JPH0294843A - Network system and error detection processing circuit - Google Patents

Abstract

PURPOSE:To attain error detection in a connection network and to improve transfer efficiency in the connection network by using an error detection method used in an individual network in an exchange connecting to a transmission and a reception end. CONSTITUTION:Exchanges at the transmission end and the reception end are provided with both an MAC layer processing function for LANs 421, 422 accommodated and an exchange function in the connection network being a function of a relay connection exchange. A 4-byte FCS (frame check sequence) in the MAC layer is used for error detection and connection exchanges 431, 433 use the FCS to guarantee the completeness of the information in the connection network. Thus, an error detection request different from a medium is implemented easily and the transmission efficiency is increased by 10-20% more in the multi-medium packet network accommodating also a voice packet.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a plurality of individual networks such as LANs (or communication devices)
The present invention relates to a connection network that interconnects networks, and in particular to a network system suitable for detecting errors within the network in a connection network using a frame relay exchange method, and an error detection processing circuit used in the system.

[Conventional technology]

Conventionally, as a network system having a connection network that interconnects individual networks (LAN) of this type, for example,
As shown in Japanese Patent Application Laid-open No. 60-26.1143,
The connecting network independently adds error detection information to the individual network information,
A system is known in which this error detection information is used to detect errors within a connected VTM (between individual networks) to maintain data safety.

[L’i problem that the invention seeks to solve]

Normally, in a connection network, information of different nature, such as traffic distribution and transfer request time, as well as information of 7M+a (multimedia) such as computer data and telephone voice information, are usually used.
The packets are multiplexed and transferred in the same procedure and on the same transmission path.

In order to realize a multimedia river that can process all information at high speed, especially when audio information is transferred in packets, in order to avoid deterioration of audio quality due to transfer delay, several 10 bytes are required within the connection network. It is advantageous to use short packets of fixed length. In other words, real-time transfer of voice information such as telephone calls requires a shorter transfer delay (call delay) than computer data, so it is necessary to avoid long packets and use short packets. .

However, when attempting to shorten packets using a method in which the connecting network adds its own error detection information as in the conventional technology described above, error detection is required for each short packet used within the connected Vt network. As information is added,
There is a problem in that the ratio of information length to packet length decreases, resulting in a decrease in information transfer efficiency.

On the other hand, voice information has more redundancy than computer data, so even if some packets are dropped, it is less likely to cause sound quality deterioration, and with the recent use of optical fiber, the error rate in connection networks has increased. Error checking tends to decrease in the connected network (
Studies are underway to detect errors in packets using only the packet header. A bucket like this
The format is CCTTT's 5GXVIII" arching group, and it is an ATM (Asynchronous Tra
nsfer Mode). However, if error checking is not performed within the connection network, the reliability of the data for the user will be lost.

For example, in a connected MUM, it is common to use fixed-length short packets of about 10 bytes;・Assuming that a packet with a check sequence) is used, 12
．． A reduction in transfer efficiency of 5% (4/32) will occur.

Therefore, an object of the present invention is to provide a network system that solves the problems of the prior art described above and improves data reliability by making it possible to detect errors within a connection network without reducing transfer efficiency. It's about doing.

[Means to solve the problem]

In order to achieve the above object, the inventors of the present invention have conducted various studies and found that if the error detection method used in individual networks is applied as is to error checking in connected networks, it would be possible to improve the conventional technology. We focused on the fact that errors in the connecting network can be detected without adding unique error detection information to each packet.

In other words, in the network system of the present invention, both when a sending end connection switching device receives a switching frame from an individual network under its control and when a receiving end connection switching device receives a switching frame from a connecting network, Error detection is performed using an exchange frame error detection method (e.g. FC3 method),
As a result, only frames with no errors are output.

