JPH02226840A - Looped network - Google Patents

Abstract

PURPOSE:To simplify the constitution, to improve the flexibility, and to obtain speediness and high reliability hy constituting a looped transmission line So that the line speed of a trunk line is equal to the transmission speed of branch lines and the branch lines pass through the line of the trunk line at need. CONSTITUTION:The trunk line optical looped network consists of stations 10a and 10b on the trunk line, the optical transmission line (trunk) 11, and stations 12a, 12b... 13a and 13b... on the branch lines. The looped transmission line is constituted so that the line speed of the trunk line 11 is equal to the transmission speed of the branch lines and the branch lines pass through the line of the trunk line 11 at need. Consequently, the system of the simple constitution has flexibility and the looped network which has the speediness and high reliability is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to a loop network used, for example, as a trunk line of a private network.

(Prior Art) Since around 1970, loop networks have been used as private networks. Initially, 1.5Mb
Electric wires were used for transmission speeds of PS, but later on, when used in the field of process control, optical technology was adopted, and transmission speeds increased to 10Mbps, 32Mbps, and in recent years 100Mbps transmission. Used for speed.

It began to be used in OA equipment around 1980, and is expected to be used in engineering automation and factory automation in the future.

Another trend in recent years is multimedia. In other words, it was initially used to transfer code data, but recently it has started to be used to transfer audio and still images.
It is expected that video transfer will be implemented in the future.

Therefore, it is thought that a large number of computers, including personal computers, will be used in the future, and the amount of communication, including still image transmission and voice communication, will continue to increase.

By the way, when a system becomes large-scale, systems having certain functions may be scattered over a wide area. For example, in a complex, large-scale system such as an airport, buildings and ground facilities are scattered over a wide site, each of which plays a specific role.

In such a large-scale system, a plurality of different types of loop networks described above may be used. For example, in one loop network, voice is the main transmission data, in another loop network, the main transmission data is packet data, and in some cases, a large amount of video data is sent. In such cases, each cable may be laid separately and in parallel, but if possible, it is more scalable and economical to lay a main line connecting each network and use that. be.

However, until now, there has been no trunk loop network that integrates such different types of loop networks.

However, as a similar network, there is a loop network as shown in FIG. 9, for example.

The loop network shown in the figure is constructed by connecting a plurality of stations 1 in a loop through transmission lines 2, and connecting each station 1 to a bus type LAN 4 via a bridge 3.

However, when the loop network shown in FIG. 9 is used for the purpose of integrating the above-mentioned different types of loop networks as a trunk loop network, there are various problems as listed below.

■The above objective can be achieved by installing multiple loop networks in parallel, but in this case, installation costs and management/operation problems will occur. That is, the installation cost is high, and it is necessary to physically install many cables in parallel.

■Loopbacks and system switching have been used to counter failures in transmission lines, and loopbacks and bypasses have been used to deal with station failures. However, in order to realize these methods, complicated software is required, and due to the complexity, there is a risk of malfunction. (2) Regarding the generation of frames, a supervisory station (master station) has performed control to generate frames and to ensure that a positive number of frames exist on the loop network. However, such a system is complicated and becomes expensive as the transmission path becomes faster.

■The supervisory station has been responsible for allocating channels. That is, the supervisory station has a list of channels, and when a channel is requested, it allocates an empty channel. However, while this method is simple in principle and has few errors, it depends on the reliability of the supervisory station, which poses a difficult problem when implementing supervisory station duplication. come.

■As for the clocking method, a regenerative relay method has been adopted. That is, the supervisory station generates a clock, and the other stations extract the clock from the received signal and use it for transmission. This method allows synchronization between each station, but it is difficult to extract the clock from the high-speed signal.
Furthermore, as the number of relays increases, there is a problem in that jitter occurs in the clock and abnormalities occur in the relay operation.

(2) A centralized method has been adopted for the generation of control packets used for loop network management, etc., but this method also assumes the existence of a supervisory station, and reliability is an issue.

Another similar network is a wide area network.

