Predication-based intelligence routing on telecommunications

The routing being one of key issues of telecommunication networks management influences the switch rate and load balance directly, and becomes more and more important as the telecommunication traffic increases at full speed. It was said that raising one percent of switch rate of current Chinese telecommunication networks would result in revenue about one billion Yuan RMB. On the basis of demerit-analysis of routing scheme being used, a new intelligent routing strategy based on multi-agent systems and recurrent neural network predication is presented, including routing strategy, Calls generating and agent, recurrent neural network computation, simulation software design, results and discussion. The results show that the new one is better by virtue of its upstanding distribution and intelligence characters, and provides excellent solution to increase network switch rate and balance network load. Meanwhile, its applications will go beyond the scope of telecommunication networks. But whether the strategy will be perfect depends on the cooperation among operator, manufacturer and researchers.