A preferred error detection processing circuit used in the network system of the present invention includes an exchange frame reproducing unit comprising a memory for storing a received exchange frame and a detector for detecting final information of the exchange frame; an error detection unit comprising an arithmetic unit that performs a new calculation and a comparator that compares the output of the arithmetic unit with the exchange frame blank detection information; an output of the detector in the exchange frame reproduction unit; and an output of the comparator in the error detection unit. and an individual network connection unit that generates a request for transmission of the exchange frame stored in the memory according to the above.

[Effect]

The operation based on the above configuration will be explained.

In the transmitting end connection switching equipment, when forwarding information is received from an individual network, error detection is performed using an error detection method applied within the individual network (for example, FC3 determination method within a LAN), and then it is determined that there is no error. Only forwarding information is sent to the connecting network.

On the other hand, when the receiving end connection switching device receives transfer information from the connection network, it calculates new error detection information using the error detection method within the individual network (intra-LAN FC3 determination method) that is the destination of the transfer information. However, if the transfer information is received, it can be determined that an error has occurred within the connection network.

Within the connecting network, exchange frames used within the individual network are stored and transferred. Therefore, the separate intra-network error detection methods for the source and destination match.

In addition, in a connection network that uses a frame relay method that exchanges at a layer equivalent to the data link layer in the O3I reference model, the relay connection switching device in the connection network uses a unique It is not necessary to add error detection information.

As a result, the ratio of information length to packet length increases,
Information transfer efficiency is improved.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

First, an embodiment of the system configuration of the present invention is shown in FIG. LAN421 and LAN422 are IEEE80
It is a LAN standardized by two committees, and has connection exchange equipment 4.
31, a connection exchange device 432, and a connection exchange device 433, it is positioned as an individual network. L
AN421 is a LA that executes a LAN communication protocol.
Through the N nodes 411 and 412, the L A N terminal 4
01 and 402. Similarly, LAN 422 accommodates LAN terminal 403 through LAN node 413. Furthermore, telephones 441 and 442 that handle voice information are also
It is accommodated via BX443.444.

When exchanging information between the LAN terminal 401 and the LANJ terminal 402, it is possible to exchange within the LAN 421, but
When transferring information from the LAN terminal 402 to the LAN terminal 403, the LAN 421 and the LAN terminal via the connection network are transferred.
This is communication between AN422 LANs. In this case, the connection exchange device 431 is the sending end connection exchange device, the connection exchange device 432
is a relay connection switching device, and connection switching device 333 is a receiving end connection switching device.

The transmitting end and receiving end connection switching equipment accommodates the LAN4
21 and [, MAC layer processing function of AN422,
It has both the connection network switching function, which is the function of a relay connection switching device. Within this MAC layer, 4 bytes of Fe2
(frame check sequence) is used for error detection, and the connection switching devices 431 and 433 use this Fe2 to guarantee the integrity of the information in the connection aX 14.

In the connection network C, exchange is performed using a frame relay method, and the frame format of each part is shown in FIG.

A frame format 501 in FIG. 5 is a frame format within the communication source LAN, and only the main components are shown here.

The length of one frame is usually about 4 to 4 bytes long. Destination address (DA) 512, source address (SA) 513, and information section (1) 514 each of 6 bytes
and a 4-byte FC5515 are sandwiched between 1-byte start and end flags (F) 511 and 516.

This frame format is used in LAN intercommunication tA(a) in FIG.

The frame format 502 is an intra-network frame format,
A 4-byte header (HD>517 and 31-byte divided transfer information (SDU
)518. The 5DU 518 is obtained by dividing the LAN frame contents into 31-byte transfer information while preserving the contents.

As shown in an enlarged 503, the header (HD) 517 includes a call number (LC) that can uniquely identify the communication destination LAN and relay route, which is assigned in advance to the LAN@terminal pair for communication.
A) 519, a packet type (PI) 520 indicating the type of voice packet or data packet, and the above LCA 519.
and error detection information (CHK) 52 for PI320
1 and a division number (AN) 522. Note that the error detection information 521 includes LCA 519 and P I 520
In particular, the frame format 502, which detects errors only in CA 519 and does not check errors in the entire data content, is used in connection intra-plane communication (b) in FIG. 4.

