem. If not greater than ID,
This subroutine is terminated assuming that error processing should be executed in the main routine. On the other hand, new me
m. ID is the current mem. If larger than ID, the process advances to step S0604.

In step S0604, the field (DA) of the incoming address stored in the payload portion of the cell is set in step S0603 or step S061.
The individual address (memIA) read in 1 is overwritten.

In the next step S0605, step S
Based on the LT value read in 0603 or step S0611, it is checked whether the individual address (memIA) indicates the inside of the service area (inside LATA) of the network 1 or the outside of the service area (outside LATA). If it is in LATA, the transfer route number (TPS) read from the member IA / TPS table in step S0603 or step S0611 is registered in step S0606, and the process proceeds to step S0607.

On the other hand, if it is outside LATA, in step S0613 it is indicated whether the TPS support information (S) written in the cell header portion of the cell is supported (S = 1) or not (S = 1). Check S = 0). And if it is not supported (S = 0)
If so, the process proceeds to step S0615. On the other hand, if it is supported (S = 1), step S061
4, the transfer route number (TPS) written in the cell header portion of the cell is registered, and the process proceeds to step S0607.

In this step S0607, step S
0603 or member IA / TP in step S0611
The member ID read from the S table is registered.
In the next step S0608, the member ID registered in step S0607 and the transfer route number (T) registered in step S0606 or step S0614.
Link number table (Fig. 2)
7) is searched, and the support information (S) and the link number described in the corresponding entry are read. Then, it is checked whether this support information (S) indicates support of the link number (S = 1) or not (S = 0). And if the link number is not supported (S = 0)
If so, the process proceeds to step S0615. On the other hand, if it is supported (S = 1), the processing is performed in step S.
Proceed to 0609.

In this step S0609, step S
The link number read in 0608 is written in the cell header portion of the cell (see FIG. 32). Subsequent step S0
In 610, the individual address (memIA) given to the cell header of the cell in step S0604 and the source address (S) stored in the payload of the cell (S).
Compare with A). This is to prevent the cell from returning to the source. Then, the individual address (memI
If A) and the source address (SA) match, the process advances to step S0615.

On the other hand, if the individual address (memIA) and the source address (SA) do not match, the process proceeds to step S0616. This step S061
In step 6, the ICI / ISSI valid / invalid table (FIG. 28) is searched using the physical link number read in step S0608 as an address, and the valid / invalid information (A) described in the corresponding entry is read. And I
If CI / ISSI is valid (A = 1), the process proceeds directly to step S0618. On the other hand, ICI
If / ISSI is invalid (A = 0), an error cell is created in step S617. That is, the signal line O
Set the EFIU flag on EF2 to the error state (= 0). Thereafter, the process proceeds to step S0618.

In a succeeding step S0618, the value of the copy counter 17a is written in the cell header portion of the cell as a copy number field (COPY CONT.) (See FIG. 32). Then, in subsequent step S0619, the cell is output from the inter-station server device 8 or the intra-station server device 7 via the physical link and the transfer route corresponding to the physical link number written in step S0609. The contents of the cell (BOM) at this point are as shown in FIG.

In the next step S0620, the value of the copy counter 17a is incremented by "1". afterwards,
The process advances to step S0621. On the other hand, when the process proceeds to step S0615,
The invalid cell transfer control having the same content as that of step S0310 is executed, and the process proceeds to step S0621.

In step S0621, step S06 is executed.
03 or step S0611, the pointer (Pointer) and the C bit read from the member IA / TPS table are registered. However, when the process proceeds from step S0602 through step S0615 to step S0621, initial values (Pointer = NUL, C = 0) are set in the register for registering the pointer value and the register indicating cell copy continuation / end. Will be registered as is.

In the next step S0622, step S
At 0621, it is checked whether the C bit registered in the cell copy continuation / end instruction register indicates copy continuation (C = 1) or copy end (C = 0). Then, when the end of copying (C = 0) is indicated, the process proceeds to step S0623. Step S0623
Then, the value of the copy counter 17a is written in the copy number table selected in step S0504 or step S0505. Then, in the next step S0624, "0" is written in the member ID register in which the member ID is registered in step S0607. Step S
When 0624 is completed, the process returns to the main routine of FIG. 11 to read the next cell (step S030
9), Step S0301 is re-executed.

On the other hand, when it is determined in step S0622 that the copying is continued (C = 1),
The process is returned to the main routine as it is, and the same cell is read (step S0315). That is, in the next step S0316, the contents of the copy buffer 16 are read out, and the process is returned to step S0303.

