Record: 23 AUTHOR: Minkoff, J. TITLE: The operation of multichannel
feedforward adaptive systems. SOURCE: IEEE Transactions on Signal Processing
v. 45 (Dec. '97) p. 2993-3005 bibl diags. STANDARD NO: 1053-587X DATE:
1997 PLACE: United States RECORD TYPE: art CONTENTS: feature article
ABSTRACT: The author presents a generalized formulation, applicable
to both random and deterministic signals, to describe the operation of
multichannel feedforward adaptive systems. Expressions for the optimum
multichannel adaptive-filter transfer functions and for the minimized cost
function were derived. In addition, the maximum achievable error-reduction
performance of the systems was obtained. SUBJECT: Cost functions. Adaptive
signal processing.
Record: 25 AUTHOR: Schodorf, Jeffrey B.; Williams, Douglas B. TITLE:
Array processing techniques for multiuser detection. SOURCE: IEEE Transactions
on Communications v. 45 (Nov. '97) p. 1375- 8 bibl diags. STANDARD NO:
0090-6778 DATE: 1997 PLACE: United States RECORD TYPE: art CONTENTS: feature
article
ABSTRACT: Techniques employed in the area of adaptive array signal
processing were applied to the multiuser detection problem. A robust detector
that is suitable for use in the presence of modeling errors and a reduced-rank
detector with improved transient behavior relative to full-rank detectors
were derived. Bit-error-rate curves and least-mean-square learning curves
were utilized to show algorithm performance. SUBJECT: Multiuser detection.
Constrained optimization methods. Adaptive signal processing.
Record: 26 AUTHOR: Feng, Jinwei.; Gan, Woon-Seng. TITLE: A broadband
self-tuning active noise equaliser. SOURCE: Signal Processing v. 62 no2
(Oct. '97) p. 251-6 bibl diags. STANDARD NO: 0165-1684 DATE: 1997 PLACE:
Netherlands RECORD TYPE: art CONTENTS: feature article
ABSTRACT: The authors propose a broadband self-tuning active noise
equalizer (SANE) to extend the applicability of the active noise equalizer
(ANE) technique to the practical situations where the primary noise power
varies with time. Using a variable gain factor, the proposed SANE can shape
the residual noise spectrum as the existing ANE does and can also automatically
adjust the residual noise power. Computer simulations are used to validate
the proposed algorithm. SUBJECT: Active noise control. Adaptive signal
processing. Self-tuning controllers.
Record: 27 AUTHOR: Li, Qiang.; Fan, H. (Howard) TITLE: On properties of information matrices of delta-operator based adaptive signal processing algorithms. SOURCE: IEEE Transactions on Signal Processing v. 45 (Oct. '97) p. 2454-67 bibl. STANDARD NO: 1053-587X DATE: 1997 PLACE: United States RECORD TYPE: art CONTENTS: feature article ABSTRACT: The authors investigated the properties of information matrices of the difference operator or the so-called delta (d) operator-based algorithms for adaptive signal processing. It was demonstrated that the conditioning of a transformed information matrix in the d domain is always better than that of the original information matrix in the conventional q domain for ill-conditioned problems. The results provide an explanation of the advantages of using delta operator algorithms for adaptive signal processing. The problem of the effect of D in the delta operator algorithms is also discussed. SUBJECT: Operators (Mathematics). Adaptive signal processing. Matrices (Mathematics). Record:
28 AUTHOR: Charleston, Sonia.; Azimi-Sadjadi, Mahmood R.; Gonzalez-Camarena,
Ramon. TITLE: Interference cancellation in respiratory sounds via a multiresolution
joint time-delay and signal-estimation scheme. SOURCE: IEEE Transactions
on Biomedical Engineering v. 44 (Oct. '97) p. 1006-19 bibl diags. STANDARD
NO: 0018-9294 DATE: 1997 PLACE: United States RECORD TYPE: art CONTENTS:
feature article
ABSTRACT: The authors address the problem of separating out heart sounds
from acquired respiratory sounds using a new joint time-delay and signal
estimation (JTDSE) procedure. The primary characteristic of the JTDSE method
is that it is based on the common information about the time delay between
reference and acquired signals that is shared by all the levels and information-bearing
subbands. Other advantages of the proposed method include robustness in
the presence of noise and accuracy in time-delay estimation. SUBJECT: Adaptive
filters - Mathematical models. Wavelet transforms. Time delay estimation.
Signal processing.
