Wireless local area networks are becoming very popular in many scenarios because they are very simple, convenient and cheap. This paper focuses on multimedia traffic management in wireless networks, where we consider to provide differentiated Quality of Service (QoS) levels. We address the complex task of traffic scheduling with multi-objective requirements in the presence of errors introduced by the radio channel. In particular, we focus on managing downlink traffic in both wireless ATM and WiFi scenarios, referring to an infrastructure wireless access network where a central coordinator takes scheduling decisions for the mobile users in its cell. Our scheduler is based on an Artificial Neural Network (ANN) with reinforcement learning. The ANN is trained from examples to behave as an “optimal” scheduler, according to an Actor-Critic model. The results obtained in scheduling concomitant voice, video and Web traffic classes permit to show the significant capacity improvement that can be achieved by our scheme with respect to other techniques previously proposed in the literature.

Pasquale, F., Giambene, G., & Trentin, E. (2007). Neural-based downlink scheduling algorithm for broadband wireless networks. COMPUTER COMMUNICATIONS, 30(2), 207-218 [10.1016/j.comcom.2006.08.009].

Neural-based downlink scheduling algorithm for broadband wireless networks

GIAMBENE, GIOVANNI;TRENTIN, EDMONDO
2007

Abstract

Wireless local area networks are becoming very popular in many scenarios because they are very simple, convenient and cheap. This paper focuses on multimedia traffic management in wireless networks, where we consider to provide differentiated Quality of Service (QoS) levels. We address the complex task of traffic scheduling with multi-objective requirements in the presence of errors introduced by the radio channel. In particular, we focus on managing downlink traffic in both wireless ATM and WiFi scenarios, referring to an infrastructure wireless access network where a central coordinator takes scheduling decisions for the mobile users in its cell. Our scheduler is based on an Artificial Neural Network (ANN) with reinforcement learning. The ANN is trained from examples to behave as an “optimal” scheduler, according to an Actor-Critic model. The results obtained in scheduling concomitant voice, video and Web traffic classes permit to show the significant capacity improvement that can be achieved by our scheme with respect to other techniques previously proposed in the literature.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11365/24511
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo