Table of Contents
Advertisement
W
R
X
S
AVE
UNNER
I
An effective way to reduce these ef
conditions the effective sampling rate of the oscilloscope and therefore determines the frequency resolution and
span at which spectral analysis can be carried out.
FFT Algorithms
A summary of the algorithms used in the oscilloscope's FFT computation is given here in a few steps:
The data are multiplied by the selected w
FFT is computed, using a fast implementation of the DFT (Discrete Fourier Transform):
where: x
is a complex array whose real part is the modified source time domain waveform, and whose imaginary
k
part is 0; X
is the resulting complex frequency-domain waveform;
n
x
and X
.
k
n
The generalized FFT algorithm, as implemented here, works on N, which need not be a power of 2.
The resulting complex vector X
the loss of the signal energy due to windowing
isolated spectrum peaks.
The real part of X
is symmetric around the Nyquist frequency, that is
n
R
= R
n
N-n
while the imaginary part is asymmetric, that is
I
= –I
n
N-n
The energy of the signal at a frequency n is distributed e
spectrum; the energy at frequency
The first half of the spectrum (Re, Im), from 0 to the Nyquist frequency is kept for further processing and dou
in amplitude:
R'
= 2 x R
_0 n < N/2
n
n
I' = 2 x I __0 n < N/2
n
n
The resultant waveform is computed for the spectrum type selected.
If "Magnitude" is selected, the magnitude of the complex vector is computed as:
Steps 1 to 5 lead to the following result:
An AC sine wave of amplitude 1.0 V with an integral number of periods N
the rectangular window, results in a fundamental peak of 1.0 V magnitude in the spectrum at frequency N
f. However, a DC component of 1.0 V, transformed with the rectangular
magnitude at 0 Hz.
The waveforms for the other available spectrum types are computed as follows:
Phase: angle = arctan (I
M ≤ M
angle = 0
n
Where M
is the minimum magnitude, fixed at about 0.001 of the full scale at any gain setting, below which the
min
angle is not well defined.
The dBm Power Spectrum:
138
ERIES
fects is to maximize the acquisition record length. Record length directly
indow function.
is divided by the cohe
n
. This compensation provides accurate amplitude values for
0 is completely contained in the 0 term.
/R
> M
)_M
n
n
n
min
min
rent gain of the window function, in order to compensate for
qually between the first and the second halves of the
in the time window, transformed
p
window, results in a peak of 2.0 V
; and N is the number of points in
WRXi-OM-E Rev C
bled
with
x D
elta
p
Advertisement
Table of Contents
