In this paper we present a non-uniform filter bank (NUFB) matched to a given signal. To obtain matched Mchannel NUFB, first, we choose the decimation set having M-down sampling/decimation factors for which perfect reconstruction NUFB exist and then using novel approach proposed in this paper, M-channel signal matched analysis bank is estimated. The outputs of all filters at the analysis side of...

A new approach to implement computationally efficient reconfigurable filter banks for software defined radio receivers is presented in this paper. If the coefficients of a finite impulse response (FIR) filter are decimated by M, i.e., if every M-th coefficient of the filter is kept unchanged and remaining coefficients are replaced by zeros, a multi-band frequency response will be obtained. The ...

COMPLEX DIRECTIONAL WAVELET TRANSFORMS: REPRESENTATION, STATISTICAL MODELING AND APPLICATIONS AN PHUOC NHU VO, Ph.D. The University of Texas at Arlington, 2008 Supervising Professor: Soontorn Oraintara The thesis presents an new image decomposition for feature extraction, which is called the pyramidal dual-tree directional filter bank (PDTDFB). The image representation has an overcomplete ratio...

In this paper, a technique to determine the filter bank coefficients of Wavelets db3 and coif2 using mean square error adaptive filter RLS algorithm has been presented. Filter bank coefficients of the wavelet are treated as weight vector of adaptive filter and are changed iteratively to reach the desired value after few iteration. It is found that of the three adaptive algorithms viz. Least Mea...

We consider optimal orthonormal filter banks for subband coding of wide sense cyclostationary signals, with -periodic second order statistics. An -channel over decimated uniform filter bank, with -periodic analysis and synthesis filters, is used as the subband coder. An average variance condition is used to measure the output distortion. We show that for at least three potential bit allocation ...

Automatic detection of real life negative emotions in speech has been evaluated using Linear Discriminant Analysis, LDA, with “classic” emotion features and a classifier based on Gaussian Mixture Models, GMMs. The latter uses Mel-Frequency Cepstral Coefficients, MFCCs, from a filter bank covering the 300-3400 Hz region to capture spectral shape and formants, and another in the 20-600 Hz region ...

In this paper, a new image indexing and retrieval algorithm using multi-resolution local binary patterns (LBP) with joint histogram is proposed. The existing LBP extracts the relationship between the center pixel and its surrounding neighbors in an image. The proposed method encodes the joint histogram between the multiresolution LBPs which are calculated using Gaussian filter bank with differe...

In this contribution we present a method to design prototype filters of oversampling uniform complex-modulated FIR filter bank pairs. Especially, we present a noniterative two-step procedure: (i) design of analysis prototype filter with minimum group delay and approximately linear-phase frequency response in the passband and the transition band and (ii) Design of synthesis prototype filter such...

In this paper we describe experiments with the acoustic front{end of our large vocabulary speech recognition system. In particular, two aspects are studied: 1) linear transforms for feature extraction and 2) the modelling of the emission probabilities. Experiments are reported on a 5000{word task of the ARPA Wall Street Journal database. For the linear transforms our main results are: Filter{ba...

In this paper, a technique to identify the filter bank coefficients of Wavelets db4 and coif5 using adaptive filter NLMS algorithm is presented. Filter bank coefficients of the wavelet are treated as the weight vector of adaptive filter, changes with each iteration and approach to the desired value after little iteration. When we compare the two adaptive algorithms viz. Least Mean Square (LMS) ...