Abstract
Our earlier Low-on-Latency (dubbed as LoL) solution offered an accurate bandwidth prediction and rate adaptation algorithm tailored for live streaming applications that targeted an end-to-end latency of up to two seconds. While LoL was a significant step forward in multi-bitrate low-latency live streaming, further experimentation and testing showed that there was room for improvement in three areas. First, LoL used hard-coded parameters computed from an offline training process in the rate adaptation algorithm and this was seen as a significant barrier in LoL’s wide deployment. Second, LoL’s objective was to maximize a collective QoE function. Yet, certain use cases have specific objectives besides the singular QoE and this had to be accommodated. Third, the adaptive playback speed control failed to produce satisfying results in some scenarios. Our goal in this paper is to address these areas and make LoL sufficiently robust to deploy. We refer to the enhanced solution as LoL <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^+$</tex-math></inline-formula> , which has been integrated to the official dash.js player in v3.2.0.
Highlights
With the rise of low-latency live (LLL) streaming applications such as Twitter’s Periscope, Amazon’s Twitch and Facebook’s Live, and users’ growing interest in eSports and streaming of live sports, the demand for low-latency services is higher than ever
We demonstrate player enhancements that are effective for media delivery in LLL streaming
The adaptive bitrate (ABR) algorithm is based on a self-organizing map (SOM) model that considers multiple QoE metrics as well as variability in network conditions in the ABR formulation. ( ) The playback speed control module implements a hybrid playback speed algorithm that combines the current latency and buffer level to control the playback speed. ( ) The throughput measurement module accurately calculates the throughput by removing the idle times between the chunks of a segment through a three-step algorithm. ( ) The QoE evaluation module computes the QoE considering five key metrics: selected bitrate, number of bitrate switches, 1As shown in [11], the SOM model is sensitive to the initial weight values
Summary
With the rise of low-latency live (LLL) streaming applications such as Twitter’s Periscope, Amazon’s Twitch and Facebook’s Live, and users’ growing interest in eSports and streaming of live sports, the demand for low-latency services is higher than ever. As proprietary solutions such as Adobe’s Real-time Messaging Protocol (RTMP) [3] fade away, HTTP adaptive streaming (HAS), using primarily the open Dynamic Adaptive Streaming over HTTP (DASH) standard and Apple’s HTTP Live Streaming (HLS) protocol, dominates the market today. The latency (measured from the moment of capturing to the moment of rendering) becomes part of this trade-off. As opposed to the accustomed 30–60 seconds of latency for the traditional HAS, the target for LLL streaming applications is mostly five seconds or less [33]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
