JP5323729B2 - Optical subscriber line termination equipment - Google Patents

Description

  The present invention is an optical access system composed of a PON (passive optical network) and a demultiplexing device (L2SW: layer 2 switch), and is highly reliable for a PON optical line terminal (OLT). Technology and link power saving technology.

  With the spread of full-scale FTTH, access and backbone networks play a role as an infrastructure of information society, and high reliability is desired. The amount of information handled in the information society is expected to increase approximately 200 times (compared to 2006) in 2025. As the number of network devices greatly increases with this information explosion, the power consumption of network devices is expected to increase approximately five times (compared to 2006). For this reason, high reliability and power saving are important issues in the network device itself and the network system.

  To save power in the network system, the LPI (low power idle) method (Non-Patent Document 1) that shifts some network device functions to the sleep state depending on the presence or absence of traffic, or the link rate is changed according to traffic An ALR (adaptive link rate) method (Non-Patent Document 2) has been proposed. The LPI method is a method for causing the link to sleep when there is no input traffic. When entering the sleep mode, the transmission side transmits a sleep signal and causes the transmitter to sleep. The receiving side receives the sleep signal and causes the receiver to sleep. In order to cope with the resumption of traffic, the reception side resumes reception after a certain sleep time, and the transmission side transmits a wake signal or a refresh signal (sleep continuation message) in accordance with the timing. On the other hand, the ALR system is a system that switches between a high-speed and high-power consumption mode and a low-speed and low-power consumption mode. The system that sleeps part of the parallel interface and adjusts the total speed is also a category of the ALR system. It is.

  In link aggregation (LAG), a technology that aggregates multiple physical links into one logical link, the LPI method and ALR method are applied to dynamically change the number of physical links according to traffic. Have been proposed (link power saving technique) (Non-Patent Documents 3 and 4).

  FIG. 1 shows a configuration example of a conventional L2SW incorporating a link power saving method. The L2SW includes a switch 11 that performs frame transfer, a control unit 12 that performs LAG control and link power saving control, and a physical port 13 that configures the LAG. Although the number of physical ports 13 of the L2SW in FIG. 1 is four, some devices have four or more physical ports. The switch 11 temporarily stores received frames, a forwarding table that is a correspondence table between destination addresses (or VID: VLAN ID) and physical ports, and a LAG forwarding that is a correspondence table between LAG ports and physical ports. Has a table. The LAG port is used as an identifier for managing a plurality of physical ports constituting the LAG. The control unit 12 sets each table and CPU (central processing unit) that instructs the physical port to LAG, LAG management DB (data base) that manages the LAG state, and dynamically determines the number of physical links according to traffic. It consists of a link power saving control unit to be changed.

  Next, the LAG setting of the control unit 12 will be described. When a plurality of physical ports are logically aggregated and used by the LAG, the administrator registers the physical ports desired to be aggregated between the opposing devices in the LAG management DB of the control unit 12 for each device. In FIG. 1, the administrator registers the physical ports 1, 2, and 3 of L2SW-1 and the three ports of physical ports 1, 2, and 3 of L2SW-2 in each LAG management DB. Although the same physical port number is connected between the opposing devices here, the physical port numbers to be connected need not be the same.

  Next, an L2SW link power saving method will be described. The link power saving control means of the control unit 12 periodically monitors the amount of frames and traffic accumulated in the frame buffer, and controls the number of effective links of the physical ports belonging to the LAG accordingly. This algorithm for determining the number of effective links is a core element of the link power saving method. In Non-Patent Document 3 and Non-Patent Document 4, the MIAD (multiplcative increase additive decrease) method that periodically monitors the frame buffer and determines the number of effective links by increasing or decreasing the buffer usage rate and the frame amount accumulated in the frame buffer. In addition, a STA (simple traffic average) method has been proposed that observes a plurality of past traffic and determines the number of effective links based on the average traffic. These systems have a feature that predicts future traffic from the amount of frames currently accumulated in the frame buffer and past traffic, and determines the number of effective links.

  FIG. 2 shows a configuration example of the optical access system. The optical access system is composed of a PON and an L2SW. In the PON, one OSU (optical subscriber unit) 21 passes through a plurality of ONUs (optical network units) 22, an optical fiber 23, and a 1 to N optical splitter 24. Thus, the network performs point-to-multipoint communication (N is a natural number). The OLT 20 includes a plurality of OSUs 21 and one control unit 25. The control unit 25 is used when the administrator sets each OSU or an ONU connected to each OSU. The L2SW 26 is connected to each OSU 21 and is a large-scale demultiplexing device that multiplexes frames from a plurality of OSUs 21.

