Abstract: A method of assigning capacity and routing flow in a bi-directional line switched SONET ring based upon ring topology and demand data defines for each pair of links of the ring a two-edge cut. Each two-edge cut divides the ring into two sets of nodes. For each two-edge cut, the capacity assignment method calculates a demand equal to the sum of all demands between nodes on opposite sides of the two-edge cut. The capacity assignment method then determines the maximum demand and sets the capacity of each link equal to one-half the maximum demand plus one-half of one demand unit. The flow routing method of the present invention calculates a cut difference for each two-edge cut. A critical cut is a two-edge cut having a cut difference equal to or less than one. If there is a critical cut with demands greater than zero on the same side of the critical cut, the method performs a first processing routine.

Abstract: A method of optimizing a network includes a plurality of nodes interconnected by high multiplex level links, in which at least some of the links carry a number of low multiplex level remaining demands that is less than the capacity of the links, by rerouting lower multiplex level remaining demands through the network to reduce the total number of higher multiplex level links in the network. The method eliminates from consideration any bridges and spurs in the network, thereby forming one or more connected components of the network. After the spurs and bridges have been eliminated from consideration, the method reroutes remaining low multiplex level demands through each of the remaining connected components of the network to reduce the total number of high multiplex level links in each remaining portion.

Abstract: A method of optimizing a telecommunications network in terms of transmission and equipment cost by multiplexing lower bandwidth level telecommunications demand-routes to form higher bandwidth level bundles, by determining, for each demand-route, whether the demand-route includes a patching subroute. For each patching subroute, the method finds a best bundle of demand-routes that includes the patching subroute. The best bundle is the one that has the highest cost gradient between the sum of the costs associated with the demand-routes included in the best bundle and the total cost associated with the best bundle. The method then multiplexes the demand-routes according to the best bundles.

Abstract: A method of routing lower multiplex level demands in a telecommunications network including a plurality of interconnected nodes to form higher multiplex level demand-routes. The method defines an initial lower multiplex level demand-route for each pair of nodes of the network. The initial lower multiplex level demand-route carries all lower multiplex level demands originating or terminating at a node of the pair. The method then splits the demand of each initial lower multiplex level demand-route between an express demand and an overflow demand and routes express demands in higher multiplex level express routes. The method patches the overflow demands into higher multiplex level routes according to a predetermined patching policy.

Abstract: An apparatus and method is described that is used to design a minimal spares network for a given communications network. Three data files are used as input which describe the communications network. The information comprising such data files include the number and configuration of the nodes and spans, the current demand in terms of the required capacity between the source and destination nodes, current spare capacity (if any), and optionally current paths used to carry the demand traffic (including the presence or absence of glass-throughs). A three pass method is used to design an efficient and economical spares network from the given input data. A simulated cut is performed on each span in the communications network. From each simulated span cut, the affected demand traffic is identified and rerouted via spare capacity that is `purchased` on an incremental cost basis. Pre-purchased spares are used before new spares are purchased.

Abstract: In a telecommunications network having a plurality of intelligent nodes interconnected by multiple communication channels, alternate paths and spare channels are set up for restoring traffic disrupted by failure to one or more of the communications channels. The alternate paths may be set up irrespective of the number of custodial node pairs, also referred to as leader/follower or sender/chooser pairs, that could simultaneously be involved in the restoration operation. The interconnected nodes have bi-directional working and/or spare channels. There is stored at each node in the network participating in the recovery operation a spare channel manifest for each of the leader/follower combinations. The first end node, or leader node, of every identified failed channel initiates a request message for each disrupted working channel to set up an alternate path.