Topology design and bandwidth allocation of embedded ATM networks using genetic algorithm

A genetic algorithm (GA) is proposed for the design of an asynchronous transfer mode (ATM) network embedded in a digital cross connect network. This highly constrained optimization problem is shown to be better solved using GA. Improved results, both in network performance and computation speed, are obtained when comparing with the existing heuristic approach. A similar methodology can be applied for the design of other communication networks.