Table 1 Using different kernels, Gaussian, Polynomial and Hermite functions, the NL-decorrelation is applied on source and mixed signals
From: Enhancement of acoustic tomography using spatial and frequency diversities
Signals | Mixture model | NL-decorrelation of sources | NL-decorrelation of mixed signals |
|---|---|---|---|
i.i.d uniform PDF | Instantaneous | Kernel ‘Gaussian’ − 23.4 | Kernel ‘Gaussien’ −5.8319 |
Kernel ‘poly’ − 25.5 | Kernel ‘poly’ 8.1 | ||
Kernel ‘hermite’ − 22.4 | Kernel ‘hermite’ −20.4 | ||
Four acoustic signals | Instantaneous convolutive | Kernel ‘poly’ − 33.4 | Kernel ‘poly’ 3.2 |
2000 samples | Kernel ‘poly’ −14.9817 | ||
Four acoustic signals | Instantaneous convolutive | Kernel ‘poly’ − 31.3 | Kernel ‘poly’ 8.8 |
4∗105 samples | Kernel ‘poly’ −13.2 |