  As a PON standard, EPON (Ethernet (registered trademark) passive optical network) defined in IEEE 802.3ah is standardized. In EPON, since signals transmitted from a plurality of ONUs are shared by the optical fiber from the optical splitter, a protocol (MPCP :) that controls signal transmission timing so as to avoid collision of signals transmitted from different ONUs. multi point control protocol). In uplink communication in MPCP, the OLT instructs the transmission start time and the transmission frame amount of each ONU, and each ONU transmits an uplink frame according to the instruction. Therefore, the OLT knows the upstream frame amount from each ONU in advance.

  Further, as shown in FIG. 3, there is also known a mode in which a switch 11, a control unit 12 and a physical port 13 are added to a conventional OLT to provide a multiplexing function, and the link between the OLT and the L2SW is connected by LAG. The advantages of this form are that (1) the bandwidth of the logical link is the sum of the bandwidths of the aggregated individual links, so that the bandwidth is expanded, and (2) of the aggregated individual links, Even if a failure occurs in some links, the communication can be continued on the remaining links, so that reliability is improved.

IEEE P802.3az (1) http://www.ieee802.org/3/az/public/mar08/chou 01 0308.pdf (2) http://www.ieee802.org/3/az/public/mar08 / healey 01 0308.pdf (3) http://www.ieee802.org/3/az/public/may08/taich 02 0508.pdf C. Gunaratne, K. Christensen, B. Nordman and S. Suen, “Reducing the Energy Consumption of Ethernet with Adaptive Link Rate (ALR),” IEEE Transaction on Computers, Vol.57, No.4, pp.448-461 , 2008. Shinji Aihara, Yutaka Fukuda, Kenji Kawahara, Yuji Oie, “Switch Power Saving Method Using Link Aggregation Technology”, IEICE Technical Report, Vol.106, No.461, pp.55-60, 2007. Hideaki Imaizumi, Masateru Minami, Tomohiro Nagata, Goro Kunigami, Kenichi Yamazaki, “Link Power Saving Method for Green Networks”, IEICE Technical Report, Vol.108, No.476, pp.105-110, 2009 . Tomoya Kanno, Yasunobu Kasahara, Shinichi Yoshihara, Toshiyuki Iwata, Yoichi Maeda, "Dynamic L2 distribution method to suppress frame order reversal", IEICE Technical Report, Vol.105, No.512, pp.19-24, 2006.

  A conventional OLT having a switch 11 that aggregates each OSU 21 and connecting the link between the OLT and the L2SW with LAG is effective for widening the band and increasing the reliability. Moreover, applying the link power saving method between the OLT and the L2SW is effective for power saving of the optical access system.

  However, when the power saving function as shown in FIG. 1 is added to the conventional OLT as shown in FIG. 3, even when the OLT switch 11 detects a failure, the OLT OSU 21 operates normally. The service was sometimes stopped. Further, in the conventional link power saving method, when the physical link that has been put to sleep is determined to be woken up by prediction, there is a possibility that performance may be deteriorated when traffic different from the prediction occurs. For example, if less traffic than expected occurs, the maximum power saving cannot be achieved, and if more traffic than expected, frame discard may occur. In addition, when power saving control is performed according to the arrived frame without using the prediction, it is necessary to store the arrived frame in the frame buffer until the physical link wakes up, which causes frame discard and delay. There was a problem.

  In order to solve the above-described problems, the present invention divides the OLT switch into two, connects the OLT and the L2SW with LAG, and applies a link power saving technique to achieve high reliability and high power saving effect. The object is to provide OLT.

