臺灣博碩士論文加值系統

網際網路通訊隨著現代科技的進步而有著迅速的發展，但是網路會發生壅塞的問題依然還是沒有徹底地被解決，這個令網路使用者困擾的問題早在二十多年前就已出現，迄今仍未改善。過去也有許多相關的學者與研究人員紛紛針對這種情況提出一些壅塞控制機制的解決方式，如TCP、ECN、RED…等等，但是卻也因為網路環境的發展與變化過快而有所不適宜。
當網路遭遇壅塞的情況發生時，在TCP原有的Retransmission Timeout機制中，在發送端正確地將資料封包傳送出去後，如果發送端無法於機制所預先設定的Retransmission Timeout時間內，準時地接收到由接收端的所回覆的Acknowledgement Packet的話，那麼發送端就會認為該已被傳送出去的資料封包並沒有成功的被送達目的端，而且是在傳送的過程中被丟失，因此重新傳送該被認定是丟失的封包，但是封包沒有在預設的時間內，被回覆Acknowledgement Packet的原因可能是因為封包傳遞過程中遭遇嚴重的壅塞情況，而使得封包傳遞速度過慢，但是封包還是在網路上傳遞並未被丟失；或者是封包已經正確地被接收端接收，且接收端已經對該封包回覆了 Acknowledgement Packet，只是因為網路的壅塞情況嚴重，而使得 Acknowledgement Packet傳遞速度過慢，造成逾時的狀況發生，在這些情況下，倘若發送端再次重送封包，就會造成更大量封包在網路上傳遞，形成更多的網路負載，使得壅塞的程度增加，甚至會造成Congestion Collapse 的發生。
在本論文中，我們藉由自行設計一套能夠根據Round Trip Time與Round Trip Time變化量的數值並以模糊理論的方式，動態調整合適Retransmission Timeout Timer的機制；當網路壅塞情況發生時，便動態的調整Retransmission Timeout Timer的數值大小，以使得有較適合的時間讓網路處理在等待的封包，減少網路的負載，繼而達成較有效舒緩網路壅塞的情況。
在本論文中，我們藉由自行設計一套能夠根據Round Trip Time與Round Trip Time變化量的數值並以模糊理論的方式，動態調整合適Retransmission Timeout Timer的機制；當網路壅塞情況發生時，便動態的調整Retransmission Timeout Timer的數值大小，以使得有較適合的時間讓網路處理在等待的封包，減少網路的負載，繼而達成較有效舒緩網路壅塞的情況。

For guaranteeing successful transmission of packets, timers are set when packets transmitted and acknowledgements of receiving packets at destinations should be returned within the period corresponding to values of timers. If the waiting time of an acknowledgement expires the value of the timer known as retransmission timeout timer (RTO), the corresponding packet is viewed as lost and the packet is retransmitted.
A possible problem making the Internet congested is then considered. If network load is heavy, packet transmission delay increases; similarly transmission delay of packet acknowledgements from destinations to sources increases. As a result, Round Trip Time (RTT) of packets increases. Larger RTT may expire the retransmission timer, and packets not yet reaching their destinations are thought as lost and retransmitted
It results in more retransmissions. Consequently, the traffic with retransmitted packets on the Internet increases and transmission delay and the corresponding RTT then become much longer. The procedure is repeated and the traffic load becomes more and more heavy. The Internet may become, therefore, congested and blocked. This is called congestion collapse.
In this thesis, we propose a method based on the fuzzy theory to adapt retransmissions timers. The fuzzy theory is widely used in controlling nonlinear, timer-varying, and ill-defined systems. The traffic on the Internet is time-varying and hard to be predicted and modeled. The uncertainty properties make the traffic control over the Internet seem solvable with the fuzzy theory. Based on fuzzy rules, larger RTT may form a conjecture of a congested condition on the Internet. The values of retransmission timer then increase. Simulations show that some unnecessary retransmissions are actually avoided and then throughput increases with our proposed mechanism.

