視頻傳輸技術伴隨著電視和電影廣泛傳播與快速發展，早期僅在特定的時間、有限的地點並且已經存儲相應的媒體資料才能觀看流暢的影音串流，近年來隨著科學技術的蓬勃發展，互聯網時代的日新月異以及日常生活節奏的逐步加快，人們對視頻的品質要求越來越高，從而視頻傳輸技術更重視體驗特質 (Quality of Experience, QoE)。
影音串流傳輸從靜態照片或單純語音的傳輸,到今天已實現了聲畫動態同步傳輸，但由於網路技術發展的滯後性，在綫視頻播放過程中流暢和卡頓之現象時有發生，這極大的影響了QoE，包括等待時間、流暢程度等要素。傳統的研究通過僅設置用戶端的緩存空間，或進通過提前向用戶端預存並推薦其可能喜好的視頻等方案來提高QoE，本研究結合兩者的優勢，最優化QoE。本文以視頻提供者的角度，研究使用者端視頻的播放過程，通過在使用端動態設置緩存空間和推薦給使用者喜好視頻的空間，同時重點控制上溢影音串流，控制下溢視頻流等方法，來提高QoE。本論文提出了一種新的資源管理方案，以最佳化技術為基礎優化影音串流而達到最優化QoE。

The technology of video transmission with television and movies wide boom so that promote quickly development .In early time, users only in fixed time, fixed location, already perfected full media data can enjoy fluently movies. Recent year, with the progress of science and technology, the fast development of internet and the pace of life gradually accelerate, more and more users need to high quality of service in video, lead to the technology of video transmission was more and more attention to QoE .
The technology of video transmission from static picture or only sound deliver to both sound and picture synchronization. During the user enjoy video time; video streaming maybe happens broke off or fluently. It is huge impact to QoE. It including waiting time,video streaming fluently and so on. Traditionally methods is only set the buffer size for user, or only Prefetching user’s favorite video. In this research, we proposal a new resource management that combination set the buffer size and Prefetching users favorite video to maximize QoE. The paper the perspective of media player, to set dynamic buffer size for user and buffer size for Prefetching favorite video, at the same time we focus on control of overflow video streaming and control of under of underflow video streaming. It is improve QoE. We propose an optimization-based resource management strategy to maximize QoE for real-time video streaming services.

Table of Contents
誌謝 ii
中文摘要 iii
ABSTRACT iv
Table of Contents v
LIST OF FIGURES vii
LIST OF TABLES ix
Chapter 1 Introduction 1
1.1 Background 1
1.2 Problem Description 2
1.3 Motivation 3
1.4 Thesis Organization 6
Chapter 2 Literature Review 7
Chapter 3 Problem Formulation 10
3.1 System Description 10
3.2 Model 14
Chapter 4 Solution Approaches 18
4.1 Introduction to Continuous Time Markov chains 18
4.1.1 First passage Time 18
4.1.2 Introduction to M/M/1 20
4.1.3 Mean Busy period 20
4.2 Introduction to Lagrangian Relaxation Method 21
4.3 Lagrangian Relaxation 24
4.4 The Dual problem and the Subgradient Method 26
Chapter 5 Computational Experiments 28
5.1 Simple Algorithm 28
5.2 Experimental Environment 28
5.3 Experimental Scenarios 29
5.4 Experiment Result 35
5.4.1 To Compared Lagrangian Relaxation Method Accuracy with LINGO Global Solution Accuracy 35
5.4.2 To Compared Lagrangian Relaxation Method Efficiency with LINGO Global Solution Efficiency 43
5.4.3 The Relationship between Power Ratio and Decision Variable 47
5.4.4 The Relationship between Overflow Time and Decision Variables 54
5.4.5 The Relationship between Underflow Time and Decision Variables 56
5.4.6 The Relationship between the number of Prefetching video M and Decision Variables 58
5.4.7 The Relationship among the Internet Speed, Overflow Time and Underflow Time 60
Chapter 6 Conclusion and Future work 64
6.1 Conclusion 64
6.2 Future Work 65
REFERENCES 66