The frame format 504 is a frame format within the destination LAN, and is the same format as the frame format 501 within the source LAN.

FIG. 1 shows a functional block diagram of a transmitting end and receiving end connection switching device. The received information 120 from r, A N is
This frame format is 501 in FIG. 5. This frame format is input to the LAN reception control unit 101 that processes LAN MAC layer protocols such as FCS check, and the LAN reception control unit 101 processes the frame. Health is checked. After this, the address selection section 102 selects L
It selects whether the received information from the AN is intra-LAN information or inter-LAN transfer line information, and only if this is inter-LAN transfer line information, the received information is transferred to the next stage.

This address selection method can be realized by, for example, comparing the destination address DA of the received information with the LAN terminal address (in the own connection switching device) that is stored in advance in the address selection unit and is directly accommodated. .

As a first function, the destination determination/information division unit 103 determines, based on the destination address DA in the received information, the transmitting end connection switching device in which the destination LAN is accommodated and the relay connection switching device leading to the device.

The connection device PAD 104 is instructed to allocate the LCA 519 shown in FIG. 5 to this logical communication path and notify the relay connection switching device of the relay route and LCA in the frame format 502 of FIG. The fact that the packet sent from 121 at this time is a notification packet means that the packet type (
PI) 520. Furthermore, this packet is sent to connection device B.
It is created in PAD 104.

The second function of the destination determination/information division unit 103 is to divide the transfer information into 31-byte transfer 4 [518 lengths] and assign a division number (AN) 522 after sending the notification packet.

Thereafter, the packet is assembled into a frame format 502 packet by the connection device PAD104, and then sent to the switch 105. The switch 105 includes a buffer, inputs packets from other similar connected devices PAD104'104#, rearranges the packets as appropriate, sorts them by destination, and multiplexes them according to each destination. Transmission section 10
Send to 6°106'... Multiplex transmitter 106.106
', those in the same direction are the same outgoing line 121°12
1'... are multiplexed and sent out while controlling the sending order. In this way, packet multiplex communication is performed.

On the other hand, regarding packets of frame format 502 received from the incoming line 122, the packets to be relayed are exchanged again at the switch 105 so as to be sent to the next-stage connection switching device.

The received packet is sent to the information assembling unit 108. The information rfVI assembling unit 108 includes a memory for storing received packets in exchange frame units and a detector for detecting that the packet is the final information of the exchange frame. After it is removed, it is separated into LCAs, and the ordering is checked and the final information reception is determined based on the division number (AN) for each LCA. This check is performed on packets 5 with the same LCA.
This is achieved by comparing the counter and the division number each time 02 is received.

The intra-LAN FC3 determination unit 109 calculates new error detection information (FCS information) based on the connection intra-LAN error detection method using the transfer information for each LCA, and compares the calculation result with the received FC5 information. The failure characteristics are compared, and based on this result and the final information reception signal from the information assembly 108, a transfer data transmission request signal is issued to the LAN transmission control section 110. L.A.
The N transmission control unit 110 performs acquisition and release control of LAN access rights, and transmits information associated with acquisition of access rights.

FIG. 2 shows an error detection processing circuit which is a main circuit of the intra-LAN FC3 determination section 109.

The signal a is the transfer information received from the information assembling unit 108, and the signal S is the exchange frame error detection information f8 in the signal a.
This is the output of the comparator 211 that compares the arithmetic unit output f, which is new error detection information calculated by the FC3 arithmetic unit 210 based on g-1, with the received exchange frame error detection information g-2.

Signal C is the final information signal from the information assembling section 10B, and the AND of signal C and signal d becomes transfer data transmission request signal 6. The latch circuits 201 to 205 are temporary memories for matching the phases. In the circuit shown in Figure 2, FC
S check result (signal b) and end packet reception (signal C)
) is configured to generate a transmission request to the LAN.

