JP4587309B2 - Path switching method in case of optical network failure - Google Patents

Description

  The present invention relates to a path switching method at the time of failure in an optical network, and more particularly to a path switching method at the time of a failure in an optical network that can quickly and efficiently switch to a backup path when a failure occurs in an active path.

  The optical signal from the client device is inserted into the line in the GMPLS optical network using an optical cross-connect device (PXC), and the currently used working line (working path) and the optical signal are not inserted. However, there is a backup line (backup path) reserved for backup. When the working path is normal, information (packets) is transmitted using the working path, and when a failure occurs in the working path, the information is switched to the backup path.

  FIG. 6 shows the simplest two-node network configuration (a) in an optical network and a path switching procedure (b) when a unidirectional (rightward) failure occurs. Two nodes 1 and 2 are interposed between the client apparatuses A and B. Node 1 (left side in the figure) includes PXC3 and WDM transmission devices (hereinafter referred to as WDM) 5 and 7, and node 2 (right side in the figure) includes PXC4 and WDM6 and 8. Here, node 1 is an initiator, node 2 is a terminator, and a working path passing through WDM 5 and 6 and a backup path passing through WDM 7 and 8 are set between them.

  The client devices A and B transmit and receive information using the active path when normal, and transmit and receive information using the backup path when a failure occurs in the active path. When a rightward failure in the direction from node 1 to node 2 occurs, light in the right direction is blocked from failure point C. A portion where no light is present when this failure occurs is indicated by a broken line in FIG.

  In FIG. 6 (b), when light in the right direction is blocked from the failure point C due to the failure of the working path, the WDM2 or PXC4 of the node 2 located downstream of the failure point C detects the light break. , Send a status (received light interruption) notification signal to PXC4 or WDM2. The PXC 4 transmits / receives a status notification signal to / from the WDM 6, and further transmits its own status (received light interruption) notification signal to the PXC 3 of the upstream node 1. When PXC3 receives the status notification signal from PXC4, PXC3 sends its status (transmitted light normal) notification signal to PXC4.

  With the above procedure, the failure point C can be identified as between the nodes 1-2, and the PXC 4 located downstream of the failure point C sends a failure notification signal to the PXC 3 of the initiator node 1. When the PXC 3 receives the failure notification signal from the PXC 4, the PXC 3 transmits an active path deletion signal and blocks light in the PXC 4 direction. When the PXC4 receives the working path deletion signal, the PXC4 blocks light in the PXC3 direction.

  After that, PXC3,4 sends a working path monitoring end signal to WDM5,6, PXC3 sends a protection path switching setting signal to PXC4, and PXC3,4, the protection path to WDM5,6 Send monitoring start signal. Finally, PXC4 sends a backup path switch completion signal to PXC3.

  The above is a case where a rightward failure in the direction from PXC3 to PXC4 occurs. When a leftward failure in the direction from PXC4 to PXC3 occurs, first, PXC3 or WDM5 of node 1 detects a light break. Subsequently, a failure notification point is identified by transmitting / receiving a status notification signal, and the active path is switched to the backup path. In this case, the PXC 3 of the initiator node 1 does not send out a failure notification signal and performs processing in its own station. FIGS. 7A and 7B show a network state and a path switching procedure when a leftward failure occurs.

  The above-described path switching procedure from the active path to the backup path is ideal. However, in reality, in the case of a rightward failure, when PXC4 receives the working path deletion signal and blocks the light in the PXC3 direction, WDM5 of node 1 detects this light break and a leftward failure has occurred. There is a possibility that a situation notification signal is sent to the PXC 3 by mistakenly determining that the PXC 3 has received this situation notification signal and further sends a status notification signal to the WDM 5 and PXC 4. Based on this status notification signal, the active path deletion signal may be sent again from the PXC 3.

  This is because WDM3 has not yet received the working path monitoring end signal when the light in the PXC3 direction is cut off, and the working path monitoring function of WDM3 is still working. FIG. 6B also shows a status notification signal (broken line portion) transmitted / received between PXC3 and WDM6, PXC4 due to light interruption in the PXC3 direction.

  Sending the failure notification signal from WDM5 and the status notification signal from PXC3 again and the working path deletion signal when the light in the PXC3 direction is blocked by the working path deletion signal is the path switching procedure when a failure occurs. There is a problem that the path switching time is increased by this extra procedure, and rapid and efficient path switching is hindered.

