在無線網路上因為有比較高的位元錯誤率(Bit Error Rate, BER)，造成TCP的傳送端誤認為有線網路發生壅塞，而不當地把壅塞視窗調降下來，使得無線網路上TCP的傳輸率大量下降。另外，以往在有線網路的排程器同樣沒有考慮到無線網路環境的問題，若是直接將有線網路上面的排程器用在無線網路上，會使得網路效能大打折扣。
因此，本論文針對無線網路的高BER，在基地台加入代替重傳機制，以增加TCP的傳輸率。並且利用在無線網路上所量測到的封包遺失率(Packet Loss Ratio, PLR)來做分類與佇列權重(weight)的調整。在第一階段先將封包分成有PLR與沒有PLR兩種等級，第二階段再根據PLR設定的範圍分成7種等級。當基地台發生壅塞時，調降有PLR佇列的無線頻寬，讓那些通道狀況比較好的移動式工作站(Mobile Station)能使用較多的無線頻寬，以提昇無線網路整體的傳輸率。在我們的機制中，佇列長度被用來判斷無線網路是否發生壅塞的依據，當佇列長度超過設定的臨界值時，表示無線網路即將發生壅塞，此時會啟動調降佇列權重的機制，並且將調降的權重給沒有封包遺失的工作站使用。另外，因為更換佇列將造成封包的亂序(out of sequence)發生，我們在TCP連線更換佇列後，會先清空之前使用的佇列的封包，來解決封包的out-of-sequence問題。
為了驗證我們所提出的機制，我們在Linux平台上實作完成，並且透過實驗結果證明了代替重傳機制確實可以改進了TCP的傳輸率。而當無線網路發生壅塞時，我們的機制能有效地增加整體的throughput，並且避免封包亂序的發生。

Traditional TCP when applied in wireless networks may encounter two limitations. The first limitation is the higher bit error rate (BER) due to noise, fading, and multipath interference. Because traditional TCP is designed for wired and reliable networks, packet loss is mainly caused by network congestions. As a result, TCP may decrease congestion window inappropriately upon detecting a packet loss. The second limitation is about the packet scheduling, which mostly does not consider wireless characteristics.
In this Thesis, we propose a local retransmission mechanism to improve TCP throughput for wireless networks with higher BER. In addition, we measure the packet loss ratio (PLR) to adjust the queue weight such that the available bandwidth for each queue can be changed accordingly. In our mechanism, the queue length is used to determine whether there is a congestion in wireless networks. When the queue length exceeds a threshold, it indicates that the wireless networks may have congestion very likely. We not only propose the dynamic weight-adjustment mechanism, but also solve the packet out-of-sequence problem, which results form when a TCP flow changes to a new queue.
For the purpose of demonstration, we implement the proposed weight-adjustment mechanisms on the Linux platform. Through the measurements and discussions, we have shown that the proposed mechanisms can effectively improve the TCP throughput in wireless networks.

目 錄

第一章 導論……………………………………………………………………………1
1.1 研究動機……………………………………………………………………1
1.2 研究及實作方法……………………………………………………………1
1.3 論文架構……………………………………………………………………3

第二章 改進無線網路效能的相關研究……………………………………………4
2.1 IEEE 802.11無線網路………………………………………………………4
2.2 改善TCP連線Throughput下降的相關研究……………………………6
2.2.1 SNOOP………………………………………………………………6
2.2.2 其他改進Wireless TCP 效能的方法……………………………7
2.3 Wireless Scheduling 的相關研究…………………………………………9
2.3.1 Channel State Prediction………………………………………9
2.3.2 Explicit Compensation……………………………………………10
2.4各種改善Wireless TCP效能和Wireless Scheduling的方法比較……13
2.5本論文的動態調整頻寬佇列……………………………………………14

第三章 根據PLR動態調整頻寬佇列………………………………………………15
3.1 代替重傳機制……………………………………………………………15
3.1.1 封包複製…………………………………………………………17
3.1.2 Mobile Stations 的ACK封包……………………………………18
3.1.3 代替重傳造成RTT量測錯誤………………………………………21
3.2 動態調整頻寬佇列………………………………………………………23
3.2.1 PDWRR的演算法…………………………………………………24
3.2.2 調整時機與ping-pong效應……………………………………27
3.2.3 out-of-sequence的解決方法……………………………………28

第四章 Linux平台的實作及量測結果……………………………………………31
4.1 Linux平台上的實作……………………………………………………31
4.1.1 在Linux Kernel當機的解決方法…………………………………31
4.1.2代替重傳機制的實作………………………………………………32
4.1.3 PDWRR實作………………………………………………………35
4.1.4 Indirect-TCP實作…………………………………………………40

