JP2004007802A - Communication network and fault recovery system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To build up a low-cost ring system wherein the total length of a ring is long, optical paths are efficiently contained, and an optical signal is subjected to insertion, demultiplexing, multiplexing. <P>SOLUTION: Active optical paths are assigned to active rings 101, 103 without fail and standby rings 102, 104 are used for the respective shared resources. On the occurrence of a fault in an active optical path 401, for example, occurrence of a broken fault of an optical fiber between nodes 105 and 106, a termination node 108 transmits a switching request to a source node 106 without loopback changeover so as to recover the fault by operating a switch of the source node 106 to take a bypass to a standby optical path 402. <P>COPYRIGHT: (C)2004,JPO

Description

The present invention relates to a communication network and a failure recovery method.

In the following, the term “separation” means to output a signal obtained by decomposing a signal transferred in a network node to another communication device in the own node. Insertion means that, in a network node, a signal from another communication device in the own node is multiplexed into a transmission signal and transmitted to another node. Passing means that some or all of the transmitted signals are not separated or inserted into other communication devices in the own node, and wavelengths and time slots are not replaced as they are, or are spatially connected. This means that the transmission is performed to another node by changing the wavelength or changing the time slot. In the following, an optical path is defined as a period from a time when an electric signal is converted into an optical signal at a certain node and sent to another node until the signal is converted again into an electric signal.

In order to respond to the demand for increasing the capacity of communication, in optical communication networks, means for increasing the capacity in one optical transmission line by performing wavelength multiplexing is employed. In order to operate such a network efficiently, a communication network node switches an optical signal wavelength unit, and separates and inserts an optical signal. Optical ADM ring systems that are configured and connected are being considered. As an optical ADM ring system, a four-fiber bidirectional ring and a two-fiber unidirectional ring are considered.

The # 4 fiber ring is a system having four rings constituted by fibers, and the two-fiber ring is a system having two rings constituted by fibers.

The # 4 fiber ring has been conventionally used as a bidirectional ring. A bi-directional ring means a ring that communicates with each other by clockwise and counterclockwise signals on the same route when communication between certain nodes is considered. The four rings are used as a working ring transmitting clockwise working signal light, a counterclockwise working ring, a counterclockwise spare ring for a clockwise working ring, and a spare ring for a counterclockwise working ring. . When a failure occurs in all the fibers between certain nodes, as shown in FIG. 10, the connection is switched to a spare fiber that transmits in the opposite direction at the node before the failure point (a loopback switch). Failure recovery can be performed (for example, see Non-Patent Document 1). In FIG. 10, reference numerals 1005 to 1008 denote communication nodes. Reference numeral 1021 denotes a working optical path passing through the working ring 1001, and is terminated at the node 1007 from the node 1006 through the nodes 1005 and 1008. Now, when a breakage fault occurs in the fiber between the node 1005 and the node 1008, the nodes 1005 and 1008, which are the nodes closest to the fault point, switch to the spare ring 1002 while multiplexing the wavelength-division multiplexed signal so as to be looped back (loop). Back switching), configure the bypass 1022, and perform failure recovery. Eventually, at the time of failure recovery, the optical signal passes through the route of the node 1006, the node 1005, the node 1006, the node 1007, the node 1008, and the node 1007, so that the optical transmission is performed for a longer distance than one round of the ring. At this time, in the case of a failure in the bidirectional ring, the wavelength multiplexed signal light is switched collectively for each fiber.

The # 2 fiber ring was previously used as a unidirectional ring. The unidirectional ring means, for example, that communication between nodes is normally performed by all clockwise signals. In a unidirectional ring, a 1 + 1 protection scheme is used as a failure recovery scheme (for example, see Non-Patent Document 2). FIG. 11 is a diagram for explaining failure recovery using the 1 + 1 protection method. Reference numerals 1101 and 1103 indicate working rings, and reference numerals 1102 and 1104 indicate backup rings. As shown in FIG. 11, in the 1 + 1 protection scheme, the transmitting side node (source node) previously transmits a protection optical path 1122 clockwise clockwise on the protection ring 1102 and a light transmission clockwise on the working ring 1101 in the counterclockwise direction. To the working optical path 1121 to be transmitted. The receiving node can receive the clockwise signal and the counterclockwise signal by switching the switch, so that when a failure occurs, the reception node is switched to receive the signal without the failure. By doing so, it is possible to perform failure recovery. Since both clockwise and counterclockwise signals are always flowing in communication between certain nodes, it can be said that working signals are always flowing in both clockwise and counterclockwise directions, without distinction from working signals and spare signals. .

Since the 1 + 1 protection scheme can be applied by using the 2-fiber unidirectional ring, it is possible to perform failure recovery at a very high speed. When the 1 + 1 protection method is used for a two-fiber unidirectional ring, the loopback is not performed from the viewpoint of optical transmission, so that the optical transmission distance does not become longer than one round of the ring.

In addition, there is a method of recovering a failure of a SONET ring (for example, see Non-Patent Document 3) by using a two-fiber unidirectional ring to loop back a failed section (loop back) (for example, Non-Patent Document 3). Since loopback is performed in the same manner as described for the four-fiber ring, the total transmission distance may be longer than one round of the ring.

On the other hand, by using a four-fiber bidirectional ring, it is possible to perform fault recovery to some extent at high speed (about 50 msec in the case of SONET).

通信 By using the above configuration, it is possible to configure a communication network that performs high-speed failure recovery.

A.F.Elrefaie, "Multiwavelength survivable ring network architectures" in in Proc. ICC '93, pp. 1245-1251, 1993.

H. Toba et al., “An optical FDM-based self-healing ring network employing arrayed waveguide grating filters and EDFA's with level equalizers,” IEEE J. on Select. Areas Commun. Vol. 14, Vol. 5, pp. 800. -813

T-H Wu, “Fiber Network Service Survivability,” Artech house, 1992

However, by using a two-fiber unidirectional ring (1 + 1 protection scheme), a protection path is prepared for the working path in a reverse direction on the ring in a one-to-one correspondence, and the optical signal is always transmitted. Need to be kept, the use efficiency is reduced and the cost is high.

On the other hand, if a system that performs loopback is used when recovering from a failure such as a four-fiber bidirectional ring, if a path close to one round of the ring is used as the working path, optical transmission close to two rounds is performed by loopback. Must be done. Even when a path of about half a ring is used as a working path, optical transmission for about one and a half rounds must be performed.

