Noise Mismatch Research Articles

On the Identification of Stable, Integrating, and Unstable Processes with Unknown Orders and Delays.

This article addresses the complete identification of stable, integrating, and unstable processes. Despite the stability of the process, a biased relay experiment can be performed to provide abundant process input-output data in stable operation. Based on a time-weighted integral filter, an effective method is developed to estimate the orders and time delay as well as the model parameters from the relay-induced data in a straightforward fashion. The method utilizes a small amount of experimental data and is robust with respect to measurement noise and model structure mismatch. Furthermore, it can easily be applied as a model reduction technique by estimating an apparent delay with user-specified orders.

SYNCHRONIZING CHAOS BY IMPULSIVE FEEDBACK METHOD

In this letter, the impulsive feedback synchronization method is suggested. Synchronization condition is given by investigating the mechanism of the impulsive feedback synchronization. Furthermore, we consider the influences of noise and parameter mismatch since they inevitably exist in the practical experiment. Finally, a visual circuit implementation of impulsive feedback synchronization is performed. The research demonstrates that impulsive operator can spread frequency spectrum of chaotic solution. Impulsive feedback can synchronize chaotic systems and has the robustness to noise and parameter mismatch. It shows the potential application in chaotic communication.

Compensation of channel and noise distortions combining normalization and speech enhancement techniques

This paper introduces two techniques to obtain robust speech recognition devices in mismatch conditions (additive noise mismatch and channel mismatch). The first algorithm, adaptive Gaussian attenuation algorithm (AGA), is a speech enhancement technique developed to reduce the effects of additive background noise in a wide range of signal noise ratio (SNR) and noise conditions. The technique is closely related to the classical noise spectral subtraction (SS) scheme, but in the proposed model the mean and variance of noise are used to better attenuate the noise. Information of the SNR is also introduced to provide adaptability at different SNR conditions. The second algorithm, cepstral mean normalization and variance-scaling technique (CMNVS), is an extension of the cepstral mean normalization (CMN) technique to provide robust features to convolutive and additive noise distortions. The requirements of the techniques are also analyzed in the paper. Combining both techniques the relative channel distortion effects were reduced to 90% on the HTIMIT task and the relative additive noise effects were reduced to 77% using the TIMIT database mixed with car noises at different SNR conditions.

SPATIO-TEMPORAL SYNCHRONIZATION OF COUPLED PARAMETRICALLY EXCITED PENDULUM ARRAYS

In this paper, parametrically excited chaotic pendulum arrays are considered with the spatial diffusive coupling. The drive array undergoes extremely irregular behavior, both temporally as well as spatially. To synchronize two coupled spatio-temporal chaotic arrays, the periodic feedback method is used. The synchronization condition is given to determine the critical values of periodic characteristic time and feedback weight. Furthermore, the influences of noise and parameter mismatch on synchronization are investigated. It is found that the periodic feedback method has strong robustness to noise and parameter mismatch, which promises its potential application in synchronizing parametrically excited pendulums.

Techniques for preventing tonal behavior of data weighted averaging algorithm in /spl Sigma/#x03A3;-/spl Delta/#x0394; modulators

Two techniques for prevention of in-band tones in DWA mismatch noise shaping algorithm are presented. "offset technique" moves in-band tones away from baseband, improving SNR by /spl sim/10 dB, and attenuates maximum in-band tone by as much as /spl sim/25 dB. "Randomized DWA" is a combination of simultaneous randomization and DWA, which attenuates tonal behavior in all frequencies, preventing aliasing of tones or tone shifting for any arbitrary input signal.

N-path dynamic element matching for multibit bandpass ΣΔ modulators

We discuss different dynamic element-matching (DEM) techniques resulting in a noise-shaped mismatch spectrum. The frequency domain operation of data-weighted averaging (DWA), which results in first-order highpass mismatch noise shaping, is explained and the extension of DWA to a general mismatch noise-shaping transfer function is considered. An N-path DEM technique, which results in a simple implementation of bandstop mismatch noise shaping, is presented and compared with the case of bandstop shaping derived from general DWA. An implementation of the proposed N-path DEM for a bandpass sigma–delta modulator is presented with simulation results. © 1997 by John Wiley & Sons, Ltd.

