Determining the Convergence of Synchronous Measurements for Embedded Industrial Applications
| Autor: | David R. Buttsworth, John Leis |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Engineering
Noise measurement business.industry 020208 electrical & electronic engineering Spectral density 02 engineering and technology Relative stability symbols.namesake Control and Systems Engineering Control theory Gaussian noise 0202 electrical engineering electronic engineering information engineering symbols Entropy (information theory) 020201 artificial intelligence & image processing Electrical and Electronic Engineering business |
| Zdroj: | IEEE Transactions on Industrial Electronics. 64:7392-7398 |
| ISSN: | 1557-9948 0278-0046 |
| Popis: | One technique which may be employed in certain types of sensing application is the phase-sensitive or lock-in method. Since this method essentially trades off longer measurement time for a greater accuracy, it is sometimes necessary to capture samples over a long time period to ensure sufficient stability in the calculated parameter estimate. This may adversely affect safety-critical systems, and some method of determining the relative stability of the measured parameter is needed to permit this useful approach to be more widely employed. In this paper, we propose and evaluate a novel entropy-based method for ascertaining stability. The contributions of this paper are to first highlight the need for a convergence measure which is reliable, automatic, and easily computed; second, we propose one such measure, and place it on a theoretical footing; finally, we give results both with simulated Gaussian noise having a typical sensor power spectrum, as well as experimental results using the type of optical sensor which would benefit from the proposed method. It is demonstrated that this approach produces a reliable asymptotic figure for convergence, both under simulated noise conditions as well as with real measurements in the field. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|