STM-1:Automatic Protection Switching

Revision as of 15:39, 7 November 2012

Automatic Protection Switching on STM-1 connections is supported by Tmedia media gateways and Tdev development platforms.

SONET/SDH topologies

Ring Topology

The Automatic Protection Switching scheme differs depending on whether the equipment is part of a SONET/SDH ring or if it is used as terminal equipment. For a ring network, traffic is diverted from the failed data path toward the opposite direction of the ring through a backup path. Thus, the data can reach its destination by moving the other way around the ring (refer to Figure 12). Linear topology is simpler as both devices are facing each other and decide (mutually or not according to configuration option) to switch the data traffic to a backup fiber upon detection of a failure. As the TelcoBridges device is a terminating device (not an add/drop/mux), only the linear configuration is supported.

Linear Topology: 1+1, 1:n and 1:1

Linear topology leads to different configuration of protection fibers depending on the type of equipments and their capabilities (see Figure 13). The ‘1+1’ configuration means that a supplementary fiber pair (called the ‘protection channel’) is dedicated to protect the primary fiber pair (referred to as ‘working channel 1’). Depending of the operation (unidirectional or bidirectional) and switching configuration (revertive or non-revertive), the data/voice traffic is redirected to the protection fiber pair in case of problems detected on the primary fiber pair. Equipment of this type usually can terminate the totality of the SONET/SDH payload content to the local network. Some equipment has the capability to support more than one time the SONET/SDH payload (10 times an STM-1 rate for example) and to terminate it to a local network. In such product, more than one fiber pair is required in order to transport the payload. For these system, using a 1+1 redundancy scheme would be too much expensive as there would be a protection fiber pair for each working channel. For these applications, the ‘1:N’ scheme is used dedicated the supplementary fiber pair (the protection channel) to protect all of the N working channels (N=10 in the previous mentioned example). When one of the working channels is in problem, the voice/data traffic is redirected to the protection fiber pair. As an option, the protection channel may also transport extra traffic when it is not actively protecting a working channel (with the consequence of loosing this extra channel when a protection switch is required). In modular 1:N system, the case where N=1 exists but still behaves as a simplification of a 1:N system. Thus, extra traffic would be still be allowed providing the hardware platform supports it.

The TelcoBridges equipment supports the STM-1/OC3 level of transports (155Mbps) in a single fiber. Since this data rate can contains the complete voice/data payload from the local side of the blade, the only supported redundancy scheme is 1+1.

Unidirectional vs bidirectional

This configuration parameter refers to the APS switching protocol established between the two facing devices. In a unidirectional operation mode, the device that detects a failure on a working channel no. 1 (let’s assume ‘Equipment E’) requests to the remote equipment (let’s assume ‘Equipment F’ to drive the signal that specific working channel data to the protection line.

Once this request is acknowledged, Equipment E now starts to ‘listen’ to the protection channel but still drives its signal on to the working channel. This configuration option makes the actual bidirectional voice/data path to be split into two different fibers (TX on working channel 1 and RX on protection channel). This methodology has the potential of surviving a dual fiber failure as long as both failures are not in the same directions. That means that Equipment F could later requests for the working channel 3 to be driven to the protection channel. In such event, the protection channel would transport working channel 1 in one direction and working channel 3 is the other direction.

When configured for bidirectional switch-over, the APS protocol will make sure that both direction of the failed working channel is switched to the protection fiber pair (regardless if one of the directions is still working). This makes the management of the fiber a little easier as a fiber can carry (or not) traffic rather than be transporting traffic in a single direction.

The TelcoBridges equipment supports both of these operation modes per a configuration setting when allocating the OC3/STM1 line interface.

Revertive vs non-revertive

This configuration options only specifies if the traffic is automatically switched back to its original channel once the failure is cleared. When in ‘non-revertive’ mode, the traffic will remain on the protection fiber until manually switched back or until a failure is detected on the protection fiber (the failure would need to be of higher priority than any failure already present on the working channel). That later case would then be considered as yet another protection switching (from the protection channel to the working channel). In ‘revertive’ mode, the switch-back operation is done automatically by the system. To avoid creating oscillation situation (where the voice/data is constantly switched back and forth to the protection channel), a configurable probing period (called the ‘wait-to-restore’) is applied. If no error occurred during the probing period (which could last from seconds to hours depending on user configuration), the switch-back will occur.

The TelcoBridges equipment supports both of these switching modes per a configuration setting when allocating the OC3/STM1 line interface. This setting applies to both the unidirectional and bidirectional mode of operations.