A problem to be solved by the present invention is to construct a ring system having a path recovery efficiency and a failure recovery function capable of setting a long overall ring length, thereby reducing the cost of a communication network. It is to be.

A first invention is a communication network, comprising: a plurality of communication node means for inserting and separating signals; and a plurality of transmission paths used as communication paths for communication between the communication node means. And a second communication node existing in the communication network from an input end of a first communication node means existing in the communication network, wherein the backup resources constituting the communication network are shared by a plurality of working signals. When a failure occurs in the communication to the output end of the means, the second communication node means detects the failure and sends a failure recovery request message to the first communication node when the failure is detected. When the first communication node receives the request message, the first communication node switches the communication path to the bypass communication path.

A second invention is the communication network according to claim 1, wherein the communication node means is an optical communication node means, the transmission path is an optical transmission path, and the communication is optical communication. I do.

A third invention is the communication network node device according to the second aspect, wherein the optical communication is wavelength-division multiplexed optical communication.

A fourth invention is a method for recovering from a failure, wherein a backup resource constituting a detour communication path exists from an input end of a first communication network node device existing in a ring network shared by a plurality of working signals. When a failure occurs in communication to the output terminal of the second communication network node device existing in the network, the second communication network node device detects the failure and performs recovery of the failure when the failure is detected. Sending a request message to the first communication network node device, wherein the first communication network node device switches the communication path to the bypass communication path when receiving the request message. I do.

A fifth invention is a failure recovery method, wherein a backup resource constituting a detour communication path is present from an input end of a first communication network node device existing in a ring network shared by a plurality of working signals. If a failure occurs in communication to the output end of the second communication network node device existing in the network, and when the second communication network node device detects the communication failure, a failure recovery request message is generated. The first communication network node device sends the message to the first communication network node device, and when the first communication network node device receives the failure recovery request message, the first communication network node device uses the switch means of the first communication network node device. It is characterized in that the communication is recovered from the failure by switching to a detour path in the opposite direction to the communication.

A sixth invention is the failure recovery method according to claim 4 or 5, wherein the communication network node device is an optical communication network node device, and the communication is optical communication. .

A seventh aspect of the present invention is the failure recovery method according to the sixth aspect, wherein the optical communication is wavelength multiplexed optical communication.

Hereinafter, the operation of the present invention will be described.

In the system described in the present invention, when a failure occurs, a shared backup resource is used to set and switch a detour in a direction opposite to the path in which the failure has occurred in the unit of an optical path to perform the recovery. There is no need to perform switching, and there is no need to perform one or more rounds of optical transmission. Further, in the present invention, although the path switching method is used, since the spare resources are shared, the path accommodation efficiency is increased. This is because, when the conventional 1 + 1 protection scheme is used, the backup path must always be operated, so that a maximum of two paths (up and down directions between certain nodes) can be used for one wavelength in one round of the ring. However, in the system described in the present invention, since the spare resource is shared among all the working resources, the number between the maximum number of adjacent nodes in one wavelength in one working ring (the number of nodes) ) Alone can accommodate the path. Since the loopback is not performed and the path accommodation efficiency is simultaneously realized, the cost of the communication network is reduced.

Also, especially in the case of a wavelength division multiplexing system, it is difficult to monitor wavelengths in units originally bundled, and it is necessary to perform management in units of wavelengths. The cost can be reduced because it can be introduced. In particular, a system having a small number of paths, such as a wavelength multiplexing system (the number of paths does not increase because there is a physical restriction on the number that can be wavelength multiplexed) or a system handling a high speed signal in which many low speed signals are multiplexed. When the present invention is applied to the present invention, the number of paths to be managed can be reduced, the management cost is reduced, and the effect is further increased.

If the present invention is applied, the fault recovery is performed without performing the loopback switch, so that the total transmission distance of the optical signal can be reduced. Therefore, when the distance that can be transmitted as light is fixed, it is possible to configure a ring having a longer overall length than a system that performs loopback. Also, unlike the 1 + 1 protection, the spare resources are not used exclusively and the spare resources are shared, so that the number of working optical paths to be operated can be increased. Accordingly, it is possible to realize a ring system having a failure recovery function having characteristics of both path accommodation efficiency and a long entire ring length, and a communication network can be constructed at low cost.

Next, embodiments of the present invention will be described with reference to the drawings.

(1) The first embodiment will be described with reference to FIG. FIG. 1 is a block diagram of an optical wavelength division multiplexing communication network according to a first embodiment of the present invention. 105 to 108 are optical communication network nodes. These nodes form a four fiber ring by connecting the fibers to form a ring topology. 101 and 103 represent working rings, and 102 and 104 represent spare rings. Each node includes a clockwise working signal processing unit (121 in node 108) for processing a clockwise working ring signal, and a counterclockwise working signal processing unit (node 108 for processing a counterclockwise working signal). Has 122). The clockwise working processor handles the signal of the working ring 101 and the signal of the counterclockwise spare ring 102 (at the time of failure). The counterclockwise working signal processing unit handles the signal of the counterclockwise working ring 103 and the signal of the clockwise backup ring 104 (when a failure occurs). Each node separates and inserts the wavelength-multiplexed demultiplexed signal. For example, from the node 105, the optical signals of 109 to 112 are wavelength division multiplexed and output. Reference numerals 109 and 110 denote optical signals wavelength-division-multiplexed and demultiplexed from the working ring 101, and reference numerals 111 and 112 denote optical signals wavelength-multiplexed and demultiplexed from the working ring 103. Optical signals 113 to 116 are inserted. 113 and 114 are optical signals to be inserted into the working ring 101, and 115 and 116 are optical signals to be inserted into the working ring 103. During the ring, two wavelengths λ1 and λ2 in the 1.5 μm band are used as the wavelength of the main signal light for transmitting data. Therefore, for example, it is possible to assign wavelengths of λ1 to 113 and 115 and wavelengths of λ2 to 114 and 116. In addition to the main signal light, a 1.3 μm band control signal light (wavelength: λs) for exchanging control signals between adjacent nodes is wavelength-multiplexed with the main signal light and transmitted. Although only the node 105 is shown for the separation signal and the insertion signal, the other nodes 106 to 108 also have the same functional configuration.

