JP6590787B2 - Optical ring network system and redundancy method thereof - Google Patents

Description

  The present invention relates to an optical ring network system and a redundancy method thereof.

The power consumption of NE (Network Element), which is a network (NW) device, has increased rapidly in proportion to the amount of Internet traffic in recent years, and it is increasingly important to build a network (network) with low power consumption and low cost. It has become. Among them, access / metro network devices with a large number of devices occupy most of the NW power consumption. Therefore, it is important to reduce power consumption including radical architectural changes in the access / metro network.
As an approach in an optical TDM (Time Division Multiplexing) method, a bandwidth, that is, a time slot (hereinafter also referred to as TS) is set for each path between arbitrary nodes on the same wavelength of a WDM (Wavelength Division Multiplex) transmission line in a network. It has been proposed that wavelength resources can be effectively utilized and power can be saved by configuring the optical layer path handling (Add / Drop).

  Therefore, in an optical ring network system in which a master node serving as a control subject and a plurality of slave nodes serving as objects to the control subject are connected by an optical transmission line, a control method in the power-saving optical TDM is reduced in cost. It is conceivable to improve the reliability. For example, a PON (Passive Optical Network) is a P2MP (Point to Multipoint) type optical network configured by passive elements such as an optical splitter, and thus can be realized at low cost. In the PON, an OLT (Optical Line Terminal) functioning as a master node and a plurality of ONUs (Optical Network Units) functioning as slave nodes are connected by P2MP.

  Conventionally, in order to improve the reliability of PON, there is an N: 1 PON protection method that avoids communication disconnection by switching to one standby system when a failure occurs in any of the N facilities. It is known (see Non-Patent Document 1).

  In the method described in Non-Patent Document 1, the OLT includes N normal system OSUs (Optical Subscriber Units), one standby system OSU, a line switch, and an OLT control unit. The OLT control unit has a function of managing each OSU, line switching switch, and optical switch control unit. The line changeover switch built in the OLT has a normal OSU / backup OSU path changeover function and a downlink data buffer storage function. In this configuration, the OLT, in particular, the OLT control unit functions as a PON control function unit.

Yasutaka Kimura, 4 others, "Proposal of N: 1 PON protection method-Link and interface between OLT and optical switch unit-", IEICE technical report, IEICE, January 2012 19th, vol.111, no.410, CS2011-97 (2012-1), p.85-90

However, in the prior art, even if the master node has a redundant configuration inside, if the building or the like where the master node is installed suffers from a catastrophic disaster such as an earthquake, it is inevitable to stop the user service. Therefore, there is a demand for improving reliability.
Also, for example, the current method PON employs an independent synchronization method, so the master node synchronizes the time by periodically updating the time of the slave node, so when a failure occurs in the master node There is a problem that time update cannot be performed and time slot interference may occur.

  Accordingly, the present invention provides an optical ring network system and a redundancy method thereof that can solve the above-described problems, increase reliability, and prevent time slot interference when a failure occurs in a master node. Let it be an issue.

  In order to solve the above-described problems, an optical ring network system according to the present invention terminates a signal transmitted / received to / from an external device and is disposed as a control entity at a first position, and the control entity And an optical ring network system in which a plurality of slave nodes as objects are connected in a ring by an optical transmission line, wherein at least one slave node has at least a part of the function of the master node. A spare device arranged at a second position different from the first position, wherein the master node synchronizes with a reference time at its own master node and uses a synchronized transmission time commonly used by each network device at each slave node. In addition, the round trip time of the active system, which is the round trip time between the own master node and each slave node, is measured. A time synchronization function unit, a path setting processing unit that sets a transmission time of a data signal in each slave node based on a reference time in the own master node of the path and a round-trip time of the working system, and a round-trip time of the working system A round-trip time notifying unit for notifying the backup device, the backup device receiving a round-trip time of the active system from the master node, and a storage for storing the received round-trip time of the active system Between the standby device and each slave node, which is calculated from the round-trip time of the working system when the master node fails. When the synchronized transmission time used in common by each network device is periodically set to each slave node based on the round trip time of the standby system, which is the round trip time A synchronization function unit, and the time synchronization function unit of the standby device sets the transmission time set in the master node as it is for the slave node located downstream from the backup device, For the slave node located upstream from the spare device, the time is set by transmitting a control signal at a time obtained by subtracting the round trip time of the spare system from the transmission time set by the master node. It is characterized by that.

  Also, the redundancy method according to the present invention terminates a signal transmitted to and received from an external device, a master node disposed at a first position as a control subject, and a plurality of slaves as objects to the control subject A redundancy method of an optical ring network system in which nodes are connected in a ring shape by an optical transmission line, wherein at least one slave node has at least a part of functions of the master node and has the first position. Is a spare device arranged at a second position different from the master node, and the master node sets a synchronized transmission time used in common in each network device in synchronization with a reference time in its own master node in each slave node Measuring a round trip time of the active system, which is a round trip time between the own master node and each slave node; and And a step of notifying the standby device of the round trip time of the system, wherein the spare device receives the round trip time of the active system from the master node, and stores the received round trip time of the active system. A step of setting the transmission time set in the master node as it is to a slave node located downstream from the spare device when the master node fails, and upstream from the spare device. For the slave node located, the round trip time of the standby system, which is the round trip time between the standby device and each slave node, calculated from the round trip time of the active system, is set in the master node. And a step of transmitting a control signal at a time obtained by subtracting from the time.

According to the optical ring network system having such a configuration or the redundancy method of such a procedure, a standby device that functions as a slave node and performs the function of the master node when the active master node fails is the active master node. Because it is installed at different locations, even if a building with network equipment is damaged due to a catastrophic disaster such as an earthquake, the reliability of the active master node and the spare device is reduced and the reliability is reduced. A network can be provided.
Further, according to this configuration or such procedure, when the standby device functions as a slave node, the active device uses the round trip time of the active system, which is the round trip time between the active master node and each slave node. It can be acquired from the system master node, held in its own device, and shared with the active system master node. Then, the round trip time of the standby system, which is the round trip time between the standby apparatus and other slave nodes, can be calculated from the round trip time of the active system. Then, the standby device sets the transmission time set in the master node as it is for the processing of the slave node located downstream from the standby device in the annular ring. In addition, the standby apparatus sets the time for the slave node located upstream in consideration of the round trip time of the standby system as compared with the process for the slave node located downstream. Therefore, in the event of a failure of the active master node, the spare device uses the information held in the own device to substitute the time setting of the master node, and continues time synchronization by periodically updating the slave node time. be able to.

