U1084A Datasheet, PDF (17/20 Pages) Keysight Technologies – Acqiris High-Speed PCIe Digitizers with On-Board Signal Processing

Powerful peak detection algorithm
Peak interpolation
Due to the incredible sampling speeds of the U1084A, the
peak analysis firmware uses a very simple yet powerful
algorithm to detect signal peaks at up to the full 4 GS/s
sampling rate.
Nine samples are used to validate if a sample corresponds
to a peak; the sample itself, along with the four samples
before and after that sample are considered.
To improve the measurement, peak time and amplitude
are determined using an interpolation routine. Fitting a
12-bit quadratic spline to the 3 points around the extrema
found by the peak detection algorithm, the position of a
peak maximum is found and defined by its time (Tmax) and
amplitude (Vmax).
This interpolation method provides measurement with
improved time and vertical resolution. The resulting his-
togram that is created, can contain up to 2 Mbins of data,
with each bin width down to 1/16th of the sampling period,
peak amplitudes encoded with the bins, to 1/16th of the
ADC resolution.
Adjustable through software commands, the user can
balance the increased timing and vertical resolution against
increased memory usage and reduced measurement
throughput, to best fit requirements.
Figure 4. Peak detection algorithm in action
The sample will be considered as a peak, if within the 9
samples analyzed: a rising edge is found before the sample
that exceeds the programmable delta ∆Rise, a falling edge
is found after the sample that exceeds the programmable
delta ∆Fall, and the sample is the maximum of all points
between the two defined rising and falling edges. There is
additional verification to make sure that only one peak is
found within the rising and falling edge region.
Using this algorithm, peaks can be found on-the-fly in
signals with frequencies up to half the sample rate.
Max
P-1
V _max
P
V_p
Quadratic Spline
T _{max T p}
P+1
Figure 5. Peak interpolation with quadratic spline
17