JP2000196642A - Node device, network using the same and its transmission controlling method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a node device for a loop type network which stops a packet loss and quickly avoids a fault when loopback control and bypass control are performed according to the self-decision of the signal relaying part of each communication node and to provide a network using it and its transmission controlling method. SOLUTION: This node device connected to a network consisting of two loops being the current system and a spare system where data are respectively made to flow in reverse directions has parallel output buffers 165 to 168 which parallelly store the same signal as a signal to be outputted to an output port for the number of output packets, management packet inserting parts 171 to 174 which return a reception signal including the number of received packets information through the spare system, management packet separating parts 181 to 184 which detect a reception signal from a downstream node device of the current system and selectors 131 to 134 which perform loopback of output packets stored in the parallel output buffering means by using the spare system when they detect a fault due to the absence of a reception signal from the downstream node device of the current system or the shortage of the number of receiving packets.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a node device, a network using the same, and a transmission control method thereof, and more particularly, to a node device for connecting a plurality of terminal devices and a network for connecting a plurality of the node devices. And its transmission control method.

[0002]

2. Description of the Related Art As is well known, a loop type network has a plurality of nodes arranged on a ring transmission line. One node (control node) serving as a reference station sends a reference signal to the transmission line, and this reference The remaining communication nodes access the transmission path according to the signal. The signal relay unit of each communication node takes in the signal from the upstream transmission line into the node, sends the signal from the node to the downstream transmission line, and relays the signal from the upstream transmission line to the downstream transmission line. Do.

In recent years, with the increase in the amount of information, a network system in which node devices are connected by a parallel multiplex transmission line is being studied in order to cope with a high-speed and large-capacity network connecting terminal devices. The following describes a node device proposed by Yamamoto et al. And a network system using the same (Japanese Patent Application Laid-Open Nos. 8-172394 and 8-23730).
No. 6) will be described.

FIG. 6 is a configuration diagram of a node device in a conventional network, and shows an example in which terminals 551 to 558 are connected to a node device 500 via a sub-transmission line. Reference numerals 501 to 508 denote separation / insertion units that detect addresses of packets input from the parallel multiplex transmission path, separate the packets into packets to be transmitted to the terminal via the sub transmission path and packets to be input to the buffer, and Has a function of inserting a packet transmitted from a multiplexed transmission line into a packet stream input from a parallel multiplex transmission line. Reference numeral 511
518 are buffers, which are separation / insertion units 501 to 508
Has a function of temporarily storing a packet output from the switch 541 in a storage area corresponding to the output terminal of the switch 541. Reference numerals 521 to 528 and 531 to 538 denote parallel multiplex transmission lines for connecting nodes, for example, a plurality of spatially separated optical fiber transmission lines, or wavelength division on one optical fiber. And multiplexed wavelength multiplexed transmission lines.

[0005] Reference numeral 541 denotes a switch, which is controlled by the switch control unit 542 and connects a packet input to the input terminals IN1 to IN8 to an arbitrary output terminal OUT1 to OUT8. When a plurality of optical fiber transmission lines are used for the parallel multiplex transmission line, the switch 541 performs switching using a space switch or the like. When a wavelength division multiplex transmission line is used, although the configuration is slightly different from the figure, a transmission unit including a plurality of variable wavelength laser diodes and a multiplexer is connected to the wavelength division multiplex transmission line, and the wavelength division multiplex transmission line is used. A switch is configured between nodes by separating each wavelength by a demultiplexer at the receiving unit, and the transmission wavelength of the tunable laser diode is set to the wavelength λ1
Exchange is performed by setting the wavelength to an arbitrary wavelength of λ8. Code 5
Reference numeral 42 denotes a switch control unit, which controls the switch 541 according to, for example, the control pattern of FIG. Reference numeral 543 denotes a buffer control unit, which controls so as to read out the stored packets from the buffers 511 to 518 when the input terminals of the switches connected to the buffers 511 to 518 are connected to the desired output terminals. Things.

FIG. 5 is a control pattern showing an input / output connection relationship of the switch 542. The input / output connection relationship of the switch is changed by sequentially changing control addresses A1 to A8 from the switch control unit 542. Input terminals IN1 to IN
8 corresponds to buffers 511 to 518, and has an output terminal O
UT1, OUT8 (or transmission wavelengths λ1 to λ8) correspond to the storage areas 1 to 8 of the buffers 511 to 518, respectively.

FIG. 7 shows a configuration example of a network system using the node devices shown in FIG. 6, in which four node devices 601 to 604 are connected in a ring type by parallel multiplex transmission lines 605 to 608, Are connected to eight terminals via eight sub-transmission paths. The terminals 611 to 618 correspond to the terminals 551 to 558 in FIG. 5, and similarly, 621 to 628, 631 to 638, 641 to 621 to 618.
648 also corresponds to the terminals 551 to 558 in FIG.

FIG. 8 is a diagram for explaining the communication principle of this network. Reference numerals 701 to 704 denote node devices, 7.
Reference numerals 05 to 708 denote exchange switches corresponding to the switches 541, 709 to 712 denote buffers corresponding to the buffers 511 to 518, 721 to 736 denote terminals, and A, B, C and D denote ring-shaped parallel transmission lines. In order to make the explanation easy to understand, FIG.
A description will be given using a network of one terminal.

