近年來網路環境發展迅速，對於網路傳輸的頻寬需求也逐漸增加，而多路徑傳輸是一種有效提升網路傳輸時可用頻寬的方式，目前已經有許多多路徑傳輸機制被研究提出並運用於網路環境中，如ECMP、鏈路聚合技術、MPTCP等，但這些機制大多有所限制。隨著軟體定義網路發展愈趨成熟，主流網路環境也漸漸將網路傳輸控制權從傳統轉發設備分離出來，由控制設備來集中管理，而在軟體定義網路與網路功能虛擬化的架構下，網路管理者能夠開發設計不同網路功能，並交由控制器來統一透過軟體實作，因此，本論文提出一個運用軟體定義網路與網路功能虛擬化，結合多路徑傳輸與網路編碼的傳輸架構，多路徑傳輸能夠增加傳輸可用頻寬，避免使用單一路徑可能產生的可用頻寬不足的問題，而網路編碼機制因為其解碼時不受收到的封包順序影響，只要收到足夠數量的封包即可解碼出原始資料的特性，能夠克服封包亂序(Out of order)與封包遺失問題。實驗結果顯示，運用網路編碼機制後，確實能夠克服多路徑傳輸時因為不同路徑間延遲有所差異造成的封包亂序問題，也能克服傳輸時封包遺失所造成的影響，而提高網路傳輸的效能。

In recent years, the network environment has developed rapidly, and the demand for the transmission bandwidth of the network environment has gradually increased. Multi-path transmission is an effective way to increase the transmission bandwidth of the network. At present, many multi-path transmission mechanisms have been studied and proposed, such as ECMP, link aggregation technology, MPTCP, etc., but these mechanisms are mostly limited. As the development of software-defined networks becomes more and more mature, the mainstream network environment gradually separates transmission control from traditional forwarding equipment and is centrally managed by the control equipment. Under the framework of software-defined networks and network functions virtualization, Network administrators can develop and design different network functions. Therefore, this paper proposes a software-defined network and network function virtualization that combines multi-path transmission and network coding. Using multi-path transmission can increase the overall transmission bandwidth to avoid the problem of insufficient bandwidth when using single path transmission, and the network coding mechanism is not affected by the order of received packets when decoding, as long as a sufficient number of packets are received the original data can be decoded, which can overcome the problems of packet out of order and packet loss. The experimental results show that the network coding mechanism can indeed overcome the problem of packet out-of-order caused by the delay difference between different paths during multi-path transmission, and can also overcome the impact of packet loss during transmission.

目錄
摘要 i
Abstract ii
目錄 iii
表目錄 v
圖目錄 vi
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 論文架構 2
第二章 相關研究與技術 3
2.1 軟體定義網路Software Defined Networks 3
2.2 OpenFlow 協定 4
2.3 網路功能虛擬化(Network Function Virtualization) 4
2.4 多路徑隨機交換(Stochastic Switching) 5
2.5 網路編碼(Network Coding) 6
2.6 隨機線性網路編碼(Random Linear Network Coding, RLNC) 7
第三章 研究方法 9
3.1 多路徑權重計算 9
3.2 運用多路徑傳輸提升整體最大可用頻寬 10
3.3 網路編碼之編碼率理想下限討論 11
3.4 實驗架構 12
第四章 實驗模擬 17
4.1 實驗模擬環境與工具 17
4.1.1 Mininet 17
4.1.2 Kodo 17
4.1.3 均方誤差(Mean Square Error, MSE) 與峰值信噪比(Peak signal-to-noise ratio, PSNR) 17
4.1.4 實驗模擬環境建置 18
4.2 實驗 18
4.2.1實驗一：多路徑傳輸中封包亂序的觀察與影響 19
4.2.2實驗二：在Out of order狀況下運用網路編碼之多路徑傳輸 26
4.2.3實驗三：多路徑傳輸中封包遺失的觀察 31
4.2.4實驗四：在封包遺失狀況下運用網路編碼之多路徑傳輸 34
4.3 實驗結果整理 39
第五章 結語 42
參考文獻 43

表目錄
表4- 1實驗一(B)結果整理 26
表4- 2實驗二(B)結果整理 31
表4- 3實驗三(B)影像傳輸結果整理 34
表4- 4實驗四(A)結果整理(封包遺失率為2.2%) 37
表4- 5實驗四(B)結果整理 38
表4- 6封包亂序相關實驗之實驗環境與方法整理 39
表4- 7封包遺失相關實驗之實驗環境與方法整理 40

圖目錄
圖 2- 1 SDN架構圖[9] 3
圖 2- 2 OpenFlow Switch[11] 4
圖 2- 3 多路徑隨機交換機制[15] 6
圖 2- 4 網路編碼概念 7
圖 2- 5 隨機線性網路編碼[22] 8
圖 3- 1 實驗架構 13
圖 3- 2 網路編碼之編碼示意圖 14
圖 3- 3 解碼機制示意圖 15
圖 3- 4 多路徑傳輸資料示意圖 15
圖 3- 5 網路編碼傳輸資料示意圖 16
圖 4- 1 實驗一之網路拓樸 19
圖 4- 2 s1實驗一實驗前各port轉發情況 21
圖 4- 3 s1實驗一實驗後各port轉發情況 21
圖 4- 4 實驗一實驗過程畫面 21
圖 4- 5 實驗一實驗結果畫面 22
圖 4- 6 路徑延遲差0.1ms之實驗結果 22
圖 4- 7 延遲差0.2ms之實驗結果 23
圖 4- 8 延遲差0.3ms之實驗結果 23
圖 4- 9 不同延遲時間差平均亂序封包數量 24
圖 4- 10 資料大小為64bytes之結果影像圖((a)路徑延遲差0.1ms、(b)路徑延遲差0.2ms、(c)路徑延遲差0.3ms) 25
圖 4- 11 資料大小為1024bytes之結果影像圖((a)路徑延遲差0.1ms、(b)路徑延遲差0.2ms、(c)路徑延遲差0.3ms) 25
圖 4- 12 實驗二網路拓樸 27
圖 4- 13 延遲差0.1ms資料傳輸結果畫面 28
圖 4- 14 延遲差0.2ms資料傳輸結果畫面 28
圖 4- 15 延遲差0.3ms資料傳輸結果畫面 29
圖 4- 16 原始影像與實驗結果影像比對圖((a)原始影像、(b)實驗結果影像) 30
圖 4- 17 實驗二實驗結果影像之MSE值 30
圖 4- 18 實驗三網路拓樸 31
圖 4- 19 實驗三(A)接收結果畫面 32
圖 4- 20 實驗三(B)傳輸結果影像 33
圖 4- 21 實驗三(B)傳輸結果之MSE與PSNR值 33
圖 4- 22 實驗三不同封包資料大小結果影像比較((a) 傳輸資料大小為64bytes之結果影像、(b) 傳輸資料大小為1024bytes之結果影像) 34
圖 4- 23 實驗四之網路拓樸 35
圖 4- 24 傳輸結果畫面 36
圖 4- 25 實驗結果影像比較圖((a)編碼率1.03、(b)編碼率1.05、(c)編碼率1.1、(d)編碼率1.2、(e)編碼率1.3) 38

