Identifiability issues in noisy ICA

@article{Davies2004IdentifiabilityII,
  title={Identifiability issues in noisy ICA},
  author={M. Davies},
  journal={IEEE Signal Processing Letters},
  year={2004},
  volume={11},
  pages={470-473}
}
We consider the identifiability of the statistical model for noisy independent component analysis showing that while the mixing process is identifiable, the noise covariance is only partially so. This raises questions as to the performance of certain maximum-likelihood algorithms for blind source separation in the presence of noise. 
Highly Cited
This paper has 37 citations. REVIEW CITATIONS