  Similarly, even when a leftward failure occurs, an extra procedure for transmitting / receiving an extra status notification signal between PXC4 and WDM6 and between PXC4 and PXC3 may be performed. FIG. 7B also illustrates transmission / reception (broken line portion) of an extra status notification signal between PXC4 and WDM6 and between PXC4 and PXC3.

  SUMMARY OF THE INVENTION The object of the present invention is to solve the above-mentioned problems, prevent an unnecessary procedure from being performed when switching from the active path to the backup path, and prevent an optical network failure that enables rapid and efficient path switching. It is to provide a time path switching method.

In order to solve the above problems, the present invention is, each node includes a transmission device provided on each the initiator and terminator side optical cross-connect device to the optical cross-connect device, has failed working path In the path switching method at the time of failure of the optical network that switches to the protection path and recovers from the failure when it occurs, it detects the interruption of the optical signal on the working path due to the failure itself in the transmission device on the initiator side of each node The transmission apparatus other than the transmission apparatus, from when detecting the interruption of the optical signal of the active path from the adjacent node on the initiator side at the time of the active path failure, until receiving the active path monitoring end signal from the optical cross-connect apparatus of its own node The status notification signal is not sent within the first protection time that minimizes the time required for transmission, and the terminator side transmission of each node The transmission device other than the transmission device that detects the interruption of the optical signal on the working path due to the failure itself, receives the working path deletion signal from the own node when the working path failure occurs, Minimize the time from the detection of the blocking of the optical signal on the working path from the adjacent node on the terminator side that blocks the optical signal of the path to the reception of the working path monitoring end signal from the optical cross-connect device of the local node The first feature is that the status notification signal is not transmitted within the second protection time .

Further, the present invention relates to the time required to transfer the working path deletion signal between the optical cross-connect devices of adjacent nodes as T, and the time required to transfer the working path deletion signal in each optical cross-connect device. ft, t is the time until the working path deletion signal arrives from the optical cross-connect device to the transmission device at the node, and the number of nodes from the initiator node to which the working path deletion signal is expected to be transferred to the own node is N (N When ≧ 2), the second feature is that the first protection time in the own node is set to at least T × (N−1) + ft × (N−2) + t.

Furthermore, the present invention has a third feature in that the first protection time and the second protection time can be variably set.

According to the present invention, among the transmission devices on the initiator side of each node, the transmission devices other than the transmission device that detects the interruption of the optical signal on the working path due to the failure itself are the adjacent nodes on the initiator side when the working path fails The status notification signal within the first protection time that minimizes the time from when the interruption of the optical signal of the working path from the time of detection to the reception of the working path monitoring end signal from the optical cross-connect device of the own node is detected. Is not sent. Of the transmission devices on the terminator side of each node, the transmission devices other than the transmission device that detects the interruption of the optical signal on the active path due to the failure itself will cause the active path deletion signal from the own node when the active path failure occurs. The optical path connection end signal is received from the optical cross-connect device of the local node when it detects the interruption of the optical signal of the active path from the adjacent node on the terminator side. The status notification signal is not transmitted within the second protection time that minimizes the time until reception. Therefore, the transmission device does not send out an extra status notification signal, and the optical cross-connect device does not send out an extra status notification signal or an active path deletion signal again. Since a procedure for sending an extra signal is not performed in this way, it is possible to quickly and efficiently switch paths when a failure occurs. In addition, since unnecessary fault notification is not performed at the time of path switching, the load on the control plane can be reduced.

The present invention can be applied to various optical networks assuming the number of nodes from an initiator node to its own node. Also, since the WDM failure notification protection time is considered to depend on the network configuration and the performance of the network equipment, the first protection time and the second protection time should be variably set so as to cope with it. Is preferred.

  The present invention will be described below with reference to the drawings. FIG. 1 is an explanatory diagram of the simplest two-node network configuration (a) in an optical network and a path switching procedure (b) when a unidirectional (rightward) failure occurs according to the present invention. In FIG. 1, the same or equivalent parts as in FIG.

  In the two-node network configuration of FIG. 1 (a), the active WDMs 5 and 6 have a WDM failure notification protection time. Here, the WDM failure notification protection time is a time that minimizes the time from the moment when the WDM 5 or 6 detects a light break until the working path monitoring end signal arrives.