First, the communication principle of this network will be described with reference to FIG. This network has a plurality of rings A, B, C, and D, and the rings are mutually connected by exchange switches 705 to 708. Each terminal is connected to one ring transmission line among the parallel transmission lines A, B, C, and D. When communicating with a terminal connected to another ring, at least once, an arbitrary switching switch The communication is performed by exchanging with another ring. Although the position where the exchange is performed is not specified, the communication control is simplified if the node device immediately before the destination node switches to the destination transmission line and the other node devices switch to an arbitrary transmission line. In order to simplify the node device in this network, the exchange switches 705 to 708 change the input / output connection relationship to a fixed cycle according to a specific cyclic pattern regardless of the input signal, and the buffers 709 to 712
To temporarily store the input signal, and when the input / output connection relationship of the exchange switch becomes a desired relationship, the buffers 709 to 7
The exchange is performed such that the packet is read out from T.12.

For example, when communication is performed from the terminal 722 to the terminal 732, the packet output from the terminal 722 is stored in the buffer 709 of the node device 701,
For example, when the input terminal IN2 of the switch 05 is read out from the buffer when the input terminal IN2 is connected to the output terminal OUT2 and is output to the transmission line B, and is input to the buffer 710 of the node device 702 and the IN2 and OUT4 of the switch 706 are connected Is read from the buffer 710, the packet is output to the transmission path D, and the packet is sent from the node device 703 to the terminal 732.

As described above, communication is performed by switching to an arbitrary ring transmission line in each node device.

[0012]

In such a node device and a network using the same, a transmission line is constituted by a metal cable or an optical fiber cable. When a failure such as a failure is detected, or when a parallel multiplex transmission line is broken, the ring transmission line is duplicated in preparation for a failure such as a transmission line disconnection, and when a failure occurs, the location is determined and a predetermined signal relay is performed. And a loopback control that causes the transmission path to be folded back in the section. An example of this type of network configuration is proposed by Nakata et al. (Japanese Patent Application No. 9-355028).
No.).

FIG. 9 is a block diagram of a node device used in the network according to the invention of Nakata et al., Showing input / output ports PORT1 to PORT4 for connecting to four terminals via a transmission path such as a twisted pair cable, etc. The configuration of a node device having eight input terminals IN1 to IN8 and eight output terminals OUT1 to OUT8 for connecting one another via a parallel transmission line such as a multi-core optical fiber cable is shown.

In FIG. 9, reference numerals 101 to 108 denote input terminals IN1 to IN8 for inputting signals from a parallel transmission line, and include an optical receiver for converting an optical signal into an electric signal. In addition to photoelectric conversion, the optical receiver performs clock extraction, decoding of encoded signals, and AI as fault detection means.
Detection of S (A1arm Indicated Signal) and detection of interruption of light input are performed. In the fault detection means, 105 to 108 include a first fault detection means, and 101 to 104 include a second fault detection means.

Reference numerals 111 to 118 denote output terminals OUT1 to OUT8 for outputting signals to the parallel transmission lines, and include optical transmitters for converting electric signals into optical signals. The optical transmitter is not only a light-to-light converter, but also has an
S insertion is performed.

Reference numerals 121 to 124 denote input / output ports PORT1 for connecting terminals via a twisted pair cable or the like.
PORT4. Each port is composed of a circuit for transmission and a circuit for reception. Reference numerals 131 to 134 denote two-input / one-output selectors for selecting and outputting any of the signals from the management packet separation units 181 to 184 and the output signal of the switch 161. Normally, the management packet separation units 181 to 184
Is selected, and when at least one of the input terminals 101 to 104 detects AIS or optical input interruption, the signal from the switch 161 is selected.

Reference numerals 135 to 138 denote management packet separation units 18
The selector is a two-input one-output selector that selects and outputs one of signals from 5-188 and the signals from the selectors 131-134. Normally, the management packet separation unit 185
188 is selected, and the input terminals 105-10
If at least one of the detectors 8 detects AIS or interruption of the optical input, a signal from the selectors 131 to 134 is selected.

Reference numerals 145 to 148 separate packets to be sent to the terminal from the packet stream transmitted from the parallel transmission line, insert packets sent from the terminal into the packet stream, and separate and insert control packets. For the separation and insertion. The separation / insertion units 145 to 148 detect the destination addresses of the packets from the selectors 135 to 138, and output the packets to the input / output ports 121 to 124 if they are addressed to the terminal connected to each separation / insertion unit. If it is a packet, it is output to the communication controller 162, and the others are output to the buffers 151 to 154. Also, packets from the input / output ports 121 to 124 are transmitted to the selectors 135 to 135.
It is inserted into the gap of the packet flow from 138, and is output to the buffers 151 to 154 if it is a data packet, and is output to the communication control unit 162 if it is a control packet for the node. Further, it has a function of inserting a control packet from the communication control unit 162 into a gap in a packet flow.

Reference numerals 151 to 154 denote buffers for temporarily storing input data, and a storage area is divided for each output terminal of a destination. Each of the buffers 151 to 154 is divided into four destination designation storage areas (CH5 to CH8) corresponding to the output terminals OUT1 to OUT4 of the switch 161 and one destination non-designation storage area not designating a destination. The destination address is detected, and if the destination is addressed to a terminal connected to the adjacent downstream node device, the destination address is stored in a destination designation storage area corresponding to the transmission channel (CH5 to CH8) to which the terminal is connected, and In the case of (1), it is stored in the destination non-designated storage area.

Reference numeral 161 denotes a 4 × 4 switch for connecting a signal input to an input terminal to an arbitrary output terminal. The switch 161 is controlled by the exchange control unit 163, and sequentially changes the input / output connection relationship in accordance with a certain repetition pattern. For example, using the pattern of FIG.
At a certain time, the connection relation of the switch 161 is changed to the phase 1 (IN1
→ OUT1, IN2 → OUT2, IN3 → OUT3, I
Assuming that N4 → OUT4, the phase 2 (IN1 → OUT2, IN2 → OUT3, I2
N3.fwdarw.OUT4, IN4.fwdarw.OUT1, and after a long time of two packets, phase 3 (IN1.fwdarw.OUT3, IN2.fwdarw.O)
UT4, IN3 → OUT1, IN4 → OUT2), and after a long time of 3 packets, phase 4 (IN1 → OUT)
4, IN2 → OUT1, IN3 → OUT2, IN4 → O
UT3) and returns to phase 1 after a length of 4 packets, and this pattern is repeated thereafter. The phase change cycle and the repetition pattern are not limited to the above, but the phase change cycle is set to an integral multiple of the packet length.

A communication control unit 162 performs assembling / disassembling of a management packet and assembling / disassembling of a control packet. 163
Indicates connection control of the switch 161, buffers 151 to 154
This is an exchange control unit that performs read control of. Switch 16
The first connection control method is as described above.

Here, the management packet is a packet for mainly transmitting physical layer management information of the network, and the management information is assembled into a packet periodically or as needed and transmitted. The transmitted management packet is decomposed by the communication control unit 162, and the management information is extracted and used for managing the node device. The management packet is
The data is transmitted to and managed on all transmission channels between adjacent node devices. Information indicating that the management packet is a management packet is described in the header of the packet, and is distinguished from a data packet and a management packet.

The control packet is mainly used for transmitting information used in an upper layer such as control information at the time of establishing a connection, and is transmitted / received between a terminal and a node device or between an arbitrary node device. The control packet has information indicating that it is a control packet written in the header of the packet, and is distinguished from a data packet and a management packet.

The AIS signal is a signal for notifying an adjacent node device when the node device detects a failure, and forcibly converts a signal from an output terminal into an AIS signal (for example, all 1s). When an AIS signal is detected, a countermeasure against a failure such as a loopback is performed.

Reference numerals 171 to 178 denote management packet insertion units for inserting the management packets assembled by the communication control unit 162 into the packet stream. 181 to 188 separate the transmitted management packet from the transmission path, and
2 is a management packet separating unit for notifying the management packet 2.

In the node device of FIG. 9, when the failure is recognized from the management packet returned from the downstream node device, the selector 131-134 outputs the OUT of the switch 161.
1 to 4 are selected to loop back the packet from the upstream node device and stop the output from the output terminals OUT5 to OUT8. Therefore, the stop of the input to the input terminals IN5 to IN8 is recognized by the downstream node device, and the selectors 131 to 138 are selected by the selectors 135 to 138 in the downstream node device.
4 and loops back packets from downstream nodes. Therefore, a new transmission loop is formed in which the node devices are cut off on the transmission line where the failure has occurred.

However, in the above-described loop network failure avoidance system, packet loss may occur in most cases. The packet loss is solved by a retransmission request from an upper protocol layer, but increases the recovery time for returning to a normal state.

The present invention has been made in view of the above problems, and has as its object to eliminate packet loss when performing loopback control or bypass control based on the self-determination of the signal relay unit of each communication node. Another object of the present invention is to provide a loop-type network node device capable of at least minimizing and quickly avoiding a failure, a network using the same, and a transmission control method therefor.

[0029]

In order to solve the above-mentioned problems, the present invention is connected to a full-duplex reverse loop type network comprising two loops, a working system and a standby system, in which data flows in opposite directions. And a parallel output buffer means for storing the same signal as the signal output to the output port in parallel for the number of output packets, and periodically transmits the received packet number information via a standby system. Return means for returning the received signal including the received signal; management packet recognition means for detecting and interpreting the received signal from the active downstream node device; and absence of the received signal from the active downstream node device, or counting of the number of received packets. A first loop for looping back output packets stored in the parallel output buffer means using the standby system when a failure is detected from the shortage; Tsu provide click means and the network using the node apparatus and which is characterized by having a the transmission control method.

Here, the first loopback means includes:
If it is determined from the received signal from the active downstream node device that the number of received packets is insufficient, the parallel output buffer means sends out the insufficient packets through the standby system and the active downstream node. Means for transmitting all packets from the parallel output buffer means through the standby system when a reception signal from the apparatus is not obtained. An output stop unit for stopping output to the active downstream node device when a failure is detected due to no received signal from the active downstream node device or a shortage of the number of received packets; and A second loopback unit configured to loop back the backup packet to the active system when detecting an output stop in the upstream node device of the system. Also, the packet has a fixed length. The packet is an ATM cell, and the received signal is a physical cell.

[0031]

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

<Example of Configuration of Node Device According to this Embodiment> FIG. 1 is a configuration diagram of a node device used in a network according to this embodiment. The node device is connected to four terminals via a transmission line such as a twisted pair cable. Input / output port PORT for
Eight input terminals IN1 to IN1 for connecting node devices 1 to PORT4 (121 to 124) and node devices of the present invention via parallel multiplex transmission lines such as multi-core optical fiber cables.
8 (101-108) and eight output terminals OUT1-OU
4 shows a configuration example of a node device having T8 (111 to 118).

In FIG. 1, reference numerals 101 to 108 denote input terminals IN1 to IN1 for inputting signals from a parallel multiplex transmission line.
N8, which includes an optical receiver for converting an optical signal into an electric signal. The optical receiver performs not only photoelectric conversion, but also clock extraction, decoding of the encoded signal, detection of AIS (Alarm Indicated Signal) as a failure detecting means, and detection of optical input disconnection. The fault detection means is first in 105-108
, And 101 to 104 include a second fault detecting means.

Reference numerals 111 to 118 denote output terminals OUT1 to OUT8 for outputting signals to the parallel multiplex transmission line, and include an optical transmitter for converting an electric signal to an optical signal. The optical transmitter performs AIS insertion at the time of transmission path coding and failure in addition to electro-optical conversion.

