There has been a remarkable increase in download of content provided by both traditional commercial suppliers as well those by amateurs. Users have very diverse computer equipment ranging from handheld devices to desktops, and downloaded items range from simple stock quotes to news stories to full movies. Such highly heterogeneous circumstances will be predominant in the networks of the near future. Current schemes for content download are inefficient at addressing the mentioned heterogeneity. The goal of this project is to obtain new and improved schemes that would result in better access to content for the users while improving the network utilization.
The research will focus on three coding techniques to attack the problem: rateless coding at the application layer, network coding at the network layer, and collaborative coding at the physical layer. Quantitative and qualitative limitations of network and rateless coding in heterogeneous environments will be studied to develop new practical coding schemes appropriate in such circumstances. Another research direction is to investigate and analyze mechanism to make networks more uniform using physical layer collaborative transmission schemes. An assessment will be made on how this process affects the performance of the higher layer coding schemes.
Broader Impact: The research, if successful, will improve the efficiency and availability of data in networks. This has the potential to benefit directly and indirectly the population at large. In addition, this collaborative research has the potential to obtain fundamental new insights at the interface of coding and information theory and combinatorial optimization. Several PhD students will be supported and trained in interdisciplinary areas and they would also obtain valuable industrial experience.