The present invention accommodates a plurality of OSUs for controlling the transmission start time and the amount of transmission frames of uplink signals transmitted by each ONU for a plurality of ONUs connected via an optical fiber and an optical splitter. n is a natural number of 2 or more), a switch for transferring frames between each OSU and n physical ports, and n physical links corresponding to the n physical ports into one logical link In an optical subscriber line terminating device (OLT) having a control unit that performs LAG control to be aggregated, switches are divided into two switches, SW1 and SW2, respectively, and SW1 is m of n physical ports (m is n > M and a natural number of 1 or more), SW2 is connected to the remaining mn physical ports, and the OSU has selector means for switching the connection with SW1 or SW2. The control unit sets a threshold value smaller than the maximum switching capacity of SW1, and when the uplink signal capacity of the entire optical subscriber line terminating device is below the threshold value, distributing a signal to the SW1, and includes a power-saving control unit which performs control to sleep n-m number of physical ports connected to SW2 and SW2.

  In the optical subscriber line terminating device of the present invention, the control unit transmits the upstream signal to half of the selector means of the plurality of OSUs when the upstream signal capacity of the entire optical subscriber line terminating device exceeds the threshold value. Is assigned to SW1, and control is performed to assign the upstream signal to SW2 for the remaining half of the selector means of the plurality of OSUs.

  In the optical subscriber line terminating device of the present invention, the control unit gives priority to the selector means of the plurality of OSUs when the uplink signal capacity of the entire optical subscriber line terminating device exceeds the threshold value. In this configuration, a high uplink signal is assigned to SW1, and the remaining uplink signal exceeding the maximum switching capacity of SW1 is assigned to SW2.

  By fixing the high-priority uplink signal path to SW1 in this way, the high-priority uplink signal path is not switched in the sleep or failure detection of SW2, so the order of the high-priority uplink signals is changed. Reversal can be prevented. In addition, since the SW that processes the high-priority uplink signal does not enter the sleep mode, there is no possibility that delay or frame discard due to the sleep will occur in the high-priority frame.

  In the optical subscriber line terminating device of the present invention, the control unit is configured to switch SW1 and SW2 to the selector means of a plurality of OSUs when the uplink signal capacity of the entire optical subscriber line terminating device exceeds the threshold value. In this configuration, the upstream signal is controlled to be distributed to SW1 and SW2 so that the upstream traffic rate is the same. Thus, by distributing the loads of SW1 and SW2, it is possible to improve the band use efficiency.

  Note that priority control and control for distributing the loads of SW1 and SW2 may be combined.

  In the optical subscriber line terminating apparatus of the present invention, when the control unit detects a failure related to SW1 or SW2, the control unit performs control to distribute the upstream signal to the non-failure SW to the selector means of a plurality of OSUs. is there.

In the optical subscriber line terminating apparatus of the present invention, the OSU includes reception buffers for buffering downstream signals to be transmitted to the ONU corresponding to SW1 and SW2, and the selector means selects the reception buffer corresponding to each SW. a structure in which the control unit, to the selector means of the plurality of OSU, when SW2 is operated, a configuration for performing control to select with priority the receiving buffer corresponding to SW1.

Furthermore, when the control unit detects a failure related to SW1 or SW2 , the control unit gives the plurality of OSU selectors priority to the reception buffer until the reception buffer corresponding to the failure SW becomes empty. select in degrees, after the reception buffer is empty, which is configured to perform control to select only receive buffer corresponding to the SW of the non-disabled.

  In the optical subscriber line terminating device of the present invention, SW1 and SW2 have the same switching capacity.

  In the optical subscriber line terminating apparatus of the present invention, the function of the control unit is incorporated in each of SW1 and SW2, and the control unit function in SW1 and the control unit function in SW2 are connected to exchange control information. It is the structure to perform.

  Further, the SW1 link may be used in order to cause the transmission / reception port of the L2SW facing the SW2 and its physical port to sleep in conjunction with the SW2.

  In order to reduce delay due to sleep, for the upstream frame, the ONU may use the transmission waiting frame capacity for the OLT to estimate the upstream frame capacity received by the OLT in the next DBA cycle. In the case of EPON, the ONU reports the transmission waiting frame capacity to the OLT by a REPORT frame.

  When the transmission / reception capacity is relatively small, the OLT of the present invention can conserve power by putting the other switch path into a sleep state by concentrating on one switch path. Furthermore, it is possible to prevent the order of high priority frames from being reversed. There is no possibility of delay or frame discard due to sleep in high priority frames.

  Further, even when one of the SWs fails, the service is not stopped by performing the degeneration operation, so that a highly reliable OLT can be realized. Even when no failure occurs, SW1 and SW2 communication can be intentionally shifted to one of them, so that maintenance can be performed without stopping the service.

