ASViS: A Protocol for Efficient Video Streaming Using SVC and Deadline-based Criteria.

Abstract

Multimedia video streaming is the most used data consumption service on the Internet. Maintaining a data flow rate that provides the best video quality possible without overflowing the pipeline is challenging. Dynamic Adaptive Streaming over HTTP (DASH) is a commonly used protocol that sends video data in small packages called chunks. However, this technique can lead to video buffering issues and video pausing due to an empty video buffer. Adaptive Scalable Video Streaming (ASViS) is a more versatile protocol that uses scalable video coding, a flow-controlled user datagram protocol (UDP), and deadline-based criteria to manage the buffered data. Like transmission control protocol (TCP), ASViS adjusts the flow based on available bandwidth, and it can discard data based on video frame deadlines to prevent outdated data transmission. ASViS is compliant with RFC 8085, preventing it from overwhelming the Internet. The scalability feature allows ASViS to have the highest image quality possible in every frame allowed by SVC with the data layers available. A theoretical model is presented that predicts ASViS behavior under different network conditions. This work also shows how different parameters affect its performance and how it can be easily configured to optimize it.