In a communication network, a channel connection establishment is typically a process of the physical layer that defines a logical path for the information transmitted in that network. When transmitting signals across wired networks, copper wire, optical fiber, coaxial cable, wireless digital link or satellite links are employed as the physical medium. The channel connection establishment process usually includes three stages: call establishment, channel connection and data transmission. In the call establishment stage, signaling and call establishment are performed by Layer 1 and Layer 2 of the network, respectively. In the channel connection stage, the logical channel connection, including channels, ports, DPs (dynamic paths), transmission rate, and other parameters are defined. Lastly, in the data transmission stage, signals are transmitted over these logical channels using the physical layer (Layer 1 or Layer 2). For a given communications channel between a source node and a destination node in a network, there are usually multiple paths between the two nodes. The multiple paths may be created by having more than one way to send data between the source and destination nodes. The different paths may be referred to as links, channels or routes. When a fault occurs in the network, such as on a particular link or a particular node, the data packets may be diverted from their original paths to alternate paths, often resulting in service disruptions and traffic congestion, referred to as “route-flapping”. Therefore, there is a need to protect the network against the disruption of service, such as route-flapping, that may result from a fault in the network. Network service disruption may occur on a point-to-point link or may occur on an end-to-end connection. Point-to-point links, as the name implies, connects a single source node with a single destination node. However, such a link may be composed of multiple channels, such as one channel for sending the data, and another channel for receiving the data. In a communications network, when multiple channels are used to connect the source node and the destination node, the signals sent from the source node are sent through the appropriate one of the multiple channels to the destination node, and vice versa. Often a single physical link may be implemented using two or more channels, with each channel providing an independent logical connection between the same source node and the same destination node. By using independent channels, failures in a link can be remedied by switching the communication over to another channel. One situation in which the use of multiple channels may be beneficial is when one of the channels is carrying more traffic than the others on a single link. End-to-end connections, as the name implies, establish a connection between two physical or logical network nodes, with all other nodes within the network being considered the intermediate nodes of the connection. To ensure that the traffic is transported between the two end-nodes without major disruptions in the transmission of the data, many networks have two distinct paths between the end nodes. Often, a single physical link or a single channel may support one or more logical end-to-end connections. The transmission of data across a network may involve multiple logical connections and multiple physical links (or channels). For example, each logical connection may be established using two channels for increased bandwidth. Each channel may be represented by a network link, or a virtual channel or a logical channel. When a fault occurs in a network, it is desirable to provide fast fault detection and routing around the fault. The primary cause of end-to-end network service disruption is due to the disruption of data transmission as a result of a point-to-point link failure or an end-to-end connection failure. When a particular path has a link failure, the data will be diverted along a different path. However, the alternate paths may not be established as quickly as needed by the network and, therefore, packets may be lost. If only one path is available between the source node and the destination node, then, upon detecting the link failure, the network would attempt to reroute the data from the source node to the destination node. This type of network fault recovery is usually referred to as “restoring data transmission after a fault occurs”. However, when there are multiple possible paths between the source node and the destination node, a network fault recovery technique has to be implemented to select the proper path and to select the proper port in a multi-channel connection for data transmission. A number of restoration methods have been used or proposed in an effort to respond to this type of failure. Some of the methods use a single failure criterion and assume that upon detection of a failure, the network service is to be restored by rerouting the traffic through a single available path. In other methods, multiple failure criteria are considered in a network fault recovery scheme and the network service is restored by choosing an available path that satisfies all the failure criteria. One known prior art solution is to define “pre-constituted” restoration sets that are formed by identifying links that appear on all paths. Thus, if all paths included a particular link, then the link would be included in all of the restoration sets. The “pre-constituted” sets are often referred to as “virtual links” or “virtual paths.” As shown in FIG. 1, this solution defines the restoration set r(A, B) to be the set of all links on every path between source node A and destination node B. Restoration Sets may be based on different criteria, such as a pre-constituted list of candidate links (i.e. “pre-constituted virtual links”). If a link exists in all of the restoration sets then that link is included in all of the sets. A restoration set is not pre-constituted by the network. An example of one implementation of virtual links is the use of MPLS-TE tunnels that pre-establish the path from source to destination node. Thus, traffic can be rerouted via an alternative virtual path should a link fail. One drawback of virtual links is that they do not consider multi-path capabilities. For example, if multiple paths were available in a restoration method that used a single failure criterion and a single virtual link, and the network service was being restored by rerouting the traffic, data packets could not be rerouted over one or more of the other paths. One way to address this drawback is to redefine the pre-constituted restoration set to include multiple links. However, this leads to a drawback in that the data transmission over the restored link is not guaranteed. As such, the term “pre-constituted” restoration set can be misleading, because it implies a restoration set that is pre-determined. In reality, there are several problems with restoring network service using pre-constituted restoration sets. For example, if two parallel paths existed between a pair of nodes, with one path having one less link than the other path, and a fault occurred on the one link, restoration would be implemented using only the path with one less link, leading to an un-utilized path and wasted resources. Thus, there is a need for a fault detection and management system that provides an efficient way to protect the transmission of data within a network, that protects against route-flapping, and that responds to multiple failures without having to re-constitute the restoration set, that increases the probability of restoring the network service, that does not waste network resources, and that is implementable in all types of networks.