Mathematical Modeling and Analysis
Recent results have shown that it is possible to obtain exact reconstruction of a large class of signals from far less data than is suggested by traditional sampling theory. Now we have shown how to extend the method to decrease the necessary amount of data still further. The underlying phenomenon is very general, and applies to signals of all types. Decoding highly corrupted bit streams, image or 3-D object reconstruction from very few radiographs, and network tomography are just a few of the many possible applications in all aspects of LANL's mission.