Adaptive scalable video streaming based on deadline-sensitive criteria
Professor Advisor
dc.contributor.advisor
Estévez Montero, Claudio
Author
dc.contributor.author
Peña Ancavil, Eliecer Ramón
Associate professor
dc.contributor.other
Becerra Yoma, Néstor
Associate professor
dc.contributor.other
Kamal, Shaharyar
Associate professor
dc.contributor.other
Tang, Xianfeng
Associate professor
dc.contributor.other
Huenupan Quinan, Fernando
Admission date
dc.date.accessioned
2024-09-12T14:51:55Z
Available date
dc.date.available
2024-09-12T14:51:55Z
Publication date
dc.date.issued
2024
Identifier
dc.identifier.other
10.58011/q5qt-za87
Identifier
dc.identifier.uri
https://repositorio.uchile.cl/handle/2250/201013
Abstract
dc.description.abstract
El streaming de video domina el uso de datos en Internet, pero entregar una calidad constante es complicado. El método Dynamic Adaptive Streaming over HTTP (DASH), a pesar de su popularidad, sufre interrupciones por búferes vacíos. Para mejorar esto, se propone el protocolo Adaptive Scalable Video Streaming (ASViS), que utiliza codificación de video escalable y un enfoque User Datagram Protocol (UDP), priorizando la entrega puntual. ASViS se adapta a la red como Transmission Control Protocol (TCP), pero es innovador al descartar datos basándose en su relevancia temporal para evitar información obsoleta, cumpliendo con el Request for Comments (RFC) 8085 y previniendo la saturación de la red. Se desarrolló un modelo para prever su comportamiento en distintas condiciones de red y evaluar la influencia de varios parámetros en su rendimiento. La ventaja de ASViS radica en maximizar la calidad de imagen ajustándose a la Scalable Video Coding (SVC) y a las capas de datos disponibles, superando así las limitaciones de DASH y apuntando a un servicio de streaming más fiable y de alta calidad. Esta innovación promete un significativo avance para el streaming de video, buscando mejorar la experiencia del usuario.
es_ES
Abstract
dc.description.abstract
Multimedia video streaming, identified as the dominant internet data consumption service, brings forth challenges in consistently delivering optimal video quality. Dynamic Adaptive Streaming over HTTP (DASH), while prevalent, often encounters buffering problems,
causing video pauses due to empty video buffers. This study introduces the Adaptive Scalable
Video Streaming (ASViS) protocol as a solution. ASViS incorporates scalable video coding,
a flow-controlled User Datagram Protocol (UDP), and deadline-based criteria. A model is
developed to predict the behavior of ASViS across varying network conditions. Additionally,
the effects of diverse parameters on ASViS performance are evaluated. ASViS adjusts data
flow similarly to the Transmission Control Protocol (TCP), based on bandwidth availability.
Data are designed to be discarded by ASViS according to video frame deadlines, preventing
outdated information transmission. Compliance with RFC 8085 ensures the internet is not
overwhelmed. With its scalability feature, ASViS achieves the highest possible image quality
per frame, aligning with Scalable Video Coding (SVC) and the available data layers. The
introduction of ASViS offers a promising approach to address the challenges faced by DASH,
potentially providing more consistent and higher-quality video streaming.
Lenguage
dc.language.iso
en
es_ES
Publisher
dc.publisher
Universidad de Chile
es_ES
Type of license
dc.rights
Attribution-NonCommercial-NoDerivs 3.0 United States