EMD Performance Comparison: Single vs Double Floating Points
| Autor: | Slawomir J. Nasuto, O. Cadenas, Dawid Laszuk |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Floating point
QA76.76 QA76.87 010504 meteorology & atmospheric sciences Computer science Stability (learning theory) 020206 networking & telecommunications 02 engineering and technology Key issues 01 natural sciences Signal Hilbert–Huang transform Performance comparison 0202 electrical engineering electronic engineering information engineering Algorithm 0105 earth and related environmental sciences |
| Zdroj: | International Journal of Signal Processing Systems. :349-353 |
| ISSN: | 2315-4535 |
| DOI: | 10.18178/ijsps.4.4.349-353 |
| Popis: | Empirical mode decomposition (EMD) is a data-driven method used to decompose data into oscillatory components. This paper examines to what extent the defined algorithm for EMD might be susceptible to data format. Two key issues with EMD are its stability and computational speed. This paper shows that for a given signal there is no significant difference between results obtained with single (binary32) and double (binary64) floating points precision. This implies that there is no benefit in increasing floating point precision when performing EMD on devices optimised for single floating point format, such as graphical processing units (GPUs). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|