Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries

Andreas I. Koutrouvelis; Aki Härmä; Athanasios Mouchtaris

doi:10.1109/ICDSP.2013.6622834

Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries

Andreas I. Koutrouvelis, Aki Härmä, Athanasios Mouchtaris

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

Abstract

The application of Compressive Sensing is explored in three signal categories; footstep sounds, hand tremors and speech. An investigation of the reconstruction performance of various dictionaries is undertaken. It is demonstrated that these signal categories are reconstructed with higher SNR performance using K-SVD dictionaries than other fixed dictionaries. In particular, for footstep sounds and hand tremors, the K-SVD dictionaries outperform the fixed dictionaries; Discrete Cosine Transform (DCT), Wavelet Symlet with order 8 Transform and the union of DCT and Discrete Sine Transform. Moreover, in speech reconstruction, the use of a codebook of K-SVD dictionaries instead of a codebook of impulse response matrices improves performance.

Original language	English
Title of host publication	2013 18th International Conference on Digital Signal Processing, DSP 2013
DOIs	https://doi.org/10.1109/ICDSP.2013.6622834
Publication status	Published - 2013
Externally published	Yes
Event	18th International Conference on Digital Signal Processing - Santorini, Greece Duration: 1 Jul 2013 → 3 Jul 2013 Conference number: 18

Conference

Conference	18th International Conference on Digital Signal Processing
Abbreviated title	DSP 2013
Country/Territory	Greece
City	Santorini
Period	1/07/13 → 3/07/13

Keywords

Compressive sensing
Impulse response matrix
K-SVD
Sparse signal reconstruction

Access to Document

10.1109/ICDSP.2013.6622834

Cite this

@inproceedings{31be98f3480e4df790e369b6cae21dd6,

title = "Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries",

abstract = "The application of Compressive Sensing is explored in three signal categories; footstep sounds, hand tremors and speech. An investigation of the reconstruction performance of various dictionaries is undertaken. It is demonstrated that these signal categories are reconstructed with higher SNR performance using K-SVD dictionaries than other fixed dictionaries. In particular, for footstep sounds and hand tremors, the K-SVD dictionaries outperform the fixed dictionaries; Discrete Cosine Transform (DCT), Wavelet Symlet with order 8 Transform and the union of DCT and Discrete Sine Transform. Moreover, in speech reconstruction, the use of a codebook of K-SVD dictionaries instead of a codebook of impulse response matrices improves performance.",

keywords = "Compressive sensing, Impulse response matrix, K-SVD, Sparse signal reconstruction",

author = "Koutrouvelis, {Andreas I.} and Aki H{\"a}rm{\"a} and Athanasios Mouchtaris",

year = "2013",

doi = "10.1109/ICDSP.2013.6622834",

language = "English",

isbn = "9781467358057",

booktitle = "2013 18th International Conference on Digital Signal Processing, DSP 2013",

note = "18th International Conference on Digital Signal Processing, DSP 2013 ; Conference date: 01-07-2013 Through 03-07-2013",

}

Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries. / Koutrouvelis, Andreas I.; Härmä, Aki; Mouchtaris, Athanasios.
2013 18th International Conference on Digital Signal Processing, DSP 2013. 2013. 6622834.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

TY - GEN

T1 - Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries

AU - Koutrouvelis, Andreas I.

AU - Härmä, Aki

AU - Mouchtaris, Athanasios

N1 - Conference code: 18

PY - 2013

Y1 - 2013

N2 - The application of Compressive Sensing is explored in three signal categories; footstep sounds, hand tremors and speech. An investigation of the reconstruction performance of various dictionaries is undertaken. It is demonstrated that these signal categories are reconstructed with higher SNR performance using K-SVD dictionaries than other fixed dictionaries. In particular, for footstep sounds and hand tremors, the K-SVD dictionaries outperform the fixed dictionaries; Discrete Cosine Transform (DCT), Wavelet Symlet with order 8 Transform and the union of DCT and Discrete Sine Transform. Moreover, in speech reconstruction, the use of a codebook of K-SVD dictionaries instead of a codebook of impulse response matrices improves performance.

AB - The application of Compressive Sensing is explored in three signal categories; footstep sounds, hand tremors and speech. An investigation of the reconstruction performance of various dictionaries is undertaken. It is demonstrated that these signal categories are reconstructed with higher SNR performance using K-SVD dictionaries than other fixed dictionaries. In particular, for footstep sounds and hand tremors, the K-SVD dictionaries outperform the fixed dictionaries; Discrete Cosine Transform (DCT), Wavelet Symlet with order 8 Transform and the union of DCT and Discrete Sine Transform. Moreover, in speech reconstruction, the use of a codebook of K-SVD dictionaries instead of a codebook of impulse response matrices improves performance.

KW - Compressive sensing

KW - Impulse response matrix

KW - K-SVD

KW - Sparse signal reconstruction

UR - http://www.scopus.com/inward/record.url?scp=84888877594&partnerID=8YFLogxK

U2 - 10.1109/ICDSP.2013.6622834

DO - 10.1109/ICDSP.2013.6622834

M3 - Conference article in proceeding

AN - SCOPUS:84888877594

SN - 9781467358057

BT - 2013 18th International Conference on Digital Signal Processing, DSP 2013

T2 - 18th International Conference on Digital Signal Processing

Y2 - 1 July 2013 through 3 July 2013

ER -

Compressive sensing in footstep sounds, hand tremors and speech using K-SVD dictionaries

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this