Reference numerals 121 to 124 denote input / output ports PORT for connecting terminals via a twisted pair cable or the like.
1 to PORT4. Each port is composed of a circuit for transmission and a circuit for reception.

Reference numerals 131 to 134 denote management packet separating units 1
Signals from 81 to 184 and parallel output buffers 165-1
68 is a two-input one-output selector for selecting and outputting one of the buffered signals. Normally, signals from the management packet separation units 181 to 184 are selected.
When at least one of the input terminals 101 to 104 detects AIS or optical input interruption, or when packet loss is detected from information on the number of packets received within a period included in the received management packet, the parallel output buffer 165 is used.
168 is selected and looped back.

Here, if a packet loss is detected from the packet number information, the parallel output buffers 165 to 168 are output.
Out of the packets stored in the downstream nodes, only the shortage packets not received by the downstream nodes
4 to automatically restore the network so that there is no excess or deficiency of packets. Therefore, the parallel output buffer 1
65 to 168 use a memory that can be accessed randomly, but can also be realized by discarding already transmitted packets using a FIFO buffer and sending them to the selectors 131 to 134.

Reference numerals 135 to 138 denote management packet separation units 1
Signals from 85 to 188 and the selectors 131 to 134
Is a two-input, one-output selector that selects and outputs one of the signals from. Management packet separation units 185 to 188
Includes packet counting means (not shown) for counting the number of packets received in the period to be described in the management packet sent periodically. Management packet separation unit 1
Reference numerals 85 to 188 indicate the number of packets to the communication control unit 162 through a path (circuit) different from the packet, and the communication control unit 1
Reference numeral 62 describes information on the number of packets received within a cycle in a management packet as a reception signal, and notifies the upstream node device via the management packet insertion units 171 to 174.
Based on this, the upstream node device determines whether the data packet has been sent normally. Normally, signals from the management packet separation units 185 to 188 are selected, and if at least one of the input terminals 105 to 108 detects AIS or optical input disconnection, or based on packet number information received in a cycle, When a loss is detected, signals from the selectors 131 to 134 are selected and looped back.

145 to 148 separate packets to be sent to the terminal from the packet stream transmitted from the parallel transmission path, insert packets sent from the terminal into the packet stream, and separate and insert control packets. It is a separation insertion part for performing. The separation / insertion units 145 to 148 detect the destination addresses of the packets from the selectors 135 to 138, and output the packets to the ports 121 to 124 if the packets are addressed to the terminals connected to the respective separation / insertion units. If it is a packet, it is output to the communication control unit 162, and the others are output to the buffers 155 to 158. Also, packets from the input / output ports 121 to 124 are transmitted to the selectors 135 to 135.
The packet is inserted into the gap of the packet flow from 138, and is output to the buffers 151 to 154 if it is a data packet, and is output to the communication control unit 162 if it is a control packet addressed to the node. Further, it has a function of inserting a control packet from the communication control unit 162 into a gap in a packet flow.

As described above, the management packet insertion units 171 to 178 select the management packets including the management information including the reception signal from the communication control unit 162 by the selectors 131 to 13.
8 into the gap of the packet flow from
It operates as if it were output to 18.

Reference numerals 155 to 158 denote buffers for temporarily storing input data, and a storage area is divided for each destination output terminal. Reference numeral 161 denotes a 4 × 4 switch which can connect a signal input to an input terminal to an arbitrary output terminal. Each of the buffers 155 to 158 is a switch 16
It is divided into four destination designation storage areas (CH5 to CH8) corresponding to one output terminal OUT5 to OUT8 and one destination non-designation storage area not designating a destination. If the destination is for a terminal connected to an adjacent downstream node device, the destination is stored in a destination specifying storage area corresponding to the transmission channel (CH5 to CH8) to which the terminal is connected. Store in the area.

The switch 161 is controlled by the exchange control unit 163, and sequentially changes the input / output connection relationship according to a certain repetition pattern. For example, using the pattern of FIG. 4, at a certain time, the connection relation of the switch 161 is changed to the phase 1 (IN5 → OUT5, IN6 → OUT6, IN7 → O
Assuming that UT7, IN8 → OUT8, the phase 2 (IN5 → OUT6, IN6 →
OUT7, IN7 → OUT8, IN8 → OUT5), and after 54 packets long time, phase 3 (IN5 → OUT5)
7, IN6 → OUT8, IN7 → OUT5, IN8 → O
UT6), and after a long time of 81 packets, phase 4 (I
N5 → OUT8, IN6 → OUT5, IN7 → OUT
6, IN8 → OUT7), and returns to phase 1 after a length of 108 packets, and this pattern is repeated thereafter. The phase change cycle and the repetition pattern are not limited to the above, but the phase change cycle is set to an integral multiple of the packet length.

A communication control unit 162 performs assembling / disassembling of a management packet and assembling / disassembling of a control packet. Reference numeral 163 denotes connection control of the switch 161 and buffer 1
An exchange control unit that performs read control of 51 to 154.
The connection control method of the switch 161 is as described above.

In the buffer read control method, at the time of the phase 1, the destination designation storage area CH5 of the buffer 155, the destination designation storage area CH6 of the buffer 156, the destination designation storage area CH7 of the buffer 157, and the destination designation storage area CH8 of the buffer 158, respectively. The stored packet is read out, and when the phase is 2, the destination designation storage area CH6 of the buffer 155, the destination designation storage area CH7 of the buffer 156, the destination designation storage area CH8 of the buffer 157, and the buffer 15
8 is read out, and at the time of phase 3, the destination designation storage area CH7 of the buffer 155, the destination designation storage area CH8 of the buffer 156, and the destination designation storage area CH of the buffer 157 are read.
5, the packet stored in the destination designation storage area CH6 of the buffer 158 is read out, and when the phase is 4, the destination designation storage area CH8 of the buffer 155 and the buffer 1 are read out.
The stored packets are read from the 56 destination designation storage areas CH5, the destination designation storage area CH6 of the buffer 157, and the destination designation storage area CH7 of the buffer 158, respectively. If no packet is stored in the destination storage area in each phase, control is performed as if a packet was read from the destination non-designated storage area.