4.2 實驗環境架構與設定……………………………………………………41
4.3 實驗效能量測與分析討論………………………………………………42
4.3.1 代替重傳的Throughput量測與討論……………………………43
4.3.2 PDWRR效能量測與討論…………………………………………46
4.3.2.1 基地台的代替重傳封包量測………………………………47
4.3.2.2 封包out-of-sequence的量測………………………………48
4.3.2.3 平均Throughput的量測……………………………………49
4.4.3 增加Indirect-TCP的效能量測…………………………………51

第五章 結論與未來工作……………………………………………………………52
5.1 結論………………………………………………………………………52
5.2 未來工作…………………………………………………………………53

Reference……………………………………………………………………………54
Index…………………………………………………………………………………57

圖表目錄

圖2-1 IEEE 802.11e的Priority功能……………………………………………5
圖2-2 IEEE 802.11e 的HCF功能…………………………………………………6
圖2-3 Congestion Coherence………………………………………………………8
圖2-4 SBFA example……………………………………………………………11
圖2-5 I-CSDP example………………………………………………………12
圖3-1 傳統TCP在無線網路的封包遺失………………………………………16
圖3-2 代替重傳機制……………………………………………………………16
圖3-3 封包複製演算法…………………………………………………………18
圖3-4 ACK進到基地台處理演算法……………………………………………20
圖3-5 RTT估算錯誤……………………………………………………………22
圖3-6 PDWRR的基本架構……………………………………………………23
圖3-7 動態頻寬調整的演算法…………………………………………………25
圖3-8 解決ping-pong效應………………………………………………………28
圖3-9 解決封包out-of-sequence的演算法……………………………………29
圖3-10 解決封包的out-of-sequence…………………………………………30
圖4-1 代替重傳modules在Linux Kernel的載入位置………………………32
圖4-2 修改過的sk_buff資料結構………………………………………………33
圖4-3 RTO timer的資料結構…………………………………………………35
圖4-4 Qdisc包含多個classes的架構…………………………………………36
圖4-5 封包進到enqueue後的kernel函式流程……………………………37
圖4-6 Linux平台上的WRR工作流程…………………………………………38
圖4-7 PDWRR的工作流程……………………………………………………39
圖4-8 Indirect-TCP基本架構…………………………………………………40
圖4-9 實驗網路拓樸……………………………………………………………41
圖4-10 代替重傳的Throughput量測結果………………………………………43
圖4-11 Congestion Window的量測結果………………………………………44
圖4-12 SACK和TCP-Reno的重傳比較…………………………………………44
圖4-13 代替重傳機制的封包連續發生錯誤……………………………………45
圖4-14 PDWRR實驗的環境設定………………………………………………46
圖4-15 基地台的重傳封包量測…………………………………………………48
圖4-16 out-of-sequence的封包量測………………………………………………49
圖4-17 基地台的平均throughput量測…………………………………………49
圖4-18 增加Indirect-TCP的效能比較…………………………………………51



表2-1 表2.1各種改善Wireless TCP效能的方法比較…………………………13
表2-2 Wireless Scheduling的方法比較…………………………………………14
表3-1 每個class分配的PLR範圍………………………………………………25
表4-1 實驗機器設備……………………………………………………………41
表4-2 實驗中資料流的PLR設定值……………………………………………46

