Minimum BER Criterion and Adaptive Moment Estimation Based Enhanced ICA for Wireless Communications

Recently, blind source separation (BSS) methods have gained significant attention in the area of signal processing. Independent component analysis (ICA) and second-order blind identification (SOBI) are two popular BSS methods that have been applied to modal identification of mechanical and structural systems. Published results by several researchers have shown that ICA performs satisfactorily for systems with very low levels of structural damping, for example, for damping ratios of the order of 1% critical. For practical structural applications with higher levels of damping, methods based on SOBI have shown significant improvement over ICA methods. However, traditional SOBI methods suffer when nonstationary sources are present, such as those that occur during earthquakes and other transient excitations. In this paper, a new technique based on SOBI, called the modified cross-correlation method, is proposed to address these shortcomings. The conditions in which the problem of structural system identification can be posed as a BSS problem is also discussed. The results of simulation described in terms of identified natural frequencies, mode shapes, and damping ratios are presented for the cases of synthetic wind and recorded earthquake excitations. The results of identification show that the proposed method achieves better performance over traditional ICA and SOBI methods. Both experimental and large-scale structural simulation results are included to demonstrate the applicability of the newly proposed method to structural identification problems.

<span>The purpose of this research is to observe the effectiveness of independent component analysis (ICA) method for denoising raw EEG signals based on word imagination, which will be used for word classification on unspoken speech. The electroencephalogram (EEG) signals are signals that represent the electrical activities of the human brain when someone is doing activities, such as sleeping, thinking or other physical activities. EEG data based on the word imagination used for the research is accompanied by artifacts, that come from muscle movements, heartbeat, eye blink, voltage and so on. In previous studies, the ICA method has been widely used and effective for relieving physiological artifacts. Artifact to signal ratio (ASR) is used to measure the effectiveness of ICA in this study. If the ratio is getting larger, the ICA method is considered effective for clearing noise and artifacts from the EEG data. Based on the experiment, the obtained ASR values from 11 subjects on 14 electrodes amounted are within the range of 0,910 to 1,080. Thus, it can be concluded that ICA is effective for removing artifacts from EEG signals based on word imagination.</span>

Nonnegative matrix factorization (NMF) is known as a parts-based linear representation for nonnegative data. This method has been applied for blind source separation (BSS) when the sources are nonnegative. This paper presents a new NMF method for independent component analysis (ICA), which is useful for BSS without the nonnegativity constraint. Using this method, we transform the sources by their cumulative distribution functions (CDFs) and perform the nonparametric quantization to construct a nonnegative matrix where each entry represents the joint probability density of two transformed signals. The NMF procedure is accordingly realized to find the ICA demixing matrix. The independence between sources is maximized towards attaining the uniformity in the joint probability density. In the experiments on the separation of signal and music signals, we show the effectiveness of the proposed NMF-ICA compared to the infomax ICA and FastICA algorithms.

Recently, principal component analysis (PCA) and independent component analysis (ICA) have attracted increasing attention for fault detection in industrial processes. However, industrial processes usually have multimodal characteristics, in which case, the performance of traditional PCA and ICA methods is poor. A fault detection method based on the local entropy double subspace (LEDS) is proposed that effectively supervises the multimodal process with multiple steady modes. The proposed method uses the local probability density (LPD) to eliminate the multimodal characteristics of the data. Furthermore, local entropy (LE) is applied to mine the statistical information associated with the process data. The process variables are classified into Gaussian and non-Gaussian subspaces by the Kolmogorov–Smirnov (KS) test. PCA and ICA models are built in Gaussian and non-Gaussian subspaces, respectively, for fault detection. A Bayesian decision method is employed to transform the detection results into fault probabilities, and the detection results are combined with the final statistical information. This method effectively eliminates the multicentre structure of the data and improves the fault detection performance of traditional PCA and ICA. The proposed method is applied to the Tennessee Eastman multimodal process. The simulation results show that the fault detection rate of LEDS is better than the rates of traditional methods when the false alarm rate is low for most faults.

The electroencephalogram (EEG) is a set of data measured by electrodes placed on the scalp and is often under the influences of artifacts. Mental EEG is recorded when a person performs different mental tasks. In this article, we separated the mental EEG signals into independent components with individual meanings based on Independent Component Analysis (ICA) method, and the EEG was reconstructed by excluding those components related to ocular and line artifacts for further feature extraction. In the end, we made a simulation to compare denoising of mental EEG using ICA with the result using wavelet packet analysis. The ICA method was found to be more efficient than the classical methods.

Automatic facial feature extraction is one of the most important and attempted problems in computer vision. It is a necessary step in face recognition, facial image compression. There are many methods have been proposed in the literature for the facial feature extraction task. However, all of them have still disadvantage such as not complete reflection about face structure, face texture. Therefore, a combination of different feature extraction methods which can integrate the complementary information should lead to improve the efficiency of feature extraction stage. In this paper we describe a methodology for improving the efficiency of feature extraction stage based on the association of two methods: geometric feature based method and Independent Component Analysis (ICA) method. Comparison of two methods of facial feature extraction: geometric feature based method combined with PCA method (called GPCA) versus geometric feature based method combined with ICA method (called GICA) on CalTech dataset has demonstrated the efficiency of GICA method. Our results show that GICA achieved good performance 96.57% compared to 94.70% of GPCA method. Furthermore, we compare two methods mentioned above on our dataset, with performance of GICA being 98.94% better 96.78% of GPCA method. The experiment results have confirmed the benefits of the association geometric feature based method and ICA method in facial feature extraction.KeywordsFace recognitionindependent component analysis (ICA)principal component analysis (PCA)geometric features

Iterative multiuser receivers constitute an effective solution for transmission over Multiple Access Interference (MAI) infested channels, when invoking a combined multiuser detector and channel decoder. Most reduced-complexity methods in this area use the Complex-valued Minimum Mean Squared Error (CMMSE) Multiuser Detector (MUD). Since the desired output of BPSK systems is real-valued, minimizing the Mean Square Error (MSE) between the beamformer's desired output and the real part of the beamformer output has the potential of significantly improving the attainable Bit Error Rate (BER) performance. We refer to this MMSE design as the Real-valued MMSE (RMMSE) receiver. In this paper, we explore a new Soft-Input Soft-Output (SISO) interference cancellation multiuser detection algorithm based on the novel Minimum BER (MBER) criterion. We demonstrate that the MBER turbo receiver outperforms both the CMMSE and the RMMSE algorithms, particularly in so-called `overloaded' beamforming systems, where the number of receiver antennas is lower than the number of users supported.

SCTICA: Sub-packet constrained temporal ICA method for fMRI data analysis

A novel deep learning model based on the ICA and Riemannian manifold for EEG-based emotion recognition

The analysis and characterization of atrial tachyarrhythmias requires, in a previous step, the extraction of the atrial activity (AA) free from ventricular activity and other artefacts. This contribution adopts the blind source separation (BSS) approach to AA estimation from multilead electrocardiograms (ECGs). Previously proposed BSS methods for AA extraction--e.g., independent component analysis (ICA)--exploit only the spatial diversity introduced by the multiple spatially-separated electrodes. However, AA typically shows certain degree of temporal correlation, with a narrowband spectrum featuring a main frequency peak around 3.5-9 Hz. Taking advantage of this observation, we put forward a novel two-step BSS-based technique which exploits both spatial and temporal information contained in the recorded ECG signals. The spatiotemporal BSS algorithm is validated on simulated and real ECGs from a significant number of atrial fibrillation (AF) and atrial flutter (AFL) episodes, and proves consistently superior to a spatial-only ICA method. In simulated ECGs, a new methodology for the synthetic generation of realistic AF episodes is proposed, which includes a judicious comparison between the known AA content and the estimated AA sources. Using this methodology, the ICA technique obtains correlation indexes of 0.751, whereas the proposed approach obtains a correlation of 0.830 and an error in the estimated signal reduced by a factor of 40%. In real ECG recordings, we propose to measure performance by the spectral concentration (SC) around the main frequency peak. The spatiotemporal algorithm outperforms the ICA method, obtaining a SC of 58.8% and 44.7%, respectively.

