The probabilistic approaches for real-time systems are based on the estimation of the probabilistic-WCET distribution. Such estimation is naturally subject to errors, caused by both systematic and estimation uncertainties. To solve this problem, statistical tests are applied on the resulting distribution to check whether such errors affect or not the output validity. In this paper, we show that the reliability of these tests depends on the statistical power that must be estimated in order to select the proper sample size. This a priori analysis is required to obtain a reliable result of the probabilistic-WCET.
Work-in-Progress: Why statistical power matters for probabilistic real-time
F. Reghenzani;W. Fornaciari
2019-01-01
Abstract
The probabilistic approaches for real-time systems are based on the estimation of the probabilistic-WCET distribution. Such estimation is naturally subject to errors, caused by both systematic and estimation uncertainties. To solve this problem, statistical tests are applied on the resulting distribution to check whether such errors affect or not the output validity. In this paper, we show that the reliability of these tests depends on the statistical power that must be estimated in order to select the proper sample size. This a priori analysis is required to obtain a reliable result of the probabilistic-WCET.File | Dimensione | Formato | |
---|---|---|---|
stat-power-short-esweek.pdf
accesso aperto
:
Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione
465.14 kB
Formato
Adobe PDF
|
465.14 kB | Adobe PDF | Visualizza/Apri |
a3-reghenzani_emsoft19_pub.pdf
accesso aperto
Descrizione: versione pubblicata
:
Publisher’s version
Dimensione
432.82 kB
Formato
Adobe PDF
|
432.82 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.