IMPLEMENTATION OF A DECENTRALIZED PARETO-OPTIMAL ALGORITHM

iterative flow control algorithms to achieve a Pareto optimal point ha ve been proposed in the telecommunications literature. In this paper, we present several relaxation and estimation techniques to improve the performance of the algorithms. We also explore several possible imple mentations that extend support to the proposed analytical models. Simu lations of these possible implementations are carried out, and the per formance of the flow control algorithms studied. The proposed implemen tations for the Pareto algorithms achieve optimization of the product of powers of individual user classes. The effect of additional factors like the length of the observation cycle, the structure of the observ ation cycle, and the relaxation factors are also studied. The simulati on results suggest that better convergence is achieved when the observ ation windows are maintained at the node level and are packet count ba sed. The window size has to be sufficiently long. The relaxation techn iques are useful with the right choice of relaxation factors. The gree dy algorithms using estimation techniques to achieve complete decentra lizability are very appealing.