2 shows a clockwise working signal processing unit 200 (121 in FIG. 1) which is a block constituting the node 108. 201 and 205 represent external input terminals, and 204 and 208 represent external output terminals, each of which is connected to another node using an optical fiber. The optical fiber of the working ring 101 is connected from the node 105 to the external input terminal 201, and is connected from the external output terminal 208 to the node 107. The optical fiber of the spare ring 102 is connected from the node 107 to the external input terminal 205, and is connected from the external output terminal 204 to the node 105. Reference numerals 221 and 223 denote control signal separators which separate the optical signal input from the external input terminal, and transmit the wavelength-multiplexed main signal light of 1.5 μm band to the optical ADM units 209 and 210, respectively. The control signal light (λs) is input to the monitoring control devices 215 and 216. As the control signal separators 221 and 223, it is possible to use a WDM coupler that separates a wavelength in the 1.3 μm band and a wavelength in the 1.5 μm band. Reference numerals 202 and 203 denote demultiplexing output terminals that output wavelength-division-multiplexed demultiplexed optical signals (λ1 or λ2), and denote demultiplexing input terminals 206 and 207 that input one-wave optical signals (λ1 or λ2). Other network devices such as a SONET terminator and an ATM switch (for example, see TH Wu, "Fiber Network Service Survivability," Artech house, 1992) are connected. 209 and 210 are optical ADM units. The optical ADM unit 209 wavelength-multiplexes and demultiplexes the wavelength-division multiplexed light input from the external input terminal 201 and outputs it to 218 and 220, or multiplexes and outputs it to the external output terminal 208. The optical ADM unit 210 wavelength-multiplexes and demultiplexes the wavelength-division multiplexed light input from the external input terminal 205 and outputs it to 217 and 219, or multiplexes and outputs it to the external output terminal 204. Reference numerals 217 to 220 denote optical splitters, which tap a part (for example, 10% of optical power) of an optical signal output by wavelength multiplexing and demultiplexing from the optical ADM unit, and connected to the monitoring controllers 215 and 216. And outputs most of the remaining optical signals (for example, 90% optical power) to the optical switches 213 and 214.

# 211 to 214 are 2 × 1 optical switches, and a mechanical optical switch can be used. The optical switches 213 and 214 are connected to the wavelength division multiplexed outputs from the optical ADM units 209 and 210 using an optical fiber, and the optical signals input to the external input terminal 201 are wavelength division multiplexed or demultiplexed. One of the wavelength division multiplexed optical signals input to the terminal 205 is selected and output to the demultiplexing output terminals 202 and 203, respectively. Similarly, an optical signal of one wavelength is input to each of the optical switches 212 and 211, and by switching the optical switches 212 and 211, the wavelengths are multiplexed by the optical ADM units 209 and 210, respectively, and the external output terminals 208 and 204 are output. Can be selected to output to either of the above.

# 215 and 216 are monitoring controllers, which monitor the tapped optical signals, and transmit switching control signals to the optical switches 211 to 214. The monitoring controllers 215 and 216 monitor the bit error rate of the input optical signal and monitor the transmission quality of the optical signal by installing an optical receiver at the input terminal of the monitoring controller (SONET frame as the optical signal). , It is possible to monitor the bit error rate by monitoring the B1 byte; for example, TH \ Wu, {"Fiber Network Service Survivability," {Artech house, 1992). In the node 108, the clockwise working signal processing unit normally monitors the error rate of the optical signal transmitted from the external input terminal 201 to the working ring 101, and determines whether the optical signal is transmitted normally. to manage. The monitoring control unit is connected to the optical switches 211 to 214, and can switch the optical switches 211 to 214 according to information of the monitoring control unit.

Control signal multiplexers 222 and 224 are connected to the front stages of the external output terminals 204 and 208, respectively, and control signal light (1.3 μm band) transmitted from the monitoring controllers 215 and 216 to other nodes and transmitted to the other nodes. The signal light (1.5 μm band) is wavelength-multiplexed. As the control signal multiplexer, a WDM coupler can be used similarly to the control signal separators 221 and 223. By superimposing and separating the control signal light on the main signal light using the control signal separators 221 and 223 and the control signal multiplexers 222 and 224, it is possible to exchange control signals with other nodes. is there.

(4) Since the control signal light from the other node is also input to the monitoring control unit, both the switching based on the control information from the other node and the switching based on the monitoring result of the optical signal of the own node are possible.

The counterclockwise working signal processing unit can also use the same configuration as 200 in FIG. Similarly, it is possible to connect the clockwise working signal processing unit and the counterclockwise working signal processing unit to the working ring 103 and the spare ring 104. Nodes other than the node 108 can similarly constitute nodes and be connected to the optical fibers of the ring.

FIG. 3 shows blocks of the optical ADM units 209 and 210 used in FIG. Reference numeral 300 denotes an optical ADM unit. Reference numeral 301 denotes a multiplexed signal input terminal for inputting a wavelength-multiplexed signal light, and reference numeral 306 denotes a multiplexed signal output terminal for outputting a wavelength-multiplexed optical signal. Reference numerals 302 and 303 denote demultiplexed signal output terminals for wavelength multiplexing and demultiplexing the optical signal input to the multiplexed signal input terminal 301. Reference numerals 304 and 305 denote insertion signal input terminals for inputting a single-wave optical signal. Reference numeral 314 denotes a wavelength multiplexing / demultiplexing device, and 307 denotes a wavelength multiplexing / wavelength-multiplexing device. , Vol.31, no.17, pp.1464-1465.) Can be used. Reference numerals 310 and 311 denote optical gate switches, and a mechanical optical switch or a gate switch using a semiconductor optical amplifier can be used. Numerals 312 and 313 denote optical splitters that split the power of the input light into two, output one to separated output terminals 302 and 303 and output the other to optical gates 310 and 311 respectively. Reference numerals 308 and 309 denote optical couplers, which output signals obtained by combining the signal light from the insertion signal input terminal 304 and the insertion signal 305 and the outputs from the optical gates 310 and 311 respectively. Reference numeral 307 outputs wavelength multiplexed light obtained by multiplexing the outputs from the optical couplers 308 and 309. By turning on or off the optical gate 310 and the optical gate 311, the signal to be input to the wavelength division multiplexer 307 is output from the optical splitter or output from the insertion signal input terminal. It is possible to choose. In the configuration of FIG. 3, since the optical signal is branched by the optical branching devices 312 and 313, an optical signal is always output to the branch signal output terminal.

Next, the failure recovery operation when the node configuration of FIG. 2 and the network of FIG. 1 are used will be described with reference to FIGS.