[1] B. N. Astuto, M. Mendonça, X. N. Nguyen, K. Obraczka and T. Turletti, “A survey of software-defined networking: Past, present, and future of programmable networks”, Communications Surveys Tutorials, IEEE, vol. 16, no. 3, pp. 1617–1634, Third 2014.
[2] C. Hopps, “Analysis of an equal-cost multi-path algorithm”, RFC 2992, IETF, Nov. 2000.
[3] J. Touch and R. Perlman, “Transparent Interconnection of Lots of Links (TRILL): Problem and Applicability Statement”, RFC 5556, May 2009.
[4] M. Scharf and A. Ford, “Multipath TCP (MPTCP) application interface considerations,” RFC 6824, IETF, Jan. 2013
[5] D. Wischik, C. Raiciu, A. Greenhalgh and M. Handley, “Design, implementation and evaluation of congestion control for multipath TCP”, 8th USENIX NSDI, 2011.
[6] S. Barré, C. Paasch, and O. Bonaventure, “Multipath TCP: from theory to practice,” International Conference on Research in Networking, pp. 444–457, 2011.
[7] A. Ford, C. Raiciu, M. Handley, and O. Bonaventure, “TCP extensions for multipath operation with multiple addresses”, RFC6824, 2013 [Online]. Available: http://tools.ietf.org/html/rfc6824
[8] Software-Defined Networking: The New Norm for Networks - Open Networking Foundation,
https://www.opennetworking.org/images/stories/downloads/sdn-resources/white-papers/wp-sdn-newnorm.pdf
[9] Software-Defined Networking (SDN) Definition- Open Networking Foundation, Retrieved from https://www.opennetworking.org/sdn-resources/sdn-definition
[10] N. McKeown, T. Anderson, H. Balakrishnan, G. Parulkar, L. Peterson, J. Rexford, S. Shenker and J. Turner, “OpenFlow: enabling innovation in campus networks”, ACM SIGCOMM Computer Communication Review 38 (2) 69–74, 2008.
[11] OpenFlow 1.5.1 Specification,
https://www.opennetworking.org/wp-content/uploads/2014/10/openflow-switch-v1.5.1.pdf
[12]“Network Functions Virualization – Update White Paper”, European Telecommunications Standards Institute (ETSI), Oct. 2013
[13] Network Functions Virtualisation, https://portal.etsi.org/NFV/NFV_White_Paper.pdf
[14] Network Functions Virtualisation (NFV); Infrastructure Overview, https://www.etsi.org/deliver/etsi_gs/NFV-INF/001_099/001/01.01.01_60/gs_NFV-INF001v010101p.pdf
[15] Network Functions Virtualisation (NFV); Management and Orchestration, https://www.etsi.org/deliver/etsi_gs/NFV-MAN/001_099/001/01.01.01_60/gs_NFV-MAN001v010101p.pdf
[16] K. S. Shourmasti, “Stochastic Switching Using OpenFlow”, Master’s thesis, Norwegian University of Science and Technology, Department of Telematics, July 2013.
[17] R. Ahlswede, N. Cai, S.-Y. R. Li, and R. W. Yeung. “Network information flow”, IEEE Transactions on information theory, 46(4), 1204-1216, 2000
[18] T. Ho, R. Koettert, M. Médard, D. R. Karger, and M. Effros, “The benefits of coding over routing in a randomized setting”, 2003
[19] T. Ho, M. Médard, R. Koetter, D. R. Karger, M. Effros, J. Shi, and B. Leong, “A random linear network coding approach to multicast”, IEEE Transactions on Information Theory, 52(10), 4413-4430, 2006
[20] L. Lima, M. M ́edard, and J. Barros, “Random linear network coding: A free cipher?”, IEEE International Symposium on Information Theory, IEEE, pp. 546–550, 2007
[21] S. Kim, W. S. Jeong, W. W. Ro, and J.-L. Gaudiot, “Design and evaluation of random linear network coding accelerators on fpgas”, ACM Transactions on Embedded Computing Systems (TECS), vol. 13, no. 1, p. 13, 2013
[22] Steinwurf Technical Documentation
http://docs.steinwurf.com/nc_intro.html
[23] B. Lantz, B. Heller, and N. McKeown, “A network in a laptop: rapid prototyping for software-defined networks”, In Proceedings of the 9th ACM SIGCOMM Workshop on Hot Topics in Networks, pp. 1-6, October 2010
[24] M. V. Pedersen, J. Heide, and F. H. P. Fitzek, “Kodo: An open and research oriented network coding library”, International Conference on Research in Networking, pp. 145-152, May 2011.
[25] S. Jaiswal, G. Iannaccone, C. Diot, J. Kurose, and D. Towsley, “Measurement and classification of out-of-sequence packets in a tier-1 IP backbone”, IEEE/ACM Transactions on networking, 15(1), 54-66, 2007.