Mismatched finite-complexity MMSE decision feedback equalizers

Closed-form expressions that quantify the degradation in the decision-point signal-to-noise ratio of the finite-length minimum mean-square-error decision feedback equalizer (MMSE-DFE) due to channel and noise mismatch conditions are derived. The analysis is further extended to other receiver structures, namely, the MMSE-DFE with an adaptive feedforward filter, the MMSE linear equalizer, and the discrete multitone transceiver, all under a finite-complexity constraint. The limiting case of infinite-length filters is also analyzed. In addition, we present computer simulation results that compare the performance of the various receiver structures under study, assuming a particular channel and noise mismatch model for the high-bit-rate digital subscriber loop environment. Finally, several methods that can be used to mitigate the effects of mismatch are outlined.

Equalization and detection in storage channels

In this work, a model for digital magnetic recording on thin-film media is developed, and the performances of several detector/run-length-limited code combinations are compared under different densities, media noise, offtrack interference, and channel mismatch conditions. The study is based on computer simulations using an experimentally extracted transition response of a magnetoresistive read head. The detection algorithms considered include the partial response maximum likelihood detectors, decision feedback equalizer (DFE), and fixed-delay tree search with decision feedback (FDTS/DF) detector. It was found that the DFE with a (0,k) run-length-limited (RLL) code and the FDTS/DF with a (1,7) RLL code show the least bit error rate degradation, as the user density increases beyond 2.5 bits/PW50. The PRML and DFE suffer less from offtrack interference. The (0,k) coded DFE and the (1,7) coded EPRML and FDTS/DF channels are relatively robust under PW50 mismatch. At a user density of 2.5, FDTS/DF (1,7) and DFE (0,k) are found to yield the best bit error performance under varying intensities of white Gaussian noise and media noise.

The impact of white gaussian noise mismatch on linear and decision feedback equalizers

Analyzes the sensitivity of equalizers to white gaussian noise mismatch, which exists whenever the noise power observed during training differs from that observed during detection. A closed-form solution for the mean-square error is obtained for linear and decision feedback equalizers as a function of the channel spectrum and degree of noise mismatch. The results show that increasing equalizer complexity may actually degrade performance in the presence of channel mismatch. >

Recursive least-squares sequence estimation

A family of adaptive communication receivers based on recursive least-squares sequence estimation (RLSSE) algorithms is proposed which provides performance comparable to that of conventional linear receivers, but with reduced complexity and less sensitivity to channel mismatch. A software-implemented version of the linear member of the family is shown to have performance equivalent to that of standard transversal equalizers under ideal conditions, yet offers a drastic improvement in white Gaussian noise mismatch environments. An analogous performance improvement for several test channels is also shown for software-implemented versions of the constrained (nonlinear) members of the family over decision feedback equalizers. Another advantage of the RLSSE family of receivers may be its ease of implementation, since it should be possible to combine the functions of channel estimation and sequence estimation on the same chip.

Enhanced Rational Signal Modeling

Abstract In various signal processing applications involving theoretical or empirical considerations, it is desired to appropriately modify a given data set so that the modified data set possesses prescribed properties. These properties are usually chosen so as to identify information signals believed to be contained within the given data set. The modification of the given data set then serves as a cleansing process whereby corrupting noise, measurement distortion or theoretical mismatch present in the given data set is removed. In this paper, a recently developed signal enhancement algorithm is described which achieves this objective. Particular attention is directed towards properties that are describable using a singular value decomposition (SVD) of a data generated matrix. Examples are given demonstrating a significant improvement in the performance of subspace-based frequency estimation techniques. Next, the behavior of these high resolution algorithms with respect to the relative phase of closely spaced sinusoids is examined. This behavior is related to the singular values of the data matrix. The predicted behavior is confirmed in the examples.

Performance analysis of total least squares methods in three-dimensional motion estimation

An algorithm is presented to obtain the total least squares (TLS) estimates of the motion parameters of an object from range/stereo data or perspective views in a closed form. TLS estimates are suitable when data in both time frames are corrupted by noise, which is an appropriate model for motion analysis in practice. The robustness of different linear least squares methods is analyzed for the estimation of motion parameters against the sensor noise and possible mismatches in establishing object feature point correspondence. As the errors in point correspondence increase, the performance of an ordinary least squares (LS) estimator was found to deteriorate much faster than that of the TLS estimator. The Cramer-Rao lower bound (CRLB) of the error covariance matrix was derived for the TLS model under the assumption of uncorrelated additive Gaussian noise. The CRLB for the TLS model is shown to be always higher than that for the LS model. >