視訊串流已成為網際網路流量的主要來源，面對其對頻寬需求的持續攀升，能夠聚合異質網路頻寬的多路徑傳輸成為極具吸引力的解決方案。Multi-Path QUIC（MP-QUIC）擴充了 QUIC，可在單一連線中同時透過多條路徑傳輸資料，藉以聚合多條路徑的頻寬。然而，各路徑在延遲、頻寬與封包遺失率上的異質特性，會造成封包亂序（Out-of-Order, OFO）抵達與隊頭阻塞（Head-of-Line Blocking, HoLB）現象，進而降低多路徑的使用效率與視訊串流的體驗品質（Quality of Experience, QoE）。特別是傳統分段式（segment-based）串流概念必須在目前分段完全下載後才會請求下一個分段，導致較快的路徑在等待慢速路徑上延遲封包的期間閒置，因而造成頻寬使用率不足，以及對後續視訊分段之傳輸頻寬的低估。為解決上述問題，本論文針對 MP-QUIC 多路徑網路環境下的非 HTTP 視訊串流，提出「分段視窗式視訊串流結合主動暫停機制」（Segment Window-oriented Video Streaming with Proactive Suspending, SWVS-PS）方法。SWVS-PS 整合三項機制：(i) 分段視窗傳輸機制，同時維護兩個可下載的分段，使較快的路徑得以在視窗前端分段於慢速路徑延遲時傳輸視窗後端分段的視訊資料，藉以提升多路徑的頻寬使用率；(ii) 分段視窗導向且具緩衝區感知的位元率自適應機制，依據分段視窗導向頻寬估測與所預測的應用層緩衝區佔用量，決定後續分段的品質等級；(iii) 主動暫停機制，於視窗前端分段之傳輸在一時間週期內無法完成時，(a) 暫時暫停該慢速路徑，並 (b) 將被暫停路徑上仍在傳輸中的封包透過較快路徑重複重傳，以確保分段能及時且循序地完成下載。本論文以 ns-3 於 LTE、Wi-Fi 與乙太網路所組成的三路徑異質 MP-QUIC 拓樸中評估所提方法，其可用頻寬在背景流量的適中（moderate）與繁重（heavy）兩個階段之間交替變化，使播放緩衝區在串流過程中反覆被耗盡與再填補。效能評估結果顯示，僅採用分段視窗機制即可將平均播放位元率由基準方法的 3.314 Mbps 提升至 3.706 Mbps，約提升 11.83%，並使播放停頓總時間縮短約 39%。在搭配主動暫停機制且設定 θ = 0.7 時，SWVS-PS 進一步達到最高的平均播放位元率與最高的平均播放品質等級，較基準方法分別高出約 12.01% 與 7.53%，同時使播放停頓總時間相較基準方法縮短約 67%、相較僅使用分段視窗機制縮短約 47%。與其他多路徑視訊串流方法相比，SWVS-PS 提供最高的播放品質等級，且其品質切換次數與切換幅度皆與基準方法相當，所付出的額外代價僅為相較於將可用頻寬用於冗餘重注（re-injection）而非預先下載後續分段之方法時，略高的播放停頓。綜上所述，受惠於聚合頻寬的更佳使用與避免使用被暫停的路徑，SWVS-PS 能提升 MP-QUIC 多路徑網路環境下視訊串流的頻寬使用率，並提供最佳的播放品質。

Video streaming has become the dominant source of Internet traffic, and its ever-increasing bandwidth demand makes multi-path transmission an attractive solution for aggregating the bandwidth of heterogeneous network paths. Multi-Path QUIC (MP-QUIC) extends QUIC to transmit data simultaneously over multiple paths within a single connection, which is able to aggregate the bandwidth over those multiple paths. However, the heterogeneity of path characteristics in delay, bandwidth, and packet loss rate introduces Out-oF-Order (OFO) packet arrivals and the Head-of-Line Blocking (HoLB) phenomena, which degrade the utilization of multiple paths concurrently and the Quality of Experience (QoE) of video streaming. In particular, the traditional segment-based streaming concept requests the next segment only after the current segment has been completely downloaded, for which the faster paths are left idle while the client waits for the delayed packets on a slow path. It thus results in bandwidth underutilization and an under-estimated bandwidth for transmitting subsequent video segments. To address these problems, this work proposes the Segment Window-oriented Video Streaming with Proactive Suspending (SWVS-PS) method for non-HTTP-based video streaming over the MP-QUIC multi-path networking environment. SWVS-PS integrates three mechanisms: (i) A segment window transmission mechanism that maintains two segments to be downloaded concurrently. It allows the faster paths to able to transmit the video data of the segment window’s rear segment while the segment window’s front segment is delayed in the slow path, and thus can improve multi-path’s bandwidth utilization; (ii) a segment window-oriented buffer-aware bitrate adaptation scheme that determines the quality level of the follow-up segment according to the segment window-oriented bandwidth estimation and the predicted application-layer buffer occupancy; (iii) a proactively suspending