Improving The Quality Of Streaming Video In Slow Networks – New Method

October 3, 2017 | Posted By Lakshay Anand

Streaming video is one of the most popular tasks in mobile networks. Users are demanding high quality of the content, and the slightest stumble or artifact can spoil all the fun of watching the video. Network engineers are constantly working on new ways to maximize the quality of videos that could help even with the congestion of wireless networks, which are limited by bandwidth. It turns out that it is enough to transmit the first most important video packets.

There are already methods to maintain a certain quality of video and audio transmission with a limited speed. This is a “quality of service” protocols, Quality of Service (QoS). These methods work well in many cases, but they usually require a bandwidth allocation for this resource from each user, which is not always possible in the crowded mobile networks.

Pen Huey Tan (Peng Hui Tan) and Lee Maodong (Maodong Li) and his colleagues at the Institute of Infocomm Research, Singapore Agency for Science, Technology and Research (A * STAR) have developed a different approach to solving this problem, due to which it becomes possible to estimate the importance of “discrete video packets, which allows you to reduce the bandwidth required to maintain a certain” quality of experience »(quality of experience, QoE). That is for the user to look more quality than in the same bandwidth, but without the use of a new method.

Researchers have developed an efficient method for determining the “importance index” for each video packet. It is based on the bit rate, which varies from package to package, depending on how much new information is required to display. For example, the slow scene with small number of motions require less bit rate. At this stage, in which the action takes place rapidly, characterized by a very high need for bit rate.

The priority of video packets based on the bit rate and other parameters of the network and allows to follow these priorities QoS-circuit in real time. The development team expects that the new approach will provide a significant improvement in the user’s perception of quality video streaming. For the end user, the new technology will reduce the number of distortions in the process of watching the video, while the service providers with the same amount of network resources will be able to provide services to more users.