A New Approach to Signal Processing of Spatiotemporal Data

  title={A New Approach to Signal Processing of Spatiotemporal Data},
  author={J. Slawinska and A. Ourmazd and D. Giannakis},
  journal={2018 IEEE Statistical Signal Processing Workshop (SSP)},
  • J. Slawinska, A. Ourmazd, D. Giannakis
  • Published 2018
  • Computer Science
  • 2018 IEEE Statistical Signal Processing Workshop (SSP)
  • We present a method combining ideas from the theory of operator-valued kernels with delay-coordinate embedding techniques in dynamical systems capable of identifying spatiotemporal patterns, without prior knowledge of the state space or the dynamical laws of the system generating the data. The approach is particularly powerful for systems in which characteristic patterns cannot be readily decomposed into temporal and spatial coordinates. Using simulated and observed sea-surface temperature data… CONTINUE READING
    2 Citations

    Figures and Topics from this paper

    Data-driven Koopman operator approach for computational neuroscience
    • 2
    • PDF
    Operator-theoretic framework for forecasting nonlinear time series with kernel analog techniques.
    • 12
    • PDF


    Vector-Valued Spectral Analysis of Space-Time Data
    • 4
    • PDF
    Spatiotemporal analysis of complex signals: Theory and applications
    • 326
    • Highly Influential
    Spatiotemporal Pattern Extraction by Spectral Analysis of Vector-Valued Observables
    • 19
    • PDF
    Spatiotemporal Feature Extraction with Data-Driven Koopman Operators
    • 42
    • PDF
    Nonlinear Laplacian spectral analysis for time series with intermittency and low-frequency variability
    • 145
    • PDF
    Data-driven spectral decomposition and forecasting of ergodic dynamical systems
    • 80
    • PDF
    Dynamic mode decomposition of numerical and experimental data
    • 2,159
    • PDF
    Delay-Coordinate Maps and the Spectra of Koopman Operators
    • 43
    • PDF
    Diffusion maps
    • 1,612
    • Highly Influential
    • PDF
    Dynamics from noisy data with extreme timing uncertainty
    • 32