Here, the management packet is a packet mainly for transmitting physical layer management information of the network. In the present embodiment, the management packet also functions as a reception signal indicating the number of received packets, The management information is assembled into a packet or transmitted as needed. The sent management packet is decomposed in the communication control unit, and the management information and the number of received packets are taken out.
Used to manage node devices. Management packet separation unit 1
If the number of received packets described in the management packets separated by 81 and 184 is insufficient, it is regarded as abnormal as described above, and the selector 131 performs loopback.
To 134 are selected so as to output the packets from the parallel output buffers 165 and 168, and the missing packets are sequentially inserted into the packet stream from the switch 161 from the packet sequence arranged in the buffer in order. .

The management packet is periodically transmitted to and managed on all transmission channels between adjacent node devices. Information indicating that the management packet is a management packet is described in the header of the packet, and is distinguished from a data packet and a management packet. Normally, the management packet of the standby system includes an acknowledgment signal (Acknowledgment signal: hereinafter, abbreviated as Ack signal), returns the management packet to the transmitting (upstream) node, and determines the number of packets received during the phase period. Notice. The number of received packets to be notified is measured by the management packet separation units 185 to 188. Information indicating that the management packet is a management packet is described in the header of the packet, and is distinguished from a data packet and a management packet.

The control packet is mainly used for transmitting information used in an upper layer such as control information at the time of establishing a connection, and is transmitted and received between a terminal and a node device or between an arbitrary node device. The control packet has information indicating that it is a control packet written in the header of the packet, and is distinguished from a data packet and a management packet.

The AIS signal is a signal for notifying an adjacent node device when the node device detects a failure.
The signal from the output terminal is forcibly converted to an AIS signal (for example, all 1s). If an AIS signal is detected, loopback is performed as a measure against a failure.

Reference numerals 171 to 178 denote management packet inserting units which are management packet inserting means for inserting the management packet assembled by the communication control unit 162 into the packet stream. Reference numerals 181 to 188 denote management packet separation units which are management packet separation units for separating the transmitted management packet from the transmission path and notifying the communication control unit 162 of the number of received packets. It has packet counting means for counting.

<Example of Network Configuration According to the Present Embodiment>
FIG. 2 is a configuration example of a network using the node device according to the present embodiment.

Reference numerals 201 to 204 denote node devices according to the present embodiment, 211 to 214, 221 to 224, and 231 to 23.
Reference numerals 4,241 to 244 denote terminals. The output terminals OUT1 to OUT4 of each node device are connected to the input terminal IN1 of the adjacent node device.
To IN4 through a second parallel transmission line, and an output terminal OUT
5 to OUT8 are input terminals IN5 to IN8 of adjacent node devices.
Are connected by a first parallel transmission line to form a double ring network.

The parallel multiplex transmission line connecting the node devices is, for example, a plurality of spatially separated optical fiber transmission lines or one wavelength multiplexed optical fiber transmission line multiplexed by wavelength. Further, the node device and the terminal are connected by two pairs of uplink and downlink twisted pair cables, optical fiber cables, and the like. Here, the number of channels in the parallel transmission path is four for each of the uplink and downlink, but is not limited thereto. Further, in the present embodiment, only the data signal is transmitted through the parallel transmission path. However, for example, a channel for transmitting a clock may be provided.

The node device and the terminal are connected by two pairs of upstream and downstream twisted pair cables or optical fiber cables.

In the present embodiment, like a general ring network, a double ring configuration of an active system and a standby system is used.
During normal operation, communication is performed using the clockwise active system ring (system 0), and in the event of a failure, loopback is performed in the node device, and both the active system and the counterclockwise backup system ring (system 1) are single-layered. And continue communication. However, this network is a special two-way multiplex ring type network that performs communication using a ring in both directions because a control packet is sent to a backup ring even during normal times, and is a feature of the present embodiment when a failure occurs. Loopback without packet loss is performed, and communication is performed by forming a unidirectional multiplex ring in both systems.

The system 0 in FIG.
08 to management packet separation units 185 to 188, selectors 135 to 138, separation / insertion units 141 to 144, buffers 151 to 154, switch 161, management packet insertion units 175 to 178, and output terminals 115 to 118. The first system is defined from the input terminals 101 to 104 to the management packet separating units 181 to 184, the selectors 131 to 134,
Management packet insertion units 171 to 174 and output terminal 11
The route from 1 to 114 is referred to. System switching is input terminal 10
In steps 1 to 108, the selectors 131 to 138 are switched when the optical input is interrupted, the AIS is detected, and the number of received packets is insufficient.

<Example of Operation of Node Device and Network According to Embodiment> Details of the communication operation will be described below.

First, the flow of signals in this network will be described. Normally, each node device sets the signal from the terminal to 0.
Insert into the ring of the system. In the ring of the first system, the reception notification signal (Ac) is synchronized with the phase of the switch described later.
A management packet including a knowledgment signal (hereinafter abbreviated as an Ack signal) is returned to the transmission-side (upstream) node device to notify the number of packets received during the phase. The number of received packets is measured by the management packet separation units 185 to 188.

Next, the communication operation of this network will be described by taking as an example a case where communication is performed from the terminal 231 to the terminal 213.