mechanism ensures the timely sequential completion of segment downloading, which temporarily suspends a path once the amount of the in-sequence acknowledged QUIC frames of the rear segment exceeds the threshold determined by the given parameter θ, duplicated-ly retransmits the on-the-fly packets of the suspended path over the active paths, and reactivates the suspended path when its monitored smoothed RTT becomes better than the smoothed RTT recorded at the moment of its suspension. Consequently, segment downloading is completed through the fast paths without being blocked by the slow path, while no path is permanently abandoned once its condition has recovered. The proposed method is evaluated using ns-3 over a three-path heterogeneous MP-QUIC topology, i.e., LTE, Wi-Fi, and Ethernet, whose available bandwidth alternates between a moderate and a heavy background traffic phase, so that the presentation buffer is repeatedly drained and replenished during the streaming. The evaluation results show that the segment window mechanism alone raises the average playout bitrate from the baseline's 3.314 Mbps to 3.706 Mbps, i.e., about 11.83 % higher, and shortens the total playback stall time by about 39 %. Cooperating with the proactively suspending mechanism configured at θ = 0.7, SWVS-PS further attains the highest average playout bitrate of 3.712 Mbps and the highest average playout quality level, which are about 12.01 % and 7.53 % higher than those of the baseline, respectively, while reducing the total playback stall time by about 67 % compared with the baseline and by about 47 % compared with using the segment window mechanism alone. Comparing with the other multi-path video streaming methods, SWVS-PS provides the highest playout quality level while keeping the amount and the magnitude of quality switches comparable with those of the baseline, for which the additional cost is only a slightly higher playback stall than the method that spends its available bandwidth on redundant re-injection rather than on pre-downloading the follow-up segments. Benefiting from the better utilization of the aggregated bandwidth and the avoidance of using the suspended path, SWVS-PS enhances the bandwidth utilization and provides the highest playout quality for video streaming over the MP-QUIC multi-path networking environment.

[1] Nokia and KGPCo, "Global Network Traffic Report: 2023-2033 Forecasts and Trends," Nokia, 2023. [Online]. Available: https://www.nokia.com/asset/213660/
[2] C. Raiciu, C. Paasch, S. Barre, A. Ford, M. Honda, F. Duchene, O. Bonaventure, and M. Handley, "How Hard Can It Be? Designing and Implementing a Deployable Multipath TCP," in Proceedings of the 9th USENIX Symposium on Networked Systems Design and Implementation (NSDI), San Jose, CA, USA, 2012, pp. 399-412.
[3] A. Ford, C. Raiciu, M. Handley, O. Bonaventure, and C. Paasch, "TCP Extensions for Multipath Operation with Multiple Addresses," RFC 8684, Mar. 2020. [Online]. Available: https://www.rfc-editor.org/info/rfc8684
[4] Q. De Coninck and O. Bonaventure, "Multipath QUIC: Design and Evaluation," in Proceedings of the 13th ACM International Conference on Emerging Networking Experiments and Technologies (CoNEXT), New York, NY, USA, 2017, pp. 160-166.
