Neural Networks for Spectral Analysis of Unevenly Sampled Data

1999/06/10 by R. Tagliaferri, A. Ciaramella, Tagliaferri, R. +5
Computer Science · Engineering · Physics and Astronomy · #Advanced Adaptive Filtering Techniques #Blind Source Separation Techniques #Speech and Audio Processing #astro-ph

paper · pdf · doi:10.48550/arxiv.astro-ph/9906181

9 pages, wicsbook.sty macro file, accepted for publication in the proceedings of the 11th Italian Workshop on Neural Networks, WIRN Vietri 99, M. Marinaro and R. Tagliaferri eds., Springer-Verlag 1999

arxiv created 1999/06/11 · arxiv updated 2009/12/01

Abstract

In this paper we present a neural network based estimator system which performs well the frequency extraction from unevenly sampled signals. It uses an unsupervised Hebbian nonlinear neural algorithm to extract the principal components which, in turn, are used by the MUSIC frequency estimator algorithm to extract the frequencies. We generalize this method to avoid an interpolation preprocessing step and to improve the performance by using a new stop criterion to avoid overfitting. The experimental results are obtained comparing our methodology with the others known in literature.

Related

• Spectral analysis of signals by time-domain statistical characterization and neural network processing: Application to correction of spectral amplitude alterations in pulse-like waveforms
• Basic Filters for Convolutional Neural Networks Applied to Music:\n Training or Design?
• Data-driven Estimation of Sinusoid Frequencies
• Extract fundamental frequency based on CNN combined with PYIN
• Rapid parameter estimation of discrete decaying signals using autoencoder networks