Record: 29 AUTHOR: Reif, Konrad.; Luo, Fa-Long.; Unbehauen, Rolf. TITLE:
The exponential stability of the invariant-norm PCA algorithm. SOURCE:
IEEE Transactions on Circuits and Systems. Part II, Analog and Digital
Signal Processing v. 44 (Oct. '97) p. 873-6 bibl. STANDARD NO: 1057-7130
DATE: 1997 PLACE: United States RECORD TYPE: art CONTENTS: feature article
ABSTRACT: A recently proposed principal component analysis algorithm
was investigated. The solutions of the corresponding differential equations
were shown to converge to the principal eigenvectors of the autocorrelation
matrix. An exponential decaying bound for the error was computed. SUBJECT:
Principal components analysis. Adaptive signal processing. Unsupervised
learning.
Record: 30 AUTHOR: Omoto, A.; Takashima, K.; Fujiwara, K. TITLE: Active
suppression of sound diffracted by a barrier: an outdoor experiment. SOURCE:
The Journal of the Acoustical Society of America v. 102 (Sept. '97) p.
1671-9 bibl il diags. STANDARD NO: 0001-4966 DATE: 1997 PLACE: United States
RECORD TYPE: art CONTENTS: feature article
ABSTRACT: The active control method was used to suppress the sound
diffracted by an outdoor barrier. This method operated by the cancellation
of the sound pressure at the diffraction edge of the barrier, which normally
behaves like the virtual source of the diffracted field. The results of
two experiments are shown in this paper. In the first experiment, we employed
two independent controllers that utilized multi-channel adaptive signal
processing to minimize the sum of the mean square of the sound pressure
at four and six points along the diffraction edge. Measurement of sound
pressure levels at various distances from the barrier showed effective
sound suppression, with about 6-dB excess attenuation over the barrier's
insertion loss at the receiver at a distance of 50 m. A practically realistic
noise source, a fan blower, was used as a primary source in the second
experiment and the moderate attenuation could be achieved at almost all
the receiver points. The results obtained in these two experiments provided
the experimental verification of the strategy for the active suppression
of sound diffracted by a noise barrier. Reprinted by permission of the
publisher. SUBJECT: Active noise control. Noise barriers. Adaptive signal
processing.
Record: 36 AUTHOR: Strobach, Peter. TITLE: Fast recursive orthogonal
iteration subspace tracking algorithms and applications. SOURCE: Signal
Processing v. 59 no1 (May '97) p. 73-100 bibl. STANDARD NO: 0165-1684 DATE:
1997 PLACE: Netherlands RECORD TYPE: art CONTENTS: feature article
ABSTRACT: The author introduces a class of fast recursive subspace
tracking algorithms that is based on the orthogonal iteration principle.
This includes the derivation of realizations with O(Nr2) and O(Nr) complexity,
where N is the model order and r(less than or equal)N is the rank of the
underlying data covariance matrix. Comparisons indicate that the algorithms
require fewer operations and offer a better angle performance than the
recently introduced transposed QR singular value decomposition subspace
tracker. Also described are complete quasi-code tables of the algorithms
and applications to adaptive frequency estimation and rank adaptive subspace
filtering. SUBJECT: Adaptive signal processing. Iterative algorithms. Tracking
algorithms.
Record: 39 AUTHOR: Goldstein, J. Scott.; Reed, Irving S. TITLE: Reduced-rank
adaptive filtering. SOURCE: IEEE Transactions on Signal Processing v. 45
(Feb. '97) p. 492- 6 bibl diags. STANDARD NO: 1053-587X DATE: 1997 PLACE:
United States RECORD TYPE: art CONTENTS: feature article
ABSTRACT: A new rank reduction scheme for adaptive filtering problems
is presented. The scheme utilizes a cross-spectral metric for selecting
the optimal lower dimensional subspace for reduced- rank adaptive filtering
as a function of the basis vectors of the full-rank space. It is found
that the steady-state solution in the subspace opted for by the cross-spectral
metric provides an upper bound for the solution found in the space spanned
by the eigenvectors that correspond with the largest eigenvalues of the
full-rank correlation matrix. SUBJECT: Adaptive signal processing.
Record: 51 AUTHOR: Bollini, P.; Chisci, L.; Farina, A. TITLE: QR versus IQR algorithms for adaptive signal processing: performance evaluation for radar applications. SOURCE: IEE Proceedings. Radar, Sonar and Navigation v. 143 (Oct. '96) p. 328-40 bibl diags. STANDARD NO: 1350-2395 DATE: 1996 PLACE: United Kingdom RECORD TYPE: art CONTENTS: feature article SUBJECT: Adaptive signal processing. Beamforming. CORDIC algorithms.