  In the optical ring network system according to the present invention, the master node calculates a round trip time of the standby system, which is a round trip time between the standby device and each of the other slave nodes, based on the round trip time of the active system. A backup device round trip time calculation unit, and a storage unit that stores the active system round trip time and the standby system round trip time, and the round trip time notifying unit includes the active system round trip time and the standby round trip time. The spare device may be notified of the round trip time of the system.

  Further, in the redundancy method according to the present invention, the master node calculates a round trip time of the standby system that is a round trip time between the standby device and each of the other slave nodes based on the round trip time of the active system. The step of notifying the spare device may be executed.

  According to the optical ring network system having such a configuration or the redundancy method of such a procedure, the master node of the active system not only has the round trip time of the active system, but also the round trip time between the standby device and each of the other slave nodes. Since the backup system round-trip time is also calculated and notified, it is possible to save the trouble of calculating the backup system round-trip time in the backup device when the active master node fails.

  In the optical ring network system according to the present invention, when the master node fails, the standby time uses the data signal transmission time set in the master node as it is by using the round trip time of the active system. Then, a path setting processing unit that sets each slave node may be provided.

  In the redundancy method according to the present invention, the master node sets a transmission time of a data signal in each slave node based on a reference time in a local master node of a path and a round trip time in the active system, and When the master node fails, the apparatus may set each slave node by using the transmission time of the data signal set in the master node as it is by using the round trip time of the active system.

  According to the optical ring network system having such a configuration or the redundancy method of such a procedure, when a failure occurs in the active master node, the spare device uses the transmission time of the data signal set in the active master node as it is. Since it can be set in each slave node, the processing load on the spare device can be reduced.

  ADVANTAGE OF THE INVENTION According to this invention, while improving reliability, the time slot interference at the time of the failure generation to a master node can be prevented.

1 is a schematic configuration diagram schematically showing an optical ring network system according to an embodiment of the present invention. It is a schematic diagram of the optical ring network system of FIG. It is a functional block diagram which shows a network device typically. 2 is a functional block diagram schematically showing a configuration example of a path setting processing unit, where (a) shows a path setting processing unit of a master node, and (b) shows a path setting processing unit of a slave node. FIG. It is a time chart explaining the example of a path configuration in an optical ring network system, (a) is a system configuration and a path, (b) is a system configuration and path route information expressed on a straight line, (c) is a time synchronization state, (D) shows main signal conduction. It is a time chart explaining the path | pass structure example in a comparative example, (a) is a system configuration | structure and a path | pass, (b) is the system configuration | structure and path | route path information represented on the straight line, (c) is a time asynchronous state, ) Indicates main signal conduction. 4A and 4B are explanatory diagrams of master node switching in an optical ring network system, where FIG. 5A shows an operation at a normal time, and FIG. 5 is a time chart when a failure occurs in the optical ring network system, where (a) is a main signal conduction before and after a failure of the master node control unit, (b) is a time setting by a standby master node, and (c) is a time setting. The main signal conduction of (d) shows the retime setting. It is a time chart before main signal conduction in an optical ring network system, where (a) is a system configuration, (b) is a system configuration represented on a straight line, (c) is an RTT measurement, and (d) is an RTT measurement value notification. , (E) shows RTT calculation value notification. It is a time chart of the operation | movement of the standby system master node in an optical ring network system, (a) is a system structure represented on the straight line, (b) is time setting, (c) is path setting, (d) is a main signal. Indicates continuity.

Hereinafter, an optical ring network system and a redundancy method thereof according to the present invention will be described in detail with reference to the drawings.
(First embodiment)
[Configuration of optical ring network system]
As shown in FIG. 1, the optical ring network system 1 is configured by a plurality of network elements (network devices, hereinafter abbreviated as NE) connected in a ring shape with a ring R formed by an optical transmission line. Here, NE # 0 is usually a master node (hereinafter referred to as MN) that becomes a controlling entity, and terminates signals transmitted to and received from external devices arranged outside the ring R. The object for this control subject is a slave node (hereinafter referred to as SN). In this optical ring network system 1, at least one slave node (SN) has at least a part of the function of a master node (MN).

  Hereinafter, NE # 0 may be referred to as the MN_0 system. Here, NE # 1, NE # 3, and NE # 4 are slave nodes (SN). The NE # 2 is normally a slave node (SN), but a master node that can act as a proxy for the MN_0 system when NE # 0 cannot function as a master node (MN). MN) spare device. Hereinafter, NE # 2 may be referred to as the MN_1 system or simply a spare device.

  Although details will be described later, each NE is sent from an optical switch (optical SW), an interface (IF) for input / output of a client signal sent from an external device such as a router (not shown), or a control device (not shown). An interface (IF) or the like for inputting control signals is provided.

In FIG. 1, the flow of the optical signal counterclockwise (counterclockwise) when the ring R is viewed from above is indicated by three types of arrows.
The thin line arrow represents the Gigabit Ethernet (registered trademark) client signal (1G). This signal is added (inserted) to the ring R at NE # 3, and dropped from the ring R at NE # 0. Thus, a path for uplink traffic from NE # 3 to NE # 0 is set. In addition, since Add / Drop itself is well-known conventionally, it is not demonstrated in particular below.

The thick line arrow represents a 100 GbE client signal (100 G), and this signal is added at NE # 0 and dropped at NE # 3 for downstream traffic from NE # 0 to NE # 3. The path is set. The uplink traffic is traffic in the direction from the SN to the MN, and the downlink traffic is traffic in the direction from the MN to the SN.
The broken arrow represents a 10 GbE client signal (10G). This signal is added by NE # 4, passes through NE # 0, and drops at NE # 1 so that NE # 4 and NE # 1. A path is set between # 1 and # 1.