FIG. 4 shows a main signal, a control signal after a failure has occurred, and an operation step in each node in the network of FIG. Hereinafter, an optical path is defined as a period from when an electric signal is converted into an optical signal at a certain node and sent to another node, until the electric signal is converted again into an electric signal. One wavelength corresponds to the optical path. Reference numeral 401 denotes a working optical path for transferring the working main signal light, which is terminated from the node 106 (source node: transmission node) to the node 108 via the node 105, and uses the wavelength of λ1. Normally, the spare ring is not used, and an optical path is set in the spare ring and used only when a failure occurs. In the spare ring, all nodes are set in advance so that all optical signals arriving from other nodes pass through as they are. This can be realized by setting the optical gates 310 and 311 in FIG. 3 to the On state during the spare ring. Now, a failure recovery operation when a break failure occurs in all the optical fibers between the nodes 106 and 105 will be described. Since the optical fiber breaks, the optical path 401 does not reach the terminal node 108. First, the monitoring controller 215 in the clockwise working signal processing unit of the node 108 detects the deterioration of the bit error rate, and A failure 401 is recognized (step 1).

When the supervisory controller detects a failure in the working optical path, the node 108 switches the optical switch 213 to select and output the optical signal from the spare ring 102 (external input terminal 205) (step 2) and switch to the source node 106. The monitoring controller is set in advance so that the request message is transmitted in the direction in which no failure has occurred using the control signal light (λS) (step 3). The control signal light can carry a destination node, an optical path name, and control contents as information. For example, it can be realized by allocating information to the position and value of a framed bit like the section overhead of SONET. For example, the first 8 bits of the framed bit string are assigned to the destination node name, the next 8 bits are assigned to the identifier of the optical path, and the next 1 bit is assigned whether or not a switching request is required. When a frame configuration in which a total of 17 bit strings are connected by the number of wavelengths is used, optical path switching request messages for the number of wavelengths can be transmitted at a time. In this case, it is possible to transmit a message amount that can cope with a single failure that an optical fiber between certain nodes is broken.

Using the control system set as described above, when the node 108 recognizes a failure in (step 1), the node 108 selects a signal from the backup ring 102 (the same wavelength as the working optical path: λ1). The optical switch 213 is switched (step 2), and sends a working optical path identifier and a switch request message (λS) to the source node (step 3).

The node 107 receives the control signal light, but transfers the control signal light to the node 106 as it is not a message addressed to the own node (step 4). When the control signal light arrives at the node 106, it is a message addressed to the own node. Therefore, the node 106 switches the optical switch corresponding to the optical switch 212 in FIG. Is transmitted to the spare ring 102 (step 5). The node 107 is not set to receive the light having the wavelength of λ1 of the spare ring 102, and is in a state where the optical signal is allowed to pass directly to another node in advance, and the node 108 is configured to receive the light of λ1 of the spare ring 102. Since the wavelength is set to be received (step 2), the optical switch 213 of the node 108 selectively outputs the optical signal of the wavelength λ1 from the protection ring 102 (formation of the protection optical path 402), and The fault is recovered by using the backup light path 402.

In the present embodiment, after step 2 (switching of the optical switch 213), step 3 (sending a switch request message to the source node) is executed, but the order of (step 2) and (step 3) is Even if the reverse is true, the present invention can be implemented without hindrance. When the time required for message transmission governs the failure recovery speed, the message is transmitted first, so that the overall failure time is reduced.

FIG. 5 shows a sequence chart of this inter-node communication and operation at the node. The vertical axis is the time axis, and the lower the time, the later the time.

FIG. 6 shows an example of a flowchart of control that the monitoring controller of each node should have to implement this sequence. Reference numeral 601 denotes a branch, which is classified depending on whether a failure of the terminal signal of the own node is detected or not. The act of recognizing the failure of the own node termination signal is defined as (step 1). A procedure 602 confirms that no failure has occurred in the own node and recognizes that the failure recovery has been completed (step 6). Reference numeral 605 denotes a procedure, which makes it possible to receive an optical signal from the spare ring by switching an optical switch (step 2). A procedure 606 transmits a control message to the source node of the failed optical path (step 3). Reference numeral 603 denotes a branch, which determines whether a control signal sent from another node is addressed to the own node. Reference numeral 607 denotes a procedure. When a control message addressed to another node arrives, the control message is directly transferred to the other node (step 4). A branch 604 determines whether the arrived control signal is a switch request addressed to the own node. Reference numeral 608 denotes a procedure, which refers to the identifier of the optical path specified by the arrived control signal, and switches to transmit the corresponding optical path to the backup ring.

(4) If such a flowchart is applied to each node, a failure recovery as shown in FIG. 4 can be performed.

In the above, the failure recovery method of the working optical path having the wavelength of λ1 has been described. However, according to the configuration and method of the present invention, any single failure in a wavelength multiplexed system can be passed through the failure unit. It is possible to perform failure recovery for all the existing optical paths (including those having different source nodes and terminal nodes). This will be described below. When a single fault occurs in a fiber or a node, a fault occurs in optical paths corresponding to the number of wavelength multiplexes. Since the protection ring is shared by the working signals of the working ring, the protection ring is not used when no failure occurs. Therefore, if the same wavelength as that of the working optical path is assigned to the spare ring that transmits a signal in the direction opposite to the transmission direction of the working ring, wavelength collision (the same wavelength is assigned to the optical path in one optical fiber and can be separated). It is possible to allocate a backup optical path without any loss. Therefore, for any single failure, the failure of all the optical paths passing therethrough can be recovered. Further, even when multiple failures occur, it is possible to cope with the situation where a route opposite to the working optical path is safe.

The method of recovering the failure of the working optical path of the working ring 101 using the spare ring 102 as a shared spare resource and the node configuration thereof have been described above. The working ring 103 (the signal in the opposite direction to the working ring 101) has been described. The same node configuration and fault recovery method can be applied to the transmission) and the backup ring 104. After the failure recovery operation, the failure point of the optical fiber is confirmed, and if the repair of the working ring 101 is completed by fusion splicing of the optical fiber, the spare resource is shared so that in preparation for the next failure, The original path is restored so as to be transmitted using the working optical path 401 without using the optical path 402.

