Intelligent allocation of network bandwidth: A comparison of two generalized particle models

There are two types of generalized particle models (GPM) that can be used to allocate network bandwidth efficiently. One is the “original” generalized particle models (OGPM) [1] with which the allocation is not changed until the price reaches equilibrium. The other one is the economic generalized particle model (EGPM) [2], where the allocation is changed dynamically even when the allocation is being calculated. EGPM therefore suits bandwidth allocation in dynamic environments. The two models can be classified as evolutionary intelligent swarm optimization approaches. The EGPM is an important extension and further development of the OGPM, which comprises two major components: (1) dynamic allocation of network bandwidth based on GPM; and (2) dynamic modulation of price and demands of network bandwidth. The two components combined can be readily implemented by a distributed algorithm. In this paper, by simulations, we compare EGPM and OGPM, and then with the ant colony optimization approach, the utility function algorithm, and the max-min fairness algorithm.