Some of the wavelengths in the link may be sections of a larger single wavelength network such as a Sonet/SDH ring or IP mesh. Services transported over this wavelength (T1 service in the diagram), may be protected by the network's inherent recovery mechanism. Other wavelengths within the link may carry high-speed point-to-point traffic as a "leased line" service. Services of this type may be combined to use one DWDM wavelength. These are called "sub-lambda services". If a single service utilizes a dedicated DWDM wavelength, it is referred to as a "wavelength service". As can be seen from the figure above, clients of these services are not protected and any failure in the DWDM link will interrupt their traffic.
The high availability of these critical services is achieved through the use of redundant resources (equipment and fiber) and protection systems that perform automatic protection switching (APS) when failure of a working resource is detected. There are several solutions for protection of DWDM networks. The choice of solution depends on the redundant resources being used as well as the required protection scheme. Resource redundancy may vary according to various factors: cost and geographical limitations, detection and repair time, etc. The protection scheme is determined by deployment considerations. For example, paths in which there are large length differences between the working and protection links may require a "dual-ended" protection scheme, to avoid problems associated with latency imbalance.
Lynx provides the following solutions for DWDM network protection:
- Optical channel/path protection
- DWDM line protection
- 1:n DWDM tributary protection
- In-line amplifier protection
This mechanism provides end-to-end protection of an entire DWDM channel, from one client site to the other. It is based on complete channel redundancy, including fibers, inline equipment and transponder line-cards that interface with the CPE.
Figure 2 shows an example of optical channel protection.
Figure 2: Optical channel/path protection
In some optical channels, which contain transport equipment and interconnecting fibers, failures can be detected by monitoring the optical signal. In such cases, performing 1+1 single-ended protection can be done by fiber protection systems. For other protection schemes, such as dual-ended protection, Lynx provides special in-band signaling solutions. In cases where the transport equipment does not support shut-off during failure, but sends AIS signals instead, failure detection must be done at the protocol level. Lynx offers Optical Failure Monitoring (OFM) modules, capable of detecting failures in all these situations. Lynx has also developed a unique technology, LynxSense™, to ensure that protection switches will indeed switch immediately when required.
Some Lynx protection systems allow users to carry extra traffic, thereby improving the utilization of the network redundancy. This extra traffic is carried over the protection channel while both channels are operational.
DWDM Line Protection
In some cases, deploying a redundant DWDM line may be more cost-effective than using several redundant channels: this depends on the number of active channels in a DWDM link that require protection (e.g., not segments of SDH/Sonet rings), and the amount of in-line transport equipment. Assuming that the terminal transponders can be protected through a 1+N protection scheme (Figure 3), and that the passive Mux/Demux is highly reliable, carriers may choose to protect only the DWDM line. Figure 4 shows such an application.
Compared to electro-optical (OEO) protection mechanisms, optical switching is a far more cost-effective way of protecting DWDM lines. Electro-optical protection requires double multiplexing/demultiplexing, and N termination and protection systems (where N is the number of potential wavelengths over the line).
In-line Amplifier Protection
In some cases, due to cost or geographical limitations, the DWDM link cannot be redundantly diverse; and the protection solutions described above cannot be used. In other cases, where some form of link protection exists, there may be long lag times in identifying and repairing in-line amplifiers, leaving the network vulnerable for unacceptably long periods. In such cases, carriers can use Lynx's EDFA systems to protect some of the in-line amplifiers locally, providing an option for immediate recovery while repair crews are dispatched. Lynx 1:n (2+1) add-on protection systems are used to protect an EDFA node (two EDFAs, one in each direction) with the use of a single spare EDFA. A special EDFA failure monitoring module, built into the protection system, is capable of detecting EDFA failures—including those in which the EDFA continues to transmit optical power.
Figure 5 shows an example of EDFA protection.
Figure 5: In-line amplifier (EDFA) protection