[5] I. Sodagar, "The MPEG-DASH Standard for Multimedia Streaming over the Internet," IEEE MultiMedia, vol. 18, no. 4, pp. 62-67, Apr. 2011.
[6] T. Y. Huang, R. Johari, N. McKeown, M. Trunnell, and M. Watson, "A Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Service," ACM SIGCOMM Computer Communication Review, vol. 44, no. 4, pp. 187-198, Oct. 2014.
[7] M. Scharf and S. Kiesel, "Head-of-Line Blocking in TCP and SCTP: Analysis and Measurements," in Proceedings of the IEEE Global Communications Conference (GLOBECOM), San Francisco, CA, USA, 2006, pp. 1-5.
[8] J. Iyengar and M. Thomson, "QUIC: A UDP-Based Multiplexed and Secure Transport," RFC 9000, May 2021. [Online]. Available: https://www.rfc-editor.org/info/rfc9000
[9] M. Bishop, Ed., "HTTP/3," RFC 9114, Jun. 2022. [Online]. Available: https://www.rfc-editor.org/info/rfc9114
[10] A. Langley, A. Riddoch, A. Wilk, A. Vicente, C. Krasic, D. Zhang, F. Yang, F. Kouranov, I. Swett, J. Iyengar, J. Bailey, J. Dorfman, J. Roskind, J. Kulik, P. Westin, R. Tenneti, R. Shade, R. Hamilton, V. Vasiliev, W. T. Chang, and Z. Shi, "The QUIC Transport Protocol: Design and Internet-Scale Deployment," in Proceedings of the ACM SIGCOMM Conference, Los Angeles, CA, USA, 2017, pp. 183-196.
[11] J. Iyengar and I. Swett, "QUIC Loss Detection and Congestion Control," RFC 9002, May 2021. [Online]. Available: https://www.rfc-editor.org/info/rfc9002
[12] C. Cui, Y. Lu, S. Li, J. Li, and Z. Ruan, "DASH+: Download Multiple Video Segments with Stream Multiplexing of QUIC," in Proceedings of the 10th International Conference on Advanced Cloud and Big Data (CBD), Guilin, China, 2022, pp. 66-72.
[13] M. Nguyen, D. Lorenzi, F. Tashtarian, H. Hellwagner, and C. Timmerer, "DoFP+: An HTTP/3-Based Adaptive Bitrate Approach Using Retransmission Techniques," IEEE Access, vol. 10, pp. 109565-109579, Oct. 2022.
[14] L. Guillen, S. Izumi, T. Abe, and T. Suganuma, "SAND/3: SDN-Assisted Novel QoE Control Method for Dynamic Adaptive Streaming over HTTP/3," Electronics, vol. 8, no. 8, p. 864, Aug. 2019.
[15] A. Rabitsch, P. Hurtig, and A. Brunstrom, "A Stream-Aware Multipath QUIC Scheduler for Heterogeneous Paths," in Proceedings of the Workshop on the Evolution, Performance, and Interoperability of QUIC (EPIQ), 2018, pp. 29-35.
[16] B. Jonglez, M. Heusse, and B. Gaujal, "SRPT-ECF: Challenging Round-Robin for Stream-Aware Multipath Scheduling," in Proceedings of the IFIP Networking Conference, Paris, France, 2020, pp. 719-724.
[17] Y. Xing, K. Xue, Y. Zhang, J. Han, J. Li, D. S. L. Wei, R. Li, Q. Sun, and J. Lu, "A Stream-Aware MPQUIC Scheduler for HTTP Traffic in Mobile Networks," IEEE Transactions on Wireless Communications, vol. 22, no. 4, pp. 2775-2788, Apr. 2023.
[18] B. Han, C. Xu, Y. Li, X. Wang, and P. Xun, "MPR-QUIC: Multi-Path Partially Reliable Transmission for Priority and Deadline-Aware Video Streaming," Journal of Systems Architecture, vol. 153, Aug. 2024, Art. no. 103195.
[19] J. Han, J. Liu, K. Xue, Q. Sun, and J. Lu, "Toward High-Quality Real-Time Video Streaming: An Efficient Multi-Stream and Multi-Path Scheduling Framework," IEEE Transactions on Networking, vol. 33, no. 4, pp. 1826-1839, Aug. 2025.
[20] H. Kang, S. Hong, D. Kang, and D. An, "Deadline-Aware Redundant Transmission on Multi-Paths for Real-Time Service," IEEE Access, vol. 13, pp. 51589-51602, Mar. 2025.
[21] X. Shi, L. Wang, F. Zhang, B. Zhou, and Z. Liu, "PStream: Priority-Based Stream Scheduling for Heterogeneous Paths in Multipath-QUIC," in Proceedings of the 29th International Conference on Computer Communications and Networks (ICCCN), Honolulu, HI, USA, 2020, pp. 1-8.
[22] X. Liang, B. Zhao, W. Peng, and T. Wang, "Towards Effective Multipath Scheduling with Multipath QUIC in Heterogeneous Paths," in Proceedings of the 10th International Conference on Information Systems and Computing Technology (ISCTech), Guilin, China, 2022, pp. 472-479.
[23] H. Zeng, L. Cui, F. P. Tso, and Z. Zhang, "Optimizing Multipath QUIC Transmission over Heterogeneous Paths," Computer Networks, vol. 215, Art. no. 109198, Oct. 2022.
[24] S. Ferlin, Ö. Alay, O. Mehani, and R. Boreli, "BLEST: Blocking Estimation-Based MPTCP Scheduler for Heterogeneous Networks," in Proceedings of the IFIP Networking Conference and Workshops, Vienna, Austria, 2016, pp. 431-439.
[25] Y. S. Lim, E. M. Nahum, D. Towsley, and R. J. Gibbens, "ECF: An MPTCP Path Scheduler to Manage Heterogeneous Paths," in Proceedings of the 13th ACM International Conference on Emerging Networking Experiments and Technologies (CoNEXT), New York, NY, USA, 2017, pp. 147-159.
[26] Y. Liu, Y. Ma, Q. De Coninck, O. Bonaventure, C. Huitema, and M. Kühlewind, "Multipath Extension for QUIC," Internet Eng. Task Force, Internet-Draft draft-ietf-quic-multipath-10, Mar. 2024, Work in Progress. [Online]. Available: https://datatracker.ietf.org/doc/draft-ietf-quic-multipath/10/
[27] C. Paasch, S. Ferlin, Ö. Alay, and O. Bonaventure, "Experimental Evaluation of Multipath TCP Schedulers," in Proceedings of the ACM SIGCOMM Workshop on Capacity Sharing (CSWS), Chicago, IL, USA, 2014, pp. 27-32.
[28] S. Nandakumar, V. Vasiliev, I. Swett, Ed., and A. Frindell, Ed., "Media over QUIC Transport," Internet Eng. Task Force, Internet-Draft draft-ietf-moq-transport-18, May 2026, Work in Progress. [Online]. Available: https://datatracker.ietf.org/doc/draft-ietf-moq-transport/18/
[29] M. H. Dwijaksara, M. H. Hilman, C. Lee, and W. Lee, "Rate Adaptation Technique for Media Streaming over QUIC with Limited Backhaul," IEEE Access, vol. 12, pp. 139028-139041, Sep. 2024.
[30] A. C. Freeman, M. Rudolph, T. Redoy, F. Schnier, S. Afzal, H. Hassler, and A. Rizk, "Point Cloud Streaming with Latency-Driven Implicit Adaptation Using MoQ," in Proceedings of the 36th Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV), New York, NY, USA, 2026, pp. 57-63.
