國立高雄大學圖資館 |

Language: English

Back

Optimal network resource allocation ...

Li, Ying.

Optimal network resource allocation for heterogeneous traffic .

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Optimal network resource allocation for heterogeneous traffic .
Author:	Li, Ying.
Description:	129 p.
Notes:	Advisers: A. Robert Calderbank; Mung Chiang.
Notes:	Source: Dissertation Abstracts International, Volume: 69-10, Section: B, page: .
Contained By:	Dissertation Abstracts International69-10B.
Subject:	Engineering, Electronics and Electrical.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3333857
ISBN:	9780549863168

Optimal network resource allocation for heterogeneous traffic .
Li, Ying.

Optimal network resource allocation for heterogeneous traffic . - 129 p.

Advisers: A. Robert Calderbank; Mung Chiang.

Thesis (Ph.D.)--Princeton University, 2008.

In the work of exploring the new value for network applications brought by the physical layer innovations, we provide approaches to make the physical innovation to support different applications over a network and to jointly maximize their utilities, where we focus on the recent innovation of diversity embedded channel coding, which combines high rates with high reliability. During the process, we develop new techniques to tackle nonconvex optimization problem.

ISBN: 9780549863168Subjects--Topical Terms:

226981
Engineering, Electronics and Electrical.

Optimal network resource allocation for heterogeneous traffic .
LDR:05213nam _2200361 _450 001 206971
005 20090413130353.5
008 090730s2008 ||||||||||||||||| ||eng d
020 $a 9780549863168
035 $a 00372183
040 $a UMI $c UMI
100 $a Li, Ying. $3 321909
245 1 0 $a Optimal network resource allocation for heterogeneous traffic .
300 $a 129 p.
500 $a Advisers: A. Robert Calderbank; Mung Chiang.
500 $a Source: Dissertation Abstracts International, Volume: 69-10, Section: B, page: .
502 $a Thesis (Ph.D.)--Princeton University, 2008.
520 $a In the work of exploring the new value for network applications brought by the physical layer innovations, we provide approaches to make the physical innovation to support different applications over a network and to jointly maximize their utilities, where we focus on the recent innovation of diversity embedded channel coding, which combines high rates with high reliability. During the process, we develop new techniques to tackle nonconvex optimization problem.
520 $a In the work of new design on content-aware collaborative multimedia communication, we deliver better end-to-end video quality and achieve more efficient communication resource sharing, by distributed optimization in video content adaptation and resource allocation, for users with heterogenous video clips. In the work of new design of architecture and network operations for delay stringent, reliable multimedia communication over diverse (multiple) paths, we study the architecture and operations in a network with competing multiple-description (MD) streams, to provide realtime multimedia services over a network due to the stringent delay requirements. Our study reveals that for large number of users competing for resources, the system could benefit through opportunistic (hierarchical) strategies.
520 $a Our new design yields stable and efficient networks with heterogenous traffic, including data traffic, multimedia traffic, etc., and users with more happiness in terms of the larger utility achieved. Our work also contributes to quantitative understanding to guide a systematic, rather than an ad hoc, process of designing layered protocol stack for wired and wireless networks, by providing frameworks for systematic networking with multimedia traffic.
520 $a Overall, we show that using optimization theory as a foundation, simulations as a building block, and engineering intuition as a guide can be a principled approach to architecture and protocol design.
520 $a The dissertation performs new designs to support heterogeneous traffic (including multimedia traffic) with different requirements of the rate, reliability and delay within limited network resources such as limited link bandwidth. To provide end users with satisfied quality of service while keeping the system efficient, robust, stable, etc., we propose utility based optimal resource allocation framework.
520 $a The utility is the measure of user's happiness about the quality of service. Traditionally for the delay insensitive data traffic, the utility can be a function of rate only, but the utility for multimedia traffic can be a function of delay and loss as well as rate. The optimality in this dissertation is mainly about maximizing the total utility or maximizing the minimum utility of all users. The optimality can be achieved by jointly or separately adjusting the design knobs, which are the network operations including the video source coding at the application layer, congestion control at the transport layer, channel coding at the physical layer, and so on. The optimal operating point of all these design knobs and optimal connection of the design knobs can be found by applying optimization theory (especially the decomposition) to come up with the distributed solution which networks favor of. By some examples of our novel design, we show the effectiveness of our new networking framework.
520 $a Wired and wireless networks are trying to support many new applications, such as online gaming, video conference, etc., hosting a variety of traffic including data, voice, and multimedia traffic. With all these new applications, some traffic is delay sensitive, and much traffic is dominated in volume by video, but the traditional networks do not support realtime traffic and multimedia traffic well. Given the new applications, this dissertation takes a different perspective on design from the traditional networks. Our new design reflects that in the protocol stacks based on Open Systems Interconnection (OSI), the innovations in the application layer should be supported by new protocols in the lower layers and similarly, new applications should take advantage of lower layer innovations.
590 $a School code: 0181.
650 $a Engineering, Electronics and Electrical. $3 226981
650 $a Operations Research. $3 227148
650 $a Computer Science. $3 212513
690 $a 0544
690 $a 0796
690 $a 0984
710 $a Princeton University. $3 212488
773 0 $g 69-10B. $t Dissertation Abstracts International
790 $a 0181
790 1 0 $a Calderbank, A. Robert, $e advisor
790 1 0 $a Chiang, Mung, $e advisor
791 $a Ph.D.
792 $a 2008
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3333857 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3333857