On the other hand, if this subroutine ends as it is from step S0612, the process returns to the main routine of FIG. 11 and the error cell transfer control of step S0311 is executed.

FIG. 17 shows a subroutine of this error cell transfer control. In the first step S0701 after entering this subroutine, the relevant cell is converted into an error cell. That is, the EFGA flag on the signal line OEF1 is set to the error state (= 0). In the next step S0702, the value of the copy counter 17a is written in the cell header portion of the cell as a copy number field (COPY CONT.). In the next step S0703, the cell is output from the inter-station server device 8 or the intra-station server device 7. After that, the process is returned to the main routine of FIG.

In the main routine after the processing returns, in the next step S0312, the value of the copy counter 17a is written in the copy number table selected in step S0504 or step S0505. Then, in order to read the next cell (step S0309), step S
0301 is re-executed. In this case, what is written in the copy number table is the number of copies that were successful before the error occurred. Therefore, subsequent cells (CO
M, EOM) is copied by this number of copies. << First Cell (BOM), Copy Number Table = 0, No Error, GA = 0 >> As a premise, the RGA field is not set and the TPS field is also set in the cell header part of the cell in this case. Not not.
Therefore, the initial value (LinkNo.) Of the link number is written with the initial configuration of the input cell (BOM) shown in FIG.

In this case, the process proceeds from step S0307 to step S0323, and the BOM individual address processing subroutine shown in FIG. 18 is executed. In the first step S0801 after entering this subroutine, the cell (BO
The link number initially set in the cell header part of M) is registered.

At the following step S0802, step S
Using the link number registered in 0801 as an address, the ICI / ISSI valid / invalid table (FIG. 28) is searched and the valid / invalid information (A) described in the corresponding entry is read. If ICI / ISSI is valid (A = 1), the process proceeds directly to step S0804. On the other hand, if the ICI / ISSI is invalid (A = 0), an error cell is created in step S0803. That is, the EF on the signal line OEF2
Set the IU flag to error state (= 0). afterwards,
The process advances to step S0804.

In step S0804, the value of the copy counter 17a is written in the cell header portion of the cell as the copy number field (COPY CONT.) (FIG. 3).
3). Then, in the following step S0805,
The cell is output from the inter-station server device 8 or the intra-station server device 7 via the physical link and the transfer route corresponding to the physical link number written in the cell header.
The contents of the cell (BOM) at this point are as shown in FIG.

At the next step S0806, the value of the copy counter 8a is incremented by "1". After that, the process advances to step S0807. This step S08
In 07, step S0504 or step S0505
In the copy number table selected in
Write the value of a. That is, this copy number table
A value of "1" will be written. Therefore, the subsequent cells (COM, EOM) are output only by "1".
After that, the process returns to the main routine of FIG. 12, and step S0301 is re-executed for the next cell read (step S0309). << First Cell (BOM), Copy Number Table = 0, Error Cell >> In this case, the process proceeds from step S0306 to step S0321, and the error transfer control processing subroutine shown in FIG. 17 is executed. This subroutine has the same content as that executed in step S0311, and therefore its explanation is omitted. However, in this case, in the following step S0322, step S05
04 or "0" is written in the copy number table selected in step S0505. Therefore, in this case, subsequent cells (COM, EOM) are not output. afterwards,
Step S0301 is re-executed for the next cell read (step S0309). << When top cell (BOM) and copy number table> 0 >>
As described above, the value of the copy number table should remain “0” as long as the copy of the head cell (BOM) is not completed, that is, the reading process of cells other than the head cell (BOM) is not started. Is. Therefore, at the time when the head cell (BOM) is still the processing target, the value in the copy number table cannot be other than "0", and it is considered that some error condition has occurred.

Therefore, in this case, the process proceeds to step S.
It advances from 0305 to step S0313. In this step S0313, this error state is masked. In the following step S0314, the dummy cell transfer control processing subroutine shown in FIG. 19 is executed. This subroutine is a process executed to return the value of the copy number table that was the check target in step S0305 to "0".

In the first step S0901 after entering this subroutine, the relevant cell is once made into an error cell. That is, the EFMS flag on the signal line OEF1 is set to the error state (= 0).

In the next step S0902, the value of the copy counter 17a is written in the cell header portion of the error cell as a copy number field (COPY CONT.). Then, in subsequent step S0903, the error cell is output from the inter-station server device 8 or the intra-station server device 7 via the physical link and the transfer route corresponding to the physical link number written in the cell header. Since this error cell is set to the error state, it is eventually discarded even if it is transmitted.

