by Xiaoqing Wang. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2000. / Includes bibliographical references (leaves 43-45). / Abstracts in English and Chinese. / Acknowledgments --- p.ii / Abstract --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- The General Architecture of the VoD System and the Related Issues --- p.4 / Chapter 2.1 --- A Brief Description of VoD System --- p.4 / Chapter 2.2 --- Why Video Streams in VoD Service are VBR in Nature? --- p.6 / Chapter 2.3 --- The Video Storage Media in the VoD Systems --- p.8 / Chapter 2.4 --- The Data Placement Scheme in the VoD System --- p.9 / Chapter 2.5 --- An Overview of Disk Scheduling in VoD System --- p.10 / Chapter 2.6 --- The Admission Control in VoD System --- p.12 / Chapter 3 --- Our Admission Control Algorithm for VoD System --- p.14 / Chapter 3.1 --- QoS Requirements We Choose --- p.14 / Chapter 3.2 --- System Model --- p.15 / Chapter 3.3 --- The Admission Control for the Storage Sub-system --- p.19 / Chapter 3.4 --- The Admission Control for Network Sub-system --- p.21 / Chapter 3.4.1 --- Preliminaries --- p.22 / Chapter 3.4.2 --- The Admission Control Algorithm for Network Sub-system --- p.27 / Chapter 4 --- Experiment --- p.33 / Chapter 5 --- Conclusion and Future Work --- p.41 / Chapter 5.1 --- Conclusion --- p.41 / Chapter 5.2 --- Future Work --- p.42

隋著移動數據網絡的迅速發展，世界各地部署著不同制式的3G，HSPA和LTE網絡，這給移動互聯網應用帶來新的機遇和挑戰。不像他們的對手, 有線網絡，移動數據網絡有較長的延遲，較高的包率，以及急劇波動的頻寬。為了應付這些挑戰，移動無線基站往往配備較大的緩衝區（KBs以至數百多MBs），以吸收短期的頻寬波動，並促進鏈路層重傳。然而，由於大部分互聯網協議，並特別是TCP，全都假設路由器只擁有較小緩衝區，所以在移動網絡的大型緩衝區下，往往導致TCP展現出次優的性能。本論文解決兩個根本性的問題來克服這一項挑戰。首先，我們開發了一種新算法 (SoD）專門來估計網絡的鏈路緩衝區的大小和隊列長度。在今天的3G移動數據網絡，本文以大量網絡數據驅動的模擬結果證明 SoD 表現優於現有的算法，如 Max-min 和 Loss-pair。另外，我們採用 SoD算法來創造新的TCP擁塞控制模塊以解決移動網絡的大型緩衝區以及的頻寬波動問題。這個新的TCP變種，稱為 TCP- QA，在模擬情況以及移動數據網絡下均表現出大大優於現有的TCP變種，包括TCP CUBIC ，TCP Vegas，TCP Westwood，和FAST TCP。 / The rapidly emerging mobile data networks fueled by the world-wide deployment of 3G, HSPA, and LTE networks created new opportunities and challenges for developing mobile Internet applications. Unlike their wired counterpart, mobile data networks are known to exhibit longer delay, higher packet loss rate, and rapidly fluctuating bandwidth. To tackle these challenges mobile radio base stations are often equipped with large buffers (from hundreds of KBs to multi-MBs) to absorb short-term bandwidth fluctuations and to facilitate link-layer retransmissions. However as most Internet protocols in general, and TCP in particular, were designed with the assumption of small router buffer size, the large buffer in mobile networks can and do interact, often negatively, with the Internet protocols, leading to sub-optimal performance. This work tackles two fundamental problems in overcoming this challenge. First, we developed a novel Sum-of-Delays (SoD) algorithm specifically designed to estimate the link buffer size and queue length of bandwidth-varying networks. Extensive trace-driven simulation results showed that SoD outperforms existing algorithms such as max-min and loss-pair by orders of magnitude in today’s 3G mobile data networks. Second, we apply the SoD algorithm to TCP’s congestion control module to incorporate and compensate for the mobile network’s large buffer size. This new TCP variant, called TCP-Queue-length-Adaptive (TCP-QA), substantially outperforms existing TCP variants including TCP CUBIC, TCP Vegas, TCP Westwood, and FAST TCP. / Detailed summary in vernacular field only. / Chan, Chi Fung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references. / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts also in Chinese. / Abstract --- p.ii / 摘要 --- p.iii / Acknowledgements --- p.iv / Contents --- p.v / Chapter Chapter 1 --- Introduction --- p.1 / Chapter Chapter 2 --- Background and Related Work --- p.5 / Chapter 2.1 --- Link Buffer Size Estimation --- p.6 / Chapter 2.2 --- Queue Length Estimation --- p.8 / Chapter 2.3 --- Performance over Fixed Networks --- p.10 / Chapter 2.4 --- Performance over Bandwidth-Varying Networks --- p.15 / Chapter Chapter 3 --- Sum-Of-Delay Method --- p.19 / Chapter Chapter 4 --- Performance Evaluation --- p.22 / Chapter 4.1 --- Link Buffer Size Estimation in Fixed Bandwidth Networks --- p.23 / Chapter 4.2 --- Link Buffer Size Estimation in Variable Bandwidth Networks --- p.25 / Chapter 4.3 --- Queue Length Estimation in Fixed Bandwidth Networks --- p.28 / Chapter 