Output details

Gradient pursuits

Significance of output:

Sparse signal approximations have had wide ranging impact on applications such as power network monitoring; distributed sensor networks; distributed video sensing; audio decomposition and coding; synthetic aperture radar imaging; communication channel estimation; image de-noising; tomographic particle image velocimetry; de-noising of images; communications signals and underwater acoustic imaging. However, they tend to be computationally very demanding which sometimes limits their use. In this paper we propose a class of algorithms that are computationally extremely efficient, sometimes by orders of magnitude, thereby considerably extending the application of these algorithms.