How Coherence is Used i n C urve Fitting
Coherence is a function of frequency. Each frequency point in the coherence function is a real
value between 0.0 and 1.0. A value of 1.0 implies that noise does not interfere with the
measurement at that frequency. Averaging allows the measured coherence to converge on the true
coherence, which is a measure of the noise in the system, not the measurement. Additional
averaging gives a better estimate of the coherence and a smaller measurement variance. Thus, as
the coherence at a specific frequency converges to a value farther below 1.0, additional averaging is
needed to obtain a small measurement variance at that frequency.
The results of either the s-domain or z-domain curve fitter can be improved by reducing the
measurement variance across the fit region.
.999 5 ,
across the fit region is
average value of the coherence drops to .95, then 200 averages may be needed for a good fit. There
is not a specific relationship between the number o f averages and the average value of the
coherence needed to obtain a good fit.
Note
Coherence is not used by the curve fitter under any of the following conditions:
•
When the data comes from a swept-sine measurement.
displayed if
•
When the data is synthesized.
•
When the measured data is the result of fewer than four averages.
As
an example, if the average value of the coherence
only 10 averages may be needed to obtain a good fit. However, if the
LAST MEAS
is selected.
How Coherence is Used in Curve Fitting
(
Coherence is still
)
Curve Fit
1 6-35