Recursive Overcomplete Signal Representations

For representing stationary signals, several well-established methods are available. For nonstationary signals, however, these approaches can be used only with serious limitations. If the signal can be characterized as a sequence of stationary intervals, overcomplete signal representations help to handle such problems. This paper introduces the concept of recursive overcomplete representations using different recursive signal processing algorithms. The novelty of the paper is that an on-going set of signal transformations together with appropriate (e.g., L-1 norm) minimization procedures can provide optimal on-going representations, on-going signal segmentations into stationary intervals, and on-going feature extractions for immediate utilization in diagnostics, or other applications.