A Linear Non-Gaussian Acyclic Model for Causal Discovery
The Journal of Machine Learning Research
Blind separation of piecewise stationary non-Gaussian sources
Signal Processing
IEEE Transactions on Signal Processing
On the convergence of ICA algorithms with symmetric orthogonalization
IEEE Transactions on Signal Processing
Performance analysis of the fastica algorithm in ICA-based co-channel communication system
WiCOM'09 Proceedings of the 5th International Conference on Wireless communications, networking and mobile computing
IEEE Transactions on Neural Networks
The deflation-based FastICA estimator: statistical analysis revisited
IEEE Transactions on Signal Processing
Speeding up FastICA by mixture random pruning
ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions
ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
Comparative speed analysis of fastICA
ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
Blind instantaneous noisy mixture separation with best interference-plus-noise rejection
ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
IEEE Transactions on Signal Processing
Expert Systems with Applications: An International Journal
A comparative study of ICA algorithms for ECG signal processing
ACAI '11 Proceedings of the International Conference on Advances in Computing and Artificial Intelligence
Effect of source kurtosis on MIMO information rate
Digital Signal Processing
Testing significance of mixing and demixing coefficients in ICA
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Consistency and asymptotic normality of FastICA and bootstrap FastICA
Signal Processing
Cramér-Rao bound for circular complex independent component analysis
LVA/ICA'12 Proceedings of the 10th international conference on Latent Variable Analysis and Signal Separation
Distributional convergence of subspace estimates in FastICA: a bootstrap study
LVA/ICA'12 Proceedings of the 10th international conference on Latent Variable Analysis and Signal Separation
On the convergence of ICA algorithms with weighted orthogonal constraint
Digital Signal Processing
Hi-index | 35.70 |
The FastICA or fixed-point algorithm is one of the most successful algorithms for linear independent component analysis (ICA) in terms of accuracy and computational complexity. Two versions of the algorithm are available in literature and software: a one-unit (deflation) algorithm and a symmetric algorithm. The main result of this paper are analytic closed-form expressions that characterize the separating ability of both versions of the algorithm in a local sense, assuming a "good" initialization of the algorithms and long data records. Based on the analysis, it is possible to combine the advantages of the symmetric and one-unit version algorithms and predict their performance. To validate the analysis, a simple check of saddle points of the cost function is proposed that allows to find a global minimum of the cost function in almost 100% simulation runs. Second, the Crame´r-Rao lower bound for linear ICA is derived as an algorithm independent limit of the achievable separation quality. The FastICA algorithm is shown to approach this limit in certain scenarios. Extensive computer simulations supporting the theoretical findings are included.