The terminal 231 divides the data to be transmitted into fixed-length packets such as ATM (Asynchronous Transfer Mode) cells, and describes the destination address (destination: node device 201, PORT3) in the header. And send it out. The packet input from the input / output port 121 through the twisted pair cable is sent to the selector 171.
Then, the destination of the header is detected, and the packet is output to the separation / insertion unit 145 for insertion into the 0 system. The packet is inserted into the gap of the packet flow from the selector 135 by the separation / insertion unit 145, and is temporarily stored in the buffer 155.

Each of the buffers 155 to 158 has a switch 16
It is divided into four destination designation storage areas (CH5 to CH8) corresponding to one output terminal OUT5 to OUT8 and one destination non-designation storage area not designating a destination. If the destination is addressed to a terminal connected to an adjacent downstream node device of system 0, the destination is stored in a destination designation storage area corresponding to the transmission channel (CH5 to CH8) to which the terminal is connected. It is stored in the unspecified storage area. Here, the buffer 155
Packet (destination: node device 201, P
Since the destination of the ORT 3) is not downstream and is not adjacent, it is stored in the destination non-designated storage area, and is read out to the management packet inserting units 175 to 178 and the parallel output buffers 165 to 168 according to an instruction from the exchange control unit 163.

The exchange control unit 163 controls the opening / closing of the input / output connection of the switch 161 periodically according to a certain repetition pattern, and the packet stored from the storage area determined in each buffer by the connection relation of the switch. Is read. For example, assuming that the change period of the connection relation is 27 packets, the connection relation of the switch 161 changes to phase 1 (IN5 → OUT5, IN6 → OUT) at a certain time as shown in FIG.
6, IN7 → OUT7, IN8 → OUT8), the phase 2 (IN5 → OU) after 27 packets long time
T6, IN6 → OUT7, IN7 → OUT8, IN8 →
OUT5), and after 54 packets long, phase 3
(IN5 → OUT7, IN6 → OUT8, IN7 → OU
T5, IN8 → OUT6, and after 81 packets long time, phase 4 (IN5 → OUT8, IN6 → OUT5)
IT7 → OUT6, IN8 → OUT7) and 108
After the packet length, the phase returns to phase 1, and this pattern is repeated thereafter. The phase change cycle and the repetition pattern are not limited to the above, but the phase change cycle is set to an integral multiple of the packet length.

At the time of phase 1, the communication control unit 162 checks the Ack signal from the management packet separation units 181 to 184, and confirms that the adjacent downstream node device has received all 27 packets in the previous phase. 155, a destination designation storage area CH6 of the buffer 156, a destination designation storage area CH7 of the buffer 157,
The packets stored respectively in the destination designation storage area CH8 of the buffer 158 are read, and at the time of phase 2, the Ack signals from the management packet separation units 181 to 184 are checked, and the adjacent downstream node device receives all 27 packets at the phase 1 After confirming that the packet is stored, the packets stored from the destination designation storage area CH6 of the buffer 155, the destination designation storage area CH7 of the buffer 156, the destination designation storage area CH8 of the buffer 157, and the destination designation storage area CH5 of the buffer 158 are respectively stored. At the time of reading and phase 3, after checking the Ack signals from the management packet separation units 181 to 184 and confirming that the adjacent downstream node device has received all 27 packets in phase 2, the destination designation storage area CH7, Destination designation storage area C of buffer 156
H8, the destination storage area CH5 of the buffer 157, and the destination storage area CH6 of the buffer 158, respectively, read out the packets stored therein, and when the phase is 4, check the Ack signals from the management packet separation units 181 to 184 to check the adjacent downstream nodes. After confirming that the device has received all 27 packets in phase 3, the destination designation storage area CH8 of the buffer 155 and the destination designation storage area CH
5. The stored packets are read from the destination designation storage area CH6 of the buffer 157 and the destination designation storage area CH7 of the buffer 158, respectively. Here, when no packet is stored in the destination designation storage area in each phase, the packet is read from the destination non-designation storage area.

By controlling in this way, a packet addressed to an adjacent downstream node device is switched to a target transmission channel through the switch 161, and other packets are switched to an empty transmission channel and dispersed. Therefore, the packet previously stored in the destination non-designated storage area of the buffer 155 is read out because there is no packet in the destination specified storage area CH8 at the timing of the phase 4, for example.
Output terminal 1 through IN5 to OUT8 of switch 161
At 18, the signal is converted into an optical signal and output to the transmission channel CH8 of the parallel multiplex transmission path. The packet transmitted from the node device 203 is input to the input terminal 108 of the node device 202, converted into an electric signal, and input to the selector 138.
The selector 138 selects this signal and selects the separation / insertion unit 148
To the buffer 158. Since the destination of this packet is the transmission channel CH7 of the adjacent downstream node device, it is stored in the destination designation storage area CH7 of the buffer 157. This packet is read at the time of phase 4 and is input from IN8 of the switch 161, and
Output to UT7. The output packet is output terminal 11
The signal is converted into an optical signal at 7 and output to the transmission channel CH7 of the parallel multiplex transmission path.

The packet transmitted from the node device 202 is input to the input terminal 107 of the node device 201, converted into an electric signal, and input to the selector 137. Selector 13
7 selects this signal and outputs it to the separation / insertion unit 147.
The separation / insertion unit 147 separates this packet from the packet stream and sends it to the selector 147 because the destination of the packet is the terminal 233 connected to the separation / insertion unit 147 via a transmission line. The packet is output from the input / output port PORT3 through the selector 147 and transmitted to the terminal 233 through the twisted pair cable. Communication is performed in this manner.