  The ring R is a transmission path using an optical fiber, and assumes that one-way communication is performed. In the present embodiment, in order to enable two-way communication, the ring R as viewed from above in FIG. 1 is doubled as shown in FIG. For example, in FIG. 2, the inner ring is a counterclockwise (counterclockwise) path, and the outer ring is a clockwise (clockwise) path. In FIG. 1, a clockwise (clockwise) ring is omitted.

  As shown in FIG. 2, NE # 0 (MN_0 system) is installed at, for example, the first position in the building 2, and NE # 2 (MN_1 system) is in the building 3 different from the building 2. It is installed at the second position. Since the ring of the metro network has a circumference of, for example, several tens to several hundreds km, the second position is preferably a remote place separated from the first position by at least the order of km. By doing so, it is possible to reduce the risk that NE # 0 (MN_0 system) and NE # 2 (MN_1 system) fail at the same time in an unexpected situation such as a large-scale disaster.

[Configuration of NE]
A configuration example of the NE will be described with reference to FIG.
The NE includes an optical switch 21, optical couplers 22 and 23, a burst receiver 31, a header analyzer 32, a switch 33, an IF transmitter 34, an IF receiver 35, a switch 36, and a buffer. 37, a burst transmission unit 38, a memory 40, a time synchronization function unit 50, a path setting processing unit 60, and a timing control unit 70.

The optical switch 21 transmits (ON) or blocks (OFF) an input optical signal that is a main signal or a control signal that conducts the ring R.
The optical coupler 22 branches an optical signal input from an optical fiber, which is an optical transmission line constituting the ring R, to the optical switch 21 and the burst receiving unit 31.
The optical coupler 23 combines the optical signal output from the optical switch 21 and the optical signal transmitted in burst from the burst transmitter 38. The burst signal is a signal with a difference in light intensity.

The burst receiving unit 31 electrically converts an optical signal transmitted from another network device (NE) and branched by the optical coupler 22. The converted electrical signal is input to the header analysis unit 32.
The header analysis unit 32 analyzes the header of the input signal and determines whether the input signal is a control signal or a main signal. The control signal is sent to the path setting processing unit 60, and the main signal is sent to the switch unit 33.
The switch part 33 is a switch means similar to normal electric SW (L2-SW). The switch unit 33 transfers the packet data to the external device side such as a router in accordance with a table set in advance.

The IF transmission unit 34 is an Ethernet (registered trademark) interface, converts an electrical signal sent from the switch unit 33 into an optical signal, and sends the optical signal to an external device such as a router.
The IF receiver 35 is an Ethernet (registered trademark) interface, terminates a client signal from an external device such as a router, electrically converts an optical signal, which is a client signal, and sends the optical signal to the switch unit 36.

  The switch part 36 is the same switch means as normal electric SW (L2-SW). The switch unit 36 transfers the packet data to the ring R side according to a preset table.

  The buffer 37 accumulates data (client signal) input from an external device such as a router. Data is transmitted from the buffer 37 to the burst transmitter 38. Further, the buffer 37 transmits the control signal input from the path setting processing unit 60 from the burst transmission unit 38 to another network device (NE).

  The burst transmitter 38 converts a data signal and a control signal to be transmitted to another network device (NE) from an electric signal to an optical signal, and transmits it as an optical burst signal in accordance with an instruction from the timing controller 70. The optical burst signal is transmitted to the ring R which is an optical transmission path via the optical coupler 23.

The burst reception unit 31, header analysis unit 32, switch unit 33, IF transmission unit 34, IF reception unit 35, switch unit 36, buffer 37, and burst transmission unit 38 described above are routers or the like. A client signal processing unit for performing input / output processing of a client signal sent from the external device.
In FIG. 3, the optical switch 21 corresponding to the clockwise (clockwise) ring, the optical couplers 22 and 23, and the client signal processing unit are omitted, but each NE also has these configurations. Yes.

  Among the NE components shown in FIG. 3, the memory 40, the time synchronization function unit 50, the path setting processing unit 60, and the timing control unit 70 are collectively referred to as a master node control unit or MN control hereinafter. Also called a department. Then, among the units constituting the MN control unit, when the functions differ depending on the master node (MN), the slave node (SN), and the standby device, the function corresponding to the device will be described.

The memory 40 is a storage unit that holds path setting information, and is composed of, for example, a general semiconductor memory. Here, the path setting information includes an Add time and a time slot time for each time slot in the network device (NE). The time slot time is a time slot time corresponding to a set bandwidth for each path and does not include a so-called guard time.
The memory 40 of the master node (MN) stores path setting information of the MN and path setting information of each SN.
The memory 40 of the slave node (SN) stores path setting information of the SN.
The memory 40 of the spare device can store other information notified from the MN (NE # 0) in addition to the path setting information of the spare device.

  The time synchronization function unit 50 of the master node (MN) sets the synchronized transmission time used in common in each NE in synchronization with the reference time in the self MN in each SN. The time synchronization function unit 50 increments the reference value (reference time) set in the MN according to a clock from an internal clock unit (not shown). The incremented value (counter value) represents the current time of the MN. In time adjustment with the SN, a time stamp of the current time of the MN is added to the control signal. The time synchronization function unit 50 of the master node (MN) performs time synchronization periodically. The reason is to correct the time synchronization shift due to the clock deviation of each SN. For this reason, transmission times are periodically transmitted according to a schedule at regular intervals within a range that does not cause time slot interference due to clock deviation.

  The time synchronization function unit 50 of the master node (MN) measures the round-trip time (RTT) of the active system, which is the round-trip time between the own MN and each SN. The time synchronization function unit 50 measures the round trip time between each SN from the time stamp value described in the control signal from each SN and the counter value in the time synchronization function unit 50. The round trip time measured by MN (NE # 0) is expressed as the round trip time (RTT_0) of the active system.

  The time synchronization function unit 50 of the slave node (SN) sets the transmission time synchronized with the MN (NE # 0). The time synchronization function unit 50 receives the reference value (reference time) set by the MN, and increments the reference value according to a clock from an internal clock unit (not shown). Further, when receiving the control signal from the MN, the time synchronization function unit 50 performs time synchronization and then transmits a control signal to which a time stamp of the transmission time of the control signal addressed to the MN is added from the SN.

