JP2012105363A - Office-side apparatus, control method therefor and computer program therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an office-side apparatus capable of performing prompt redundancy switchover.SOLUTION: OSU 1 to N (12-1 to 12-N) are connected to a plurality of PON lines 3-1 to 3-N, respectively. A control unit 14 performs registration processing and band allocation processing on a plurality of customer-side devices connected to the plurality of PON lines 3-1 to 3-N. Further, the control unit 14 performs OSU switchover by modifying communication channels between the plurality of PON lines 3-1 to 3-N and the plurality of OSU 1 to N (12-1 to 12-N), while maintaining the registration state of the plurality of customer-side devices. Thus, prompt redundancy switchover can be performed.

Description

  The present invention relates to a PON (Passive Optical Network) that is a medium-shared communication in which a plurality of home-side devices share a medium and transmit data, and in particular, a plurality of optical line units (hereinafter referred to as OSUs) that terminate at a station side. (Referred to as "Optical Subscriber Unit"). The present invention relates to a station-side device having a fault-tolerant function enhanced by making it redundant, a control method thereof, and a computer program thereof.

  In recent years, the Internet has become widespread, and users can access various information on sites operated in various parts of the world and obtain the information. Along with this, devices capable of broadband access such as ADSL (Asymmetric Digital Subscriber Line) and FTTH (Fiber To The Home) are rapidly spreading.

  As prior arts related to this, there are technologies disclosed in Patent Document 1, Patent Document 2, and Non-Patent Document 1 below. In the invention disclosed in Patent Document 1, the working optical termination unit communicates with a plurality of user optical termination devices via an optical transmission line. The unit includes a management information storage device that stores management information of a plurality of user optical terminal devices. The backup optical termination unit includes a storage device capable of storing the management information transferred from the active optical termination unit. The control device controls switching from the working optical termination unit to the backup optical termination unit.

  In the invention disclosed in Patent Document 2, an L3 switch having a port mirroring function is arranged upstream of an OLT (Optical Line Terminal), and a 2: 1 optical coupler is arranged downstream. Management information and setting information are transferred from the first OLT to the second OLT before the start of switching. When a special frame indicating the start of switching is input from the switch to the first and second OLTs, the first OLT stops taking downlink data, and the second OLT starts taking downlink data. The bandwidth allocation to the ONU (Optical Network Unit) is temporarily stopped, the second OLT is instructed to take in uplink data, and the bandwidth allocation to the ONU is resumed. When there is no downlink data of the first OLT, the management information is transferred from the first OLT to the second OLT. The second OLT is operated in a normal operation.

  Non-Patent Document 1 describes one method of PON. All information including user information passing through the PON and control information for managing and operating the PON is in the form of an Ethernet (registered trademark) frame. EPON (Ethernet (registered trademark) PON) to be communicated, an access control protocol (MPCP (Multi-Point Control Protocol)), and an OAM (Operations, Administration and Maintenance) protocol are defined. By exchanging MPCP frames between the station side device and the home side device, the home side device joins and leaves, and uplink access multiplexing control is performed.

  Even in 10GEPON (EPON corresponding to a communication speed of 10 Gbps) that is standardized as IEEE 802.3av, MPCP is assumed as the access control protocol.

  Patent Document 3 shows that a layer 2 switch (L2SW) is generally used to further aggregate the uplinks of a plurality of PONs. In L2SW, relay processing is performed in units of terminal MAC addresses.

JP 2007-036926 A JP 2007-067601 A JP 2004-201013 A

IEEE Std 802.3ah (registered trademark) -2004

  The inventions described in Patent Document 1 and Patent Document 2 described above transfer management information necessary for OSU operation when switching OSUs. Therefore, there are problems that the switching time becomes long and that the management information transferred itself is not correct and does not operate normally after switching.

  In addition, a CPU is arranged for each OSU, and it is not possible to deal with abnormalities in the OSU including the CPU. Therefore, there is a problem that OSU abnormality is vulnerable.

  Furthermore, in the configuration in which the management information storage unit is shared and the OSU is switched without transferring the management information, there is a problem that an abnormal OSU may destroy the management information of other OSUs. .

  Further, in the configuration in which the OSU uplink concentrating is performed by the L2SW, even if the OSU is switched, the terminal MAC address under the OSU is stored in association with the port to the OSU before switching. Until the frame is aged out or re-learned by arrival of an upstream frame, the downstream frame is not delivered to the switched OSU, resulting in a long switching time. Even if the relationship between the terminal MAC address and the OSU port is forcibly reset, there is still a problem that it takes a long switching time because the number of terminal MAC addresses is large.

  The present invention has been made to solve the above problems, and a first object of the present invention is to provide a station-side apparatus capable of quickly performing redundant switching.

  The second object is to provide a station side device capable of improving the fault tolerance performance without significantly impairing the economy.

  According to one aspect of the present invention, a station-side device that accommodates a plurality of passive optical networks, a plurality of optical line units connected to each of the plurality of passive optical networks, and a plurality of passive optical networks Control means for performing registration processing and bandwidth allocation processing for a plurality of home-side devices connected to the network, and the control means changes communication paths between the plurality of passive optical networks and the plurality of optical line units. Thus, the optical line unit is switched while maintaining the registration state of the plurality of home-side devices.

  Since the control means changes the communication path between the plurality of passive optical networks and the plurality of optical line units, it switches the optical line units while maintaining the registered state of the plurality of home-side devices. Switching can be performed quickly.

  Preferably, the station-side apparatus further includes a concentrator that multiplexes uplink frames from a plurality of optical line units and transmits them to the uplink, and distributes downlink frames from the uplink to the respective optical line units. Changes the communication path between the plurality of passive optical networks and the plurality of optical line units, and changes the setting of the concentrator, thereby maintaining the registered state of the plurality of home-side devices. Switching is performed.

  The control means changes the communication path between the plurality of passive optical networks and the plurality of optical line units and changes the setting of the concentrating means, thereby maintaining the registered state of the plurality of home devices. Since the line units are switched, redundant switching of the optical line units can be quickly performed even in a configuration in which the uplinks of a plurality of optical line units are concentrated.

  Preferably, the station side device further includes an optical switch means for switching the connection between the plurality of optical line units and the plurality of passive optical networks, the plurality of optical line units includes a spare optical line unit, and the control means includes The optical switch means is controlled to switch the at least one of the plurality of passive optical networks to be connected to the spare optical line unit while maintaining the registration state of the plurality of home-side devices.

  Therefore, it is possible to improve the fault tolerance performance without significantly impairing the economy.

  Preferably, the plurality of optical line units include an active optical line unit and a standby optical line unit, and the control means maintains a registered state of the plurality of home-side devices while maintaining a plurality of passive optical networks. Is switched from the active optical line unit to the standby optical line unit.

  Therefore, even if a failure occurs in a plurality of optical line units, communication can be maintained, so that the fault tolerance performance can be further improved.

  Preferably, the station side device includes an active system control unit and a standby system control unit, and the active system control unit maintains a registered state of the plurality of home side devices from the active system control unit. Switching to the control means.

  Therefore, it is possible to improve the fault tolerance performance of the control means without significantly impairing the economy.

