6533b831fe1ef96bd12990cc

RESEARCH PRODUCT

A genetic algorithm for the design of a fuzzy controller for active queue management

Alfonso UrsoFrank HoffmannG. Lo ReG. Di Fatta

subject

Settore ING-INF/05 - Sistemi Di Elaborazione Delle InformazioniRouterQueue management systemComputer sciencePID controllerFuzzy control systemRandom early detectionTCP Congestion ControlActive queue managementNetwork CongestionFuzzy logicComputer Science ApplicationsHuman-Computer InteractionNetwork congestionControl and Systems EngineeringControl theoryElectrical and Electronic EngineeringTail dropActive Queue ManagementSoftwareFuzzy Controllers.Information Systems

description

Active queue management (AQM) policies are those\ud policies of router queue management that allow for the detection of network congestion, the notification of such occurrences to the\ud hosts on the network borders, and the adoption of a suitable control\ud policy. This paper proposes the adoption of a fuzzy proportional\ud integral (FPI) controller as an active queue manager for Internet\ud routers. The analytical design of the proposed FPI controller is\ud carried out in analogy with a proportional integral (PI) controller,\ud which recently has been proposed for AQM. A genetic algorithm is\ud proposed for tuning of the FPI controller parameters with respect\ud to optimal disturbance rejection. In the paper the FPI controller\ud design metodology is described and the results of the comparison\ud with random early detection (RED), tail drop, and PI controller\ud are presented.

yearjournalcountryeditionlanguage
2003-08-01IEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews)
https://doi.org/10.1109/tsmcc.2003.818946