A Simulation-Based Performance eva‎luation Framework For Transporting DASH Video Over Multipath TCP

2/13/2024 12:00:00 AM

Today, HTTP Adaptive Streaming (HAS) is widely used on the Internet. With the help of dynamic adaptation, the quality of video playback can be adapted to the conditions of the end user. Dynamic Adaptive Streaming over HTTP (DASH) is a new standard that aims to bring together techniques comparable to HAS. Multipath TCP is an extension of the classic TCP protocol that is capable of establishing multiple connections simultaneously and presenting it as if it were a single connection to the application layer. When there are multiple network interfaces in a system (like the case we face in today's mobile devices), the use of multi-path capability can improve the quality of video playback. The use of multiple interfaces at the same time as well as the use of packet sending scheduling methods lead to better quality of service for programs with high bandwidth consumption such as online video streaming and online games. Most of the proposed approaches are aimed at improving the quality of the multimedia stream to the servers or to the client. Some studies have been on MPTCP performance investigation, which has led to obtaining a series of important characteristics for further eva‎luations. In the studies conducted, we noticed the lack of a comprehensive platform to test and eva‎luate the proposed algorithms. In this thesis, we intend to introduce a specific framework for simulating and eva‎luating MPTCP efficiency for video streaming under different network scenarios. With the help of the proposed framework, researchers can test and eva‎luate their own data and compare the different results obtained with each other. In the proposed framework, we create a simulation environment and examine the quality of experience (QoE) perceived by the end user by combining DASH and MPTCP and analyze the results. Obviously, the most important differences between our proposed eva‎luation framework and existing frameworks are: • The possibility of applying specific settings such as the number of switches, the number of hosts, the number of links, the number of paths, etc. as input • The possibility of checking the specific characteristics and requirements of the researcher, such as the user's quality experience, the resulting congestion, etc. as an output • The possibility of graphically viewing the network based on the researcher's inputs • The possibility of testing different scenarios by the researcher In general, it can be said that the proposed framework can be a reliable and practical tool for researchers to test and review their proposed scenarios related to the network and dynamic adaptive flow through HTTP and MPTCP.

DASH , MPTCP , TCP , پخش ويديو شبكه هاي سرعت بالا , TCPچند مسيري , جريان تطبيقي پويا از طريق HTTP , كيفيت تجربه

DASH , MPTCP , TCP , Video Streaming High Speed Networks , Multipath TCP , Dynamic Adaptive Streaming Over HTTP , QoE

Jenen Alemeri

Vesal Hakami