  Preferably, the station side device includes an active system control unit and a standby system control unit, an active system line collecting unit and a standby system line collecting unit, and the active system control unit registers a plurality of home side devices. While maintaining the state, switching from the active control means to the standby control means, or switching from the active concentrator to the standby concentrator is performed.

  Therefore, in addition to improving the fault tolerance performance of the control means, it is possible to improve the fault tolerance performance of the concentrating means without significantly impairing the economy.

  According to another aspect of the present invention, there is provided a method for controlling a station-side device that accommodates a plurality of passive optical networks, wherein the control means registers a plurality of home-side devices connected to the plurality of passive optical networks. A step of sequentially performing the processing and the bandwidth allocation processing, and a step of changing a communication path of a control message for performing the registration processing and the bandwidth allocation processing while maintaining the registration state of the plurality of home-side devices.

  According to still another aspect of the present invention, there is provided a computer program for causing a computer to execute a method for controlling a station-side device that accommodates a plurality of passive optical networks. Sequentially performing registration processing and bandwidth allocation processing of a plurality of home-side devices connected to the network, determining communication paths of control messages for performing registration processing and bandwidth allocation processing by OSU conversion, and a plurality of home-side devices And changing the OSU conversion while maintaining the registered state of the device.

  According to an aspect of the present invention, the control unit changes the communication path between the plurality of passive optical networks and the plurality of optical line units, thereby maintaining the registered state of the plurality of home-side devices. Since the line unit is switched, redundant switching can be performed quickly.

It is a block diagram which shows schematic structure of the station side apparatus in the 1st Embodiment of this invention. It is a figure which shows the structural example of the optical switch 11a. It is a block diagram which shows the structural example of OSU. It is a block diagram which shows the structural example of the concentrating part 13a. 3 is a block diagram illustrating a configuration example of a control unit 14. FIG. 4 is a flowchart for explaining a procedure of initialization processing by a control unit 14; 4 is a flowchart for explaining a procedure of an interrupt processing routine by a control unit 14; It is a flowchart for demonstrating the detail of the discovery process (S26) shown in FIG. It is a flowchart for demonstrating the detail of RP timeout process (S27) shown in FIG. It is a flowchart for demonstrating the detail of the management communication master process (S28) shown in FIG. It is a flowchart for demonstrating the detail of the management communication slave process (S31) shown in FIG. 12 is a flowchart for explaining details of a scheduled confirmation process (S72) shown in FIG. It is a flowchart for demonstrating the detail of the control system switching (on) process (S32) shown in FIG. It is a flowchart for demonstrating the detail of the message reception process (S23) shown in FIG. 15 is a flowchart for explaining details of a register request process (S103) shown in FIG. 15 is a flowchart for explaining details of a register confirmation process (S104) shown in FIG. It is a flowchart for demonstrating the detail of the deregister process (S105) shown in FIG. It is a flowchart for demonstrating the detail of the report reception process (S106) shown in FIG. It is a flowchart for demonstrating the detail of a RTT update process. It is a flowchart for demonstrating the detail of the band allocation process (S107) shown in FIG. 15 is a flowchart for explaining details of OAM message processing (S108) shown in FIG. 14; 15 is a flowchart for explaining details of OSU management message processing (S109) shown in FIG. 14; It is a flowchart for demonstrating the detail of the OAM process (S24) shown in FIG. It is a flowchart for demonstrating the detail of the operation IF process (S25) shown in FIG. It is a flowchart for demonstrating the detail of the OSU switching process (S221) shown in FIG. It is a flowchart for demonstrating the detail of the OSU switch-back process (S222) shown in FIG. FIG. 25 is a flowchart for explaining details of a control system switching (off) process (S223) shown in FIG. 24. FIG. It is a block diagram which shows schematic structure of the station side apparatus in the 2nd Embodiment of this invention. It is a block diagram which shows the structural example of the concentrating part 13b in the 2nd Embodiment of this invention. It is a block diagram which shows schematic structure of the station side apparatus in the 3rd Embodiment of this invention. It is a figure which shows the structural example of the optical switch 11b in the 3rd Embodiment of this invention. It is a block diagram which shows schematic structure of the station side apparatus in the 4th Embodiment of this invention. It is a block diagram which shows the other example of a connection of the station side apparatus in the 2nd Embodiment of this invention.

  Hereinafter, the details of the embodiment of the present invention will be described. As a premise common to the embodiments, the PON is an Ethernet (registered trademark) -based PON (EPON), and is based on an MPCP frame defined in IEEE 802.3ah. It is assumed that ONU registration, withdrawal, bandwidth allocation to ONU, bandwidth request from ONU, and the like are performed.

(First embodiment)
FIG. 1 is a block diagram showing a schematic configuration of a station-side apparatus according to the first embodiment of the present invention. This station side device (hereinafter also referred to as OLT) 1a is connected to N PON lines 1 to N (3-1 to 3-N), and can terminate N PON lines. Each of the PON lines 1 to N (3-1 to 3-N) is connected to each of the optical couplers 4-1 to 4-N, and is connected to a plurality of ONUs 2 via the optical coupler.

  The station side device 1a includes an optical switch 11a, N + 1 OSU1 to N + 1 (12-1 to 12-N + 1), a concentrator 13a, and a control unit 14 that performs overall control of the station side device 1a. . This station-side apparatus 1a has a redundant configuration including one spare OSU N + 1 for N OSUs 1 to N (hereinafter, this redundant configuration is referred to as an N: 1 redundant configuration).

  In response to an instruction from the control unit 14, the optical switch 11a includes N + 1 OSU1 to N + 1 (12-1 to 12-N + 1), N PON lines 1 to N (3-1 to 3-N), Switch the connection.

  The concentrator 13a multiplexes the uplink frames from the OSU1 to N + 1 (12-1 to 12-N + 1) and transmits them to the upper network (hereinafter referred to as uplink) and appropriately receives the downlink frames received from the uplink. The process of distributing to the appropriate OSU is performed.

  FIG. 2 is a diagram illustrating a configuration example of the optical switch 11a. In the optical switch 11a, N optical fibers on the PON side and N optical fibers on the OSU side are arranged to face each other. Hereinafter, these N optical fibers 3-1 to 3-N are referred to as normal optical fibers.

  Collimating lenses 23-1 to 23-N are arranged near the end face of each optical fiber on the OSU side, and collimating lenses 24-1 to 24-N are arranged near the end face of each optical fiber on the PON line side. In the state, optical space transmission is performed between the optical fibers facing each other. The optical axes of the N optical fibers are arranged to be parallel on the same plane.

  The movable mirror 22 is driven by an actuator 21 and can move on an axis orthogonal to the optical axes of the N optical fibers 3-1 to 3-N. The movable mirror 22 is inclined by 45 ° with respect to the optical axis between the normal optical fibers, and reflects the light beam from the optical fiber on the PON line side in the moving axis direction of the movable mirror 22. The light beam reflected by the movable mirror 22 enters the standby optical fiber 3-N + 1 via the collimator lens 23-N + 1.

  Further, since the light beam from the standby optical fiber 3-N + 1 is reflected by the movable mirror 22 and enters the optical fiber on the PON line side where the movable mirror 22 is located, the PON where the movable mirror 22 is located. Optical space transmission can be performed between the optical fiber on the line side and the backup optical fiber.

