Improving QoS of VoWLAN via Cross-Layer-Based Adaptive Approach

Abstract

The next generation technology of Voice over Internet Protocol (VoIP) allows voice to be delivered over the Internet Protocol based networks. The integration of VoIP and Wireless Local Area Network (WLAN) has become popular. However, WLAN architecture was primarily designed to carry data and not for real-time applications, such as VoIP. Hence, there is a need to enhance the quality of VoIP transmission over the WLAN. On the other hand, IEEE 802.11 standard of WLAN provides support to link adaptive (LA) technique. LA leads to multi-rate transmissions, which causes degradation on voice quality. In this paper, cross-layer based adaptive approach is proposed to improve the quality of voice delivery over WLAN by minimizing the multi-rate effects on the quality of service (QoS) of VoIP. The approach in achieving this is by introducing a communication agent between MAC sub-layer and Application layer, by which congestion caused by the multi-rate could be reduced, if not eliminated. Moreover, Encoder adaptation is utilized at the Application layer as codec parameters are adapted according to the change of rate occurs at the MAC sub-layer. Simulation of different scenarios for different codec rates at different transmission rates has been implemented using OPNET. The result of QoS improvements is the evidence for the proposed cross-layer adaptation approach.

Survey on application-layer mechanisms for speech quality adaptation in VoIP

“…On the other hand, some adaptive mechanisms monitor parameters at the MAC layer, such as transmission rate, in order to control the sending bitrate [Alshakhsi and Hasbullah 2011;Bilbao et al 2009;Kalama et al 2008;Kawata and Yamada 2006;Ksentini 2009;Servetti and Martin 2003;Sfairopoulou et al 2007;Tüysüz and Mantar 2010]. This can be performed by the sender alone without consulting the receiver, which allows for a faster reaction time against quality degradation.…”

“…This can be performed by the sender alone without consulting the receiver, which allows for a faster reaction time against quality degradation. Such a cross-layer approach may also use RTCP RR reports for providing a refined adaptation [Alshakhsi and Hasbullah 2011;Sfairopoulou et al 2007;Tüysüz and Mantar 2010]. However, in satellite systems, because of the large end-to-end delay, the use of RTCP RR reports by the adaptation mechanism is not feasible [Kalama et al 2008].…”