Record: 53 AUTHOR: Liu, Shih-Chii.; Mead, Carver. TITLE: Continuous-time adaptive delay system. SOURCE: IEEE Transactions on Circuits and Systems. Part II, Analog and Digital Signal Processing v. 43 (Nov. '96) p. 744-51 bibl diags. STANDARD NO: 1057-7130 DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Adaptive signal processing. Delay lines - Mathematical models. Phase locked loops.
Record: 57 AUTHOR: Rabideau, Daniel J. TITLE: Fast, rank adaptive subspace tracking and applications. SOURCE: IEEE Transactions on Signal Processing v. 44 (Sept. '96) p. 2229-44 bibl diags. STANDARD NO: 1053-587X DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Tracking algorithms. Subspaces. Adaptive signal processing.
Record: 58 AUTHOR: Guerci, Joseph R.; Feria, Erlan H. TITLE: Application of a least squares predictive-transform modeling methodology to space-time adaptive array processing. SOURCE: IEEE Transactions on Signal Processing v. 44 (July '96) p. 1825-33 bibl diags. STANDARD NO: 1053-587X DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Moving target indicators. Radar signal processing.
Record: 60 AUTHOR: Jin, Qu.; Luo, Zhi-Quan (Tom); Wong, Kon Max. TITLE: Optimum filter banks for signal decomposition and its application in adaptive echo cancellation. SOURCE: IEEE Transactions on Signal Processing v. 44 (July '96) p. 1669-80 bibl diags. STANDARD NO: 1053-587X DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Adaptive signal processing. Multidimensional filters - Design. Echo suppression.
Record: 61 AUTHOR: Yu, Shiann-Jeng.; Lee, Ju-Hong. TITLE: Adaptive array beamforming for cyclostationary signals. SOURCE: IEEE Transactions on Antennas and Propagation v. 44 (July '96) p. 943-53 bibl diags. STANDARD NO: 0018-926X DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Beamforming - Mathematical models. Adaptive signal processing. Adaptive algorithms.
Record: 62 AUTHOR: Candy, James V.; Sullivan, Edmund J. TITLE: Model-based identification: an adaptive approach to ocean- acoustic processing. SOURCE: IEEE Journal of Oceanic Engineering v. 21 (July '96) p. 273- 89 bibl diag. STANDARD NO: 0364-9059 DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature article SUBJECT: Underwater sound. Acoustic signal processing. Parameter estimation.
Record: 65 AUTHOR: Bermudez, Jose Carlos M.; Bershad, Neil J. TITLE:
A nonlinear analytical model for the quantized LMS algorithm-- the arbitrary
step size case. SOURCE: IEEE Transactions on Signal Processing v. 44 (May
'96) p. 1175- 83 bibl diags. STANDARD NO: 1053-587X DATE: 1996 PLACE: United
States RECORD TYPE: art CONTENTS: feature article
ABSTRACT: An investigation of the quantization effects in the finite
precision least mean squares (LMS) algorithm with arbitrary step sizes.
The LMS algorithm is unquestionably one of the most popular algorithms
for digital implementation of real- time high-speed adaptive filters. Deterministic
nonlinear recursions are developed for the mean and second moment matrix
of the weight vector about the Wiener weight for white Gaussian data models
and small algorithm step sizes. These recursions are solved numerically
and demonstrated to be in very close agreement with the Monte Carlo simulations
during all phases of the adaptation process. A design example is used to
illustrate how the theory can be employed to select the number of quantizer
bits and the adaptation stepsize m to yield a desired transient behavior
and cancellation performance. SUBJECT: Adaptive signal processing. Least
square algorithms.
Record: 66 AUTHOR: Doroslovacki, Milos.; Fan, H. (Howard) TITLE: Wavelet-based
linear system modeling and adaptive filtering. SOURCE: IEEE Transactions
on Signal Processing v. 44 (May '96) p. 1156- 67 bibl diags. STANDARD NO:
1053-587X DATE: 1996 PLACE: United States RECORD TYPE: art CONTENTS: feature
article
ABSTRACT: The authors report on wavelet-based linear system modeling
and adaptive filtering. Adaptive filtering of signals attempts to generate,
from an input signal, an output signal with desired properties, while learning
to do it in the best possible way. Optimal coefficients and the corresponding
minimum mean square error were found, and they were, in general, shown
to be time varying. Least-mean-square adaptive filtering algorithms were
developed for on-line filtering and system identification. Theoretically
and by simulations, the advantages of using wavelet-based filtering were
demonstrated: