Skip to Main Content
We discuss the issues that a network adaptive application faces on current network infrastructure and the role a programmable network such as the active network can play in facilitating them. We present a novel cognizant video transcoding system, which is capable of negotiating local network state based rate and let the video propagate over an extreme network with highly asymmetric link and node capacities. For core operation of rate transformation it utilizes knowledge at three levels - about the network, about the content protocol and about the content itself. As a platform the stream uses the computing power of a quasi-active network. As a result the passing video stream appears as a self-organizing stream, which automatically senses the network asymmetry and adapts itself as the packets diffuse via the active subnet. We discuss the issues of network adaptation with the exposition of this novel diffusion computing over a quasi-active network.
Date of Conference: 2002