In a wide area network! , a 5 Mbps line is transmitted via an 8 Mbps line, and an 8 Mbps line is transmitted via a 32 Mbps line. That is, in this case, 6Mbps, 3
The 2Mbps line is used as a trunk line. However, although the wide area network is similar to the loop network described above, the scale and method used to achieve the above objectives are essentially different.

(Problems to be Solved by the Invention) It has been difficult to realize a trunk loop network formed by integrating different types of loop networks for various reasons.

The present invention was made in order to solve the above-mentioned problems, and its purpose is to provide a loop network that has a simple configuration, has a flexible system, and can realize high speed and high reliability. Ru.

[Structure of the Invention] (Means for Solving the Problem) In order to solve the problem, the present invention provides a transmission line that connects a main line consisting of a loop-shaped transmission line including a plurality of lines and a terminal or bridge. The line speed of the main line is equal to the transmission speed of the branch line, and the branch line forms a loop-shaped transmission line via the line of the main line as necessary.

The invention according to claim 2 provides a loop network in which a plurality of stations are connected in a loop through transmission paths, in which each station generates its own frame, and each bit of the generated frame is assigned to each channel. It constitutes.

According to the third aspect of the invention, a plurality of stages are connected in a loop through a transmission path, each station generates its own frame, and each bit of the generated frame constitutes each channel. In the loop network, when a control packet having an empty/busy table indicating empty/busy channels of the line is circulated on the transmission path, and each station searches for an empty channel from the empty/busy table on the control packet, The used section of the line and the name of the station that uses the section are written on the line empty/blocked table, and the used section is recognized when the written content is found on the control packet that has made a round of the transmission path. It is judged as a thing.

(Function) In the invention described in claim 1, the line speed of the main line and the transmission speed of the branch line are equal, and the branch line forms a loop-shaped transmission line via the main line as necessary.

In the invention set forth in claim 2, each station generates its own frame, and each bit of the generated frame constitutes each channel.

In the invention as claimed in claim 3, when a control packet having an empty/busy table indicating empty/busy channels of the line is circulated on the transmission path, and each station searches for an empty channel from the empty/busy table on the control packet, Enter the section of the line used and the name of the station that uses the section on the line empty/blocked table.
It is determined that the usage section has been recognized when the written content is found on the control packet that has made one circuit around the transmission path.

Therefore, according to these inventions, the system has flexibility with a simple configuration, and can also realize high speed and high reliability.

(Example) Next, an example of the present invention will be described in detail based on the drawings.

The functions of the trunk optical loop network according to an embodiment of the present invention include having packet switching and circuit switching functions, accommodating branch lines, and accommodating high-speed terminals such as video terminals and cluster controllers accommodating various interfaces. There is a particular thing.

Each component in the trunk optical loop network according to this embodiment will be explained below.

FIG. 1 is a diagram showing the configuration of a trunk optical loop network according to this embodiment.

In the figure, % 10as 10b... are mainline stations, 11 are these stations 10a110b
It is a double-loop optical transmission line (main line) that connects...

12a, 12b=... are branch line stations with a transmission speed of 1OOMbps, and these stations 1
2a and stations 12b, . . . can be directly connected via a trunk line 11. That is, the frame that has passed through station 12a is once on main line 11, and then passes through station 12b.

13a113b... are branch line stations with a transmission speed of 400Mbps, and these stations 1
Similarly, frames pass through 3a, 13b, and so on.

The trunk stations 10a, 10b, . . . each have a 1.6 GHz clock. The accuracy of the clock owned by each stage generator is within a predetermined manufacturable tolerance. Stations 12a, 12b-
, , possess a 100MHz clock source. Stations 13a, 13b, . . . own a 400 MHz clock source. These loOMHz.

The 400 MHz clock is also within a predetermined manufacturable tolerance, as in the case of the 1.BIlz clock mentioned above.

FIG. 2 is a diagram showing the structure of a frame format circulating on the main line 11.

As shown in figure (a), the frame format is 16
Consists of bits.

