Common analysis and display functions
R&S
®
FSVA3000/ R&S
®
FSV3000
585
User Manual 1178.8520.02 ─ 08
Logarithmic averaging is recommended if sinewave signals are to be clearly visible
against noise since with this type of averaging noise suppression is improved while the
sinewave signals remain unchanged.
For noise or pseudo-noise signals, the positive peak amplitudes are decreased in loga-
rithmic averaging due to the characteristic involved. The negative peak values are
increased relative to the average value. If the distorted amplitude distribution is aver-
aged, a value is obtained that is smaller than the actual average value. The difference
is -2.5
dB.
This low average value is usually corrected in noise power measurements by a 2.5
dB
factor. Therefore the R&S
FSV/A offers the selection of linear averaging. The trace
data is linearized before averaging, then averaged and logarithmized again for display
on the screen. The average value is always displayed correctly irrespective of the sig-
nal characteristic.
Trace smoothing
A Video Bandwidth Filter (VBW) is a hardware-based method of smoothing the trace
(see also
Chapter 6.5.1.2, "Smoothing the trace using the video bandwidth"
on page 465). However, other sweep and bandwidth settings can be coupled to the
VBW. For some signals, a VBW may not be freely selectable to obtain the required
smoothing effect. Therefore, a software-based trace smoothing function is also availa-
ble.
(Software-based)
smoothing
is a way to remove anomalies visually in the trace that
can distort the results. The smoothing process is based on a moving average over the
complete measurement range. The number of samples included in the averaging proc-
ess (the
aperture
size) is variable and is a percentage of all samples that the trace con-
sists of.
Figure 7-13: Sample size included in trace smoothing
Effects of smoothing on post-processing functions
Note that in Spectrum mode, all functions performed after the sweep, such as limit
checks, markers, or channel power measurements, are based on the smoothed trace
data. Thus, the results differ from results based on the original trace.
You can turn trace smoothing on and off for all traces individually and compare, for
example, the raw and the smooth trace.
Linear smoothing is based on the following algorithm:
Trace configuration
