In this paper, sparsely-connected neural networks synthesized for recognizing discrete-time sinusoidal signals as a possible support in children’s acoustic rehabilitation are considered. In particular, a robustness analysis of this kind of sparsely-connected neural networks is developed, both in the presence of perturbed input patterns and of network parameter variations. Performances are evaluated by choosing simple acoustic signals, that is, sinusoidal signals with different values of angular frequency, width and phase. Test cases are reported to illustrate the capabilities of designed networks

On Performances of Sparsely-Connected Neural Networks for Acoustic Signal Recognition

Leonarda Carnimeo;Antonio Giaquinto
2004

Abstract

In this paper, sparsely-connected neural networks synthesized for recognizing discrete-time sinusoidal signals as a possible support in children’s acoustic rehabilitation are considered. In particular, a robustness analysis of this kind of sparsely-connected neural networks is developed, both in the presence of perturbed input patterns and of network parameter variations. Performances are evaluated by choosing simple acoustic signals, that is, sinusoidal signals with different values of angular frequency, width and phase. Test cases are reported to illustrate the capabilities of designed networks
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11589/9136
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact