The enormous and ever-growing amount of web images has brought great challenges and opportunities to computer vision. A lot of recent research efforts have been devoted to the effective access of web images (e.g., web image re-ranking, content-based image retrieval), as well as to making better use of web images as auxiliary data sources. A key problem that needs to be solved is to model the semantic and visual structure of web images and to effectively organize the web image collections. This thesis proposes Visual Semantic Complex Network (VSCN), a automatically generated graph structure to model their structure and organize the web image collections. The nodes of the VSCN are compact clusters of images with visual and semantic consistency, named semantic concepts. These nodes are connected based on the visual and semantic correlations. The VSCN consists of 33,240 semantic concepts and cover 10 million images, and is expandable given more computational resources. / The VSCN enables a macroscopic understanding of the structure of web image collection. We study its structural properties from the viewpoint of complex network, and derive a great deal of valuable information. We study the connectivity and efficiency of the VSCN, and show that it is a well connected and efficient network. The study of in-degrees reflects the unequal importance of the semantic concepts. We also show the presence of community structures in the VSCN through analyzing the clustering coefficient. These studies not only gives us a better understanding of the VSCN and the web image collection at a macroscopic level, but also inspires a number of import applications of the VSCN in traditional vision tasks. / The VSCN models the structure of web images, it can be used to assist a variety of vision tasks and provide effective access to web images. We demonstrate three applications of the VSCN in this thesis. In the first application, we devise Anchor Concept Graph Distance (ACGD Distance) based on the VSCN and use this distance for web image re-ranking. As the VSCN effectively models the structures of web images, the ACG distance is able to reduce the semantic gap and better represents similarities between images. We demonstrate its effectiveness on image re-ranking benchmarks. The second application leverages the VSCN to improve the content-based image retrieval algorithms. As the relevant images on the web are connected via inter-concept correlations and grouped by community structures, we can therefore effectively reduce the search space by exploiting the structures of web images encoded in the VSCN. Therefore, the image retrieval task can greatly benefit from the VSCN. In the third application, we develop an novel image browsing scheme based on the VSCN. With the help of the VSCN structure, our proposed browsing scheme allows image browsing over the entire network via navigating the inter-concept connections. The browsing process not only bridges relevant images indexed under different textual queries, it is also guided by natural and meaningful transitions of images. We demonstrate the the VSCN-based scheme leads to better user experience for image browsing. Apart from benefiting better access to web image collections, the VSCN can be also used as an large, structured auxiliary database to assist other vision tasks. In this thesis, we present an example application of object recognition. Our method employs both imagery and textual data indexed by the VSCN, as well as its structural information. Experiments on benchmark datasets demonstrate the effectiveness of our method and the potential of the VSCN database. / 海量並且不斷增長的互聯網圖片給計算機視覺的研究帶來了很大的挑戰和機遇。許多近期的研究致力於網絡圖片的有效獲取（例如網絡圖片重排序，基於內容的圖片檢索即CBIR），以及更有效利用網絡圖片作為輔助數據源。其中一個亟待解決的關鍵問題是如何對網絡圖片間的語義和視覺結構進行建模，並據此組織這些圖片。本論文提出一個自動生成的網絡結構，即視覺語義復雜網絡（VSCN）來建模網絡圖片的結構。VSCN上的節點由一些在視覺和語義層面具有一致性的圖片聚類組成，稱之為語義概念。這些節點又由它們之間的視覺和語義相關性所連接。VSCN共包含33,240個語義概念並囊括了超過一千萬的圖片，並且可以隨著計算資源的增加而不斷增長。 / VSCN使得對網絡圖片結構的宏觀描述成為可能。我們從復雜網絡的角度對它的結構性質進行了研究，並且得出了許多有價值的信息。我們研究了VSCN的連通性和效率，說明它是一個聯通性和效率較好的網絡。對網絡入度的研究反應了語義概念具有不均等的重要性。我們通過分析聚類系數，說明了VSCN存在較明顯的社區結構。這些研究不僅僅為我們提供了對於VSCN和網絡圖片結構的宏觀理解，並且啟發了許多VSCN在計算機視覺任務中的重要應用。 / 因為VSCN建模了網絡圖片結構，它可以被用於許多計算機視覺的任務中並提供對網絡圖片有效的獲取。在本論文中我們展示了三個VSCN的應用。在第一個應用中，我們基於VSCN設計了用於網絡圖片重排序的ACG距離。由於VSCN有效地建模了網絡圖片的結構，ACG距離能夠減少語義鴻溝並更好地表達圖片之間的相似性。我們在基准數據集上展現了此方法的有效性。第二個應用利用VSCN改進了CBIR系統的。由於網絡中的相關圖片被VSCN中的語義概念間相關性所聯系，並組成了社區結構，我們可以利用此種結構信息來有效地減小搜索空間。這使得圖片檢索任務獲益，並使檢索結果得到很大改善。在第三個應用中，我們基於VSCN設計了一個新穎的圖片瀏覽模式。我們的瀏覽模式可以允許通過語義概念間的連接對VSCN進行瀏覽。這個瀏覽過程不僅聯系了屬於不同文本檢索詞的相關圖片，並且被具有指導性的圖片過渡所輔助。我們證明這種基於VSCN的瀏覽方式能夠帶來圖片瀏覽中更好的用戶體驗。除去能夠使得獲取互聯網圖片獲益之外，VSCN同樣可以被用作一個大規模的具有結構的輔助數據庫，並輔助其他計算機使用的任務。在本論文中，我們介紹了一個在物體識別的應用例。我們的方法利用的VSCN所檢索的圖片和文本信息，以及VSCN本身的結構信息。在標准數據集上的實驗證明了我們方法以及VSCN作為輔助數據庫的有效性。 / Qiu, Shi. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2014. / Includes bibliographical references (leaves 145-164). / Abstracts also in Chinese. / Title from PDF title page (viewed on 04, October, 2016). / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only.

by Ngai Tin Yee. / Bibliography: leaves [28]-[29] / Thesis (M.Ph.)--Chinese University of Hong Kong, 1984

by Chun-kit Chan. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1995. / Includes bibliographical references (leaves 68-[73]). / Chapter 1 --- A High-Speed All-Optical Tunable-Channel Multi-Access (TCMA) Network --- p.1 / Chapter 1.1 --- Introduction --- p.1 / Chapter 1.2 --- Tunable-Channel Multi-Access (TCMA) Networks --- p.4 / Chapter 1.3 --- Protocols For TCMA Networks --- p.5 / Chapter 1.3.1 --- Media Access Procedure For ACTA Protocol --- p.6 / Chapter 1.3.2 --- Cyclc Utilization & Adaptive Algorithm --- p.8 / Chapter 1.3.3 --- Advantages of ACTA Protocol --- p.9 / Chapter 1.4 --- Proposed High-Speed Photonic TDM Implementation --- p.10 / Chapter 1.4.1 --- Centralized Pulse Source --- p.12 / Chapter 1.4.2 --- Channel-Tunable Transmitter --- p.14 / Chapter 1.4.3 --- High Speed All-Optical Demultiplexing --- p.22 / Chapter 1.4.4 --- Timing and Synchronization --- p.23 / Chapter 1.4.5 --- Advantages of the Proposed Scheme --- p.24 / Chapter 1.5 --- Other Network Issues --- p.25 / Chapter 1.5.1 --- Scalability --- p.25 / Chapter 1.5.2 --- Surviability --- p.26 / Chapter 1.5.3 --- Cost-Effectiveness --- p.27 / Chapter 1.6 --- Conclusion --- p.28 / Chapter 2 --- Theoretical Analysis of High Repetition Rate Optical Pulse Multiplication using Fiber Coupler Loop Configuration --- p.29 / Chapter 2.1 --- Introduction --- p.29 / Chapter 2.2 --- Single Coupler Loop Configuration --- p.30 / Chapter 2.3 --- Cascaded Coupler Loop Configuration --- p.34 / Chapter 2.4 --- Discussion --- p.36 / Chapter 2.5 --- Conclusion --- p.38 / Chapter 3 --- Spectrum Analysis of Multichannel CATV Systems --- p.39 / Chapter 3.1 --- Introduction --- p.39 / Chapter 3.2 --- Considerations of Analysis of Multichannel CATV Systems --- p.41 / Chapter 3.3 --- Effects of Incomplete-cycle Sampling --- p.43 / Chapter 3.4 --- Methods to Alleviate the Incomplete-cycle Sampling Effcct --- p.48 / Chapter 3.4.1 --- Windowing --- p.48 / Chapter 3.4.2 --- Correction --- p.54 / Chapter 3.5 --- Nonlinear Distortion in Multichannel CATV Systems --- p.55 / Chapter 3.5.1 --- Two-tone Third Order Distortion Estimation --- p.56 / Chapter 3.5.2 --- Composite Triple Beat Estimation --- p.62 / Chapter 3.6 --- A Procedure of Spectrum Analysis for Multichannel CATV Systems --- p.64 / Chapter 3.7 --- Conclusion --- p.67 / Bibliography --- p.68 / Chapter A --- Implementation of a high-speed TCMA optical network --- p.74 / Chapter A.1 --- System Setup --- p.74 / Chapter A.2 --- Channel-tunable Delay Line Circuit --- p.75 / Chapter A.3 --- DFB Laser diode --- p.76 / Chapter A.4 --- Erbium-doped Fiber Amplifier (EDFA) --- p.76 / Chapter A.5 --- 1 Gb/s to 16 Gb/s Fiber Multiplexer --- p.77 / Chapter A.6 --- Nonlinear Amplifying Loop Mirror (NALM) --- p.77 / Chapter A.7 --- Decision Circuit --- p.78 / Chapter B --- Frequency Assignment Scheme of CATV Systems --- p.79 / Chapter C --- Derivation --- p.80 / Chapter C.1 --- Channel Carrier Power Level Variation with Rectangular Window --- p.81 / Chapter C.2 --- Channel Carrier Power Level With Hanning Window Function ´Ø --- p.82 / Chapter C.3 --- Correction Factor for the Channel Carriers --- p.83 / Chapter C.4 --- Correction Scheme for Distortion Terms --- p.83

Wong, Ho Yin Starsky. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2002. / Includes bibliographical references (leaves 81-85). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.1 / Chapter 1.2 --- Problems of IP multicast --- p.2 / Chapter 1.3 --- End-system multicast --- p.3 / Chapter 1.4 --- The Challenge of ESM --- p.3 / Chapter 1.5 --- Document Roadmap --- p.4 / Chapter 2 --- System Architecture --- p.5 / Chapter 3 --- ESM Protocol --- p.8 / Chapter 3.1 --- ESM: Tree Formation Protocol --- p.8 / Chapter 3.1.1 --- Example of Tree Formation Protocol --- p.14 / Chapter 3.1.2 --- "The proof of ""Tree Formation Protocol always main- tains a tree topology""" --- p.15 / Chapter 3.1.3 --- "The proof of ""Tree Formation Protocol guarantees that there is no partition in the ESM-tree""" --- p.16 / Chapter 3.1.4 --- State Transition Diagram for Tree Formation Protocol --- p.16 / Chapter 3.2 --- ESM: Data Transfer --- p.28 / Chapter 3.3 --- ESM: Tree Optimization Protocol CONTENTS --- p.30 / Chapter 3.3.1 --- Example of Tree Optimization Protocol --- p.37 / Chapter 3.3.2 --- "The proof of ""Distributed Locking Protocol avoids loop formation and tree partition""" --- p.38 / Chapter 3.3.3 --- State Transition Diagram for Tree Optimization Protocol --- p.39 / Chapter 3.4 --- ESM: Node Leaving Protocol --- p.46 / Chapter 3.4.1 --- Example of ESM: Node Leaving Protocol --- p.51 / Chapter 3.4.2 --- State Transition Diagram for Node Leaving Protocol --- p.53 / Chapter 4 --- Performance Evaluation --- p.60 / Chapter 4.1 --- Experiment 1 - Comparisons between IP Unicast and ESM --- p.61 / Chapter 4.2 --- Experiment 2 - Comparisons between different ESM topologies --- p.64 / Chapter 4.3 --- Experiment 3 - Comparison between different thresholds for tree optimization operation --- p.67 / Chapter 4.4 --- Experiment 4 - NS2 Simulation --- p.69 / Chapter 5 --- Related Work --- p.74 / Chapter 6 --- Concluding Remarks --- p.78 / Chapter 6.1 --- Contributions --- p.78 / Chapter 6.2 --- Future Work --- p.79 / Chapter 6.2.1 --- Large-scale Experiments --- p.79 / Chapter 6.2.2 --- Evaluation for non-reliable data transfer --- p.79 / Chapter 6.2.3 --- Investigation of tree-optimization activation threshold --- p.80

Lee, Pak-Ching. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2003. / Includes bibliographical references (leaves 80-83). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Related Work --- p.5 / Chapter 3 --- Tree-Based Group Diffie-Hellman --- p.9 / Chapter 4 --- Interval-Based Distributed Rekeying Algorithms --- p.14 / Chapter 4.1 --- Rebuild Algorithm --- p.15 / Chapter 4.2 --- Batch Algorithm --- p.16 / Chapter 4.3 --- Queue-batch Algorithm --- p.19 / Chapter 5 --- Performance Evaluation --- p.22 / Chapter 5.1 --- Mathematical Analysis --- p.22 / Chapter 5.1.1 --- Analysis of the Rebuild Algorithm --- p.24 / Chapter 5.1.2 --- Analysis of the Batch Algorithm --- p.25 / Chapter 5.1.3 --- Analysis of the Queue-batch Algorithm --- p.30 / Chapter 5.2 --- Experiments --- p.31 / Chapter 5.3 --- Discussion of the experimental results --- p.35 / Chapter 6 --- Authenticated Tree-Based Group Diffie-Hellman --- p.43 / Chapter 6.1 --- Description of A-TGDH --- p.44 / Chapter 6.2 --- Security Analysis --- p.47 / Chapter 7 --- Implementation and Applications --- p.50 / Chapter 7.1 --- Leader and Sponsors --- p.51 / Chapter 7.1.1 --- Leader --- p.51 / Chapter 7.1.2 --- Sponsors --- p.53 / Chapter 7.1.3 --- Rekeying Operation --- p.56 / Chapter 7.2 --- System Architecture --- p.57 / Chapter 7.2.1 --- System Preliminaries --- p.57 / Chapter 7.2.2 --- System Components --- p.58 / Chapter 7.2.3 --- Implementation Considerations --- p.64 / Chapter 7.3 --- SGCL API --- p.65 / Chapter 7.4 --- Experiments --- p.67 / Chapter 7.5 --- Applications --- p.72 / Chapter 7.6 --- Future Extensions --- p.75 / Chapter 8 --- Conclusions and Future Directions --- p.76 / Chapter 8.1 --- Conclusions --- p.76 / Chapter 8.2 --- Future Directions --- p.77 / Chapter 8.2.1 --- Construction of a Hybrid Key Tree with the Physical and Logical Properties --- p.77 / Chapter 8.2.2 --- Extended Implementation --- p.79 / Bibliography --- p.80

by Lin Hon Man. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1998. / Includes bibliographical references (leaves 59-[63]). / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Organization of Thesis --- p.3 / Chapter 2 --- Principles of Multicast Cross-Path Switches --- p.4 / Chapter 2.1 --- Introduction --- p.4 / Chapter 2.2 --- Unicast Cross-Path switch --- p.5 / Chapter 2.2.1 --- Routing properties in Clos networks --- p.5 / Chapter 2.2.2 --- Quasi-static routing procedures --- p.5 / Chapter 2.2.3 --- Capacity and Route Assignment --- p.7 / Chapter 2.3 --- Multicast Cross-Path Switch --- p.8 / Chapter 2.3.1 --- Scheme 1 - Cell replication performed at both input and output stages --- p.10 / Chapter 2.3.2 --- Scheme 2 - Cell replication performed only at the input stage --- p.10 / Chapter 3 --- Architectures --- p.14 / Chapter 3.1 --- Introduction --- p.14 / Chapter 3.2 --- Input Module Design (Scheme 1) --- p.16 / Chapter 3.2.1 --- Input Header Translator --- p.16 / Chapter 3.2.2 --- Input Module Controller --- p.17 / Chapter 3.2.3 --- Input Replication Network (Scheme 1) --- p.19 / Chapter 3.2.4 --- Routing Network --- p.23 / Chapter 3.3 --- Central Modules --- p.24 / Chapter 3.4 --- Output Module Design (Scheme 1) --- p.24 / Chapter 3.5 --- Input Module Design (Scheme 2) --- p.25 / Chapter 3.5.1 --- Input Header Translator (Scheme 2) --- p.26 / Chapter 3.5.2 --- Input Module Controller (Scheme 2) --- p.27 / Chapter 3.5.3 --- Input Replication Network (Scheme 2) --- p.28 / Chapter 3.6 --- Output Module Design (Scheme 2) --- p.29 / Chapter 4 --- Performance Evaluations --- p.31 / Chapter 4.1 --- Introduction --- p.31 / Chapter 4.2 --- Traffic characteristics --- p.31 / Chapter 4.2.1 --- Fanout distribution --- p.31 / Chapter 4.2.2 --- Middle stage traffic load and its calculation --- p.32 / Chapter 4.3 --- Throughput Performance --- p.34 / Chapter 4.4 --- Delay Performance --- p.37 / Chapter 4.4.1 --- Input Stage Delay --- p.38 / Chapter 4.4.2 --- Output Stage Delay --- p.39 / Chapter 4.5 --- Cell Loss Performance --- p.43 / Chapter 4.5.1 --- Cell Loss due to Buffer Overflow --- p.44 / Chapter 4.5.2 --- Cell Loss Due to Output Contention --- p.45 / Chapter 4.6 --- Complexities --- p.50 / Chapter 5 --- Conclusions --- p.57 / Bibliography --- p.59