Next, a communication operation when a failure such as disconnection of the parallel multiplex transmission line occurs will be described. The case where the parallel multiplex transmission line of the 0 system between the node device 202 and the node device 203 is disconnected will be described.

In this case, the input terminal 10 of the node device 203
To 1 to 104, a management packet indicating that the number of received packets is insufficient arrives from the node device 202 via the one-system parallel multiplex transmission path. Therefore, the communication control unit 162 inserts a packet not transmitted in the parallel output buffers 165 to 168 into the cell stream, and switches the switched selectors 131 to 13.
4, the output signal of the switch 161 which is the packet flow of the 0 system is selected and controlled so as to return to the 1 system (loop back). If the management packet has not arrived from the node device 202, control is performed so that all packets stored in the parallel output buffers 165 to 168 are returned to the first system. At this time, the communication control unit 162 notifies the loopback information to the output terminals 115 to 118 and stops the transmission of the optical signal. In the above description, the management packet indicating that the number of received packets is insufficient is received from the node device 202. However, the received packet number itself is received from the node device 202, and the communication control unit 162 of the node device 203 transmits the packet number. The shortage may be determined.

When the optical signals from the output terminals 115 to 118 stop, the input terminals 105 to 10
At 8, the interruption of the optical signal input is detected and notified to the communication control unit 162, the selectors 135 to 138 are switched, and the output signals of the selectors 131 to 134, which are the packet flows of the 1 system, are selected and returned to the 0 system. By doing so, the ring of the 0 system and the 1 system that does not use the faulty transmission path is reconfigured, and the communication is restarted. This is shown in FIG. 2 by arrows X and Y.
Indicated by

The types of fault detection include disconnection of the parallel multiplex transmission path, a case where the optical transmitters at the output terminals 111 to 118 have failed and the optical transmission output power has decreased, and a case where the input terminals 101 to 118 have failed.
The present invention can also be applied to a case where out of synchronization occurs due to failure of the optical receiver 108. In addition, since the maximum communication capacity is halved in the event of a failure compared to the normal state, control such as call admission control is performed by band management so that the communication capacity does not exceed the maximum value. By doing so, it is possible to take measures against a failure without having a spare line.

FIG. 3 shows that the communication control unit 162
In the case of a computer composed of M and RAM,
It is a flowchart which shows an example of a control procedure. FIG.
Shows only parts corresponding to the present invention. Selector 1
31 to 138 are connected to the management packet separating units 181 to 188 when there is no instruction from the communication control unit 162.

First, in step S2, the input disconnection from the management packet separating units 185 to 188 is checked.
If the input is disconnected, the selectors 135 to 13 are selected in step S4.
8, the output of the selectors 131 to 134 is selected, and if the input is not turned off, the management packet including the number of received packets (or the shortage) is inserted into the management packet insertion unit 171 in step S6.
Inserted at ~ 174 and returned to the upstream node device.

Next, in step S8, it is checked whether or not the management packet has been returned from the downstream node device through the management packet separation units 181 to 184. If there is no reception, the outputs of the parallel output buffers 165 to 168 are selected by the selectors 131 to 134 in step S10, the output from the output terminals OUT5 to OUT8 of the 0 system is stopped in step S12, and the parallel output buffer 165 is selected in step S14. ~ 168
Loops back all packets.

On the other hand, if a management packet has been received from the downstream node device, it is checked in step S16 whether or not the number of received packets is insufficient.
In 18, the parallel output buffer 1 is selected by the selectors 131 to 134.
The outputs from the output terminals OUT5 to OUT8 of the 0 system are stopped in step S20.
At 22, the packet not sent from the parallel output buffers 165 to 168 is looped back.

The management packet separation units 181 to 18
The reception of the management packet and the number of received packets from the management packet separation unit 8 may be started by an interrupt from the management packet separation units 181 to 188.

Further, the present invention may be applied to a system composed of a plurality of devices or an apparatus composed of one device.

Further, an object of the present invention is to supply a storage medium storing program codes of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (or CPU) of the system or apparatus.
And MPU) read and execute the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, based on the instructions of the program code, It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

When the present invention is applied to the storage medium, the storage medium stores program codes corresponding to the flowcharts described above.

[0080]

As described above, according to the present invention, when performing loopback control or bypass control based on the self-determination of the signal relay unit of each communication node, packet loss is eliminated, or at least minimized, and packet failure is prevented. It is possible to provide a loop-type network node device capable of quickly avoiding the problem, a network using the same, and a transmission control method therefor.
(Hereinafter abbreviated as an Ack signal) is returned to the transmitting node, and the number of packets received during the phase is notified to notify the disconnection of the parallel multiplex transmission line and the circuit of the input / output unit of the parallel multiplex transmission line Even if a failure occurs, communication can be performed without packet loss, so that the reliability of the network is improved.

[0081]

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration example of a node device according to an embodiment;

FIG. 2 is a diagram illustrating a configuration example of a network according to the present embodiment.

FIG. 3 is a flowchart illustrating an operation example of a communication control unit of the node device according to the present embodiment.

FIG. 4 is a diagram illustrating an example of switch control.

FIG. 5 is a diagram illustrating another example of switch control.

FIG. 6 is a diagram illustrating a configuration of a conventional node device.

FIG. 7 is a diagram showing a configuration of a conventional network.

FIG. 8 is a diagram illustrating a communication principle of a conventional network.

FIG. 9 is a diagram illustrating a configuration example of a previously proposed node device.

