Batch Learning Techniques for Blind Signal Separation

Blind signal separation method is the method that the source signals are obtained by separating only the mixed signals. Generally, separation precision is lower when the basic method of blind signal separation is used for the sound signals. It is the cause that the power of the sound signals is concentrated in the low-frequency band. For the problem, we already considered that the method of the blind signal separation using QMF. Specifically, the mixed signals are divided into two frequency bands, and we perform suitable separation for each two bands. As the result, the conventional method has higher separation precision than the basic method. However, in the low-frequency band, power spectrum is not constant for frequency. Therefore, it is expected that the separation precision is higher than the conventional method by performing the suitable separation for each of signals based on the division to several narrow frequency bands on the low-frequency band. Then, in this paper, we propose the method that has higher separation precision and we evaluate the proposed method by computer simulation.

An adaptive semiblind signal separation algorithm for the separation of a class of arbitrarily distributed but temporally correlated source signals is introduced in this paper. The proposed adaptive algorithm is based only on second-order statistical(SOS) information and exploits the assumption that the source signals are not strictly spatial statistically independent, and temporal do not identical distributed too. Furthermore, the novel adaptive batch blind signal separation(BSS) method give an adaptive algorithm for correlated source signals. In order to get the ideal separation, the source signals must have different power spectra shapes. Simulation results for real image recordings show that the proposed algorithm has the ability to restore source signals, and its performance is comparable to that of the BSS algorithm such as AMUSE and JADE algorithm. The proposed algorithm is also shown to alleviate the limitation that the classic independent component analysis(ICA) algorithm can separate at most one Gaussian distributed source.

This paper introduces a new source separation technique exploiting the time coherence of the source signals. The proposed approach relies only on stationary second order statistics. Blind Signal Separation (BSS) method using trilinear decomposition is proposed in this paper. Simulation results reveal that our proposed algorithm has the better blind signal separation performance than joint diagonalization method. Our proposed algorithm does not require whitening processing. Moreover, our proposed algorithm works well in the underdetermined condition, where the number of sources exceeds than the number of sensors.

Blind Signal Separation II. Linear, Convolutive Combinations: II. Linear, Convolutive Combinations

The main goal of automatic speech and speaker recognition (ASSR) is to transcribe natural speech and recognize its speaker. Recognizing a spoken sentence is obviously a knowledgeintensive process, which must take into account all variable information about the speech communication process. Out of several approaches, artificial intelligence approach has been observed to give remarkable results by visualizing, analysing and finally making a decision on the acoustic patterns. The present chapter deals with two basic processes for ASSR: modelling process and understanding process. The modelling process involves three stages: enhancement, segmentation and pre-processing. The understanding phase involves the recognition of the speech and the speaker through the known knowledge-based model. It has been observed from the literature that, in any speech processing system, because of channel coupling, many source signals mix together. Thus there may be a chance of variability in speech signals affecting the performance of the speech systems due to some factors like: background and channel noise, electrical noise from different sources, meaningless sounds (a sneeze) or filler words (‘uh’ or ‘um’) between words, different speaking rate, mood and styles of speakers. Due to this some signals may not be observed and may result to out-of-vocabulary (OOV) word problem during recognition or understanding process. Thus speech enhancement has played a vital role in the development of a perfect acoustic model or noise – free artificial word model (AWM) and vowel diphthong model (VDM). To process for speech enhancement in frequency – domain, the speech signal has to be split into frames using blind signal separation (BSS) method (Yilmaz and Richard 2004; Yamashita and Hirai 2004; Araki et al 2002; Murata et al 2001; G. J. Jang and Lee 2003; F. Bach and Jordan 2005; B. A. Pearl Mutter and Olsson 2006). Further the spectrum of the background noise has to be estimated by subtracting the noise spectrum from the spectrum of the frame (M. Berouli et al 1979; Yasser Ghanbari and Md. Reza 2004). It has been observed that some of the noise remains in the spectrum when the value of noise is greater than its mean. At the same time, some of the speech spectrum gets removed when noise is greater than the actual value of noise. In the spectrum this produces negative values and have to be set to zero. The overall effect puts a noise in the output signal known as

The compound noises in circuit deteriorate the output signals. This paper proposes a mutual information based separation (MISEP) method of the blind signal separation (BSS) problem for the circuits with the instant mixing model. Both the linear and the nonlinear circuits are taken as the circuits under experiment, and both the signals and the noises are regarded as the inputs of the circuits. Experimental results verify the effectiveness of the proposed method, because the different inputs of the circuits are identified by the time domain waveforms and the power spectrum characteristics.

Blind separation of non-stationary sources using continuous density hidden Markov models

A new method to perform blind separation of chaotic signals is articulated in this paper, which takes advantage of the underlying features in the phase space for identifying various chaotic sources. Without incorporating any prior information about the source equations, the proposed algorithm can not only separate the mixed signals in just a few iterations, but also outperforms the fast independent component analysis (FastICA) method when noise contamination is considerable.

Blind source separation (BSS) consists in processing a set of observed mixed signals to separate them into a set of original components. Most of the current blind separation methods assumes that the source signals are “as statistically independent as possible” given the observed data. In many real-world situations, however, this hypothesis does not hold. In order to cope with such signals, a first geometric method was proposed that separates statistically dependent signals, provided that they are nonnegative and locally orthogonal.This paper presents a new geometric method for the separation of nonnegative source signals which relies on a working assumption that is weaker than local orthogonality. The separation problem is expressed as the identification of relevant facets of the data cone. After a rigorous proof of the proposed method, the details of the separation algorithm are given. Experiments on signals from various origins clearly show the efficiency of the new procedure.

Electroencephalogram (EEG) brain maps provide useful and reliable neurodiagnostic information. Accurate reconstruction of the data requires an efficient separation of the electrode signals. Although autoregressive (AR) spectrum estimation highly refines the signals, it cannot remove the effect of adjacent electrode signals. This report describes an efficient blind signal separation (BSS) method. The algorithm identifies the coefficients of an adaptive FIR filter by minimization of a cost function in terms of the corresponding fourth-order cumulants. Applying this method, the quality of the results is far superior to the traditional methods.

Suppressive interference is a common interference signal for synthetic aperture radar (SAR) that can seriously affect the target identification and imaging results of SAR. This paper proposes a method for suppressing suppressive jamming using blind source separation (BSS) for singular value and eigenvalue decomposition based on information entropy. First, we developed an airborne SAR imaging geometry model and a suppressive interference signal mixing model. Next, we perform blind signal separation of the interfered mixed signal by means of BSS based on singular value and eigenvalue decomposition. Then, we image the different signals we have extracted. Finally, we extract the features of the image domain for the separated signals and set the information entropy threshold by the difference of information entropy to identify the jamming signal and the source signal and obtain the source signal. This method uses eigenvalue and singular value decomposition for BSS and extracts the image domain features of the signal after BSS by information entropy and identifies the source signal by information entropy thresholding. This method compensates for the uncertainty in the decomposition of the signal by means of BSS. The signal loss is minimal and the similarity of the separated signal and the original signal is very high. Simulated and measured data demonstrate the feasibility of this algorithm.

Combining the advantages of local mean decomposition and variational Bayesian hidden Markov model, a blind source separation method for mechanical faults based on local mean decomposition and variational Bayesian hidden Markov model is proposed. In the proposed method, the non-stationary signals are decomposed into a series of production functions by the local mean decomposition, then the obtained production functions and the original observed signals are used to construct new observation signals, therefore the underdetermined blind source separation problem is transformed into the overdetermined blind source separation problem. Finally, the source signals are estimated by the variational Bayesian hidden Markov model method. The proposed method has been successfully applied to the blind separation of bearing faults, and the experiment results verify the effectiveness of the proposed method. The obtained outcome in this paper provides an effective separation method for mechanical fault diagnosis.

A new second-order method for blind signal separation from dynamic mixtures

Blind signal separation methods are diverse and almost studied in no noise situations, while the noise exists in practice. In this paper, firstly we designed an FIR filter and applied it to remove the added Gaussian noise in the observed signals. And then we improved the standard natural gradient algorithm (SNG) and named it as an improved natural gradient algorithm (ING). Lastly we used Fast ICA algorithm, the SNG and the ING to separate the mixed signals which were processed by the FIR filter. We contrasted and analogized the signal interference ration (SIR) of results of the three algorithms above. We obtained that it's helpful for separation to filter the observed signals under low signal to noise ration (SNR) circumstance, however under the high SNR circumstance, the filtering operation doesn't improve the separation property.

In this paper we propose a new algorithm for solving the blind source extraction (BSE) problem when the desired source signals are sparse. Previous approaches for solving this problem are based on the independent component analysis (ICA ) technique, that extracts a source signal by finding a separating vector that maximizes the non-Gaussianity of the extracted source signal. These algorithms are general purpose algorithms and are not designed specifically for extracting sparse signals. In this paper we propose a new algorithm for extracting sparse source signals. The proposed algorithm is based on finding a separating vector that maximizes the sparsity of the extracted source signal. In the proposed algorithm, a nonconvex objective function that measures the sparsity of the separated signal is locally replaced by a quadratic convex function. This results in an iterative algorithm in which a new estimate of the separating vector is obtained by solving an eigenvalue decomposition problem. A numerical example is presented to investigate the superiority of the proposed algorithm in comparison with one of the well known ICA algorithm for extracting sparse sources.