A reconstruction of band-limited signals from their non-uniformly distributed samples is considered. The problem of reconstruction is based upon estimation of discrete wavelet transform coefficients and is formulated as an optimization procedure of finding the signal that matches the given samples with smallest norm. This procedure generates discrete wavelet transform coefficients in a form suitable for further processing of the reconstructed signal, such as compression and de-noising. The method of conjugate gradients is used in the approximation phase. Sparse matrix techniques are applied to reduce the number of floating point operations.

Biography

Momcilo Bogdanov received his B.S. and M.S. from Faculty of Electrical Engineering, Belgrade (Yugoslavia), and his Ph.D. from Faculté Polytechnique de Mons (Belgium), all in electrical engineering. He is now a professor in the Faculty of Electrical Engineering at Skopje (R. Macedonia). He has also been a faculty member at the University of Oran (Algeria) and the Technical Faculty of Pristina (Yugoslavia). His actual research interests include the design and applications of filter banks.