  The movable mirror 22 can take any position among N + 1 positions of the intersection position with the optical axes of the N normal optical fibers and the standby position. The actuator 21 moves the movable mirror 22 to an appropriate position within the range of the N + 1 position in accordance with a control signal from the control unit 14. The control signal from the control unit 14 indicates the presence / absence of OSU redundancy switching and the number of the PON line to be switched, and the actuator 21 moves the movable mirror 22 accordingly.

  FIG. 3 is a block diagram illustrating a configuration example of the OSU. The OSU 12 includes a concentrator IF (Interface) unit 31, a control IF unit 32, a reception processing unit 33, a transmission processing unit 34, a PON transmission / reception unit 35, a local control unit 36, and a FIFO 1 (37 ) And FIFO2 (38) for accumulating downstream frames.

  3 to 5, the configurations of the OSU, the concentrator, and the controller are described in a format including duplexing described in third to fourth embodiments described later. In FIG. 3, the concentrating IF unit 31 and the control IF unit 32 are not connected to the duplex system, but are connected to one concentrating unit 13a and the control unit 14, respectively. The PON transceiver unit 35 is connected to the optical switch 11a. The same applies to FIGS. 4 and 5.

  The PON transmission / reception unit 35 is connected to the optical switch 11a by a single optical fiber, receives an upstream optical signal of a specific wavelength, for example, a 1310 nm band so that bidirectional communication can be performed on the optical fiber, and transmits an electrical signal. And is output to the reception processing unit 33, and the electric signal output from the transmission processing unit 34 is multiplexed as a downstream optical signal of another wavelength, for example, 1490 nm band.

  The reception processing unit 33 reconstructs a frame from the electrical signal received from the PON transmission / reception unit 35 and distributes the frame to the control IF unit 32, the local control unit 36, or the FIFO 1 (37) according to the type of the frame. Specifically, the user frame is output to the FIFO 1 (37), a special control frame such as a loopback test is output to the local control unit 36, and other general control frames are output to the control IF unit 32.

  Further, the reception processing unit 33 may receive grant information indicating from which logical link the frame is received and when it is received from the transmission control unit 34, and may filter received frames that are not granted. Further, the time stamp at the time of reception may be overwritten on the MPCP frame and output.

  The concentration IF unit 31 sends the upstream frame stored in the FIFO 1 (37) to the concentration unit 13a and stores the downstream frame received from the concentration unit 13a in the FIFO 2 (38). At this time, the concentrating IF unit 31 performs conversion between the signal format of the concentrating unit 13a and the internal signal format.

  When the FIFO 2 (38), the control IF unit 32, or the local control unit 36 has a frame / message to be transmitted, the transmission processing unit 34 receives the frame / message according to the priority order, assembles the frame, and sends it to the PON transmission / reception unit 35. Output. At this time, the time stamp at the time of transmission may be overwritten on the MPCP frame and output. In addition, the transmission processing unit 34 outputs the grant information described in the gate message from the control IF unit 32 to the reception processing unit 33.

  The control IF unit 32 outputs the control message received from the reception processing unit 33 to the control unit 14 and outputs the control message received from the control unit 14 to the transmission processing unit 34. At this time, the control IF unit 32 performs conversion between the signal format of the control unit 14 and the internal signal format.

  As a general rule, the control unit 14 terminates the control protocol for managing and operating the PON. However, in order to reduce the processing load on the control unit 14, the local control unit 36 terminates a specific protocol. In the present embodiment, a local control unit 36 performs a loopback test for an ONU, which is a kind of OAM, in accordance with an instruction from the control unit 14. That is, the local control unit 36 sets the loopback test mode, generates a loopback test frame, inspects the frame returned by the loopback, notifies the result, and cancels the loopback mode.

  The PON transmission / reception unit 35 collects statistical information such as the number of transmission frames, the number of received frames, and the number of received signal code errors, and notifies the control unit 14 via the control IF unit 32. This statistical information is used when determining whether to switch the system when the control unit or the like is set to a dual system as described later. For example, when the number of code errors in the received signal is large, control such as switching to another system is performed.

  Further, the PON transmission / reception unit 35 monitors its own transmission light level, and when the transmission light level is out of the specified range due to failure or aging degradation of the light emitting element, the control unit 14 via the control IF unit 32. Notify the alarm.

  The reception processing unit 33 collects statistical information such as the number of received frames for each frame type and notifies the control unit 14 via the control IF unit 32. Similarly, the transmission processing unit 34 collects statistical information such as the number of transmission frames for each frame type, and notifies the control unit 14 via the control IF unit 32.

  FIG. 4 is a block diagram illustrating a configuration example of the concentrator 13a. The line concentrator 13a includes an uplink transmitter / receiver 41, a downlink distributor 42, a line concentrator 43, a control IF 44, an OSU IF 1 to N + 1 (45-1 to 45-N + 1), and a filter. 1 to N + 1 (46a-1 to 46a-N + 1), FIFOs 47-1 to 47-N + 1, and a selector 48.

  The OSU IF unit i (i = 1 to N + 1) converts the upstream frame sent from the corresponding OSUi into an internal signal format, and temporarily buffers it in the FIFOi via the filter unit i.

  The line concentrator 43 manages the state of the FIFOs 47-1 to 47-N + 1, determines the output order to the uplink for the non-empty FIFO, and transfers the upstream frame from the FIFO to the uplink transceiver 41. Instruct. At this time, the line concentrator 43 controls the selector 48 and performs setting so that the uplink frame output from the FIFO reaches the uplink transceiver 41 via the selector 48.

  The uplink transmission / reception unit 41 transmits the uplink frame output from the selector 48 to the uplink, and outputs the downlink frame received from the uplink to the downlink distribution unit 42.

  The downlink distribution unit 42 copies the downlink frame received from the uplink transmission / reception unit 41 and outputs it to the filter units 1 to N + 1 (46a-1 to 46a-N + 1).

  The filter unit i (i = 1 to N + 1) refers to VLAN (Virtual LAN) header information stored in the downstream frame, and determines which PON line i the frame is to be transmitted to. The filter unit i filters downstream frames that should not be transmitted, performs necessary processing such as header conversion on the downstream frames to be transmitted, and outputs the result to the OSU IF unit i. Further, the filter unit i connects or disconnects the uplink path and the downlink path based on information regarding the redundant configuration of the control unit 14.

  The filter unit N + 1 (46a-N + 1) corresponding to OSU N + 1 may hold the settings of all other filter units and select and apply the setting corresponding to the switched line. Here, the setting of the filter unit refers to information used for filtering downstream frames such as VLAN information and special reserved MAC addresses, and connection / disconnection of the entire path according to the redundant configuration. Information used for the purpose.

  The control IF unit 44 communicates with the control unit 14, and performs setting of each unit of the concentrating unit 13a, notification of a state, transfer of an alarm, and the like.

  Further, the uplink transmission / reception unit 41 collects statistical information such as the number of transmission frames, the number of received frames, and the number of code errors in the received signal, and sends these statistical information to the control IF unit 44 in response to an inquiry from the control unit 14. To the control unit 14. In addition, the uplink transmission / reception unit 41 detects link establishment / disconnection at the physical layer level with the opposite device in the uplink, and detects transmission abnormality by monitoring its own output signal. The uplink transmission / reception unit 41 notifies a corresponding alarm to the control unit 14 via the control IF unit 44 when an uplink disconnection or transmission abnormality is detected.