The time synchronization function unit 50 of the standby device (NE # 2, MN_1 system) sets the transmission time synchronized with the MN (NE # 0), and at the time of failure of the MN (NE # 0), the active system Based on the round trip time (RTT_1) of the standby system, which is the round trip time between the standby device (NE # 2) and each SN, calculated from the round trip time (RTT_0) of The transmission time is periodically set to each SN. Note that the time synchronization function unit 50 of the standby apparatus performs time synchronization periodically in the same manner for the same reason as the time synchronization function unit 50 of the MN.
The time synchronization function unit 50 uses the transmission time set in the MN (NE # 0) as it is for the SN located downstream from the spare device (NE # 2), and sets the spare device. A control signal is transmitted to the SN located upstream from (NE # 2) at the time obtained by subtracting the round trip time (RTT_1) of the standby system from the transmission time set by MN (NE # 0). To set the time.

The path setting processing unit 60 sets path setting information based on a control signal between network devices (NE) and stores it in the memory 40.
In the present embodiment, as an example, as illustrated in FIG. 4A, the MN path setting processing unit 60A includes a route information acquisition unit 61, a path setting information generation unit 62, a path setting information update unit 63, Path setting information notifying means 64, spare device RTT calculating means (spare device round trip time calculating section) 65, and RTT notifying means (round trip time notifying section) 66 are provided.

  The route information acquisition means 61 acquires path route information from a control device (not shown) arranged outside the MN via a control interface (not shown in FIG. 2). The path route information includes NE information for performing Add. In the path route information, the operator of the control device can set a desired route. In this embodiment, it is assumed that the path route information is stored in advance in the MN before setting a path. Note that this path route information may be input and updated from the outside each time it is processed.

  The path setting information generating unit 62 of the MN (NE # 0) configures a time slot (TS) for each path set between arbitrary NEs on the same wavelength, and a reference in the path's own MN (NE # 0) Based on the time and the round trip time (RTT_0) of the active system, the transmission time of the data signal in each SN is set as path setting information.

The path setting information updating unit 63 sets the path setting information generated for the MN in the memory 40. When the path is updated, the path setting information held in the memory 40 is updated.
The path setting information notifying unit 64 notifies a corresponding SN of a control signal including path setting information generated for each SN.

The spare device RTT calculation means 65 calculates the round trip time (RTT_1) of the spare system, which is the round trip time between the spare device (NE # 2) and each other SN, based on the round trip time (RTT_0) of the active system. To do.
The RTT notification means 66 notifies the standby apparatus (NE # 2) of the round trip time (RTT_0) of the active system. In the present embodiment, the RTT notification unit 66 also has a function of notifying the standby system (NE # 2) of the standby system round-trip time (RTT_1).

In the present embodiment, the round trip time (RTT_0) of the active system and the round trip time (RTT_1) of the standby system are stored in the memory 40.
Further, the path setting information generation unit 62 of the standby device (NE # 2) is set by the MN (NE # 0) by using the working round-trip time (RTT_0) when the MN (NE # 0) fails. The transmission time of the received data signal is used as it is and set to each SN.

As an example, the path setting processing unit 60B of the slave node (SN) includes a path setting information receiving unit (path setting information receiving unit) 67 and a path setting information updating unit 68, as shown in FIG. I have.
The path setting information receiving unit 67 receives path setting information notified from the MN (NE # 0) by a control signal.

  The path setting information update unit 68 sets the received path setting information in the memory 40.

  As described above, when the MN path setting processing unit 60A has the configuration shown in FIG. 4A and the SN path setting processing unit 60B has the configuration shown in FIG. The path setting processing unit 60 includes a route information acquisition unit 61, a path setting information generation unit 62, a path setting information update unit 63, a path setting information notification unit 64, a path setting information reception unit 67, and a path setting information update. Means 68.

The standby unit path setting information receiving means 67 also has a function of receiving the working round-trip time (RTT_0) and the standby round-trip time (RTT_1) from the MN (NE # 0).
The path setting information update unit 68 of the standby apparatus also has a function of setting in the memory 40 the round trip time (RTT_0) of the active system and the round trip time (RTT_1) of the standby system acquired from the MN (NE # 0). Note that the memory 40 of the standby device also has a function of storing the round trip time (RTT_0) of the active system received by the path setting information receiving unit 67.

The timing control unit 70 performs control for transmitting (ON) or blocking (OFF) the optical switch 21 based on the time set by the path setting information.
The timing control unit 70 compares the counter value (current time) in the time synchronization function unit 50 with the timing value (allocation time) described in the path setting information according to the set path setting information, and the burst transmission unit 38. The TS transmission operation is controlled.

Hereinafter, the redundancy method in the optical ring network system 1 according to the embodiment of the present invention will be described in comparison with the current PON system.
[Assumption]
First, the common premise of this embodiment and a comparative example is demonstrated with reference to FIG.
In the system configuration shown on a straight line in FIG. 5B, NE # 0 (MN) is arranged at the left end, and NEs are arranged in order counterclockwise (counterclockwise) on the ring in FIG. 5A. It is. Note that NE # 0 at the right end is the same MN as NE # 0 at the left end.

In the example shown in FIG. 5B, in an optical signal having a predetermined wavelength, for TS # 1, which is a time slot, data from an external device is added at NE # 4 and dropped at NE # 3. Assume path route information.
Further, path route information is assumed such that data from an external device is added at NE # 3 and dropped at NE # 2.
Further, path route information is assumed such that data from an external device is added at NE # 2 and dropped at NE # 1. The above is the SN-SN path.
Furthermore, as the path between MN and SN, path route information is assumed in which data from an external device is added at NE # 1 and dropped at NE # 0.

Also, it is assumed that TS # 2 of the same wavelength has path route information in which data from an external device is added at NE # 2 and dropped at NE # 0 as an MN-SN path.
Further, it is assumed that TS # 3 of the same wavelength has path route information in which data from an external device is added at NE # 3 and dropped at NE # 1 as an SN-SN path.

