Search

We present a general nonlinear Bayesian filter for high-dimensional state estimation using the theory of reproducing kernel Hilbert space (RKHS). By applying the kernel method and the representer theorem to perform linear quadratic estimation in a functional space, we derive a Bayesian recursive state estimator for a general nonlinear dynamical system in the original input space. Unlike existing nonlinear extensions of the Kalman filter where the system dynamics are assumed known, the state-space representation for the Functional Bayesian Filter (FBF) is completely learned online from measurement data in the form of an infinite impulse response (IIR) filter or recurrent network in the RKHS, with universal approximation property. Using a positive definite kernel function satisfying Mercer’s conditions to compute and evolve information quantities, the FBF exploits both the statistical and time-domain information about the signal, extracts higher-order moments, and preserves the properties of covariances without the ill effects due to conventional arithmetic operations. We apply this novel kernel adaptive filtering (KAF) to recurrent network training, chaotic time-series estimation and cooperative filtering using Gaussian and non-Gaussian noises, and inverse kinematics modeling. Simulation results show FBF outperforms existing Kalman-based algorithms. [ABSTRACT FROM AUTHOR]

This letter introduces a new framework for quantifying predictive uncertainty for both data and models that relies on projecting the data into a gaussian reproducing kernel Hilbert space (RKHS) and transforming the data probability density function (PDF) in a way that quantifies the flow of its gradient as a topological potential field (quantified at all points in the sample space). This enables the decomposition of the PDF gradient flow by formulating it as a moment decomposition problem using operators from quantum physics, specifically Schrödinger's formulation. We experimentally show that the higher-order moments systematically cluster the different tail regions of the PDF, thereby providing unprecedented discriminative resolution of data regions having high epistemic uncertainty. In essence, this approach decomposes local realizations of the data PDF in terms of uncertainty moments. We apply this framework as a surrogate tool for predictive uncertainty quantification of point-prediction neural network models, overcoming various limitations of conventional Bayesian-based uncertainty quantification methods. Experimental comparisons with some established methods illustrate performance advantages that our framework exhibits. [ABSTRACT FROM AUTHOR]

We propose a novel nearest-neighbors approach to organize and curb the growth of radial basis function network in kernel adaptive filtering (KAF). The nearest-instance-centroid-estimation (NICE) kernel least-mean-square (KLMS) algorithm provides an appropriate time-space tradeoff with good performance. Its centers in the input/feature space are organized by quasi-orthogonal regions for greatly simplified filter evaluation. Instead of using all centers to evaluate/update the function approximation at every new point, a linear search among the iteratively-updated centroids determines the partial function to be used, naturally forming locally-supported partial functionals. Under this framework, partial functionals that compose the adaptive filter are quickly stored/retrieved based on input, each corresponding to a specialized “spatial-band” subfilter. The filter evaluation becomes the update of one of the subfilters, creating a content addressable filter bank (CAFB). This CAFB is incrementally updated for new signal applications with mild constraints, always using the past-learned partial filter sums, opening the door for transfer learning and significant efficiency for new data scenarios, avoiding training from scratch as have been done since the invention of adaptive filtering. Using energy conservation relation, we show the sufficient condition for mean square convergence of the NICE-KLMS algorithm and establish the upper and lower bounds of steady-state excess-mean-square-error (EMSE). Simulations on chaotic time-series prediction demonstrate similar levels of accuracy as existing methods, but with much faster computation involving fewer input samples. Simulations on transfer learning using both synthetic and real-world data demonstrate that NICE CAFB can leverage previously learned knowledge to related task or domain. [ABSTRACT FROM AUTHOR]

Goal: We demonstrate an algorithm to automatically learn the time-limited waveforms associated with phasic events that repeatedly appear throughout an electroencephalogram. Methods: To learn the phasic event waveforms we propose a multiscale modeling process that is based on existing shift-invariant dictionary learning algorithms. For each channel, waveforms at different temporal scales are learned based on the assumption that only a few waveforms occur in any window of the time-series, but the same waveforms reoccur throughout the signal. Once the waveforms are learned, the timing and amplitude of the phasic event occurrences are estimated using matching pursuit. To summarize the waveforms learned across multiple channels and subjects, we analyze their frequency content, their similarity to Gabor–Morlet wavelets, and perform shift-invariant k-means to cluster the waveforms. A prototype waveform from each cluster is then tested for differential spatial patterns between different motor imagery conditions. Results: On multiple human EEG datasets, the learned waveforms capture key characteristics of signals they were trained to represent, with a consistency in waveform morphology and frequency content across multiple training sections and initializations. On multichannel datasets, the spatial amplitude patterns of the waveforms are also consistent and can be used to distinguish different modalities of motor imagery. Conclusion: We explored a methodology that can be used for modeling the recurrent waveforms in EEG traces. Significance: The methodology automatically identifies the most frequent phasic event waveforms in EEG, which could then be used as features for automatic evaluation and comparison of EEG during sleep, pathology, or mentally engaging tasks. [ABSTRACT FROM PUBLISHER]

