Dynamic Routing Strategies Based on Local Topological Information of Scale-free Network

Abstract

In this paper, we first propose a traffic model based on scale-free network to investigate the connections between the topology of a network and the routing strategies work on it. Then, we put forward a topology function with a tunable exponential parameter to represent the static part of local routing strategies. In order to make routing algorithms still work well even when the network falls into a congested state, we add the dynamic ingredient into them so that packets could avoid those congesting nodes when traffic jam occurs. In order to maximize the throughput of the network, we investigate the performance of a class of routing strategies that based on degree of nodes and put forward a new class of dynamic local routing strategies based on probabilistic betweenness of nodes. Experimental results show that the new class of algorithms work well even when the traffic of the network is very heavy.