In order to set the path of the path information shown in FIG. 5B, the time setting and the RTT measurement are performed in advance between NE # 0 (MN) and another NE (SN), as will be described later in detail. There is a need.
As shown in FIG. 5C, NE # 0 (MN) sets the current time to the reference time of NE # 0 (step S100), and sends a control signal including the time stamp of this reference time to NE # 1. The time at which NE # 1 receives this control signal is set as the reference time for NE # 1 (step S101).

Similarly, NE # 0 (MN) transmits a control signal including a time stamp of the reference time to NE # 2, NE # 3, NE # 4, thereby NE # 2, NE # 3, NE # 4. Are set (steps S102, S103, S104).
Here, although the transmission time of the control signal to each NE is actually different, in FIG. 5C, it is indicated that each NE's timer (indicated by a clock hand) is in a time synchronization state in which it matches the MN's timer. This is indicated by the arrow 105.

To set the above path, TS # T ₁ the reference time 1 data is Drop at NE # 0, TS # T ₂ the reference time 2 data is Drop at NE # 0, TS # 3 data Let T _{3 be} the reference time for dropping at NE # 0.

In this case, NE # 0 (MN) is a control signal for setting the time t for the Add / Drop operation (T ₁ -RTT ₀₁₎ with respect to TS # 1 in NE # 1, the Drop operation with respect to TS # 3 and a control signal for setting the time t (T ₃ -RTT ₀₁₎ by sending the NE # 1, passes set to NE # 1. Here, the subscript (01) of RTT means the round trip time measured from NE # 0 to NE # 1, and if it is RTT _ij , the round trip measured from NE # i to NE # j. Means time.

Further, NE # 0 (MN), the time for the control signal to set the time t (T ₁ -RTT ₀₂₎ for the Add / Drop operation with respect to TS # 1 in NE # 2, the Add operation with respect to TS # 2 and a control signal for setting t to (T ₂ -RTT ₀₂₎ by sending the NE # 2, passes set to NE # 2.

The time NE # 0 (MN) is for a control signal for setting the time t for the Add / Drop operation (T ₁ -RTT ₀₃₎ with respect to TS # 1 in NE # 3, the Add operation with respect to TS # 3 A path is set to NE # 3 by transmitting to NE # 3 a control signal that sets t to (T ₃ -RTT ₀₃ ).

Further, NE # 0 (MN) by sending respect TS # 1 in NE # 4 control signals to set the time t to the Add operation (T ₁ -RTT ₀₄₎ to NE # 4, NE # 4 Set the path to.

With these settings, for TS # 1 of a predetermined wavelength, when the time t in the NE # 4 is (T ₁ -RTT _04), the NE # 4 is the Add operation to the main signal, FIG. as shown in 5 (d), when the time in the NE # 3 is (T ₁ -RTT _03), with NE # 3 to Drop data from NE # 4, NE # 3 is Add to the main signal To work.
Then, NE # 2, when the time in the NE # 2 is (T ₁ -RTT _02), while Drop data from NE # 3, NE # 2 is the Add operation to the main signal.
Then, NE # 1, when the time in the NE # 1 is (T ₁ -RTT _01), while Drop data from NE # 2, NE # 1 to the Add operation to the main signal.
Then, NE # 0 drops the data from NE # 1 when the time of NE # 0 is T ₁ .

Also, if you set in this way, for the TS # 2 of the same wavelength, when the time t in the NE # 2 is (T ₂ -RTT _02), the NE # 2 is the Add operation to the main signal as shown in FIG. 5 (d), when NE # 0 is a time T _2, NE # 0 to Drop data from NE # 2.

Also, if you set in this way, for the TS # 3 of the same wavelength, when the time t in the NE # 3 is (T ₃ -RTT _03), the NE # 3 to the Add operation to the main signal As shown in FIG. 5D, NE # 2 is transmitted when the time at NE # 2 is (T ₃ -RTT ₀₂ ), and NE # 1 is transmitted at the time at NE # 1 (T ₃ -RTT). ₀₁ ), data from NE # 3 is dropped.

  A rectangle denoted by reference numeral 106 in FIG. 5D schematically shows the time slot time. For example, the time slot time of TS # 3 and the time slot time of TS # 2 do not overlap in a normal control state.

[In case of failure (comparative example)]
Next, in the current PON system as a comparative example, NE # 0 (MN) will be described with reference to FIG. Note that FIG. 6A and FIG. 6B are the same as FIG. 5A and FIG.

As shown in FIG. 6 (c), when the NE # 0 MN control unit fails (step S110), periodic time synchronization cannot be set by NE # 0, and MN control is performed as schematically shown by a broken line. The influence of the partial failure spreads to other NEs (SN).
When the control signal for time synchronization control does not reach from NE # 0, NE # 1 enters a self-running state, and the timer (displayed by a clock hand) is disturbed as time elapses (step S111).

Similarly, NE # 2, NE # 3, and NE # 4 also enter a free-running state, and the timer is disturbed as time elapses (steps S112, S113, and S114).
In FIG. 6C, the arrow of reference numeral 115 indicates that the timers of the NEs match if normal. However, at this time point, NE # 1 is operating based on the time when the internal clock is generated in a free-running state with a predetermined delay. Further, NE # 2 and NE # 4 operate based on a time delayed from NE # 1. In this example, NE # 3 operates based on the time when the internal clock is advanced and generated in a free-running state.

Therefore, when NE # 0 (MN) has set the path information path setting operation shown in FIG. 6B before failure, the operation shown in FIG. 5D cannot be realized.
Specifically, as shown in FIG. 6D, for TS # 1 having a predetermined wavelength, when the time t in NE # 4 is (T ₁ -RTT ₀₄ ) (see FIG. 5D). Even if NE # 4 tries to perform the Add operation to the main signal, the NE # 4 timer is delayed, so the Add operation is performed at the delayed time (step S124).

Further, (see FIG. 5 (d)) at time of NE # 3 is (T ₁ -RTT _03), even if NE # 3 is to the Add operation to the main signal, NE # 3 timers willing Therefore, the Add operation is performed at the advanced time (step S123).
Then, (see FIG. 5 (d)) at time of NE # 2 is (T ₁ -RTT _02), even if NE # 2 is to the Add operation to the main signal, the NE # 2 timer delay Therefore, the Add operation is performed at the delayed time (step S122).
Then, (see FIG. 5 (d)) at time of NE # 1 is (T ₁ -RTT _01), even if NE # 1 is to the Add operation to the main signal, the NE # 1 timer delay Therefore, the Add operation is performed at the delayed time (step S121).

Also, the TS # 2 of the same wavelength, (see FIG. 5 (d)) at time t in the NE # 2 is _{(T 2 -RTT 02), NE} # 2 is trying to Add operation to the main signal However, since the NE # 2 timer is delayed, the Add operation is performed at the delayed time (step S132).
Furthermore, for the TS # 3 of the same wavelength, (see FIG. 5 (d)) at time t in the NE # 3 is _{(T 3 -RTT 03), NE} # 3 is trying to Add operation to the main signal However, since the NE # 3 timer is advanced, the Add operation is performed at the advanced time (step S133). As a result, in NE # 2, interference between the time slot time of TS # 3 and the time slot time of TS # 2 (time slot interference) occurs.

[Outline of processing at the time of failure]
Next, in the present embodiment, an outline of processing when NE # 0 (MN) fails after, for example, path setting will be described with reference to FIGS. 7 (a) and 7 (b).
In the case of the present embodiment, as shown in FIG. 7A, NE # 0 (MN_0 system) exchanges control signals such as time synchronization with other NEs (SN) during normal operation. At this time, NE # 0 (MN_0 system) shares transmission time information for each other NE (SN) with NE # 2 (MN_1 system).
When the NE # 0 MN control unit fails, as shown in FIG. 7B, the master node (MN) is switched, and the NE # 2 that has been functioning as the slave node (SN) until then becomes the MN_1 system. Control signals such as time synchronization are exchanged with the remaining SNs.

[Time chart of processing at the time of failure]
Next, in the present embodiment, a processing time chart when NE # 0 (MN) fails will be described with reference to FIG.
Since FIG. 8A is the same as FIG. 5C and FIG. 6C, description thereof is omitted.
When each NE is in a time asynchronous state, the NE # 2 determines that the master node (MN) becomes the master node (MN) when a predetermined time-out period elapses from the periodic current schedule for the transmission time in the own node. The MN_0 system is switched to the MN_1 system. That is, NE # 2 starts computation as the MN_1 system.

  In this case, as shown in FIG. 8B, NE # 2 (MN_1 system) sets the current time to the reference time of NE # 2 (step S140), and NE # 1, NE # 3, NE #. 4 is set (steps S141, S143, S144). Details of the time setting will be described later. In FIG. 8B, the arrow 145 indicates that each NE timer (indicated by a clock hand) is in a time-synchronized state corresponding to the NE # 2 (MN_1 system) timer.

When a time elapses for a while after the control signal for time synchronization control arrives from NE # 2 and is synchronized, NE # 1 enters a self-running state, and the timer (indicated by a clock hand) is disturbed (step S151). ). Similarly, NE # 2, NE # 3, and NE # 4 also enter a free-running state, and the timer is disturbed as time elapses (steps S152, S153, and S154).
In FIG. 8C, the arrow 155 indicates that the timers of the NEs match if they are normal. At this time, however, the NE # 1 timer is delayed due to self-running. The NE # 2 and NE # 4 timers are delayed from the NE # 1 timer. In this example, the NE # 3 timer is advanced because of self-running.

As described above, when each NE is in the time asynchronous state, the NE # 2 re-enters the MN_1 system when a predetermined time-out period elapses from the periodic current schedule for the transmission time in the own node. The operation as is started.
The processes in steps S160, S161, S163, and S164 in FIG. 8D are the same as the processes in steps S140, S141, S143, and S144 in FIG. In addition, the code | symbol 165 of FIG.8 (d) represents the meaning similar to the code | symbol 145 of FIG.8 (b).
Thereafter, NE # 2 repeats the same processing. In the present embodiment, NE # 2 periodically performs time synchronization in this way, so that time slot interference between paths can be avoided.

[Processing time chart between master nodes]
Next, in the present embodiment, a time chart of processing between NE # 0 (MN_0 system) and NE # 2 (MN_1 system) will be described with reference to FIG. Note that FIG. 9 and the explanation thereof are performed without displaying the subscripts in order to make it easy to understand the explanation of the round-trip times relating to the active system and the standby system. Attached display is omitted.
In the system configuration represented on a straight line in FIG. 9B, NE # 0 (MN_0 system) is arranged at the left end, and NEs are arranged in order counterclockwise (counterclockwise) on the ring in FIG. 9A. Is. Note that NE # 2 is the MN_1 system.

First, time setting and RTT measurement by NE # 0 (MN_0 system), which is an active MN, will be described with reference to FIG.
The NE # 0 (MN_0 system), for example, transmits a control signal including a time stamp of the transmission time (reference time) T1 to the NE # 1, so that the time T2 when the control signal is received by the NE # 1 (illustrated) Time t) on the time axis is set to a common reference time T1 that is handled in common (step S171). Then, NE # 1 transmits a control signal including a time stamp of transmission time T4 to be handled in common at time T3 (time t = T3 on the time axis shown) after a predetermined time has elapsed from reception time T2. (Step S172). Then, NE # 0 calculates a round trip time RTT01 between NE # 0 and NE # 1. Incidentally, RTT01 has the same meaning as RTT ₀₁ described above, if RTTij, means a round trip time measured addressed NE # j from NE # i. This round trip time RTT01 is obtained as a difference between the time T5 when NE # 0 receives the control signal including the time stamp of the transmission time T4 from NE # 1 to NE # 1 and the transmission time T4 (RTT01 = T5). -T4).

  Further, NE # 0 transmits the control signal including the time stamp of the reference time T6 to NE # 2, for example, so that the reception time T7 of NE # 2 (time t = T7 on the time axis shown) is used as the reference. Set at time T6. Then, NE # 1 transmits a control signal including a time stamp of transmission time T9 to NE # 0 at time T8 (time t = T8 on the time axis shown) after a predetermined time has elapsed. Then, NE # 0 calculates a round trip time RTT02 between NE # 0 and NE # 2 from the difference between the received time T10 and the transmission time T9 (RTT02 = T10−T9).