[Explanation of symbols]

101-108 Input terminal 121-124 I / O port 131-138, 171-174 Selector 141-148, 501-508 Separation / insertion unit 151-158, 511-518, 709-712 Buffer 161,541,705-708,801 Switches 165 to 168 Parallel output buffers 201 to 205, 601 to 604, 701 to 704, 9
01 to 904 Node device 211 to 254, 551 to 558, 611 to 648, 7
21 to 736, 911 to 948 Terminals 521 to 528, 531 to 538, 605 to 608 Parallel multiplex transmission line 542 Switch control unit 543 Buffer control unit

Continued on the front page F term (reference) 5K002 AA05 DA03 DA05 DA11 EA32 EA33 FA01 5K014 AA01 AA04 CA06 DA01 FA01 GA01 HA05 HA08 5K030 GA12 HA10 HC01 JA11 JL07 LE10 MA01 MB04 MD02 5K031 AA08 BA06 CB12 CC02 DA12 EA01 CB07 CC EB09 CC09 DD10 EE01 KK56 LL06 LL07

Claims (17)

[Claims] 1. A node device connected to a full-duplex reverse loop type network comprising two loops of a working system and a protection system in which data flows in opposite directions, respectively, wherein the signal is the same as the signal output to an output port. Output buffer means for accumulating the received packet number information in parallel for the number of output packets, a return means for periodically returning a reception signal including information on the number of received packets via a standby system, and a downstream node device of the active system A management packet recognition means for detecting and interpreting a reception signal from the server, and when a failure is detected from a lack of a reception signal from the downstream node device of the active system, or a shortage of the number of received packets, the standby system is used. A first loop-back of output packets stored in the parallel output buffer means
And a loopback means. 2. The first loop-back means, when judging that there is a shortage of the number of received packets from a signal received from the active downstream node device, the first loop-back means outputs the shortage packet from the parallel output buffer means. Means for transmitting through the standby system, and means for transmitting all packets through the standby system from the parallel output buffer when the received signal from the downstream node device of the active system is not obtained. The node device according to claim 1, wherein 3. An output stop means for stopping output to the active downstream node device when a failure is detected due to no received signal from the active downstream node device or an insufficient number of received packets. 2. The apparatus according to claim 1, further comprising a second loop-back unit configured to loop back the backup system packet to the active system when detecting an output stop in the active upstream node device. Node device. 4. The node device according to claim 1, wherein the packet has a fixed length. 5. The node device according to claim 1, wherein the packet is an ATM cell, and the received signal is a physical cell. 6. A full-duplex reverse loop type network constituted by connecting a plurality of node devices to two loops of an active system and a standby system in which data flows in opposite directions, wherein each of the node devices is A parallel output buffer means for storing the same signal as the signal output to the output port in parallel for the number of output packets, and periodically returning a reception signal including the number of reception packets via a standby system Replying means, management packet recognizing means for detecting and interpreting a signal received from the active downstream node device, and detecting a failure from absence of a received signal from the active downstream node device or shortage of the number of received packets. A first loop-back of output packets stored in the parallel output buffer means using the standby system.
And a loopback means. 7. The first loop-back means, when judging that there is a shortage of the number of received packets from a signal received from the active downstream node device, the first loop-back means outputs the shortage packet from the parallel output buffer means. Means for transmitting through the standby system, and means for transmitting all packets through the standby system from the parallel output buffer when the received signal from the downstream node device of the active system is not obtained. 7. The network according to claim 6, wherein: 8. When each of the node devices detects a failure due to no reception signal from the active downstream node device or an insufficient number of received packets, the node device outputs the output to the active downstream node device. Output stop means for stopping, and second loopback means for looping back the packet of the standby system to the active system when output stop at the upstream node device of the active system is detected. 7. The network according to claim 6, wherein: 9. The network according to claim 6, wherein the packet has a fixed length. 10. The network according to claim 6, wherein the packet is an ATM cell and the received signal is a physical cell. 11. A transmission control method in a full-duplex reverse loop type network constituted by connecting a plurality of node devices to two loops of a working system and a protection system in which data flows in opposite directions, respectively. Each of the node devices periodically returns a reception signal including the reception packet number information via the standby system, and accumulates the same signal as the signal output to the output port in parallel for the number of output packets, When a certain node device detects a failure due to no reception signal from the downstream node device of the working system or a shortage of the number of received packets of the reception signal, the certain node device performs the storage using the standby system. Loopback of the output packet thus performed to stop output to the downstream node device of the active system, and a node device downstream of the certain node device , Transmission control method detects the output stop at the certain node device, and wherein the loop back packets for the standby system to the active system. 12. In the loopback of the certain node device, when it is determined from the received signal from the active downstream node device that there is a shortage of the number of received packets, the shortage of the stored packets is compared with the stored packets. 12. The transmission according to claim 11, wherein if the signal transmitted through the protection system is not received from the downstream node device of the working system, all the stored packets are transmitted through the protection system. Control method. 13. A node device connectable to a network having a first loop and a second loop, wherein the switch outputs a signal to be output to the first loop to the first loop and stores the signal in a memory. Means, determining means for determining the state of the first loop, and output means for outputting a signal stored in the storage means to the second loop in accordance with the determination by the determining means. Node device to perform. 14. The first loop has a plurality of channels, the switch means includes a plurality of input terminals for inputting signals from each of the plurality of channels of the first loop,
A plurality of output terminals for outputting a signal to each of a plurality of channels of the first loop, and connecting each of the plurality of input terminals and each of the plurality of output terminals according to a predetermined pattern. 14. The node device according to claim 13, wherein: 15. The node device according to claim 14, wherein the determination unit determines whether there is a failure in the first loop. 16. The node device according to claim 15, wherein the determination unit performs the determination based on a control signal received from the second loop. 17. The method according to claim 1, wherein the determining unit determines whether there is a failure in the first loop based on the number of signals received by another node device that has received the signal from the switch unit, based on the control signal. 17. The node device according to claim 16, wherein: