Service manual

WLAN TX Measurements
R&S
®
FSV-K91/91n/91ac/91p
25Operating Manual 1176.7649.02 ─ 04
The transition to the frequency domain is achieved by an FFT of length 64. The FFT is
performed symbol-wise for every of the "nof_symbols" symbols of the payload. The
calculated FFTs are described by r
l,k
with:
●
l = [1, nof_symbols] as the symbol index
●
k = [–31, 32] as the channel index
In case of an additive white Gaussian noise (AWGN) channel the FFT is described by
[4], [5]
kl
phasephasej
klkl
neHgaKr
kl
common
l
kl
,
(
,mod
)timing(
,
)(
,


Equation (10) (3 - 1)
with:
●
k
mod
: the modulation-dependant normalization factor
●
a
l,k
: the symbol of sub-carrier k at symbol l
●
g
l
: the gain at the symbol l in relation to the reference gain g = 1 at the long symbol
(LS)
●
H
k
: the channel frequency response at the long symbol (LS)
●
l
(common)
: the common phase drift phase of all sub-carriers at symbol l (see Equa-
tion (11))
●
phase
l,k
(timing)
: the phase of sub-carrier k at symbol l caused by the timing drift (see
Equation (11))
●
n
l,k
: the independent Gaussian distributed noise samples
The common phase drift in Equation (10) is given by:
Equation (11) (3 - 2)
with
●
N
s
= 80: the number of Nyquist samples of the symbol period
●
N = 64: the number of Nyquist samples of the useful part of the symbol
●
Δ f
rest
: the (not yet compensated) frequency deviation
● dϒ
l
: the phase jitter at the symbol l
In general, the coarse frequency estimate Δ
coarse
(see) Signal processing of the IEEE
802.11a application) is not error-free. Therefore the remaining frequency error Δf
rest
represents the frequency deviation in r
l,k
not yet compensated. Consequently, the over-
all frequency deviation of the device under test (DUT) is calculated by:
Δf = Δ
coarse
+ Δf
rest
The only motivation for dividing the common phase drift in Equation (11) into two parts
is to be able to calculate the overall frequency deviation of the DUT.
Signal Processing of the IEEE 802.11a Application