[1] J. Budomo, I. Ahmad, D. Habibi, and E. Dines, "4G LTE-A systems at vehicular speeds: Performance evaluation," 2017 International Conference on Information Networking (ICOIN), Da Nang, Vietnam, pp. 321-326,Jan 2017.
[2] F. Cavaliere, L. Giorgi, and L. Potì, "Transmission and Switching Technologies for 5G Transport Networks," 2018 IEEE Optical Interconnects Conference (OI), Santa Fe, NM, USA, pp. 47-48,Feb 2018.
[3] R. Abuhadra and B. Hamdaoui, "Proactive In-Network Caching for Mobile On-Demand Video Streaming," 2018 IEEE International Conference on Communications (ICC), Kansas City, MO, USA, pp. 1-6,Dec 2018
[4]Netflix (Wilmot Reed Hastings, Jr.). 2018. [Online]. https://www.youtube.com/user/NewOnNetflix [Accessed: 28- Dec- 2018]
[5] D. Chitimalla, M. Tornatore, S. Lee, H. Lee, S. Park, H. Chung, and B.Mukherjee, "QoE enhancement schemes for video in converged OFDMA wireless networks and EPONS," in IEEE/OSA Journal of Optical Communications and Networking, vol. 10, no. 3, pp. 229-239, March 2018.
[6]T. Kimura, M. Yokota, A. Matsumoto, K. Takeshita, T. Kawano, K. Sato, H. Yamamoto, T. Hayashi, K. Shiomoto, and K. Miyazaki,"QUVE: QoE Maximizing Framework for Video-Streaming," in IEEE Journal of Selected Topics in Signal Processing, vol. 11, no. 1, pp. 138-153, Feb. 2017.
[7]M. Li, S. Jianbin, and L. Hui, "A Determining Method of Frame Rate and Resolution to Boost the Video Live QoE," 2017 2nd International Conference on Multimedia and Image Processing (ICMIP), Wuhan, China, pp. 206-209,Dec,2017.
[8]M. Gao, W. Zhou, and Z. Hu, "A QoE Estimation Model Considering Video Popularity for Video Streaming Services," 2018 IEEE International Conference on Big Data and Smart Computing (BigComp), Shanghai, China, pp. 354-359,Dec 2018.
[9]Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update, 2016–2021 White Paper (Cisco Mobile VNI). [Online]. https://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/mobile-white-paper-c11-520862.html [Accessed: 28- Dec- 2018]
[10] Statista (Cisco Systems). [Online]. https://www.statista.com/statistics/271405/global-mobile-data-traffic-forecast/ [Accessed: 28- Dec- 2018]
[11] Percentage of internet users who watch online video content on any device as of January 2018, by country (Google Consumer Survey). [Online]. https://www.statista.com/statistics/272835/share-of-internet-users-who-watch-online-videos/ [Accessed: 28- Dec- 2018]
[12]Y. Huang, S. Mao and S. F. Midkiff, "A Control-Theoretic Approach to Rate Control for Streaming Videos," in IEEE Transactions on Multimedia, vol. 11, no. 6, pp. 1072-1081, Oct. 2009.
[13]Moshe Zukerman” Introduction to Queueing Theory and Stochastic Teletraffic Models”,pp.1-182.Apr 2018
[14]G. Gao, Y. Wen and H. Hu, "QDLCoding: QoE-differentiated low-cost video encoding scheme for online video service," IEEE INFOCOM 2017 - IEEE Conference on Computer Communications, Atlanta, GA, USA, pp. 1-9,Dec,2017.
[15]M. Kovacevic, B. Kovacevic, D. Stefanovic and S. Novak, "Automated monitoring of HTTP live streaming QoE factors on Android STB," 2015 IEEE 1st International Workshop on Consumer Electronics (CE WS), Novi Sad, Serbia, pp. 72-75.Dec,2015.
[16]M. Seufert, N. Wehner and P. Casas, "Studying the Impact of HAS QoE Factors on the Standardized QoE Model P.1203," 2018 IEEE 38th International Conference on Distributed Computing Systems (ICDCS), Vienna, Austria, pp. 1636-1641.Dec,2018.
[17]Y. Ben Youssef, M. Afif and S. Tabbane, "Novel AHP-based QoE factors'' selection approach," 2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA), Agadir, Morocco, pp. 1-6.Dec,2016.
[18]W. Yu and M. Zhang, "Super Resolution Reconstruction of Video Images Based on Improved Glowworm Swarm Optimization Algorithm," 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), Chongqing, China, pp. 331-335.Dec,2018.