Files
Citation
Share
Date
Abstract
Notas del catalogo: César Vargas Rosales, Ph.D., Asesor. Bibliography: l. [39]. Desc. física: 38 leaves : ill. ; 28 cm
Nowadays the number ofusers in the Internet is growing in an exponentially way and there arc several applications that the Internet offers and these require to be handled in different ways to run efficiently and improve the use ofthe network resources. We proposed a revenue function that allows the service provider to know their maximum revenue based on some parameters such as — the average number of packets, the average delay, the arrival rates, the capacity in the network links- and associate it with costs. In order to obtain the revenue function, we established a network topology with two different classes oftraffic, the delay permissible for each one and the capacity in the links ofthe network. We used the Broyden-Fletcher-Goldfarb-Shanno (BFGS) method to maximize the revenue function. We compare the optimized values versus non optimized values to observe their differences and how they rebound in the network performance and it is possible to control the network based on the prices.