  FIG. 5 is a block diagram illustrating a configuration example of the control unit 14. The control unit 14 is shared by a CPU (Central Processing Unit) 51, a ROM (Read Only Memory) 52, a RAM (Random Access Memory) 53, an IO (Input Output) control unit 54, and an OSU IF unit 55. A RAM 56 and a clock / timer 57 are included.

  The OSU IF unit 55 is connected to the OSUs 1 to N + 1 (12-1 to 12-N + 1), and is transmitted / received to / from the ONU. The OSU IF unit 55 manages the PON and the OSU for managing and controlling the OSU. Send and receive information via message communication. These messages are interfaced with the CPU 51 via an input message queue and an output message queue realized by the shared RAM 56.

  The IO control unit 54 receives a command from the CPU 51, performs setting and status management of the concentrator 13a and the optical switch 11a, and provides an operation interface (hereinafter, operation IF) of the station side device 1a to the operator. . Further, when receiving an event such as a response from the operation IF, the concentrator 13a and the optical switch 11a or an alarm, the IO control unit 54 notifies the CPU 51 as an interrupt.

  The clock / timer 57 manages a clock and various timers for managing the PON, and outputs a time and timer end interrupt to the CPU 51.

  The ROM 52 stores a program for controlling the entire control unit 14 and fixed data. The RAM 53 is used as a work area for temporarily storing data.

  The CPU 51 reads and executes a program from the ROM 52, and reads the fixed data stored in the ROM 52 and reads / writes data to / from the RAM 53 to execute processing described later.

  The reference clock for driving the PON clock is also given to each OSU, and the PON clock of each OSU is synchronized with the periodic time notification by management communication.

  The processing in the control unit 14 is basically the same as the processing in the well-known OSU local control unit, except that all PONs are controlled by a single CPU and program, and quick OSU redundancy switching / switchback. In order to implement, the concept of virtual OSU and OSU conversion is introduced. Note that switching back refers to returning to the original state after performing redundancy switching.

  The OSU that normally interfaces with the CPU 51 executing the program is a virtual OSU. That is, the virtual OSU is interfaced with the real OSU by performing OSU conversion. For example, when the end of the PON line i is switched from OSUi to OSU N + 1, the OSU conversion is changed from (virtual OSUi⇔real OSUi) to (virtual OSUi⇔real OSU N + 1). As a result, the virtual OSUi can be made to correspond to the PON line i regardless of the OSU redundancy switching / switchback state.

  Such implementation of OSU conversion may be software processing by the CPU 51 or hardware processing by the OSU IF unit 55. However, regarding OSU control information for managing and controlling the OSU, the CPU 51 can execute a program and directly specify the real OSU to perform message communication.

  FIG. 6 is a flowchart for explaining the procedure of the initialization process by the control unit 14. This initialization process performs necessary initial settings for each part of the station side apparatus 1a. Here, a logical link table LLT and a final allocation time TE are provided for each virtual OSU. Each element of the logical link table LLT stores information unique to the logical link. Specifically, it includes a logical link state LLstat, report information RPinfo, latest report reception time RPtime, round-trip propagation time RTT, and latest OAM link connectivity confirmation frame reception time OAMt.

  First, the input message queue Qin and the output message queue Qeg are emptied, the logical link tables LLTij (i = 1, 2,..., N: j∈ {LLID of OSU i}) are all NULL, and the final allocation time TEi is all The current time is ctime, and the OSU conversion is OSUmap (i) = i (i = 1, 2,..., N) (S11).

  Next, by performing management communication with another system control unit (S12), it is determined whether the control unit is duplexed or a standby system control unit (S13). In this embodiment, the control unit is not duplicated. However, in order to explain the duplication of the control unit described in the third and fourth embodiments together, FIGS. Such a procedure is also included. Therefore, in the present embodiment, operational processing is always performed.

  If the control unit is duplicated and the control unit is a standby system (S13, Yes), the management information of the active control unit is copied, and the current time ctime is set as the active system ctime. Match to the operational system (S14). Then, the management communication timer (TMC) is set (S15), and the process ends.

  If the control unit is not duplexed or the control unit is an active system (S13, No), the configuration definition information is taken in via the operation IF and the optical switch 11a is set to the normal state. Then, based on the transfer rule included in the configuration definition information, for example, information indicating the correspondence between the VLAN ID and the PON line in the upper network via the uplink, the setting of each filter unit in the concentrator 13a is performed (S16). ).

  Finally, the discovery timer (TD) is set, the management communication timer (TMC) is set (S17), and the process ends.

  FIG. 7 is a flowchart for explaining the procedure of the interrupt processing routine by the control unit 14. Since most of the processes in the control unit 14 occur irregularly, interrupts are used, and priority is given to each interrupt process. In the processes of steps S23 to S28, the process of S23 has the highest priority, and the process of S28 has the lowest priority. Further, in the processes of steps S30 to S32, the process of S30 has the highest priority, and the process of S32 has the lowest priority.

  First, when an interrupt occurs, it is determined whether or not the control unit 14 is an active system (S21). If the control unit 14 is an active system (S21, Yes), the process branches depending on the interrupt type (S22). If the interrupt type indicates that there is a message in the input message queue (S22, Qin is not empty), a message reception process is performed (S23), and the process is terminated.

  If the interrupt type is the expiration of the OAM process activation timer (S22, TOAM expiration), the OAM process is performed (S24), and the process is terminated. If the interrupt type is an interrupt from the operation IF (S22, operation IF), the operation IF process is performed (S25), and the process ends.

  If the interrupt type is the expiration of the discovery timer (TD) (S22, TD expiration), the discovery process is performed (S26), and the process is terminated. If the interrupt type is expiration of the timer TLij (S22, TLij expiration), a report timeout process is performed (S27), and the process is terminated. If the interrupt type is the expiration of the management communication timer TMC (S22, TMC expiration), the management communication master process is performed (S28), and the process is terminated.

  In step S21, if the control unit 14 is a standby system (S21, No), the process branches depending on the interrupt type (S29). If the interrupt type indicates that there is a message in the input message queue (S29, Qin non-empty), a message reception process is performed (S30), and the process ends.

  If the interrupt type is management communication (S29, management communication), a management communication slave process is performed (S31), and the process ends. If the interrupt type is the expiration of the management communication timer TMC (S29, TMC expiration), the system switching process is performed (S32), and the process is terminated.

  FIG. 8 is a flowchart for explaining details of the discovery process (S26) shown in FIG. This discovery process may not be performed if all the contracted logical links are in the registered state, and if all the logical links contracted for a certain PON are in the registered state, steps S41 and S42 are performed. The transmission of the discovery gate message for the PON (virtual OSU) may be skipped.

  First, discovery gate messages for OSUk (k = 1, 2,..., N) are sequentially constructed and placed in the output message queue Qeg. At this time, the start time is based on TEk or the later of the current time (S41). Then, the end of the discovery period is set as a new TEk (S42). That is, the arrangement of the discovery window is determined on the basis of the immediately preceding TEk or the later of the current time, taking into account the assumed RTT range, the congestion status of register requests, and the like.