  A light interruption due to a rightward failure between the nodes 1-2 is detected by the PXC 4 of the node 2. The PXC 4 transmits / receives a status notification signal to / from the PXC 3 of the node 1 to identify a failure between the nodes 1-2 and sends a failure notification signal to the PXC 3 of the initiator node 1. Since WDM 6 of node 2 also has a WDM failure notification protection time, even if a light interruption is detected or a status notification signal is received from PXC 4, a status notification signal is not transmitted. Upon receiving the failure notification signal from PXC4, PXC3 sends an active path deletion signal to PXC4. When the PXC4 receives the working path deletion signal, the PXC4 blocks light in the PXC3 direction.

  WDM5 detects this light interruption and may send a status (reception light interruption) notification signal if this is left as it is, but since it has a WDM failure notification protection time, at least the working path monitoring ends The status notification signal is not sent until the signal arrives. In addition, since the PXC 3 has already sent a status notification signal, it does not send a status notification signal itself even if the light is interrupted. When the PXC 3 receives the status notification signal from the WDM 5, the PXC 3 transmits the status notification signal in response thereto, but since the WDM 5 does not transmit the status notification signal, the PXC 3 does not transmit the status notification signal.

  Suppression of status notification signal transmission in WDM5 is, for example, by setting a timer with a WDM failure notification protection time set in WDM5, triggering the timer at the timing when WDM5 detects a light break, and starting timing. This can be realized by suppressing the transmission of the status notification signal during the time measurement. Although a specific example of the WDM failure notification protection time will be described later, the time depends on the network configuration and the performance of the network equipment, and it is preferable that the time can be set variably.

  Next, with reference to FIG. 1 (b), operations from the time of occurrence of a failure will be sequentially described. When light in the PXC4 direction is interrupted from the failure point C due to the failure of the active path, the PXC4 located downstream of the failure point C detects the light break, and the PXC3 of the initiator node 1 Send a notification signal. When PXC3 receives the status notification signal from PXC4, PXC3 sends its status (transmitted light normal) notification signal to PXC4.

  With the above procedure, the failure point C can be identified as between the nodes 1-2, and the PXC 4 located downstream of the failure point C sends a failure notification signal to the PXC 3 of the initiator node 1. When the PXC 3 receives the failure notification signal from the PXC 4, the PXC 3 transmits an active path deletion signal and blocks light in the PXC 4 direction. When the PXC 4 of the terminator node 2 receives the working path deletion signal, it blocks light in the PXC 3 direction.

  The WDM 5 detects this light break in the left direction, but does not transmit a status notification signal or a failure notification signal at least within the WDM failure notification protection time until the working path monitoring end signal arrives from the PXC 3.

  Thereafter, when the PXC 3 and 4 reach the WDM 5 and 6 to the active path monitoring end signal, the WDM 5 and 6 end the active path monitoring. When the active path monitoring is completed, the WDM 5 does not transmit a status notification signal or a failure notification signal even if it is outside the WDM failure notification protection time.

  Next, PXC 3 sends a protection path switching setting signal to PXC 4, and PXC 3 and 4 send a protection path monitoring start signal to WDM 7 and 8. When receiving the protection path monitoring start signal, the WDMs 7 and 8 start monitoring the protection path. Finally, PXC4 sends a backup path switch completion signal to PXC3.

  FIG. 2 is a flowchart showing the above-described path switching procedure when a failure occurs. When a failure occurs in the working path from PXC3 to PXC4 (S1), the downstream PXC4 detects the failure caused by the failure and sends a failure notification signal to the upstream PXC3 (S2). ). PXC3 receives the failure notification signal and sends an active path deletion signal to PXC4. At this time, the light in the direction from PXC3 to PXC4 is blocked (S3).

  PXC4 receives the working path deletion signal from PXC3 and deletes the working path. As a result, the light in the PXC3 direction is blocked (S4). The WDM 5 has a WDM failure notification protection time of a certain time, and even if an optical interruption is detected, the status notification signal and the failure notification signal by the active path monitoring function are not transmitted within the WDM failure notification protection time (S5).

