Дисертації з теми "Multimedia Optimization"

Thesis (M.S.)--University of Florida, 2000.
Title from first page of PDF file. Document formatted into pages; contains xii, 154 p.; also contains graphics. Vita. Includes bibliographical references (p. 152-153).

At present, there are many hardware and software implementations that provide reasonable quality real-time multimedia communication such as video telephony and conferencing. Since sofiware-based video telephony can provide lower cost and more flexible implementations than hardware-based systems, many systems are implemented in software. Due to the computationally intensive algorithms in video telephony, there are some real-time computation bottlenecks that lead to difficulties in improving performance to achieve better quality. With the emerging of a new group of multimedia-related processor instructions, a significant improvement to the real-time implementation of multimedia communication systems can be made. One such a group of instructions is the Streaming SIMD Extensions of the Pentium® III processor. This thesis describes multimedia coding related optimizations using the Pentium® III Streaming SIMD Extensions in a video telephony system based on the ITU—T (International Telecommunication Union — Telecommunication Standardization Sector) H.324 and H.323 standards. It focuses on the implementation of three computationally intensive algorithms, the Sum of Absolute Difference (SAD) used in Video motion estimation, the Discrete Cosine Transform (DCT) used in hybrid Video codecs (e.g. H.261 and H.263), and the Dot Product (DP) in many audio codecs (e.g. G.723.1). Depending on whether algorithms use SSIMDEs instructions, integer or floating point, the performance improvement is not certain. The experimental results presented in this thesis show that SAD and DP achieve improvements as expected, whereas DCT does not benefit from using SSIMDEs. The thesis discusses performance improvements and analyses the results.

The objective of this thesis is to formulate a generic dictionary learning method with the guiding principle that states: Efficient representations lead to efficient estimations. The fundamental idea behind using transforms or dictionaries for signal representation is to exploit the regularity within data samples such that the redundancy of the representation is minimized subject to a level of fidelity. This observation translates to rate-distortion cost in compression literature, where a transform that has the lowest rate-distortion cost provides a more efficient representation than the others. In our work, rather than using as an analysis tool, the rate-distortion cost is utilized to improve the efficiency of transforms. For this, an iterative optimization method is proposed, which seeks an orthonormal transform that reduces the expected value of rate-distortion cost of an ensemble of data. Due to the generic nature of the new optimization method, one can design a set of orthonormal transforms either in the original signal domain or on the top of a transform-domain representation. To test this claim, several image codecs are designed, which use block-, lapped- and wavelet-transform structures. Significant increases in compression performances are observed compared to original methods. An extension of the proposed optimization method for video coding gave us state-of-the-art compression results with separable transforms. Also using the robust statistics, an explanation to the superiority of new design over other learning-based methods such as Karhunen-Loeve transform is provided. Finally, the new optimization method and the minimization of the "oracle" risk of diagonal estimators in signal estimation is shown to be equal. With the design of new diagonal estimators and the risk-minimization-based adaptation, a new image denoising algorithm is proposed. While these diagonal estimators denoise local image patches, by formulation the optimal fusion of overlapping local denoised estimates, the new denoising algorithm is scaled to operate on large images. In our experiments, the state-of-the-art results for transform-domain denoising are achieved.

Mobile communication systems have continued to evolve beyond the currently deployed Third Generation (3G) systems with the main goal of providing higher capacity. Systems beyond 3G are expected to cater for a wide variety of services such as speech, data, image transmission, video, as well as multimedia services consisting of a combination of these. With the air interface being the bottleneck in mobile networks, recent enhancing technologies such as the High Speed Downlink Packet Access (HSDPA), incorporate major changes to the radio access segment of 3G Universal Mobile Telecommunications System (UMTS). HSDPA introduces new features such as fast link adaptation mechanisms, fast packet scheduling, and physical layer retransmissions in the base stations, necessitating buffering of data at the air interface which presents a bottleneck to end-to-end communication. Hence, in order to provide end-to-end Quality of Service (QoS) guarantees to multimedia services in wireless networks such as HSDPA, efficient buffer management schemes are required at the air interface. The main objective of this thesis is to propose and evaluate solutions that will address the QoS optimization of multimedia traffic at the radio link interface of HSDPA systems. In the thesis, a novel queuing system known as the Time-Space Priority (TSP) scheme is proposed for multimedia traffic QoS control. TSP provides customized preferential treatment to the constituent flows in the multimedia traffic to suit their diverse QoS requirements. With TSP queuing, the real-time component of the multimedia traffic, being delay sensitive and loss tolerant, is given transmission priority; while the non-real-time component, being loss sensitive and delay tolerant, enjoys space priority. Hence, based on the TSP queuing paradigm, new buffer managementalgorithms are designed for joint QoS control of the diverse components in a multimedia session of the same HSDPA user. In the thesis, a TSP based buffer management algorithm known as the Enhanced Time Space Priority (E-TSP) is proposed for HSDPA. E-TSP incorporates flow control mechanisms to mitigate congestion in the air interface buffer of a user with multimedia session comprising real-time and non-real-time flows. Thus, E-TSP is designed to provide efficient network and radio resource utilization to improve end-to-end multimedia traffic performance. In order to allow real-time optimization of the QoS control between the real-time and non-real-time flows of the HSDPA multimedia session, another TSP based buffer management algorithm known as the Dynamic Time Space Priority (D-TSP) is proposed. D-TSP incorporates dynamic priority switching between the real-time and non-real-time flows. D-TSP is designed to allow optimum QoS trade-off between the flows whilst still guaranteeing the stringent real-time component’s QoS requirements. The thesis presents results of extensive performance studies undertaken via analytical modelling and dynamic network-level HSDPA simulations demonstrating the effectiveness of the proposed TSP queuing system and the TSP based buffer management schemes.

The transition from analogue TV to digital TV has led a fast-growing of available multimedia contents. In recent decades, there has been a great development of digital broadcasting systems, in addition to those relating to the existing DVB consortium other examples just as representative as DAB (Digital Audio Broadcasting) and DRM (Digital Radio Mondiale). The purpose of the Part I of this project therefore is as follows: the study and implementation of an algorithm for optimizing the planning of digital broadcasting networks based on COFDM modulation for use in single frequency networks (SFN). This algorithm manipulates the information on a broadcast network and estimates degree of coverage and interference that offers a particular network, and does it the best way possible and at minimum cost. Subsequently the study and implementation of this algorithm optimization, we have studied in terms of functionality and performance, to evaluate the usefulness of them and if conclusions drawn from their application can be applied reliably. Part II of this thesis proposes two novel algorithms: the Adaptive Eeal-time Multi-user Access Network Selection (ARMANS) and Prioritized ARMANS (P-ARMANS). ARMANS is a load balancing algorithm that takes into account not only the real-time global traffic load on each network, but also considering the different classes of traffic, voice, video, best-effort, and background. The proposed algorithm improves both QoS and load balancing in comparison with the case when a classic network selection with no traffic type load balancing is employed. The novel P-ARMANS load balancing algorithm Markov decision process (MDP)-based multi-agent framework was introduced to perform the bandwidth allocation on users with different typology and priority and considering and classifying the type of traffic involved in a multimedia data user demand. The proposed approach performs the available bandwidth on users with different screen resolution and priority. We demonstrate that P-ARMANS framework ensures high QoS distributing different type of traffic load among typical and business users compared to a classic no-priority model. Both ARMANS and P-ARMANS operate over heterogeneous wireless networks (HWNs), 4G-LTE and 802.11n.

This thesis proposes a convex network utility maximization (NUM) problem that can be solved to optimize a cross-layer network based on user and system defined requirements for quality and link capacity of multimedia applications. The problem can also be converged to a distributed solution using dual decomposition. Current techniques do not address the changing system's requirements for the network in addition to the user's requirements for an application when optimizing a cross-layer network, but rather focus on optimizing a dynamic network to conform to a real-time application or for a specific performance. Optimizing the cross-layer network for the changing system and user requirements allows a more accurate optimization of the overall cross-layer network of any given multi-node, ad-hoc wireless application for data transmission quality and link capacity to meet overall mission demands.

As a large volume of digital video data becomes available, along with revolutionary advances in multimedia technologies, demand related to efficiently retrieving and recounting multimedia data has grown. However, the inherent complexity in representing and recognizing multimedia data, especially for large-scale and unconstrained consumer videos, poses significant challenges. In particular, the following challenges are major concerns in the proposed research. One challenge is that consumer-video data (e.g., videos on YouTube) are mostly unstructured; therefore, evidence for a targeted semantic category is often sparsely located across time. To address the issue, a segmental multi-way local feature pooling method by using scene concept analysis is proposed. In particular, the proposed method utilizes scene concepts that are pre-constructed by clustering video segments into categories in an unsupervised manner. Then, a video is represented with multiple feature descriptors with respect to scene concepts. Finally, multiple kernels are constructed from the feature descriptors, and then, are combined into a final kernel that improves the discriminative power for multimedia event detection. Another challenge is that most semantic categories used for multimedia retrieval have inherent within-class diversity that can be dramatic and can raise the question as to whether conventional approaches are still successful and scalable. To consider such huge variability and further improve recounting capabilities, a per-exemplar learning scheme is proposed with a focus on fusing multiple types of heterogeneous features for video retrieval. While the conventional approach for multimedia retrieval involves learning a single classifier per category, the proposed scheme learns multiple detection models, one for each training exemplar. In particular, a local distance function is defined as a linear combination of element distance measured by each features. Then, a weight vector of the local distance function is learned in a discriminative learning method by taking only neighboring samples around an exemplar as training samples. In this way, a retrieval problem is redefined as an association problem, i.e., test samples are retrieved by association-based rules. In addition, the quality of a multimedia-retrieval system is often evaluated by domain-specific performance metrics that serve sophisticated user needs. To address such criteria for evaluating a multimedia-retrieval system, in MFoM learning, novel algorithms were proposed to explicitly optimize two challenging metrics, AP and a weighted sum of the probabilities of false alarms and missed detections at a target error ratio. Most conventional learning schemes attempt to optimize their own learning criteria, as opposed to domain-specific performance measures. By addressing this discrepancy, the proposed learning scheme approximates the given performance measure, which is discrete and makes it difficult to apply conventional optimization schemes, with a continuous and differentiable loss function which can be directly optimized. Then, a GPD algorithm is applied to optimizing this loss function.

Le conflit entre la demande de services multimédia en multidiffusion à haut débit (MBMS) et les limites en ressources radio demandent une gestion efficace de l'allocation des ressources radio (RRM) dans les réseaux 3G UMTS. À l'opposé des travaux existant dans ce domaine, cette thèse se propose de résoudre le problème de RRM dans les MBMS par une approche d’optimisation combinatoire. Le travail commence par une modélisation formelle du problème cible, désigné comme Flexible Radio Resource Management Model (F2R2M). Une analyse de la complexité et du paysage de recherche est effectuée à partir de ce modèle. Tout d’abord on montre qu'en assouplissant les contraintes de code OVSF, le problème de RRM pour les MBMS peut s'apparenter à un problème de sac à dos à choix multiples (MCKP). Une telle constatation permet de calculer les limites théoriques de la solution en résolvant le MCKP similaire. En outre, l'analyse du paysage montre que les espaces de recherche sont accidentés et constellés d'optima locaux. Sur la base de cette analyse, des algorithmes métaheuristiques sont étudiés pour résoudre le problème. Nous montrons tout d'abord que un Greedy Local Search (GLS) et un recuit simulé (SA) peuvent trouver de meilleures solutions que les approches existantes implémentées dans le système UMTS, mais la multiplicité des optima locaux rend les algorithmes très instables. Un algorithme de recherche tabou (TS) incluant une recherche à voisinage variable (VNS) est aussi développé et comparé aux autres algorithmes (GLS et SA) et aux approches actuelles du système UMTS ; les résultats de la recherche tabou dépassent toutes les autres approches. Enfin les meilleures solutions trouvées par TS sont également comparées avec les solutions théoriques générées par le solveur MCKP. On constate que les meilleures solutions trouvées par TS sont égales ou très proches des solutions optimales théoriques
The high throughputs supported by the multimedia multicast services (MBMS) and the limited radio resources result in strong requirement for efficient radio resource management (RRM) in UMTS 3G networks. This PhD thesis proposes to solve the MBMS RRM problem as a combinatorial optimization problem. The work starts with a formal modeling of the problem, named as the Flexible Radio Resource Management Model (F2R2M). An in-depth analysis of the problem complexity and the search landscape is done from the model. It is showed that, by relaxing the OVSF code constraints, the MBMS RRM problem can be approximated as a Multiple-Choice Knapsack Problem (MCKP). Such work allows us to compute the theoretical solution bounds by solving the approximated MCKP. Then the fitness landscape analysis shows that the search spaces are rough and reveal several local optimums. Based on the analysis, some metaheuristic algorithms are studied to solve the MBMS RRM problem. We first show that a Greedy Local Search (GLS) and a Simulated Annealing (SA) allow us to find better solutions than the existing approaches implemented in the UMTS system, however the results are instable due to the landscape roughness. Finally we have developed a Tabu Search (TS) mixed with a Variable Neighborhood Search (VNS) algorithm and we have compared it with GLS, SA and UMTS embedded algorithms. Not only the TS outperforms all the other approaches on several scenarios but also, by comparing it with the theoretical solution bounds generated by the MCKP solver, we observe that TS is equal or close to the theoretical optimal solutions

Wireless operators adopting third-generation (3G) technologies and those migrating from second-generation (2G) to 3G face a number of challenges related to traffic modeling, demand characterization, and performance analysis, which are key elements in the processes of designing, dimensioning and optimizing their network infrastructure. Traditional traffic modeling assumptions used for circuit-switched voice traffic no longer hold true with the convergence of voice and data over packet-switched infrastructures. Self-similar models need to be explored to appropriately account for the burstiness that packet traffic is expected to exhibit in all time scales. The task of demand characterization must include an accurate description of the multiple user profiles and service classes the network is expected to support, with their distinct geographical distributions, as well as forecasts of how the market should evolve over near and medium terms. The appropriate assessment of the quality of service becomes a more complex issue as new metrics and more intricate dependencies have to be considered when providing a varying range of services and applications that include voice, real-time, and non-real time data. All those points have to be considered by the operator to obtain a proper dimensioning, resource allocation, and rollout plan for system deployment. Additionally, any practical optimization strategy has to rely on accurate estimates of expected demand and growth in demand. In this research, we propose a practical framework to characterize the traffic offered to multimedia wireless systems that allows proper dimensioning and optimization of the system for a particular demand scenario. The framework proposed includes a methodology to quantitatively and qualitatively describe the traffic offered to multimedia wireless systems, solutions to model that traffic as practical inputs for simulation analysis, and investigation of demand-sensitive techniques for system dimensioning and performance optimization. We consider both theoretical and practical aspects related to the dimensioning of hybrid traffic (voice and data) for mobile wireless networks. We start by discussing wireless systems and traffic theory, with characterization of the main metrics and models that describe the users’ voice and data demand, presenting a review of the most recent developments in the area. The concept of service class is used to specify parameters that depend on the application type, performance requirements and traffic characteristics for a given service. Then we present the concept of “user profile,“ which ties together a given combination of service class, propagation environment and terminal type. Next, we propose a practical approach to explore the dynamics of user geographical distribution in creating multi-service, multi-class traffic layers that serve as input for network traffic simulation algorithms. The concept of quality-of-service (QoS) is also discussed, focusing on the physical layer for 3G systems. We explore system simulation as a way to dimension a system given its traffic demand characterization. In that context, we propose techniques to translate geographical distributions of user profiles into the actual number of active users of each layer, which is the key parameter to be used as input in simulations. System level simulations are executed for UMTS systems, with the purpose of validating the methodology proposed here. We complete the proposed framework by applying all elements together in the process of dimensioning and optimization of 3G wireless networks using the demand characterization for the system as input. We investigate the effects of modifying some elements in the system configuration such as network topology, radio-frequency (RF) configuration, and radio resource management (RRM) parameters, using strategies that are sensitive to traffic geographical distribution. Case study simulations are performed for Universal Mobile Telecommunications System (UMTS) networks, and multiple system variables (such as antenna tilts, pilot powers, and RRM parameters) are optimized using traffic sensitive strategies, which result in significant improvements in the overall system capacity and performance. Results obtained in the case studies, allied to a generic discussion of the trade-offs involved in the proposed framework, demonstrate the close dependence between the processes of system dimensioning and optimization with the accurate modeling of traffic demand offered to the system.
Ph. D.

En raison de leur fort potentiel applicatif dans différents domaines innovants (télésurveillance, télémédecine, etc.), les réseaux de capteurs multimédias sans fil (RCMSF) suscitent l’intérêt de nombreux travaux de recherche. En outre des contraintes soulevées par les réseaux de capteurs scalaires, les RCMSF imposent de nouvelles contraintes liées à la nature même des données capturées et manipulées. En effet, les données multimédias sont, sans aucune mesure, très largement volumineuses en comparaison aux données scalaires. De plus, leur contenu sémantique, très riche, dépendent de la qualité de l’acquisition. Dans le cadre de cette thèse, nous nous sommes intéressés à la problématique pratique d’un réseau de capteurs multimédias permettant de renseigner les automobilistes en temps réel sur les places de parking disponibles au niveau d’une ville, voire d’une agglomération. Cependant, de manière générale, les approches proposées dans nos travaux concernent tout RCMSF de surveillance.Dans ce contexte, l’objectif principal reste de maximiser la durée de vie du réseau tout en assurant une qualité perçue acceptable au niveau de la destination et ce sous un contrôle distribué (pour des raisons de passage à l’échelle évidentes). Deux axes sont à considérer : le traitement des données à la source et leur routage.Dans l'axe traitement de données, le problème principal réside dans la « qualité » des données à transmettre. De manière générale, plus la qualité est importante, plus les données sont volumineuses et conséquemment la consommation énergétique est importante et vice-versa. Il s’agit donc de trouver un équilibre qui préserve les ressources énergétiques, c-à-d. maximiser sa durée de vie tout en assurant une qualité acceptable des données envoyées. Cette dernière est le résultat d’un processus d’encodage au niveau de la source.Ainsi, nous avons d'abord abordé l'axe de traitement de données et proposé un algorithme complètement distribué qui maximise la durée de vie du réseau tout en assurant de manière optimale un équilibre entre la puissance d'encodage au niveau de la source et la qualité vidéo exigée au niveau de la destination. Contrairement aux approches existantes, notre algorithme, de nature distribuée, est assuré de trouver un tel compromis quelle que soit la configuration initiale du réseau.En raison de la complexité de ce problème, notamment dans un contexte décentralisé, les travaux antérieurs n’ont traité que la partie traitement de données indépendamment du routage. En d’autres termes, le routage a été considéré comme une entrée.Dans les travaux de recherche de cette thèse, nous avons par la suite montré que le routage impacte directement les résultats du processus de prolongation de la durée de vie du réseau. En effet, nous avons analysé le comportement de plusieurs protocoles de routage dans les RCMSF et les résultats obtenus ont mis en exergue cette influence. Nous avons donc proposé un modèle analytique intégrant de facto et le codage des données au niveau des sources et leur routage jusqu’à la station de traitement. Nous avons développé une résolution semi-distribuée de ce problème. Les résultats obtenus étaient très encourageants.Ainsi, dans la deuxième partie, une solution entièrement distribuée a été proposée, dans laquelle, l'axe de routage ne peut pas être réalisé sans les paramètres déterminés et mis à jour par l'axe de traitement de données, et inversement. La solution proposée permet: a) un routage de bout en bout avec des décisions locales dans chaque nœud capteur et b) de déterminer le nombre suffisant de chemins nécessaires pour assurer une transmission fiable de données.Pour la suite, nous avons complété nos travaux en considérant plus de contraintes réalistes, notamment la fiabilité des liens ainsi que la variation de leurs capacités (en fonction de l'énergie restante des nœuds intermédiaire). Les résultats de simulation ont montré une économie d’environs 25% de l'énergie totale
Due to their high application potential in various innovative fields (telemonitoring, telemedicine, etc.), Wireless Multimedia Sensor Networks (WMSN) arouse the interest of numerous research projects. In addition to inherent constraints of scalar sensor networks in terms of energy limitation, deployment, coverage, reliability, ..., WMSNs impose new constraints related to the captured data. Indeed, multimedia data are very voluminous in comparison to scalar data and, in addition, have a time constraint (real-time delivery). Moreover, their semantic content, very rich, is subject to different perceptions and interpretations depending on the quality of the acquisition. As a target application, this dissertation focuses detecting available car parking spaces within a large city or a metropolis. Nevertheless, the proposed approaches can be used for a wide variety of WMSN applications for surveillance purposes.In this context, the main objective remains the network lifetime maximization while ensuring an acceptable perceived quality at the destination station. The studied approaches are of a distributed nature for scalability reasons, required in WMSN. Two main axes have been targeted: data processing at source nodes and data routing toward the destination.In the data processing axis, the main problem lies in the quality of the data to be transmitted. In general, the higher the quality is, the larger the data are, and consequently more important is the energy consumption and vice versa. It is therefore a question of finding a balance that preserve the energy resources; i.e. maximize the network lifetime while ensuring an acceptable quality of the sent data. The latter is the result of an encoding process at the source level.Thus, we proposed a fully distributed algorithm that maximizes the network lifetime by optimally balancing the encoding power and the source rate at the source node in order to meet a desired visual quality at the destination station. In opposition to existing approaches, our algorithm, of distributed nature, is ensured to find such a trade-off whatever the initial network configuration is.As a second step, we focuses on data routing. In fact, due to the complexity of this problem, especially in a decentralized context, literature works have not considered jointly data processing and routing. In other words, routing was considered as a network input.In the research work of this thesis, we have subsequently shown that the routing directly impacts the results of the network lifetime maximization process. Indeed, we have analyzed the behavior of several routing protocols in WMSN and the obtained results highlighted this influence. We have therefore proposed an analytic model integrating simultaneously the encoding of data at the source nodes and their routing to the base station. We have developed a semi-distributed resolution of this problem. The results obtained were very encouraging.Thus, in the second part, a fully distributed solution was proposed, in which, the routing axis cannot be achieved without the parameters, that should be determined and updated by the data processing axis. On the other hand, the data processing axis cannot be achieved without the routing tables updated by the routing axis. The proposed solution allows: a) an end-to-end routing with local decisions at each video sensor node and b) the choose of the sufficient number of paths needed to ensure a reliable data transmission.For the rest, we have completed our work by considering more realistic constraints, in particular the dynamic reliability of the links as well as the variation of their capacities (according to the remaining energy of the intermediate nodes). The simulation results showed savings of around 25% of the total energy

Currently a growing number of users depend on the Edge Cloud Computing Paradigm in a Metro Optical Network (MON). This has led to increased competition among the Cloud Service Providers (CPs) to supply incentives for the user through guaranteed Quality of Service (QoS). If the CP fails to guarantee the QoS for the accepted request, then the user will move to another CP. Making an informed decision dynamically in such a sensitive situation demands that the CP knows the user's application requirements. The Software Defined Networking (SDN) paradigm enabled the CP to achieve such desired requirement. Therefore, a framework called Virtual Network Embedding on SDN-based Metro Optical Network (VNE-MON) is proposed in this Thesis. The use of SDN paradigm in the framework guarantees profit to the CP as well as QoS to the user.\par The design concept of the SDN control plane, raises concerns regarding its scalability, reliability and performance compared to a traditionally distributed network. To justify concerns regarding the SDN, the performance of VNE-MON and its possible dependancy on the controller location is investigated. Several strategies are proposed and formulated using Integer Linear Programming to determine the controller location in a MON. Performance results from the assessment of the VNE-MON illustrates that it is more stable compare to GMPLS-based network. It is evident that the controller location's attributes have a significant effect on the efficacy of the accepted VN request.

This thesis describes the development and setup of hybrid index structures. They are access methods for retrieval techniques in hybrid data spaces which are formed by one or more relational or normalised columns in conjunction with one non-relational or non-normalised column. Examples for these hybrid data spaces are, among others, textual data combined with geographical ones or data from enterprise content management systems. However, all non-relational data types may be stored as well as image feature vectors or comparable types. Hybrid index structures are known to function efficiently regarding retrieval operations. Unfortunately, little information is available about reorganisation operations which insert or update the row tuples. The fundamental research is mainly executed in simulation based environments. This work is written ensuing from a previous thesis that implements hybrid access structures in realistic database surroundings. During this implementation it has become obvious that retrieval works efficiently. Yet, the restructuring approaches require too much effort to be set up, e.g., in web search engine environments where several thousands of documents are inserted or modified every day. These search engines rely on relational database systems as storage backends. Hence, the setup of these access methods for hybrid data spaces is required in real world database management systems. This thesis tries to apply a systematic approach for the optimisation of the rearrangement algorithms inside realistic scenarios. Thus, a measurement and evaluation scheme is created which is repeatedly deployed to an evolving state and a model of hybrid index structures in order to optimise the regrouping algorithms to make a setup of hybrid index structures in real world information systems possible. Thus, a set of input corpora is selected which is applied to the test suite as well as an evaluation scheme. To sum up, it can be said that this thesis describes input sets, a test suite including an evaluation scheme as well as optimisation iterations on reorganisation algorithms reflecting a theoretical model framework to provide efficient reorganisations of hybrid index structures supporting multimedia search criteria.

En los últimos años, el consumo de servicios multimedia ha aumentado y se prevé que esta tendencia continúe en un futuro próximo, convirtiendo el tema de la evaluación de la Calidad de la Experiencia (QoE) en un tema muy importante para valorar el servicio de los proveedores. En este sentido, la optimización de la QoE recibe cada vez más atención ya que las soluciones actuales no han tenido en cuenta, la adaptación, la viabilidad, la rentabi-lidad y la fiabilidad. La presente memoria se centra en la caracterización, diseño, desarrollo y evaluación de diferentes aplicaciones multimedia, con el fin de optimizar la QoE. Por tanto, este trabajo investiga la influencia que la infraestructura de redes, las características de los videos y los terminales de los usuarios, presentan en la QoE de los servicios multimedia actuales en Internet. Esta tesis se basa en la investigación exhaustiva de la evaluación subjetiva y objetiva de QoE en redes heterogéneas. Los desafíos y cuestiones relacionados con el estado de la técnica y se discuten en esta disertación. En la primera fase, diseñamos una metodología de prueba para evaluar la QoE en la transmisión de video en directo y a través de plataformas de video bajo demanda en redes Wi-Fi y celulares. A partir de esta fase inicial, propondremos los problemas a investigar y las preguntas para resolver a lo largo de esta disertación. Nuestra metodología hace uso de métricas subjetivas y objetivas para evaluar la QoE percibida por los usuarios finales. Se realiza un conjunto de experimentos en laboratorio donde nuestra metodología de pruebas es aplicada. Los resultados obtenidos se recopilan y analizan para extraer las relaciones entre la Calidad de servicio (QoS) y QoE. A partir de estos resultados, se propone un mapeo de QoS-QoE que permite predecir la QoE. En la siguiente fase de la investigación, desarrollamos los algoritmos de optimización de QoE basados en la administración del sistema de red para redes Wi-Fi y celulares. Los algoritmos usan los parámetros clave que se tuvieron en cuenta para la evaluación de QoE. El objetivo de estos algorit-mos es proporcionar un sistema de gestión flexible para las redes con el ob-jetivo de lograr un equilibrio controlado entre la maximización de QoE y la eficiencia del uso de los recursos. Por último, se diseña el banco de pruebas del sistema para evaluar el rendimiento de las aplicaciones de servicios multimedia genéricos en los diferentes entornos de prueba. El banco de pruebas del sistema se basa en el enfoque de virtualización; usa los recursos compartidos de un hardware fí-sico para virtualizar todos los componentes. El banco de pruebas virtualiza-do proporciona funciones de red virtualizadas para diferentes escenarios, como Internet (las redes de distribución de contenido - CDNs) y redes inalámbricas. Por lo tanto, se adoptan protocolos livianos y mecanismos ágiles en el sistema, para proporcionar un mejor servicio a los usuarios fina-les. Los resultados de QoE son proporcionados a los proveedores de servi-cios de acuerdo con los parámetros que se definen en el proceso de la eva-luación. Como resultado hemos obtenido un sistema que presenta un servi-cio rentable como una forma factible para la evaluación de la prueba.
In recent years, multimedia services consumption has increased and it is expected that this trend will continue in the near future, becoming the evaluation of Quality of Experience (QoE) as a very important issue for assessing the quality of providers' services. In this sense, the optimization of the QoE is progressively receiving much attention considering that current solutions are not based on the adaptation, feasibility, cost-effectiveness, and reliability. The present dissertation is focused on the characterization, design, development and evaluation of different multimedia applications aimed to optimize the QoE. Therefore, this work investigates the influence that the networks infrastructure, the videos' characteristics and the users' terminals present on QoE of the current Internet multimedia services. The work is based on a comprehensive research of subjective and objective assessments in heterogeneous networks. Challenges and research questions related to the state of the art are discussed in this dissertation. In the first phase of this dissertation, we design a test methodology for assessing QoE of live video streaming and video on demand platforms to be transmitted over Wi-Fi and cellular networks. From this initial step, we will propound the related research issues and questions to solve this dissertation. Our methodology considers the use of subjective and objective metrics to evaluate the QoE perceived by end-users. A set of laboratory experiments is conducted where our proposed methodology is applied. The obtained results are gathered and analyzed to extract the relations between Quality of Service (QoS) and QoE. From the results, we propose a QoS-QoE mapping which allows predicting QoE. In the next phase of the research, we develop QoE-optimization algorithms based on network system management for Wi-Fi and cellular networks. The algorithms use the key parameters that were taken into account for QoE assessment. The goal of these algorithms is to provide a flexible management system for the networks in order to achieve the desirable trade-off between QoE maximization and resource usage efficiency. Lastly, the system testbed is designed in order to evaluate the performance of generic multimedia services applications for the different environments under test. The system testbed is based on virtualization approach; it uses the shared resources of a physical hardware to virtualize all components. The virtualized testbed provides virtualized network functions for the different scenarios such as the Internet (Content Delivery Networks - CDNs) and wireless networks. Therefore, lightweight protocols and agile mechanisms are adopted in the system to provide enhanced service to end-users. The QoE results are reported to the service providers according to the parameters defined in the evaluation process. As a result, we have obtained a cost-effective system, which is considered as a feasible way for test evaluation.
En els últims anys, el consum de serveis multimèdia ha augmentat i es preveu que aquesta tendència continue en un futur pròxim, convertitnt el tema de l'avaluació de la Qualitat d'Experiència (QoE) una tasca molt im-portant per a valorar el servei dels proveïdors. En aquest sentit, l'optimització de la QoE rep cada vegada més atenció degut a que les solucions actuals no tenen en compte, l'adaptació, la viabilitat, el rendiment i la fiabilitat. La present memòria se centra en la caracterització, disseny, desenvolupament i avaluació de diferents aplicacions multimèdia, amb la finalitat d'optimitzar la QoE. Per tant, aquest treball investiga la influència que la infraestructura de les xarxes, les característiques dels videos i els terminals dels usuaris tenen sobre la QoE dels serveis multimèdia actuals d'Iinternet. Aquesta tesi es basa en una recerca exhaustiva de l'avaluació subjectiva i objectiva de QoE en xarxes heterogènies. Els desafiaments i preguntes relacionats amb l'estat de la tècnica es discuteixen en aquesta dissertació. En la primera fase, dissenyem la metodologia de prova per a avaluar la QoE de transmissió de video en directe i de plataformes de video baix demanda en xarxes Wi-Fi i cel·lulars. A partir d'aquest primer pas, proposem els problemes de recerca relacionats i les preguntes a resoldre a través d'a-questa tesi. La nostra metodologia fa ús de mètriques subjectives i objecti-ves per a avaluar la QoE dels usuaris finals. Es realitzen un conjunt d'expe-riments en laboratori on s'aplica la nostra metodología. Els resultats obtin-guts es recopilen i analitzen per a extraure les relacions entre la QoS i la QoE. A partir d'aquests resultats, esproposa un mapatge de QoS-QoE que ens permetrà predir la QoE. En la següent fase de la recerca, desenvolupem els algoritmes d'optimi-tzació de la QoE per a l'administració de xarxes Wi-Fi i cel·lulars. Els algo-ritmes utilitzen els paràmetres clau que es van tenir en compte per a l'ava-luació de QoE. L'objectiu d'aquests algoritmes és proporcionar un sistema de gestió flexible per ales xarxes que permetrá aconseguir un equilibri con-trolat entre la maximització de la QoE i l'us eficient dels recursos. Finalment, el banc de proves del sistema està dissenyat per a avaluar el rendiment de les aplicacions de serveis multimèdia genèrics en els diferents entorns de prova. El banc de proves del sistema es basa en l'enfocament de virtualització; usa els recursos compartits d'un equip físic que virtualitza tots els components. El banc de proves virtualitzat proporciona les funcions de xarxa virtualitzades per a diferents escenaris, com Internet (les xarxes de distribució de continguts - CDNs) i xarxes sense fils. Per tant, s'adopten protocols lleugers i mecanismes àgils en el sistema per a proporcionar un millor servei als usuaris finals. Els resultats de QoE son proporcionats als proveïdors de serveis d'acord amb els paràmetres que es defineixen en el procés de l'avaluació. Com a resultat, hem obtés un sistema que presenta un servei rendible i com a viable per a l'avaluació de la prova.
Abdullah, MT. (2018). Smart Client-Server Protocol and Architecture for Adaptive Multimedia Streaming [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/103324
TESIS

Information and communication technologies are rapidly disrupting the way we interact with each other in this new connected world. The transmission of multimedia information is becoming mainstream also thanks to the availability of several commercial systems. Also in the educational environment, these novel technologies are making it possible to create and manage learning objects in ways which were unthinkable a few years back and this represents the driver for the creation of new approaches to education. Seeing the ubiquitous nature of the environments where web users interact, this work starts by analyzing the state of the art in the multimedia communication field focusing on those technologies which allow to reach a context awareness. Being able to sense the conditions where the device is used and quickly react to its changes is important in order to guarantee a flawless quality of experience. Part of this work focuses on the dynamic multimedia aspects, analyzing the latest trends in web streaming technologies and proposing a new framework for tackling some of the upcoming issues for such systems. In modern learning environments the presence of multimedia is massive. Multimedia objects are constantly created worldwide and this introduces new needs, such as the creation of proper repositories to store them and platforms to share them through the web. After analyzing the state of the art in educational technologies, focusing in particular in the content repository field, this work drafts a set of minimum requirements to follow in order to create a novel educational platform, ready to face the increasing number of needs of the future learners. In this context, the author proposes FARE, a novel architecture designed for both educators and learners in order to ease their everyday learning experience. The modularity of such a platform allows to benefit from a large set of modules, most of which present some of the multimedia aspects highlighted beforehand. Both qualitative and quantitative evaluations are presented exploiting the fully working prototype of the FARE e-learning platform deployed in the University machines. This work is meant to highlight how the advances in the multimedia field represent a clear advantage for the community of learners, creating novel ways of interacting with the learning objects and increasing the overall quality of experience.

Distributed three-dimensional (3D) graphics applications exhibit both resemblance and uniqueness in comparison with conventional streaming media applications. The resemblance relates to the large data volume and the bandwidth-limited and error-prone transmission channel. The uniqueness is due to the polygon-based representation of 3D geometric meshes and their accompanying attributes such as textures. This specific data format introduces sophisticated rendering computation to display graphics models and therefore places an additional constraint on the streaming application. The objective of this research is to provide scalable, error-resilient, and time-efficient solutions for high-quality 3D graphics applications in distributed and resource-constrained environments. Resource constraints range from rate-limited and error-prone channels to insufficient data-reception, computing, and display capabilities of client devices. Optimal resource treatment with transmission and rendering scalability is important under such circumstances. The proposed research consists of three milestones. In the first milestone, we develop a joint mesh and texture optimization framework for scalable transmission and rendering of textured 3D models. Then, we address network behaviors and develop a hybrid retransmission and error protection mechanism for the on-demand delivery of 3D models. Next, we advance from individual 3D models to 3D scene databases, which contain numerous objects interacting in one geometric space, and study joint application and transport approaches. By properly addressing the properties of 3D scenes represented in multi-resolution hierarchies, we develop a joint source and channel coding method and a multi-streaming framework for streaming the content-rich 3D scene databases toward optimized transmission and rendering scalability under resource constraints.

Video communications play a vital role in present and future wireless ad hoc networks. One of the key requirements for a successful deployment of multimedia applications in multihop wireless networks is the ability to provide an acceptable video quality, even under a highly dynamic and perhaps unfriendly (or hostile) environment (e.g., in the presence of frequent node/link failure, interference, shadowing, fading, and so forth). Existing ad hoc routing protocols work well for data communications, but are not optimized for video, which is sensitive to latency and packet loss. Moreover, traditional end system based error control mechanisms alone cannot guarantee a sustainable video quality. Conventional QoS approaches typically optimize one or more network layer metrics, but they are usually agnostic to any kind of application layer performance. Consequently, new methodologies must be explored to improve the performance of video applications in multihop wireless networks. This dissertation directly addresses this important problem area by leveraging recent advances in video coding techniques along with novel cross-layer formulations and powerful optimization techniques. We follow an application centric cross-layer approach to address multimedia service provisioning over ad hoc networks. Our research efforts show that video communications over multihop wireless networks can substantially benefit from a cross-layer design principle by factoring in application layer video quality into routing algorithmic designs at the network layer. There are three components in this investigation, namely, (1) concurrent routing, (2) path selection and rate allocation, and (3) multipath routing for multiple description video. Each component addresses one or more unique challenges that hinder video communications in multihop wireless networks. Although we expect that a cross-layer approach will be more effective than a network centric (single-layer) approach in addressing application performance, it also brings in complex problems that cannot be effectively solved using traditional methods, and thus, calls for the design of customized algorithms. In concurrent routing, we focus on issues that arise while supporting multiple concurrent video communication sessions in an ad hoc network. These sessions compete for limited network resources (such as bandwidth) while interacting with each other. Such inter-session interactions couple the performance of an individual flow with that of other flows. Applying a video centric cross-layer design principle, we model the end-to-end video distortion as a function of network layer behavior, and formulate a network-wide optimal routing problem that minimizes the total video distortion. Results based on computational experiments performed using randomly generated network topologies establish the relative efficacy and robustness of the proposed genetic algorithm based solution approach. Specifically, we demonstrate that our approach outperforms other trajectory based metaheuristic approaches as well as with conventional network centric routing algorithms such as shortest path and disjoint shortest path routing. The joint path selection and rate allocation problem considers not only selecting the best set of paths for video communication, but also, computing the optimal video encoding rate and partitioning it among the chosen set of paths. The end-to-end video distortion is modeled as a function of network layer resources by capturing the tight coupling that exists between the optimal encoding rate for each video session, the selection of paths for video transmission, and the allocation of traffic among these selected paths. This problem is formulated as a nonlinear nonconvex programming problem, for which a tight linear programming relaxation is constructed via the Reformulation-Linearization Technique (RLT). This construct is embedded within a specialized branch-and-bound algorithm to achieve global optimality. Computational experience is reported for various problem instances, and the results validate the robustness of the proposed algorithmic procedure. The results exhibit the advantage of the solution approach over the popularly used max-min rate allocation scheme. The emergence of Multiple Description (MD) coding technique offers great potential for multipath routing of video in multihop wireless networks. In studying multipath routing for MD coding, we show that MD coded video, when used in combination with multipath routing in wireless networks, has tremendous advantages over traditional layered video coding techniques. We discuss how to implement an MD video codec and formulate a cross-layer optimization problem that can find a set of optimal paths, (one for each description) such that the overall video quality at the receiver is maximized. We further devise a specialized RLT-based branch-and-bound solution procedure for the ensuing 0-1 mixed integer nonconvex optimization problem. Convergence behavior of the proposed solution procedure is observed for various network topologies and the results further demonstrate the performance advantage of the proposed cross-layer approach over non-cross-layer approaches. The scope of this research is highly interdisciplinary. It intersects video communication, networking, optimization, and algorithm design. We expect that the theoretical and algorithmic results of this investigation will serve as important building blocks in developing a comprehensive methodology for addressing complex cross-layer problems in the area of wireless ad hoc networks.
Ph. D.

Multimedia services are the killer applications on next generation convergent networks. Video contents are the most resource consuming part of a multimedia flux. Video transmission, video multicast and video conferencing services are the most popular types of video communication with increasing difficulty levels. Four main parts of the distributed cross-layer scalable multimedia services over next generation convergent networks are considered in this research work, both from the architecture and performance point of views. Firstly, we evaluate the performance of scalable multimedia transmissions over an overlay network. For that, we evaluate the performance of scalable video end-to-end transmissions over EvalSVC. It is capable of evaluating the end-to-end transmission of SVC bit-streams. The output results are both objective and subjective metrics of the video transmission. Through the interfaces with real networks and an overlay simulation platform, the transmission performance of different types of SVC scalability and AVC bit-streams on a bottle-neck and an overlay network will be evaluated. This evaluation is new because it is conducted on the end-to-end transmission of SVC contents and not on the coding performance. Next, we will study the multicast mechanism for multimedia content over an overlay network in the following part of this PhD thesis. Secondly, we tackle the problems of the distributed cross-layer scalable multimedia multicast over the next generation convergent networks. For that, we propose a new application-network cross layer multi-variable cost function for application layer multicast of multimedia delivery over convergent networks. It optimizes the variable requirements and available resources from both the application and the network layers. It can dynamically update the available resources required for reaching a particular node on the ALM's media distribution tree. Mathematical derivation and theoretical analysis have been provided for the newly proposed cost function so that it can be applied in more general cases of different contexts. An evaluation platform of an overlay network built over a convergent underlay network comprised of a simulated Internet topology and a real 4G mobile WiMAX IEEE802.16e wireless network is constructed. If multicast is the one-to-many mechanism to distribute the multimedia content, a deeper study on the many-to-many mechanism will be done in the next part of the thesis through a new architecture for video conferencing services. Thirdly, we study the distributed cross-layer scalable video conferencing services over the overlay network. For that, an enriched human perception-based distributed architecture for scalable video conferencing services is proposed with theoretical models and performance analysis. Rich theoretical models of the three different architectures: the proposed perception-based distributed architecture, the conventional centralized architecture and perception-based centralized architecture have been constructed by using queuing theory to reflect the traffic generated, transmitted and processed at the perception-based distributed leaders, the perception-based centralized top leader, and the centralized server. The performance of these three different architectures has been considered in 4 different aspects. While the distributed architecture is better than the centralized architecture for a scalable multimedia conferencing service, it brings many problems to users who are using a wireless network to participate into the conferencing service. A special solution should be found out for mobile users in the next part of the thesis. Lastly, the distributed cross-layer scalable video conferencing services over the next generation convergent network is enabled. For that, an IMS-based distributed multimedia conferencing services for Next Generation Convergent Networks is proposed. [...]

Advances in wireless technology and the growing popularity of multimedia applications have brought about a need for energy efficient and cost effective portable supercomputers capable of delivering performance beyond the capabilities of current microprocessors and DSP chips. The SIMPil architecture currently being developed at Georgia Institute of Technology is a promising candidate for this task. In order to develop applications for SIMPil, a high level language and an optimizing compiler for the language are essential. However, with the recent trend of interconnect latency becoming a major bottleneck on computer systems, optimizations focusing on reducing latency are becoming more important, especially with SIMPil, as it is highly scalable. The compiler tracks the path of data through the network and buffers data in each processor to eliminate redundant communication. With a buffer size of 5, the compiler was able to eliminate 96 percent of the redundant communication for a 9x9 convolution and 8x8 DCT algorithms. With 5x5 convolution, only 89 percent elimination was observed. In terms of performance, 106 percent speedup was observed with 9x9 convolution at buffer size of 5 while 5x5 convolution and 8x8 DCT which have a much lower number of communication showed only 101 percent speedup.

Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2012.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 85-89).
Multimedia applications are the most dominant workload in desktop and mobile computing. Such applications regularly process continuous sequences of data and can be naturally represented under the stream programming domain to take take advantage of domain-specific optimizations. Exploiting characteristics specific to multimedia programs can provide further significant impact on performance for this class of programs. This thesis identifies many multimedia applications that maintain induction variable state, which directly inhibits data parallelism for the program. We demonstrates it is essential to recognize and parallelize filters with induction variable state to enable scalable parallelization. We eliminate such state by introducing a new language construct that automatically returns the current iteration number of a target filter. This thesis also exploits the fact that multimedia applications are tolerant in the accuracy of the program output. We apply a memoization technique that exploits this tolerance and the repetitive nature of multimedia data. We provide a runtime system that automatically tunes the memoization capabilities for performance and output quality. These optimizations are implemented in the StreamIt programmming language. The necessity of parallelizing induction variable state and performance improvements and quality control of our memoization technique is demonstrated by a case study of the MPEG benchmark.
by Eric Wong.
M. Eng.

The explosive growth of the Internet during the last few years, has been propelled by the TCP/IP protocol suite and the best effort packet forwarding service. However, quality of service (QoS) is far from being a reality especially for multimedia services like video streaming and video conferencing. In the case of wireless and mobile networks, the problem becomes even worse due to the physics of the medium, resulting into further deterioration of the system performance. Goal of this dissertation is the systematic development of comprehensive models that jointly characterize the performance of transport protocols and media delivery in heterogeneous wireless networks. At the core of our novel methodology, is the use of analytical models for driving the design of media transport algorithms, so that the delivery of conversational and non-interactive multimedia data is enhanced in terms of throughput, delay, and jitter. More speciffically, we develop analytical models that characterize the throughput and goodput of the transmission control protocol (TCP) and the transmission friendly rate control (TFRC) protocol, when CBR and VBR multimedia workloads are considered. Subsequently, we enhance the transport protocol models with new parameters that capture the playback buffer performance and the expected video distortion at the receiver. In this way a complete end-to-end model for media streaming is obtained. This model is used as a basis for a new algorithm for rate-distortion optimized mode selection in video streaming appli- cations. As a next step, we extend the developed models for the aforementioned protocols, so that heterogeneous wireless networks can be accommodated. Subsequently, new algorithms are proposed in order to enhance the developed media streaming algorithms when heterogeneous wireless networks are also included. Finally, the aforementioned models and algorithms are extended for the case of concurrent multipath media transport over several hybrid wired/wireless links.

L’utilisation de plusieurs cœurs pour l’exécution des applications mobiles sera l’approche dominante dans les systèmes embarqués pour les prochaines années. Cette approche permet en générale d’augmenter les performances du système sans augmenter la vitesse de l’horloge. Grâce à cela, la consommation d’énergie reste modérée. Toutefois, la concurrence entre les tâches doit être exploitée afin d’améliorer les performances du système dans les différentes situations où l’application peut s’exécuter. Les applications multimédias comme la vidéoconférence ou la vidéo haute définition, ont de nombreuses nouvelles fonctionnalités qui nécessitent des calculs complexes par rapport aux normes précédentes de codage vidéo. Ces applications créent une charge de travail très importante sur les systèmes multiprocesseurs. L’exploitation du parallélisme pour les applications multimédia, comme le codec vidéo H.264/AVC, peut se faire à différents niveaux : au niveau de données ou bien au niveau tâches. Dans le cadre de cette thèse de doctorat, nous proposons de nouvelles solutions pour une meilleure exploitation du parallélisme dans les applications multimédia sur des systèmes embarqués ayant une architecture parallèle symétrique (ou SMP pour Symmetric Multi-Processor). Des approches innovantes pour le décodeur H.264/AVC qui traitent des composantes de couleur et des blocs de l’image en parallèle sont proposées et expérimentées
Parallel computing is currently the dominating architecture in embedded systems. Concurrency improves the performance of the system rather without increasing the clock speed which affects the power consumption of the system. However, concurrency needs to be exploited in order to improve the system performance in different applications environments. Multimedia applications (real-Time conversational services such as video conferencing, video phone, etc.) have many new features that require complex computations compared to previous video coding standards. These applications have a challenging workload for future multiprocessors. Exploiting parallelism in multimedia applications can be done at data and functional levels or using different instruction sets and architectures. In this research, we design new parallel algorithms and mapping methodologies in order to exploit the natural existence of parallelism in multimedia applications, specifically the H.264/AVC video decoder. We mainly target symmetric shared-Memory multiprocessors (SMPs) for embedded devices such as ARM Cortex-A9 multicore chips. We evaluate our novel parallel algorithms of the H.264/AVC video decoder on different levels: memory load, energy consumption, and execution time

Les communications professionnelles et critiques sont établies soit entre utilisateurs du secteur de la sûreté publique soit entre acteurs opérants des infrastructures critiques. Du fait des fortes exigences en termes de couverture, de priorité d’accès, de fiabilité et de résilience, sans oublier les services supplémentaires pour les utilisateurs professionnels, ces communications utilisent généralement les technologies PMR (Professional Mobile Radio). Vu la croissance des demandes de services, des changements importants sont attendus dans le domaine de la PMR. Les technologies PMR historiques échouent en effet à fournir des services à débits de données élevés, tels que les services vidéos et le transfert de photos. Ainsi, l’adaptation des technologies utilisées par les opérateurs commerciaux à la PMR apparaît comme une solution prometteuse. D’autre part, la prochaine génération de réseaux cellulaires prévoit une nouvelle variété d’applications et de services, dont les exigences de performances sont extrêmement hétérogènes. Ils se classent en trois groupes: enhanced Mobile BroadBand (eMBB), massive Machine-Type Communications (mMTC) et Ultra-Reliable Low Latency Communications (URLLC). Récemment, les communications critiques ont été classées dans dans la famille URLLC des cas d’usage car elles sont prioritaires par rapport aux autres types de communications dans le réseau.Dans ce contexte, nous concentrons à renforcer la couverture des réseaux radio fournissant des communications de groupe, service essentiel fournit par les technologies PMR, afin de satisfaire les besoins. Tout d’abord, on évalue la performance des transmissions unicast et multicast, c’est à dire, les transmissions Multicast/Broadcast Single Frequency Network (MBSFN) et Single-Cell Point-To-Multipoint (SC-PTM), en termes de qualité radio, d’efficacité spectrale du système et de couverture de cellules, tout en considérant des configurations MBSFN statiques. Puis, nous étudions un modèle analytique pour estimer le Signal to Interference plus Noise Ratio (SINR) dans un réseau MBSFN.En outre, nous proposons un algorithme simple de répétitions sans requête, comme alternative à l’algoritgme Hybrid Automatic Repeat re-Quest (HARQ), afin d’améliorer la couverture du réseau en présence de communications de groupe. En considérant les caractéristiques du canal radio, ainsi que les contraintes de délai de service, nous justifions que notre modèle fournit un important gain par rapport aux algorithmes de répétitions traditionnels.Enfin, on évalue le compromis entre la couverture et la capacité d’un réseau utilisant les transmissions multicast, qui évolue en fonction de la taille du cluster de stations serveuses. On formule alors un problème d’optimisation dont l’objectif est de maintenir une probabilité de blocage acceptable du système, tout en maximisant le SINR moyen du groupe d’utilisateurs. Pour chaque groupe, on choisit le cluster de cellules d’une manière dynamique, en se fondant sur la minimisation d’une fonction sous-modulaire, qui prend en compte le trafic de chaque cellule du réseau à travers certains poids, ainsi que le SINR moyen du groupe. Ces poids sont optimisés au moyen de la méthode Nelder-Mead, dans le but de diriger la probabilité de blocage vers un certain seuil. Les résultats obtenus montrent l’importance du regroupement dynamique des cellules dans l’amélioration de la capacité et la couverture du système
Business- and mission-critical communications are communications between professional users either from the public safety sector or operating critical infrastructures. Owing to special coverage, priority access, reliability and resilience requirements, as well as additional services for professional users, these communications are conveyed by Professional Mobile Radio (PMR) networks. Driven by the demand growth, significant changes are taking place in the PMR industry. The existing PMR technologies are indeed not well suited to provide high data rates mobile services like video and photo transfers; hence, the adoption of commercial technologies for mission-critical communications is gaining strong momentum. On the other hand, the next generation cellular networks are envisioned to support a large variety of applications and services with heterogeneous performance requirements, i.e., enhanced Mobile BroadBand (eMBB), massive Machine-Type Communications (mMTC) and Ultra-Reliable Low Latency Communications (URLLC). Recently, mission-critical communications have been classified in a URLLC use case family, characterized by the need to a higher priority over other communications in the networks.In this context, we focus on enhancing the coverage of wireless networks providing group communications, the main service allowed by PMR networks, taking advantage of the current technologies (e.g. Multimedia Broadcast/Multicast Service), to meet the mission-critical communications needs. First, we evaluate the performance of unicast and multicast transmission techniques, i.e., the Multicast/Broadcast Single Frequency Network (MBSFN) and Single-Cell Point-To-Multipoint (SC-PTM), in terms of radio quality, system spectral efficiency and cell coverage, assuming static MBSFN configurations. Then, we introduce an analytical model to derive an approximate closed-form formula of the Signal to Interference plus Noise Ratio (SINR) in a MBSFN network.Furthermore, we propose a simple repetition scheme without request, as an alternative to Hybrid Automatic Repeat re-Quest (HARQ), in the aim of improving the network coverage in presence of group communications. By considering the wireless channel characteristics, as well as the service delay constraints, we show that our proposed scheme provides significant gains over traditional repetition schemes.Finally, we assess the trade-off in the cluster’s size of serving cells which arises between network coverage and capacity in multi-point transmissions. We formulate an optimization problem to maintain an acceptable system blocking probability, while maximizing the average SINR of the multicast group users. For group calls, a dynamic cluster of cells is selected based on the minimization of a submodular function that takes into account the traffic in every cell through some weights and the average SINR achieved by the group users. Traffic weights are then optimized using a Nelder-Mead simplex method with the objective of tracking a blocking probability threshold. Results show the importance of dynamic clustering in improving system capacity and coverage

Submitted by Cleide Dantas (cleidedantas@ufpa.br) on 2014-07-30T16:35:12Z No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Tese_CrossLayerOptimizations.pdf: 20257418 bytes, checksum: 96559fd5f83b5752bc63ce073eb1c700 (MD5)
Rejected by Edisangela Bastos (edisangela@ufpa.br), reason: on 2014-07-30T16:37:23Z (GMT)
Submitted by Cleide Dantas (cleidedantas@ufpa.br) on 2014-07-30T16:41:46Z No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Tese_CrossLayerOptimizations.pdf: 20257418 bytes, checksum: 96559fd5f83b5752bc63ce073eb1c700 (MD5)
Approved for entry into archive by Ana Rosa Silva (arosa@ufpa.br) on 2014-09-04T14:34:02Z (GMT) No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Tese_CrossLayerOptimizations.pdf: 20257418 bytes, checksum: 96559fd5f83b5752bc63ce073eb1c700 (MD5)
Made available in DSpace on 2014-09-04T14:34:02Z (GMT). No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Tese_CrossLayerOptimizations.pdf: 20257418 bytes, checksum: 96559fd5f83b5752bc63ce073eb1c700 (MD5) Previous issue date: 2014
CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico
University of Bern
A proliferação de conteúdo multimídia bem como a demanda por novos serviços de áudio ou vídeo promoveram o desenvolvimento de uma nova era baseada em informações multimídia, o que permitiu a evolução das Redes de Sensores Sem Fio Multimídia (RSSFM) e também das redes ad-hoc desenvolvimento formadas por VANTs (FANETs). Desta forma, serviços multimídia em tempo real requerem transmissões de vídeo em tempo real com uma baixa taxa de perda de quadros, atraso fim-a-fim tolerável, para apoiar a disseminação de vídeo com qualidade de Experiência (QoE) assegurada. Desta forma, um princípio fundamental de uma abordagem ciente de QoE é a transmissão de quadros de vídeo com alta prioridade, baixa taxa de perda de pacotes, bem como baixa sobrecarga da rede, a fim de protegê-los. Além disso, o conteúdo multimídia devem ser transmitidos a partir de um determinado nó de origem para um nó de destino através de nós intermediários com alta confiabilidade em um cenário de grande escala. O serviço de roteamento deve lidar com topologias dinâmicas causadas por falha de um nó ou a mobilidade do mesmo, bem como as mudanças no canal sem fio, a fim de continuar a operar mesmo em casos de mudanças de topologia durante a transmissão multimídia. Por fim, o mapeamento da satisfação do usuário ao assistir um determinado vídeo está se tornando um requisito fundamental para a entrega de conteúdos multimídia com suporte à QoE. Com estes objetivos em mente, soluções envolvendo transmissões multimídia deve levar em conta as características de vídeo e do usuário para melhorar a entrega de vídeo com qualidade assegurada. As principais contribuições desta tese são conduzidos pela seguinte questão de pesquisa: como para fornecer distribuição de multimídia com alta eficiência energética, confiabilidade, robustez, escalabilidade e suporte à QoE em redes sem fio ad hoc. A tese aborda vários domínios de problemas com contribuições em diferentes camadas da pilha de comunicação. Na camada de aplicação, apresentamos um mecanismo de redundância de pacotes ciente de QoE para reduzir o impacto da não confiabilidade do canal sem fio, e assim prover disseminação de conteúdo multimídia em tempo real com suporte à QoE. Na camada de rede, apresentamos dois protocolos de roteamento, ou seja, o video-aware Multi-hop and multi-path hierarchical routing protocol for Efficient VIdeo transmission for static WMSN scenarios (MEVI) e o cross-layer link quality and geographical-aware beaconless OR protocol for multimedia FANET scenarios (XLinGO). Ambos os protocolos de roteamento permitem a disseminação de conteúdo multimídia com eficiência energética, confiabilidade e suporte à QoE. Isto é alcançado através da combinação de métricas de múltiplas camadas para a tomada de decisão para o roteamento de pacotes, e assim estabelecer rotas confiáveis.
The proliferation of multimedia content and the demand for new audio or video services have fostered the development of a new era based on multimedia information, which allowed the evolution of Wireless Multimedia Sensor Networks (WMSNs) and also Flying Ad-Hoc Networks (FANETs). In this way, live multimedia services require realtime video transmissions with a low frame loss rate, tolerable end-to-end delay, and jitter to support video dissemination with Quality of Experience (QoE) support. Hence, a key principle in a QoE-aware approach is the transmission of high priority frames (protect them) with a minimum packet loss ratio, as well as network overhead. Moreover, multimedia content must be transmitted from a given source to the destination via intermediate nodes with high reliability in a large scale scenario. The routing service must cope with dynamic topologies caused by node failure or mobility, as well as wireless channel changes, in order to continue to operate despite dynamic topologies during multimedia transmission. Finally, understanding user satisfaction on watching a video sequence is becoming a key requirement for delivery of multimedia content with QoE support. With this goal in mind, solutions involving multimedia transmissions must take into account the video characteristics to improve video quality delivery. The main research contributions of this thesis are driven by the research question how to provide multimedia distribution with high energy-efficiency, reliability, robustness, scalability, and QoE support over wireless ad hoc networks. The thesis addresses several problem domains with contributions on different layers of the communication stack. At the application layer, we introduce a QoE-aware packet redundancy mechanism to reduce the impact of the unreliable and lossy nature of wireless environment to disseminate live multimedia content. At the network layer, we introduce two routing protocols, namely video-aware Multi-hop and multi-path hierarchical routing protocol for Efficient VIdeo transmission for static WMSN scenarios (MEVI), and cross-layer link quality and geographical-aware beaconless OR protocol for multimedia FANET scenarios (XLinGO). Both protocols enable multimedia dissemination with energy-efficiency, reliability and QoE support. This is achieved by combining multiple cross-layer metrics for routing decision in order to establish reliable routes.

Les applications multimédias, dans les réseaux de capteurs sans fil, véhiculent des données volumineuses, qui nécessitent un taux de transmission élevé et un traitement intensif et par conséquent une consommation d'énergie importante. Transmettre efficacement ces flux hétérogènes, tout en assurant leur fiabilité et garantir les exigences de QoS, avec les ressources limitées disponibles, en particulier dans les contextes critiques, demeure un verrou scientifique ouvert. C’est pourquoi, dans le cadre de cette thèse, nous nous sommes intéressés aux aspects liés : au routage, à la tolérance aux pannes et à la gestion des interférences dans les RCMSF.Compte tenu, du très faible nombre de protocoles de routage, qui ont approché la phase expérimentale et encore moins ceux qui l’ont abordé pendant la transmission des données en temps réel, nous avons développé dans ce contexte, un protocole routage géographique baptisé GNMFT (Geographic Non-interfering Multipath Fault-tolerant),En premier lieu, nous avons amélioré le mode de sélection des nœuds, utilisé par le transfert glouton, pour faire face au problème du Minimum-local, où une fonction objective multicritères (distance, énergie et puissance de réception) relative au choix du prochain saut a été définie. Puis, nous avons introduit une phase d’optimisation des chemins construits afin d’éliminer les boucles et une approche adaptative gérant les transmissions simultanées des différents trafics.Par la suite, nous nous sommes orientés vers la tolérance aux pannes, pour assurer la fiabilité des données transmises ainsi que la connectivité du réseau. A cet effet, nous avons modélisé le nombre des paquets perdus durant la livraison des paquets et proposé deux mécanismes : un curatif pour réparer les défaillances soudaines et un préventif afin d’anticiper l’épuisement des batteries. Les deux mécanismes sont combinés avec une stratégie de basculement dynamique lors de la construction des chemins alternatives.En dernier lieu, nous avons présenté un modèle d’interférence et un troisième mécanisme qui limite les interactions entre les liens adjacents. De plus, nous avons défini également un modèle de perte de chemins dans un environnement multi-paires (source,sink) et nous avons calculé le coefficient de probabilité d’erreurs basée sur une fonction de distance qui sépare les flux de données. Une nouvelle métrique a été ajouté dans la fonction objective, relative à la somme des bruits des nœuds actifs qui interférent sur les nœuds du ForwardingSet du nœud courant.Les résultats obtenus montrent l’efficacité des approches proposées qui ont été étudiées et validées à la fois par simulation et sur un banc d’essai expérimental
Multimedia applications in WSNs convey large data (image, audio and video) that requires high transmission rate and intensive treatment and therefore high energy consumption. Effectively transmit these heterogeneous flows, while ensuring their reliability and guaranteeing QoS requirements, with the limited resources available, especially in critical contexts, remains an open scientific problem. That is why, in this thesis, we are interested in aspects related to : routing, fault tolerance and interference management in WMSNs.Given the very low number of routing protocols, that have approached the experimental phase and still less those who approached it during data transmission in real time, we developed in this context, a geographic routing protocol baptised GNMF (Geographic Non-interfering Multipath Fault-tolerant),First, we improved node's selection mode used by the greedy-forwarding, to deal with local minimum problem, where a multi-criteria objective function (distance, energy and reception power) related to next-hop choice has been defined. Then, we introduced an optimization phase of built paths to eliminate loops and an adaptive approach to manage simultaneous traffic transmissions.After that, we oriented towards fault tolerance, to ensure transmitted data reliability and network connectivity. To this end, we modeled the number of lost packets during package delivery and proposed two mechanisms. The curative is used when sudden failures occurs and the preventive to anticipate batteries depletion. Both are combined with a dynamic failover strategy during alternative paths construction.Finally, we presented an interference model and a third mechanism that limits interactions between adjacent links. In addition, we also defined a path loss model in a multipairs environment (source, sink) and computed the error probability coefficient based on a distance function that separates the data flows. A new metric has been added in the objective function, related to noise sum of the active nodes that interferes on forwarding set nodes of the current node.Obtained results show the effectiveness of the proposed approaches that have been studied and validated both by simulation and on an experimental testbed

博士
國立東華大學
電機工程學系
102
Cognitive Radio (CR) is an innovative wireless sensing technology which can improve efficiency of spectrum usage. Based on CR, Cognitive network (CN) further focus on end-to-end communications for the global optimization and policy enforcement. However, there is a lack of global control channel or middleware for cognitive optimization in CN environment. In this thesis, we proposed a Cognitive IP Multimedia Subsystem framework called CogIMS, which consists of benefits from CN and IP multimedia subsystem (IMS). In this framework, the cognitive optimization is modeled as a service over IMS architecture. By adopting the proposed service-oriented framework, the global optimization is simple to achieve under standardized architecture and the computation requirement is reduced from users to core networks. In recent years, multimedia network services have moved from a single service to rich services which integrate multiservice capabilities. If all service requests require the user to send the request by himself, this will result to a huge control function load and complex service collaboration. In order to address service interaction and reuse the service capability, 3GPP proposes a Service Capability Interaction Manager (SCIM), which can provide service capabilities invocation and service interaction management between Application Servers (ASs) and Serving-Call Session Control Function (S-CSCF). However, its architecture may cause joint and cooperation problems between the different service providers. In this thesis, we propose a scalable service capability interaction framework called ServiceChord that can process multiple service capabilities with different ASs and reduce the call set-up delay while communicating with the S-CSCF. The Chord DHT technique is used to improve the framework, leading to a reduction of message redundancy on the S-CSCF while achieving an efficient service capability interaction, and providing scalability for CogIMS services and CogASs.

Chow Lik Hang.
Thesis (M.Phil.)--Chinese University of Hong Kong, 2003.
Includes bibliographical references (leaves 100-102).
Abstracts in English and Chinese.
Chapter 1 --- Introduction --- p.1
Chapter 1.1 --- Multimedia Streaming Background --- p.2
Chapter 1.2 --- Streaming over the Internet --- p.2
Chapter 1.3 --- Traditional Approaches --- p.4
Chapter 1.4 --- Document Road-map --- p.6
Chapter 2 --- Related Work --- p.8
Chapter 3 --- Our Multi-path Streaming Approach --- p.11
Chapter 3.1 --- Potential Benefits --- p.14
Chapter 3.2 --- Performance Metrics --- p.15
Chapter 3.3 --- Visual Quality of Data --- p.15
Chapter 3.3.1 --- Experiment: Effect of Correlated Bursty Losses on Video Quality --- p.16
Chapter 4 --- Performance Evaluation using Gilbert Model --- p.18
Chapter 4.1 --- Mathematical analysis --- p.18
Chapter 4.1.1 --- Model --- p.19
Chapter 4.1.2 --- Performance Analysis of SP vs. Multi-path Streaming (without FEC) --- p.20
Chapter 4.1.3 --- Performance Analysis of SP vs. Multi-path Streaming (with FEC) --- p.26
Chapter 4.2 --- Analytical Model Based Evaluation --- p.32
Chapter 5 --- Functional Gilbert Model and Optimization --- p.45
Chapter 5.1 --- Functional Gilbert Model --- p.45
Chapter 5.2 --- Optimal Traffic Splitting --- p.48
Chapter 5.2.1 --- Optimization Based on Achieved Loss Rate --- p.49
Chapter 5.2.2 --- Optimization Based on Lag-1 Autocorrelation --- p.54
Chapter 5.3 --- Experiments --- p.58
Chapter 5.3.1 --- Type A Experiment: Without an Erasure Code --- p.59
Chapter 5.3.2 --- Type B Experiment: with an Erasure Code --- p.62
Chapter 6 --- NS Simulations --- p.68
Chapter 6.1 --- Simulation Setup --- p.68
Chapter 6.2 --- Simulation Result --- p.70
Chapter 7 --- Quantization of Traffic Splitting Vector --- p.80
Chapter 8 --- Prototype Implementation and Experiments --- p.86
Chapter 8.1 --- Multi-path Streaming Prototype --- p.86
Chapter 8.2 --- Experiments --- p.87
Chapter 9 --- Other Design Issues and Considerations --- p.93
Chapter 9.1 --- Requirements and Overheads --- p.93
Chapter 9.2 --- Share Point of Congestion (SPOC) --- p.96
Chapter 10 --- Conclusion --- p.98
Bibliography --- p.100

The general use of Internet access and user equipments, such as smartphones, tablets and personal computers, is creating a new wave of video content consumption. In the past two decades, the Television broadcasting industry went through several evolutions and changes, evolving from analog to digital distribution, standard definition to high definition TV-channels, form the IPTV method of distribution to the latest set of technologies in content distribution, OTT. The IPTV technology introduced features that changed the passive role of the client to an active one, revolutionizing the way users consume TV content. Thus, the clients’ habits started to shape the services offered, leading to an anywhere and anytime offer of video content. OTT video delivery is a reflection of those habits, meeting the users’ desire, introducing several benefits discussed in this work over the previous technologies. However, the OTT type of delivery poses several challenges in terms of scalability and threatens the Telecommunications Operators business model, because OTT companies use the Telcos infrastructure for free. Consequently, Telecommunications Operators must prepare their infrastructure for future demand while offering new services to stay competitive. This dissertation aims to contribute with insights on what infrastructure changes a Telecommunications Operator must perform with a proposed bandwidth forecasting model. The results obtained from the forecast model paved the way to the proposed video content delivery method, which aims to improve users’ perceived Quality-of-Experience while optimizing load balancing decisions. The overall results show an improvement of users’ experience using the proposed method.
A generalização do acesso à Internet e equipamentos pessoais como smartphones, tablets e computadores pessoais, está a criar uma nova onda de consumo de conteúdos multimedia. Nas ultimas duas décadas, a indústria de transmissão de Televisão atravessou várias evoluções e alterações, evoluindo da distribuição analógica para a digital, de canais de Televisão de definição padrão para alta definição, do método de distribuição IPTV, até ao último conjunto de tecnologias na distribuição de conteúdos, OTT. A tecnologia IPTV introduziu novas funcionalidades que mudaram o papel passivo do cliente para um papel activo, revolucionando a forma como os utilizadores consumem conteúdos televisivos. Assim, os hábitos dos clientes começaram a moldar os serviços oferecidos, levando à oferta de consumo de conteúdos em qualquer lugar e em qualquer altura. A entrega de vídeo OTT é um reflexo destes hábitos, indo ao encontro dos desejos dos utilizadores, que introduz inúmeras vantagens sobre outras tecnologias discutidas neste trabalho. No entanto, a entrega de conteúdos OTT cria diversos problemas de escalabilidade e ameaça o modelo de negócio das Operadoras de Telecomunicações, porque os fornecedores de serviço OTT usam a infraestrutura das mesmas sem quaisquer custos. Consequentemente, os Operadores de Telecomunicações devem preparar a sua infraestrutura para o consumo futuro ao mesmo tempo que oferecem novos serviços para se manterem competitivos. Esta dissertação visa contribuir com conhecimento sobre quais alterações uma Operadora de Telecomunicações deve executar com o modelo de previsão de largura de banda proposto. Os resultados obtidos abriram caminho para o método de entrega de conteúdos multimedia proposto, que visa ao melhoramento da qualidade de experiência do utilizador ao mesmo tempo que se optimiza o processo de balanceamento de carga. No geral os testes confirmam uma melhoria na qualidade de experiência do utilizador usando o método proposto.
Mestrado em Engenharia de Computadores e Telemática

Rateless codes, also known as fountain codes, are a class of erasure error-control codes that are particularly well suited for broadcast/multicast systems. Raptor codes, as a particularly successful implementation of digital fountain codes, have been used as the application layer forward error correction (FEC) codes in the third generation partnership program (3GPP) Multimedia Broadcast and Multicast Services (MBMS) standard. However, the application of rateless codes to wireless multimedia broadcast/multicast communications has yet to overcome two major challenges: first, wireless multimedia communications usually has stringent delay requirements. In addition, multimedia multicast has to overcome heterogeneity. To meet these challenges, we propose a rateless code design that takes the layered nature of source traffic as well as the varying quality of transmission channels into account. A convex optimization framework for the application of unequal error protection (UEP) rateless codes to synchronous and asynchronous multimedia multicast to heterogeneous users is proposed. A second thread of the thesis addresses the noisy, bursty and time- varying nature of wireless communication channels that challenge the assumption of erasure channels often used for the wired internet. In order to meet this challenge, the optimal combination of application-layer rateless code and physical layer FEC code rates in time-varying fading channels is investigated. The performance of rateless codes in hybrid error-erasure channels with memory is then studied, and a cross-layer decoding method is proposed to improve decoding performance and complexity.
Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2011-06-12 16:26:36.136

碩士
大同大學
資訊工程學系(所)
97
Multi-processor system-on-a-chip (MPSoC) architectures have received a lot of attention in the past years. As the gradually growing computation amount of data, using single core based systems is not feasible any more to meet the demands of embedded applications. In order to solving this problem, it’s a better choice to use multiple processor. By adopting multiple processor architectures, several works can be executed parallel and fast. A couple of key optimization techniques exist and have been demonstrated on SoC architecture. This paper presents these techniques and their effect on a multi-core environment and brings forward improvements. These techniques allow for optimization of memory space, reduction of processing time simultaneously. We choose a multimedia application as the optimization target. After the optimization of multimedia application architecture, we further improve the execution performance by assigning its works to dual-core and quad-core processor with using OpenMP techniques. It leads the execution performance to 73% and 201% improvement than single-core enviroonment.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1998.
Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 148-161).