(4) By using the first embodiment, since the failure recovery is performed without performing the loopback switching, the transmission distance of the optical signal can be reduced. Therefore, when the distance that can be transmitted as light is fixed, it is possible to configure a ring having a longer overall length than a system that performs loopback. Also, unlike the 1 + 1 protection, the spare resources are not used exclusively and the spare resources are shared, so that the number of working optical paths to be operated can be increased. In the 1 + 1 method, when the shortest route is set as the communication in the up direction, the down direction always uses the route around the same direction, so that the accommodation of the optical path becomes inefficient. For example, in the 1 + 1 method, only two optical paths can be formed per ring at one wavelength (the optical path 1122 and the optical path from the node 1107 to the node 1106 in FIG. 11). By using this configuration, for example, as shown in FIG. 7, up to four optical paths can be configured with one wavelength. In FIG. 7, reference numerals 701 to 704 denote working optical paths having a wavelength of λ1, 101 denotes a working ring, and 102 denotes a spare ring. The protection resource for the working optical paths 701 to 704 is the protection ring 102, which is shared between the working optical signals. For example, the backup optical path for the working optical path 701 uses the wavelength λ1 on the backup ring 102 in the path of node 106 → node 105 → node 108 → node 107. As a backup optical path for the working optical path 702, the wavelength λ1 can be used on the backup ring 102 in a path of node 107 → node 106 → node 105 → node 108. In the section from the node 106 to the node 105 to the node 107, the same wavelength λ1 is used and shared as a backup optical path. Since these backup optical paths are independent events (for a single failure), the backup resource λ1 in the backup ring 102 can be shared between the working optical paths 701 and 702. is there. Similarly, in the case of FIG. 7, after all, the working optical paths 701 to 704 share the wavelength λ1, which is the spare resource of the spare ring 102. The same applies to optical paths of other wavelengths. In addition, since the spare resources are shared, the optical path can be set independently for the clockwise working ring and the counterclockwise working ring in communication between certain nodes. Setting to makes efficiency higher.

Also, messaging for failure recovery only requires one round of the ring at most, so it is possible to perform failure recovery at a speed similar to the operation speed of failure recovery of a SONET 4-fiber bidirectional ring. .

In addition, in the 1 + 1 protection method, since the optical signal is always transmitted to the protection path, the protection resource is used even when no failure occurs. On the other hand, when the present configuration and method are used, when no failure occurs, a spare resource can be used, and a low-priority optical path can flow there (standby access). When a failure occurs because the priority is low, it may be used as a backup optical path for another high-priority optical path, but it is possible to configure a low-priority optical path when no failure has occurred There are advantages.

Further, in the SONET system, if the signal that bundles the paths is monitored in units of monitoring lines (units of signals in which paths between adjacent nodes are multiplexed), the quality of the path signal (for example, error rate) can be checked. Was possible. However, in a wavelength division multiplex system, it is necessary to manage wavelengths between nodes from the beginning, and it is difficult to operate a management system only by managing a bundle of optical paths. Therefore, the configuration and method of the present invention can be used in a management and monitoring system in the case of performing a fault recovery on a per optical path basis, which is more effective.

Also, in the current SONET system, 50 Mb / s is treated as a path unit, but switching in a path group unit (for example, a 2.5 Gb / s unit in which 50 Mb / s signals are bundled) is bundled. When handling is performed, the number of paths to be managed is reduced, and the effect of applying this method is further increased. Even in the case of light, since the number of wavelength multiplexes is limited to some extent due to physical restrictions, the number of paths does not increase very much, which is more effective.

Further, by using the first embodiment, even if a failure occurs, only the terminal node and the source node of the optical path need to switch to recover from the failure of the optical path, and the path may be interrupted in the optical path. Since the node only needs to transfer the switching request message sent from the terminal node to the source node, the control is simple and the failure recovery speed is high.

Next, a second embodiment will be described. In the first embodiment, the configuration and the method of the four-fiber ring have been described, but in the second embodiment, the case of the two-fiber ring will be described. The two-fiber ring has a clockwise ring and a counterclockwise ring. Assuming that four wavelengths λ1 to λ4 are wavelength-multiplexed, λ1 and λ2 are allocated to the wavelength of the working optical path and λ3 and λ4 are allocated to the wavelengths of the protection optical path in both rings. It is possible to allocate a spare resource corresponding to the working optical path configured using λ1, λ2 in the clockwise ring to λ3, λ4 of the counterclockwise ring, and to assign λ1, λ2 in the clockwise ring. It is possible to allocate the spare resources corresponding to the working optical path configured using λ3 and λ4 of the clockwise ring. Therefore, a two-fiber ring can be considered in the same way as a four-fiber ring. The resources of λ1, λ2 of the clockwise ring correspond to the working ring 101 of FIG. 1, the λ3, λ4 of the counterclockwise ring correspond to the spare ring 102 of FIG. 1, and the counterclockwise rings λ1, λ2 correspond to the working ring of FIG. When the clockwise ring λ3 and λ4 are made to correspond to the spare ring of FIG. 1, it can be understood that the same operation as the four-fiber ring described in the first embodiment can be logically performed. . Since the node configuration uses λ1 and λ2 for the working optical path and λ3 and λ4 for the backup optical path, for example, the output end of the optical switch 212 and the optical ADM are compared with the four-fiber node configuration of FIG. A wavelength converter for converting an input optical signal into λ1 is inserted between the units 209, and a wavelength converter for converting the input optical signal into λ3 between the output terminal of the optical switch 212 and the optical ADM unit 210. A wavelength converter for converting an input optical signal into λ2 is inserted between the output terminal of the optical switch 211 and the optical ADM unit 209, and the wavelength converter is inserted between the output terminal of the optical switch 211 and the optical ADM unit 210. It is necessary to insert a wavelength converter for converting the input optical signal into λ4. As a wavelength converter, use a method in which an optical signal is once converted to an electric signal using a photodiode, and then the laser signal of a desired wavelength is modulated using the electric signal and converted to another wavelength. Is possible.

を By using the second embodiment, there is an effect similar to that of the first embodiment. The difference from the first embodiment is that the number of fibers used (the number of rings) is half, so that the optical fiber laying cost accounts for a large part of the cost or is particularly effective when only two fiber rings can be constituted. Is larger.

In the second embodiment, a wavelength converter having a fixed wavelength output is inserted into the node configuration shown in FIG. 2. However, it is obvious that the present invention can be applied to a wavelength converter having a variable wavelength output. In this case, the method of allocating the backup optical path can be flexibly changed, so that it is more effective to cope with multiple failures than to use a fixed wavelength converter.