Functional positron emission tomography (fPET) is a neuroimaging method involving continuous infusion of 18-F-fluorodeoxyglucose (FDG) radiotracer during the course of a PET examination. Compared with the conventional bolus administration of FDG in a static PET scan, which provides an average glucose uptake into the brain over an extended period of up to 30 ​min, fPET offers a significantly higher temporal resolution to study the dynamics of glucose uptake. Several earlier studies have applied fPET to investigate brain FDG uptake and study its relationship with functional magnetic resonance imaging (fMRI). However, due to the unique characteristics of fPET signals, modelling of the fPET signal is a complex task and poses challenges for accurate interpretation of the results from fPET experiments. This study applied independent component analysis (ICA) to analyse resting state fPET data, and to compare the performance of ICA and the general linear model (GLM) for estimation of brain activation in response to tasks. The fPET signal characteristics were compared using GLM and ICA methods to model fPET data from a visual activation experiment. Our aim was to evaluate GLM and ICA methods for analysing task fPET datasets, and to apply ICA methods to the analysis of resting state fPET datasets. Using both simulation and in-vivo experimental datasets, we show that both ICA and GLM methods can successfully identify task related brain activation. We report fPET metabolic resting state brain networks revealed by application of the fPET ICA method to a cohort of 28 healthy subjects. Functional PET provides a unique method to map dynamic changes of glucose uptake in the resting human brain and in response to extrinsic stimulation.

Two distinct topics are dealt with. First, a new method for independent component analysis (ICA) has been constructed that exploits the invariance of criteria under component-wise scaling, which is intrinsic to ICA. This practical and simple ICA method is called the nested Newton's method. When the number of the channel of observation is less than a certain level, factor analysis (FA) is ineffective (bound for FA). The target of this paper is these cases. Three of many concrete advantages of the nested Newton's method are addressed. i) It is robust against Gaussian noise and outperforms existing methods, such as JADE and Fast ICA, especially under Gaussian noise conditions. ii) It is highly stable globally. iii) Each step resolves itself into two-dimensional (2-D) matrix problems. There is thus no need to deal with gigantic matrices, which means that fewer computational resources are required. Second, a method called "post factor analysis (post-FA)" is described that is aimed to be useful as post-processing for ICA. Although it is functionally similar to conventional FA, post-FA is a completely new method and is more powerful than conventional FA in compensation for its stronger assumption that there are mutually independent sources behind observations. By fully making use of this assumption, post-FA is capable of estimating the noise variance beyond the known limit for FA. Furthermore, it improves the accuracy of ICA to a considerable extent. Any ICA algorithm without prewhitening (pre-WH) or pre-factor-analysis (pre-FA) can be used for preprocessing, although the nested method is a good candidate.

A novel genetic algorithm (GA) assisted minimum bit error rate (MBER) beamforming technique is introduced. The performance of the proposed GAs is characterised by the probability density function (PDF) and the mean value of the achievable bit error rate (BER) at the beamformer's output. The results are also compared to the theoretical bounds. It is shown that GAs are suitable for MBER beamforming and that a trade-off between the complexity of the GA and its robustness against system parameter variations can be found.

This paper proposes a robust independent component analysis (ICA) approach for noise reduction. Noise reduction is a difficult problem in ICA model. In general signal processing applications, there is more than one interference signal which may have unknown characteristics. In these situations, traditional linear ICA may lead to poor results. Hence, noise reduction is preferred to be performed with nonlinear adaptive filtering. In this paper, a radial basis function network (RBFN) is employed to transform the observed signals into output space in a nonlinear manner. The weights of RBFN are updated by utilizing a modified fixed‐point algorithm. The proposed method has not only the capacity of recovering the mixed signals, but also reducing noise with unknown characteristics from observed signals. The simulation results and analysis show that the proposed algorithm is suitable for practical unsupervised noise reduction problem. © 2008 Wiley Periodicals, Inc. Electron Comm Jpn, 91(3): 45–52, 2008; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecj.10073

Epilepsy is a brain disorder that 1% of people population are suffering from. One of the proper non-invasive equipment for diagnosis and analysis of this disease is electroencephalogram (EEG) recordings. However, EEG signals are often contaminated with noises and artifacts that hide epileptic signals of interest. Independent Component Analysis (ICA) is a common Blind Source Separation (BSS) method to denoise EEG signals. ICA has been proved as a worthwhile method to separate the signals of interest from noise and artifacts; nevertheless, it also has some weaknesses. In this work, to improve ICA performance in denoising context, we present an algorithm based on combination of ICA and Time Varying AutoRegressive (TVAR) model for denoising of interictal EEG signals. TVAR model is used serially after ICA method for interictal spike enhancement. The coefficients of TVAR model are estimated using Kaiman filter. The results indicate the proposed algorithm is better than ICA in terms of performance for very low Signal-to-Noise Ratio (SNR) values.