From Theory to Practice

Abstract

Modern video streaming uses adaptive bitrate (ABR) algorithms that run inside video players and continually adjust the quality (i.e., bitrate) of the video segments that are downloaded and rendered to the user. To maximize the quality-of-experience (QoE) of the user, ABR algorithms must stream at a high bitrate with low rebuffering and low bitrate oscillations. Further, a good ABR algorithm is responsive to user and network events and can be used in demanding scenarios such as low-latency live streaming. Recent research papers provide an abundance of ABR algorithms but fall short on many of the above real-world requirements. We develop Sabre, an open-source publicly available simulation tool that enables fast and accurate simulation of adaptive streaming environments. We empirically validated Sabre to show that it accurately simulates real-world environments. We used Sabre to design and evaluate BOLA-E and DYNAMIC, two novel ABR algorithms. We also developed a FAST SWITCHING algorithm that can replace segments that have already been downloaded with higher-bitrate (thus, higher-quality) segments. The new algorithms provide higher QoE to the user in terms of higher bitrate, fewer rebuffers, and lesser bitrate oscillations. In addition, these algorithms react faster to user events such as startup and seek, and they respond more quickly to network events such as improvements in throughput. Further, they perform very well for live streams that require low latency, a challenging scenario for ABR algorithms. Overall, our algorithms offer superior video QoE and responsiveness for real-life adaptive video streaming, in comparison to the state-of-the-art. Importantly, all three algorithms presented in this article are now part of the official DASH reference player dash.js and are being used by video providers in production environments. While our evaluation and implementation are focused on the DASH environment, our algorithms are equally applicable to other adaptive streaming formats such as Apple HLS.

Universal access: user needs for immersive captioning

“…Due to the large file size of immersive videos, the player was updated to support both HTTP Live Streaming (HLS) [20] using hls.js [21] and Dynamic Adaptive Streaming over HTTP (DASH) [22] streams using dash.js [23]. This massively improves the performance of video playback, where network bandwidth was limited and therefore improves the user experience.…”

Using Machine Learning and In-band Network Telemetry for Service Metrics Estimation

“…It's important to highlight that the calculation of rewards does not occur immediately after an action is taken in the current state. This delay in reward calculation is attributed to the fact that the effects of the agent's action do not manifest instantly, primarily due to the inherent control mechanisms incorporated within TCP and ABR systems, as detailed in [26]. Consequently, the computation of rewards is deferred until the subsequent state's observation.…”

“…To gain a deeper understanding of these results, it's important to clarify some aspects of the Adaptive Bitrate (ABR) adaptation logic employed by the DASH.js player, as described in [26]. The adaptation logic used in DASH.js, known as DYNAMIC, employs two different algorithms at different stages of video playback.…”

“…From this point on, DYNAMIC switches to BOLA which chooses a bitrate at least as high as the bitrate chosen by THROUGHPUT. DYNAMIC switches back to THROUGH-PUT when the buffer level falls below 10s and BOLA chooses a bitrate lower than THROUGHPUT [26].…”

Study on Dynamic Adaptive Bitrate Selection Algorithm for Mobile Streaming Media

“…In DASH, the dynamic adaptive bitrate selection algorithm is the core technology to enables high quality streaming of media content over the Internet de-livered from conventional HTTP web servers. A good adaptive bitrate algorithm not only continually adjust the bitrate of the video segments that are downloaded and rendered to the user, but also is responsive to user and network events and can be used in demanding scenarios such as low-latency live streaming [2].…”