Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/54679
Title: Resource allocation and scheduling for scalable video streaming over wireless networks
Authors: Li, Maodong
Keywords: DRNTU::Engineering::Electrical and electronic engineering::Wireless communication systems
Issue Date: 2013
Source: Li, M. (2013). Resource allocation and scheduling for scalable video streaming over wireless networks. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: Scalable video, evolved in a long way of more than 20 years, has attracted significant interest in the past decades in both academia and industry due to its versatility and potential in providing flexible storage and communication services. In contrast to the conventional video coding which encodes a video clip once and for all, the scalable video enables the composition of multiple clips in different specifications into a single bitstream and provides feasible mechanisms to adapt to the changing needs of networks or applications. With the finalization of the scalable extension of legacy H.264/AVC, i.e., Scalable Video Coding (SVC), the exploration on the potentials of coding, adaptation, scheduling and resource allocation becomes more interesting. Especially with the era of mobile internet, the exponential growth in demand for wireless video puts the scalable video to the spotlight, considering its high coding efficiency and feasible rate adaptation capability. Due to the error-prone and resource scarcity nature of the wireless networks, it is critical that efficient and effective resource allocation and packet scheduling schemes can be developed to deal with the underlying challenges. A comprehensive scheme always lies in the joint investigations on the video characteristics and the network specifications. Video content analysis focuses mainly on coding and the optimization of rate-utility (rate-distortion), scalability-rate, scalability-quality (PSNR, MOS, QoE), and so on. Meanwhile, resource allocation and packet scheduling normally consider the specifications of wireless networks from different Open Systems Interconnection (OSI) layers. In this thesis, scalable video streaming over wireless networks is investigated and evaluated in several typical wireless systems, through which we hope to inspire new research on this topic and foster its deployment in practical systems. The main contributions and achievements of this thesis are summarized below: 1. The first contribution of this thesis is the proposal of efficient packet prioritization and scheduling schemes for scalable video streaming over WLANs. The work provides an inspiration on how to apply the conventional unequal error protection strategy into scalable video transmissions and shows a novel way to deal with the network congestion issue in best effort networks. 2. This thesis also devotes to the fast scalable video rate adaptation. The fast adaptation scheme answers for how to obtain the maximal video quality in real-time scalable video adaptation and also provides effective tool to facilitate the video streaming applications. 3. The third contribution concerns a scalable resource allocation framework for SVC video streaming over MIMO-OFDM networks. By skillfully utilizing the feature of scalable video, the scalable framework renders a novel way to tackle the conventional dilemma in fair versus efficient design in multiuser resource allocation problems. 4. Based on a subjective video quality assessment database, the thesis summarizes various scalability adaptation tracks for QoE-aware rate adaptation. This thesis also constructs a rate-QoE model, which favors the understanding of scalable video adaptation and also facilitates the analysis in QoE-aware video streaming.
URI: http://hdl.handle.net/10356/54679
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
thesis.pdf8.09 MBAdobe PDFThumbnail
View/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.