In the second embodiment, as the wavelength converter, a method of converting an optical signal into an electric signal and then converting it again into an optical signal is used. However, a wavelength converter which remains light (for example, a mutual gain of a semiconductor optical amplifier) is used. It is obvious that the present invention can be implemented even by using a modulation effect or a wavelength converter using the effect of mutual phase modulation.

Next, a third embodiment will be described as an application method of the present invention. The third embodiment is a case of a two-fiber ring as in the second embodiment, and the first ring and the second ring transmit optical signals in opposite directions. Λ1 and λ2 are used as the wavelengths for transmitting the working signal of the first ring, and the wavelength λ1 of the second ring and the wavelength of the first ring are used as the spare resources for the working optical path of the wavelength λ1 of the first ring. Λ2 of the second ring is used for the working optical path of wavelength λ2. Λ3, λ4 are used as wavelengths for transmitting the working signal of the second ring, and λ3 of the first ring and λ3 of the second ring are used as protection resources for the working optical path of the wavelength λ3 of the second ring. λ4 of the first ring is used for the working optical path of λ4. When the same wavelength is assigned to the two fiber rings as the working wavelength and the protection wavelength as the working wavelength and the protection wavelength, the wavelength converter used in the second embodiment does not need to be used. In the second embodiment, a wavelength converter is required because a wavelength λ1 is used for a working optical path and a wavelength λ3 is used for a backup optical path in a certain source node. This is because if used, the wavelength used for the working optical path and the wavelength used for the backup optical path are the same, so that there is no need for wavelength conversion.

用 い る Using the third embodiment has the same effect as that described in the second embodiment, except that the wavelength converter is not required.

In the embodiment of the present invention, the case of a fiber failure has been described. However, it is obvious that failure recovery can be performed by the same method for other failures such as a node failure.

In the embodiment of the present invention, the method of monitoring the bit error rate is used as the monitoring of the optical path. However, the monitoring may be performed by using the method of monitoring the optical power. This can be realized by installing a photodiode at the input end and monitoring its photocurrent. In addition, monitoring S / N (signal-to-noise ratio) of light can be applied. The S / N of light can be obtained by obtaining the ratio between ASE (spontaneous emission noise) and signal light.

In the embodiment of the present invention, the sequence as shown in FIG. 5 is used. However, for example, the present invention can be implemented without any trouble even if the order of step 2 and step 3 is changed.

(6) In the embodiment of the present invention, a flowchart as shown in FIG. 6 is used for controlling each node, but it is apparent that it is not always necessary to use the same one. For example, even if the order in which the branch 603 and its associated procedure 607 are grouped together and the group in which the branch 604 and its associated procedure 608 are grouped are reversed (after the branch procedure 602, the branch 604 is created first). It is obvious that the present invention can be implemented without any trouble.

In the embodiments of the present invention, a ring using an optical path in a wavelength division multiplexing system has been described. However, it is obvious that the present invention is applicable to a system in which paths such as SONET and SDH are time-multiplexed. is there. However, since the transmission distance of the optical signal can be reduced by not performing the loopback switch, it is possible to increase the ring length. Therefore, the present invention is applicable to an optical network in which an optical signal passes through a node as light. Is more effective in the case of applying (the SONET ring converts an optical signal into an electric signal for each node and reproduces the signal). Also, since the optical path is a single optical path, whether it is an optical signal of 2.5 Gb / s or an optical signal of 10 Gb / s, the optical path of 2.5 Gb / s and the optical path of 10 Gb / s Even in a system in which paths are mixed, the same node configuration and failure recovery method as in the first embodiment can be used, and the system has high flexibility.

In the embodiment of the present invention, a method using an optical path in a wavelength division multiplexing system has been described. However, the present invention is applicable to a VP (Virtual @ Path) or VC (Virtual @ Channel) of an ATM if it is a ring network. Is self-evident.

In the embodiment of the present invention, the configuration of FIG. 3 is used as the configuration of the optical ADM unit, but the configuration of FIG. 8 and the configuration of FIG. 9 can be used.

FIG. 8 shows another embodiment of the configuration shown in FIG. 3, in which a 2 × 2 optical switch is inserted between the wavelength division multiplexer 314 and the wavelength division multiplexer 307, and Terminal or a separated signal output terminal. In the configuration of FIG. 3, an optical signal is always output to the separation signal output terminal. However, in this configuration, when the distribution selection type (multicast type) is not used as the 2 × 2 optical switch, the 2 × 2 optical switch is used. It is output to the separation signal output terminal only when it is in the cross state.

FIG. 9 shows another embodiment of the configuration shown in FIG. 3, in which part of the output of the wavelength division multiplexer is directly connected to the wavelength division multiplexer, and the other part is separated signal. It is directly connected to the output end. These cannot switch the operation of separation and insertion, but can be applied to the optical ADM unit in FIG. 2 to perform the failure recovery operation of the present invention.

Even if other configurations or a combination of these configurations are used, a multiplexed signal is input, a part of the multiplexed signal is output, a part of the multiplexed signal is input to the multiplexer, and the inserted signal is input to the multiplexer. It is obvious that the present invention can be applied to any configuration that allows input.

In the embodiment of the present invention, an optical signal having a wavelength of 1.5 μm band is used for the main signal system and an optical signal having a wavelength of 1.3 μm band is used for the control signal system. However, the main signal system and the control signal system can be separated. It is obvious that the wavelength is not limited to using these wavelengths.

In the embodiment of the present invention, a frame structure is used as a method of transferring a control signal to another node, a destination node name is set in the first 8 bits, an optical path identifier is set in the next 8 bits, and a switching request is set in the next 1 bit. The presence / absence of the bit is assigned, but even if they are not the same, any bit may be assigned as long as the request for path failure recovery is transmitted to the source node. In addition, there is no need to assign information to bits, and message-oriented communication can be used. It is also possible to use packet communication, frame relay, or communication using ATM.

In the embodiment of the present invention, an optical signal having a wavelength different from that of the main signal is used as the control signal transfer means. However, it is not necessary to use a different wavelength from the main signal, and any medium that can transfer control information is used. It is self-evident. For example, the present invention can be applied to a case where control information is exchanged between nodes using a system in which a radio signal or a subcarrier is superimposed on an optical signal and transmitted, or a control signal is exchanged using a telephone line. it is obvious.

In the embodiment of the present invention, the event of detecting the failure of the own node termination signal is used as a trigger for starting the failure recovery operation, but the failure recovery operation may be started by a failure notification from another node or another network device. Obviously, the present invention can be implemented without any trouble. For example, a method may be used in which a node preceding the node terminating the optical path (wavelength: λ1) detects a failure of the optical path having the wavelength of λ1 and notifies the terminal node of the abnormality to cause a failure recovery operation. The present invention can be practiced without any trouble.

In the configuration of the present invention, when no failure has occurred, the spare ring is set to pass all optical signals, but the present invention can also be performed by performing this setting when a switch request message arrives from the terminal node. It is self-evident. However, when this method is used, the optical gate is switched after the switch request message arrives, so that the failure recovery time may be delayed.

In the embodiment of the present invention, the case has been described where the traffic between nodes is symmetrical in the upstream and downstream directions. However, when the traffic between nodes is asymmetric in the upstream and downstream directions (for example, when the downstream It is self-evident that the present invention can be applied to a system having only the same.

In the embodiment of the present invention, a system using one failure recovery system in one ring system has been described. However, the present invention can be realized by combining the configuration and method of the present invention with other systems such as the conventional 1 + 1 protection system. For example, for each wavelength, λ1 and λ2 can be used for the failure recovery method by the 1 + 1 method, and λ3 and λ4 can be used for the failure recovery according to the present invention. Further, it is not always necessary to detour in the direction opposite to the transmission direction of the working optical path. For example, if a failure occurs only in the working ring between nodes and the protection ring is safe, a detour is assigned to the shortest path on the protection ring. In this case, the switch request message is sent in both clockwise and counterclockwise directions.

In the embodiment of the present invention, a mechanical optical switch is used as the optical switches 211 to 214. However, if the optical switch satisfies performance such as crosstalk and loss, an optical switch using an electro-optic effect, a thermo-optical switch, The present invention can be implemented by an optical switch using an effect or an optical gate switch using a semiconductor optical amplifier.

In the embodiment of the present invention, an optical switch in which a signal is not distributed to another path when a certain path in the switch is made conductive is used as the optical switches 211 to 214. For example, a semiconductor gate switch is provided on the branch side of the optical coupler. It is obvious that the present invention can be applied even if a distribution selection type switch (switch capable of multicasting) having a connected configuration is used, if the multicast function is blocked by a gate.

In the embodiment of the present invention, 2 × 1 optical switches are used as the optical switches 211 to 214. However, the present invention is applicable to switches having a different size and configuration from the 2 × 1 switches. For example, it is possible to apply a 2 × 2 switch as the optical switch 212 and connect a spare network device to the input terminal of the optical switch to which the separate input terminal 207 is not connected. In addition, a distribution selection type 2 × 2 switch is used as the optical switch 213, one of the output terminals of the optical switch is connected to the separation output terminal 202, and the other is connected to the optical signal monitoring device to monitor the optical signal. Obviously, the present invention can be practiced without any problem.

The distribution-selection type optical switch used as the 1 × 2 optical switches 211 and 212 for switching the transmission side can be realized by a configuration in which an optical gate switch (a switch for passing or not transmitting light) is connected to a branch portion of the coupler. It is. In addition, it is also possible to use a configuration in which the optical gate switch is connected to one output terminal of the optical gate switch and the optical gate switch is not connected to the other output terminal. The output end of the optical switch to which the optical gate switch is connected may be connected to the spare ring, and the output end of the optical switch to which the optical gate switch is not connected may be connected to the working ring. Normally, it is necessary to prevent signal light from flowing through the spare ring, so it is necessary to connect an optical gate switch and turn on / off. However, even if an optical signal flows through the working ring, the optical switches 213 and 213 of the terminal nodes are not used. This is because it is possible to select an optical signal of either the working ring or the backup ring by using the 214.

Also, in the case of a system having n spare rings having shared spare resources for working signals, it is necessary to use (n + 1) × 1 optical switches in order to enable switching to all working rings to working rings. . It is obvious that the present invention can be applied to a general mxn switch in which a plurality of such switches are integrated.

In the embodiment of the present invention, the optical switches 211 to 214 are used as the switches on the transmitting side and the receiving side. However, the optical switches 211 to 214 are directly connected to the demultiplexed output terminal and the demultiplexed input terminal without switching, and the optical signals are electrically connected. It is obvious that the present invention can also be implemented by performing protection by an electric switch after converting into a signal.

As the electric switch, an electric switch for spatially switching, an electric switch for switching a time-division multiplexed signal obtained by time-division multiplexing, and an ATM for switching a connection established by a cell like an ATM switch. The present invention can be practiced with a switch without any trouble.

In the embodiment of the present invention, an optical coupler having a branching ratio of 10:90 is used for monitoring an optical signal. However, if there is no problem in optical level design, the optical power branching ratio and the coupling ratio are particularly limited. It is self-evident.

In the embodiment of the present invention, a case of a four-node, two-wavelength ring has been described. However, it is obvious that the present invention can be applied to a system having other numbers of nodes and wavelength multiplexing.

In the embodiment of the present invention, a configuration is used in which all optical signals can be inserted and separated. However, it is apparent that the present invention can be applied to a configuration in which insertion and separation of all wavelengths is not possible.

で は While the embodiments of the present invention are based on a wavelength multiplexed system, it is clear that the present invention can be implemented even when the number of wavelength multiplexes is one.

In the embodiment of the present invention, the case where the wavelength multiplexing technique is applied as the optical multiplexing technique has been examined. However, the present invention can be implemented by applying other multiplexing techniques such as polarization multiplexing, time multiplexing, and spatial multiplexing. It is clear. To apply the present invention to a spatial multiplexing system, a bundle of a plurality of optical fibers is treated as an optical fiber group, nodes are connected to the ring topology by the optical fiber group, and a ring constituted by the optical fiber group is connected to one. The present invention can be applied by treating it as a ring. For example, if there are four rings in the fiber group, it is possible to perform failure recovery in the same manner as in the first embodiment. If there are two rings in the fiber group, the second embodiment can be used. This is because it can be handled in the same manner as in the third embodiment.

と し て As an embodiment of the present invention, a case of two fibers and a case of four fibers have been described, but the present invention is not limited to this. For example, if a spare ring serving as a shared spare resource is increased clockwise and counterclockwise one by one from a four-fiber system and a switch used for failure recovery is a 3 × 1 switch, the present invention can be applied to a six-fiber ring. it can. Also, as described in the second and third embodiments, if a part of the band resource can be used as the working resource and the rest can be used as the spare resource, there is no need to use a two-fiber ring. The present invention is applicable to a three-fiber ring and a four-fiber ring.

