Abstract

The evolution of IEEE 802.11 wireless local area network (WLAN) has emerged as the most preferred choice for organizations as well as end users and consumers to access the network, especially the Internet. Today, WLAN technology has becoming more mature and slowly taking over the job for information sharing which was once borne by wired networks. WLAN has not only been used to transfer data packets, but also multimedia packets that are more time sensitive. Given the nature of WLAN which is prone to interference, several challenges need to be encountered for WLAN to keep abreast with the ability of wired network, especially in terms of performance and user experience. Among the challenges of WLAN is the support of Quality of Experience (QoE) to stream video which demands special attention and immediate solution. Failing to address this demand in a timely and effective manner will eventually limit the success of WLAN. This thesis presents a novel QoE provisioning technique in the IEEE 802.11e networks. With these techniques, video traffic can be prioritized so that the end users will have better video quality especially during wireless network congestion. This is done by looking at the types of videos frames and prioritises the more important ones (I-Frame). These techniques use two different algorithms, namely congestion-triggered based and PSNR-triggered based. Within the former method, I-Frames are prioritized when the system detects that the network is congesting. Less important video frames (B-Frames) will be dropped from the queue whenever an I-Frame enters the MAC queue. Meanwhile in PSNR-triggered based, I-Frames will be given priority over B-Frames when the end user’s predicted PSNR level is below a certain threshold. Simulations results done using NS-3 showed that improvements on the proposed schemes, both in objective and subjective evaluation. On implementing the proposed techniques, better QoE can be achieved where better PSNR and MOS were recorded. Finally, a brief discussion on the constraints during experiment implementations and future directions of QoE in WLAN conclude this thesis.

Document Type

Thesis

Publication Date

2022-01-01

DOI

10.24382/1029

Download

Share

COinS