In the next step S0904, the value of the copy counter 17a is incremented by "1". afterwards,
The process advances to step S0905. This step S0
In 905, it is checked whether the value of the copy counter 8a after the increment matches the value of the copy number table which was the comparison target in step S0305.

If the two values do not match,
The process directly returns to the main routine of FIG. In this case, in steps S0315 and S0316, the same cell is read from the copy buffer 16, the process proceeds from step S0305 to step S0314, and the subroutine is entered again.

As a result of repeating such processing, when it is determined in step S0905 that the two values match, "0" is written in the corresponding entry of the copy number table in step S0906, and step S090
7, the copy number counter 17a is set to "0". Then, the process returns to the main routine. Also in this case, the same cell is read from the copy buffer 16 in steps S0315 and S0316, but in the next step S0305, the value of the copy number table is determined to be "0". << In case of cells other than top cell (BOM) >> As a premise,
In this case, in the cell header part of the cell, FIG.
, The segment type (IST) is set to indicate COM or EOM and has the same message identifier (MID) and VCI as attached to the BOM based on the same variable length message. .

In this case, it is premised that the reading and transfer processing of the first cell (BOM) has been completed. Therefore, the corresponding entry (MID
And an entry determined by VCI), a value of 1 or more is set.

In this case, when the processing advances from step S0304 to step S0317, it is checked whether or not the value of the corresponding entry in the copy number table is "0".
As described above, in this case, a value of "1" or more should have been set, so if a value of "0" is set, it can be considered that some kind of error has occurred. it can. Therefore, the process proceeds to step S0321. Since the processing contents after step S0321 have already been described, the duplicate description will be omitted here.

On the other hand, when a value of "1" or more is set in the corresponding entry of the copy number table, in the next step S0318, whether an error cell is received during the copying of COM or EOM (EFMS = 0). Check it out. When an error cell is received (EFMS = 0), error cell conversion is performed in step S0319.
That is, the EFGA flag on the signal line OEF1 is set to the error state (= 0).

If it is determined in step S0318 that an error cell has not been received (EFMS = 1), and if the processing in step S0319 is completed, step S0
At 320, a subroutine for processing other than BOM shown in FIG. 20 is executed. In the first step S1001 after entering this subroutine, the value of the copy counter 17a is changed to
The copy number field (COPY CONT.) Is written in the cell header of the error cell (see FIG. 35). The initial value of this value is "0".

In the next step S1002, the cell is made to follow the head cell (BOM) based on the same variable length message by the message identifier (MID) written in the cell header, and the inter-station server device 8 or the intra-station server device Output from 7. The configuration of the cell at this point is shown in FIG.

In the next step S1003, the value of the copy counter 17a is incremented by "1". afterwards,
The process advances to step S1004. This step S1
In 004, the copy counter 17a after the increment
It is checked whether or not the value of is the same as the value described in the corresponding entry of the copy number table.

If the two values do not match,
Since the required number of cell copies have not been completed, the process proceeds to step S0315 in the main routine of FIG. Then, in step S0316, the same cell is read from the copy buffer 16.

On the other hand, if it is determined in step S1004 that the two values match, it is considered that copying of the cell has been completed. Therefore, in step S1005, it is checked whether the cell is an error cell. Then, if the cell is an error cell,
In step S1007, "0" is written in the corresponding entry of the copy number table, and then the process proceeds to step S0309 in the main routine of FIG. 11 to read the next cell. However, since the corresponding entry "0" of the copy number table is set in step S1007, the next cell is not output. This is to prevent omission of variable length messages (packets).

On the other hand, if it is determined in step S1005 that the cell is not an error cell, it is checked in step S1006 whether the cell is the last cell (EOM). Then, the cell is the final cell (EO
If not M), to read the next cell,
In step S0309 of the main routine shown in FIG. 11, the process of step S0301 is re-executed.

On the other hand, if it is determined in step S1006 that the cell is the last cell, 0 is written in the copy number table in step S1007. This is because the read processing (and copy processing) of all cells regarding the variable length message has been completed, so that the initial values are returned to the read processing of cells forming another variable length message. Therefore, in the subsequent processing of the main routine, a new head cell (BOM) is read.

[0140]

As described above, according to the present invention, group address data can be delivered to a terminal which is not directly accommodated in each exchange forming a public communication network.

If the message to be transferred is divided into ATM cell bases and transferred, the group address transfer control in the connectionless communication service using the ATM network and the cell transfer related to the control. Control (illegal processing) etc. can be processed efficiently and at high speed.
Further, by accommodating the group address in the exchange, it becomes possible to reduce the amount of table memory for storing the destination address provided in the server of the originating LAN, thus reducing the size of the server. Can contribute.