If a system for transmitting an optical signal in two directions in one fiber is used, there is only one ring physically, but it can be logically regarded as two rings rotating in opposite directions. The configuration and method are applicable. When this technique is used, the present invention can be applied physically using a smaller number of rings than the number of rings described in the embodiment of the present invention.

In the embodiment of the present invention, the receiving node switches the ring to be received by using a switch. However, when a failure occurs, the optical signal of the failed one is not input to the receiving node (the source node switches to send a signal to the detour), so the receiving node: Only the optical signal from the non-failed ring is input to the signal termination node. Therefore, it is not necessary to select which ring to receive using an optical switch, and the present invention can be implemented without any trouble by using an optical coupler. Therefore, as an example of the merging means in the claims of the present invention, a coupler type such as an optical coupler for adding power or a switching type such as the optical switch described in the embodiment of the present invention can be used. is there.

In the embodiment of the present invention, an AWG is used as a wavelength multiplexer and a wavelength demultiplexer. However, a filter using a diffraction grating or a fiber Bragg grating (a filter having a periodic structure in a fiber to constitute a filter) is used. It is obvious that the present invention can be implemented without any trouble if a device having a function of multiplexing wavelengths or performing wavelength multiplexing / demultiplexing, such as a combination of the above-mentioned combinations, is used.

で は In the embodiment of the present invention, the optical amplifier is not used in the optical communication node or the optical transmission line, but it is obvious that the present invention can be implemented without any problem even in a system using the optical amplifier.

In the embodiment of the present invention, the optical communication network is described in which the optical signal passes through a node on the way without being converted into an electric signal without converting the optical signal into an electric signal. It is obvious that the present invention can be implemented without any trouble even when the device is inserted. By incorporating such a device, the length of the ring can be increased.

In the embodiment of the present invention, an optical path having no wavelength conversion is used as an optical path.However, a wavelength converter may be inserted into a ring network, and an optical path in which wavelength conversion is performed may be treated as an optical path. The present invention can be implemented without any trouble. As a wavelength converter, a method of once converting an optical signal to an electric signal and then converting it again to an optical signal using a light source of a desired wavelength, a method using mutual gain modulation, mutual phase modulation, four-wave mixing, etc. But it can be applied. By using the wavelength converter, it is possible to reuse the wavelength in the spare ring by allocating the spare optical path well (to reuse the same wavelength in the same ring), so that multiple faults such as double faults can be caused. Resistance is improved.

In the embodiment of the present invention, light was not transmitted when the failure did not occur in the spare ring.However, in order to confirm whether a failure has occurred in the transmission system using the spare ring, the occurrence of the failure has occurred. The present invention can be applied even when a method of transmitting an optical signal is used even when the method is not performed. For example, the backup ring is periodically operated so as to configure all the backup paths, the backup optical path is monitored periodically, and when a failure is detected or a switch request message is received, the backup path for monitoring is May be stopped and a method of configuring only a spare optical path for recovery from a failure may be used.

In the embodiment of the present invention, the description has been given of the recovery from a failure in one-way communication of either the left-handed or right-handed working path. However, even if both the right-handed communication and the left-handed communication fail simultaneously, The present invention is applicable. This is because, in the present invention, each of the shared spare resources is allocated independently, and a detour can be formed independently of each other.

FIG. 1 is a block diagram showing a first embodiment of the present invention. FIG. 2 is a block diagram illustrating a clockwise working signal processing unit used in FIG. 1. FIG. 3 is a block diagram illustrating an optical ADM unit used in FIG. 2. FIG. 4 is a diagram illustrating a failure recovery operation used in the first embodiment. 5 is a sequence chart illustrating a failure recovery operation used in the first embodiment. 5 is a flowchart in one node illustrating a failure recovery operation used in the first embodiment. FIG. 5 is a diagram for describing effects of the system used in the first embodiment. FIG. 4 is a block diagram showing another embodiment of FIG. 3. FIG. 4 is a block diagram showing another embodiment of FIG. 3. FIG. 9 is a block diagram showing a conventional example. FIG. 9 is a block diagram showing a conventional example.

Explanation of reference numerals

101, 103 working ring 102.104 spare ring 200 clockwise working signal processing unit 211-214 optical switch 215, 216 monitoring controller 217-220 optical splitter 310, 311 optical gate 401 working optical path 402 spare optical path 1021 working light Path 1022 Backup optical path 1121 Working optical path 1122 Backup optical path

Claims (7)

A plurality of communication node means for inserting and separating signals and a plurality of transmission paths used as communication paths for communication between the communication node means, and a spare resource constituting a bypass communication path is shared by a plurality of working signals. A ring-shaped communication network,
When a failure occurs in the communication from the input terminal of the first communication node means present in the communication network to the output terminal of the second communication node means present in the communication network,
The second communication node means detects the failure, and sends a failure recovery request message to the first communication node when the failure is detected;
The communication network according to claim 1, wherein said first communication node means switches said communication path to said bypass communication path upon receiving said request message. 2. The communication network according to claim 1, wherein said communication node means is an optical communication node means, said transmission path is an optical transmission path, and said communication is optical communication. 3. The communication network according to claim 2, wherein the optical communication is wavelength multiplexed optical communication. A second communication network node device existing in the ring network from an input end of a first communication network node device in a ring network in which spare resources constituting a detour communication path are shared by a plurality of working signals If communication to the output end of the
The second communication network node device detects the failure, and sends a failure recovery request message to the first communication network node device when the failure is detected;
The failure recovery method according to claim 1, wherein the first communication network node device switches the communication path to the bypass communication path when receiving the request message. A second communication network node device existing in the ring network from an input end of a first communication network node device in a ring network in which spare resources constituting a detour communication path are shared by a plurality of working signals If communication to the output end of the
When the second communication network node device detects the communication failure, sends a failure recovery request message to the first communication network node device;
When the first communication network node device receives the failure recovery request message, the first communication network node device switches to a detour path in the reverse direction to the communication by using a switch means of the first communication network node device. And a failure recovery method. 6. The failure recovery method according to claim 4, wherein the communication network node device is an optical communication network node device, and the communication is optical communication. 7. The method according to claim 6, wherein the optical communication is wavelength multiplexed optical communication.

Images