[31] S. Patil and S. Gaikwad, "Proactive Congestion Detection and Video QoE Preservation over QUIC Using LSTM," in Proceedings of the IEEE International Conference on Blockchain and Distributed Systems Security (ICBDS), Kolhapur, India, 2025, pp. 1-6.
[32] J. S. Sidhu, A. Sahu, and A. Bentaleb, "TAROT: Towards Optimization-Driven Adaptive FEC Parameter Tuning for Video Streaming," in Proceedings of the ACM Multimedia Systems Conference (MMSys), New York, NY, USA, 2026, pp. 226-237.
[33] Y. Cao, H. Zhang, M. Jiang, Y. Jiang, and J. Nie, "When Multipath QUIC Meets Model Predictive Control and Band Sparse Network Coding: A Novel Multipathing Solution for Video Streaming over Heterogeneous Wireless Networks," IEEE Transactions on Broadcasting, vol. 71, no. 3, pp. 756-773, Sep. 2025.
[34] D. Q. Mayne, "Model Predictive Control: Recent Developments and Future Promise," Automatica, vol. 50, no. 12, pp. 2967-2986, Dec. 2014.
[35] Y. Li, J. Zhu, and Z. Bao, "Sparse Random Linear Network Coding with Precoded Band Codes," IEEE Communications Letters, vol. 21, no. 3, pp. 480-483, Mar. 2017.
[36] B. Kimura, S. Ferlin, T. Paiva, T. Mahmoodi, A. Brunstrom, and O. Alay, "Evaluating Adaptive Video Streaming over Multipath QUIC with Shared Bottleneck Detection," ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 21, no. 9, Art. no. 246, Sep. 2025.
[37] Z. Wei, Q. Li, T. Zhao, C. Luo, C. Ouyang, X. Yu, D. Zhao, and Y. Jiang, "EMVOD: Elastic Multi-Path QUIC Scheduling for CDN Video-on-Demand Service," in Proceedings of the 21st European Conference on Computer Systems (EuroSys), New York, NY, USA, 2026, pp. 532-547.
[38] Z. Zheng, Y. Ma, Y. Liu, F. Yang, Z. Li, Y. Zhang, J. Zhang, W. Shi, W. Chen, D. Li, Q. An, H. Hong, H. H. Liu, and M. Zhang, "XLINK: QoE-Driven Multi-Path QUIC Transport in Large-Scale Video Services," in Proceedings of the ACM SIGCOMM Conference, New York, NY, USA, 2021, pp. 418-432.
[39] G. Lv, Q. Wu, Y. Liu, Z. Li, Q. Tan, F. Yang, W. Chen, Y. Ma, H. Guo, Y. Chen, and G. Xie, "Chorus: Coordinating Mobile Multipath Scheduling and Adaptive Video Streaming," in Proceedings of the 30th Annual International Conference on Mobile Computing and Networking (MobiCom), New York, NY, USA, 2024, pp. 246-262.
[40] B. Hu, T. Li, C. Qiao, and J. Cao, "CROSS: A Dual-Sided Scheduling Framework for Efficient Multipath Video Streaming," in Proceedings of the 3rd Workshop on Emerging Multimedia Systems (EMS), New York, NY, USA, 2025, pp. 13-18.
[41] A. Sepahi, L. Cai, W. Yang, and J. Pan, "LiveStream Meta-DAMS: Multipath Scheduler Using Hybrid Meta Reinforcement Learning for Live Video Streaming," IEEE Transactions on Cognitive Communications and Networking, vol. 11, no. 4, pp. 2739-2754, Aug. 2025.
[42] Y. Zanichkovskyy, L. Hlinenko, and V. Fast, "Contextual Multi-Arm Bandit Framework for Adaptive Video Streaming," in Proceedings of the IEEE 18th International Conference on Advanced Trends in Radioelectronics, Telecommunications and Computer Engineering (TCSET), Lviv, Ukraine, 2026, pp. 1-4.
