Ehud Weinstein

New criteria for blind deconvolution of nonminimum phase systems (channels)

A necessary and sufficient condition for blind deconvolution (without observing the input) of nonminimum-phase linear time-invariant systems (channels) is derived. Based on this condition, several optimization criteria are proposed, and their solution is shown to correspond to the desired response. These criteria involve the computation only of second- and fourth-order moments, implying a simple tap update procedure. The proposed methods are universal in the sense that they do not impose any restrictions on the probability distribution of the (unobserved) input sequence. It is shown that in several important cases (e.g. when the additive noise is Gaussian), the proposed criteria are essentially unaffected. >

Parameter estimation of superimposed signals using the EM algorithm

A computationally efficient algorithm for parameter estimation of superimposed signals based on the two-step iterative EM (estimate-and-maximize, with an E step and an M step) algorithm is developed. The idea is to decompose the observed data into their signal components and then to estimate the parameters of each signal component separately. The algorithm iterates back and forth, using the current parameter estimates to decompose the observed data better and thus increase the likelihood of the next parameter estimates. The application of the algorithm to the multipath time delay and multiple-source location estimation problems is considered. >

Convergence analysis of LMS filters with uncorrelated Gaussian data

Statistical analysis of the least mean-squares (LMS) adaptive algorithm with uncorrelated Gaussian data is presented. Exact analytical expressions for the steady-state mean-square error (mse) and the performance degradation due to weight vector misadjustment are derived. Necessary and sufficient conditions for the convergence of the algorithm to the optimal (Wiener) solution within a finite variance are derived. It is found that the adaptive coefficient μ, which controls the rate of convergence of the algorithm, must be restricted to an interval significantly smaller than the domain commonly stated in the literature. The outcome of this paper, therefore, places fundamental limitations on the mse performance and rate of convergence of the LMS adaptive scheme.

Super-exponential methods for blind deconvolution

A class of iterative methods for solving the blind deconvolution problem, i.e. for recovering the input of an unknown possibly nonminimum-phase linear system by observation of its output, is presented. These methods are universal do not require prior knowledge of the input distribution, are computationally efficient and statistically stable, and converge to the desired solution regardless of initialization at a very fast rate. The effects of finite length of the data, finite length of the equalizer, and additive noise in the system on the attainable performance (intersymbol interference) are analyzed. It is shown that in many cases of practical interest the performance of the proposed methods is far superior to linear prediction methods even for minimum phase systems. Recursive and sequential algorithms are also developed, which allow real-time implementation and adaptive equalization of time-varying systems. >

Multi-channel signal separation by decorrelation

Identification of an unknown system and recovery of the input signals from observations of the outputs of an unknown multiple-input, multiple-output linear system are considered. Attention is focused on the two-channel case, in which the outputs of a 2*2 linear time invariant system are observed. The approach consists of reconstructing the input signals by assuming that they are statistically uncorrelated and imposing this constraint on the signal estimates. In order to restrict the set of solutions, additional information on the true signal generation and/or on the form of the coupling systems is incorporated. Specific algorithms are developed and tested. As a special case, these algorithms suggest a potentially interesting modification of Widrows (1975) least-squares method for noise cancellation, where the reference signal contains a component of the desired signal. >

Fundamental limitations in passive time delay estimation--Part I: Narrow-band systems

Time delay estimation of a noise-like random signal observed at two or more spatially separated receivers is a problem of considerable practical interest in passive radar/sonar applications. A new method is presented to analyze the mean-square error performance of delay estimation schemes based on a modified (improved) version of the Ziv-Zakai lower bound (ZZLB). This technique is shown to yield the tightest results on the attainable system performance for a wide range of signal-to-noise ratio (SNR) conditions. For delay estimation using narrow-band (ambiguity-prone) signals, the fundamental result of this study is illustrated in Fig. 3. The entire domain of SNR is divided into several disjoint segments indicating several distinct modes of operation. If the available SNR does not exceed SNR 1 , signal observations from the receiver outputs are completely dominated by noise thus essentially useless for the delay estimation. As a result, the attainable mean-square error \bar{\epsilon}^{2} is bounded only by the a priori parameter domain. If SNR 1 2 , the modified ZZLB coincides with the Barankin bound. In this regime differential delay observations are subject to ambiguities. If SNR > SNR 3 the modified ZZLB coincides with the Cramer-Rao lower bound indicating that the ambiguity in the differential delay estimation can essentially be resolved. The transition from the ambiguity-dominated mode of operation to the ambiguity-free mode of operation starts at SNR 2 and ends at SNR 3 . This is the threshold phenomenon in time delay estimation. The various deflection points SNR i and the various segments of the bound (Fig. 3) are given as functions of such important system parameters as time-bandwidth product (WT), signal bandwidth to center frequency ratio (W/ω 0 ) and the number of half wavelengths of the signal center frequency contained in the spacing between receivers. With this information the composite bound illustrated in Fig. 3 provides the most complete characterization of the attainable system performance under any prespecified SNR conditions.

Iterative and sequential Kalman filter-based speech enhancement algorithms

Speech quality and intelligibility might significantly deteriorate in the presence of background noise, especially when the speech signal is subject to subsequent processing. In particular, speech coders and automatic speech recognition (ASR) systems that were designed or trained to act on clean speech signals might be rendered useless in the presence of background noise. Speech enhancement algorithms have therefore attracted a great deal of interest. In this paper, we present a class of Kalman filter-based algorithms with some extensions, modifications, and improvements of previous work. The first algorithm employs the estimate-maximize (EM) method to iteratively estimate the spectral parameters of the speech and noise parameters. The enhanced speech signal is obtained as a byproduct of the parameter estimation algorithm. The second algorithm is a sequential, computationally efficient, gradient descent algorithm. We discuss various topics concerning the practical implementation of these algorithms. Extensive experimental study using real speech and noise signals is provided to compare these algorithms with alternative speech enhancement algorithms, and to compare the performance of the iterative and sequential algorithms.

Criteria for multichannel signal separation

We consider the problem in which we want to separate two (or more) signals that are coupled to each other through an unknown multiple-input-multiple-output linear system (channel). We prove that the signals can be decoupled, or separated, using only the condition that they are statistically independent, and find even weaker sufficient conditions involving their cross-polyspectra. By imposing these conditions on the reconstructed signals, we obtain a class of criteria for signal separation. These criteria are universal in the sense that they do not require any prior knowledge or information concerning The nature of the source signals. They may be communication signals, or speech signals, or any other 1-D or multidimensional signals (e.g., images). Computationally efficient algorithms for implementing the proposed criteria, that only involve the iterative solution to a linear least squares problem, are presented. >

A general class of lower bounds in parameter estimation

A general class of Bayesian lower bounds on moments of the error in parameter estimation is formulated, and it is shown that the Cramer-Rao, the Bhattacharyya, the Bobrovsky-Zakai, and the Weiss-Weinstein lower bounds are special cases in the class. The bounds can be applied to the estimation of vector parameters and any given function of the parameters. The extension of these bounds to multiple parameter is discussed. >

Multichannel signal separation: methods and analysis

The problem of multichannel signal separation has attracted considerable interest in recent literature. A variety of methods and criteria have been proposed to solve the problem based on statistical independence between the source signals. Most of these criteria involve the computation of high-order statistics of the observed signals. We present a unified framework for many of these criteria and analyze their statistical variance.