Quality-aware strategies for optimizing ABR video streaming QoE and reducing data usage

Abstract

Streaming videos over cellular networks is highly challenging. Since cellular data is a relatively scarce resource, many video and network providers offer options for users to exercise control over the amount of data consumed by video streaming. Our study shows that existing data saving practices for Adaptive Bitrate (ABR) videos are suboptimal: they often lead to highly variable video quality and do not make the most effective use of the network bandwidth. We identify underlying causes for this and propose two novel approaches to achieve better tradeoffs between video quality and data usage. The first approach is Chunk-Based Filtering (CBF), which can be retrofitted to any existing ABR scheme. The second approach is QUality-Aware Data-efficient streaming (QUAD), a holistic rate adaptation algorithm that is designed ground up. We implement and integrate our solutions into two video player platforms (dash.js and ExoPlayer), and conduct thorough evaluations over emulated/commercial cellular networks using real videos. Our evaluations demonstrate that compared to the state of the art, the two proposed schemes achieve consistent video quality that is much closer to the user-specified target, lead to far more efficient data usage, and incur lower stalls.

Popularity-based transcoding workload allocation for improving video quality in live streaming systems

“…We consider seven bitrates at four resolutions as recommended by public transcoding cloud provider, Zencoder [3]. VMAF values for each bitrate version were generated according to a normal distribution based on the measured values [4]. The access probability of each channel was modeled with the parameters of the gamma distribution observed on Twitch.tv [5].…”

“…Due to the limited processing capacity, transcoding services can only be provided to premium channels. Thus there are not enough video quality versions for all channels, so this makes it harder for the adaptation logic at the client to find a close match based on estimated bandwidth availability [3,4]. In addition, live streaming requires real-time transcoding processing, so processing capacity needs to be continuously allocated to transcoding tasks for each channel [3].…”

Describing Subjective Experiment Consistency by p-Value P--P Plot

“…Therefore, numerous works focus on QoE optimization in various contexts. Many target quality improvement by better network resources utilization [2,33]. Some go beyond that and use the QoE to optimize parameters like power consumption of mobile device displays [45] or to improve user-perceived quality when watching 360 • videos.…”

DataPlanner

“…Existing approaches in industry to reducing data usage are either service-based or network-based [37]. A service-based approach is typified by a video service providing a user multiple options via the client player user interface; users can choose an option with a lower data usage at the cost of lower quality experience.…”

“…Existing ABR adaptation schemes in literature ( §7) focus mainly on maximizing the video quality under the network bandwidth constraints, and not on limiting data usage. One exception is [37], which assumes a user specified target quality and proceeds to avoid segments whose qualities exceeding the target quality, leading to bandwidth savings. The approach, however, still does not provide a user an explicit control of data usage.…”