  Thereafter, the PXC 3 sends a working path monitoring end signal to the WDM 5, and the WDM 5 stops the working path monitoring when the working path monitoring end signal arrives (S6). In addition, the PXC 4 receives the working path deletion signal and sends a working path monitoring end signal to the WDM 6. The WDM 6 stops the active path monitoring when the active path monitoring end signal arrives (S7).

  Next, the PXC 3 sends a protection path switching setting signal to the PXC 4 (S8), and further sends a protection path monitoring start signal for starting monitoring of the protection path to the WDM 7. The WDM 7 starts monitoring the protection path when the protection path monitoring start signal arrives (S9). Further, the PXC 4 receives the protection path switching setting signal and sends a protection path monitoring start signal for starting monitoring of the protection path to the WDM 8. The WDM 8 starts monitoring the protection path when the protection path monitoring start signal arrives (S10).

  Finally, the PXC 4 sends a backup path switching completion signal to the PXC 3 (S11). Thus, a series of path switching procedures from the active path to the backup path is completed.

  FIG. 2 also shows a procedure (broken line portion) that can be avoided by providing the WDM 5 with a WDM failure notification protection time. In other words, before the WDM5 receives the working path monitoring end signal, the procedure for detecting a light interruption and sending the status notification signal to the PXC3 (S12), the PXC3 receives the status notification signal from the WDM5, and accordingly The procedure (S13) of sending a status notification signal to WDM5 or PXC4 is avoided by providing WDM5 with a WDM failure notification protection time.

  The above is the operation in WDM5 when a right failure, that is, an optical interruption from PXC3 to PXC4 occurs. However, WDM6 also has a WDM failure notification protection time in the same way, so that the left failure, that is, PXC3 direction When an optical interruption occurs, it is possible to prevent the WDM 6 from sending an extra status notification signal or failure notification signal.

Next, a specific example of the WDM failure notification protection time will be described. FIG. 3 is an explanatory diagram of a WDM failure notification protection time in a two-node optical network configuration. Here, it is assumed that the PXC 3 of the initiator node 1 transmits an active path deletion signal (Path tear) and then transmits an active path monitoring end signal (CS: Not active) within time s ₁ .

When the procedure signal is transferred through the LAN or the control plane, a certain packet transfer time is required for transferring the working path deletion signal between the nodes. If this packet transfer time becomes a problem, the value of the WDM failure notification protection time is set in consideration of the delay (packet transfer time). Here, the time required to transfer the Path tear from node 1 to node 2 and T _1.

  First, as shown in FIG. 3B, if a rightward failure occurs between the nodes 1-2, the PXC 4 detects this and sends a failure notification signal to the PXC 3. The PXC3 that has received the failure notification signal sends a Path tear to the PXC4. PXC3 stops sending light in the right direction (PXC4 direction) when it sends out Path tear.

When the PXC 4 receives the Path tear, the transmission of light in the left direction (PXC 3 direction) is stopped, and the WDM 5 of the node 1 detects this light break. In this case, since WDM 5 receives CS during (s ₁ -T ₁ ) from the time of detection of light break, in order to prevent WDM 5 from sending an extra status notification signal due to light break detection, It is sufficient that the WDM _{5 has} at least (s ₁ -T ₁ ) WDM failure notification protection time. Since PXC and WDM are installed in the same station, it is also possible to trigger the WDM failure notification protection time when the PXC3 sends a Path tear. ₁ is enough.

  Since the PXC3 has already sent a status notification signal when a failure occurs, it will not send the status notification signal itself even if it is interrupted, and it will not receive a status notification signal from the WDM5. No signal is sent out.

  Next, as shown in Fig. 3 (c), if a bidirectional failure occurs between nodes 1-2, the situation between node 1 PXC3-WDM5, node 2 PXC4-WDM6, and PXC3-4 Based on the transmission / reception of the notification signal, the PXC 3 detects a bidirectional failure between the nodes 1 and 2, sends a Path tear to the PXC 4, and stops sending light in the right direction. PXC4 stops sending light in the left direction when it receives Path tear.

  Since the light to the WDMs 5 and 6 has already been turned off when a failure occurs, the WDMs 5 and 6 do not send a status notification signal again based on the transmission and reception of the Path tear. Therefore, it is not necessary to consider the WDM failure notification protection time under the assumption of a bidirectional failure in a two-node network configuration.

