Showing papers in "Digital Signal Processing in 2003"

TL;DR: A new algorithm for electrocardiogram (ECG) compression based on the compression of the linearly predicted residuals of the wavelet coefficients of the signal, which reduces the bit rate while keeping the reconstructed signal distortion at a clinically acceptable level.

TL;DR: The invertibility of the TT-transform points to the possibility of filtering and signal to noise improvements in the time domain, and some insight into the localized spectra of the S-transform.

TL;DR: This paper presents and contrast three problem of fully exploiting the spatial diversity of a multi-antenna scheme (using exact CSI at the receiver), for reasonably slow fading scheme and in fact quite close to the performance of the clairvoyant schemes that avoid the previous requirement of STBC.

TL;DR: The lomo scale-space is introduced, a novel two-dimensional generalization of the concept of locally monotonic (lomo) signals that is applicable to several multiscale image processing tasks such as segmentation, object-based image compression, content-based retrieval, and video tracking.

TL;DR: By adjusting frame size with field displacement, the effective restoration of images blurred by a laboratory-grade spherical lens is demonstrated, and mean-square-error and restoration improvement are reported as a function of field angle.

TL;DR: It is shown that the subband technique can provide an unbiased estimate of the second-order statistics of a noisy observed signal vector.

TL;DR: The proposed model may be regarded as a stochastic optimization approach to the problem of searching for an optimal composition of dictionary elements, and the main motivation is the application of this model to construct adaptive atomic analyzers, based on overcomplete but sparse representations.

TL;DR: An algorithm for determining an optimum image thresholding value is proposed and phase correlation between the gray-level image and its binary counterpart is defined as a function of the thresholding parameter.

TL;DR: Training sequence selection for frequency offset estimation in frequency selective channels is studied for the first time with real-time data and it shows clear trends in training sequence selection.

TL;DR: An overview of the basic concepts, current state and future directions of uniform and nonuniform multirate filter banks and their applications, and some selected applications of multidimensional filter banks are described.

TL;DR: This paper considers the least mean-squared error linear estimation problems, using covariance information, in linear discrete-time stochastic systems with uncertain observations for the case of white plus coloured observation noises.

TL;DR: A nonoscillating iterative method of Gold deconvolution, generalized for multidimensional data is presented and a new optimized algorithm aimed to reduce the number of computer operations is derived.

TL;DR: A careful comparison of three numeric techniques for estimation of the curvature along spatially quantized contours is reported, with results important not only for characterizing the relative performance of the considered methods, but also for providing practical guidelines for those interested in applying those techniques to real problems.

TL;DR: This paper proves that the only attractors of the algorithm correspond to the points where perfect separation is achieved and proposes novel strategies to remove the permutation and the amplitude indeterminacies that appear when the sources are recovered in a different order or with different amplitude in some frequency bins.

TL;DR: Channel order and RMS delay spread estimation with application to AC power line communications and how this affects decision-making in the electricity industry is studied.

TL;DR: A hybrid of the proposed algorithm and the TDLMS is described that offers significant performance improvements since it can be tuned to achieve very fast convergence while allowing low misadjustment.

TL;DR: The proposed filter design method is based on integer programming (IP) and can be directly applied to any fixed-point FIR design specifications which can be formulated in a form of linear constraints on the filter coefficients.

TL;DR: The theory of tempered distributions provides the mathematical foundation for much of signal theory, especially DSP as mentioned in this paper, but signals processing texts do not contain careful expositions of the theory, preferring instead to treat the subject as a collection of rules to be memorized and formally manipulated.

TL;DR: A quadtree based method is proposed that allows for more efficient classification by using variable-sized blocks in order to maximize the classification gain, while maintaining a limit on the classification overheads.

TL;DR: An explicit formula for the impulse response of a discrete-time causal linear time-varying infinite-impulse-response system in terms of the system parameters is obtained from the solution of a linear difference equation with variable coefficients.

TL;DR: Fundamental principles for the design of a data transmission scheme that obtains spectrally efficient orthogonal frequency division multiplexing of digital vestigial sideband signals are presented and how bandwidth on demand by factors of 2 can be obtained is shown.

TL;DR: The well-known method of frames approach to the signal decomposition problem is reformulated as a certain bilevel goal-attainment linear least squares problem and a numerically robust variant of the method is proposed on the basis of a certain minimal Euclidean norm approximating splitting pseudo-iteration-wise method.

TL;DR: Analysis of clinical data indicates that a FECG with a favorable signal-to-noise ratio could be acquired in real-time from the IC after maternal ECG cancellation using the proposed method, and could be used for determining fetal HR and electrophysiological information.

TL;DR: A unified analytical approach for analysing cumulant-based AR and ARMA identification methods based on the replacement of poles of identified systems by expansions to infinite geometric series, which shows that a non-identifiable class of systems can always be defined for the relationship between the system poles and zeros.

TL;DR: An eigenvalue-based residuum criterion for the detection of the number of sinusoids in white Gaussian noise is presented and gives a superior performance over the Akaike information criterion (AIC) and the minimum description length principle (MDL), especially for situations where the analyzed signals exhibit a low signal-to-noise ratio (SNR), short data records, and a high number of Sinusoids.

TL;DR: A very low-complexity audio codec that provides audio playback quality similar to the MPEG-I/audio level 3 codec at 64 Kbps for a monophonic-channel signal and an adaptive arithmetic coder with multiplication-free adaptation is presented.

TL;DR: A Bayesian framework is used to derive a solution that can be viewed as an approximate MAP estimator for the optimal filter response, where the desired response signal is contaminated by noise.

TL;DR: In this article, a new method for temporal decomposition (TD) of speech parameters for very low rate coding applications is developed, where phonetic relevance is not considered here, instead, the spectral parameters of speech using pre-defined interpolation functions.

TL;DR: In this article, the use of the symmetrical extension as an alternative to wraparound in order to avoid the necessity of overlapping is discussed, and a new algorithm is presented for reducing sharp variations in quantization noise that appear at the border of frames when overlapping is avoided.

TL;DR: It is shown that, in the colored noise case, the amplitude information can also help improving the performance of the frequency estimator, and vice versa, and an efficient FFT-based implementation structure of the optimal estimators is proposed.