[Brief description of drawings]

FIG. 1 is a principle diagram of the present invention.

FIG. 2 is a block diagram of a B-ISDN according to an embodiment of the present invention.

FIG. 3 is a format diagram showing the configuration of cells transferred between the exchanges of FIG.

4 is a detailed configuration block diagram of the exchange shown in FIG.

5 is an explanatory diagram of conversion of a variable length message and a cell, which is executed by the conversion unit of FIG.

6 is a detailed configuration block diagram of the interoffice server of FIG.

FIG. 7 is a detailed configuration block diagram of the CL server of FIG.

8 is a flowchart showing a process for writing a cell, which is executed by a DA analysis unit and a carrier detection unit of FIG. 7.

9 is a flowchart showing the contents of a carrier detection processing subroutine executed in step S0004 of FIG.

10 is a flowchart showing the contents of a buffer write control processing subroutine executed in step S0005 of FIG.

11 is a flowchart showing a process for reading and outputting a cell, which is executed by the read control unit and the transfer control unit of FIG. 7.

FIG. 12 is a flowchart showing a process for reading and outputting a cell, which is executed by the read control unit and the transfer control unit of FIG.

13 is a flowchart showing the contents of a buffer read processing subroutine executed in step S0301 of FIG.

14 is a flowchart showing the contents of a setting processing subroutine executed in step S0303 of FIG.

15 is a flowchart showing the contents of a BOM copy processing subroutine executed in step S0308 of FIG.

16 is a flowchart showing the contents of a BOM copy processing subroutine executed in step S0308 of FIG.

17 is a flowchart showing the contents of an error cell transfer control subroutine executed in step S0311 of FIG. 11 and step S0321 of FIG.

FIG. 18 is a flowchart showing the contents of a BOM IA processing subroutine executed in step S0323 of FIG.

FIG. 19 is a flowchart showing the contents of a dummy cell transfer control processing subroutine executed in step S0314 of FIG.

FIG. 20 is a flowchart showing the contents of a processing subroutine other than BOM executed in step S0320 of FIG.

21 is a configuration diagram of a carrier table stored in the carrier valid / invalid storage RAM of FIG. 7. FIG.

22 is a configuration diagram of a TPS table stored in the carrier valid / invalid storage RAM of FIG. 7.

23 is a configuration diagram of a write ok table stored in the write ok information storage RAM in FIG. 7.

24 is a configuration diagram of a copy number table stored in the RAM for storing GA copy number in FIG. 7.

FIG. 25 is a block diagram of a member start table stored in the RAM for storing GA member start address in FIG. 7.

FIG. 26 is a configuration diagram of a member IA / TPS table stored in the RAM for storing GA member IA etc. in FIG. 7.

FIG. 27 is a configuration diagram of a link number table stored in the RAM for storing link numbers and the like in FIG. 7.

28 is a configuration diagram of an ICI / ISSI valid / invalid table stored in the RAM for storing the link number and the like in FIG. 7.

FIG. 29 is a format diagram of the top cell (BOM) at the time of input to the CL server.

FIG. 30 is a format diagram of the top cell (BOM) when writing to the buffer section.

FIG. 31 is a format diagram of the top cell (BOM) when reading from the buffer section.

FIG. 32 is a format diagram of a head cell (BOM) when outputting from a CL server for group address transfer.

FIG. 33 is a format diagram of a head cell (BOM) when outputting from a CL server as individual address transfer.

FIG. 34 is a format diagram of cells other than the first cell (BOM) when writing to and reading from the buffer unit.

FIG. 35 is a format diagram of cells other than the head cell (BOM) at the time of output from the CL server.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 exchange 2 conversion part 3 ATM cell switch 7 intra-server server device 8 inter-station server device 8a CL server 9 LAN 10 individual terminal 11 DA analysis part 12 carrier detection part 13 write control part 14 buffer part 15 read control part 16 copy buffer 17 Transfer control unit 18a Copy counter

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Internal reference number FI Technical display location 9466-5K H04L 11/20 102 A (72) Inventor Hideaki Yamada Hakata Ekimae, Hakata-ku, Fukuoka-shi, Fukuoka 22-22-8 Fujitsu Kyushu Digital Technology Co., Ltd. (72) Inventor Mitsuharu Wakayoshi 3-22-8 Hakataekimae, Hakata-ku, Fukuoka City, Fukuoka Prefecture Inventor at Fujitsu Kyushu Digital Technology Co., Ltd. (72) Maehara Showa 3-22-8 Hakataekimae, Hakata-ku, Fukuoka City, Fukuoka Prefecture, Fujitsu Limited Kyushu Digital Technology Co., Ltd. (72) Inventor Katsuya Shirota 3-28-1 Joto, Oyama City, Tochigi Prefecture Fujitsu Digital In Technology Co., Ltd. (72) inventor Ken Kawasaki Kanagawa Prefecture, Nakahara-ku, Kawasaki, Kamikodanaka 1015 address Fujitsu within Co., Ltd. (72) inventor Yasuhiro Aso Kawasaki City, Kanagawa Prefecture Nakahara-ku, Kamikodanaka 1015 address Fujitsu within Co., Ltd.

Claims (14)

[Claims] 1. An exchange forming a communication network using digital communication data, wherein a receiving unit for receiving the digital communication data, and transfer information to be distributed to a terminal not directly accommodated in the communication network, A transfer information detection unit that detects from the digital communication data received by the reception unit, an address storage unit that stores in advance address information of the terminal corresponding to the transfer information, and the transfer information detection unit that transfers the transfer information. When information is detected, a transfer control unit for reading the address information of the terminal corresponding to this transfer information from the address storage unit and writing this address information in the digital communication data is provided. Digital exchange. 2. The digital exchange according to claim 1, wherein the terminal is connected to the communication network via a regional information communication network. 3. The address storage unit stores each address information of a plurality of terminals corresponding to one piece of the transfer information, and the transfer control unit stores a plurality of address information corresponding to the transfer information. When there are a plurality of the pieces of address information, the digital communication data is copied according to the number of the address information, and the address information is written one by one to each of the copied digital communication data, respectively. switch. 4. A copy buffer for storing digital communication data received by the receiving section, further comprising: when the digital communication data is copied, the transfer control section uses the digital communication from the copy buffer. The digital exchange according to claim 3, wherein the data is read. 5. The receiving unit, upon receiving the digital communication data, divides the digital communication data into a plurality of segments, and stores the transfer information in the first segment of the plurality of segments. The digital exchange according to claim 1. 6. The transfer information detector detects the transfer information from the first segment, and the transfer controller detects the transfer information when the transfer information is detected by the transfer information detector. 6. The digital exchange according to claim 5, wherein address information of the corresponding terminal is read from the address storage unit and the address information is written in the first segment. 7. The digital exchange according to claim 5, wherein the receiving unit attaches the same identification information to each of the plurality of segments based on the same digital communication data. 8. The address storage unit stores address information of a plurality of terminals corresponding to one piece of the transfer information, and the transfer control unit stores a plurality of address information corresponding to the transfer information. In the case of the number of the address information, the head segment is copied according to the number of the address information, and the address information is written one by one to each of the copied head segments, and has the same identification information as the head segment. 8. The digital exchange according to claim 7, wherein the other segments are copied by the same number as the number of copies of the first segment. 9. The transfer control unit has a counter for measuring the number of copies of the segment and a copy number table for storing the measured value of the counter, and the counter when the copying of the first segment is completed. Is stored in the copy number table, and when the other segment is copied, copying is performed until the measured value of the counter becomes the same as the value stored in the copy number table. The digital exchange according to claim 8. 10. The transfer control unit, when an error occurs while copying the segment, stores the measured value of the counter up to that point in the copy number table. The digital exchange according to claim 9. 11. The digital exchange according to claim 1, wherein the transfer information is written at a position for writing an incoming address in the digital communication data. 12. The transfer information detecting section displays an error in the segment when an error occurs in the segment, and the transfer control section, when the error display is attached to the segment, 10. The digital exchange according to claim 9, wherein a minimum value of the counter is set in the copy number table. 13. The transfer control unit, when a value equal to or greater than the minimum value of the counter is stored in the copy number table at the time of receiving the head segment, sets a dummy segment in the copy number table. 10. The digital exchange according to claim 9, wherein the address information is written into the leading segment after the output of the difference between the value and the minimum value of the counter is output. 14. A method of transferring the digital communication data in an exchange forming a communication network using the digital communication data, wherein the digital communication data is received and delivered to a terminal not directly accommodated in the communication network. Transfer information to that effect is detected from the digital communication data received by the receiving unit, and address information previously stored as address information of the terminal corresponding to the transfer information is written in the digital communication data. A method of transferring digital communication data, comprising outputting the digital communication data according to address information.