Distributed precomputation of network signal paths with...

Details Distributed precomputation of network signal paths with... Distributed precomputation of network signal paths with...

1999-11-24

2001-03-20

Olms, Douglas W. (Department: 2732)

Multiplex communications

Fault recovery

Bypass an inoperative channel

C370S225000

Reexamination Certificate

active

06205117

ABSTRACT:

RELATED APPLICATIONS
The present application is related to U.S. Patent Applications entitled “Distributed Precomputation of Signal Paths in an Optical Network,” “Distributed Precomputation of Network Signal Paths with Improved Performance Through Parallelization,” “Multi-Phase Process for Distributed Precomputation of Network Signal Paths,” and “Hybrid Centralized/Distributed Precomputation of Network Signal Paths,” each filed concurrently herewith in the names of inventors Bharat T. Doshi, Subrahmanyam Dravida, P. Harshavardhana and Yufei Wang, and assigned to the assignee of the present application.
FIELD OF THE INVENTION
The present invention relates generally to techniques for restoring communication in a network after a failure in a link, span or node of the network, and more particularly to restoration techniques in which restoration paths are precomputed at nodes distributed throughout the network.
BACKGROUND OF THE INVENTION
All-optical networks using wavelength division multiplexing (WDM) are increasingly being deployed for a wide variety of communication applications. WDM techniques allow optical signals having different wavelengths to be multiplexed into a single optical fiber. Current WDM deployments allow multiplexing of up to about 16 different wavelengths on a single fiber, but systems multiplexing 32 or more different wavelengths on a single fiber are expected to become available soon. Each of the wavelengths serves as an optical carrier and can be used independently of the other wavelengths, such that different wavelengths may use different modulation formats to carry different signal types. In a simple example, each wavelength may carry a modulation signal representing a synchronous optical network/synchronous digital hierarchy (SONET/SDH) client payload, where each client is a SONET-rate TDM application and the common carried signals are in an OC-48 or an OC-192 format.
FIG. 1
shows a conventional optical routing device
10
which includes a wavelength selecting cross-connect (WSCC)
12
, two input optical fibers
14
-
1
,
14
-
2
and two output optical fibers
14
-
3
,
14
-
4
. The routing device
10
in this embodiment is configured to route incoming optical signals at wavelengths &lgr;
1
and &lgr;
2
on fiber
14
-
1
to output fibers
14
-
4
and
14
-
3
, respectively, and to route incoming optical signals at wavelengths &lgr;
1
′ and &lgr;
2
′ on fiber
14
-
2
to output fibers
14
-
4
and
14
-
3
, respectively. The WSCC
12
thus serves to cross-connect incoming wavelengths on a given input fiber to different output fibers, but does not provide any transformation in wavelength. When only this type of routing device is present in an optical network, the network typically routes a given end-to-end demand using a single wavelength. If a primary network path assigned to the given demand fails, the demand generally must be carried on a secondary or restoration path using exactly the same wavelength as the primary path.
FIG. 2
illustrates an optical network
20
in which wavelength transformations may be provided for signals traversing the network, but only at the interface between a client and the optical network. A first client equipment (CE) device
18
-
1
communicates with a second CE device
18
-
2
. The first CE device
18
-
1
uses wavelength &lgr;
1
and the second CE device uses wavelength &lgr;
3
. The first CE
18
-
1
transmits a signal at &lgr;
1
to a wavelength adapter
22
which maps the incoming wavelength &lgr;
1
to an outgoing wavelength &lgr;
2
. A wavelength adapter (WA) is a device which allows conversion of wavelength at the client-network interface. The wavelength &lgr;
2
is used to carry the modulation signal of CE
18
-
1
from an access node
24
of network
20
to an egress node
26
of network
20
. The egress node
26
delivers the &lgr;
2
signal to a second WA
28
which maps the wavelength &lgr;
2
to wavelength &lgr;
3
for transmission to the second CE
18
-
2
. In the event of a failure in the primary path through optical network
20
from CE
18
-
1
to CE
18
-
2
, a secondary or restoration path with a different wavelength, such as &lgr;
4
, may be used to transport the customer demand through the network
20
. Other types of optical network elements combine features of the WSCC
12
of FIG.
1
and the WAs
22
,
28
of FIG.
2
. For example, a wavelength interchange device may be used to cross-connect incoming wavelengths onto different output fibers while also providing transformation of wavelengths. Such devices are called wavelength interchanging cross-connects (WICCs).
An important issue in the design of large-scale optical networks including WSCCs, WAs, WICCs and other optical signal routing devices relates to traffic restoration in the event of a failure in a link, span or node. A simplistic approach to restoration in an optical network is to provide complete redundancy, such that the network includes a dedicated back-up or secondary connection for each primary connection of the network. When a link, span or node of the primary connection fails, traffic may then be switched onto the corresponding elements of the secondary connection. Unfortunately, this approach uses a large amount of restoration capacity and therefore may be undesirable in many networks. More sophisticated approaches involve the use of a path restoration algorithm to provide automatic restoration of network traffic in the event of a primary path failure, while sharing restoration capacities whenever possible.
It should be noted that large-scale optical networks typically include a large number of spans, and two different point-to-point links may share a common span section.
FIG. 3
illustrates a shared span section in a portion of a network including nodes A, B and C. The dotted lines AC and AB represent two distinct optical links. The physical layout, shown by solid lines, is such that both of these links share the span AS. If this span fails due to a fiber cut or other problem, then both the links AC and AB will fail. Thus a demand using link AB on its primary path cannot be restored on a route using link AC. It is therefore important that a given restoration algorithm achieve restoration of network traffic in the event of span failures as well as link failures, by providing distinct spans and links for the restoration path. Furthermore, to decrease vulnerability of the network to node failures, it is also desirable to perform automatic restoration in the event of single node failures. Thus the overall goal of an effective restoration algorithm should be to perform automatic restoration in the event of single link, span or node failures. The term “automatic” connotes restoration by control computers in the network, rather than by manual intervention, thus permitting fast restoration.
FIG. 4
shows a portion of a network including nodes A, B, C and D providing a bidirectional path at a wavelength &lgr;
1
between a first CE
18
-
1
and a second CE
18
-
2
. In simple optical networks, failures are generally discovered through signal strength measurements, which may be collected for each individual wavelength at each node of the network. If a link failure occurs between nodes B and C as shown in
FIG. 4
, the bidirectional nature of the path allows each of the nodes A, B, C and D to detect a loss of signal (LOS) condition, but, with only the LOS information, none of these nodes will know the exact location of the failure. As a result, local restoration around the failed link, by the nodes connecting the failed link, is generally not possible, assuming that the optical network under consideration does not employ any other mechanism to isolate failures. This inability to determine the exact location of the failure from LOS information also requires that the restoration path be disjoint from the primary path. Depending on whether a network includes WSCCs, WAs or WICCs, additional restrictions may be imposed on the restoration and primary paths of a demand. The network path of
FIG. 4
includes only WSCCs, and the secondar

Affiliated with

Also associated with

Rating

Distributed precomputation of network signal paths with... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Distributed precomputation of network signal paths with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Distributed precomputation of network signal paths with... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2449518