Search

Nonnegative Matrix Factorization (NMF) models are widely used to recover linearly mixed nonnegative data. When the data is made of samplings of continuous signals, the factors in NMF can be constrained to be samples of nonnegative rational functions, which allow fairly general models; this is referred to as NMF using rational functions (R-NMF). We first show that, under mild assumptions, R-NMF has an essentially unique factorization unlike NMF, which is crucial in applications where ground-truth factors need to be recovered such as blind source separation problems. Then we present different approaches to solve R-NMF: the R-HANLS, R-ANLS and R-NLS methods. From our tests, no method significantly outperforms the others, and a trade-off should be done between time and accuracy. Indeed, R-HANLS is fast and accurate for large problems, while R-ANLS is more accurate, but also more resources demanding, both in time and memory. R-NLS is very accurate but only for small problems. Moreover, we show that R-NMF outperforms NMF in various tasks including the recovery of semi-synthetic continuous signals, and a classification problem of real hyperspectral signals., Comment: 13 pages

The canonical polyadic decomposition (CPD) can be used to extract meaningful components from a tensor. Most existing optimization methods for fitting the CPD use as cost function the least-squares distance between the tensor and its CPD. While the minimum of this cost function coincides with the maximum likelihood estimator for data with additive i.i.d. Gaussian distributed noise, for other noise distributions, better-suited cost functions exist. For such cost functions, first-order, gradient-based optimization methods have been proposed. However, (approximate) second-order methods, which additionally use information from the Hessian of the cost function to achieve faster convergence, are still largely unexplored. In this paper, we generalize the Gauss–Newton nonlinear least-squares algorithm to twice differentiable entry-wise cost functions. The low-rank structure of the problem is exploited to keep the computational cost low. As a special case, $\beta$ -divergence cost functions are examined. We show that quadratic convergence can be obtained close to the solution with a reasonable extra cost in memory and computation time, making the proposed method particularly useful when high accuracy of the decomposition is desired.

IEEE The Canonical Polyadic Decomposition (CPD) plays an important role for signal separation in array processing. The CPD model requires arrays composed of several displaced but identical subarrays. Consequently, it is less appropriate for more complex array geometries. In this paper we explain that coupled CPD allows a much more flexible modeling that can handle multiple shift-invariance structures, i.e., arrays that can be decomposed into multiple but not identical displaced subarrays. Both deterministic and generic identifiability conditions are presented. We also point out that, under mild conditions, the signal separation problem can in the exact case be solved by means of an eigenvalue decomposition. This is similar to ESPRIT, although the working conditions are much more relaxed. Borrowing tools from algebraic geometry, we derive generic uniqueness bounds for L-shaped, frame-shaped and triangular-shaped arrays that come close to bounds that are necessary for uniqueness. Recognizing multiple shift-invariance can be a bit of an art by itself. We explain that any centro-symmetric array processing problem can be interpreted in terms of a coupled CPD. In addition, we demonstrate that the coupled CPD model allows us to significantly relax the far-field assumption commonly used in CPD-based array processing. ispartof: IEEE Transactions on Signal Processing vol:66 issue:14 pages:3665-3680 status: published

Many real-life signals can be described in terms of much fewer parameters than the actual number of samples. Such compressible signals can often be represented very compactly with low-rank matrix and tensor models. The authors have adopted this strategy to enable large-scale instantaneous blind source separation. In this paper, we generalize the approach to the blind identification of large-scale convolutive systems. In particular, we apply the same idea to the system coefficients of finite impulse response systems. This allows us to reformulate blind system identification as a structured tensor decomposition. The tensor is obtained by applying a deterministic tensorization technique called segmentation on the observed output data. Exploiting the low-rank structure of the system coefficients enables a unique identification of the system and estimation of the inputs. We obtain a new type of deterministic uniqueness conditions. Moreover, the compactness of the low-rank models allows one to solve large-scale problems. We illustrate our method for direction-of-arrival estimation in large-scale antenna arrays and neural spike sorting in high-density microelectrode arrays.

Several applications in biomedical data processing, telecommunications, or chemometrics can be tackled by computing a structured tensor decomposition. In this paper, we focus on tensor decompositions with two or more block-Hankel factors, which arise in blind multiple-input-multiple-output (MIMO) convolutive system identification. By assuming statistically independent inputs, the blind system identification problem can be reformulated as a Hankel structured tensor decomposition. By capitalizing on the available block-Hankel and tensorial structure, a relaxed uniqueness condition for this structured decomposition is obtained. This condition is easy to check, yet very powerful. The uniqueness condition also forms the basis for two subspace-based algorithms, able to blindly identify linear underdetermined MIMO systems with finite impulse response.

Several problems in signal processing have been formulated in terms of the Canonical Polyadic Decomposition of a higher-order tensor with one or more Vandermonde constrained factor matrices. We first propose new, relaxed uniqueness conditions. We explain that, under these conditions, the number of components may simply be estimated as the rank of a matrix. We propose an efficient algorithm for the computation of the factors that only resorts to basic linear algebra. We demonstrate the use of the results for various applications in wireless communication and array processing.