As a result of the calculation of expanded uncertainty, the measurement uncertainty defines an interval within which the measurand lies. Expanded uncertainty is generally calculated by using z-distribution in conjunction with coverage factor k . But, by using z-distribution, correctly finding the interval which the measurand lies within, with a specified confidence level, requires too many measurements which must be done repeatedly. In an ideal case, the number of measurements must approach infinity. Since this is not the case in practice, if the number of measurements is small, using Student’s factor tν,p instead of coverage factor k results in the definition of a relatively wider interval and more accurate results. This situation is shown, as a result of an analysis done on a biomedical system, by calculating the measurement uncertainty and the interval within which the measurand lies in a confidence level of 95%, comparing two different approaches.
Key Words: Biomedical calibration, Measurement uncertainty, t-distribution, z-distribution
Primary Language | English |
---|---|
Journal Section | Electrical & Electronics Engineering |
Authors | |
Publication Date | March 29, 2010 |
Published in Issue | Year 2007 Volume: 20 Issue: 3 |