Quality of Service Dynamic Routing Schemes for Real Time Systems in IP Network

Many types of routing algorithms have been proposed, such as shortest-path, centralized, distributed, flow-based, etc., for optimally using the network resources. The resolution of this problem, considered as a necessary condition in a high performance networks, is naturally formulated as a dynamic programming problem, which, however, is too complex to be solved exactly. Making globally optimal routing decisions requires that as the load levels, traffic patterns and topology of the network change, the routing policy also adapts a decision’s router in the goal to take into account the dynamic’s change communication network. We proposed here an overview for these methods and we focused on neurodynamic programming to construct dynamic statedependent routing policies. These policies offer several advantages, including a stochastic modelization of the environment (especialy links, link costs, traffic, and congestion), learning and evaluation are assumed to happen continually (do not have an explicit learning phase followed by evaluation), multipaths routing and minimizing state overhead.