Complete Blind Subspace Deconvolution
نویسنده
چکیده
Cocktail-party Problems (increasing generality): • Independent component analysis (ICA) [1, 2]: onedimensional sound sources. • Independent subspace analysis (ISA) [3]: independent groups of people. • Blind source deconvolution (BSD) [4]: one-dimensional sound sources and echoic room. • Blind subspace deconvolution (BSSD) [5]: independent source groups and echoes. Separation Theorem: • ISA ([3], proof for certain distribution types [5]): ISA = ICA + clustering. It forms the basis of the state-of-the-art ISA algorithms. • Undercomplete BSSD (uBSSD): – uBSSD = temporal concatenation + ISA [5]: but ‘high dimensional’ associated ISA problem. – uBSSD = LPA (linear predictive approximation) + ISA [6]. Based on: an undercomplete polynomial matrix has a polynomial matrix left inverse (with prob. one).
منابع مشابه
A novel framework method for non-blind deconvolution using subspace images priors
Non-blind deconvolution has been an active challenge in the research fields of computer vision and computational photography. However, most existing deblurring methods conduct direct deconvolution only on the degraded image and are sensitive to noise. To enhance the performance of non-blind deconvolution, we propose a novel framework method by exploiting different sparse priors of subspace imag...
متن کاملMultichannel Blind Deconvolution of the Short-Exposure Astronomical Images
In this paper we present a new multichannel blind deconvolution method based on so-called subspace technique that was originally proposed by Harikumar and Bresler. When at least two differently degraded images (channels) of the original scene are provided, the method is better conditioned than classical single channel ones. In comparison with earlier multichannel blind deconvolution techniques ...
متن کاملUndercomplete Blind Subspace Deconvolution
Here, we introduce the blind subspace deconvolution (BSSD) problem, which is the extension of both the blind source deconvolution (BSD) and the independent subspace analysis (ISA) tasks. We treat the undercomplete BSSD (uBSSD) case. Applying temporal concatenation we reduce this problem to ISA. The associated ‘high dimensional’ ISA problem can be handled by a recent technique called joint f-dec...
متن کاملGradient Adaptive Paraunitary Filter Banks for Spatio-Temporal Subspace Analysis and Multichannel Blind Deconvolution
Paraunitary filter banks are important for several signal processing tasks, including coding, multichannel deconvolution and equalization, adaptive beamforming, and subspace processing. In this paper, we consider the task of adapting the impulse response of a multichannel paraunitary filter bank via gradient ascent or descent on a chosen cost function. Our methods are spatio-temporal generaliza...
متن کاملFundamental Limits of Blind Deconvolution Part II: Sparsity-Ambiguity Trade-offs
Blind deconvolution is an ubiquitous non-linear inverse problem in applications like wireless communications and image processing. This problem is generally ill-posed since signal identifiability is a key concern, and there have been efforts to use sparse models for regularizing blind deconvolution to promote signal identifiability. Part I of this two-part paper establishes a measure theoretica...
متن کامل