目前使用即時性多媒體應用的需求增加，然而網路技術所供給的頻寬有限，在大量資料傳輸的情況下無法提供良好的多媒體播放品質(multimedia playback quality)。由於多路徑(multi-path)傳輸可聚合頻寬(aggregate bandwidth)，故能更有效的利用網路資源來提供即時多媒體應用使用。大多數的多路徑傳輸相關文獻著重於處理封包失序(packets out-of-order)現象及處理重要影格類型遺失的問題，但皆未考量傳輸具時效性的多媒體應用應及時將資料送達接收端，因此本研究提出適合多媒體應用之多路徑傳輸排程法－頻寬感知排程法(Bandwidth Aware Scheduling Algorithm, BASA)，其根據多媒體應用資料組成的特性並依多路徑傳輸時網路環境的變化將影格加以重新排程。BASA依照影格類型的重要性依序分配於傳輸時間最短的路徑上，同時調整影格傳送的順序以縮短傳輸時間，隨後將預期無法在時限前送達接收端的影格予以丟棄，以期望能將較重要的影格及時送達接收端，提升多媒體播放品質。本方法與多路徑傳輸文獻裡的SMOS(Sender-based Multipath Out-of-order Scheduling)法比較，模擬結果顯示BASA在不同的環境中皆能有效改善SMOS的排程結果，在不同頻寬環境下BASA能提升SMOS之DFR (Decodable Frame Ratio)可達131.3%，而在不同傳播延遲下能提升SMOS之DFR可達241.3%。

The needs using real-time multimedia applications continue increasing. However, bandwidth supported by current network technology is still limited, causing that satisfactory multimedia playback quality can not be provided when a large amount of data is transmitted. Multi-path transmission with aggregating bandwidth can well-utilize network resources for real-time multimedia applications. Most previous papers conquered with the problems of out-of-order packets and the important frame losses. However, they do not consider that data should arrive to the receiver in time because of the real-time characteristics of multimedia applications. Therefore, the thesis proposes a multi-path transmission scheduling which is suitable for multimedia applications, called Bandwidth Aware Scheduling Algorithm (BASA). BASA considers the composition of multimedia and schedules frames according to network variation at transmitting. Depending on the order of frame importance, BASA distributes a frame to the path which can transmits it with the shortest time and adjusts the transmitted orders of frames to shorten their transmission time. After that, BASA drops the frames which can not arrive to the receiver in time. As the results, the receiver can receive most important frames before deadlines to raise the playback quality. Comparing BASA with a related work, SMOS (Sender-based Multipath Out-of-order Scheduling), the simulation results show that BASA can improve the scheduling results of SMOS under various environments. Under different bandwidth, BASA can increase SMOS DFR (Decodable Frame Ratio) up to 131.3%. Under different propagation delay, BASA can increase SMOS DFR up to 241.3%.

[1] R. Stewart, “RFC 4960 - Stream Control Transmission Protocol,” Sept. 2007.
[2] M. Bagnulo, “RFC6181-Threat Analysis for TCP Extensions for Multipath Operation with Multiple Addresses,” Mar. 2011.
[3] Farhan H. Mirani, Mahmoud Kherraz, and Nadia Boukhatem, “Forward Prediction Scheduling: Implementation and Performance Evaluation,” 2011 18th International Conference on Telecommunications (ICT). Pages 321-326, May 2011.
[4] Han Ah Kim, Bong-hwan Oh, and Jaiyong Lee, “Improvement of MPTCP Performance in heterogeneous network using packet scheduling mechanism,” 2012 18th Asia-Pacific Conference on Communications (APCC). Pages 842-847, Oct. 2012.
[5] Chung-Ming Huang, Yih-Chung Chen, and Shih-Yang Lin, “The Qos-aware Order Prediction Scheduling (QOPS) Scheme for Video Streaming Using the Multi-path Datagram Congestion Control Protocol (MP-DCCP),” 2012 15th International Conference on Network-Based Information Systems (NBiS). Pages 276-283, Sept. 2012.
[6] Golam Sarwar, Roksana Boreli, Emmamuel Lochin, Ahlem Mifdaoui, and Guillaume Smith, “Mitigating Receiver’s Buffer Blocking by Delay Aware Packet Scheduling in Multipath Data Transfer,” 2013 27th International Conference on Advanced Information Networking and Applications Workshops (WAINA). Pages 1119-1124, Mar. 2013.
[7] Allen L. Ramaboli, Olabisi E. Falowo, and Anthony H. Chan, “Using Multiple Links Simultaneously to Increase Capacity for Multi-homed Terminals in Heterogeneous Wireless Networks,” 2013 IEEE 27th International Conference on Advanced Information Networking and Applications (AINA). Pages 788-793, Mar. 2013.
[8] Feng Zhong, Chai Kiat Yeo, and Bu Sung Lee, “Adaptive Load Balancing Algorithm for Multiple Homing Mobile Nodes,” Journal of Network and Computer Applications. Volume 35, Issue 1, Pages 316-327, Jul. 2011.
[9] Arpad Huszak and Sandor Imre, “Content-aware Interface Selection Method for Multi-Path Video Streaming in Best-effort Networks,” 2009. ICT '09. International Conference on Telecommunications. Pages 196-201, May 2009.
[10] Wang Xia and Jiang Hong-an, “Packet Scheduling Scheme for MPEG Video Transportation,” 2011 International Conference on Multimedia Technology (ICMT). Pages 5013-5016, Jul. 2011.
[11] Varun Singh, Saba Ahsan, and Jorg Ott, “MPRTP: Multipath Considerations for Real-time Media,” Proceedings of the 4th ACM Multimedia Systems Conference. Pages 190-201, Feb. 2013.
[12] Code Diop, Guillaume Dugue, Christophe Chassot, and Ernesto Exposito, “Qos-oriented MPTCP extensions for multimedia multi-homed systems,” 2012 26th International Conference on Advanced Information Networking and Applications Workshops (WAINA). Pages 1119-1124, Mar. 2012.
[13] Jianxin Liao, Jingyu Wang, Tonghong Li, and Ping Zhang, “Sender-based Multipath Out-of-order Scheduling for High-definition Videophone in Multi-homed Devices,” IEEE Transactions on Consumer Electronics. Pages 1466-1472, Aug. 2010.
[14] Jianxin Liao, Jingyu Wang, and Tonghong Li, “OSIA: Out-of-order Scheduling for In-order Arriving in concurrent multi-path transfer,” Journal of Network and Computer Applications. Volume 35, Issue 2, Pages 633-643, Sept. 2011.
[15] The Network Simulator-ns-2, 2011, http://www.isi.edu/nsnam/nsh