Synchronous optical network using a ring architecture

Details Synchronous optical network using a ring architecture Synchronous optical network using a ring architecture

2000-07-07

2002-09-17

Patel, Ajit (Department: 2664)

Multiplex communications

Pathfinding or routing

Switching a message which includes an address header

Reexamination Certificate

active

06452931

ABSTRACT:

BACKGROUND
1. Field of the Invention
The invention relates to synchronous optical networks (SONET) and specifically to the use of ring architectures that use stacked rings.
2. Description of the Prior Art
At present, proposed architectures for implementing SONET over relatively large geographic areas are Digital Cross-connect System (DCS) based. The relative size of these areas is larger than a Local Access and Transport Area (LATA), or larger than a metropolitan area. An example of such a network is an Interexchange Carrier (IXC).
In
FIG. 1
, network
10
is depicted without actual connections. For clarity, the network is shown as encompassing a geometric area with short routes, but clearly, networks may span entire countries and continents. Network
10
nodes
20
-
29
are shown. A node is a site in the network where traffic is processed often, this processing involves switching, providing access, and grooming. Additionally, physical routes
30
-
42
are shown between nodes
20
-
29
. The physical routes do not represent actual connections, but they represent the physical space that the actual connections may occupy. For example, the two lines between nodes
20
and
21
define physical route
30
between nodes
20
and
21
. These physical routes are typically optical fibers. Logical connections, or spans, occupy these physical fiber routes. A network will have many more nodes and longer routes than are shown on
FIG. 1
, but the amount shown is restricted for clarity.
The prior art DCS architecture for SONET deployment in a relatively large network is depicted in FIG.
2
. DCS architecture is well-known in the art and is based on point-to-point connections which employ cross-connect switching at the network nodes where point-to-point connections intersect. In
FIG. 2
, nodes
20
-
29
and routes
30
-
42
are again shown as in FIG.
1
. In
FIG. 2
, they are shown connected by DCS switches
50
-
59
over spans
60
-
72
occupying physical routes
30
-
42
. The current selection of a DCS architecture for SONET in a relatively large network is dictated by the SONET standards. These standards make rings impractical for these larger networks which encompass areas greater than a LATA or a metropolitan area. ANSI T1.105.XX Series requires that a SONET ring may contain a maximum of only 16 ring terminals.
In the large network environment, this standard puts a severe limitation on the number of ring terminals that may be placed on a ring. A large network, such an IXC, will require hundreds or even thousands of ring terminals to deploy SONET over the large geographic area covered by the network. These large geographic areas are greater than a LATA or a metropolitan area. At only 16 ring terminals per ring, the network is required to implement a high number of rings.
For the IXC deploying SONET, ring terminals will be required at all points of presence (POPs). A POP is where the IXC provides access to its network. Additionally, ring terminals are required at points where switching or grooming capability is located. An IXC network may cover thousands of square miles and contain thousands of POPs, switching, and grooming sites. This requires thousands of ring terminals. For an IXC to comply with SONET standards, the SONET architecture would include a very large number of rings. This is because the thousands of POPs, switching, and grooming sites can only be connected at 16 ring terminals per ring.
The large number of rings coupled with the great geographic distances involved represent a costly amount of overbuild. This overbuild is caused by the fact that ring connections require return spans to complete the ring. The logical connections between ring terminals are called spans. These spans, in turn, require physical fiber routes to complete the ring. At present, a relatively large network is required to add an excessive amount of physical fiber routes to facilitate the high number of spans required to close the large number of rings. The rings must accommodate a large number of POPs, switching, and grooming sites.
In the local environment, this overbuild is not nearly as severe because the geographic areas are restricted within the LATA. As such, the use of ring architectures for SONET has been restricted to small geographic areas such LATAs and individual metropolitan areas. Additionally, networks may employ a single large ring which covers a large area because only one ring must be closed instead of the several rings implicated in large networks.
The large network using rings faces the problem of the extra spans required to close rings, the large number of rings, and the large geographic distances to span. These geographic distances comprise areas larger than a LATA or a metropolitan area. This problem is exacerbated by the constraint of using existing physical routes. If possible, the network tries to re-use its current physical routes in order to avoid having to acquire more physical space for its routes. Real estate costs, as well as, construction and equipment costs are a significant deterrent to acquiring new physical territory for spans. Additionally, due to the terrain problems on long routes, such as mountains, small rings may just not be possible.
The resulting inefficiency has driven the choice to use DCS architecture in the networks larger than LATAs or metropolitan areas. A DCS based network is point to point and requires no return connections. DCS architecture reduces the number of spans required to deploy SONET, and the spans required for DCS adapt well to the existing physical routes. As a result, DCS architecture is the choice at present for large network SONET architectures.
However, there are also problems caused by DCS architectures. DCS survivability is controlled by a centralized device called a Digital Cross-connect Management System (DCMS). The DCMS is well-known in the art. In
FIG. 2
, DCMS
80
is shown and is connected to DCS switches
50
-
59
by signaling links
81
. When there is an interruption in a DCS network: 1) a DCS switch must sense the interruption, 2) the DCS switch must signal the DCMS of the condition, 3) the DCMS must determine alternate routing, 4) the DCMS must signal the alternate instructions to the DCS switches, and 5) the DCS switches must implement the alternate re-route instructions. At present, this sequence takes several minutes in a large network, such as IXC. The several minute loss of service is a serious problem.
In contrast, rings may be self-healing. Self-healing SONET rings are detailed in ANSI Standard T1.105.XX Series. Survivability is achieved despite an interruption by routing traffic around the operational side of the ring to complete the connection. No communication with a central control device is needed. No complex re-route instructions need to be determined. This is one reason rings are the choice for networks covering small geographic areas. The small overbuild is offset by the improvement in survivability time. A network can restore service with self-healing rings in milliseconds.
At present, large networks implementing SONET face a dual problem. Ring architectures require grossly impractical overbuild for such a network in order to close the high number of large rings. These are rings which encompass areas larger than a LATA or metropolitan area. The problem is due in part to the SONET standards, the large number of network nodes, and the length of existing physical routes. Although DCS architectures relieve the overbuild problem, the survivability of a DCS based network takes several minutes for a large network. This amount of time is unacceptable. For the above reasons, relatively large networks need a SONET system that does not require impractical overbuild, yet also has millisecond survivability.
SUMMARY
The present invention is a SONET system that satisfies the need of a large network architecture that efficiently complies with the SONET standards and offers acceptable survivability. The SONET system includes SONET ring terminals which are connected by SONET spans to form a ring architecture. The ring arch

Affiliated with

Also associated with

Rating

Synchronous optical network using a ring architecture does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Synchronous optical network using a ring architecture, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Synchronous optical network using a ring architecture will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2846899