Reference

[1] IEEE 802.11 WG. IEEE Std Standard 802.11, 1999 edition International Standard [for] Information Technology-Telecommunications and Information Exchange between systems-Local and metropolitan networks. http://standards.ieee.org/getieee802/802.11.html , 1999.
[2] IEEE 802.11 WG. Draft Supplement to International Standard [for] Information Technology-Telecommunications and Information Exchange between systems LAN/MAN Specific Requirements. IEEE 802.11e/D2.0, http://standards.ieee.org/getieee802/802.11.html , Nov 2001.
[3] S. Mangold, S. Choi, P. May, O. Klein, G. Hiertz, and L. Stibor, “IEEE 802.11e Wireless LAN for Quality of Service”, In Proceedings of the European Wireless, Vol. 1, pp. 32-39, Florence, Italy, February 2002.
[4] H. Balakrishnan, S. Seshan, and R. H. Katz, “Improving Reliable Transport and Handoff Performance in Cellular Wireless Networks”, ACM Wireless Networks Journal, vol. 1, no. 4, pp. 469-481, Dec. 1995.
[5] A. Bakre and B. R. Badrinath, “I-TCP: Indirect TCP for mobile hosts”, Proc. 15th International Conf. On Distributed Computing Systems(ICDCS), May 1995, pp. 136-143.
[6] W. Ding and A. Jamalipour, “A New Explicit Loss Notification with Acknowledgment for Wireless TCP”, Proc. 12th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, vol. 1, pp. 65-69, 2001.
[7] C. Liu and R. Jain, “Approaches of Wireless TCP Enhancement and A New Proposal Based on Congestion Coherence”, the 36th Hawaii International Conference on System Sciences, Quality of Service in Mobile and Wireless Network minitrack, Big Island, Hawaii, pp. 307-316, 2003.
[8] S. Floyd, “TCP and Explicit Congestion Notification”, ACM Computer Commn. Review, V.24,No. 5, October 1994, pp. 10-23.
[9] B. S. Bakshi, P. Krishna, N. H. vaidya, and D. K. Pradhan, “Improving Performance of TCP over Wireless Networks”, In 17th international conference on distributed computing systems(ICDCS), May 1997, pp. 365-373.
[10] J. H. Hu, G. Feng, and K. L. Yeung, “Hierarchical Cache Design for Enhancing TCP over Heterogeneous Networks with Wired and Wireless Links,” IEEE Transaction on Wireless Communications, Volume: 2 Issue: 2 , pp. 205 –217, 2003 .
[11] J. H. Hu and K. L. Yeung, “FDA: A Novel Base Station Flow Control Scheme for TCP over Heterogeneous Networks”, INFOCOM 2001, pp. 142-151.
[12] P. Bhagwat, A. Krishna, and S. Tripathi, “Enhancing throughput over wireless LAN’s using channel state dependent packet scheduling”, in Proc. INFOCOM96, Mar. 1996, pp. 1133-1140.
[13] H. Aida, Y. Tamura, Y. Tobe, and H. Tokuda, “Wireless Packet Scheduling with Signal-to-Noise Ratio Monitoring”, IEEE Annual Conference on Local Computer Networks (LCN 2000), Tampa, FL, Nov. 2000, pp. 32-41.
[14] P. Ramanathan and P. Agrawal, ”Adapting packet fair queueing algorithms to wireless networks”, in ACM/IEEE MOBICOM ’98, Dallas, TX, pp. 1-9.
[15] J. Gomez, A. T. Campbell, and H. Morikawa, “The Havana framework for supporting application and channel dependent QoS in wireless networks”, in Proc. ICNP ’99, Nov. 1999, pp. 235-244.
[16] M. Shreedhar and G. Varghese, “Efficient Fair Queueing using Deficit Round Robin”, In Proceedings of ACM SIGCOMM, 1995, pp. 231-242.
[17] S. Lu, V. Bharghavan, and R. Srikant, “Fair Scheduling in Wireless Packet Networks”, in Proc. Of SIGCOMM ’97, Aug. 1997, pp. 63-74.
[18] M. R. Jeong, H. Morikawa, and T. Aoyama, “Fair Scheduling Algorithm for Providing QoS in Wireless Packet Networks”, in Proc. Of ICCC ’99.
[19] T. S. Eugene Ng, I. Stoica, and H. Zhang, “ Packet fair queueing algorithms for wireless networks with location-dependent errors”, in Proc. INFOCOM98, Mar. 1998, pp. 1103-1111.
[20] D.A. Eckhardt and P. Steenkiste, “Effort-Limited Fair (ELF) Scheduling for Wireless Networks”, INFOCOM 2000, Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies, Proceedings. IEEE, vol. 3, pp. 1097-1106, Mar. 2000.
[21] D. A. Rusling, “The Linux Kernel”, http://www.tldp.org/LDP/tlk/tlk.html ,
1999.
[22] A. Rijsinghani, “Computation of the Internet Checksum via Incremental Update”, RFC 1624, May 1994.
[23] B. Hubert and G. Maxwell, “Linux Advanced Routing & Traffic Control Howto”, http://lartc.org/ , December 17, 2001.
[24] W. Almesberger, “Linux Network Traffic Control – Implementation Overview”, EPFL ICA, February 4, 2001.
[25] M. Devera, ”Devik’s Linux QoS Development”, http://luxik.cdi.cz/~devik/qos/qos.htm
[26] A. Tirumala, F. Qin, J. Dugan, and J. Ferguson, “Iperf Version 1.70”, http://dast.nlanr.net/Projects/Iperf/ , March 2003
[27] Web100 projects, http://www.web100.org/rprojects/
[28] O. Pomerantz, ”Linux Kernel Module Programming Guide”, http://www.tldp.org/LDP/lkmpg/mpg.html , 1999.
[29] M. Mathis, J. Mahdavi, S. Floyd, and A. Romanow, “TCP Selective Acknowledgment Options”, RFC-2018, 1996.