Minimum classification error using time-frequency analysis

C. Breakenridge, M. Mesbah

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

For certain classes of signals, such as time varying signals, classical classification algorithms are not suitable. Hence, time-frequency based techniques are employed for classification of these types of signals. In this paper we propose data-driven time frequency representations kernel optimization, that leads to the minimum classification error (MCE) for nonstationary signal classification. Our central issue is to determine the optimal kernel parameters and best distance measure to achieve the MCE performance measure. The minimum classification error achievable using optimized kernels is investigated for two types of nonstationary signals; namely simulated chirp signals and real-life newborn EEG signals. For the EEG signals a classification error as low as 4.6% was achieved.

Original languageEnglish
Title of host publicationProceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages717-720
Number of pages4
ISBN (Electronic)0780382927, 9780780382923
DOIs
Publication statusPublished - 2003
Event3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003 - Darmstadt, Germany
Duration: Dec 14 2003Dec 17 2003

Other

Other3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003
CountryGermany
CityDarmstadt
Period12/14/0312/17/03

Fingerprint

Electroencephalography

Keywords

  • Algorithm design and analysis
  • Chirp
  • Electroencephalography
  • Kernel
  • Optimization methods
  • Pediatrics
  • Signal analysis
  • Signal processing algorithms
  • Testing
  • Time frequency analysis

ASJC Scopus subject areas

  • Signal Processing

Cite this

Breakenridge, C., & Mesbah, M. (2003). Minimum classification error using time-frequency analysis. In Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003 (pp. 717-720). [1341221] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSPIT.2003.1341221

Minimum classification error using time-frequency analysis. / Breakenridge, C.; Mesbah, M.

Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003. Institute of Electrical and Electronics Engineers Inc., 2003. p. 717-720 1341221.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Breakenridge, C & Mesbah, M 2003, Minimum classification error using time-frequency analysis. in Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003., 1341221, Institute of Electrical and Electronics Engineers Inc., pp. 717-720, 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003, Darmstadt, Germany, 12/14/03. https://doi.org/10.1109/ISSPIT.2003.1341221
Breakenridge C, Mesbah M. Minimum classification error using time-frequency analysis. In Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003. Institute of Electrical and Electronics Engineers Inc. 2003. p. 717-720. 1341221 https://doi.org/10.1109/ISSPIT.2003.1341221
Breakenridge, C. ; Mesbah, M. / Minimum classification error using time-frequency analysis. Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003. Institute of Electrical and Electronics Engineers Inc., 2003. pp. 717-720
@inproceedings{adc562e427804c058d728650245527cf,
title = "Minimum classification error using time-frequency analysis",
abstract = "For certain classes of signals, such as time varying signals, classical classification algorithms are not suitable. Hence, time-frequency based techniques are employed for classification of these types of signals. In this paper we propose data-driven time frequency representations kernel optimization, that leads to the minimum classification error (MCE) for nonstationary signal classification. Our central issue is to determine the optimal kernel parameters and best distance measure to achieve the MCE performance measure. The minimum classification error achievable using optimized kernels is investigated for two types of nonstationary signals; namely simulated chirp signals and real-life newborn EEG signals. For the EEG signals a classification error as low as 4.6{\%} was achieved.",
keywords = "Algorithm design and analysis, Chirp, Electroencephalography, Kernel, Optimization methods, Pediatrics, Signal analysis, Signal processing algorithms, Testing, Time frequency analysis",
author = "C. Breakenridge and M. Mesbah",
year = "2003",
doi = "10.1109/ISSPIT.2003.1341221",
language = "English",
pages = "717--720",
booktitle = "Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Minimum classification error using time-frequency analysis

AU - Breakenridge, C.

AU - Mesbah, M.

PY - 2003

Y1 - 2003

N2 - For certain classes of signals, such as time varying signals, classical classification algorithms are not suitable. Hence, time-frequency based techniques are employed for classification of these types of signals. In this paper we propose data-driven time frequency representations kernel optimization, that leads to the minimum classification error (MCE) for nonstationary signal classification. Our central issue is to determine the optimal kernel parameters and best distance measure to achieve the MCE performance measure. The minimum classification error achievable using optimized kernels is investigated for two types of nonstationary signals; namely simulated chirp signals and real-life newborn EEG signals. For the EEG signals a classification error as low as 4.6% was achieved.

AB - For certain classes of signals, such as time varying signals, classical classification algorithms are not suitable. Hence, time-frequency based techniques are employed for classification of these types of signals. In this paper we propose data-driven time frequency representations kernel optimization, that leads to the minimum classification error (MCE) for nonstationary signal classification. Our central issue is to determine the optimal kernel parameters and best distance measure to achieve the MCE performance measure. The minimum classification error achievable using optimized kernels is investigated for two types of nonstationary signals; namely simulated chirp signals and real-life newborn EEG signals. For the EEG signals a classification error as low as 4.6% was achieved.

KW - Algorithm design and analysis

KW - Chirp

KW - Electroencephalography

KW - Kernel

KW - Optimization methods

KW - Pediatrics

KW - Signal analysis

KW - Signal processing algorithms

KW - Testing

KW - Time frequency analysis

UR - http://www.scopus.com/inward/record.url?scp=77950292140&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950292140&partnerID=8YFLogxK

U2 - 10.1109/ISSPIT.2003.1341221

DO - 10.1109/ISSPIT.2003.1341221

M3 - Conference contribution

SP - 717

EP - 720

BT - Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003

PB - Institute of Electrical and Electronics Engineers Inc.

ER -