  Furthermore, at the time of power saving control, the frame discard can be minimized in the OSU, and the delay of the high priority frame can be minimized.

  In addition, it is possible to easily wake up the transmission / reception port of L2SW that faces SW2 and its physical port.

It is a figure which shows the structural example of conventional L2SW incorporating the link power saving technique. It is a figure which shows the structural example of an optical access system. It is a figure which shows the structural example of OLT with a multiple function. It is a figure which shows the structural example of Example 1 of OLT of this invention. It is a figure which shows the structure of switch management DB. It is a figure which shows the state transition of SW part of OLT. It is a figure which shows the example 1 of the frame distribution in OSU. It is a figure which shows the example 2 of the distribution of the frame in OSU. It is a figure which shows the structural example of Example 2 of OLT of this invention. It is a figure which shows the structural example of Example 3 of OLT of this invention.

FIG. 4 shows a configuration example of the first embodiment of the OLT of the present invention.
The OLT of this configuration example is provided with two switches (SW1) 11-1 and a switch (SW2) 11-2 in addition to the conventional OLT shown in FIG. And a power saving control unit is added. SW1 and SW2 are collectively referred to as an SW unit, and SW1 is a main system and SW2 is a subsystem. A physical port 13 is connected to each of SW1 and SW2. Here, an example is shown in which one physical port is connected to each SW. However, SW1 is m (n is a natural number of 2 or more) physical ports (m is n> m and 1 or more). SW2 may be connected to the remaining mn physical ports.

  The SW unit has a function of demultiplexing a frame between each OSU 21 and the L2SW, and includes a transfer table indicating a transfer destination of the frame and a frame buffer for temporarily storing the frame.

  The power saving control unit of the control unit 12 monitors the traffic rate of the SW unit and the amount of frames accumulated in the frame buffer, transmits a power saving control command to each OSU 21, and then connects to the SW2 and SW2 physical ports. To sleep / wake. The OSU 21 has a selection circuit (SEL: selector), and distributes the upstream frame to SW1 and SW2 for transmission. Also, the OSU 21 has reception buffers for SW1 and SW2, and temporarily selects and transfers downstream frames from both systems to the ONU after temporarily storing them in the buffers.

  For example, in the configuration of the OLT in FIG. 4, the switching capacities of SW1 and SW2 are 10 Gbps, the physical ports are 10 GbE (10 gigabit Ethernet (registered trademark)), the PON is EPON, and the number of OSUs is 16-20.

  As shown in FIG. 5, the switch management DB is composed of two tables, a transfer table and a switch management table. The forwarding table is a correspondence table between VLAN VID and OSU, and the administrator sets the correspondence. The CPU registers the switch management table of the switch management DB set by the administrator in the transfer table of the SW unit. The switch management table exclusively selects the physical port number of the SW unit and the main / sub system of the SW unit (here, SW1 is the main system, SW2 is the sub system, and the sub system is set to sleep). It consists of a main system / sub system, ACT / SLP / failure, and sleep state.

  FIG. 6 shows the state transition of the SW part of the OLT. The state is represented by a combination of SW1 and SW2, and SW1 and SW2 each have a state of “ACT”, “SLP”, or “failure”. Here, the sleep transition condition is a case where the upstream frame amount of the entire OLT is lower than the switching capacity of SW1. The wake transition condition is when the upstream frame amount of the entire OLT exceeds the switching capacity of SW1.

(Frame distribution example 1 in OSU)
FIG. 7 shows a frame distribution example 1 that each OSU sends to the SW unit. Here, it is assumed that the number of OSUs is 16 in one OLT.

  When both SW1 and SW2 are in the ACT state, OSU1 to OSU8 transmit frames to SW1, and OSU9 to OSU16 transmit frames to SW2, as shown in FIG. At this time, the correspondence between each OSU and the SW unit is fixed. When SW2 is in a failure detection or SLP state, OSUs 1 to 16 transmit frames to SW1 as shown in FIG.

(Frame allocation example 2 in OSU)
FIG. 8 shows a second example of frame distribution that each OSU sends to the SW unit. Here, an example is shown in which sorting is performed according to the priority of the frame.

  When both SW1 and SW2 are in the ACT state, each OSU transmits to SW1 in order from the frame with the highest priority as shown in FIG. 8A, and the remaining frames exceeding the switching capacity of SW1 are transmitted to SW2. As a result, a high priority frame flows through SW1, and a low priority frame flows through SW2. Therefore, when SW2 is in a failure or SLP state, each OSU transmits a frame to SW1 as shown in FIG. 8 (b). Therefore, there is little possibility that a high priority frame switches the link. The SW failure means all failures related to SW, for example, SW failure, physical port failure, wiring disconnection between OSU and SW, fiber disconnection connected to physical port, and the like.

(Frame distribution example 3 in OSU)
In the frame distribution example 1 shown in FIG. 7, when excessive traffic exceeding the switching capacity is concentrated on one of the SWs, there is a possibility that the frame is discarded. Therefore, as a frame distribution example 3, a method is adopted in which the correspondence between each OSU and the SW unit is dynamically controlled so that the traffic rates of SW1 and SW2 are approximately the same. For example, the control unit monitors the frame buffer and traffic rate of the SW unit, and instructs each OSU to send a frame to the SW unit so that the traffic rates of SW1 and SW2 are approximately the same. Thus, by distributing the loads of SW1 and SW2, it is possible to improve the bandwidth utilization efficiency of the entire SW unit.

(Frame allocation example 4 in OSU)
Frame distribution example 4 is a combination of frame distribution example 2 and frame distribution example 3. In this case, control is performed so that frames with higher priority are transmitted to SW1 in order and the traffic rates of SW1 and SW2 are approximately the same.

(Example of frame distribution in L2SW)
The L2SW connected to the OLT distributes the downstream frame from the upper network to SW1 and SW2. The downlink frame distribution method is classified according to whether or not the downlink traffic rate exceeds the switching capacity of the opposite OLT SW1. If so, the L2SW transmits a downstream frame to SW1. At this time, the receiving part of the OLT SW2 facing the transmission port of the L2SW sleeps. In the case of exceeding, the downstream frame is distributed to SW1 and SW2 using the VID so as to coincide with the upstream frame distribution example in the OSU (for example, the frame distribution example 2 in FIG. 8).

  When the L2SW distributes the downstream frame in the frame allocation example 2, it is highly likely that the downstream frame of the SW1 has a high priority. Therefore, the reception frame buffer for the SW1 is selected with priority. An example of this selection method is a weighted round robin method.

  Further, if the transmission / reception port of the opposite L2SW can be put to sleep in accordance with the sleep of SW2 and the physical port, it is more effective for power saving of the entire network. If this control is performed via SW1, control is possible even if SW2 is sleeping, which is effective. For example, when the L2SW transmission / reception port connected to SW2 is waked, the SW1 link may be used to match the wakeup timing of the SW2 and L2SW ports.

  When a failure of either SW is detected, each OSU processes the two received frame buffers accumulated at that time in a weighted round-robin method, and the SW receiving buffer in which the failure is detected becomes empty. After that, the other SW reception buffer accumulated after the failure detection is selected.

FIG. 9 shows a configuration example of the second embodiment of the OLT of the present invention.
In the OLT of the present embodiment, in the OLT of the first embodiment shown in FIG. 4, the control unit receives DBA information from each OSU 21, and the power saving control unit receives an uplink frame received by the OLT in the next dynamic bandwidth allocation (DBA) cycle. Estimate total traffic. Based on the estimation, the power saving control unit is configured to determine the transmission destination of the frame of each OSU 21 as SW1 or SW2 using a power saving control command.

FIG. 10 shows a configuration example of Embodiment 3 of the OLT of the present invention.
The OLT of the present embodiment is a part of the functions of the control unit 12 (when a failure of the power saving control function and SW1 or SW2 is detected in the OLT of the first embodiment shown in FIG. 4 and the second embodiment shown in FIG. In addition, the control function for distributing the upstream signal to the non-failed SW and the switch management table) are distributed and arranged in SW1 and SW2 for the selector means of a plurality of OSUs. Here, the control unit of SW1 and the control unit of SW2 are configured to exchange control information (for example, failure detection information, DBA information, switch management table information, etc.) with each other.

  Even in this configuration, the same functions as those of the first and second embodiments can be obtained. Furthermore, since the selector means of each OSU can be controlled directly from SW1 or SW2 without going through the control unit, higher-speed control than the OLTs of the first and second embodiments is possible. Further, the SW1 control unit, the SW2 control unit, and the OSUs of the third embodiment are connected to each other in a one-to-one relationship in FIG. 10, but are connected to a common bus, for example, and can exchange control information with each other. But there is.