This paper reviews methodologies for analyzing neural potentials via frequency, time–frequency, or wavelet representations, and adaptive models that estimate the signal's spatial or temporal structure. The fundamental assumptions of each method are discussed. In particular, the Fourier transform is contrasted with overcomplete representations, which are able to precisely delineate the timing and/or frequency of neural events. Finally, a novel approach that combines overcomplete representations with adaptive signal models is presented. This approach describes a continuous signal as a linear combination of reoccurring waveforms, referred to as phasic events, which are often associated with neural processing. The new methodology automatically learns the reoccurring waveforms and quantifies the neural potentials by the set of amplitudes and timings. [ABSTRACT FROM AUTHOR]

The self-organizing map (SOM) is one of the popular clustering and data visualization algorithms and has evolved as a useful tool in pattern recognition, data mining since it was first introduced by Kohonen. However, it is observed that the magnification factor for such mappings deviates from the information-theoretically optimal value of 1 (for the SOM it is 2/3). This can be attributed to the use of the mean square error to adapt the system, which distorts the mapping by oversampling the low probability regions. In this work, we first discuss the kernel SOM in terms of a similarity measure called correntropy induced metric (CIM) and empirically show that this can enhance the magnification of the mapping without much increase in the computational complexity of the algorithm. We also show that adapting the SOM in the CIM sense is equivalent to reducing the localized cross information potential, an information-theoretic function that quantifies the similarity between two probability distributions. Using this property we propose a kernel bandwidth adaptation algorithm for Gaussian kernels, with both homoscedastic and heteroscedastic components. We show that the proposed model can achieve a mapping with optimal magnification and can automatically adapt the parameters of the kernel function. [ABSTRACT FROM AUTHOR]

Monitoring heart activity from electrocardiograms (ECG) is crucial to avoid unnecessary fatalities; therefore, detection of QRS complex is fundamental to automated ECG monitoring. Continuous, portable 24/7 ECG monitoring requires wireless technology with constraints on power, bandwidth, area, and resolution. In order to provide continuous remote monitoring of patients and fast transmission of data to medical personnel for instantaneous intervention, we propose a methodology that converts analog inputs into pulses for ultralow power implementation. The signal encoding scheme is the time-based integrate and fire (IF) sampler from which a set of signal descriptors in the pulse domain are proposed. Furthermore, a logical decision rule for QRS detection based on morphological checking is derived. The proposed decision logic depends exclusively on relational and logical operators resulting in ultrafast recognition and can be implemented using combinatorial logic hardware to guarantee power consumption orders of magnitude lower than any microprocessor device. The algorithm was evaluated using the MIT-BIH arrhythmia database and results show that our algorithm performance is comparable to the state-of-the art software-based detection. [ABSTRACT FROM PUBLISHER]

Recent studies suggest that the minimum error entropy (MEE) criterion can outperform the traditional mean square error criterion in supervised machine learning, especially in nonlinear and non-Gaussian situations. In practice, however, one has to estimate the error entropy from the samples since in general the analytical evaluation of error entropy is not possible. By the Parzen windowing approach, the estimated error entropy converges asymptotically to the entropy of the error plus an independent random variable whose probability density function (PDF) corresponds to the kernel function in the Parzen method. This quantity of entropy is called the smoothed error entropy, and the corresponding optimality criterion is named the smoothed MEE (SMEE) criterion. In this paper, we study theoretically the SMEE criterion in supervised machine learning where the learning machine is assumed to be nonparametric and universal. Some basic properties are presented. In particular, we show that when the smoothing factor is very small, the smoothed error entropy equals approximately the true error entropy plus a scaled version of the Fisher information of error. We also investigate how the smoothing factor affects the optimal solution. In some special situations, the optimal solution under the SMEE criterion does not change with increasing smoothing factor. In general cases, when the smoothing factor tends to infinity, minimizing the smoothed error entropy will be approximately equivalent to minimizing error variance, regardless of the conditional PDF and the kernel. [ABSTRACT FROM AUTHOR]

