Abstract
In this study we investigate the parallelization of a key feature extraction method called spectral correlation density (SCD) function, which is used in signal classification systems particularly under low signal-to-noise ratio conditions for classifying numerous signals. In order to reduce the computation complexity of the SCD function, we introduce a method called Quarter SCD (QSCD) that allows extracting features of a given signal by processing only quarter of the input signal data. We then parallelize the QSCD by targeting general purpose graphics processing unit (GPU) through architecture specific optimization strategies. We present experimental evaluations on identifying the parallelization configuration for maximizing the efficiency of the program architecture in utilizing the threading power of the GPU architecture. We show that algorithmic and architecture specific optimization strategies result with improving the throughput of the state of the art GPU based Full SCD from 120 signals/second to 2719 signals/second.
| Original language | English (US) |
|---|---|
| Title of host publication | 2018 Conference on Design and Architectures for Signal and Image Processing, DASIP 2018 |
| Publisher | IEEE Computer Society |
| Pages | 88-93 |
| Number of pages | 6 |
| ISBN (Electronic) | 9781538682371 |
| DOIs | |
| State | Published - Dec 31 2018 |
| Event | 12th Conference on Design and Architectures for Signal and Image Processing, DASIP 2018 - Porto, Portugal Duration: Oct 10 2018 → Oct 12 2018 |
Publication series
| Name | Conference on Design and Architectures for Signal and Image Processing, DASIP |
|---|---|
| Volume | 2018-October |
| ISSN (Print) | 2164-9766 |
Conference
| Conference | 12th Conference on Design and Architectures for Signal and Image Processing, DASIP 2018 |
|---|---|
| Country | Portugal |
| City | Porto |
| Period | 10/10/18 → 10/12/18 |
Fingerprint
Keywords
- GPGPU
- signal classification
- spectral correlation density
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition
- Hardware and Architecture
- Signal Processing
- Electrical and Electronic Engineering
Cite this
GPU Based Quarter Spectral Correlation Density Function. / Marshall, Scott; Vanhoy, Garrett; Akoglu, Ali; Bose, Tamal; Ryu, Bo.
2018 Conference on Design and Architectures for Signal and Image Processing, DASIP 2018. IEEE Computer Society, 2018. p. 88-93 8596977 (Conference on Design and Architectures for Signal and Image Processing, DASIP; Vol. 2018-October).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - GPU Based Quarter Spectral Correlation Density Function
AU - Marshall, Scott
AU - Vanhoy, Garrett
AU - Akoglu, Ali
AU - Bose, Tamal
AU - Ryu, Bo
PY - 2018/12/31
Y1 - 2018/12/31
N2 - In this study we investigate the parallelization of a key feature extraction method called spectral correlation density (SCD) function, which is used in signal classification systems particularly under low signal-to-noise ratio conditions for classifying numerous signals. In order to reduce the computation complexity of the SCD function, we introduce a method called Quarter SCD (QSCD) that allows extracting features of a given signal by processing only quarter of the input signal data. We then parallelize the QSCD by targeting general purpose graphics processing unit (GPU) through architecture specific optimization strategies. We present experimental evaluations on identifying the parallelization configuration for maximizing the efficiency of the program architecture in utilizing the threading power of the GPU architecture. We show that algorithmic and architecture specific optimization strategies result with improving the throughput of the state of the art GPU based Full SCD from 120 signals/second to 2719 signals/second.
AB - In this study we investigate the parallelization of a key feature extraction method called spectral correlation density (SCD) function, which is used in signal classification systems particularly under low signal-to-noise ratio conditions for classifying numerous signals. In order to reduce the computation complexity of the SCD function, we introduce a method called Quarter SCD (QSCD) that allows extracting features of a given signal by processing only quarter of the input signal data. We then parallelize the QSCD by targeting general purpose graphics processing unit (GPU) through architecture specific optimization strategies. We present experimental evaluations on identifying the parallelization configuration for maximizing the efficiency of the program architecture in utilizing the threading power of the GPU architecture. We show that algorithmic and architecture specific optimization strategies result with improving the throughput of the state of the art GPU based Full SCD from 120 signals/second to 2719 signals/second.
KW - GPGPU
KW - signal classification
KW - spectral correlation density
UR - http://www.scopus.com/inward/record.url?scp=85061378419&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85061378419&partnerID=8YFLogxK
U2 - 10.1109/DASIP.2018.8596977
DO - 10.1109/DASIP.2018.8596977
M3 - Conference contribution
AN - SCOPUS:85061378419
T3 - Conference on Design and Architectures for Signal and Image Processing, DASIP
SP - 88
EP - 93
BT - 2018 Conference on Design and Architectures for Signal and Image Processing, DASIP 2018
PB - IEEE Computer Society
ER -