In Uni 1+1 Sig APS switching mode the Tx data is sent on the active link only (it is not bridged to both links simultaneously). 1+1 signaling, however, is used for full interoperability with signaling-compliant 1+1 architectures.
In the ingress direction (Rx), the decision to accept data from either the working or protection circuit is based on both locally detected failures/degradation and on what circuit the far-end is listening on (as indicated in the K bytes). Although it is not required in the APS standards, the system’s implementation of Unidirectional 1+1 Signaling APS uses standards based signaling to keep both the Rx and Tx on the same circuit / port. If the far-end indicates that it has switched its active receiver, then the local node also switches its receiver (and Tx) to match the far-end. If the local Rx changes from one circuit to another it notifies the far end using the K bytes.
In the egress direction (Tx), the data is only transmitted on the active circuit. If the active Rx changes, then Tx also changes to the same circuit.
Because the router transmits on active circuits only and keeps active TX and RX on the same port, both local and remote switches are required to restore the service. For a single failure a data outage is limited to a maximum of 100 milliseconds.
The APS channel (bytes K1 and K2 in the SONET header – K bytes) exchanges requests and acknowledgments for protection switch actions. In Unidirectional 1+1 Signaling APS switching mode, the router sends correct status on the K bytes and requires the far-end to also correctly update/send the K-bytes to ensure that data is transmitted on the circuit on which the far-end has selected as its active receiver.
Line alarms are processed and generated independently on each physical circuit.
In Unidirectional 1+1 Signaling APS switching mode:
K-bytes are generated/transmitted based on local request/condition only (as required by the APS signaling).
Local request priority is compliant to 1+1 U-APS specification.
RX and TX are always forced on to the same (active) circuit (bidirectional). This has the following restrictions:
If an APS switch is performed because of a local condition, then the TX direction is moved as well to the newly selected RX circuit (old inactive). The router sends L-AIS on the old active TX circuit to force the remote end to APS switch to the newly active circuit. Note that some local request may not cause an APS switch when a remote condition prevents both RX and TX direction to be on the same circuit (for example an SD detected locally on a working circuit does not cause a switch if the protection circuit is locked out by the remote end).
If the remote end indicates an APS switch and the router can RX and TX on the circuit newly selected by the remote end, then the router moves its TX direction and performs an APS switch of its RX direction (unless the router already TX and RX on the newly selected circuit).
If the remote end indicates an APS switch and the router cannot RX and TX on the circuit newly selected by the remote end (for example because of a higher priority local request, like a force request or manual request, and so on), then L-AIS are sent on the circuit newly selected by the remote end to force it back to the previously active circuit.
The sent L-AIS in the above cases can be either momentary or persistent. The persistent L-AIS is sent under the following conditions:
On the protection circuit when the protection circuit is inactive and cannot be selected because of local SF or Lockout Request.
On the working circuit as long as the working circuit remains inactive because of a local condition. The persistent L-AIS is sent to prevent revertive switching at the other end.
In all other cases a momentary L-AIS is sent. The system provides debugging information that informs operators about the APS-induced L-AIS.