目錄
ABSTRACT
摘要
致謝辭
目錄
圖表目錄
表格目錄
1.簡介
2.相關技術
2.1 TRANSMISSION CONTROL PROTOCOL (TCP)
2.1.1 背景
2.1.2 TCP Flow Control
2.1.3 RTO ( Retransmission Timeout )：
2.2. FUZZY THEORY
2.2.1 背景
2.2.2模糊集合
2.2.3 歸屬函數定義
2.2.4 模糊推論
2.2.4.1模糊推論的方式
2.2.4.2解模糊化法
2.2.4.3模糊控制
2.2.4.4 傳統控制的限制
2.2.5模糊控制的建構
3.研究方法
4.實驗結果
4.1 實驗方法
4.2實驗結果
5.結論
6.附錄
6.1歸屬函數分類類別
6.1.1調整歸屬函數分類數目
6.1.2在C係數集合數目相同下，針對RTT與 ΔRTT的歸屬函數分類
6.2與M/M/1/N的比較
6.3RTT與RTO數值的變化
參考資料

[1]J. Postel, “Transmission control protocol,” Internet
Request for Comments (RFC) 793, September 1981.
[2]S. Floyd and T. Henderson, “The new Reno modification to TCP’s fast recovery algorithm,” Internet Request for Comments (RFC) 2582, April 1999.
[3]K. Ramakrishnan and S. Floyd, “A proposal to add explicit congestion notification (ECN) to IP,” Internet Request for Comments (RFC) 2481, January 1999.
[4]Sally Floyd and Van Jacobson, “Random early detection gateways for congestion avoidance,” IEEE/ACM Transactions on Networking, vol. 1, no. 4, pp. 397-413, August 1993.
[5]John Nagle, “Congestion Control in IP/TCP Internetworks,” Internet Request for Comments (RFC) 896, January 1984.
[6]S. Floyd, “Congestion Control Principles,” Internet Request for Comments (RFC) 2914, September 2000.
[7]R. Braden, “Requirements for Internet hosts-communication layers,” Internet Request for Comments (RFC) 1122, October 1989.
[8]J. Hadi Salim and U. Ahmed, “Performance evaluation of explicit congestion notification (ECN) in IP networks,” Internet Request for Comments (RFC) 2884, July 2000.
[9]K. Ramakrishnan, S. Floyd, and D. Black, “The addition of explicit congestion notification (ECN) to IP,” Internet Request for Comments (RFC) 3168, November 2001.
[10]Celio Albuquerque, Brett J. Vickers, and Tatsuya Suda, “Network border patrol: preventing congestion collapse and promoting fairness in the Internet,” IEEE/ACM Transactions on Networking, vol. 12 , no. 1 , pp. 173 — 186, February 2004.
[11]T. Ziegler and H.D. Clausen, “Congestion avoidance with BUC (buffer utilization control) gateways and RFCN (reverse feedback congestion notification),” in Proceedings of the IEEE International Conference on Performance, Computing, and Communications, Phoenix, USA, February 5-7, 1997, pp. 410-418.
[12]George Klir and Bo Yuan, Fuzzy Sets and Fuzzy Logic: Theory and Applications, Prentice Hall PTR, May 11, 1995.
[13]M. Allman, S. Floyd and C. Partridge, “Increasing TCP’s initial window,” Internet Request for Comments (RFC) 3390, September 2002.
[14]K. Poduri and K. Nichols, “Simulation studies of increased initial TCP windows size,” Internet Request for Comments (RFC) 2415, September 1998.
[15]K. Fall and S. Floyd , “Simulation-based comparisons of Tahoe, Reno and SACK TCP ,” Computer Communications Review, vol. 26, no. 3, pp. 5-21, July 1996.
[16]M. Allman, V. Paxson and W. Stevens, “TCP congestion control,” Internet Request for Comments (RFC) 2581, April 1999.
[17]Van Jacobson and Michael J. Karels, “Congestion Avoidance and Control,” in Proceedings of SIGCOMM, Stanford, USA, August 1988, pp. 314-329.
[18]Phil Karn and Craig Partidge, “Improving Round-Trip time estimates in reliable transport protocols,” in Proceedings of the ACM workshop on Frontiers in computer communications technology, Stowe, Vermont, USA, August 1987, pp. 2-7.
[19]Douglas E. Comer and John C. Lin, “Probing TCP implementations,” in USENIX, San Francisco, USA, January 1994, pp. 245-255.
[20]Vockler J., “Retransmission behavior of solaris 2.5.1 and 2.6,” Available: Http://sun-microsystems.org
[21]Scott Dawson, Farnam Jahanian, and Todd Mitton, “Experiments on six commercial TCP implementations using a software fault injection tool,” in Proceedings of Software Practice and Experience, December 1997, pp. 1385-1410.
[22]Fuzzy 控制，中國生產力編譯，全華圖書，民國81年.
[23]L.A. Zadeh, “A New Approach to the Analysis of Complex Systems and decision processes,” IEEE Transactions on Systems Man and Cybernetics, vol. 3, pp. 28-44, 1973.