by Jin Mai. / Thesis submitted in: June 1998. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 62-[67]). / Abstract also in Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Background and Motivation --- p.1 / Chapter 1.2 --- Organization of the thesis --- p.8 / Chapter 2 --- Principles of WDM Cross-Path Switch --- p.11 / Chapter 2.1 --- Principles of path scheduling --- p.11 / Chapter 2.2 --- Call setup and path rearrangement --- p.15 / Chapter 2.3 --- ABR control --- p.17 / Chapter 3 --- Star coupler and WDM path scheduling --- p.20 / Chapter 3.1 --- Star coupler and other WDM ATM switches --- p.20 / Chapter 3.2 --- Two schemes of implementation --- p.22 / Chapter 4 --- input/output modules and local routing --- p.26 / Chapter 4.1 --- Shared buffer memory switch --- p.26 / Chapter 4.2 --- local routing at input/output modules --- p.29 / Chapter 5 --- Multicasting --- p.32 / Chapter 5.1 --- Two multicasting schemes --- p.32 / Chapter 5.2 --- Call blocking --- p.36 / Chapter 6 --- Performance --- p.37 / Chapter 6.1 --- Introduction --- p.37 / Chapter 6.2 --- Switch complexity --- p.38 / Chapter 6.3 --- Speed up --- p.40 / Chapter 6.4 --- Two multicasting schemes --- p.41 / Chapter 7 --- Switch Model and Operation --- p.47 / Chapter 8 --- Conclusions --- p.50 / Chapter A --- Effective bandwidth and QoS guarantee --- p.52 / Chapter A.l --- ATM service categories and QoS parameters --- p.52 / Chapter A.2 --- Effective bandwidth for single source --- p.53 / Chapter A.2.1 --- Markovian on/off source approach --- p.54 / Chapter A.2.2 --- Leaky bucket regulated source --- p.55 / Chapter A.3 --- Effective bandwidth for multiplexed sources --- p.60 / Chapter A.3.1 --- Gaussian model approach --- p.60 / Bibliography --- p.62

Zhang Jialiang. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (leaves 51-52). / Abstracts in English and Chinese. / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Related Work --- p.1 / Chapter 1.2 --- Organization of the Thesis --- p.3 / Chapter Chapter 2 --- Interference Modeling for Directional Antennas --- p.5 / Chapter 2.1 --- Pair-wise Physical Link Interference Model of Generic Directional Antenna --- p.6 / Chapter 2.2 --- Potential Interference Region --- p.8 / Chapter 2.3 --- Antenna Pattern and Phased Array Antenna --- p.9 / Chapter Chapter 3 --- Null Width of Directional Antennas --- p.12 / Chapter 3.1 --- Concept of Null Width --- p.12 / Chapter 3.2 --- Effective Null Width and Interference --- p.14 / Chapter 3.2.1 --- Probability of Interference --- p.14 / Chapter 3.2.2 --- Scenario of Directional Transmission and Omni-directional Reception --- p.15 / Chapter 3.2.3 --- Scenario of Directional Transmission and Directional Reception --- p.17 / Chapter 3.3 --- Properties of General Effective Beam Width --- p.18 / Chapter 3.4 --- Numerical Scaling Law of Effective Beam Width of Some Particular Antenna Patterns --- p.23 / Chapter 3.5 --- Summary --- p.26 / Chapter Chapter 4 --- Scaling Law of Network Capacity of Wireless Random Networks with Directional Antennas --- p.27 / Chapter 4.1 --- Random Network Model and Network Capacity --- p.27 / Chapter 4.2 --- Node distribution and MAC Protocol --- p.29 / Chapter 4.3 --- Scenario of Directional Transmission and Omni-directional Reception --- p.30 / Chapter 4.3.1 --- Probability of Transmission to be Success and Per-Link (Transport) Throughput --- p.30 / Chapter 4.3.2 --- Scaling Law of Network Capacity --- p.32 / Chapter 4.3.3 --- Concluding Remark --- p.37 / Chapter 4.4 --- Scenario of Directional Transmission and Directional Reception --- p.38 / Chapter 4.4.1 --- Antenna Steering Protocol --- p.39 / Chapter 4.4.2 --- Probability of Transmission to be Success --- p.40 / Chapter 4.4.3 --- Scaling Law of Network Capacity --- p.41 / Chapter 4.4.4 --- Scaling Law of Phased Array Antennas --- p.42 / Chapter Chapter 5 --- Conclusion --- p.44 / Appendix A: Proof of equation (22) --- p.47 / Appendix B: Proof of equation (28) --- p.49 / Appendix C: Constraint on Region of Optimality for pt and r --- p.50 / References --- p.51

Lau Ka Ming. / Thesis submitted in: August 2005. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (leaves 69-72). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgements --- p.iii / Table of Contents --- p.iv / List of Figures --- p.vi / List of Tables --- p.viii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Introduction --- p.1 / Chapter 1.2 --- Wireless Ad Hoc Networks --- p.2 / Chapter 1.2.1 --- Wireless Sensor Network --- p.3 / Chapter 1.2.2 --- IEEE802.11 Ad Hoc Network --- p.3 / Chapter 1.2.3 --- Bluetooth Personal Area Network --- p.4 / Chapter 1.3 --- Power Saving in Wireless Ad Hoc Networks --- p.4 / Chapter 1.4 --- Contributions of the Thesis --- p.8 / Chapter 1.5 --- Outline of the Thesis --- p.9 / Chapter 2 --- Power Saving Mechanisms in Wireless Ad hoc Networks --- p.10 / Chapter 2.1 --- Recent Research Proposals --- p.10 / Chapter 2.1.1 --- Synchronous Power Saving Schemes --- p.11 / Chapter 2.1.2 --- Asynchronous Power Saving Schemes --- p.12 / Chapter 2.2 --- Existing Standards --- p.17 / Chapter 2.2.1 --- IEEE802.1l Ad Hoc Power Saving Mode --- p.17 / Chapter 2.2.2 --- Bluetooth Low Power Modes --- p.20 / Chapter 3 --- Analytical Framework for Designing Synchronous Wakeup Patterns --- p.22 / Chapter 3.1 --- System Model --- p.23 / Chapter 3.1.1 --- Vacation Model --- p.23 / Chapter 3.1.2 --- Optimal Wakeup Pattern --- p.25 / Chapter 3.2 --- Analytical analysis of different wakeup patterns --- p.27 / Chapter 3.2.1 --- Exhaustive Wakeup Pattern --- p.27 / Chapter 3.2.2 --- Gated Wakeup Pattern --- p.31 / Chapter 3.2.3 --- Gated Wakeup With Constant Cycle Time --- p.34 / Chapter 3.3 --- Discussion of results --- p.43 / Chapter 3.3.1 --- Performances impacts of various system parameters --- p.43 / Chapter 3.3.2 --- Performances comparison of different wakeup patterns --- p.47 / Chapter 3.4 --- Chapter Summary --- p.48 / Chapter 4 --- An improved IEEE802.1l Power Saving Mode --- p.49 / Chapter 4.1 --- Related Proposals --- p.50 / Chapter 4.2 --- Proposed Scheme --- p.52 / Chapter 4.2.1 --- Overview --- p.52 / Chapter 4.2.2 --- Beacon Sending Station --- p.53 / Chapter 4.2.3 --- Beacon Receiving Station --- p.55 / Chapter 4.2.4 --- Computing the Transmission Schedule --- p.55 / Chapter 4.2.5 --- Data Transmission Specifications --- p.56 / Chapter 4.2.6 --- Failure Conditions --- p.58 / Chapter 4.3 --- Performances Evaluation --- p.58 / Chapter 4.3.1 --- Simulation Model --- p.60 / Chapter 4.3.2 --- Simulation Results --- p.60 / Chapter 4.4 --- Chapter Summary --- p.64 / Chapter 5 --- Conclusion --- p.66

這篇論文討論了三個重要的全光網絡的網絡管理方向，分別為在全光封包交換網絡的光路追踪問題，光封包的暫存及在波分複用無源光網絡（WDM-PON）的能源節省運作。 / 使用全光編碼器進行光路追踪 / 在所有全光封包交換網絡中，數據包的路由光路是可以被通過光學交叉連接（OXC）的網絡節點重新配置。要監控任何可能的路由錯誤、信號質量下降的任何可能的原因、發現任何惡意或攻擊流量的來源、或提供服務質量（QoS）的路由策略，光學數據包的光路監測是必要的。此方案編碼光路信息到光學數據的標籤，每個網絡節點被分配一個獨特的質數為識別標記的光路信息。使用位於OXC輸出端的光學編碼器，每次當數據包通過該網絡節點時，數據包的標籤值將乘以特定的質數。網絡數據包已經走過的節點將被編碼的標籤為代表的網絡節點的所有素數的乘積。因此，在標籤的檢示器中，光數據包走過的每個網絡節點都可以很容易地通過標籤值的因式分解求得。 / 相比使用時間延遲識別技術，我們的設計大幅減少了光纖線的長度要求。此外，與導頻訊號檢測方法相比，它提供了快速檢測的好處。此外，網絡循環的問題也可以檢測。編碼標籤也可以作為的光封包的保存週期指示器（TTL）。 / 功率控制的全光封包緩衝器 / 全光封包緩衝區全光交換網絡中是非常重要的。為了實現簡單而有效的光學數據包緩衝操作，我們提出使用功率控制時間延遲的全光數據包緩衝區。的循環迴路的數量是由輸入信號的功率控制（即具有較高的功率輸入信號將經歷更長的時間延遲）。我們把偱環延遲緩衝器重新設計成信號功率依賴過濾的問題。光信號首先通過在不同的時刻，每延遲副本將減半功率與前一個通過循環生成多個副本的循環。然後，它會通過信號的功率大子而使用非線性光學效應來實施過濾。過濾器只與特定功率水平的信號，可以輸出，而別人得到減毒，因此清除的特點。因此，為了改變延遲量，我們只需要改變輸入信號功率等具體延遲的信號，將陷入電源依賴的濾波器的通帶和輸出。 / 比起使用許多的SOA來控制循環延遲或使用可調諧波長轉換器及波長依賴時滯的方案，我們的計劃提供更容易的延遲控制。 / 信號電源效率的WDM-PON的操作技巧 / 在WDM-PON中，光網絡單元（ONU）的上傳信號通過對由光線路終端（OLT）的下傳信號的重新編寫產生。傳統上，這設計產生功耗的問題。例如：如果沒有下傳信號，光網絡單元不能發送上傳數據。因此，即使沒有數據傳送，光線路終端也要不斷傳送下傳信號，為了確保光網絡單元總是能夠發送上傳數據。在這種網絡中，突發傳送模式可以提供電源效率上的節省。我們提出了一個信號傳送技術來由光網絡單元向光線路終端發送“喚醒消息，以通知光線路終端的收發器從睡眠模式恢復。這技術是通過對在RSOA放大自發輻射（ASE）噪聲進行特定導頻信號的，因而不需要重新調制下游信號。在光線路終端，額外的模塊將負責檢測不同光網絡單元的導頻信號，然後啟動相應的收發器。我們的計劃提供了在WDM-PON的簡單和高成本效益的能源節省方案。 / This thesis addresses three important network management aspects of optical networks, namely, optical lightpath tracing problem for all-optical networks, all-optical packet buffering in optical packet switching networks, power-efficient operation in wavelength division multiplexing passive optical networks (WDM-PON). / Lightpath tracing through all-optical encoder / In an all-optical reconfigurable wavelength routing network, the lightpath of the optical data packets can be reconfigured, via the optical cross connects (OXC) residing at each network node. In order to monitor any possible routing errors, any possible causes of signal quality degradation, detect any source of malicious or attack traffic and provide quality of service (QoS) aware next-hop routing strategy, monitoring of lightpath of optical data packets is necessary. This scheme encodes the lightpath information into the label of individual optical packet. Each network node is assigned with a distinct prime number as an identification tag. By using the optical encoders located at the outputs of OXCs, the label value of the packets will be multiplied by the prime number designated to the respective OXC, residing at the network node. Hence, the information of the network nodes that the packet has traversed will be encoded to the label as the product of all the prime numbers assigned to all traversed network nodes. Therefore, at the destination node, the whole physical lightpath of each received optical data packet can be easily identified through factorization of the encoded label value. / Our scheme provides substantial reduction in the requirement of fiber delay lines, as compared to the time-delay recognition techniques. It offers fast detection when comparing with pilot tone detection method. Besides, possible network looping problem can be detected and the encoded label can be acted as time-to-live (TTL) identifier of the optical packet. / All-optical power-controlled optical packet buffer / All-optical packet buffer is essential for the contention resolution in all-optical packet switching network. In order to realize simple and efficient operation of optical packet buffer, we propose the use of all-optical power-controlled packet buffer for which the number of circulating loops is controlled by the input signal power (i.e. input signal with higher power will experience longer delay). We formulate the problem of designing a re-circulating delay buffer into signal power dependent filtering problem. The optical OOK signal first passes through a re-circulating loop generating multiple copies at different time instants that each delayed copy will have halved power as the previous one. Then it will pass through the signal power dependent filter implemented by using optical nonlinear effect. The filter has the characteristics that only signal (packet copy) with specific power level can be outputted while the others get attenuated and therefore cleared. As a result, in order to change the amount of delay, we just need to change the input signal power such that the signal with specific delay will fall into the pass band of the power dependent filter and get outputted. / Compared with other delay schemes which use many SOAs as gates to control the number of re-circulating delay or implementing tunable wavelength converter and passes through wavelength dependent delay, our scheme provides easy control of the delay required. / Signaling techniques for power-efficient operation of WDM-PON / In WDM-PON, the upstream signal at the optical network unit (ONU) can be generated by re-modulating the downstream signal received from the optical line terminal (OLT). However, the conventional architecture may suffer from power consumption problem. When there is no downstream signal, the ONU is not able to send any upstream data. Thus even if there is no traffic on the line, the OLT has to send the downstream signal continuously, in order to ensure the ONU can always be able to send its upstream data. In such networks, burst-mode traffic transmission can provide improvement on power efficiency. We propose a signaling technique to send “Wake Up message from the ONU to OLT, to notify the transceiver of the OLT to recover from sleep mode. It is done by modulating the Amplified Spontaneous Emission (ASE) noise in RSOA at the ONU with the particular ONU specific pilot tone monitoring signal. Thus, it does not require the presence of the remodulating downstream signal. At the OLT, a specific module is needed for the detection of the pilot tones from different ONUs and then activate the corresponding transceiver. Our scheme offers simple and cost-effective approach for power-efficient operation in the WDM-PON. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Tse, Kam Hon. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 90-97). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / Abstract --- p.i / Acknowledgements --- p.viii / Chapter Chapter 1 --- Background --- p.1 / Chapter 1.1 --- All-Optical Packet Switching Networks --- p.3 / Chapter 1.1.1 --- Packet Buffering --- p.4 / Chapter 1.1.2 --- Route Tracing --- p.7 / Chapter 1.2 --- Wavelength Division Multiplexing Passive Optical Networks (WDM-PON) --- p.12 / Chapter 1.2.1 --- ONU Upstream Re-Modulation in WDM-PON --- p.12 / Chapter 1.2.2 --- Green Networking --- p.13 / Chapter 1.3 --- Contributions --- p.17 / Chapter 1.3.1 --- Path Tracing Scheme for All-Optical Packet-Switched Networks --- p.17 / Chapter 1.3.2 --- All-Optical Power-Controlled Optical Packet Buffer for All-Optical Packet-Switched Networks --- p.18 / Chapter 1.3.3 --- Signaling Techniques for Power-Efficient Operation of WDM-PON --- p.19 / Chapter 1.4 --- Organization of Thesis --- p.19 / Chapter Chapter 2 --- Path Tracing Scheme for All-Optical Packet-Switched Networks --- p.21 / Chapter 2.1 --- Introduction --- p.21 / Chapter 2.2 --- Path Tracing Using Prime-Number Tags --- p.26 / Chapter 2.2.1 --- Network Node Tracing --- p.26 / Chapter 2.2.2 --- Network Link Tracing --- p.28 / Chapter 2.2.3 --- Network Looping Problem --- p.29 / Chapter 2.3 --- Optical Implementation --- p.30 / Chapter 2.4 --- Experimental Results --- p.38 / Chapter 2.5 --- Multi-Level Amplitudes Of The Label --- p.41 / Chapter 2.6 --- Required Maximum Fiber Delay --- p.44 / Chapter 2.7 --- Summary --- p.47 / Chapter Chapter 3 --- A Novel Fiber-based Variable All-Optical Packet Buffer based on Self-Phase Modulation Induced Spectral Broadening --- p.48 / Chapter 3.1 --- Introduction --- p.48 / Chapter 3.2 --- Input signal power dependent delay --- p.51 / Chapter 3.3 --- Numerical Simulation Studies --- p.54 / Chapter 3.4 --- Experimental Results --- p.64 / Chapter 3.5 --- Discussion --- p.68 / Chapter 3.6 --- Summary --- p.71 / Chapter Chapter 4 --- A Cost-effective Pilot-Tone-based Monitoring Technique for Power Saving in RSOA-based WDM-PON --- p.73 / Chapter 4.1 --- Introduction --- p.73 / Chapter 4.2 --- Proposed System Architecture --- p.75 / Chapter 4.3 --- Experimental Setup and Result --- p.78 / Chapter 4.4 --- Power Saving Efficiency Calculation --- p.82 / Chapter 4.5 --- Summary --- p.86 / Chapter Chapter 5 --- Conclusions and Future Work --- p.87 / Chapter 5.1 --- Conclusions --- p.87 / Chapter 5.2 --- Future Work --- p.88 / Bibliography --- p.90 / Publications during PhD Study --- p.98