A fuzzy set delay representation for computer network routing algorithms

The authors present a novel approach for routing path delay estimation and manipulation in wide area networks. Routing is an important functional aspect of a data communication network that handles data packets during their transit from source to destination. The efficiency of a distributed dynamic routing algorithm depends on how well it adapts to the traffic and topology changes in the network. The nodes in the network periodically exchange information regarding these changes for efficient routing. The information used by the algorithms to arrive at the routing paths suffers from uncertainty due to a variety of reasons. Certain aspects of fuzzy sets are applied to model the uncertainty in the delay information and this new delay representation is used to improve the performance of one particular class of routing algorithms in computer networks. The results of some simulation studies of these algorithms are presented.<>