This diploma thesis focuses on multimedia streaming from IP cameras. Its main goal is to explain theoretical background of real-time streaming via computer networks, and describe development of a recording system. This recording system is meant to be used mainly in schools for lecture recording purposes. The thesis contains description on how a recording server application and web-based management system were developed. The theoretical part explains topics related to multimedia streaming, networking, and multimedia procesing, such as real-time streaming protocols, encoding, compression, network latency, network congestion and many others.

No description available.

Wireless broadband technologies provide ubiquitous broadband access to wireless users, enabling services that were available only to wireline users. At the same time, with the steady growth of real-time streaming applications such as video on demand (VoD), voice over IP (VoIP), massive online gaming, and so forth, the IEEE 802.16 standard (commonly known as WiMAX) has emerged as one of the strongest contenders to provide such kind of broadband wireless access services. WiMAX has specified some advanced lard features at physical (PHY) layer techniques and media access control (MAC) layer protocols, which adopted many Quality of Service (QoS) scheduling algorithms, resource allocation, Hybrid ARQ and so on. Moreover, forward error correction (FEC) and automatic repeat request (ARQ) techniques have already specified in the standard and they are used to support real-time streaming services in all kind of channel conditions. Inside this thesis, we propose an efficient fast feedback algorithm and buffer congestion control improvement scheme for data streaming over WiMAX networks. Two reserved bits in Generic MAC header (GMH) of each MPDU in WiMAX systems are utilized. The first reserved bit is used as a trigger in fast feedback strategy to add more robust coding and choose better feedback channel when the feedback message does not arrive properly within its cycle time trip (CTT) or after timeout. The second reserved bit is used to inform the base station about the serving subscriber stations¡¦ buffers states when their buffers are above the predefine-threshold value. Increasing number of retrieval of the feedback message, which means we can increase restore-bit-error probabilities within each frame, and then the throughput performance shall increase as well. In addition, by having each subscriber stations or service station¡¦s buffer states knowledge, the BS will be able to provide extra bandwidth allocation to the SSs more efficiency and accuracy. Keywords: WiMAX, PHY, MAC, Hybrid ARQ, QoS, Generic MAC header, CTT, Fast Feedback, Buffer Congestion Control Improvement, Real-Time streaming.

In recent years there has been an increasing demand for real-time streaming applications that handle large volumes of data with low latency. Examples of such applications include real-time monitoring and analytics, electronic trading, advertising, fraud detection, and more. In a streaming pipeline the first step is ingesting the incoming data events, after which they can be sent off for processing. Choosing the correct tool that satisfies application requirements is an important technical decision that must be made. This thesis focuses entirely on the data ingestion part by evaluating three different platforms: Apache Kafka, Apache Pulsar and Redis Streams. The platforms are compared both on characteristics and performance. Architectural and design differences reveal that Kafka and Pulsar are more suited for use cases involving long-term persistent storage of events, whereas Redis is a potential solution when only short-term persistence is required. They all provide means for scalability and fault tolerance, ensuring high availability and reliable service. Two metrics, throughput and latency, were used in evaluating performance in a single node cluster. Kafka proves to be the most consistent in throughput but performs the worst in latency. Pulsar manages high throughput with low message sizes but struggles with larger message sizes. Pulsar performs the best in overall average latency across all message sizes tested, followed by Redis. The tests also show Redis being the most inconsistent in terms of throughput potential between different message sizes

Le trafic de vidéo streaming étant en très forte augmentation dans les réseaux mobiles, il devient essentiel pour les opérateurs de tenir compte des spécificités de ce trafic pour bien dimensionner et configurer le réseau. Dans cette thèse, nous nous intéressons à la modélisation du trafic de vidéo streaming dans les réseaux mobiles. Pour le trafic de vidéo streaming en temps-réel, nous obtenons une forme analytique pour une métrique de qualité de service (QoS) importante, le taux de perte de paquets, et utilisons ce modèle à faire du dimensionnement. Pour le trafic de vidéo streaming de type HTTP adaptatif, nous proposons et analysons d’autres métriques de QoS comme le bitrate moyen, le taux de déficit vidéo et le surplus de buffer, afin de trouver le bon compromis entre résolution de la vidéo et fluidité de la diffusion vidéo. Nous étudions par simulation l’impact de quelque paramètres clés du systéme. Nous montrons que l’utilisation de segments de vidéo courts, d’un nombre réduit d’encodages vidéos et de l’ordonnancement de type round robin améliore la fluidité de la vidéo tout en diminuant sa résolution. Nous proposons par ailleurs d’adapter le nombre des segments téléchargés dans une requête HTTP de sorte que chaque requête corresponde au même volume de données. Enfin, nous appliquons les techniques de l’apprentissage automatique pour analyser la corrélation entre les caracteristiques du système et la qualité d’expérience (QoE) des utilisateurs. / As the traffic of video streaming increases significantly in mobile networks, it is essential for oper- ators to account for the features of this traffic when dimensioning and configuring the network. The focus of this thesis is on traffic models of video streaming in mobile networks. For real-time video streaming traffic, we derive an analytical form for an important Quality-of-Service (QoS) metric, the packet outage rate, and utilize the model for dimensioning. For HTTP adaptive video streaming traffic, we propose and evaluate other QoS metrics such as the mean video bit rate, the deficit rate and the buffer surplus, so as to find the good trade-off between video resolution and playback smoothness. We study by simulation the impacts of some key parameters of the system. We show that using smaller chunk durations, fewer video coding rates and round-robin scheduling may provide a smoother video playback but decrease the mean video resolution. We also propose to adapt the number of chunks downloaded in an HTTP request so that each HTTP request has the same data volume. Finally, we apply machine learning techniques to analyze the correlation between system characteristics and the quality of experience (QoE) of users.

No description available.