FIG. 3 shows a waveform time chart of each part of the error detection processing circuit of FIG. 2. The solid line of the signal in FIG.
(if there was no change). As a result, if there is no error, the transfer data transmission request signal e becomes logic 1 from the signal d and the signal C, and the LAN transmission transmission control function operates.

According to this embodiment, since fixed length and short packets are used as the frame format for transfer within the connection network, it is possible to store voice packets that require real-time transfer within the connection network using the same transfer procedure. In other words, in order to improve the quality of voice through real-time transmission, it is generally required to have as little delay as possible, but by using fixed length and short packets, voice packets can be inserted immediately at any position in various data packets. This makes it possible to transfer audio without delay. In addition, since voice packets have a strong tolerance to information errors (even if some of the voice packets are dropped, there is very little deterioration in quality), information error detection is generally not performed within the connection Vt1ii. is customary. On the other hand, since the quality of data deteriorates significantly due to information loss, it is required to provide an error detection function. According to the method of this embodiment, which detects the occurrence of an error within a connected network using an individual intra-network error detection method, it is possible to easily respond to different error detection requests based on such media.

According to this implementation, error detection within the connection network is performed without adding new error detection information within the connection network, so in a multimedia packet network that also accommodates voice packets,
Transfer efficiency can be improved by about 10% to 20%.

〔Effect of the invention〕

As described in detail above, according to the present invention, the error detection information is used within the individual network within the connection switching device at the sending end and the receiving end without adding new error detection information to the intra-connection transfer packet within the connection network. By using the error detection method, it is possible to detect errors within the connection network, so that excellent effects such as sufficiently improving transfer efficiency within the connection network can be achieved.

[Brief explanation of the drawing]

FIG. 1 shows a LAN of a connection switching device according to an embodiment of the present invention.
A block diagram showing the configuration of the connection part, Figure 2 is the LA of Figure 1.
A configuration diagram of the error detection and processing circuit which is the main circuit of the N-FC5 determination section. FIG. 3 is a type for explaining the operation of FIG. 2, Chars 1 to 4 are according to the embodiment of the present invention. FIG. 5, which is a block diagram of a system to which the error detection method is applied, is a diagram showing the frame format used in the system of FIG. 101・・・・・・LAN reception system J1 [1st part (L
(including the FC5 determination unit within the AN), 109...
Bow, A N internal F'C3 judgment section, 210...
...FC3 arithmetic unit, 211...Error detection information comparator, 401, 402.4 (13...
・・LAN terminal, 411, 412, 413...
...LAN node, 421. ．． 422・・・・・・・・・
・・LAN (individual network), 431・・・・・・・・・
Connection exchange device (at the sending end), 432...
(Relay) connection exchange equipment, 433...
(receiving end) connection switching device, 450...... Connection X-JT network, 501...... Frame format in source LAN, O2...... ... Frame format within the connection network, ... Destination LAN frame format. 4... Figure Figure Figure

Claims (1)

[Claims] 1. The exchange frame format used in a plurality of individual networks (or communication devices) is saved, and the individual networks (or communication devices) In a network system having a transmission end connection switching device and a reception end connection switching device configuring a connection network that performs the interconnection, the transmission end connection switching device connects the transmission source directly accommodated therein. When an exchange frame is received from an individual network (or a communication device), an error detection method is used to detect errors within the exchange frame, and only frames with no errors are transferred to the connection network, In the receiving end connection switching device,
A network system configured to perform error detection using the exchanged frame error detection method and to transfer only frames with no errors as a result to a destination individual network (or communication device) directly accommodated. . 2. An information assembly unit consisting of a memory that stores received packets in exchange frame units and a detector that detects that the packet is the final information of the exchange frame, and an arithmetic unit that newly calculates error detection information within the exchange frame. and an error detection unit comprising a comparator that compares the output of the arithmetic unit with the exchanged intra-frame error detection information; 2. An error detection processing circuit for use in a network system according to claim 1, further comprising an individual network connection unit for generating a transmission request for a stored exchange frame.