碩士
銘傳大學
電腦與通訊工程學系碩士班
100
Multimedia Content Delivery Networks (CDN) is used to improve the performance and reliability on Internet. In CDN architecture, the multimedia contents are replicated from the origin server to replica servers in order to improve the performance and minimize the use of network bandwidth. Efficient placing the multimedia contents in CDN is a challenging problem. There are five factors that can be used to determine the placement of multimedia contents, they are bandwidth availability, connection availability, storage availability, CPU availability, and memory availability. In this thesis, a Particle Swarm Optimization (PSO) algorithm is adopted to solve this issue. PSO algorithm uses these five different input parameters as different dimensions. In this five dimension space, PSO algorithm can find out the global optimal solution. With this global optimal solution, it is the most appropriate replica server that should place the multimedia content. The simulation results show that the PSO algorithm can achieve a better performance than other algorithms.

碩士
國立東華大學
電機工程學系
99
In recent years, the wireless terminal equipments growth rapidly resulted in wireless resource shortage. How to satisfy different service requirements and service quality under limited wireless resource for users to reach wireless resource allocation optimization becomes an important issue. Cognitive Radio is (CR) an innovative wireless sensing technology can improve efficiency of spectrum usage. Based on CR, Cognitive network (CN) is further focus on end-to-end communications for the global optimization and strategic decision. However, there is a lack of global control channel or middleware for cognitive optimization in CN environment. In next generation network (NGN), IP multimedia subsystem (IMS) provides a common framework with various services for users in heterogeneous network environments. Users can access to the IMS to acquire for services through various access technologies. In this thesis, we proposed Service Oriented Cognitive IP Multimedia Subsystem architecture called CogIMS, which consists of benefits from CN and IMS. CogIMS can satisfy user’s demand and optimize network efficiency by adopting the proposed service-oriented framework for conducting end-to-end communication and services. In order to realize service-oriented optimization, we use multi-objective optimization to reduce the complexity for implementing cognitive networks. The simulation results show that the proposed CogIMS can achieve the optimization of wireless resource allocation with different network situations.

abstract: The rapid escalation of technology and the widespread emergence of modern technological equipments have resulted in the generation of humongous amounts of digital data (in the form of images, videos and text). This has expanded the possibility of solving real world problems using computational learning frameworks. However, while gathering a large amount of data is cheap and easy, annotating them with class labels is an expensive process in terms of time, labor and human expertise. This has paved the way for research in the field of active learning. Such algorithms automatically select the salient and exemplar instances from large quantities of unlabeled data and are effective in reducing human labeling effort in inducing classification models. To utilize the possible presence of multiple labeling agents, there have been attempts towards a batch mode form of active learning, where a batch of data instances is selected simultaneously for manual annotation. This dissertation is aimed at the development of novel batch mode active learning algorithms to reduce manual effort in training classification models in real world multimedia pattern recognition applications. Four major contributions are proposed in this work: $(i)$ a framework for dynamic batch mode active learning, where the batch size and the specific data instances to be queried are selected adaptively through a single formulation, based on the complexity of the data stream in question, $(ii)$ a batch mode active learning strategy for fuzzy label classification problems, where there is an inherent imprecision and vagueness in the class label definitions, $(iii)$ batch mode active learning algorithms based on convex relaxations of an NP-hard integer quadratic programming (IQP) problem, with guaranteed bounds on the solution quality and $(iv)$ an active matrix completion algorithm and its application to solve several variants of the active learning problem (transductive active learning, multi-label active learning, active feature acquisition and active learning for regression). These contributions are validated on the face recognition and facial expression recognition problems (which are commonly encountered in real world applications like robotics, security and assistive technology for the blind and the visually impaired) and also on collaborative filtering applications like movie recommendation.
Dissertation/Thesis
Ph.D. Computer Science 2013

碩士
國立成功大學
資訊工程學系研究所
84
We develop and analyze a class of admission control algorithmsbased on the co ncept of ``reward optimization''for multimedia servers designed to provide on- demand servicesto clients in an environment where workload characteristics of clientscan change dynamically.These admission control algorithms are developed based on twostrategic choices, namely, ``deterministic'' or ``best-effort'' r egardingquality of service (QoS) control, and ``priority-reservation'' or ``no priority-reservation'' regarding reservation control. We first formulate the design of admission control algorithms forreal-time multimedia servers as a r eward optimization problem, with``reward'' referring to the ``value'' which th e system receives afterservicing prioritized clients based on the QoS levels r equested and delivered. Then, we derive close-form solutions for the reward ra te whichthe system can obtain under these admission control algorithms andvali date the analytical results with a simulated variable bit rate(VBR) video serv er. A physical interpretation is given forthe best combination of strategic ch oices under whichthe system can obtain the best reward as a function ofworkloa d characteristics of the clients.We discover that the reward rate which the sy stem can obtain under ourproposed admission control algorithms is much higher than that undertraditional admission control algorithms which do not consider reward optimization. For the simulation results, we also compare other perform ance metrics,including the number of clients which the system can admit, the u tilization of the server, and the effective QoS delivered to the clients.

碩士
國立中正大學
資訊工程所
96
Nowadays, most embedded systems support different kinds of multimedia applic ations and real-time tasks. The computational requirements of multimedia applications are usually very expensive and embedded systems may not be powerful enough to support multiple multimedia applications. As a result, real-time tasks will miss their deadlines and thus get rejected. To solve this problem, we propose a Dynamic WorkloadShaping(DWS) methodby extending aDynamicBufferingTechnique to multiple multimedia tasks andintegrating two optimization methods. The scheduling method dynamically changes the playout buffer fill level of multimedia applications to periodically free up sufficient processor bandwidth to support other concurrently running real-time applications. The original scheduling method can only guarantee that theplayout devices can always read outdataitemsfrom theplayout buffers, but cannot guarantee any QoS. Two algorithms are proposed in this work to optimize and guarantee the quality of multimedia applications. The first algorithm optimizes the multimedia playout device rates, in terms of frames per second, while allowing a real-time task to meet its deadline. The second algorithm optimizes the processor bandwidth allocations,interms of MHz, given to each of the multimedia tasks, while allowing a real-time task to meet its deadline. We use several examples to illustrate how both algorithms can be applied to real systems.

博士
國立暨南國際大學
資訊工程學系
96
In a Digital Library/Museum system, media presentation is an important task that directly affects a contact with the user. How to transmit several media items effectively and accurately to reduce the user's waiting time (wait latency) is one of the main concern at this stage. Similar to a television broadcast, automatically and continuously present the media items in sequence enables the user to avoid downloading, and playing back repeatedly. This automatic presentation of the media is referred to as TV-like presentation and is user-preferred. According to this presentation, for reducing the wait latency, the server should be capable of deciding a proper order of the media items and effectively delivering. This dissertation focuses on optimizing the media transmission and playback technology by using the Prefetch mechanism under complicated environments. Aiming to seek the shortest time to arrange the sequence of the media items, this research applies the conventional scheduling algorithm to the multimedia file transmission and playback problem. First, an extensive discussion of the relevant issues related to reducing the presentation latency under various different types of constraints. Problems with object precedence constraints, objective due dates, and client-side buffer limitations are considered and solved with branch and bound solutions. In particular, the player-side buffer is “processing time dependence” and is different from the conventional item-based intermediate buffer constraints discussed in previous flow-shop scheduling problems; this new problem is then proven to be strongly NP-hard. In addition, various heuristic algorithms are proposed to approximate the optimal solution for the problem with the buffer constraint. From theoretical and practical 3G transmission experiments, a better scheduling algorithm can significantly reduce the wait latency efficiently.

博士
國立中正大學
資訊工程所
98
With the advances in the video coding algorithms, there is more and more demanded computational cost for the multimedia applications. The high complexity of multimedia applications enables processor architectures using parallelism with multithreading technique for performance improvement. In this dissertation, we propose a hardware/software optimization methodology on the UniCore visoMT processor, which is a multi-threading and multi-core processor with a main thread and several accelerative threads executed at the same time to realize multimedia codec. In the hardware optimization, we design fine-grained generic media functional units (GMFUs) as the physical accelerative threads. Since the multimedia applications typically process narrow data types (for example, 8-bit per pixel), GMFU is designed as SIMD architecture for providing the data-level parallelism. In addition, the scalability of GMFU makes the thread-level parallelism be feasible. We also provide the firmware libraries of GMFU for users to develop their multimedia applications quickly and easily. In the software optimization, we analyze the multimedia software to extract parallelisms and complex computations and speed them up with the proposed GMFUs. UniCore visoMT processor provides the applications programming interface (API) for programmers to utilize the firmware libraries of GMFU. In addition to utilize the firmware libraries, we also optimize the multimedia software algorithmically and modify the software architecture to improve the performance. With the proposed techniques, we evaluate the UniCore visoMT processor using a broad range of media kernels and show that its performance achieves encoding/decoding 30/70 frames per second of MPEG-4 video Codec and encoding 17 frames per second of H.264 video in CIF (352x288) resolution at the clock rate of 200 MHz.

碩士
明新科技大學
土木工程與環境資源管理研究所
100
Indoor air quality is good or bad in the classroom for students to attend class or learn a considerable impact, the student about 10 hours a day in the campus, of which about 8 hours of time in the classroom, therefore indoor air quality healthy learning environment for students should be of considerable attention in the classroom. The purpose of this study was to explore the classroom concentrations of CO2 (Carbon Dioxide), in Situation 1: Air-conditioning is not open, doors and windows open to the indoor air does not produce convection. Situation 2: air-conditioned open the doors and windows closed both case, the best time to exhaust. In this study MingHsin University of Science and Technology Department of Civil Engineering Hall of wireless communication and multimedia teaching classrooms for the experimental site, the classroom volume is 412.21m3, this study was conducted by using the CDU440 Carbon Dioxide Monitor detector, a data record every 5 minutes for 3 hours, to detect the concentration of CO2 in the classroom, through references to CO2 emissions estimate for the hardware in the classroom, final the specification in accordance with the indoor air quality, CO2 concentration recommended value of 600ppm as the evaluation criteria. The results is, in Situation 1: there are 32 people, CO2 concentration in class after 30 minutes, reach 650 ppm, has exceeded the recommended value should be opened to the exhaust fan to exhaust. In Situation 1: there are 24 people, CO2 concentration in class after 10 minutes, reach 640 ppm, has exceeded the recommended value should be opened to the external air circulation；The estimation results of the theoretical value of the carbon dioxide emissions in the human respiratory, per minute exhaled 0.603 g of CO2, measured values in the classroom for 40 people on average0.0204 (g/m3/min), the theoretical estimated value of 0.0585 (g/m3/min), 36 people when the measured value the average .0070 (g/m3/min), theoretical estimated value of .0527 (g/m3/min).

博士
國立臺灣大學
電機工程學研究所
89
Low-power, high-throughput and small-area products with a short development schedule are general design aims in the 3C (computer, communication and consumption) era. The IAs ( Information Appliances) are the typical applications which expand the desktop information applications into everywhere as small as it can exist. It is recommended that hardward and software design issues should be considered simultaneously, co-operatively and interactively. Especially for the multiple design objectives (e.g. low power, high throughput, small area, etc.), the MO ( Multicriteria Optimization) technique can be applied to make a trade-off decision. FACE/FACE++ (Fine-tuned Architecture Codesign environment/Version 1 & 2) was developed to achieve these goals, including the organization of a design flow, a frequency-driven information analyzer, compiler techniques (code generator and instruction optimization) and a hierarchical object design library. This thesis explores the design space of a retargetable compiler and a fine-tuned hardware emulator, which combine both software and hardware reprogrammability. The environment, FACE, we have developed allows us to quickly move the functions between software and hardware in a state of flux. It generates the ASIP (Application-Specific Integrated Processor) and a compiler for the new ASIP architecture. The experiment demonstrates the efficiency in ASIP design of FACE. As for the applications in the new century, many information appliance products will replace traditional electronic appliances to help people in smart, efficient and low-cost ways. These successful products must be capable of communicating multimedia information, which is embedded into the electronic appliances with high integration, innovation, and power-throughput tradeoff. We have developed a codesign procedure based on FACE, called FACE++, to analyze, compare and emulate the multimedia communication applications to find the candidate implementations under different criteria. The results show that in general, memory technology dominates the optimal tradeoff and ALU improvements impact greatly on particular applications. This thesis combines lots of technologies into the design environment to evaluate the tradeoff among multiple objectives and solves the evaluation problem efficiently and effectively.