[43] L. Li, W. Chu, J. Langford, and R. E. Schapire, "A Contextual-Bandit Approach to Personalized News Article Recommendation," in Proceedings of the 19th International Conference on World Wide Web (WWW), Raleigh, NC, USA, 2010, pp. 661-670.
[44] Z. Deng, Y. Liu, J. Liu, A. Argyriou, and D. Liu, "BBR-Based and Fairness-Guaranteed Congestion Control and Packet Scheduling for MPQUIC over Heterogeneous Networks," Computer Communications, vol. 224, pp. 213-224, Jun. 2024.
[45] N. Cardwell, Y. Cheng, C. S. Gunn, S. H. Yeganeh, and V. Jacobson, "BBR: Congestion-Based Congestion Control," ACM Queue, vol. 14, no. 5, pp. 20-53, Sep.-Oct. 2016.
[46] J. S. Wejin, A. A. Adewale, and K. O. Okopkujie, "MPS-OF-AS: An SDN-Based MPQUIC Scheduler for Software-Defined Cloud Services," Scientific African, vol. 31, Art. no. e03134, 2026.
[47] Y. Ren, C. Wu, A. Shan, J. Chen, Y. Lin, and Z. Du, "A QoS-Driven Fuzzy Logic-Based Adaptive MPQUIC Scheduler for Heterogeneous Networks," IEEE Transactions on Network Science and Engineering, vol. 13, pp. 2586-2603, 2026.
[48] H. Joo, N. Yoon, T. Yoo, and H. Kim, "Time-Efficient Multipath Transmission for Reliable Internet-of-Things Networks," in Proceedings of the IEEE Annual Congress on Artificial Intelligence of Things (AIoT), Osaka, Japan, 2025, pp. 99-106.
[49] K. Cai, Z. Chen, J. Zhang, and J. C. S. Lui, "OLMS: A Flexible Online Learning Multi-Path Scheduling Framework," IEEE Transactions on Network Science and Engineering, vol. 12, no. 3, pp. 2277-2291, May-Jun. 2025.
[50] P. Auer, N. Cesa-Bianchi, Y. Freund, and R. E. Schapire, "The Nonstochastic Multiarmed Bandit Problem," SIAM Journal on Computing, vol. 32, no. 1, pp. 48-77, 2002.
[51] M. H. Vu and A. Nakao, "Logrα: A DRL-Based MPQUIC Scheduler for Managing Periodicity and Heterogeneity in Satellite Networks," in Proceedings of the 23rd IEEE Consumer Communications and Networking Conference (CCNC), Las Vegas, NV, USA, 2026, pp. 1-4.
[52] H. Ott, K. Miller, and A. Wolisz, "Simulation Framework for HTTP-Based Adaptive Streaming Applications," in Proceedings of the Workshop on ns-3 (WNS3), 2017, pp. 95-102.
[53] G. F. Riley and T. R. Henderson, "The ns-3 Network Simulator," in Modeling and Tools for Network Simulation, K. Wehrle, M. Güneş, and J. Gross, Eds. Berlin, Heidelberg: Springer, 2010, pp. 15-34.
[54] S. Shu, W. Yang, J. Pan, and L. Cai, "A Multipath Extension to the QUIC Module for ns-3," in Proceedings of the Workshop on ns-3 (WNS3), 2023, pp. 86-93.
[55] R. Khalili, N. Gast, M. Popovic, and J.-Y. Le Boudec, "MPTCP Is Not Pareto-Optimal: Performance Issues and a Possible Solution," IEEE/ACM Transactions on Networking, vol. 21, no. 5, pp. 1651-1665, Oct. 2013.
[56] YouTube, "Recommended Upload Encoding Settings." 2023. [Online]. Available: https://support.google.com/youtube/answer/1722171