2010/07/31 by Huy Nguyen, Huy T. Nguyen, Rong Zheng · 35 citations
Computer Science · Engineering · Mathematics · Neuroscience · #Algorithm #Artificial intelligence #Binary number #Blind Source Separation Techniques #Component (thermodynamics) #Computer science #EEG and Brain-Computer Interfaces #Gaussian #Independence (probability theory) #Independent component analysis #Latent variable #Mathematics #Matrix (chemical analysis) #Random variable #Sparse and Compressive Sensing Techniques #Statistics #cs.IT #cs.NI #math.IT
paper · pdf · doi:10.1109/tsp.2011.2144975
published in IEEE Transactions on Signal Processing 59(7), 3168-3181 (Institute of Electrical and Electronics Engineers) · Manuscript submitted to IEEE Transactions on Signal Processing
arxiv created 2010/12/19 · openalex publication_date 2011/04/26 · arxiv updated 2015/05/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Independent component analysis (ICA) is a computational method for separating a multivariate signal into subcomponents assuming the mutual statistical independence of the non-Gaussian source signals. The classical independent components analysis (ICA) framework usually assumes linear combinations of independent sources over the field of real-valued numbers R. In this paper, we investigate binary ICA for or mixtures (bICA), which can find applications in many domains including medical diagnosis, multi-cluster assignment, Internet tomography and network resource management. We prove that bICA is uniquely identifiable under the disjunctive generation model, and propose a deterministic iterative algorithm to determine the distribution of the latent random variables and the mixing matrix. The inverse problem to infer the values of latent variables is also considered for noisy measurements. We conduct an extensive simulation study to verify the effectiveness of the propose algorithm and present examples of real-world applications where bICA can be applied.