Title :
A Model Based RL Admission Control Algorithm for Next Generation Networks
Author :
Mignanti, Silvano ; Giorgio, Alessandro Di ; Suraci, Vincenzo
Author_Institution :
Dept. of Inf. & Syst. Theor., Univ. of Rome "La Sapienza", Rome
Abstract :
In this paper we study the call admission control problem to optimize the network operators revenue guaranteeing quality of service to the end users. We consider a network scenario where each class of service is characterized by a different constant bit rate and an associated revenue. We formulate the problem as a semi-Markov decision process, and we use a model based reinforcement learning approach. Other traditional algorithms require an explicit knowledge of the state transition models while our solution learn it on-line. We will show how our policy provides better solution than a classic greedy algorithm.
Keywords :
Markov processes; greedy algorithms; learning (artificial intelligence); quality of service; telecommunication congestion control; call admission control problem; greedy algorithm; model based RL admission control algorithm; next generation networks; quality of service; reinforcement learning; semi-Markov decision process; state transition models; Admission control; Bit rate; Call admission control; Electronic mail; Greedy algorithms; Informatics; Learning; Network topology; Next generation networking; Quality of service; Connection Admission Control; Model based; Quality of Service; Reinforcement Learning; Semi Markov Decision Process;
Conference_Titel :
Networks, 2009. ICN '09. Eighth International Conference on
Conference_Location :
Gosier, Guadeloupe
Print_ISBN :
978-1-4244-3470-1
Electronic_ISBN :
978-0-7695-3552-4
DOI :
10.1109/ICN.2009.39