  Next, the WDM failure notification protection time in an optical network configuration with three or more nodes will be described. Here, in order to make the explanation easy to understand, a 6-node optical network configuration shown in FIG. 4A is assumed. However, it can be generalized not only to 6 nodes but also to an optical network configuration of 3 nodes or more.

The packet transfer time between the PXC of the node k and the PXC of the node (k + 1) is T _k (1 ≦ k ≦ 5), and the PXC of the next node (k + 1) after the PXC of the node k receives the Path tear Ft _k (2 ≦ k ≦ 5), the time required for the PXC of node k to process the path tear transfer, until CS is sent to the initiator-side WDM after the PXC of node k receives the Path tear. Let t _k (2 ≦ k ≦ 6) be the time to reach, and let s _k (1 ≦ k ≦ 5) be the time from when the PXC of node k transmits Path tear until CS reaches the terminator WDM.

  First, the WDM failure notification protection time for a unidirectional (rightward) failure will be described. As shown in FIG. 4 (b), when a rightward failure occurs between nodes 5-6 in communication using node 1 as an initiator and node 6 as a terminator, the PXC of node 6 notifies the failure to the PXC of node 1. A signal is sent, and the PXC of node 1 sends a path tear toward the PXC of node 6 and simultaneously stops sending light in the right direction. As a result, all the light in the right direction between the nodes 1 to 6 is cut off. Path tear is sequentially forwarded to nodes 2-6. When receiving the Path tear, the PXCs of the nodes 2 to 6 stop sending the light in the left direction.

  First, the WDM failure notification protection time of the initiator side WDM of the nodes 2 to 5 (the WDM shown on the left side of each node) will be described. If the light in the right direction is blocked, the initiator-side WDM of the nodes 2 to 5 may send an extra status notification signal. To eliminate this possibility, allow WDM failure notification protection time until CS arrives at the initiator side WDM of nodes 2 to 5, and even if they detect an optical interruption, the status notification signal is within the WDM failure notification protection time. Should not be sent.

The WDM failure notification protection time required by the initiator-side WDM of the node k (3 ≦ k ≦ 5) can be obtained as follows. It takes a packet transfer time T _k to transfer the Path tear to the nodes k to (k + 1) (1 ≦ k ≦ 5), and the path tear transfer processing at the node k (2 ≦ k ≦ 5) Ft _k is required, and t _k is required until CS reaches the initiator-side WDM after the PXC of node k (2 ≦ k ≦ 6) receives the Path tear, so node k (3 ≦ k ≦ 5) When the PXC of the node 1 receives the Path tear, it is expressed by the equation (1) when the PXC of the node 1 sends the Path tear, that is, after the WDM of the node k (4 ≦ k ≦ 6) detects the light break. After a time, CS arrives at the initiator-side WDM of the node k (3 ≦ k ≦ 5) after the time represented by the equation (2).

Therefore, it is sufficient that the initiator-side WDM of the node k (3 ≦ k ≦ 5) has at least the WDM failure notification protection time calculated by the equation (2).
The WDM failure notification protection time required by the initiator-side WDM of the node 2 is expressed as T ₁ + t ₂ . Here, T ₁ is a packet transfer time required for the Path tear to be transferred from the node 1 to the node 2, and t ₂ is CS to the initiator WDM after the PXC of the node 2 receives the Path tear. Is the time until

  The time required for CS to reach the initiator-side WDM of the nodes 2 to 5 after the detection of light blocking in the right direction is the maximum for the WDM of the node 5, and less than that for the WDM of the nodes 2 to 4.

  Next, the WDM failure notification protection time of the terminator-side WDM (WDM shown on the right side of each node) of the nodes 1 to 5 will be described. When the PXC of nodes 2-6 receives the transmitted path tear and stops sending the light in the left direction, the terminator-side WDM of nodes 1-5 detects the light break and sends an extra status notification signal. May be sent out. To eliminate this possibility, the WDM failure notification protection time until CS reaches the terminator side WDM of nodes 1 to 5 has a WDM failure notification protection time. Should not be sent.

The terminator-side WDM of nodes 1 to 5 detects a light break in the left direction. Since CS reaches the terminator-side WDM of node k (1 ≦ k ≦ 5) from the time of detection of optical breakage to (s _k -T _k ), the WDM failure notification protection time that node k has is (s _k -T _k ). Further, it is sufficient to WDM fault notification protection time s _k in the case of a time when the PXC node k has sent the Path tear trigger the WDM failure notification protection time.