Hereinafter, similarly, NE # 0 (MN) performs time setting with other NE (SN).
For example, NE # 0 calculates the round-trip time RTT04 between NE # 0 and NE # 4 from the difference between the time T20 when the control signal is received from NE # 4 and the transmission time T19 of NE # 4 (RTT04). = T20-T19).

Under the above premise, NE # 0 (MN_0 system) transmits a control signal including all round trip times measured in the MN_0 system as RTT_0 values to NE # 2 (MN_1 system) (step S180). In this example, the RTT_0 value is a round trip time RTT01 measured from NE # 0 to NE # 1, a round trip time RTT02 measured from NE # 0 to NE # 2, and measured from NE # 0 to NE # 3. The round trip time RTT03 and the round trip time RTT04 measured from NE # 0 to NE # 4 are shown.
RTT_0 = {RTT01, RTT02, RTT03, RTT04}

Then, NE # 0 (MN_0 system) calculates a round trip time (RTT_1 value) when measuring from NE # 2 (MN_1 system) to another NE based on the round trip time measured in the MN_0 system.
In this example, the RTT_1 value is the round trip time RTT21 when measuring from NE # 2 to NE # 1, the roundtrip time RTT23 when measuring from NE # 2 to NE # 3, and NE # 2 to NE #. It represents the round trip time RTT24 when measuring to # 4.
RTT_1 = {RTT21, RTT23, RTT24}
RTT21 = RTT02-RTT01
RTT23 = RTT03-RTT02
RTT24 = RTT04-RTT02

Then, NE # 0 (MN_0 system) transmits a control signal including the calculated RTT_1 value to NE # 2 (MN_1 system) (step S190).
Through these processes, NE # 2 (MN_1 system) retains the received RTT_0 value and RTT_1 value in the memory 40 of its own device (step S200). Accordingly, the RTT_0 value and the RTT_1 value can be shared between the master nodes (MN), that is, between NE # 0 (MN_0 system) and NE # 2 (MN_1 system).

[Time chart of spare device processing when MN control unit fails]
Next, in the present embodiment, a time chart of processing performed by the NE # 2 (MN_1 system) which is a spare device when the MN control unit of NE # 0 (MN_0 system) fails will be described with reference to FIG. In FIG. 10 and subsequent figures and in the description thereof, the time code is displayed as a subscript.

FIG. 10A is the same as FIG. 9B. When the MN control unit of NE # 0 (MN_0 system) fails as shown in FIG. 10 (a), NE # 2 (MN_1 system) becomes NE # 0 (MN_0 system) as shown in FIG. 10 (b). The transmission time set for each NE is set as it is. Specifically, T ₂₀ is set as a transmission time in NE # 1 by NE # 0 (MN_0 system). Further, the times T ₃₀ and T ₄₀ are set in NE # 3 and NE # 4, respectively.

In this case, NE # 2 (MN_1 system) does not receive NE # 1 from the set transmission time (T ₂₀ ) to NE # 1, which is located upstream (NE # 0 side) from its own node. The control signal is transmitted at a time obtained by subtracting the round trip time (RTT ₂₁ ) between 2 and NE # 1. That is, by transmitting a control signal including a time stamp whose transmission time is (T ₂₀ -RTT ₂₁ ), the time at which NE # 1 receives this control signal is set as the transmission time (T ₂₀ ) at NE # 1. Set (step S201). Note that RTT ₂₁ means a round-trip time when measuring from NE # 2 to NE # 1 as in RTT21 described above.

Further, NE # 2 (MN_1 system) is controlled at the transmission time (T ₃₀ ) set by NE # 0 (MN_0 system) with respect to NE # 3 arranged downstream of the own node. Send a signal. That is, by transmitting a control signal including a time stamp whose transmission time is (T ₃₀ ), the time at which NE # 3 receives this control signal is set as the transmission time (T ₃₀ ) of NE # 3 ( Step S203).

Similarly, NE # 2 (MN_1 system) is controlled at the transmission time (T ₄₀ ) set by NE # 0 (MN_0 system) for NE # 4 arranged downstream of its own node. Send a signal. That is, by transmitting a control signal including a time stamp whose transmission time is (T ₄₀ ), the time at which NE # 4 receives this control signal is set as the transmission time (T ₄₀ ) of NE # 4 ( Step S204).

In this case, the NE # 2 (MN_1 system) uses the transmission time of the data signal set by the NE # 0 (MN_0 system) as it is by using the round trip time (RTT_0) of the active system as it is. ). The NE # 2, which is a spare device, can acquire path route information from a control device (not shown) arranged outside the MN by the route information acquisition unit 61 of the path setting processing unit 60.
Then, NE # 2 (MN_1 system) performs an Add / Drop operation on TS # 1 in NE # 1, for example, as the processing performed by NE # 0 (MN_0 system) as a setting process for each NE (SN). Transmitting to NE # 1 a control signal for setting the time t to be set at (T ₁ -RTT ₀₁ ) and a control signal for setting the time t at which the drop operation is performed for TS # 3 to (T ₃ -RTT ₀₁ ). Thus, a path is set for NE # 1 (step S211).

Further, NE # 2 (MN_1 system) is a control signal for setting the time t for the Add / Drop operation (T ₁ -RTT ₀₃₎ with respect to TS # 1 in NE # 3, the Add operation with respect to TS # 3 and a control signal for setting the time t (T ₃ -RTT ₀₃₎ by sending the NE # 3, passes set to NE # 3 (step S213).
Further, NE # 2 (MN_1 system) by sending a control signal to set the time t to the Add operation with respect to TS # 1 in NE # 4 to (T ₁ -RTT ₀₄₎ to NE # 4, NE # 4 is set (step S214). By setting in this way, the operation described with reference to FIG. 5D can be realized.