The first bit 20 is a frame header and includes a synchronization bit for frame synchronization and a control communication area. The contents of the frame header are as shown in FIG.
] and [XXI (X C "Om or 1") are mutually consecutive. Synchronization is performed by detecting the [lO] part. Note that conventional techniques are used to protect synchronization with the transmitter of the previous station. The [XXI part is a control communication area, where [XXI is extracted from the mutually consecutive lines [IO] and [XXI, and the continuous "X" part is used to transmit and receive control packets. The control packet is used for communication between the trunk stations 10a, 10b...
Used for network management such as channel allocation and failure countermeasures.

FIG. 3 is a diagram showing the structure of a frame format of a control packet.

As shown in the figure, the frame header FH (Fraa+e Header) uses a unique code at the beginning.
) 40, by which the beginning of the control packet is recognized. Next, the destination address DA (Desti
nalon Address) 41, source address 5A (Source Address) 42, priority P43, and command C (Co■ Kasumi and) 44 follow. Command C44 indicates what purpose the frame is used for. Next data section DA
The communication telegram is written in TE45. The last end limiter ED (end Delinlter) 46 uses a unique code to indicate that this is the end of the control packet.

Following the frame header 20 shown in FIG.
The part b... is a data communication area, and consists of a total of 15 channels from channel 0 to channel 14.

The transmission rate of each channel is 100 MHz, and this one channel is equal to the loOMHz branch line transmission rate within manufacturing technology tolerances. Also, 4 channels are 400M.
11z branch line transmission speed within manufacturing technology tolerances. Therefore, 100MIIz, 400MIIz and 1.
The 8GIIz clock must be created to maintain this relationship.

In this embodiment, each station independently generates such a frame.

FIG. 4 shows the configuration of the frame transmitting/receiving section in the station.

As shown in the figure, frame synchronization of the signal received by the optical receiver 30 is detected by the synchronization detection unit 32, and the control communication area and the data communication area are controlled by the control communication area receiving unit 33, respectively.
The data is stored in each dual port memory 35, 35 via the data communication area receiving section 34.

On the other hand, on the transmission side, a frame generation section 38 is separately provided, and transmits the control communication area and data communication area sent from the control communication area transmission section 36 and the data communication area transmission section 37 via the optical transmitter 31. Is going.

39 is a clock and timing management section.

In this way, this embodiment employs a method in which each station generates frames independently, so that the frame aligner required in the conventional method is no longer necessary. Therefore, there is no need for a supervisory station, and the system and software configuration are simplified.

FIG. 5 is a diagram showing the frame format of the branch line.

The branch line clocking method will be the PPCK method. The frame format of the cluster controller also follows this format.

As shown in the figure, there is a preamble 50 at the beginning. When this frame passes through the branch station,
It passes through this preamble length to absorb the difference in clock frequency between stations.

Start Delii!
ter) 51 and End Demilita E D (End De
llmlter) 53 indicate the beginning and end of the frame, respectively.

52 is a data section.

Next, a control packet relaying method will be explained based on the algorithm state transition diagram shown in FIG. Note that each station has a function of generating control packets.

If the next control packet is not received even after a certain period of time has passed after the reception of the previous control packet is completed, the generation of control packets is continued from then on using the last received data. On the other hand, when a control packet with a higher priority than the upstream station is received, the station stops generating control packets because a station higher in rank than the own station is generating control packets. When receiving a control packet generated by its own station, it continues to generate control packets.

As described above, in this embodiment, since the control packet is relayed using the distributed control method, there is no need for a supervisory station.

Next, a relay method in the data communication area will be explained.

FIG. 7 is a diagram showing the configuration of the data communication area processing circuit.

In the data communication area, frames are multiplexed from branch lines and cluster controllers. The signal received by the optical receiver 70 shown in the figure is divided into 100 Mbp by the channel dividing section 73.
s and 400 Mbps channels, respectively.

The preamble lengths of the frames output from the channel division section 73 are adjusted by respective preamble length adjustment sections 75. Thereafter, the signals are multiplexed by the channel multiplexer 74 and then transmitted by the optical transmitter 71.

It should be noted that among the 15 channels, which part is assigned to the 100 Mbps channel and which part is assigned to the 400 Mbps channel is determined by the configuration at the time of factory shipment.

Next, the failure countermeasure method will be explained.

As a measure against failures in this embodiment, a detour method is adopted in order to simplify the method.