  As shown in FIG. 4C, when a rightward failure occurs between the nodes 3-4, the PXC of the node 4 sends a failure notification signal to the PXC of the node 1. The PXC of node 1 sends out a Path tear toward the PXC of node 6 and simultaneously stops sending light in the right direction. As a result, all the light in the right direction between the nodes 1 to 6 is cut off. Path tear is sequentially forwarded to nodes 2-6. When receiving the Path tear, the PXCs of the nodes 2 to 6 stop sending the light in the left direction.

  In this case, there is a possibility that the initiator-side WDM of the nodes 2 and 3 sends an extra status notification signal. Therefore, under the assumption in this case, it is sufficient that the initiator-side WDMs of nodes 2 and 3 have a WDM failure notification protection time, and the required WDM failure notification protection time is the same as in the case of a failure between nodes 5-6. Similarly, it can be obtained by equation (2).

  Since the light to the initiator-side WDMs of the nodes 4 to 6 has already been cut off when a failure occurs, they do not send a status notification signal again based on the reception of the Path tear. Therefore, it is not necessary to consider those WDM failure notification protection times. Further, the WDM failure notification protection time of the terminator side WDM of the nodes 1 to 5 is the same as that in the case of FIG.

  Next, a WDM failure notification protection time for a bidirectional failure in a 6-node optical network configuration will be described. FIG. 5 is an explanatory diagram of the WDM failure notification protection time in this case.

  As shown in FIG. 5B, when a bidirectional failure occurs between the nodes 5-6, the PXC of the node 6 sends a failure notification signal to the PXC of the node 1, and the PXC of the node 1 sends it to the PXC of the node 6. At the same time as sending Path tear, stop sending light in the right direction. As a result, all the light in the right direction between the nodes 1 to 6 is cut off. Path tear is sequentially forwarded to nodes 2-6. When the PXC of node 6 receives the Path tear, it stops sending light in the left direction.

The time from when the initiator side WDM of the nodes 2 to 5 detects the light break in the right direction until the CS is received is expressed in the same manner as in the case of the one-way failure. Therefore, the initiator side WDM of the nodes 3 to 5 has at least the WDM failure notification protection time calculated by the expression (2), and the initiator side WDM of the node 2 has the WDM failure notification protection time of T ₁ + t ₂ You can do it. The time required for CS to reach the initiator-side WDM of the nodes 2 to 5 after the detection of light blocking in the right direction is the maximum for the WDM of the node 5, and less than that for the WDM of the nodes 2 to 4.

  Since the light to the terminator side WDM of nodes 1 to 5 has already been cut off when a failure occurs, those WDMs will not send a status notification signal again even if nodes 2 to 6 receive a Path tear. Absent. Therefore, under the assumption in this case, it is not necessary to consider the WDM failure notification protection time of the terminator side WDM of the nodes 1 to 5.

  As shown in FIG. 5 (c), when a bidirectional failure occurs between the nodes 3-4, the failure notification signal is transmitted from the PXC of the node 4 to the PXC of the node 1. The PXC of node 1 sends out a Path tear toward the PXC of node 6 and simultaneously stops sending light in the right direction. As a result, all the light in the right direction between the nodes 1 to 6 is cut off. Path tear is sequentially forwarded to nodes 2-6. The PXCs of the nodes 4 to 6 stop sending the light in the left direction when receiving the Path tear.

  In this case, there is a possibility that the initiator-side WDM of the nodes 2 and 3 sends an extra status notification signal. Therefore, it is necessary to consider the WDM failure notification protection time in those WDMs, but the WDM failure notification protection time required by the initiator-side WDM of nodes 2 and 3 is the case of a rightward failure between nodes 5-6 It can be obtained in the same way. Further, the WDM failure notification protection time of the terminator side WDM of the nodes 4 and 5 is the same as in the case of FIG.

  Since the light to the initiator side WDM of nodes 4 to 6 and the terminator side WDM of nodes 1 to 3 has already been cut off when a failure occurs, these WDMs are status notification signals anew based on the transmission or reception of path tears. Is not sent. Therefore, under the assumption in this case, it is not necessary for these WDMs to have a WDM failure notification protection time.

