Dissertations / Theses on the topic 'Hybrid Multicast'

Native Multicast plays an important role in distributing and managing delivery of some of the most popular Internet applications, such as IPTV and media delivery. However, due to patchy support and the existence of multiple approaches for Native Multicast, the support for Native Multicast is fragmented into isolated areas termed Multicast Islands. This renders Native Multicast unfit to be used as an Internet wide application. Instead, Application Layer Multicast, which does not have such network requirements but is more expensive in terms of bandwidth and overhead, can be used to connect the native multicast islands. This thesis proposes Opportunistic Native Multicast (ONM) which employs Application LayerMulticast (ALM), on top of a DHT-based P2P overlay network, and Automatic Multicast Tunnelling (AMT) to connect these islands. ALM will be used for discovery and initiating the AMT tunnels. The tunnels will encapsulate the traffic going between islands' Primary Nodes (PNs). AMT was used for its added benefits such as security and being better at traffic shaping and Quality Of Service (QoS). While different approaches for connecting multicast islands exists, the system proposed in the thesis was designed with the following characteristics in mind: scalability, availability, interoperability, self-adaptation and efficiency. Importantly, by utilising AMT tunnels, this approach has unique properties that improve network security and management.

The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Title from PDF of title page (University of Missouri--Columbia, viewed on September 14, 2009). Thesis advisor: Dr. Wenjun Zeng, Includes bibliographical references.

In recent years, we have witnessed a growing interest in the synchronous collaboration based class of applications. Several techniques for Collaborative Virtual Environments (CVE) and Collaborative Haptic, Audio and Visual Environments (C-HAVE) have been designed. However, several challenging issues remain to be resolved before CVE and C-HAVE technologies become a common place. In this thesis, we focus on applications that are based on closely coupled and highly synchronized haptic tasks that require a high-level of coordination among the participants. Four main protocols have been designed, in the literature, to resolve the synchronization issues in collaborative virtual environments: the Synchronous Collaboration Transport Protocol (SCTP), the Selective Reliable Transmission Protocol (SRTP), the Reliable Multicast Transport Protocol (RMTP) and the Scalable Reliable Multicast (SRM). However, none of these protocols has been able to meet all of the basic CVE requirements, i.e., scalability, reliability, synchronization, and minimum delay. In this thesis, we propose a hybrid multicast transport protocol, which we refer to as HMTP that aims at meeting the CVE requirements. We describe our protocol, discuss its implementation and report on the performance results we have obtained for two virtual tele-surgery training class of applications.

Dissertação apresentada para obtenção do Grau de Mestre em Informática, pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia
In recent years we have observed an increased use of the Internet as a means for transmitting large content. There have been several technology attempts to attack this problem, including costly distribution networks and, more recently, peer to peer (P2P) protocols. Amongst these P2P protocols, BitTorrent has proven itself as an effective means for transmitting large content items and today enjoys great popularity. Numerous researchers have analyzed BitTorrent and proposed concepts and models to enhance its reliability, efficiency and fairness. Further, there are proposals to extend BitTorrent to support on-demand multimedia streaming. In this Dissertation we present Bitocast, a content distribution system that combines IP Multicast and BitTorrent protocols in order to achieve a more efficient usage of an Internet Service Provider’s network and reduce download time when serving large sets of contents to large audiences.

Multimedia services envisaged for high speed networks may have large numbers of users, require high volumes of network resources and have real-time delay constraints. For these reasons, several multicast routing heuristics that use two link metrics have been proposed with the objective of minimising multicast tree cost while maintaining a bound on delay. Previous evaluation work has compared the relative average performance of some of these heuristics and concludes that they are generally efficient. This thesis presents a detailed analysis and evaluation of these heuristics which illustrate that in some situations their average performance is prone to wide variance for a particular multicast in a specific network. It concludes that the efficiency of an heuristic solution depends on the topology of both the network and the multicast, which is difficult to predict. The integration of two heuristics with Dijkstras shortest path tree algorithm is proposed, to produce a hybrid that consistently generates efficient multicast solutions for all possible multicast groups in any network. The evaluation results show good performance over a wide range of networks (flat and hierarchical) and multicast groups, within differing delay bounds. The more efficient the multicast tree is, the less stable it will be as multicast group membership changes. An efficient heuristic is extended to ensure multicast tree stability where multicast group membership is dynamic. This extension decreases the efficiency of the heuristics solutions, although they remain significantly cheaper than the worst case, a shortest delay path tree. This thesis also discusses how the hybrid and the extended heuristic might be applied to multicast routing protocols for the Internet and ATM Networks. Additionally, the behaviour of the heuristics is examined in networks that use a single link metric to calculate multicast trees and concludes one of the heuristics may be of benefit in such networks.

