Inverse Problems with Time-frequency Dictionaries and non-white Gaussian Noise

Abstract : Sparse regressions to solve ill-posed inverse problems have been massively investigated over the last decade. Yet, when noise is present in the model, it is almost exclusively considered as Gaussian and white. While this assumption can hold in practice, it rarely holds when observations are time series as they are corrupted by auto-correlated and colored noise. In this work we study sparse regression under the assumption of non white Gaussian noise and explain how to run the inference using proximal gradient methods. We investigate an application in brain imaging: the problem of source localiza-tion using magneto-and electroencephalography which allow functional brain imaging with high temporal resolution. We use a time-frequency representation of the source waveforms and a sparse regularization which promotes focal sources with smooth and transient activations. Our approach is evaluated using simulations comparing it to strategies that assume the noise is white or to simple prewhitening.
Document type :
Conference papers
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01199635
Contributor : Matthieu Kowalski <>
Submitted on : Tuesday, September 15, 2015 - 5:06:28 PM
Last modification on : Friday, June 7, 2019 - 11:07:06 AM
Long-term archiving on : Tuesday, December 29, 2015 - 7:21:41 AM

File

KG_eusipco.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01199635, version 1

Citation

Matthieu Kowalski, Alexandre Gramfort. Inverse Problems with Time-frequency Dictionaries and non-white Gaussian Noise. 23rd European Signal Processing Conference (EUSIPCO 2015), Aug 2015, Nice, France. ⟨hal-01199635⟩

Share

Metrics

Record views

879

Files downloads

583