11 Switch (SW1, SW2)
12 control unit 13 physical port 20 OLT (optical line terminal)
21 OSU (optical subscriber unit)
22 ONU (optical network unit)
23 Optical fiber 24 Optical splitter 25 Control unit 26 Demultiplexer (L2SW: layer 2 switch)

Claims (9)

For a plurality of ONUs (optical network units) connected via an optical fiber and an optical splitter, a plurality of OSUs (optical subscriber units) for controlling the transmission start time and the amount of transmission frames of upstream signals transmitted by each ONU are provided. Contain,
n physical ports (n is a natural number of 2 or more), a switch for transferring frames between each OSU and the n physical ports, and n physical links corresponding to the n physical ports In an optical subscriber line termination device (OLT) equipped with a control unit that performs LAG (link aggregation) control for aggregation into logical links,
The switches are divided into two switches, SW1 and SW2, respectively. SW1 is connected to m physical ports among the n physical ports (m is n> m and a natural number of 1 or more), and SW2 is the remaining n -Connect to m physical ports
The OSU includes selector means for switching connection with the SW1 or SW2.
The control unit sets a threshold value smaller than the maximum switching capacity of the SW1, and when the capacity of the uplink signal of the entire optical subscriber line termination apparatus is lower than the threshold value, the control unit sets the selector means of the plurality of OSUs. optical subscriber, characterized in that said distributing an uplink signal to the SW1, and equipped with a power saving control unit which performs control to sleep the n-m number of physical ports connected to the SW2 and the SW2 for Line termination equipment. In the optical subscriber line termination device according to claim 1,
The control unit distributes the upstream signal to the SW1 with respect to half of the selector means of the plurality of OSUs when the capacity of the upstream signal of the entire optical subscriber line terminating device exceeds the threshold. An optical subscriber line terminating apparatus characterized in that control is performed to distribute the upstream signal to the SW2 for the remaining half of the selector means of a plurality of OSUs. In the optical subscriber line termination device according to claim 1,
When the uplink signal capacity of the entire optical subscriber line terminating device exceeds the threshold, the control unit sends the high priority uplink signal to the SW1 for the selector means of the plurality of OSUs. An optical subscriber line terminating apparatus characterized by performing control to distribute and distribute the remaining upstream signal exceeding the maximum switching capacity of SW1 to SW2. In the optical subscriber line termination device according to claim 1,
When the uplink signal capacity of the entire optical subscriber line terminating device exceeds the threshold value, the control unit has the same upstream traffic rates of SW1 and SW2 for the selector means of the plurality of OSUs. The optical subscriber line terminating apparatus is configured to perform control to distribute the upstream signal to the SW1 and the SW2 so as to reach a certain level. In the optical subscriber line terminating device according to any one of claims 2 to 4,
The control unit is configured to perform control to distribute the uplink signal to a non-failure SW with respect to selector means of the plurality of OSUs when a failure related to the SW1 or the SW2 is detected. Optical subscriber line termination equipment. In the optical subscriber line termination device according to claim 1,
The OSU includes reception buffers that respectively buffer downstream signals to be transmitted to the ONUs corresponding to the SW1 and the SW2, and the selector means selects a reception buffer corresponding to each SW.
The control unit, said the to the plurality of OSU of selector means, if the previous SL SW2 is operating, is configured to perform control to select with priority the receiving buffer corresponding to the SW1 An optical subscriber line terminating device. In the optical subscriber line termination device according to claim 6,
When the control unit detects a failure related to the SW1 or the SW2, the control unit sets the reception buffer for the selector unit of the plurality of OSUs until the reception buffer corresponding to the failure SW becomes empty. the selected at priority, after the reception buffer is empty, optical network unit which is a configuration for performing control to select only receive buffer corresponding to the SW of the non-disabled. In the optical subscriber line termination device according to claim 1,
The SW1 and the SW2 are configured to have the same switching capacity. In the optical subscriber line termination device according to claim 1,
The function of the control unit is incorporated in each of the SW1 and SW2, and the control unit function in the SW1 and the control unit function in the SW2 are connected to exchange control information. An optical subscriber line terminating device.

Images