Le multicast overlay a pour objectif la réalisation des services multicast à grande échelle. Un réseau de recouvrement (réseau overlay) est une topologie virtuelle construite au dessus de l'infrastructure physique d'Internet. Le concept des réseaux de recouvrement permet au multicast d'être déployé comme réseau de service, permettant un déploiement sur des résaux hétérogènes sans modifications matérielles. Cette thèse concerne l'étude de la qualité de service pour les réseaux multicast applicatifs (critères et méthodes) afin de fournir des services multicast à grande échelle dans l'Internet. Les ressources et le coût réseau dans le contexte des réseaux de recouvrement sont différents de ceux des réseaux classiques, présentant de nouveaux défis et de nouveaux problèmes à résoudre. Le but de notre étude est de permettre une meilleure utilisation de réseau et d'améliorer la qualité de service d'une architecture donnée, en fonction de ses caractéristiques. Pour cela, nous allons traité la problématique d'optimisation suivant deux axes : l'optimisation de la topologie de recouvrement (par un compromis entre le délai moyen et la bande passante) et la gestion efficace du trafic de session (par un algorithme de routage dans le recouvrement). La combinaison de ces deux axes fournit un procédé complet pour la planification et le contrôle d'un réseau de multicast overlay. Cette étude a ensuite servi de base à la proposition de différents mécanismes d'optimisation de l'architecture MPNT (Multicast Proxies Network). L'objectif premier de cette architecture étant de fournir une solution multicast pour des applications à grande échelle, la scalabilité fut donc notre souci principal. Par conséquent, nous avons proposé des mécanismes d'optimisation scalables décentralisés pour construire et maintenir un recouvrement efficace, offrant une diffusion multicast robuste et s'adaptant aux différents contextes d'utilisation
Overlay multicast (OM) aims to provide multicast services on a large scale, independently the underlying network. An overlay network is a virtual topology built over the physical infrastructure of the Internet. The concept of overlay networks makes it possible to deploy multicast services over heterogeneous networks without hardware modifications. This PhD thesis work concerned the study of the quality of the service (criteria and methods) of overlay multicast networks in order to provide scalable multicast services over the Internet. The resources and the network cost in the context of overlay networks are different from those of the traditional networks, introducing new challenges as well as new problems to be solved. The purpose of our study is to allow a better use of the network and to improve the quality of service for a given architecture, according to its characteristics. We treated the problems of optimization along two axes : the optimization of the overlay topology (by a compromise between delay and bandwidth) and the effective management of the multicast traffic of a session. The combination of these two axes provided a complete process for the planning and the control of an overlay multicast network. This study was then used as a basis for the purposal of various mechanisms of optimization for the MPNT architecture (Multicast Proxies NeTwork). The first objective of this architecture being to provide a multicast solution for large scale applciations, the scalability was thus our main concern. Consequently, we proposed scalable and shared mechanisms to builk and maintain an effective overlay topology, providing a robust multipoint broadcast and adapted to the various contexts of use

Esta pesquisa investiga os fatores direcionadores para a utilização de produtos híbridos em contextos de transição tecnológica. É considerado um produto híbrido aquele que reúne, em sua arquitetura, muitos dos componentes ou subsistemas centrais pertencentes a duas gerações tecnológicas distintas para desempenhar uma mesma função o que caracteriza o produto como um híbrido intergerações. Uma destas tecnologias é a tecnologia estabelecida (a antiga) e a outra é a inovação (a nova tecnologia) lançada e/ou em teste no mercado. Apesar de o emprego de produtos híbridos não ser recente, este trabalho objetiva analisar e contribuir com a discussão sobre seus antecedentes, isto é, pesquisar os principais elementos e condicionantes do emprego de produtos híbridos, investigação bastante incipiente na literatura. Com base na revisão das literaturas de estratégias de inovação, desenvolvimento de novos produtos, cadeia de suprimentos e também de competências organizacionais, o trabalho levanta quatro proposições acerca das condições de maior propensão ao emprego da estratégia de produto híbrido. O estudo de caso foi escolhido como abordagem de pesquisa, realizado de modo retrospectivo (histórico) com a utilização de casos múltiplos descritos em profundidade, cobrindo aproximadamente cem anos de história de cada setor estudado. Três estudos de caso históricos são analisados, onde há cinco momentos de transições tecnológicas, nos quais um híbrido, ao menos, fora lançado. Cada um dos casos retratou a evolução de um produto em um setor industrial norte-americano (pneu, máquina de escrever e máquina fotográfica). A análise dos casos e a dos resultados sugerem a existência de dois tipos distintos de híbridos que estariam vinculados à estratégia da empresa no momento da descontinuidade tecnológica. O híbrido de exploração (exploitation hybrid) estaria ligado a empresas que adotam uma estratégia de cunho mais defensivo que buscam continuar se beneficiando de sua posição de mercado e estrutura instalada, porém buscando se inserir de alguma forma na transição tecnológica. O híbrido de prospecção (exploration hybrid) foi encontrado em empresas inovadoras que adotam uma postura ofensiva e que estariam dispostas a gerar a descontinuidade promovendo a adoção da inovação que desenvolveram. Os resultados propõem que este híbrido poderia ser usado neste esforço por possibilitar uma transição mais gradual em direção à nova tecnologia, reduzindo o risco, o custo de mudança e a rejeição à inovação. Esta pesquisa contribui para uma definição mais precisa para o híbrido intergeração, além de apresentar uma tipologia para defini-lo em razão da sua utilização durante a transição tecnológica e caracterizar as condições e alguns de seus principais direcionadores ou antecedentes. Outra contribuição do trabalho é a proposição de um continuum para as estratégias de gestão da inovação em períodos de transições tecnológicas e que posicionam os produtos híbridos.
This research investigates the guideline factors for the use of hybrid products in technological transition contexts. This research considered that hybrid product is the one that gathers in its architecture, many central subsystems or components belonging to two different generations of technology to perform the same thing, which characterizes the product as an intergenerational hybrid. One of these technologies is the established technology (the mature one) and the other is the innovation (the new technology) launched and/or in the test in the market. Although the use of hybrid products are not new, this paper aims to analyze and contribute to the discussion about your antecedents, i.e., search the main elements and conditions of use of hybrid products, incipient research in the literature. Based on the review of the literature of innovation strategies, new products development, supply chain as well as organizational competencies, the work raises four propositions about the more likely conditions to the use of hybrid product strategy. The case study was chosen as a research approach , carried out retrospectively (historical ) with the use of multiple cases described in depth, covering about a hundred years of history of each studied sector Three historical case studies are analyzed highlighted five technological transitions which, at least, one hybrid was released. Each one of the cases portrayed the evolution of a product in an industrial sector in the United States (tire, typewriter and camera). The analysis of the cases and the results suggest the existence of two distinct types of hybrids that would be linked to the business strategy at the time of the technological discontinuity. The exploitation hybrid would linked to companies that adopt a more defensive strategy, seeking to continue to benefit from its market position and installed structure. The exploration hybrid was found in innovative companies that adopt an offensive posture and would be willing to generate discontinuity by promoting the adoption of the innovation they developed. The results suggest that this hybrid could be used in this effort, because it allows a more gradual transition towards the new technology, reducing the risk, cost of change and the rejection of the innovation. This research contributes to a more precise definition for the intergeneration hybrid, besides presenting a typology to define it, due to its use during the technological transition and characterizing the conditions and some of its key drivers or antecedents. Another contribution of this work is the proposal of a continuum for innovation management strategies in technological transitions periods and positioning the hybrid products.

Le travail effectué aborde la problématique des services de communication multipoints fiables à grande échelle. Dans ce contexte, la possibilité de déployer un tel service au moyen d'un satellite géostationnaire émettant en bande Ka est étudiée. L'emploi de la bande $Ka$ introduit cependant une grande variabilité de la qualité de réception au niveau des utilisateurs finals, rendant nécessaire l'utilisation d'un protocole de transport mettant en oeuvre des mécanismes spécifiques. Selon une fonction de coût définie, la comparaison des solutions basées sur IP Multicast classiquement utilisées montre que l'utilisation d'une approche hybride couplant l'utilisation des réseaux satellites et terrestres est avantageuse. Le principe de la proposition, nommée Hybrid Satellite Terrestrial Reliable Multicast, consiste ainsi à choisir, en fonction de la taille du groupe, le moyen de diffusion le plus rentable --- au vu d'une fonction de coût définie. Une description détaillée de la position inclut le comportement de la source et des récepteurs, et le format des messages échangés. Bien que le principe de cette approche soit simple, plusieurs points durs sont liés à la conception des mécanismes adéquats. Ces problèmes concernent notamment la gestion de la fiabilité (utilisation de code correcteur d'erreur ou FEC), l'estimation de taille de très grands groupes, et la reprise des erreurs par voie terrestre (utilisation de réseaux de pair-à-pairs). Ces mécanismes sont étudiés de manière unitaire afin de déterminer des configurations satisfaisantes, et pour détecter des problèmes de performances. Ces mécanismes étant définis, la proposition de transport a été globalement modélisée, de manière à obtenir une vérification fonctionnelle du service proposé. Le protocole a été décrit au moyen du profil UML temps réel TURTLE. Les résultats de validation ont été obtenus grâce à la chaîne d'outils TTool-RTL, et à CADP.

La problématique des services de communication multipoints fiables à grande échelle. Dans ce contexte, un protocole nommé Hybrid Satellite Terrestrial Reliable Multicast est proposé. Le principe de ce dernier consiste à choisir, en fonction de la taille du groupe, le moyen de diffusion le plus rentable (i. E. Le réseau terrestre ou satellite) au vu d'une fonction de coût prédéfinie. Bien que le principe soit simple, plusieurs points durs sont liés à la conception de mécanismes adéquats : la gestion de la fiabilité (utilisation de code correcteur d'erreur), l'estimation de taille de très grands groupes, et la reprise des erreurs par voie terrestre (utilisation de réseaux de pair-à-pairs). Ceux-ci sont étudiés séparément afin de déterminer des configurations satisfaisantes, et pour détecter des problèmes de performance. Ces mécanismes étant définis, la proposition de transport a été globalement modélisée, de manière à entamer la validation fonctionnelle du service proposé.

Internet, the largest network of networks has evolved from a research-oriented network to one with a myriad number of commercial applications. Over the last several years, there has been an explosion in the introduction of new Internet technologies with high-end workstations being engaged in real-time and multimedia communications like video conferencing, IP telephony, streaming video broadcast, online gaming etc. Supporting both the legacy services like email, file transfer and real-time multimedia like video conferencing requires service differentiation of Internet traffic. Moreover, multipoint communication requires implementation of multicast services. Two of the emerging technologies for service differentiation in multipoint communication are Differentiated Services (DiffServ) and multicasting. Although the two technologies share complementary goals, the integration of the two technologies is a non-trivial issue due to three fundamental problems. The problems are the scalability of per-group state information, sender- versus receiver-driven QoS, and resource management. The issues surrounding how to solve these problems provide the basis for this thesis. Edge Based Multicast (EdgeCast)--has been proposed here to satisfy the requirements for scalable DiffServ multicasting architectures. In addition it also presents a new join/leave protocol to support the EdgeCast architecture. Performance simulation has been done and compared with some already developed techniques. Finally, the whole work has been summarized and comments have been made on future applications of the architecture and several potential areas for future research.

The application of FEC to multicast communication has seen a significant development in years. However, there are still numerous questions about practical implementations that make this field interesting for research. This work is an attempt to address some of these problems. In the first part of this investigation, the effect of different error control techniques to the multicast communication is analyzed. A new proactive hybrid FEC/ARQ technique is tested in the lab. This technique is based on the RSE code (Reed-Solomon Erasure Correcting Code) which mainly deals with erasures of network at packet level. The sender sends parity-encoded RSE repairs along with original data. The redundant information will allow the reconstruction of some amount of missing data at the receivers. If these proactive repairs are insufficient to reliably deliver the data to receivers, additional RSE repairs are obtained via receiver requests for the transmission (ARQ) of additional repairs. The thesis also proposes a reliable multicast implementation scheme, and develops a laboratory test-bed for this scheme. Multicast tree is built and maintained by adopting some control messages for multicast session establishment. According to dynamically established multicast routing table, data packets will be propagated from senders to receivers through Rendezvous Points. The above-mentioned proactive hybrid FEC/ARQ technique is applied to this scheme. The performance of the proposed scheme is compared to that of the traditional ARQ scheme. Significant improvements are achieved in terms of transmission time, feedback implosion, and bandwidth utilization.

碩士
國立交通大學
資訊科學系所
92
We propose a novel multicast routing protocol, called hybrid overlay multicast routing protocol (HOMRP) for MANETs (Mobile Ad hoc NETworks). Existing overlay multicast routing protocols in MANETs have the main drawback of high packet delivery delay. In order to improve this shortcoming, we integrate multicasting and unicast tunnels for efficient packet delivery. In HOMRP, it creates multiple local multicast trees. Each pair of parent node and child node in a local multicast tree is at a distance of one-hop. It uses multicasting to deliver multicast packets in local multicast trees. This will provide efficient data forwarding. Unicast tunnels are used for transmitting packets between local multicast trees. To avoid looping routes between local multicast trees, each local multicast tree is assigned a tree ID. To deal with dynamic changing of network topology in MANETs, HOMRP uses a tree consolidation scheme for highly efficient data forwarding. In addition, in order to reduce the overhead of control messages flooding, we adopt two-level flooding for member discovery by limiting the value of TTL in a packet. HOMRP does not restrict to use any specific unicast routing protocol; hence it can operate with any unicast routing protocols. Route maintenance in HOMRP is initiated by each multicast group member. Each member only needs to maintain the nearby members. The same route structure of a multicast group can be shared by multiple senders (sources); this brings high efficiency of route maintenance. Simulation results have shown that the packet delivery ratio of HOMRP is close to that of ODMRP and is 50% better than that of AMRoute. HOMRP reduces 18% control overhead compared to ODMRP and reduces 9% end-to-end packet delay compared to AMRoute. In sum, HOMRP provides lower control overhead, high packet delivery ratio and is especially suitable for multicast applications with larger number of senders, compared to other approaches.

碩士
國立臺北科技大學
資訊工程系研究所
105
As the wide spread of smart devices, the digital set-top box (STB) that conventionally only serves TV watchers with broadcast channel videos now also provides streaming videos to mobile devices. The STB users in a family may have individual or common needs. If two or more users intend to watch the same video, applying the conventional unicast technique for multimedia streaming service will cause waste in bandwidth. In this study we realized a hybrid FLUTE (File Delivery over Unidirectional Transport) multicast and unicast and MPEG-DASH (Dynamic Adaptive Streaming over Hypertext Transfer Protocol) adaptive multimedia streaming system. The video quality was adjusted using the layer coding of H.264 SVC (Scalable Video Coding). We further proposed a quality-adaptation algorithm for this system, which dynamically determined the transmitted video quality according to the user’s demand, the bandwidth, and user’s device specification. Simulation results showed that the proposed algorithm effectively reduced the waiting time and enhanced the received video quality of smart device users.

碩士
國立暨南國際大學
資訊管理學系
97
The problem for finding the minimum cost multicasting tree with multiple constraints is NP-complete. This paper describes the Quality of Service (QoS) multicast routing problem with bandwidth and delay constraints and proposed a hybrid algorithm based on metaheuristics. Our algorithm hybridizes particle swarm optimization (PSO) and ant colony optimization (ACO) with the media through pheromone and guiding best solutions. We first introduce a new encoding method for the multicasting tree and then propose a heuristic based on tabu list and feedback list in order to avoid cycling on the visited path. With this mechanism, our algorithm intends to find the minimum cost multicasting tree while satisfying the imposed constraints. The experimental results manifest that, compared to other algorithms, our algorithm is more likely to escape from the trap of infeasible solutions and reducing the probability of causing routing failure. The attained multicast routing tree has high quality.

博士
國立中正大學
電機工程研究所
99
With fast development of group communications, the requests for multimedia multicast service and application are increasingly growth over current Internet. The key points for multicasting applications are quality of service (QoS) and quality of experience (QoE) supports which involve the multicast group dynamics of membership management, multicast network dynamics with P2MP routing and resource control, and dynamic multicast traffic control. In this dissertation, the proposed multicast QoS supported network design is divided into three relevant phases of multicast session, such as static and mobile member joining in multicast group management phase, multicast QoS routing and resource reservation in multicast network control phase, and multicast traffic transmission in traffic control phase. The purpose of the centralized control and management (CCM) architecture is to setup network provisioning models and provide viable solutions for multicast QoS problems: scalability, efficiency and deployment, which is caused by dynamic behaviors in multicast session. The multicast session is created by assigning the group address for static and dynamic member joining. In multicast group management phase, the RP shared tree of CCM can be applied for the member joining and network information collection. And, the multicast agent (MA) of overlay network can be used for mobile member joining via IGMP mechanism. In multicast network control phase, the multicast session is setup by multicast tree construction, spanning all the sources and receivers. In multicast CCM architecture, we propose multicast resource broker (MRB) function with efficient resource provisioning which employs the CSPF multicast routing, and RSVP-TE resource reservation for multicast tree. Furthermore, for IPTV multicast application development, we propose HT-ERM approach for IPTV application which can improve the performance for multicast network delivery and channel control. For inter-domain multicast networking, we propose overlay tree construction algorithm to achieve all-IP multicasting. In multicast traffic control phase, the proposed HT-ERM switchover control mechanism can provide fast switchover mechanism for channel change delay, and admission control for multicast IPTV QoS traffic distribution over hybrid-tree based multicast network. In additions, to achieve the seamless multicast transport connectivity, we propose context aware handoff mobility connection approach for mobile member access service. Finally, our proposed models and algorithms can enhance the existing protocols and apply for practical framework implementation. The experimental results and performance evaluations for the reference architecture and application are validated to provide all-IP multicast transport service over Future Internet.

by Ng Chi Ho.
Thesis (M.Phil.)--Chinese University of Hong Kong, 2001.
Includes bibliographical references (leaves 64-65).
Abstracts in English and Chinese.
ABSTRACT --- p.I
ACKNOWLEDGEMENT --- p.II
TABLE OF CONTENTS --- p.III
LIST OF TABLES --- p.VI
LIST OF FIGURES --- p.VII
LIST OF ABBREVIATIONS --- p.X
Chapter CHAPTER 1 --- INTRODUCTION --- p.1
Chapter 1.1 --- Overview --- p.1
Chapter 1.2 --- Related works --- p.5
Chapter 1.3 --- Organization of this Thesis --- p.6
Chapter CHAPTER 2 --- BACKGROUND --- p.7
Chapter 2.1 --- Introduction to VOD Systems --- p.7
Chapter 2.1.1 --- Pure unicast VOD System --- p.8
Chapter 2.1.2 --- Pure multicast VOD System --- p.9
Chapter 2.1.3 --- Centralized VOD System --- p.9
Chapter 2.1.4 --- Distributed VOD System --- p.10
Chapter 2.1.5 --- Hybrid VOD System (DINA) --- p.11
Chapter 2.1.6 --- Comparisons --- p.12
Chapter 2.2 --- Interactive Functions --- p.14
Chapter 2.2.1 --- Speedup --- p.14
Chapter 2.2.2 --- Split and merge (I and S streams) --- p.14
Chapter 2.2.3 --- Prerecord --- p.15
Chapter 2.3 --- Error Recovery --- p.16
Chapter 2.3.1 --- Pure FEC --- p.17
Chapter 2.3.2 --- Pure ARQ --- p.17
Chapter 2.3.3 --- Hybrid ARQ --- p.18
Chapter 2.3.4 --- Rate-Compatible Punctured Convolutional Codes --- p.18
Chapter CHAPTER 3 --- HYBRID MULTICAST-UNICAST VOD SYSTEM --- p.21
Chapter 3.1 --- System Overview --- p.21
Chapter 3.1.1 --- VSC (Video Server Cluster) --- p.22
Chapter 3.1.2 --- DIS (Distributed Interactive Server) --- p.24
Chapter 3.1.3 --- NAK (Negative Acknowledgement Server) --- p.25
Chapter 3.1.4 --- CS (Client Stations) --- p.26
Chapter 3.1.5 --- MBN (Multicast Backbone Network) --- p.27
Chapter 3.1.6 --- LDN (Local Distribution Network) --- p.27
Chapter 3.2 --- Interactive Functions --- p.28
Chapter 3.2.1 --- Hybrid Multicast- Unicast --- p.28
Chapter 3.2.2 --- Pause --- p.30
Chapter 3.2.3 --- Slow Forward (SF) --- p.33
Chapter 3.2.4 --- Slow Backward (SB) --- p.35
Chapter 3.2.5 --- Fast Forward (FF) / Fast Backward (FB) --- p.38
Chapter 3.2.6 --- Jump Forward (JF) / Jump Backward (JB) --- p.41
Chapter 3.3 --- System Performance --- p.46
Chapter 3.3.1 --- System Model --- p.46
Chapter 3.3.2 --- Simulation Results --- p.47
Chapter 3.3.3 --- Trade off --- p.53
Chapter CHAPTER 4 --- DISTRIBUTED TYPE-II HARQ --- p.54
Chapter 4.1 --- Algorithm Description --- p.54
Chapter 4.1.1 --- Design details --- p.54
Chapter 4.1.2 --- Simulation Results --- p.59
Chapter CHAPTER 5 --- CONCLUSION --- p.62
BIBLIOGRAPHY --- p.64

碩士
國立臺北科技大學
資訊工程系
107
The video streaming platforms have recently evolved to provide diversified services, including live video, video on demand, and TV shows premiere. The traditional unicast method may cause a waste of bandwidth when many users watch the same video contents at the same time. In this study, we realized a hybrid unicast and multicast adaptive streaming system. The system uses ROUTE (Real-time Object delivery over Unidirectional Transport) combined with the MPEG-DASH (Dynamic Adaptive Streaming over HTTP) to simultaneously satisfy the needs of unicast and multicast users. We use the SHVC (Scalable High-efficiency Video Coding) to generate high-quality video of different bitrates. The QoE (Quality of Experience) function is measured by bitrate, PSNR, and server bandwidth. The proposed adaptive algorithm dynamically allocates the bitrates of unicast and multicast users and achieves optimal average viewing quality. Experimental results show that the proposed algorithm effectively improves the system’s QoE under various bandwidth models compared to the conventional method. In the hybrid multicast and unicast scenario, the PSNR for the stable, incremental, and decreasing bandwidth models are increased by 0.79 dB, 4.92 dB, and 7.88 dB, respectively. The average quality switching frequency of users is also decreased.

碩士
國立中正大學
資訊工程學系碩士在職專班
107
Due to Proliferation in mobile and other network devices increase of video streaming users, consumption of video data has an unprecedented surge, resultant in the mobile networks are facing more traffic loads. So the Multicast video streaming is an urgently needed application over mobile networks. Video on Demand (VoD) multicast service could be delivered efficiently, but in the live streaming multicast procedure and implementation has several challenges. In this thesis, we proposed a design for live streaming multicasting with the help of File Delivery over Unidirectional Transport (FLUTE) and Dynamic Adaptive Streaming (DASH) in Software Defined Network (SDN) based Long Term Evolution (LTE) network with Mobile Edge Computing (MEC) Environment. FLUTE is a file delivery multicast protocol, which can be used to provide multicast video streaming services in SDN based LTE network. Where SDN is open interface provides programmability and create topology discovery for LTE network with the help of Open-Flow protocols to control and configure data plan, and the presence of MEC servers which provides cloud computing capabilities to the edge of cellular networks. In this thesis, we explained the procedure how FLUTE protocol multicasting and MEC joined and leave multicast channel according to their need and how SDN control LTE traffic for FLUTE file delivery.

Chan Kwun-chung.
Thesis (M.Phil.)--Chinese University of Hong Kong, 2001.
Includes bibliographical references (leaves 71-73).
Abstracts in English and Chinese.
摘要 --- p.II
ABSTRACT --- p.III
ACKNOWLEDGEMENT --- p.V
TABLE OF CONTENTS --- p.VI
LIST OF FIGURES --- p.X
LIST OF SYMBOLS --- p.XII
Chapter 1. --- INTRODUCTION --- p.1
Chapter 1.1 --- Contributions --- p.3
Chapter 1.2 --- Organization of the Thesis --- p.4
Chapter 1.3 --- Publications --- p.5
Chapter 2. --- RELATED WORKS --- p.6
Chapter 2.1 --- Previous VOD System --- p.7
Chapter 2.1.1 --- Service Model --- p.7
Chapter 2.1.1.1 --- Unicast VOD --- p.7
Chapter 2.1.1.2 --- Multicast VOD --- p.8
Chapter 2.1.2 --- Architecture --- p.9
Chapter 2.1.2.1 --- Centralized Architecture --- p.9
Chapter 2.1.2.2 --- Distributed Architecture --- p.10
Chapter 2.1.3 --- Interactive Function --- p.11
Chapter 2.1.3.1 --- Limited Interactive Function --- p.11
Chapter 2.1.3.2 --- Unlimited Interactive Function --- p.11
Chapter 2.1.4 --- Split and Merge Operation --- p.12
Chapter 2.1.4.1 --- SAM Scheme (Split and Merge) --- p.12
Chapter 2.1.4.2 --- SRMDRU Scheme (Single Rate Multicast Double Rate Unicast) --- p.14
Chapter 2.2 --- Previous Caching Algorithm --- p.15
Chapter 2.2.1 --- LFU (Least Frequently Used) --- p.15
Chapter 2.2.2 --- LRU (Least Recently Used) --- p.15
Chapter 2.2.3 --- Media Stream Caching --- p.15
Chapter 3. --- DESIGN OFA NOVEL VOD SYSTEM --- p.17
Chapter 3.1 --- System Architecture --- p.18
Chapter 3.1.1 --- Multicast Video Server Cluster (MVSC) --- p.19
Chapter 3.1.2 --- Unicast Video Server Cluster (UVSC) --- p.20
Chapter 3.1.3 --- Multicast Backbone Network (MBN) --- p.20
Chapter 3.1.4 --- Local Distribution Network (LDN) --- p.21
Chapter 3.1.5 --- Distributed Interactive Server (DIS) --- p.21
Chapter 3.1.6 --- Distributed Proxy Server (DPS) --- p.22
Chapter 3.1.7 --- Client Station (CS) --- p.22
Chapter 3.2 --- Batched Multicast Transmission --- p.24
Chapter 3.3 --- Split and Merge Operation --- p.26
Chapter 3.4 --- Interactive Function --- p.31
Chapter 3.4.1 --- Pause --- p.31
Chapter 3.4.2 --- Slow Motion --- p.35
Chapter 3.4.3 --- Various Speed Fast Forward / Fast Rewind (FF/REW) --- p.37
Chapter 3.4.4 --- Jump Forward/Jump Backward (JF/JB) --- p.42
Chapter 3.5 --- Performance Analysis --- p.46
Chapter 3.5.1 --- Model --- p.46
Chapter 3.5.2 --- System Parameters --- p.49
Chapter 3.5.3 --- Results --- p.49
Chapter 4. --- DESIGN OF A VIDEO PROXY SYSTEM --- p.57
Chapter 4.1 --- Video Proxy System --- p.58
Chapter 4.1.1 --- Priority Function --- p.59
Chapter 4.1.2 --- Two-Stage Replacement Policy --- p.60
Chapter 4.1.3 --- Caching Policy --- p.61
Chapter 4.2 --- Performance Evaluation --- p.63
Chapter 4.2.1 --- Simulation Environment --- p.63
Chapter 4.2.2 --- Performance Metric --- p.64
Chapter 4.2.3 --- Results --- p.64
Chapter 5. --- CONCLUSION --- p.69
BIBLIOGRAPHY --- p.71

碩士
國立中興大學
通訊工程研究所
105
Multicast is widely implied in Wireless Mesh Networks, especially through Muilticast Tree structure, source node can collect all the member’s data from system. Hence, installing sensor on each node in the system, node can process information exchange through Multicast Tree and send the measured data to source node. It can reduce the waste of human resources. In this thesis, we use AM335Xevm Evaluation Board and WL1835MOD Wireless Module to construct a Hybrid Internet Of Things(IOT) system. Among this system, senor can connect to AM335Xevm Evaluation Board through BLE, and every AM335Xevm Evaluation Board can form a high efficiency Muilticast Tree based on Wifi Wireless Mesh Networks. After that, via Multicast Tree can send data that sensor collected to back-end server. Multicast Tree construct protocol we developed can according latest information that exchanging packets on node to adjust step by step. After counting all the weight on possible routing path, choose the lowest weight to improve efficiency for overall system message sending.

Internship Report presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics
Due to the increased availability of mature data mining and analysis technologies supporting CRM processes, several financial institutions are striving to leverage customer data and integrate insights regarding customer behaviour, needs, and preferences into their marketing approach. As decision support systems assisting marketing and commercial efforts, Recommender Systems applied to the financial domain have been gaining increased attention. This thesis studies a Collaborative- Demographic Hybrid Recommendation System, applied to the financial services sector, based on real data provided by a Portuguese private commercial bank. This work establishes a framework to support account managers’ advice on which financial product is most suitable for each of the bank’s corporate clients. The recommendation problem is further developed by conducting a performance comparison for both multi-output regression and multiclass classification prediction approaches. Experimental results indicate that multiclass architectures are better suited for the prediction task, outperforming alternative multi-output regression models on the evaluation metrics considered. Withal, multiclass Feed-Forward Neural Networks, combined with Recursive Feature Elimination, is identified as the topperforming algorithm, yielding a 10-fold cross-validated F1 Measure of 83.16%, and achieving corresponding values of Precision and Recall of 84.34%, and 85.29%, respectively. Overall, this study provides important contributions for positioning the bank’s commercial efforts around customers’ future requirements. By allowing for a better understanding of customers’ needs and preferences, the proposed Recommender allows for more personalized and targeted marketing contacts, leading to higher conversion rates, corporate profitability, and customer satisfaction and loyalty.

碩士
國立臺灣科技大學
電機工程系
102
In this paper, the on-line visual imitation of an HR for different humans with 3-D motions is developed by the proposed visual imitation principle. First, the sequence of 3-D motion of a human is captured by a stereo vision system (SVS), which skeleton algorithm can capture and estimate 3-D coordinates of fifteen joints, i.e., head, neck, torso, left shoulder, left elbow, left hand, right shoulder, right elbow, right hand, left hip, left knee, left foot, right hip, right knee, and right foot. Because the dynamic balance of the imitator (i.e., humanoid robot (HR)) is not considered, the proposed on-line visual imitation is partitioned into upper body (UB) and lower body (LB). Based on the 3-D coordinates of head, left and right feet, eleven stable motions with developed feature vectors for the LB are classified by the proposed multi-class support vector machine (MSVM). The inverse kinematics (IK) of two pairs of (hand, elbow) of the UB are respectively approximated by eight pre-trained hybrid learning models for eight sub-work spaces of two arms. The comparisons between hybrid learning model based (HLMB) IK and ordinary IK are also discussed. Combining the classified motions of LB and the IKs for two pairs of (hand, elbow) of UB with interpolation imitates the 3-D motions of different humans. Finally, the corresponding experiments confirm the effectiveness and robustness of the proposed method.

Due to the highly-varying wireless channels over time, frequency, and space domains, statistical QoS provisioning, instead of deterministic QoS guarantees, has become a recognized feature in the next-generation wireless networks. In this dissertation, we study the adaptive wireless resource allocation problems for statistical QoS provisioning, such as guaranteeing the specified delay-bound violation probability, upper-bounding the average loss-rate, optimizing the average goodput/throughput, etc., in several typical types of mobile wireless networks. In the first part of this dissertation, we study the statistical QoS provisioning for mobile multicast through the adaptive resource allocations, where different multicast receivers attempt to receive the common messages from a single base-station sender over broadcast fading channels. Because of the heterogeneous fading across different multicast receivers, both instantaneously and statistically, how to design the efficient adaptive rate control and resource allocation for wireless multicast is a widely cited open problem. We first study the time-sharing based goodput-optimization problem for non-realtime multicast services. Then, to more comprehensively characterize the QoS provisioning problems for mobile multicast with diverse QoS requirements, we further integrate the statistical delay-QoS control techniques — effective capacity theory, statistical loss-rate control, and information theory to propose a QoS-driven optimization framework. Applying this framework and solving for the corresponding optimization problem, we identify the optimal tradeoff among statistical delay-QoS requirements, sustainable traffic load, and the average loss rate through the adaptive resource allocations and queue management. Furthermore, we study the adaptive resource allocation problems for multi-layer video multicast to satisfy diverse statistical delay and loss QoS requirements over different video layers. In addition, we derive the efficient adaptive erasure-correction coding scheme for the packet-level multicast, where the erasure-correction code is dynamically constructed based on multicast receivers’ packet-loss statuses, to achieve high error-control efficiency in mobile multicast networks. In the second part of this dissertation, we design the adaptive resource allocation schemes for QoS provisioning in unicast based wireless networks, with emphasis on statistical delay-QoS guarantees. First, we develop the QoS-driven time-slot and power allocation schemes for multi-user downlink transmissions (with independent messages) in cellular networks to maximize the delay-QoS-constrained sum system throughput. Second, we propose the delay-QoS-aware base-station selection schemes in distributed multiple-input-multiple-output systems. Third, we study the queueaware spectrum sensing in cognitive radio networks for statistical delay-QoS provisioning. Analyses and simulations are presented to show the advantages of our proposed schemes and the impact of delay-QoS requirements on adaptive resource allocations in various environments.