Abstract: This paper addresses the problem of semi-blindly extracting one single desired signal using a priori information about its higher order temporal structure. Our approach is based on the maximization of the autocorrentropy function for a given time delay. The a priori information is quantified as a time delay and a size for a Gaussian kernel to set the free parameters in the correntropy function. Those values provide information which allows the proposed method to adapt a demixing vector to extract the desired signal without the indeterminacy of the permutation problem in blind source separation. Moreover, this method is different from those for Independent Component Analysis that separate all the available sources, which, in some problems, is not desirable or computationally possible. Since correntropy can be interpreted as a generalization of correlation, we demonstrate that it is a suitable measure for studying the temporal behavior of higher order statistics of a signal. Also, the flexibility brought by the kernel size selection allows the user to choose the range of statistics he is interested in. We show in simulations that correntropy achieve better or equal separation than other linear methods proposed in the literature for source extraction based on temporal structures. [Copyright &y& Elsevier]

The minimum error entropy (MEE) criterion has been receiving increasing attention due to its promising perspectives for applications in signal processing and machine learning. In the context of Bayesian estimation, the MEE criterion is concerned with the estimation of a certain random variable based on another random variable, so that the error's entropy is minimized. Several theoretical results on this topic have been reported. In this work, we present some further results on the MEE estimation. The contributions are twofold: (1) we extend a recent result on the minimum entropy of a mixture of unimodal and symmetric distributions to a more general case, and prove that if the conditional distributions are generalized uniformly dominated (GUD), the dominant alignment will be the MEE estimator; (2) we show by examples that the MEE estimator (not limited to singular cases) may be non-unique even if the error distribution is restricted to zero-mean (unbiased). [ABSTRACT FROM AUTHOR]

This paper presents a parameterized version of the stochastic information gradient (SIG) algorithm, in which the error distribution is modeled by generalized Gaussian density (GGD), with location, shape, and dispersion parameters. Compared with the kernel-based SIG (SIG-Kernel) algorithm, the GGD-based SIG (SIG-GGD) algorithm does not involve kernel width selection. If the error is zero-mean, the SIG-GGD algorithm will become the least mean p-power (LMP) algorithm with adaptive order and variable step-size. Due to its well matched density estimation and automatic switching capability, the proposed algorithm is favorably in line with existing algorithms. [ABSTRACT FROM AUTHOR]

Abstract: Point process modeling of neural spike recordings has the potential to capture with high specificity the information contained in spike time occurrence. In Brain–Machine Interfaces (BMIs) the neural tuning characteristic assessed from neural spike recordings can distinguish neuron importance in terms of its modulation with the movement task. Consequently, it improves generalization and reduces significantly computation in previous decoding algorithms, where models reconstruct the kinematics from recorded activities of hundreds of neurons. We propose to apply information theoretical analysis based on an instantaneous tuning model to extract the important neuron subsets for point process decoding on BMI. The cortical distribution of extracted neuron subsets is analyzed and the statistical decoding performance using subset selection is studied with respect to different number of neurons and compared to the one by the full neuron ensemble. With much less computation, the extracted importance neurons provide comparable kinematic reconstructions compared to the full neuron ensemble. The performance of the extracted subset is compared to the random selected subset with same number of neurons to further validate the effectiveness of the subset-extraction approach. [Copyright &y& Elsevier]

Walter Freeman in his classic 1975 book "Mass Activation of the Nervous System" presented a hierarchy of dynamical computational models based on studies and measurements done in real brains, which has been known as the Freeman's K model (FKM). Much more recently, liquid state machine (LSM) and echo state network (ESN) have been proposed as universal approximators in the class of functionals with exponential decaying memory. In this paper, we briefly review these models and show that the restricted K set architecture of KI and KII networks share the same properties of LSM/ESNs and is therefore one more member of the reservoir computing family. In the reservoir computing perspective, the states of the FKM are a representation space that stores in its spatio-temporal dynamics a short-term history of the input patterns. Then at any time, with a simple instantaneous read-out made up of a KI, information related to the input history can be accessed and read out. This work provides two important contributions. First, it emphasizes the need for optimal readouts, and shows how to adaptively design them. Second, it shows that the Freeman model is able to process continuous signals with temporal structure. We will provide theoretical results for the conditions on the system parameters of FKM satisfying the echo state property. Experimental results are presented to illustrate the validity of the proposed approach. [ABSTRACT FROM AUTHOR]