博士
國立臺灣科技大學
電機工程系
104
Since Internet Protocol (IP) is the most important protocol in Next Generation Networks (NGNs), the 3rd Generation Partnership Project (3GPP) utilizes Session Initiation Protocol (SIP) based on IP as the base and major protocol for negotiating sessions in IP Multimedia Subsystem (IMS). Unlike traditional circuit-switched network, in IMS, the media traffic flow and signaling are both delivered through IP transport, which is known as packet-switched network. The media traffic affects the signaling efficiency in the core network, due to traffic/packet collisions and the best effort packet delivery mechanism in ehternet. In addition, various emergency services based on wireless sensor network technologies have recently been proposed. However, the success of these services/networks is inherently limited by geographical restrictions, and their need to be deployed in advance. In order to overcome these problems, this thesis proposes an application level approach to enhance the service coverage and availability of emergency services, and a novel path-migration mechanism for enhancing the traffic efficiency in integrated Cooperative Emergency IMS (CE-IMS). Specifically, this thesis augments these services with All-IP network infrastructure based on an IP Multimedia Subsystem. Furthermore, the integration of IMS Emergency Services architecture with Cooperative Network technology is proposed to provide ubiquitous emergency services. The primary problems of cooperation between heterogeneous networks and IMSs are also investigated. The discussion and the experimental results of performance of the proposed CE-IMS are all examined. The simulation results also show that interference and traffic collisions can be reduced by applying the proposed path-migration mechanism, and the signaling efficiency in the core network can be improved with higher system capability and the voice quality is also guaranteed when making emergency calls. Moreover, the application of the Internet of Things (IoT) has become an emerging business in recent years. The most important concept of next-generation networks for providing a common global IT platform is combining seamless networks and networked things, objects or sensors. Thus, wireless body area networks (WBANs) are becoming mature with the widespread usage of the IoT. In order to support WBANs, the well-designed platform, scenario and emergency service flow for IoT with WBANs are necessary elements due to the sensors in a WBAN being related to wearer's life. The sensors on the body could detect bioinformatic information and medical signals, such as heartbeat and blood pressure. Thus, the integration of the IoT and network communication in daily life is important. However, there is not only a lack of common fabric for integrating the IoT with the Internet, there is also no emergency call process in the current network communication environment. In order to address such problems, the prototype integration of the IoT and an emergency call process is discussed. Then, a bootstrap platform is proposed to provide the discussion of open challenges and solutions for deploying the IoT in the Internet, and the emergency communication system is analyzed using a 3GPP IP multimedia subsystem service. Finally, the prototype for supporting a WBAN with an emergency service is also addressed, and the performance results are useful to service providers and network operators, helping them estimate their migration to the IoT by referring to this experiment.

This thesis describes the development and setup of hybrid index structures. They are access methods for retrieval techniques in hybrid data spaces which are formed by one or more relational or normalised columns in conjunction with one non-relational or non-normalised column. Examples for these hybrid data spaces are, among others, textual data combined with geographical ones or data from enterprise content management systems. However, all non-relational data types may be stored as well as image feature vectors or comparable types. Hybrid index structures are known to function efficiently regarding retrieval operations. Unfortunately, little information is available about reorganisation operations which insert or update the row tuples. The fundamental research is mainly executed in simulation based environments. This work is written ensuing from a previous thesis that implements hybrid access structures in realistic database surroundings. During this implementation it has become obvious that retrieval works efficiently. Yet, the restructuring approaches require too much effort to be set up, e.g., in web search engine environments where several thousands of documents are inserted or modified every day. These search engines rely on relational database systems as storage backends. Hence, the setup of these access methods for hybrid data spaces is required in real world database management systems. This thesis tries to apply a systematic approach for the optimisation of the rearrangement algorithms inside realistic scenarios. Thus, a measurement and evaluation scheme is created which is repeatedly deployed to an evolving state and a model of hybrid index structures in order to optimise the regrouping algorithms to make a setup of hybrid index structures in real world information systems possible. Thus, a set of input corpora is selected which is applied to the test suite as well as an evaluation scheme. To sum up, it can be said that this thesis describes input sets, a test suite including an evaluation scheme as well as optimisation iterations on reorganisation algorithms reflecting a theoretical model framework to provide efficient reorganisations of hybrid index structures supporting multimedia search criteria.

The Multi-Level Computing Architecture (MLCA) is a multiprocessor system-on-chip architecture designed for multimedia applications. It provides a programming model that simplifies the process of writing parallel applications by eliminating the need for explicit synchronization. However, developers must still invest effort to design applications that fully exploit the MLCA’s multiprocessing capabilities. We present a set of compiler techniques to streamline the process of developing applications for the MLCA. We present an algorithm to automatically partition a sequential application into tasks that can be executed in parallel. We also present code generation algorithms to translate annotated, sequential C code to the MLCA’s programming model. We provide an experimental evaluation of these techniques, performed with a prototype compiler based upon the open-source ORC compiler and integrated with the MLCA Optimizing Compiler. This evaluation shows that the performance of automatically generated code compares favourably to that of manually written code.

Η δημοφιλής τάση για ασύρματη επικοινωνία συμβολίζει την ανάγκη μας για ανεξαρτησία και ευελιξία. Σε συνδυασμό με την έμφυτη ανάγκη του ανθρώπου για επικοινωνία, ανταλλαγή σκέψεων και εμπειριών, η τάση αυτή περιλαμβάνει τα τελευταία χρόνια την απαίτηση για ασύρματα πολυμέσα. Η ασύρματη επικοινωνία πολυμέσων έχει τη δυνατότητα να αλλάξει πολλές από τις παραμέτρους της καθημερινότητάς μας. Η πληθώρα των εφαρμογών που προσφέρονται περιορίζεται μόνο από τη δυνατότητα του ασύρματου καναλιού να διαθέσει πόρους, ενέργεια και χρόνο. Η χωρητικότητα του ασύρματου διαύλου επικοινωνίας είναι το κλειδί και η έρευνα εστιάζει στη βελτιστοποίηση της μετάδοσης εφαρμογών ροής βίντεο, αξιοποιώντας όσο το δυνατόν καλύτερα το ασύρματο κανάλι. Στα πλαίσια της διατριβής προτείνεται μια νέα αναλυτική μέθοδος δημιουργίας συνθετικών βίντεο ροών MPEG-4 που μεταδίδονται σε ένα ετερογενές ασύρματο περιβάλλον. Η κίνηση που δημιουργούν πολλαπλές ροές βίντεο μοντελοποιείται από μια αλυσίδα Μαρκόφ δύο διαστάσεων. Προκειμένου να διατηρηθεί η ποιότητα παρεχόμενης υπηρεσίας, προτείνεται ένας μηχανισμός μεταγωγής των ροών βίντεο με βάση τις πιθανότητες μετάβασης καταστάσεων του μοντέλου. Επιπλέον, παρουσιάζεται ένα αναλυτικό μοντέλο εκτίμησης της αλλοίωσης της αντιλαμβανόμενης ποιότητας κωδικοποιημένου βίντεο Η.264/AVC. Το μοντέλο αυτό αξιοποιεί τα ιδιαίτερα χαρακτηριστικά της κωδικοποίησης Η.264/AVC και ιδιαίτερα τις αλληλεξαρτήσεις μεταξύ των πλαισίων βίντεο, ώστε να εξασφαλίσει την ακριβή εκτίμηση της συνολικής αλλοίωσης. Η μελέτη περιλαμβάνει συνδυασμούς πιθανών σφαλμάτων στο ασύρματο κανάλι ώστε να αποδειχθεί η δυνατότητα εκτίμησης της αλλοίωσης από το προτεινόμενο μοντέλο σε κάθε συνθήκη του δικτύου. Στη συνέχεια, αναλύονται μηχανισμοί προγραμματισμού της μετάδοσης πακέτων βίντεο και διαφοροποίησης της σημαντικότητας των πακέτων με βάση τον αντίκτυπό τους στη συνολική αλλοίωση του βίντεο. Η σημασία κάθε πακέτου εκτιμάται από το προτεινόμενο μοντέλο εκτίμησης της αλλοίωσης. Εξασφαλίζεται, έτσι, η προσαρμογή του ρυθμού μετάδοσης του βίντεο στις διαφορετικές συνθήκες του δικτύου. Η αξιολόγηση του μοντέλου εκτίμησης επεκτείνεται ώστε να συμπεριλάβει το πρωτόκολλο 802.11e το οποίο διαθέτει μηχανισμούς διασφάλισης της ποιότητας υπηρεσίας. Τέλος, παρουσιάζονται αλγόριθμοι επιλογής βέλτιστων διαδρομών μεταξύ πηγής και παραλήπτη σε ασύρματα δίκτυα αισθητήρων πολυμέσων. Προτείνεται μια επέκταση του πρωτοκόλλου LEACH ώστε να είναι δυνατή η δρομολόγηση πακέτων βίντεο από πολλαπλές διαδρομές με στόχο τη μικρότερη δυνατή αλλοίωση της ποιότητας και την εξοικονόμηση ενέργειας στους κόμβους.
The popular trend for wireless communications marks our demand for freedom and flexibility. Recently, the inherent human need for communication and sharing of ideas and experiences has been enriched with the need for wireless multimedia. Wireless multimedia communications have the ability to alter several parameters of our daily life. The great variety of offered services is only limited by the available resources of the wireless medium. The key is the capacity of the wireless medium and this research focuses on optimizing the video traffic while fully utilizing the wireless channel. This thesis introduces a new analytical method of creating synthetic MPEG-4 video streams over heterogeneous wireless networks. The aggregate video traffic generated by multiple wireless video sources is modeled by a two-dimensional Marcov chain. In order to achieve a guaranteed quality of service, a handoff mechanism is proposed based on the state transition probabilities of the statistical model. Moreover, an analytical distortion prediction model for H264/AVC coded video is presented. In order to precisely estimate the received video distortion, the proposed model utilizes the inherent characteristics of H.264/AVC coding and in particular, the correlation among neighboring video frames. The distortion prediction model’s ability to estimate the received video distortion is studied under different channel conditions. Hence, complex combinations of errors due to the wireless channel are considered in this thesis. Moreover, an analysis of video packet scheduling algorithms and traffic differentiation schemes, based on the video packet’s impact to the overall video distortion, is included. The importance of each video packet is estimated by the proposed distortion prediction model. Therefore, the transmission rate of a video source can be adapted to the varying wireless channel conditions. The evaluation of the model has been extended in order to include the study of the 802.11e standard, which incorporates mechanisms for guaranteeing quality of service. Finally, the thesis proposes algorithms for optimised video multipath routing in wireless multimedia sensor networks. A proposed extension of the LEACH protocol is implemented that ensures multipath video packet routing. Thus, significantly improved received video distortion and power consumption in the wireless sensor nodes, is achieved.

碩士
臺灣大學
電機工程學研究所
95
Platform Architect, one of the CoWare corporation’s electronic system level design tool, can be used to build and simulate various virtual System-on-chip(SoC) platforms. This thesis explores a jpeg2000 decoder on SoC platforms that are based on Transaction-Level Modeling (TLM). We improve the system execution time by using both methods of hardware and software optimizations. In the hardware part, we refine the communication architecture and make use of master ports on the SoC platform to reduce communication time. In the software part, we enable caches, compiler options and rewriting parts of the source code for the different architectures to improve the system execution time. Finally we can get the best solution from exploring the hardware and software optimizations.

碩士
國立清華大學
資訊工程學系
95
In the high performance and low power multimedia embedded system design, the VLIW-based embedded DSPs which need compiler to exploit the ILP become popular and play an important role today. For this reason, we need optimizing embedded DSP compilers that can generate capable code with efficiency in terms of performance, power, area, and productivity in order to use embedded DSPs effectively. With exploiting the specific hardware feature of DSPs, the embedded DSP compilers can avoid designer to optimize applications on hand which will increase the time-to-market without lost performance. In this paper, we show that using the proposed post compilation framework, we can exploit hardware specific features of DSPs with runtime information to optimize the results of other compilers. In the simulation results, we demonstrate that the proposed framework can optimize the programs optimized by another compiler Blackfin GCC 3.4 and VDSP++ 4.5 with optimization level 3 and get additional 17.56% and 8.8% performance on average. We also can get additional 5.8% performance on average of the real multimedia program H.264 which is optimized by hand-turned DSP library.