Skip to Main content Skip to Navigation
Conference papers

User to User QoE routing system

H. A. Tran A. Mellouk 1
1 CIR
LISSI - Laboratoire Images, Signaux et Systèmes Intelligents
Abstract : Recently, wealthy network services such as Internet protocol television (IPTV) and Voice over IP (VoIP) are expected to become more pervasive over the Next Generation Network (NGN). In order to serve this purpose, the quality of these services should be evaluated subjectively by users. This is referred to as the quality of experience (QoE). The most important tendency of actual network services is maintaining the best QoE with network functions such as admission control, resource management, routing, traffic control, etc. Among of them, we focus here on routing mechanism. We propose in this paper a protocol integrating QoE measurement in routing paradigm to construct an adaptive and evolutionary system. Our approach is based on Reinforcement Learning concept. More concretely, we have used a least squares reinforcement learning technique called Least Squares Policy Iteration. Experimental results showed a significant performance gain over traditional routing protocols.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01678772
Contributor : Yacine Amirat <>
Submitted on : Tuesday, January 9, 2018 - 1:16:31 PM
Last modification on : Friday, October 4, 2019 - 1:28:51 AM

Identifiers

  • HAL Id : hal-01678772, version 1

Collections

Citation

H. A. Tran, A. Mellouk. User to User QoE routing system. Proc. of the 9th IFIP International Conference on Wired/Wireless Internet Communications (WWIC 2011), Jun 2011, Barcelona, Spain. ⟨hal-01678772⟩

Share

Metrics

Record views

61