Search

The article presents an improved and modified version of the M-FOCal Underdetermined System Solver (FOCUSS) algorithm, which can be efficiently used for sparse and locally smooth solutions. The new algorithm is provided with the overlapping blocks and is shown to be much more accurate and efficient than the previous one.

An improved whitening scheme for blind source separation, a novel biquadratic contrast function for extraction of independent sources and an efficient alternative method for joint implementation of a set of approximate diagonalization-structural matrices, is presented. The results have shown that the performance of the improved whitening scheme is much better than the classical whitening methods.

A novel sequential algorithm for the blind separation of a class of periodic source signals is introduced, which exploits the temporal periodic nature of the source signals, and thus performs separation based only on second-order statistics. The proposed algorithm is also shown to mitigate the limitation that the equivariant adaptive source separation (EASI) algorithm and separate at most one Gaussian distributed source.

The underdetermined blind source separation (BSS) using a two-stage sparse representation approach is discussed. It is seen that several probability inequalities and probability estimates for the recoverability issue are established based on the condition and various types of source sparsity.

A new decomposition method for multichannel blind deconvolution of nonminimum-phase systems, for decomposing a doubly finite impulse response filter into a cascade form of two filters, is presented. To illustrate the validity and favorable learning performance of the proposed algorithms, simulations are illustrated.

Two algorithms for single- and multichannel equalizations and deconvolution tasks that self-whiten the input signals within the coefficient updates are described. The results showed that the algorithms exhibit quasi-Newton coefficient adaptation behavior locally about an equalizing or deconvolving solution. In addition, the complexities of the simplest algorithm were found to be only 50% greater than the corresponding least-mean-square algorithms.

Statistical analyses facilitated the derivation of two adaptive network structures for liner-phase temporal and spatial cleaning of multichannel signals. Simulations of network operation indicate the viability of the approach for decorrelating signals in space and time. The network combines an a posteriori error criterion and a system for spatial and temporal decorrelation. These methods can be useful in a variety of communications, control and signal processing applications.

In this paper, we present a new filter decomposition method for multichannel blind deconvolution of nonminimum-phase systems. With this approach, we decompose a doubly finite impulse response filter into a cascade form of two filters: a causal finite impulse response (FIR) filter and an anticausal FIR filter. After introducing a Lie group to the manifold of FIR filters, we discuss geometric properties of the FIR filter manifold. Using the nonholonomic transform, we derive the natural gradient on the FIR manifold. By simplifying the mutual information rate, we present a very simple cost function for blind deconvolution of nonminimum-phase systems. Subsequently, the natural gradient algorithms are developed both for the causal FIR filter and for the anticausal FIR filter. Simulations are presented to illustrate the validity and favorable learning performance of the proposed algorithms. [ABSTRACT FROM AUTHOR]