  Finally, the discovery timer TD is set (S43), and the process ends. The value set in the discovery timer TD is discovery_interval which is a predetermined period. That is, discovery gate messages for OSUs are sequentially issued at a predetermined cycle.

  FIG. 9 is a flowchart for explaining details of the RP timeout process (S27) shown in FIG. This RP timeout process is a process when no report message is received during the reception window period due to the expiration of the timer TLij.

  First, it is determined whether or not the logical link is being registered by referring to LLstat of LLTij (S51). If the logical link is not registered (S51, No), a value obtained by subtracting the latest report reception time RPtime from the current time ctime is substituted into RPlossT (S52). This RPlossT is the elapsed time from the latest report reception time.

  If RPlossT is greater than 1 second (S53, Yes), the process proceeds to step S56. If RPlossT is 1 second or less (S53, No), in order to receive the report message from the logical link j again, a gate message for OSUi, LLIDj is constructed and placed in the output message queue Qeg. At this time, a report forced flag is set. The reception time is assigned as the start time is based on TEi or the current time, and the grant length is an amount by which only the report frame can be transmitted (S54).

  Then, the start of laser off of the grant is set as a new TEi, a timer TLij is set at a time when a certain amount of blur time is anticipated (S55), and the process is terminated.

  If the logical link is being registered in step S51 (S51, Yes), a deregistering process described later is performed (S56), the standby control unit is notified of the deregistering of LLIDij (S57), and the process is terminated. .

  FIG. 10 is a flowchart for explaining details of the management communication master process (S28) shown in FIG. First, the current time ctime is notified to the control unit of the other system (standby system) to notify that the active system is normal, the status is inquired to the control unit of the standby system, the presence / absence of a duplex system, The normal / abnormal of (standby system) is recorded (S61).

  Next, it notifies the optical switch 11a and the concentrator 13a that the operation control unit is normal, inquires about the states of the optical switch 11a and the concentrator 13a, and records them (S62). Finally, the management communication timer (TMC) is set (S63), and the process ends.

  FIG. 11 is a flowchart for explaining details of the management communication slave process (S31) shown in FIG. It is determined whether the process is a scheduled confirmation process or an individual process (S71). If it is a regular confirmation process (S71, regular confirmation), a regular confirmation process is performed (S72), and the process is terminated. This scheduled confirmation process responds to the standby system status inquiry shown in step S61 of the management communication master process of FIG.

  If the process is an individual process (S71, individual process), the individual process is performed (S73), and the process is terminated. This individual process includes a management communication process for synchronizing the state of the standby system with the active system, and a system switching instruction. Details of these processes will be described later.

  FIG. 12 is a flowchart for explaining the details of the regular confirmation process (S72) shown in FIG. First, the current time ctime of the active system received from the active system control unit is set to the current time ctime of the standby system to match the clock with the active system and to respond to inquiries from the active system control unit (S81).

  Finally, it notifies the optical switch 11a and the concentrator 13a that the standby control unit is normal, inquires about the states of the optical switch 11a and the concentrator 13a, and records them (S82). Exit.

  FIG. 13 is a flowchart for explaining details of the control system switching (on) processing (S32) shown in FIG.

  In the control system switching (on) process, the process is shifted to the operational system (S95), the discovery timer (TD) is set, the management communication timer (TMC) is set (S96), and the process ends.

  FIG. 14 is a flowchart for explaining details of the message reception process (S23) shown in FIG. First, the head message is taken out from the input message queue Qin (S101), and branched depending on the message type (S102).

  If the message type is a register request message (S102, register request), register request processing is performed (S103), and the process proceeds to step S110. If the message type is a register confirmation message (S102, register confirmation), register confirmation processing is performed (S104), and the process proceeds to step S110.

  If the message type is a deregister request message (S102, deregister request), deregister processing is performed (S105), and the process proceeds to step S110. If the message type is a report message (S102, report), a report reception process is performed (S106), a bandwidth allocation process is performed (S107), and the process proceeds to step S110.

  If the message type is an OAM message (S102, OAM), OAM message processing is performed (S108), and the process proceeds to step S110. If the message type is an OSU management message (S102, OSU management), OSU management message processing is performed (S109), and the process proceeds to step S110.

  In step S110, it is determined whether or not the input message queue Qin is empty. If the input message queue is not empty (S110, No), the process returns to step S101, the next message is extracted, and the subsequent processing is performed. If the input message queue Qin is empty (S110, Yes), the process ends.

  FIG. 15 is a flowchart for explaining the details of the register request process (S103) shown in FIG. When a register request message is received from the home side device with OSUi and MAC address m, it is first determined whether or not the transmission source ONU is valid (S111).

  If the transmission source ONU is valid (S111, Yes), a new LLIDj is assigned to OSUi. Then, the logical link state LLstat is set during registration for LLTij, the report information RPinfo is set to NULL, and (T2-T1) is set as the round trip propagation time RTT (S112). Here, T1 is a time stamp recorded in the message by the home side device, and T2 is a time stamp added by OSUi at the time of reception.

  Next, it is determined whether or not the control unit is an active system (S113). If the control unit is a standby system (S113, No), the process is terminated as it is.

If the control unit is an active system (S113, Yes), a register message (w / Ack) for OSUi is constructed and placed in the output message queue Qeg (S114).
At this time, the transmission destination address DA of the register message is set to m, and the LLID is set to j. Note that (w / Ack) indicates a register message to which Ack is added.

  Next, a gate message for OSUi and LLIDj is constructed and placed in the output message queue Qeg. At this time, the start time is based on TEi or the current time, and the grant window is assigned as a reception window as an amount capable of transmitting only the register confirmation frame (S115).

  It should be noted that the arrangement of the reception window takes into account the accuracy error and, of course, does not overlap the reception window of other logical links even if the laser on / off overlap is allowed. The RTT and ONU processing times are arranged ahead of the scheduled transmission time.

  Then, the start of laser off of the grant is set as a new TEi, and a time when a certain amount of blur time is expected is set in the timer TLij (S116), and the process is terminated.

  If the ONU is not valid in step S111 (S111, No), a register message (w / Nack) for OSUi is constructed and placed in the output message queue. At this time, the transmission destination address of the register message is set to m (S117), and the process ends.

  FIG. 16 is a flowchart for explaining details of the register confirmation processing (S104) shown in FIG. It is assumed that a register confirmation message is received from OSUi and LLIDj. First, an RTT update process, which will be described later, is performed, and it is determined whether the RTT update process has ended normally (S121). If the RTT update process does not end normally (S121, NG), the process ends.

  If the RTT update process ends normally (S121, OK), the logical link state LLstat is set to registered for LLTij, and the current time ctime is set to the latest report reception time RPtime (S122). Then, it is determined whether or not the control unit is an active system (S123). If the control unit is a standby system (S123, No), the process is terminated as it is.

  If the control unit is an active system (S123, Yes), a gate message for OSUi and LLIDj is constructed and placed in the output message queue Qeg. At this time, a report forced flag is set. The start time is based on TEi or the current time, and the grant window is assigned as a reception window as an amount that can transmit only the register confirmation frame (S124).

  Then, the start of laser off of the grant is set as a new TEi, a time when a certain blur time is expected is set in the timer TLij (S125), the OAM process activation timer TOAMij is set (S126), and the process is terminated.