4.4 --- Queue Length Estimation in Variable Bandwidth Networks --- p.31 / Chapter 4.5 --- Convergence --- p.34 / Chapter 4.6 --- Limitations in Passive Estimation --- p.40 / Chapter Chapter 5 --- Impact of Competing Traffic --- p.42 / Chapter 5.1 --- Simulation Setup --- p.42 / Chapter 5.2 --- Link Buffer Size Estimation --- p.44 / Chapter 5.3 --- Queue Length Estimation --- p.50 / Chapter 5.4 --- Discussions --- p.54 / Chapter Chapter 6 --- TCP-Queue-Length-Adaptive (TCP-QA) --- p.56 / Chapter 6.1 --- Challenges in Mobile Networks --- p.57 / Chapter 6.2 --- Queue-Length-Adaptive Congestion Control --- p.59 / Chapter 6.3 --- Performance Comparisons --- p.65 / Chapter Chapter 7 --- Summary and Future Work --- p.76 / Bibliography --- p.78

Orientador: Gladys Dorotea Cacsire Barriga / Banca: Andreia da Silva Meyer / Banca: Glauco Henrique de Sousa Mendes / Resumo: A falta de qualidade no setor de serviços exige das empresas técnicas de controle mais efetivas e métodos de avaliação da qualidade menos genéricos, baseados nas dimensões da qualidade de cada setor. Em particular, os serviços de locação de automóveis se apresentam como um mercado promissor e competitivo, tanto no cenário nacional quanto no mundial. Dessa forma, o objetivo dessa pesquisa é analisar quais dimensões da qualidade de serviço no setor de locação de automóveis são determinantes para satisfação do consumidor brasileiro. O método de pesquisa quantitativo do tipo survey foi aplicado utilizando como instrumento de coleta de dados o questionário, elaborado com base na revisão de literatura. 196 respostas foram obtidas permitindo a aplicação das análises estatísticas: descritiva, fatorial exploratória e regressão logística. Os resultados mostraram que apesar da grande quantidade de dimensões encontradas na literatura, as dimensões comunicação com o cliente, segurança, aspectos de importância da empresa e apresentação foram as que mais influenciaram a qualidade do serviço de locação de automóveis no Brasil, na visão dos clientes. Porém, outras dimensões se mostraram relevantes como confiabilidade, acessibilidade, entrega e conforto. Portanto, as implicações dessa pesquisa para a literatura sugerem um sistema de mensuração de qualidade centralizado nessas variáveis, disponibilizando um instrumento de melhoria da qualidade dos processos de serviços de locação de automóveis.... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Orientador: Hérida Regina Nunes Salgado / Banca: Marlus Chorilli / Banca: Martin Steppe / Resumo: O ácido fusídico é um antimicrobiano esteroide, indicado no tratamento de infecções graves causadas por micro-organismos do gênero Staphylococcus e de outras espécies de bactérias Gram-positivas, incluindo espécies resistentes à penicilina e a outros antimicrobianos. Pesquisas envolvendo o desenvolvimento de métodos analíticos mais práticos são relevantes para otimizar as análises de rotina na indústria farmacêutica e garantir a qualidade do produto comercializado. Desta forma, este trabalho teve como principal objetivo o desenvolvimento de métodos analíticos mais rápidos, seguros para os operadores, econômicos e de fácil execução, em comparação aos métodos descritos na literatura para a determinação do ácido fusídico na forma farmacêutica de creme dermatológico. O ácido fusídico foi analisado quanto às suas características organolépticas e físico-químicas. Também foi realizada análise qualitativa pelo método de espectrofotometria de absorção na região do visível e por cromatografia em camada delgada. Para quantificação do ácido fusídico, foram desenvolvidos e validados dois métodos: método por cromatografia líquida de alta eficiência indicadivo de estabilidade, na faixa de concentração de 5,0 a 95,0 μg/mL, utilizando fase móvel composta por acetonitrila e água (72:28, v/v), pH 3,5 ajustado com ácido acético e comprimento de onda de 210 nm, este método apresentou tempo de retenção médio de 8,12 minutos, exatidão de 101,22% e teor de 104%; ensaio microbiológico pelo método turbidimétrico, na faixa de concentração de 0,25 a 2,25 μg/mL, utilizando caldo casoy e 6% do micro-organismo Staphylococcus epidermidis ATCC 12228, cuja exatidão e percentual médio foram de 101,95% e 104,60%, respectivamente. Os métodos foram validados de acordo com as diretrizes do ICH e da legislação brasileira e mostraram ser precisos, robustos,... / Abstract: Fusidic acid is an antimicrobial