Dynamic HTTP Live Streaming Method for Live Feeds
Prasenjit Chakraborty, Sachin Dev, Rajaram Hanumanthacharya Naganur · 2015
Streaming Feed has become one of the most used utility now-a-days. Starting from Home-Security to Business Surveillance, streaming feed is utilized everywhere. However, due to unstable network bandwidth, frequent buffering and delay occurrence usually creates degraded user experience. One of the ways to avoid this is to use Adaptive Bit rate Streaming. Among various protocols supporting this Adaptive Feed, HTTP Live Streaming (HLS) has become one of the most famous commercially used protocol choices. The main advantage achieved via this is the ability to provide support for different device categories (in terms of maximum resolution or maximum bit rate the device can support) as well as supporting of various video qualities depending upon network condition. But the downside of using HLS is the introduction of transcoding delay in Live Feed. In this paper, we are proposing Dynamic transcoding algorithm for HLS by optimizing the delay and providing a near live experience to a vast number of differently capable devices at the same time without violating the principles of the specification. Based on our experimental results, we will show that, using the proposed methodology, we can achieve better user experience by providing best suitable video with current network condition apart from supporting varied kind of devices at the same time.