September 17, 2013
A new technology developed by researchers at NYU-Poly may make spotty wireless streaming and data-hogging downloads a thing of the past.
The patent-pending technique, called streamloading, in the simplest terms makes use of a video format that splits the video into two layers—a base layer, which contains a coarse representation of the video, and an enhancement layer, which completes the image quality and includes the fine-grain details. Traditional streaming involves downloading 30 to 60 seconds of video ahead of time, with the video quality and speed varying depending on wireless signal strength.
Streamloading allows users to pre-download the enhancement layer onto their devices in a location where wireless signal is strong—at home, for example—and stream only the base layer at the time of viewing.
Shivendra S. Panwar, professor of electrical and computer engineering at NYU-Poly and the lead developer of streamloading, estimates that the technique could remove as much as 75 percent of the streaming content from increasingly overloaded cellular wireless networks, while at the same time reducing high data usage charges for consumers. Panwar explains the technology could relieve some of the bandwidth crunch for wireless carriers and significantly improve the speed and quality of streaming video, making it easier and less expensive to access content.
Panwar, along with a team of students at NYU-Poly who have been working on the prototype technology, designed streamloading to be compatible with current digital rights management (DRM) protocols. The team plans to continue testing and refining the technology and have already initiated conversations with wireless carriers.
The research is supported by the National Science Foundation and the Center for Advanced Technology in Telecommunications at NYU-Poly.