A nature-inspired algorithm for intelligent optimization of network resources

Abstract

In complex computer networks having the characteristic of social dynamics, bandwidth allocation is a fundamental problem where bandwidth has to be reserved for connections in advance. This paper presents the theory and approach of the economic generalized particle model (EGPM) for intelligent allocation of network bandwidth. This approach transforms the complicated network bandwidth allocation problem into efficient, parallel allocation of network bandwidth. This approach is an important extension and further development of the generalized particle model (GPM) [1]. EGPM emphasizes the use of pricing as the network control mechanism. For the pricing, it makes use of the tatonnement process in economics. EGPM arises from GPM but can overcome some of GPM's deficiencies for the network bandwidth allocation problem. © 2008 IEEE.