This will be explained using FIG.

8th Tl4J: C, Transmission line TL Te1. BGbps
No2 Mi NolN10 IJT loop is shown.

90a%b... is a station, 91 is a camera, and 92 is a monitor.

The video signal from the camera 91 is connected to the inner IN loop and the 0
It flows using both LIT loops.

Here, since there is a failure point shown in the figure, the video signal of the 0IJT loop cannot reach the station 90b, and therefore the monitor 92 cannot receive it either. Monitor 92 selects the IN loop.

Although the control signal from the monitor 92 is flowing using the outer 0tJT loop, this lN10tlT selection is performed learnedly. That is, the monitor 92 controls the sending of the control signal from IN.
I tried using both loop and 0LIT loop, but I
When the N loop is used, no response is returned from the camera 91, and when the OUT loop is used, a response is returned from the camera 92. Therefore, from now on, communication will be performed using only the OUT loop.

Next, the characteristic points of the trunk optical loop network of this embodiment are listed below.

1, 66ibps trunk line is loOMbps /400M
Accommodates bps branch lines. loOMbps /400
Frames on branch lines of Mbps are sometimes l, 8Gbp
It passes on the main line of s.

1. On the ecbps trunk line, l is
oOMbps/400Mbps line is available. - loOMbps /400Mbps branch line as required, 100)4bps on Bbps trunk line
/400Mbps line is used.

Since 1.8 Gbps requires high-speed processing and the hardware is expensive, a simple method is used and the processing content is minimized.

-loOMbps /400Mbps branch line o-)
P P CK (PoInt to Po
1nt clock) is adopted. In other words, each station on the branch line has its own clock source, and when a frame passes through the station, by adjusting the preamble length at the beginning of the frame, the clock frequency of each station on the branch line has its own clock source. Absorb differences.

・The loOMbps/400Mbps frame on the branch line is the loOMbps/400M frame prepared on the main line.
When using a bps line, passing through stations forming a trunk line is also possible by adjusting the preamble length. i.e. 100ML on the main line) l) S /4
00 Mbps lines also have a preamble adjustment device owned by each station on the branch line.

A frame on the φ main line consists of 16 bits.

The first bit is the frame header, and the remaining 15 bits provide 15 lines for loOMbps. It is also possible to prepare multiple loOMbps and 400M lines. The frame header is used for frame synchronization establishment and control communication area. Control packets in the control communication area are used for communication between stations, and are used for tasks such as failure recovery and line allocation.

・Except for control packets, the detour method will be used as a countermeasure against disconnections, station failures, etc. The transmission path is a double loop system, and there are two routes from one station to another. When a failure occurs, the other functioning transmission line is used.

・In the control packet area, (Loop is established/
There is an area for displaying (disconnected) and an area for displaying (name of control packet generated STN).

Therefore, each station knows whether there is a fault and the location of the fault. When there is a unilateral system failure, a living system is used.

When there is a failure in both systems, the contents of the control packet are looped back before and after the disconnection in both systems. Therefore, the control packet is always configured as a loop, and an algorithm can be used in which a channel can be secured if the control packet goes around the loop once at the time of line allocation.

・Channel assignment is performed using the table in the control packet. The table has a column for the number of lines, and the section in use for each line is listed. If the section to be used is blank, that portion of the line is unused. A station wishing to use the line searches for an unused section of the line, and when it finds one, writes in its own number. If the number you entered is returned and you do not find another station's number in that field by the time it returns, you can use that section of the line from now on. On the other hand, if another number is found, the user must give up and start searching for a free line again.

・Generation of control packets shall be performed using a distributed control method. Normally, the station adds information about its own station to the received control packet and transmits it.

If a control packet is not received even after waiting for a certain period of time, the own station transmits a control packet.

- Send control packets to both systems. The transmitting station erases the transmitted data. When the transmission line is broken and the transmission line is not looped, there is no need to erase the transmitted data.

- Line allocation for 100Mbps and 400M will be determined as specifications at the time of manufacturing.

Although the present embodiment is explained using a 1.6 Gbps system, the present invention is sufficiently effective at other speeds as well.

In this embodiment, by having the above features, the following effects can be achieved.

