Simulation study of switched circuit communication networks using learning automata routing

Traffic Routing in switched circuit communication networks is considered using both alternate or fixed rule and learning automaton routing strategies. The simulation studies using a microprocessor based real-time simulator are shown to provide confirmation that the automaton routing scheme performs at least as well as the optimal fixed rule. In addition, studies under failure mode conditions, including link, node and focussed overloads, conclusively demonstrate the superior performance of learning automata routing.