  FIG. 17 is a flowchart for explaining details of the deregister processing (S105) shown in FIG. In this process, OSUi and LLIDj are deregistered. First, a deregister message for OSUi and LLIDj is constructed and placed in the output message queue Qeg (S131). Then, the logical link state LLstat is set to NULL for LLTij, the logical link j is released (S132), and the process is terminated.

  FIG. 18 is a flowchart for explaining details of the report reception process (S106) shown in FIG. In this process, a report is received from OSUi and LLIDj. First, for the LLTij, the latest report reception time RPtime is set to the current time ctime (S141), RTT update processing described later is performed, and it is determined whether the RTT update processing has ended normally (S142).

  If the RTT update process ends normally (S142, OK), the total of the ONU upstream queue length queueK_report included in the report information extracted from the input message queue Qin is substituted into the report information RPinfo for LLTij (S143). ), The process is terminated.

  If the RTT update process does not end normally (S142, NG), the process ends.

  FIG. 19 is a flowchart for explaining details of the RTT update processing. First, a value obtained by subtracting the time stamp T1 recorded in the message by the home side device from the time stamp T2 added by the OSUi at the time of reception is set as a new RTT (newRTT) (S151). Then, it is determined whether the absolute value of the difference between newRTT and RTT exceeds the allowable drift value DRIFTmax (S152).

  If it is equal to or less than the allowable drift value DRIFTmax (S152, No), the RTT is set to newRTT with respect to LLTij (S153), and the process is terminated assuming that the RTT update process has ended normally. If the allowable drift value DRIFTmax is exceeded (S152, Yes), it is determined whether or not the control unit is an active system (S154).

  If the control unit is an active system (S154, Yes), the deregistration process shown in FIG. 17 is performed (S155), the deregistration of LLIDij is notified to the standby control unit (S156), and the RTT update process ends normally. If not, the process ends. If the control unit is a standby system (S154, No), it is assumed that the RTT update process has not ended normally, and the process ends.

  When the OSU switching process (S221) and the OSU switchback process (S222) can refer to the transition period necessary for switching the path and the reception time of the control frame corresponds to the transition period, in S152, the drift is allowed. The value may be increased by taking into account fluctuations associated with path switching, or the comparison may be ignored (S152 is always No).

  FIG. 20 is a flowchart for explaining details of the bandwidth allocation processing (S107) shown in FIG. This process allocates a bandwidth to OSUi and LLIDj. First, the time REPORT_length required for report frame transmission is added to LLTij report information RPinfo (ONU upstream queue length) to set Len, and the synchronization period Sync is added to the laser on period Ton to set the overhead time OVL. Then, the smaller value of the sum of Len, OVL and laser off period Toff and the upper limit value GLmax of the grant length is defined as the grant length GL.

  The sum of the last allocation time TEi of OSUi and the burst gap burst_gap is TSi, and the sum of the current time ctime, RTT, and the processing time proc_time of the home side device is TSc. The latest time among TSi and TSc is set as TS (S162).

  Next, a gate message for OSUi and LLIDj is constructed and placed in the output message queue Qeg. At this time, a report forced flag is set. The start time is a value obtained by subtracting RTT from TS, and a grant window is assigned as a grant length GL (S163).

  Then, the value obtained by subtracting Toff from the value obtained by adding GL to TS is set as the final allocation time TEi (S164), the timer TLij is set at the time when some fluctuation is expected in TEi (S165), and the process is terminated.

  When the OSU switching process (S221) and the OSU switchback process (S222) can refer to the transition period necessary for switching the route, the TS is adjusted so that the grant period and the transition period do not overlap. Also good.

  Further, the bandwidth allocation process may be performed in cooperation with virtual OSUs concentrated on the same uplink. In particular, if the bandwidth allocation of each PON line is performed based on uplink uplink bandwidth allocation so that the uplink frame output by the virtual OSU can pass without staying in the concentration section, the FIFO of the concentration section can be changed. The amount may be small, and the time for the user frame to pass through the station side device can be shortened.

  FIG. 21 is a flowchart for explaining the details of the OAM message processing (S108) shown in FIG. Assume that an OAM message is received from OSUi and LLIDj. First, when the OAM message is an OAM link connectivity confirmation message, OAMt is set to the current time ctime for LLTij (S171), and it is determined whether or not the control unit is an active system (S172).

  If the control unit is not the active system (S172, No), the process is terminated as it is. If the control unit is an active system (S172, Yes), processing according to the content of the OAM message is performed (S173), and the processing is terminated.

  FIG. 22 is a flowchart for explaining the details of the OSU management message processing (S109) shown in FIG. Assume that an OSU management message is received from OSUi. First, it is determined whether or not the control unit is an active system (S211). If the control unit is a standby system (S211, No), the process is terminated as it is.

  If the control unit is an active system (S211, Yes), it is determined whether a failure notification has been received (S212). If there is a failure notification (S212, Yes), an OSU switching process described later is performed (S213), and the process is terminated. If there is no failure notification (S212, No), processing according to the message content is performed (S214), and the processing is terminated.

  FIG. 23 is a flowchart for explaining the details of the OAM process (S24) shown in FIG. This OAM process is performed independently for each virtual OSU (denoted i) and logical link (denoted j), and ONU setting and status confirmation are performed by OAM communication with the corresponding ONU.

  First, it is determined whether or not OAMt of LLTij is NULL (S191). If the OAMt of LLTij is NULL (S191, Yes), this indicates that the OAM link is in the initial state, so an OAM loopback test is performed (S192), and it is determined whether or not it is successful (S193).

  If the OAM loopback test is successful (S193, Yes), the ONU corresponding to PONi and LLIDj is initialized (S194), and the OSUi is allowed to communicate between the PON side and the uplink side of LLIDj (S195). . Then, the OAM process activation timer TOAMij is set (S196), and the process ends.

  If the OAM loopback test is not successful (S193, No), deregister processing is performed (S197), the standby control unit is notified of LLIDij deregister (S198), and the processing is terminated.

  In step S191, if the OAMt of LLTij is not NULL (S191, No), it is determined whether or not the value obtained by subtracting the OAMt of LLTij from the current time ctime is smaller than OAMmaxinterval (S199). This OAM max interval is determined in advance, and this value determines whether or not the OAM communication is interrupted.

  If the value obtained by subtracting OLMt of LLTij from the current time ctime is smaller (S199, Yes), an OAM link connectivity confirmation message is sent to ONUi (S200), and an OAM process activation timer TOAMij is set (S201). ), The process is terminated.

  If the value obtained by subtracting OLTt of LLTij from the current time ctime is not smaller (No in S199), deregister processing is performed (S202), the deregistration of LLIDij is notified to the standby control unit (S203), The process ends.

  FIG. 24 is a flowchart for explaining details of the operation IF process (S25) shown in FIG. If the instruction by the operation IF is an OSU switching instruction, an OSU switching process is performed (S221), and the process ends.

  If the instruction by the operation IF is an OSU failback instruction, an OSU failback process is performed (S222), and the process ends. If the instruction by the operation IF is a control system switching instruction, a control system switching process is performed (S223), and the process is terminated. If the instruction by the operation IF is other, the process according to the instruction is performed (S224), and the process is terminated.