  Next, a specific example will be described. Here, 10 ms is the time from when the PXC of each node sends a Path tear until the CS arrives at the terminator WDM, and the time from when the PXC at each node receives the Path tear until CS arrives at the initiator WDM If the time required for path tear transfer processing at each node is 5 ms and the packet transfer time is not taken into account, as shown in Fig. 4 (b), WDM failure notification protection at the WDM on the initiator side The time may be set to at least 10 ms, 15 ms, 20 ms, and 25 ms for the nodes 2 to 5, respectively.

  If packet transfer time is taken into account, it is only necessary to add packet transfer time from node 1 to its own node, and at nodes 2 to 5, respectively, at least (10 ms + (packet transfer time from node 1 to node 2)), (15 ms + (Packet transfer time from node 1 to node 3)), (20ms + (packet transfer time from node 1 to node 4))), (25ms + (packet transfer time from node 1 to node 5)). Generally, it is at least “(5 ms × (the number of assumed nodes to the own node−2) +10 ms) + (packet transfer time from the node 1 to the own node)”.

  In addition, the WDM failure notification protection time in the terminator WDM is at least 10 ms at nodes 2 to 5 if the packet transfer time is not taken into account, and if the packet transfer time is taken into account, (10 ms-(from the own node to the next node) Packet transfer time)).

  As described above, specific examples have been described. However, in an actual optical network, it is normal that nodes serving as initiators and terminators are not fixed. In such a case, the WDM failure notification protection time for the number of nodes from the initiator assumed in various states to the own node is obtained in the same manner as described above, and the maximum time among them is set as the WDM failure notification protection time. Good.

It is an explanatory diagram of the simplest two-node network configuration (a) in an optical network and a path switching procedure (b) when a unidirectional (rightward) failure occurs according to the present invention. It is a flowchart which shows the path | route switching procedure at the time of the fault occurrence by this invention. It is explanatory drawing of the WDM failure notification protection time in the case of a two-node optical network configuration. It is explanatory drawing of the WDM failure notification protection time in the case of the unidirectional failure of 6 node optical network structure. It is explanatory drawing of the WDM failure notification protection time in the case of the bidirectional | two-way failure of 6 node optical network structure. It is a figure which shows the prior art of a 2 node optical network structure and the path switching procedure at the time of right direction failure occurrence. It is a figure which shows the prior art of a 2 node optical network structure and the path switching procedure at the time of a left direction failure occurrence.

Explanation of symbols

1, 2 ... nodes, 3, 4 ... optical cross-connect equipment (PXC), 5-8 ... transmission equipment (WDM), A, B ... client equipment

Claims (3)

Each node has an optical cross-connect device and a transmission device provided on each of the initiator side and the terminator side with respect to the optical cross-connect device . When a failure occurs in the active path, the node is switched to the standby path to recover from the failure. In the path switching method at the time of failure of the optical network to be performed,
Of the transmission devices on the initiator side of each node, the transmission devices other than the transmission device that detects the interruption of the optical signal on the active path due to the failure itself will cause the active path from the adjacent node on the initiator side to fail when the active path fails. The status notification signal is not sent within the first protection time that minimizes the time from when the optical signal cutoff is detected until the working path monitoring end signal is received from the optical cross-connect device of the own node. Of the transmission devices on the terminator side of the node, transmission devices other than the transmission device that detects the interruption of the optical signal on the active path due to the failure itself receive the active path deletion signal from its own node when the active path fails Blocking the optical signal on the working path to the own node The optical cross-connect of the own node from when the blocking of the optical signal on the working path from the adjacent node on the terminator side is detected Second disaster path switching method of the optical network, characterized in that does not transmit the status notification signal within the protection time to time from the winding device to the reception of the working path monitoring end signal a minimum. The time required to transfer the working path deletion signal between the optical cross-connect devices of adjacent nodes is T, the time required to transfer the working path deletion signal in each optical cross-connect device is ft, and the optical cross When the time until the working path deletion signal arrives from the connect device to the transmission device is t, and the number of nodes from the initiator node that is expected to transfer the working path deletion signal to its own node is N (N ≧ 2) 2. The path switching method at the time of failure of an optical network according to claim 1, wherein the first protection time in the own node is set to at least T × (N−1) + ft × (N−2) + t. 3. The path switching method for faults in an optical network according to claim 1 or 2, wherein the first protection time and the second protection time can be variably set.

Images