■By connecting one base to another base via a trunk line, installation and operating costs are economical, and maintenance, operation, and expandability are also excellent.

■The method has become simpler by using a detour method as a countermeasure against transmission line failures and station failures.

Furthermore, if the fault location is an unused section, the fault recovery operation will not affect communication.

(2) According to the conventional system, a supervisory station controls the generation of frames and the presence of integral multiples of frames on the loop network.

However, according to such a system, one supervisory station must be prepared within the system and its maintenance must also be performed. Therefore, as in this method, we made the frames as short as 1θ bits and made it possible to generate frames at each station, so the supervisor could
No station is required.

- By adopting a distributed control method for channel allocation, a supervisory station is no longer necessary.

■It is now possible to mix a 100 Mbps system and a 400 Mbps system on a BGbps trunk line.

■100Mbps/400Mbps lines on trunk lines also have the following characteristics by adopting PPCK as the clocking method.

(1) There is no accumulation of jitter in the clock.

H) In the case of a regenerative relay system, it is necessary to determine one clock source within the system and take measures in case a failure occurs in that clock source.

However, according to this system, each station owns a clock, so there is no need to determine a single clock source within the system, and there is no need to take measures against its failures.

(iii) When a regenerative relay method is adopted, there is a problem of how to synchronize with the wide area network. That is, a. The independent synchronization method is a method in which a high-precision oscillator such as a rubidium oscillator is provided within the system, but slips occur and data errors occur, albeit to a limited extent.

b. A method of synchronizing a 1.8 Gbps clock source with a wide area network is also considered, but this seems difficult to implement.

To solve the above problem, if the branch line synchronizes with the wide area network as in this embodiment, there is no need for the main line to synchronize with the wide area network.

■When a regenerative relay system is adopted, there is only one clock system in the system, so a partial failure will affect the entire system. In a large-scale system, it is not desirable to have a system in which failures spread throughout the system like this. On the other hand, in the case of the PPCK system of this embodiment, even if a part fails, it does not affect the entire system and the recovery operation is quick.

- Since control packets are also generated using a distributed control method, a supervisory station is not required.

[Effects of the Invention] As described above, according to the present invention, the system has a simple configuration, has flexibility, and can realize high speed and high reliability.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the overall system configuration of a trunk optical loop network according to an embodiment of the present invention, FIG. 2 is a diagram showing the configuration of a frame format circulating on the trunk, and FIG. 3 is a frame format of a control packet. A diagram showing the configuration of
Figure 4 is a diagram showing the frame transmitting/receiving section of the station, Figure 5 is a diagram showing the frame format configuration of the branch line, Figure 6 is an algorithm state transition diagram of control packet generation and relaying method, and Figure 7 is data communication. FIG. 8 is a diagram showing the configuration of the area processing circuit; FIG. 8 is a diagram for explaining the failure bypass method; FIG.
The figure shows a trunk LAN in which a large number of bus-type LANs are connected via bridges. 10a, 10b...Station of main line, 11...
Optical transmission line (main line), 12as 12b, 13a, 13
b... Branch line station. Applicant: Toshiba Corporation

Claims (3)

[Claims] (1) A main line consisting of a loop-shaped transmission line including a plurality of lines is connected to a branch line consisting of a transmission line to which a terminal or a bridge is connected, and the line speed of the main line and the transmission speed of the branch line are equal, and A loop network characterized in that a branch line forms a loop-shaped transmission line via the main line as necessary. (2) In a loop network in which multiple stations are connected in a loop through transmission paths, each station generates its own frame, and each bit of the generated frame constitutes each channel. Features a loop network. (3) In a loop network in which a plurality of stations are connected in a loop through a transmission path, each station generates its own frame, and each bit of the generated frame constitutes each channel, A control packet having an empty/busy table indicating whether a line channel is empty/busy is circulated on the transmission path, and when each station searches for an empty channel from the empty/busy table on the control packet, the station determines whether the line is empty/busy. Enter the used section of the line and the name of the station that uses the section above, and when the written content is found on the control packet that has made a round of the transmission path, it is determined that the used section has been recognized. Features a loop network.