  FIG. 25 is a flowchart for explaining details of the OSU switching process (S221) shown in FIG. Assume that OSUi is switched. First, the optical switch 11a is instructed to switch the PON line from OSUi to OSU N + 1 (S231), and the concentrator 13a is instructed to set the filter unit N + 1 (46a-N + 1) of the concentrator 13a to the setting of the filter unit i. (S232).

  Then, the setting related to the virtual OSUi for OSU conversion is set to OSUmap (i) = N + 1 (S233), the switching of the OSUi is notified to the standby control unit (S234), and the process ends.

  FIG. 26 is a flowchart for explaining details of the OSU switchback processing (S222) shown in FIG. Assume that OSUi is switched back. First, the optical switch 11a is instructed to return the PON line i to the original OSUi (S241). Then, the setting related to the virtual OSUi for OSU conversion is set to OSUmap (i) = i (S242), the standby control unit is notified of the OSUi switchback (S243), and the process is terminated.

  In the OSU switching process (S221) and the OSU switchback process (S222), the transition period during which the path is switched is recorded in a global variable so that it can be referred to from other processes such as the bandwidth allocation process (S107). Good.

  FIG. 27 is a flowchart for explaining details of the control system switching (off) processing (S223) shown in FIG. First, a system switching instruction is given to the control unit of the other system (standby system) (S251), and the system shifts to the standby system (S252). Then, the management communication timer (TMC) is set (S253), and the process ends.

  As described above, according to the station side apparatus in the present embodiment, the control unit 13a collectively controls the OSU1 to OSUNN + 1 (12-1 to 12-N + 1). It became possible to reduce the cost. Since the MPCP frame transmission / reception timing has a large allowable range, there is no problem even if the transmission / reception timing slightly varies. Therefore, even if the termination of the MPCP frame is performed by one control unit, there is no particular problem.

  Further, since the control unit 13a can perform the OSU redundancy switching only by changing the communication path while maintaining the ONU registration state, the redundancy switching can be performed quickly and the OSU abnormality is affected. Is gone.

  In addition, since redundant switching can be performed only by preparing one spare OSU for N OSUs, it is possible to improve the fault tolerance performance without significantly impairing the economy.

(Second Embodiment)
FIG. 28 is a block diagram illustrating a schematic configuration of a station-side apparatus according to the second embodiment of the present invention. In this station side apparatus 1b, the OSU and the PON line are duplexed, and one ONU 2 is connected to two OSUs via different PONs. Further, the station side device 1b has 2N OSUs, and can terminate N sets of duplicated PON lines. The duplexed OSU and PON lines are divided into two systems, the A system and the B system. The A system OSU and PON lines are marked with "A", and the B system OSU and PON lines are used. "B" shall be attached to.

  The station side device 1b performs overall control of the OSUs 1A to NA (12-1A to 12-NA) and the OSUs 1B to NB (12-1B to 12-NB), the concentrator 13b, and the station side device 1b. Part 14. Each configuration of the OSU is the same as the configuration of the OSU in the first embodiment shown in FIG.

  FIG. 29 is a block diagram showing a configuration example of the concentrator 13b in the second embodiment of the present invention. Compared with the concentrator 13a in the first embodiment shown in FIG. 4, the number of filter units and FIFOs is N, the number of OSU IF units is 2N, and the filter unit 1 to N (46b-1 to 46b-N) are connected to the A-system OSU IF unit and the B-system OSU IF unit. When the control unit 14 performs control via the control IF unit 44, the filter units 1 to N (46b-1 to 46b-N) use any path of the A system and the B system as the active system, Determine whether the path is a standby system. By default, the A system is the active system, and the OSU conversion is (virtual OSUi real OSUiA).

  The filter units 1 to N (46b-1 to 46b-N) connect the active system path in the downstream direction and disconnect the standby system path. Also, both paths are connected in the upstream direction. Hereinafter, this redundant configuration is referred to as a 1: 1 redundant configuration.

  Further, when both paths are connected in the down direction, the active path is connected in the up direction, and the standby path is disconnected, this redundant configuration is called a 1 + 1 redundant configuration.

  The control unit 14 assumes N virtual OSUs, and associates the virtual OSUi with the active real OSU by OSU conversion. That is, either OSUiA or OSUiB is linked to OSUi. The control unit 14 performs OSU redundancy switching according to the following procedure.

  First, an OAM message notifying the ONU of system switching is sent to the virtual OSUi. That is, it broadcasts to the ONU in operation via the PON line iA. Then, the I / O control unit 54 is instructed to the concentrator 13b to set the path connection / disconnection of the filter unit of the concentrator 13b so that the A system is the standby system and the B system is the active system.

  Then, the setting related to the virtual OSUi for OSU conversion is changed to (virtual OSUi⇔real OSUiB). Then, an OAM message notifying the ONU of the system switching is sent to the virtual OSUi again. That is, the OAM message is broadcast via the PON line iB.

  The operation of the control unit 14 described above corresponds to a 1: 1 redundant configuration, but can also correspond to a 1 + 1 redundant configuration. In this case, without performing OSU conversion, the control unit 14 directly interfaces with 2N real OSUs, establishes and operates logical links and OAM links separately from each duplexed OSU A system and B system, Communicate.

  Further, as an intermediate configuration between the 1: 1 redundant configuration and the 1 + 1 redundant configuration, the standby system establishes and maintains the logical link and the OAM link, but it is also possible not to pass the user frame. In this case, only the MPCP frame and the OAM frame are transmitted to the standby PON line.

  In the case of this redundant configuration and the 1 + 1 redundant configuration, the operating system may be determined by comparing the number of established links.

  As described above, according to the station side device 1b in the present embodiment, in addition to the effects described in the first embodiment, the PON line can also be made redundant, so that the fault tolerance performance is further improved. It became possible to make it.

  As a modification of the 1: 1 redundant configuration of the present embodiment, the configuration shown in FIG. 33 is also possible. This uses a 2 × N type optical coupler, and the PON line downstream from the coupler constitutes a single system and the OSU constitutes a duplex system. It is possible to improve only the fault tolerance performance of the OSU without significantly increasing the PON line cost.

(Third embodiment)
FIG. 30 is a block diagram showing a schematic configuration of the station side apparatus according to the third embodiment of the present invention. In the station side device 1c, the concentrator and the control unit of the station side device 1a described in the first embodiment are duplicated. That is, the 1-line concentrator 13a-1, the 2-system concentrator 13a-2, the 1-system controller 14-1, and the 2-system controller 14-2 are included.

  FIG. 31 is a diagram illustrating a configuration example of the optical switch 11b according to the third embodiment of the present invention. This optical switch 11b differs from the optical switch 11a in the first embodiment shown in FIG. 2 only in that a system selection unit 25 is added. The system selection unit 25 selects either a control signal from the first system control unit 14-1 or a control signal from the second system control unit 14-2.

  The system selection unit 25 continuously performs management communication with the first system control unit 14-1 and the second system control unit 14-2, autonomously determines the control unit that is the active system, and operates. A control signal from the control unit serving as a system is output to the actuator 21.