According to the present embodiment, when the standby device NE # 2 (MN_1 system) functions as a slave node (SN), the active system round trip time (RTT_0) and the standby system round trip time (RTT_1). Can be acquired from the active master node (MN), held in the own apparatus, and shared with the active MN. For this reason, when the working MN fails, the spare device uses the information stored in the own device to perform time setting for the master node (MN) and periodically updates the slave node (SN) time. The time synchronization can be continued.
Further, the standby device uses the round trip time (RTT_0) of the active system, and sets the transmission time of the data signal set by NE # 0 (MN_0 system) as it is to each slave node (SN). Can do.
In addition, since the active MN is located at different locations and the so-called distributed buildings are used, even if a building where network facilities are installed due to a catastrophic disaster such as an earthquake, the active MN It is possible to provide a highly reliable network in which the risk of simultaneous failure with the spare device is reduced. In addition, by making the master node (MN) redundant by building distribution, the MN control function without affecting the signal of the path that does not go through the failure location when the master node control function unit (MN control function unit) fails The part can be switched.

The present invention is not limited to the above-described embodiment, and various modifications can be made. Moreover, you may deform | transform as follows. For example, the number of slave nodes (SN) is not limited to the four illustrated, and the time slots of the same wavelength are not limited to the three illustrated.
Although the ring NW has been described as a double ring, a single ring may be used. Further, the present invention can be applied to a multi-ring system having a layer structure having an upper ring and a lower ring.

In the above embodiment, the active MN calculates the round trip time of the standby system. However, the standby apparatus may calculate the round trip time of the standby system from the acquired round trip time of the active system.
Moreover, although the said embodiment demonstrated the spare apparatus as one, you may install a several spare apparatus.
Further, the location where the MN (NE # 0) and the spare device (NE # 2) are installed is not limited to buildings such as buildings.

  Further, the NE # 2, which is a standby device, acquires path route information from a control device (not shown) by the route information acquisition means 61 of the path setting processing unit 60 of its own device when the MN control unit fails, but this is not limitative. Absent. For example, the path setting information of all slave nodes (SN) may be acquired together with information on the round trip time of the active system before the MN control unit failure. By doing in this way, the effort which produces | generates path setting information based on path route information can be saved.

DESCRIPTION OF SYMBOLS 1 Optical ring network system 21 Optical switch 22, 23 Optical coupler 31 Burst receiving part 32 Header analysis part 33, 36 Switch part 34 IF transmission part 35 IF receiving part 37 Buffer 38 Burst transmission part 40 Memory 50 Time synchronization function part 60, 60A , 60B path setting processing unit 61 path information acquiring unit 62 path setting information generating unit 63 path setting information updating unit 64 path setting information notifying unit 65 spare device RTT calculating unit (reserve device round trip time calculating unit)
66 RTT notification means (round-trip time notification unit)
67 Path setting information receiving means (path setting information receiving unit)
68 Path setting information update means 70 Timing controller R ring NE Network element (network device)

Claims (6)

Signals transmitted to and received from external devices are terminated, and a master node disposed at a first position as a control subject and a plurality of slave nodes as objects to the control subject are annularly formed by a ring by an optical transmission line A connected optical ring network system,
At least one slave node is a spare device that has a function of at least part of the master node and is disposed at a second position different from the first position;
The master node is
Synchronized with the reference time in its own master node, and sets the synchronized transmission time used in common in each network device in each slave node, and the round trip time between the own master node and each slave node. A time synchronization function unit for measuring time;
A path setting processing unit that sets the transmission time of the data signal in each slave node based on the reference time in the own master node of the path and the round trip time of the working system;
A round-trip time notifying unit for notifying the spare unit of the round-trip time of the active system,
The spare device is
A receiving unit for receiving a round trip time of the active system from the master node;
A storage unit for storing the received round trip time of the active system;
In addition to setting the synchronized transmission time with the master node, when the master node fails, the round trip time between the standby device and each slave node is calculated from the round trip time of the active system. A time synchronization function unit that periodically sets a synchronized transmission time commonly used in each network device to each slave node based on a round trip time of a certain standby system,
The time synchronization function unit of the spare device sets the transmission time set in the master node as it is for the slave node located downstream from the spare device, and is located upstream from the spare device. For the slave node, the time is set by transmitting a control signal at the time obtained by subtracting the round trip time of the standby system from the transmission time set by the master node.
An optical ring network system characterized by that. The master node is
A backup device round trip time calculating unit that calculates a round trip time of the backup system that is a round trip time between the standby device and each of the other slave nodes based on the round trip time of the active system;
A storage unit that stores the round trip time of the active system and the round trip time of the standby system;
2. The optical ring network system according to claim 1, wherein the round trip time notifying unit notifies the spare unit of the round trip time of the active system and the round trip time of the standby system. The spare device is
A path setting processing unit configured to set each slave node using the transmission time of the data signal set in the master node as it is by using the round trip time of the active system when the master node fails. The optical ring network system according to claim 1 or 2. Signals transmitted to and received from external devices are terminated, and a master node disposed at a first position as a control subject and a plurality of slave nodes as objects to the control subject are annularly formed by a ring by an optical transmission line A redundant method of connected optical ring network systems, comprising:
At least one slave node is a spare device that has a function of at least part of the master node and is disposed at a second position different from the first position;
The master node is
Setting the synchronized transmission time to be used in common in each network device in synchronization with the reference time in its own master node in each slave node;
Measuring the round trip time of the active system, which is the round trip time between the own master node and each slave node;
Performing a step of notifying the standby device of the round trip time of the active system,
The spare device is
Receiving a round trip time of the active system from the master node;
Storing the received round trip time of the active system;
For the slave node located downstream from the spare device at the time of failure of the master node, setting using the transmission time set in the master node as it is,
For the slave node located upstream from the spare device, the round trip time of the standby system, which is the round trip time between the spare device and each slave node, calculated from the round trip time of the active system, Executing a step of transmitting a control signal at a time obtained by subtracting from a transmission time set in a node;
A redundancy method characterized by that. The master node is
5. The step of calculating a backup system round-trip time, which is a round-trip time between the standby device and each of the other slave nodes, based on the round-trip time of the active system and notifying the backup device is executed. The redundancy method described. The master node sets the transmission time of the data signal in each slave node based on the reference time in the own master node of the path and the round trip time of the working system,
5. The standby apparatus sets each slave node by using the transmission time of the data signal set in the master node as it is by using the round trip time of the active system when the master node fails. Alternatively, the redundancy method according to claim 5.

Images