跳到主要內容

臺灣博碩士論文加值系統

(3.235.185.78) 您好!臺灣時間:2021/07/29 23:01
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:胡朝傑
研究生(外文):Chau-Chieh Hu
論文名稱:基於路由協助之多媒體通訊可用路徑頻寬測量的新方法
論文名稱(外文):A Novel Method for Router Assisted Available Path Bandwidth Measurement of Multimedia Communications
指導教授:賴文彬賴文彬引用關係
學位類別:碩士
校院名稱:元智大學
系所名稱:通訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:中文
論文頁數:54
中文關鍵詞:可用路徑頻寬可用頻寬探測封包時戳資訊鏈路容量路徑容量網路負載網路模擬器NS2
外文關鍵詞:Available Path BandwidthAvailable BandwidthProbing PacketTime StampLink CapacityPath CapacityLoadingNS2
相關次數:
  • 被引用被引用:0
  • 點閱點閱:151
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
本研究發展出了一套高精確度且即時性的可用路徑頻寬的測量新方法。本方法乃是基於Packet Dispersion技術,以各封包訊務支流交會時之流速競爭與主流流速限制等觀念,發展微觀的封包動力學模型,加上我們倡議引入之路由器時戳資訊的新協定架構的輔助,提供接收端計算探測封包在網路各節點的Average Dispersion Rate,然後套入我們所發展的可用路徑頻寬預測公式;如此,能以最少的額外路由負擔,以短序列探測封包求出即時且精確的可用路徑頻寬。
我們採用NS2網路模擬器輔以實驗室架設之路由網路實測驗證,證明本研究的測量方法適用於任意級數與任意拓樸形狀的網路,且不會受到鏈路瓶頸與探測封包大小所限制。當網路負載從Light Load變化至Heavy Load時,本方法的測量誤差值均可控制在1%以下。此外,本方法相較於PathChirp法能更即時且精確地反應可用路徑頻寬的動態變化,其精確度可大幅提昇約1至2個數量級(百分誤差約0.1% ~ 1%)。因此,本方法之應用對多媒體通訊速率的自適應調整與整體品質提昇有重大的意義。
This research has developed a novel method for high-precision and real-time Available Path Bandwidth (APB) measurement. The method is based on the Packet Dispersion technique. We have developed a microscopic model of packet dynamics based on the flow rate competition between different ingress traffic flows merging at a given node and the total limited capacity at the egress link. In addition, we propose a router assisted time stamp protocol to allow for the calculation of Average Dispersion Rates at all the nodes of the probing path so that these rates can be plug into the APB prediction formula we developed. Hence, it is possible to use a short sequence of probe packets to obtain the APB accurately and in real-time under the minimum overhead brought by the time stamp protocol.
To verify our method is applicable to any network topology and not limited by the bottleneck link capacity or affected by the probing packet size, we conducted simulations using the NS2 network simulator and lab experiments. When the network loading changes from light to heavy, all the measurement errors of our method can be control under 1%. Besides, compared to PathChirp, our method can detect the dynamical changes of APB more real-time and accurately with a great enhancement up to one or two orders of magnitude ( 0.1% ~ 1% equivalently). Hence, the application of our method may lead to a significant impact to adaptive rate control and integral quality enhancement of multimedia communications.
1 緒論 1
1.1 研究目標 1
1.2 研究動機 1
1.3 本研究貢獻 3
1.4 論文架構 4
2 研究背景 5
2.1 技術用語定義 5
2.2 研究背景 8
2.2.1 Packet Dispersion 探測技術系列 8
2.2.2 Self-Loading Periodic Streams探測技術系列 9
2.3 本研究的定位 10
3 路由協助之可用路徑頻寬理論建構 11
3.1 PACKET PAIR 12
3.2 PACKET TRAIN 13
3.3 佇列模型 15
3.4 測量原理 16
3.5 本研究建構之測量方法推導 19
3.5.1 1級網路拓樸APB公式推導 19
3.5.2 2級網路拓樸APB公式推導 20
3.5.3 N級網路APB公式推導 21
4 理論驗證與分析 23
4.1 實驗步驟 23
4.2 基礎網路模擬實驗驗證 24
4.2.1 1級網路拓樸模擬實驗驗證 24
4.2.2 2級網路拓樸模擬實驗驗證 26
4.3 多級網路拓樸模擬實驗驗證 27
4.3.1 Tight Link ≠ Narrow Link 27
4.3.2 Tight Link = Narrow Link 29
4.4 實驗室路由網路實測驗證 31
4.4.1 1級網路拓樸 32

5 各種探測流參數與背景流參數影響的結果與分析 34
5.1 探測封包速率的動態範圍 34
5.1.1高探測速率下,改變網路負載 35
5.1.2 低探測速率下,改變網路負載 36
5.1.3 高網路負載下,改變探測封包速率 37
5.2 在LIGHT LOAD情況下,改變探測流的封包大小 38
5.3 在HEAVY LOAD情況下,改變探測流的封包大小 39
5.4 同一個節點,多個背景流量匯流通過 40
5.4.1 多個交叉背景流量,匯流通過一段路徑 40
5.4.2 多個平行背景流量匯流通過整條通訊路徑 41
5.5動態改變之背景流量 42
5.5.1 3個背景流量 42
5.5.2 10個背景流量 44
5.6 本方法與PATHCHIRP的效能比較 47
5.6.1 精確度比較 47
5.6.2 即時性比較 48
6 結論與未來展望 50
6.1 結論 50
6.2 未來展望 51
REFERENCE 52
[1]C. Dovrolis, P. Ramanathan and D. Moore, “Packet-Dispersion techniques and a capacity-estimation methodology,” in IEEE/ACM Transaction on Networking, vol. 12, No. 6, Dec. 2004.
[2]C. Dovrolis, P. Ramanathan, and D. Moore, “What do Packet Dispersion techniques measure,” in Proc. IEEE INFOCOM, Apr. 2001, pp.905–914.
[3]Ling-Jyh Chen, Cheng-Fu Chou, Bo-Chun Wang, ”A Machine Learning-based Approach for Estimating Available Bandwidth”, IEEE TENCON 2007, Taipei, Taiwan, 2007.
[4]V. Jacobson, “Congestion avoidance and control,” in Proc. ACM SIGCOMM, Sep. 1988, pp. 314–329.
[5]S. Keshav, “A control-theoretic approach to flow control,” in Proc. ACM SIGCOMM, Sep. 1991, pp. 3–15.
[6]I. C. Bolot, “Characterizing end-to-end packet delay and loss in the internet,” in Proc. ACM SIGCOMM, Sep. 1993, pp. 289–298.
[7]R. S. Prasad, M. Jai, and C. Dovrolis, “Effects of interrupt coalescence on network measurements,” in Proc. Passive and Active Measurements (PAM) Worksop, 2004.
[8]R. S. Prasad, M. Murray, C. Dovrolis and K. Claffy, “Bandwidth estimation: metrics, measurement techniques, and tools,” in IEEE Network, vol. 17, pp. 27–35, Nov 2003.
[9]J. Strauss, D. Katabi, and F. Kaashoek, “A measurement study of available bandwidth estimation tools,” in ACM IMC, 2003.
[10]Constantinos Dovrolis. Pathrate. [Online]. Available: http://www.pathrate.org
[11]M. Jain and C. Dovrolis, “End-to-End Available Bandwidth: Measurement Methodology, Dynamics, and Relation with TCP Throughput,” in IEEE/ACM Transactions on Networking, vol. 11, No. 4, August 2003.
[12]A. Chobanuzn, Matt Mutka and Z. Cen, “One Way Delay Trend Detection for Available Bandwidth Measurement,” in IEEE Globecom, 2005.
[13]M. Jain and C. Dovrolis, “Pathload: a measurement tool for end-to-end available bandwidth,” in Proc. Passive and Active Measurements (PAM) Worksop, 2002.
[14](2009, April.)Tools for Bandwidth Estimation. [Online]. Available: http://www.icir.org/models/tools.html.
[15]Vinay J. Ribeiro, Rudolf H. Riedi, Richard G.. Baraniuk, J. Navratil and L. Cottrell. “PathChirp: Efficient available bandwidth estimation for network paths,” in Proc. Passive and Active Measurement Workshop, April 2003.
[16]UCB/LBNL/VINT, “Network Simulation – NS (Version 2.0),” October 1998.
[17]K. Fall, K. Varadhan, NS2: The NS Manual. [Online]. Available: http://www.isi.edu/nsnam/ns
[18]ASPAC Project, AWK: AWK Tutorial Guide. [Online]. Available: http://phi.sinica.edu.tw/aspac/reports/94/94011
[19]Peter Barry and Gerard Hartnett, Designing Embedded Networking Applications, INTEL Press, 2005.
[20]Iperf. [Online]. Available: http://iperf.sourceforge.net
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top