  The control IF unit 32 of the OSU 12 shown in FIG. 3 has an interface between the 1-system control unit 14-1 and the 2-system control unit 14-2. The control IF unit 32 continuously performs management communication with the first system control unit 14-1 and the second system control unit 14-2 via this interface, and autonomously controls the control unit that is in the operation system. Judgment. The control IF unit 32 processes only the signal from the active control unit, and outputs the same signal to both control units.

  Similarly, the concentrating IF unit 31 has an interface between the first system control unit 14-1 and the second system control unit 14-2. The concentrator IF31 outputs the upstream frame to both the concentrators 13a-1 and 13a-2, and outputs the downstream frame sent from both systems to the FIFO2 (38). Since no downstream frame is sent from the concentrator of the standby system, no collision occurs. Further, an input signal from the standby system may be blocked in preparation for an abnormality in the concentrator. In this case, the control unit 14-1 or 14-2 performs system selection via the control IF unit 32.

  The control IF unit 44 of the concentrator 13a shown in FIG. 4 has an interface between the first system control unit 14-1 and the second system control unit 14-2. The control IF unit 32 continuously performs management communication with the first system control unit 14-1 and the second system control unit 14-2 via this interface, and autonomously controls the control unit that is in the operation system. Judgment. The control IF unit 44 processes only the signal from the active control unit, and outputs the same signal to both control units.

  When the concentrator is a double system, the path selection is reflected in the path connection / path disconnection setting in the filter unit 46a. The active control unit gives this instruction via the control IF unit 44. For example, when the concentrating unit supports 1: 1 redundancy, both the upstream path and the downstream path of all the filter units may be set to be disconnected for the standby concentrating unit. Further, in the case of supporting 1 + 1 redundancy, it is only necessary to set the downstream paths of all the filter units to disconnection for the standby line concentrator.

  The IO control unit 54 of the control unit 14 shown in FIG. 5 can interface with the IO control unit of the other control unit. The CPU 51 performs management communication with another system CPU via the IO control unit 54 and autonomously determines whether to be an active system or a standby system. In some cases, the operation / standby system is explicitly instructed from the operation IF. Since signals from each part of the station side device are input to both systems, it is possible to trace the state change in the station side device and the state of the PON line even in the standby system.

  With the redundancy of the concentrator, 1: 1 redundancy, 1 + 1 redundancy, and load distribution are possible. Here, the load distribution means that the OSU is divided into two groups (group A and group B). In normal times, the 1-system concentrator 13a-1 collects the group A, and the 2-system controller 13a-2 Concentrate Group B. For example, when a failure occurs in the first line concentrator 13a-1 or the uplink, the second line concentrator 13a-2 switches so as to collect both groups. When performing 1: 1 redundancy or load balancing, the uplink transmission / reception unit 41 of the concentrator 13a performs management communication via the uplink, monitors the uplink state, and receives a failure notification from the opposite device. When there is an abnormality, a corresponding alarm is notified to the control unit 14.

  When recognizing an abnormality in the active line concentrator, the control unit 14 switches to another system when the other line concentrator is normal. This switching is performed by changing the path setting of the filter unit 46a of the concentrator 13a according to the redundant configuration (1: 1, 1 + 1, load distribution). In the case of load distribution, the control unit 14 receives an instruction from the outside via the operation IF, and changes the path setting of the filter unit 46a of the line concentrator 13a to restore the load distribution state.

  When the control unit 14 is switched, the band allocation is taken over. There are an operation of strictly taking over the past allocation, and an operation of performing a new bandwidth calculation without taking over the past allocation. In the former case, the active control unit notifies the standby control unit including the OSU that transmitted the control message transmitted to each PON line.

  As described above, according to the station side device in the present embodiment, the control unit and the concentrator in the station side device in the first embodiment are made redundant, so in the first embodiment In addition to the effects described, it has become possible to further improve the fault tolerance performance without significantly impairing the economy.

(Fourth embodiment)
FIG. 32 is a block diagram showing a schematic configuration of the station side apparatus according to the fourth embodiment of the present invention. In this station side apparatus 1d, the concentrating part and control part of the station side apparatus 1b demonstrated in 2nd Embodiment are duplicated. That is, the 1-line concentrator 13b-1, the 2-system concentrator 13b-2, the 1-system controller 14-1, and the 2-system controller 14-2 are included. The system selection control is the same as that described in the third embodiment. Therefore, detailed description here will not be repeated.

  Regarding the setting of the duplex configuration, the filter units 1 to N of the 1-line concentrator 13b-1 have path connection / path disconnection determined by the duplex configuration of OSU and PON lines, and path connection / path determined by the duplex configuration of the concentration unit It is sufficient to reflect the disconnection AND (path disconnection except when both are connected by a path). The same applies to the second system concentrator 13b-2.

  As described above, according to the station side apparatus in the present embodiment, the control unit and the concentrator in the station side apparatus in the second embodiment are made redundant, so in the second embodiment, In addition to the effects described, it has become possible to further improve the fault tolerance performance without significantly impairing the economy.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  1a, 1b, 1c, 1d Station side device, 2 ONU, 3 PON line, 4 optical coupler, 11a, 11b optical switch, 12 OSU, 13a, 13b Concentrator, 14 controller, 21 actuator, 22 movable mirror, 23, 24 collimating lens, 25 system selection unit, 31 concentrating IF unit, 32 control IF unit, 33 reception processing unit, 34 transmission processing unit, 35 PON transmission / reception unit, 36 local control unit, 37, 38, 47 FIFO, 41 uplink transmission / reception Unit, 42 Downstream distribution unit, 43 Concentration control unit, 44 Control IF unit, 45 OSU IF unit, 46 Filter unit, 48 Selector, 51 CPU, 52 ROM, 53 RAM, 54 IO control unit, 55 OSU IF unit, 56 Sharing RAM, 57 Clock / Timer.

Claims (4)

A station-side device accommodating a plurality of passive optical networks,
A plurality of optical line units connected to each of the plurality of passive optical networks;
Including first control means and second control means for performing registration processing and bandwidth allocation processing of a plurality of home-side devices connected to the plurality of passive optical networks,
The first control means and the second control means generate a control message for performing the registration process and a control message for performing the bandwidth allocation process,
A station-side device that switches from one control means to the other control means while maintaining the registered state of the plurality of home-side devices. A control method for a station side device accommodating a plurality of passive optical networks,
A step in which the first control means and the second control means sequentially perform registration processing and bandwidth allocation processing of a plurality of home-side devices connected to the plurality of passive optical networks by sequentially generating control messages; ,
And switching the control means while maintaining the registered state of the plurality of home-side devices. A computer program for causing a computer to execute a method for controlling a station-side device accommodating a plurality of passive optical networks, the computer comprising:
Sequentially performing registration processing and bandwidth allocation processing of a plurality of home-side devices connected to the plurality of passive optical networks by sequentially generating control messages;
A computer program for executing a control message for performing a registration process and a step of changing a communication path of a control message for performing the bandwidth allocation process while maintaining a registration state of the plurality of home-side devices. A computer program for causing a first control means and a second control means to execute a method for controlling a station-side device accommodating a plurality of passive optical networks, the first control means and the second control. As a means,
Sequentially performing registration processing and bandwidth allocation processing of a plurality of home-side devices connected to the plurality of passive optical networks by sequentially generating control messages;
A computer program for executing a step of switching control means while maintaining a registered state of the plurality of home devices.

Images