steroid, indicated in the treatment of serious infections caused by microorganisms of the genus Staphylococcus and other Gram-positive bacteria species, including resistant species to penicillin and other antimicrobials. Research involving the development of more practical analytical methods are relevant to optimize the analysis in the pharmaceutical industry and secure the quality of commercialized product. Thus, this study aimed to develop faster analytical methods, safer for operators, economical and easy to perform, compared to the methods described in the literature for the determination of fusidic acid in the pharmaceutical form of dermatological cream. Fusidic acid was analyzed for its organoleptic and physical-chemical characteristics. Qualitative analysis was performed method by spectrophotometric absorption in the visible region and by thin layer chromatography. For quantification of fusidic acid, were developed and validated two methods: stability indicative method by high performance liquid chromatography, in the concentration range from 5.0 to 95.0 μg mL-1, using mobile phase composed by acetonitrile and water (72:28, v/v), pH 3.5 adjusted with acetic acid and at a wavelength at 210 nm, this method presented average retention time of 8.12 minutes, accuracy of 101.22% and content of 104%; microbiological assay by turbidimetric method, in the concentration range from 0.25 to 2.25 μg mL-1, using casoy broth and 6% of the microorganism Staphylococcus epidermidis ATCC 12228, whose accuracy and mean percentage were 101.95% and 104.60%, respectively. The methods were validated according to ICH guidelines and Brazilian legislation and shown to be accurate, robust, selective, accurate and equivalent to quantify the fusidic acid, at a significance level of 5%. Thus, both methods can be employed in the analysis of fusidic acid in pharmaceutical form of dermatological cream in the pharmaceutical ... / Mestre

Digitized by Kansas Correctional Industries

Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management, 2004. / Includes bibliographical references (leaf 68). / In the early 1980's and 1990's, companies began to build upon the principles of Total Quality Management and developed there own unique quality systems. The most popular and well known of these systems is Six Sigma that was developed by Motorola and successfully adopted by others such as Allied Signal (now Honeywell) and most notably, General Electric. Six Sigma can be characterized as a highly formalized, process oriented improvement tool that is data focused. The Six Sigma process is normally performed by a diverse team, who attack a quality/process problem by analyzing process variation or in statistical terms, sigma. The foundations of Six Sigma are commitment from upper management, detailed training and a regimented diagnostic approach. Another quality operating system is the less known, but very successful, Achieving Competitive Excellence (ACE) operating system. This system was developed and is practiced by United Technologies Corporation (UTC). The ACE system is broader based than the Six-Sigma approach, however, ACE is not as data oriented as the Six Sigma approach. ACE revolves around the three principle categories of process improvement and waste elimination tools, decision-making tools, and problem solving tools. These tools impact issues as diverse, but not limited to, factory floor cleanliness, market feedback analysis, machine tool preventative maintenance and set up reduction. ACE is a combination of lean manufacturing and quality improvement philosophies. This paper provides an analysis of both the Six Sigma and ACE Quality Operating Systems. In the paper the systems are compared and contrasted. Further, strengths and weaknesses of each system are discussed. In particular, the analysis focuses on how ACE can leverage elements / (cont.) and aspects of Six Sigma. The analysis concludes that there are elements of Six Sigma that would benefit ACE. The paper identifies that the strength of Six Sigma's statistical approach and its positive impact on process certification could be beneficially applied to the ACE system. Further, there are recommendations for UTC to place more of an emphasis on ACE training and to accelerate its current efforts to better link quality and lean improvement to product engineering and design. / by Thomas C. Hutton. / M.B.A.

Digitized by Kansas Correctional Industries

O presente trabalho foi motivado pelas condições de comercialização da sardinha verdadeira (Sardinella brasiliensis) fresca, descongelada e processada no município de São Paulo-SP, aparentemente não adequadas. Amostras de sardinhas foram analisadas usando como parâmetros os valores de substâncias que reagem com o ácido tiobarbitúrico (TBARS) e nitrogênio de bases voláteis totais (N-BVT). As amostras foram coletadas e analisadas em três momentos diferentes da cadeia produtiva: ao desembarque na CEAGESP (Companhia de Entrepostos e Armazens Gerais do Estado de São Paulo), durante a comercialização em feiras-livres antes da vigência do defeso da espécie e durante a comercialização em feiras-livres na vigência do defeso da espécie (dezembro a março). Para a sardinha desembarcada na CEAGESP, foram observados os seguintes valores médios ± desvio-padrão: TBARS= 0,18 ± 0,17 mg de aldeído malônico (AM)/ kg e N-BVT= 15,75 ± 2,39 mg/ 100 g. Para a sardinha comercializada em feira-livre antes do defeso, os valores registrados foram TBARS= 0,82 ±0,63 mg AM/ kg e N-BVT= 27,06 ± 2,18 mg/ 100g. Para a sardinha comercializada em feira-livre durante o defeso foram detectados TBARS= 7,14 ± 5,36 mg AM/ kg e N-BVT= 27,69 ± 2,80 mg/ 100 g . Os resultados mostraram haver, para os valores de TBARS, diferença estatisticamente significativa entre a sardinha desembarcada na CEAGESP e a comercializada em feira-livre antes do defeso, bem como entre estas e as comercializadas na vigência do defeso. Já para os valores de N-BVT foi notada diferença estatisticamente significativa entre a sardinha desembarcada na CEAGESP e as comercializadas em feira-livre, porém não houve diferença estatisticamente significativa entre as sardinhas comercializadas em feira-livre antes e durante a vigência do defeso. Na sardinha salmourada foram verificados valores médios de TBARS= 4,07 ± 1,22 mg AM/ kg e N-BVT= 44,32 ± 14,38 mg/ 100 g quando não-lavada; e TBARS= 1,25 ± 0,23 mg AM/ kg e N-BVT= 39,63 ± 4,00 mg/ 100 g quando lavada. Para a sardinha anchovada foram detectados os valores médios de TBARS= 3,71 ± 0,77 mg AM/ kg e N-BVT= 62,96 ± 9,33 mg/ 100 g. Foram ainda comparados valores de TBARS e lisina biodisponível, não se observando correlação significativa entre eles (R2= 0,1732). Apenas a sardinha fresca comercializada na CEAGESP apresentou condição aceitável de consumo. / This work was motivated by the observation of the apparently not adequate trade conditions of the fresh, defrosted and processed Brazilian sardine (Sardinella brasiliensis) in the city of São Paulo-SP. Sardine samples were analysed for thiobarbituric acid reacting substances (TBARS) and total volatile base nitrogen (TVB-N). The samples were collected and analysed at three different times in the production chain: on landing at CEAGESP (the Central Warehouse Company of the State of São Paulo), while on sale at the street markets before the species-catching prohibition and while on sale at the street markets during the species-catching prohibition period (December-March). The following results were observed for the sardine landed at CEAGESP (mean ± standard deviation): TBARS= 0.18 ± 0.17 mg malondialdehyde (MDA)/ kg and TVB-N= 15.75 ± 2.39 mg/ 100 g. For the sardines collected at the street markets before the prohibition period, the results were TBARS= 0.82 ± 0.63 mg MDA/ kg and TVB-N= 27.06 ± 2.18 mg/ 100 g. The values detected for the samples collected at the street markets during the prohibition period were TBARS= 7.14 ± 5.36 mg MDA/ kg and TVB-N= 27.69 ± 2.80 mg/ 100 g. The results show a statistically significant difference between the TBARS values for the samples collected at CEAGESP and the ones collected at the street markets before the prohibition period and between the latter and the ones collected at the street markets during the prohibition period. On the other hand, a statistically significant difference was noticed between the TVB-N values for the sardines collected at CEAGESP and the ones collected at the street markets, but there was not any statistically significant difference between the samples collected at the street markets before and during the prohibition period. In unwashed brined sardines the following values were recorded: TBARS= 4.07 ± 1.22 mg MDA/ kg and TVB-N= 44.32 ± 14.38 mg/ 100 g, while for the washed samples the results were TBARS= 1.25 ± 0.23 mg MDA/ kg and TVB-N= 39.63 ± 4.00 mg/ 100 g. For ripened sardine values of TBARS= 3.71 ± 0.77 mg MDA/ kg and TVB-N= 62.96 ± 9.33 mg/ 100 g were detected. TBARS values were also compared to available lysine and no significant correlation was found (R2= 0.1732). Only the fresh sardines landed at CEAGESP showed acceptable condition for consumption.

如今，隨著信息爆炸，骨幹網絡和城域網絡的容量需求已成倍增加。因此，如何提高網絡容量正成為學術界和工業界的熱門話題。可變帶寬光網絡技術通過為不同速率的數據傳輸分配剛剛足夠的帶寬來提高網絡容量，而物理層網絡編碼技術（PNC）在沒有復雜的硬件要求下可以增加網絡容量。在這篇論文中，我們首先提出將組播應用於可變帶寬光網絡來提高網絡容量。我們進一步提出將物理層網絡編碼技術應用於時分複用光接入網絡（TDM-PON），從而來提高全光虛擬專用通信（VPN）的網絡容量。 / 可變帶寬光網絡中組播的分析 / 可變帶寬光網絡相比傳統的波分複用光網絡（WDM）可以提高骨幹網絡的頻譜利用率，因為它可以靈活地分配剛剛足夠的帶寬。另一方面，光網絡層上的組播是一種高效的支持點對多點的通信技術。在未來的許多寬帶服務中，點對多點應用服務是必不可少的，通過光組播技術可以節省頻譜帶寬和接發器的數目。為了進一步提高網絡容量，我們建議在可變帶寬光網絡中進行組播。雖然關於可變帶寬光網絡的研究已經有很多了，但據我們所知，關於可變帶寬光網絡的組播尚未被研究。我們通過兩種有效算法來解決可變帶寬光網絡組播的路由和頻譜分配問題。採用相同的路由和波長/頻譜分配算法，我們研究了有靈活帶寬分配產生的好處，通過比較可變帶寬光網絡和傳統波分複用網絡的組播。我們也探討了由非均勻帶寬分配造成的頻譜間隙對提高網絡容量的影響。 / 時分複用光接入網中(TDM-PON)的物理層網絡編碼技術（PNC） / 網絡編碼是一種很有前途的技術，可以提高網絡的容量和健全性。雖然最近有關於在時分複用光接入網中進行網絡編碼的研究，應用於同一個光接入網絡中的光網絡單元（ONU）之間的通信，但在這些研究中的最大的網絡容量提高只有33。此外，在光網路終端（OLT）和光網絡單元中還需要大量的緩衝來存儲VPN數據。在時分複用光接入網中，全光VPN網絡可以重新將VPN數據傳送到相應的ONU，實現ONU之間的直接通信，不需要在OLT進行光-電-光的轉換。在這裡，據我們所知，我們第一次用實驗驗證了一種新方案，將物理層網絡編碼技術應用於TDM-PON，使得全光VPN通信的網絡容量增加了一倍。我們也提出了在光接入網中的遠程節點處使用光環路器，以此減少VPN通信的插入損耗。當兩個ONU之間需要進行雙向通信，可以通過利用PNC來實現全雙工傳輸，相比傳統半雙工的全光VPN方案，網絡容量可以提高100。實驗結果表明，可以實現無差錯全雙工VPN通信，相比半雙工通信功率補償不超過3分貝，而且這方案中ONU間的同步是不需要的。 / Nowadays, with the information explosion, the capacity demand has been exponentially increasing in backbone networks and metro networks. Therefore, it is becoming a hot topic for both academic and industry to improve the network capacity. Elastic technologies are promising to scale up the network capacity due to just-enough bandwidth allocation for different data-rate traffic request, while physical-layer network coding (PNC) can increase the throughput without complex requirement on hardware. In this thesis, we first propose a novel scheme to improve the network capacity by implementing multicast in elastic optical networks. We further present the capacity improvement by integrating PNC in time-division multiplexing passive optical network (TDM-PON) for all-optical virtual private network (VPN) communications. / Analysis of multicast in elastic optical networks / Elastic optical networks can increase the spectrum utilization of backbone networks compared to the traditional wavelength-division multiplexing (WDM) networks due to flexible and just-enough bandwidth allocation. On the other hand, multicast over the optical layer is a bandwidth-efficient communication technique which supports point-to-multipoint applications. As many broadband services in the future can be from one source to several destinations, it is essential to enable optical multicast to save bandwidth as well as transceivers. To further improve the network throughput, we propose to implement multicast in spectrum elastic optical networks. Although many investigations on elastic optical networks have been carried out, to the best of our knowledge, the performance of multicast in elastic optical networks have not yet been studied. We develop two efficient multicast heuristics to solve the multicast routing and spectrum allocation (MC-RSA) problem in elastic optical networks. By adopting the same routing and wavelength/spectrum allocation algorithms, the benefits of elastic optical networks resulting from flexible bandwidth allocation are studied for multicast compared to the traditional WDM networks. We also investigate the impact of spectral gap caused by non-uniform bandwidth allocation on the improvement of network throughput. / Physical-layer network coding (PNC) in TDM-PON / Network coding is a promising technique to improve the network throughput and robustness. Although network coding in TDM-PON has been recently investigated for exchanging information among optical network units (ONUs) in the same PON, the maximum capacity improvement of inter-ONU communications in these schemes is only 33%. In addition, large electrical buffer is required to store the VPN traffic at both optical line terminal (OLT) and ONUs. All-optical VPN in TDM-PON can optically reroute VPN traffic to the destined ONU without optical-electrical-optical conversion at OLT, which enables direct communications among ONUs. Here, to the best of our knowledge, for the first time, we experimentally demonstrate a novel PNC scheme integrated in TDM-PON for all-optical VPN communications to double the network throughput. A unique remote node that uses optical circulators to reduce the insertion loss of VPN communications is also proposed. By transmitting two inter-ONU traffic streams of opposite direction simultaneously using PNC (full-duplex), it can improve the network throughput by 100% compared to the traditional all-optical VPN schemes (half-duplex). Experiments show that error-free full-duplex VPN communications are achieved, and the power penalty is no more than 3 dB. Synchronization of ONUs is not required for the proposed scheme. / 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. / Wang, Qike. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 43-48). / Abstracts also in Chinese. / Chapter Chapter 1 --- Background --- p.1 / Chapter 1.1 --- Elastic optical networks --- p.1 / Chapter 1.2 --- Multiscast in WDM networks --- p.5 / Chapter 1.3 --- Network coding in passive optical network (PON) --- p.7 / Chapter 1.4 --- All-optical virtual private nework (VPN) in PON --- p.11 / Chapter 1.5 --- Contribution of this thesis --- p.13 / Chapter 1.6 --- Organization of this thesis --- p.15 / Chapter Chapter 2 --- Analysis of multicast in elastic optical networks --- p.16 / Chapter 2.1 --- Introduction --- p.16 / Chapter 2.2 --- Network model and heuristics --- p.18 / Chapter 2.2.1 --- Multicast-capable node architecture --- p.18 / Chapter 2.2.2 --- Multicast goup size (MGS) factor --- p.19 / Chapter 2.2.3 --- Network resource and assumption --- p.19 / Chapter 2.2.4 --- Multicast routing and spectrum allocation (MC-RSA) heuristics --- p.20 / Chapter 2.3 --- Numerical results --- p.22 / Chapter 2.4 --- Summary --- p.27 / Chapter Chapter 3 --- Physical-layer network coding (PNC) in TDM-PON --- p.28 / Chapter 3.1 --- Introduction --- p.28 / Chapter 3.2 --- A novel PNC in TDM-PON scheme for all-optical VPN applications --- p.31 / Chapter 3.2.1 --- System architecture --- p.31 / Chapter 3.2.2 --- Implementation of PNC --- p.32 / Chapter 3.2.3 --- Management of wavelength collision --- p.33 / Chapter 3.3 --- Experiemnts and results --- p.35 / Chapter 3.4 --- Summary --- p.39 / Chapter Chapter 4 --- Conclusion and Future Works --- p.40 / Chapter 4.1 --- Conclusion of this thesis --- p.40 / Chapter 4.2 --- Future works --- p.41 / Bibliography --- p.43 / List of Publications --- p.50