Binary Independent Component Analysis With or Mixtures

Abstract

Abstract-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 realvalued 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 concerning inferring the values of latent variables are also considered along with 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.

Binary Blind Identification of Wireless Transmission Technologies for Wide-Band Spectrum Monitoring

“…Then, we introduce binary independent component analysis (bICA) [13], an inferring methodology, which plays a key role in BTI.…”

“…Next, we will give a brief introduction to bICA. A detailed description of bICA and its applications can be found in [13].…”

“…. , p 2 m −1 ] is the active probability vector of G. In [13], we have shown that p can be calculated from p 0 and p * as follows,…”

“…where d is the Hamming distance and G :,i is the i'th column in G. Due to the possible differences in the number and the order of inferred independent components in G andĜ, we need to perform the structure matching process (see details in [13]) before estimatingH g .…”

“…Note that performance of bICA is only hampered by large number of observable variables, not by large number of hidden variables as shown in[13].This full text paper was peer reviewed at the direction of IEEE Communications Society subject matter experts for publication in the IEEE Globecom 2011 proceedings.…”