Digital Signal Processing Reference
In-Depth Information
Figure 2.10 Estimated source signal constellations obtained by jade (a) and GUT
method gut5 (b). BPSK, 8-QAM and circular Gaussian source signals are clearly
discernible on the left, middle and right plots.
second-order non-circular. The above random communications signals are symmetric,
and hence any pair of scatter and spatial pseudo-scatter matrices can be employed in
the definition of the GUT matrix and DOGMA matrix. The array outputs are corrupted
by additive noise
n
with a circular complex Gaussian distribution with covariance
matrix
C
(
n
)
¼ s
n
I
. The signal to noise ratio (SNR) is 10 log
10
(
s
s
=s
n
)
¼
20 dB and
the number of snapshots is
n ¼
2000. The estimated source signal constellations
obtained by
jade
and
gut5
are shown in Figure 2.10. Both of the methods were able
to separate the sources as BPSK, 8-QAM and circular Gaussian source signals are
clearly discernible. Table 2.2. shows the average performance of all the ICA estimators
over 100 simulation runs.
jade
and
kmg
are performing the best but GUT methods are
not far behind.
fobi
and
d1
, however, stand out as they do not quite reach the same level
of performance as the others.
Table 2.2 Mean values of
2
10 log
10
[PI(G)] computed from 100 array
snapshot data sets
gut1
gut2
gut3
gut4
gut5
Fica
Jade
Cfpa
Kmg
d1
Fobi
Without
outliers
17.5
15.2
16.7
16.7
16.9
16.5
18.7
16.8
18.4
12.2
12.7
With
outliers
8.5
2.7
11.4
16.6
16.8
3.6
2.5
2.6
2.4
12.2
2.6
Search WWH ::
Custom Search