Adaptive Approaches for Blind Equalization Based on Multichannel Linear Prediction

Maurício Sol de Castro, João Marcos Travassos Romano

DOI: 10.14209/its.2002.442

Keywords:

Abstract

"In this work, the problem of blind multichannel equalization is considered. A strategy for saving computation in zero-forcing equalizers design is proposed and its performance is evaluated under an adaptive implementation of an algorithm based on multichannel forward linear prediction. Moreover, a cascade structure based on forward and backward linear prediction is regarded: an adaptive implementation of such a structure is also proposed and its performance is verified through computer simulations."

Download

A Model for the Behavior of the Least Mean Kurtosis (LMK) Adaptive Algorithm with Gaussian Inputs

Pedro I. Hübscher, José C. M. Bermudez

DOI: 10.14209/its.2002.448

Keywords:

Abstract

"The LMK algorithm is a stochastic gradient algorithm which seeks to minimize the negated kurtosis of the error signal. It outperforms the LMS algorithm in several applications of practical interest, with little increase in computational complexity. This paper presents a statistical analysis of the Least Mean Kurtosis (LMK) adaptive algorithm. Deterministic nonlinear recursive equations are derived for the mean weight behavior and for the weight error correlation matrix, for a Gaussian reference signal and slow learning. The new model describes the algorithm behavior during transient and steady-state for a white measurement noise with any even probability density function (pdf). The accuracy of the model is demonstrated by Monte Carlo simulations."

Download

The Rounded Hartley Transform

R. J. de Sobral Cintra, H. M. de Oliveira, C. O. Cintra

DOI: 10.14209/its.2002.454

Keywords:

Abstract

"A new multiplication-free transformation derived from DHT is introduced: the RHT. Investigations on the properties of the RHT led us to the identification of an approximate inverse. We showed that RHT is not involutional like the DHT, but it exhibits approximate involutional property. Thus instead of using the actual inverse transform, the RHT is taken as an involutional transformation, allowing the use of direct (multiplicationfree) to evaluate the inverse. A fast algorithm to compute RHT is presented, which offers zero multiplications and O(n log n) additions for power-of-2 blocklengths. This algorithm shows embedding properties. We also extended RHT to the two-dimensional case. This permitted us to perform a preliminary analysis on the effects of RHT on images. Despite of some SNR loss, RHT can be interesting for applications which involve image monitoring associated to decision making, such as military applications or medical imaging."

Download

Echo Canceller Based on Adaptive Interpolated FIR Filters

Orlando J. Tobias, Rui Seara Júnior, Rui Seara

DOI: 10.14209/its.2002.460

Keywords:

Abstract

"In this paper, the use of adaptive interpolated FIR-LMS structures for echo cancelling in digital telephone systems is assessed. These structures are compared with the widely used FIR-LMS one. By using interpolated structures, it is possible to achieve substantial savings in the required arithmetic for both filtering and tap updating operations. Experimental results show that the interpolated structures also outperform, in terms of residual echo, the traditional FIR structure."

Download

Q-GA: Performance Analysis in Low-pass Filter Equalization Design

Delmar B. Carvalho, Sidnei Noceti Filho, Rui Seara

DOI: 10.14209/its.2002.465

Keywords:

Abstract

"The Q-GA (modified genetic algorithm) has been proposed to design phase equalizers by using the symmetry error criterion of the impulse response. In this approach, the search space is partitioned into subspaces in which a small population evolves, regarding a reduced number of generations. This approach presents a considerable computational complexity gain, as compared with the use of a conventional GA. In this paper, we assess such an algorithm for performance in equalizer design of low-pass filters."

Download

A New Stochastic Analysis of Affine Projection Algorithm for Gaussian Inputs and Large Number of Coefficients

Sérgio J. M. de Almeida, José Carlos M. Bermudez

DOI: 10.14209/its.2002.471

Keywords:

Abstract

"This paper presents a new analytical model for predicting the behavior of Affine Projection (AP) algorithm. The model is derived for autoregressive (AR) Gaussian inputs and for unity step size (fastest convergence). Deterministic recursive equations are determined for the mean weight and the mean square error behaviors for a large number of adaptive taps N, as compared to the order P of the algorithm. Simulation are presented which show excellent agreement between theory an simulations in steady-state, and fair to good agreement during transient. The model predictions for the transient phase improve as N/P increases. These caracteristics are of special interest in applications such as acoustic echo cancellation."

Download

Optimized Cosine-Modulated Filter Banks Using the Frequency Response Masking Approach

Miguel B. Furtado Jr., Sergio L. Netto, Paulo S. R. Diniz

DOI: 10.14209/its.2002.476

Keywords:

Abstract

"An optimization procedure for cosine-modulated filter banks (CMFB) using the frequency-response masking approach is proposed. In the given method, we perform minimization of the maximum attenuation level in the filter\u2019s stopband, subject to intersymbol interference (ISI) and intercarrier interference (ICI) constraints. For optimization, a quasiNewton algorithm with line search is used, and we propose a simplified analytical expressions to impose the interference constraints. The result is lower levels of ISI and ICI for a pre-determined filter order, or reduced filter complexity for given levels of interferences."

Download

Efficient Design of CMFBs Using Two-Stage Frequency Response Masking

Luiz C. R. de Barcellos, Sergio L. Netto, Paulo S. R. Diniz

DOI: 10.14209/its.2002.481

Keywords:

Abstract

"A new method for designing narrowband cosine-modulated filter banks (CMFBs) is proposed based on the frequency-response masking (FRM) approach with masking filter decomposition. The resulting structure, the so-called FRM2-CMFB, presents a reduced computational complexity, and allows one to design filter banks with extremely high number of bands, where standard filter design methods fail to converge. Examples included, illustrate that the total number of coefficients of the CMFB prototype filter can be reduced to about 60% of the original FRM-CMFB structure, which does not use masking filter decomposition."

Download

Adaptive LCMV Beamforming Avoiding DOA Estimation for Packet-like Wireless Systems

Danilo Zanatta Filho, Charles Casimiro Cavalcante, João Marcos Travassos Romano

DOI: 10.14209/its.2002.485

Keywords:

Abstract

"A new adaptive LCMV solution that avoids DOA estimation for uplink beamforming is proposed. The use of the uplink channel covariance matrix (UCCM) instead of DOA estimation is shown to be a suitable alternative for constraints selection. The resulting LCMV adaptive solutions are then based on the LMS and RLS versions with the UCCM estimation. The new method is then evaluated for packed-like wireless systems through computational simulations."

Download

An Algorithm for Segmentation and Detection of Changes in Signals Using Diffusion Equations

Aline R. Gesualdi, Marcelo P. de Albuquerque, Márcio P. de Albuquerque, Marcello L. R. de Campos

DOI: 10.14209/its.2002.490

Keywords:

Abstract

"We will present, in details, an algorithm for the Stabilized Inverse Diffusion Equation (SIDE) technique introduced in [4]. The algorithm will be evaluated in two different application areas of interest, which are: signal detection and image segmentation. SIDEs are a family of semi-discrete evolution equations which stably sharpen edges and suppress noise. This characteristic makes SIDEs an efficient tool in noise reduction problems. This paper will focus on the detection of 1-D signals and on the segmentation of ultrasound and mammography images. Results will be presented."

Download