Abstract: Nonlinearity tests have become an essential step in system analysis and modeling due to the computational demands and complexity of analysis involved in nonlinear modeling. Standard nonlinear measures are either too complicated to estimate accurately (such as Lyapunov exponents and correlation dimension), or not able to capture sufficient but not necessary conditions of nonlinearity (such as time asymmetry). Correntropy is a kernel-based similarity measure which contains the information of both statistical and temporal structure of the underlying dataset. The capability of preserving nonlinear characteristics makes correntropy a suitable candidate as a measure for determining nonlinear dynamics. Moreover, since correntropy makes use of kernel methods, its estimation is computationally efficient. Using correntropy as the test statistic, nonlinearity tests based on the null hypothesis that signals of interest are realizations of linear Gaussian stochastic processes are carried out via surrogate data methods. Experiments performed on linear Gaussian, linear non-Gaussian, and nonlinear systems with varying in-band noise levels, data lengths, and kernel sizes confirm that correntropy can be employed as a discriminative measure for detecting nonlinear characteristics in time series. Results of tests performed on data collected from natural systems are in agreement with findings in time series analysis literature. [Copyright &y& Elsevier]

Abstract-A new spatiotemporal filtering method for single-trial event-related potential (ERP) estimation is proposed. Instead of attempting to model the entire ERP waveform, the method relies on modeling ERP component descriptors (amplitude and latency) thru the spatial diversity of multichannel recordings, and thus, it is tailored to extract signals in negative SNR conditions. The model allows for both amplitude and latency variability in the ERP component under investigation. The extracted ERP component is constrained through a spatial filter to have minimal distance (with respect to some metric) in the temporal domain from a userdesigned template component. The spatial filter may be interpreted as a noise canceller in the spatial domain. Study with both simulated data and real cognitive ERP data shows the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: The minimum average correlation energy (MACE) filter is a well known correlation filter for object recognition. Recently, a nonlinear extension to the MACE filter using the correntropy function in feature space has been introduced. Correntropy is a positive definite function that generalizes the concept of correlation by utilizing higher order moment information of signal structure. Since the MACE is a spatial matched filter for an image class, the correntropy MACE (CMACE) can potentially improve its performance. Both the MACE and CMACE are basically memory-based algorithms and due to the high dimensionality of the image data, the computational cost of the CMACE filter is one of the critical issues in practical applications. We propose to use a dimensionality reduction method based on random projections (RP), which has emerged as a powerful method for dimensionality reduction in machine learning. We apply the CMACE filter with random projection (CMACE-RP) to face recognition and show that it indeed outperforms the traditional linear MACE in both generalization and rejection abilities with small degradation in performance, but great savings in storage and computational complexity. [Copyright &y& Elsevier]

Abstract: In this paper, a multiple discrete quasi-sliding mode (QSM) control scheme is proposed for a general class of nonlinear discrete time systems with unknown dynamical equations, provided that input–output data is available for system identification. The self-organizing map (SOM) is employed to divide the state space into local regions such that it associates the operating region where a local linear model is the winner with a local quasi-sliding mode controller (QSMC). Switching of the controllers is done synchronously with the active local linear model that tracks the different operating conditions. The simulation results show that the proposed controller outperforms tracking the desired trajectory in noisy environments either with a global controller or simpler controllers based on multiple models. [Copyright &y& Elsevier]

Abstract: In this work, we analyze algorithms for adaptive filtering based on non-linear cost function of the error, which we named non-linear even moment (NEM) algorithms. We assume that this non-linear function can be generally described in a Taylor series as a linear combination of the even moments of the error. NEM is a generalization of the well-known least mean square (LMS). We study the NEM convergence behavior and derive equations for misadjustment and convergence. We found a good approximation for the theoretical results and we show that there are various combinations of the even moments which yields better results than the LMS as well as other algorithms proposed in the literature. [Copyright &y& Elsevier]

