Sparse time-frequency representations.

Center for Studies in Physics and Biology, The Rockefeller University, 1230 York Avenue, New York, NY 10021.
Proceedings of the National Academy of Sciences (Impact Factor: 9.81). 05/2006; 103(16):6094-9. DOI: 10.1073/pnas.0601707103
Source: PubMed

ABSTRACT Auditory neurons preserve exquisite temporal information about sound features, but we do not know how the brain uses this information to process the rapidly changing sounds of the natural world. Simple arguments for effective use of temporal information led us to consider the reassignment class of time-frequency representations as a model of auditory processing. Reassigned time-frequency representations can track isolated simple signals with accuracy unlimited by the time-frequency uncertainty principle, but lack of a general theory has hampered their application to complex sounds. We describe the reassigned representations for white noise and show that even spectrally dense signals produce sparse reassignments: the representation collapses onto a thin set of lines arranged in a froth-like pattern. Preserving phase information allows reconstruction of the original signal. We define a notion of "consensus," based on stability of reassignment to time-scale changes, which produces sharp spectral estimates for a wide class of complex mixed signals. As the only currently known class of time-frequency representations that is always "in focus" this methodology has general utility in signal analysis. It may also help explain the remarkable acuity of auditory perception. Many details of complex sounds that are virtually undetectable in standard sonograms are readily perceptible and visible in reassignment.


Available from: Marcelo Osvaldo Magnasco, May 30, 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Spectral decomposition, or local time-frequency analysis, tries to enhance the amount of information one can obtain from a seismic volume by finding the frequency content of the seismic data at each time sample. However, if a small amount of noise is present within the seismic amplitude volume, it has the potential to become more prominent in the spectrally decomposed data especially if high-resolution or sparsity promoting methods are utilized. To combat this problem post-processing noise removal has commonly been employed, but these techniques can potentially degrade the resolution of small-scale geological structures in their attempt to remove this noise. Rather than de-noising the spectrally decomposed data after they are generated, we propose to incorporate the ideas of f−x−y deconvolution within the spectral decomposition process to create an algorithm that has the ability to de-noise the time-frequency representation of the data as they are being generated. By incorporating the spatial prediction error filters that are utilized for f−x−y deconvolution with the spectral decomposition problem, a spatially smooth time-frequency representation that maintains its sparsity, or high-resolution characteristics, can be obtained. This spatially smooth high-resolution time-frequency representation is less likely to exhibit the random noise that was present in the more conventionally obtained time-frequency representation. Tests on a real data set demonstrate that by de-noising while the time-frequency representation is being constructed, small-scale geological structures are more likely to maintain their resolution since the de-noised time-frequency representation is specifically built to reconstruct the data.
    Geophysical Prospecting 06/2013; 61(s1). DOI:10.1111/j.1365-2478.2012.01104.x · 1.51 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present here an innovative hypothesis and report preliminary evidence that the sound of NMR signals could provide an alternative to the current representation of the individual metabolic fingerprint and supply equally significant information. The NMR spectra of the urine samples provided by four healthy donors were converted into audio signals that were analyzed in two audio experiments by listeners with both musical and non-musical training. The listeners were first asked to cluster the audio signals of two donors on the basis of perceived similarity and then to classify unknown samples after having listened to a set of reference signals. In the clustering experiment, the probability of obtaining the same results by pure chance was 7.04% and 0.05% for non-musicians and musicians, respectively. In the classification experiment, musicians scored 84% accuracy which compared favorably with the 100% accuracy attained by sophisticated pattern recognition methods. The results were further validated and confirmed by analyzing the NMR metabolic profiles belonging to two other different donors. These findings support our hypothesis that the uniqueness of the metabolic phenotype is preserved even when reproduced as audio signal and warrants further consideration and testing in larger study samples.
    Omics A Journal of Integrative Biology 03/2015; 19(3):147-156. DOI:10.1089/omi.2014.0131 · 2.73 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper considers the analysis of multicomponent signals, defined as superpositions of real or complex modulated waves. It introduces two new post-transformations for the short-time Fourier transform that achieve a compact time-frequency representation while allowing for the separation and the reconstruction of the modes. These two new transformations thus benefit from both the synchrosqueezing transform (which allows for reconstruction) and the reassignment method (which achieves a compact time-frequency representation). Numerical experiments on real and synthetic signals demonstrate the efficiency of these new transformations, and illustrate their differences.
    IEEE Transactions on Signal Processing 03/2015; 63(5):1-1. DOI:10.1109/TSP.2015.2391077 · 3.20 Impact Factor