Studies on distributed algorithm for network flow optimization problem based on tie-set flow vector space

Information networks have progressed rapidly toward large and complicated systems, and the amount of the information flowing through the networks is also increasing dramatically. Under these circumstances, it is important to establish distributed network management which utilizes network resources effectively. This paper presents a new approach to network optimization problems based on the theory of tie-set graph and tie-set flow vector space. A new distributed algorithm for network optimization is proposed, where local optimization on each tie-set leads to the global optimization of the network, autonomously. The proposed distributed algorithm can be applied to various network optimization problems.