R&S NRP2, User Manual

Page: 92 / 437

Instrument Functions
R&S
®
NRP2
74 User Manual 1173.9157.02 ─ 03
are needed. Taking a 100 µs idle time of these sensors into consideration, a reduction of
the aperture time even below 500 µs will merely speed up measurements but significantly
increase noise.
Smoothing modulated signals
Continuous average-power measurements on signals with low-frequency modulation or
on pulsed signals with low repetition rates may suffer from fluctuations in the measure-
ment result, which are due to the acquisition time not being matched to the modulation
(pulse) period. Although it is possible with R&S
NRP-Zxx Power Sensors to set the aper-
ture time identical to a multiple integer of the period of the power envelope of the signal,
this approach fails with a varying or unknown period.
In these cases it is recommended to activate smoothing. Instead of simply summing up
the samples within the acquisition interval, the samples are individually weighted prior to
taking the mean. Samples at the beginning and at the end of the acquisition interval are
downscaled whereas the size of samples in the center of the acquisition interval is
increased (von Hann window). This way, fluctuations of the measurement result are
reduced almost independently from the degree of the aperture time being matched to the
modulation period. The only aspect that must be considered is that the aperture time has
to be set to at least two or three modulation (pulse) periods in order to reach an efficient
operation. Set aperture time accordingly, particularly if you are using R&S
NRP-
Z81/85/86 wideband power sensors, which feature very low default aperture times.
The noise content of the measurement result is enlarged by approximately 30 % when
smoothing is activated. To compensate for this, the averaging number should be doubled
compared to non-active smoothing. Therefore, smoothing may be not sufficient when
measuring very low powers.
Dropout tolerance
The "Dropout" parameter facilitates the identification of the burst end of modulated signals
(e.g. NADC). Without this parameter, the end of the burst might erroneously be detected
within the burst owing to the considerable modulation-specific power drops of such sig-
nals. "Dropout" is set at least for the duration of modulation-dependent power drops within
the burst; thus, the end of the burst cannot be detected until the set time has elapsed
after the power has dropped below the trigger threshold. If the trigger threshold is excee-
ded again during the dropout time, the process restarts from the beginning.
Timing diagram of burst signals
The timing diagram below shows the relation between the following parameters:
●
Trigger threshold and trigger event
●
Trigger event, exclude intervals and measurement interval
Although the signal frequently falls below the trigger threshold during the burst, an end
of the burst is not detected. The reason for this is that these drops are shorter than the
set dropout time.
The last falling edge in the burst is detected as the end of the burst since the trigger
threshold is no longer exceeded during the dropout time indicated below
Data acquisition and parameters