The next generation of aircraft will have dynamics that vary considerably over the operating regime. A single controller will have difficulty to meet the design specifications. In this paper, a self-organizing map (SOM)-based local linear modeling scheme of an unmanned aerial vehicle (UAV) is developed to design a set of inverse controllers. The SOM selects the operating regime depending only on the embedded output space information and avoids normalization of the input data. Each local linear model is associated with a linear controller, which is easy to design. Switching of the controllers is done synchronously with the active local linear model that tracks the different operating conditions. The proposed multiple modeling and control strategy has been successfully tested in a simulator that models the LoFLYTE UAV. [ABSTRACT FROM AUTHOR]

A method for modeling the flutter response of a thin winged aircraft is presented, A hybrid physical-adaptive modeling framework is proposed to separate the autoregressive and moving average flutter components. Adaptive oscillators set at the structural free-vibration modal frequencies of the wing represent the structure (the autoregressive component). The moving average filters represent signal changes caused by aerodynamic forces encountered during flight. Connected in series, these modules form a hybrid model for wing flutter under changing flight conditions. The moving average component is trained to predict the signal produced by an accelerometer at the wing tip. The residual is segmented using an analysis of variance. The resulting family of linear moving average models provides a synthesis of the wing's response over time. Our analysis shows that this modeling paradigm performs well with data taken at increasing Mach numbers in level flight, Network parameters are shown to correlate linearly with Mach. Network components can themselves be predicted as functions of Mach number, allowing the accelerometer signals to be predicted with a high degree of accuracy. The paradigm is further validated by using an adapted model with a separate set of data from a different but similar flight condition. [ABSTRACT FROM AUTHOR]

Recent publications have proposed various information-theoretic learning (ITL) criteria based on Renyi's quadratic entropy with nonparametric kernel-based density estimation as alternative performance metrics for both supervised and unsupervised adaptive system training. These metrics, based on entropy and mutual information, take into account higher order statistics unlike the mean-square error (MSE) criterion. The drawback of these information-based metrics is the increased computational complexity, which underscores the importance of efficient training algorithms. In this paper, we examine familiar advanced-parameter search algorithms and propose modifications to allow training of systems with these ITL criteria. The well known algorithms tailored here for ITL include various improved gradient-descent methods, conjugate gradient approaches, and the Levenberg-Marquardt (LM) algorithm. Sample problems and metrics are presented to illustrate the computational efficiency attained by employing the proposed algorithms. [ABSTRACT FROM AUTHOR]

Recently, we have proposed the minimum error entropy (MEE) criterion as an information theoretic alternative to the widely used mean square error criterion in supervised adaptive system training. For this purpose, we have formulated a nonparametric estimator for Renyi's entropy that employs Parzen windowing. Mathematical investigation of the proposed entropy estimator revealed interesting insights about the process of information theoretical learning. This new estimator and the associated criteria have been applied to the supervised and unsupervised training of adaptive systems in a wide range of problems successfully. In this paper, we analyze the structure of the MEE performance surface around the optimal solution, and we derive the upper bound for the step size in adaptive linear neuron (ADALINE) training with the steepest descent algorithm using MEE. In addition, the effects of the entropy order and the kernel size in Parzen windowing on the shape of the performance surface and the eigenvalues of the Hessian at and around the optimal solution are investigated. Conclusions from the theoretical analyses are illustrated through numerical examples. [ABSTRACT FROM AUTHOR]

In this paper, we develop a spatio-temporal memory that blends properties from long and short-term memory and is motivated by reaction diffusion mechanisms. The winning processing element of a self-organizing network creates traveling waves on the output space that gradually attenuate over time and space to diffuse temporal information and create localized spatio-temporal neighborhoods for clustering. The novelty of the model is in the creation of time varying Voronoi tessellations anticipating the learned input signal dynamics even when the cluster centers are fixed. We test the method in a robot navigation task and in vector quantization of speech. This method performs better than conventional static vector quantizers based on the same data set and similar training conditions. [Copyright &y& Elsevier]

Proposes the quadratic Renyi's error entropy as an alternative cost function for supervised adaptive system training. Estimation of Renyi's entropy with parzen windowing; Gradient training algorithm for information potential; Kernel size and global optimization by functional smoothing; Conclusion.

Presents a study which investigated error-entropy-minimization algorithm for supervised training of nonlinear adaptive systems. Method of the study; Results and discussion; Conclusion.

Presents information on a study which described a novel valley seeking clustering algorithm using an information theoretic measure to estimate the cost of partitioning the data set. Divergence measures; Clustering evaluation function; Optimization algorithm; Experimental results.

Presents information on a study which described the application of support vector machines for synthetic aperture radar automatic target recognition. Learning criteria for empirical and structural risk minimization; Experimental results; Conclusions.

Presents a study which investigated the algorithm that performs competitive principal component analysis of a time series for signal processing. Temporal principal component analysis; Practical implementation issues; Simulations; Conclusions.

Provides information on a study which proposed a global optimization strategy for training adaptive systems such as neural networks and adaptive filters. Analysis of learning with added noise in the desired signal; Speaker verification using gamma neural networks; Simulation results of the study.

Develops an adaptive two-step paradigm for the super-resolution of optical images. Discussion on Kernel-based approaches; Architecture for Kernel family super-resolution; Conclusion.

Focuses on the use of linear and nonlinear adaptive networks in target discrimination in synthetic aperture radar (SAR) imagery. Neural networks for pattern classification; Performance of analysis using pattern recognition perspective; Detection of the Neyman-Person criterion.

This contribution gives the information on a useful application of principal component analysis (PCA) in the field of electroencephalogram (EEG) and laser-Doppler signal processing. The principal components are estimated by a neural network (NN) approach. [ABSTRACT FROM AUTHOR]

Presents information the development of a discriminator called the `quadratic gamma discriminator' for synthetic aperture radar imagery based in a 2-D extension of the gamma functions. Reference to parameter constant false-alarm rate; Why the discriminator was called the quadratic gamma discriminator; Features of the discriminator.

Focuses on the convolution of sinc kernel with infinite sequence of a periodic function for discrete signals. Numerically stable formulation of sinc kernel; Conclusions.

Presents information on the development of the least mean square (LMS) algorithm, which converge in a statistically sense to global minimum of the mean square error (MSE) objective function, while indicating approximation by a variable noise source to the infinite impulse response (IIR-LMS) algorithm. Performance improvement of the IIR-LMS algorithm over the Steiglitz-McBride algorithm; Investigation of the use of stochastic global-optimization methods.

This article introduces the present issue of the journal "IEEE Transactions on Neural Networks." Nonlinear signal processing using neural networks inherited the dominant mean-squared error criteria from linear adaptive filtering theory due to a number of appealing properties exhibited by second-order statistical optimality criteria under the linear model. Information theory has a long history in the design of optimal communication systems and coding. Numerous useful theoretical results relating information theoretic learning (ITL). The aim of this special issue is to present the current state of the art in the application of information theoretic concepts and approaches to learning, adaptation, and neural network theories by active experts working in the broadly defined area of ITL.

The article outlines the most salient features of the journal "IEEE Transactions on Biomedical Engineering" (T-BME) during 2005. Also, the article addresses some important modifications that will add to the quality and scope of the journal. All T-BME issues in 2005 were printed in time. Moreover, there was a decrease in the paper backlog in 2005. It is expected that with the increase of the page budget for 2006, it should drop to a residual value at the end of the year. A new type of manuscript called T-BME Letters, the first of its kind in bioengineering will be started from the year 2006. Manuscripts submitted to T-BME under the category of letters will have a special treatment.

This article discusses about the 2004 issues of the journal "IEEE Transactions on Biomedical Engineering" (T-BME). The T-BME issues in 2004 were always on time, which means they were mailed before the first of the month. The paper backlog grew in 2004 to 120 papers, for reasons that are not totally clear to the author, but are a combination of the rapid review cycle and increase in the number of submissions to the journal. One should be grateful to the EMB society for agreeing to a 500-page budget increase in mid 2004 which trimmed the number of papers in the backlog, but unfortunately did not solve it totally. As a consequence you have Probably noticed the increased thickness of the T-BME issues.

The article presents editorial, published in the January 1, 2004 issue of the biomedical journal, IEEE Transactions on Biomedical Engineering (T-BME). He is completing his third year of service as Editor in Chief of the T-BME. He is starting by highlighting some of the important achievements during the past year. The impact factor of T-BME for 2002 is 1.66, up from 1.56 in 2001. These gains in efficiency were predicated on the dedicated, knowledgeable, and hardworking group of associate editors, reviewers and office staff. The breadth of T-BME continues to grow, but most of the published papers come from modeling and signal processing.

Introduces a series of articles on statistical signal processing.