Higher-order statistics symmetry properties

• Type: Article in International Conference Proceedings Book
• Date: 1995

Authors

André Puga (Author) FEUP; Artur Pimenta Alves (Author) FEUP

Conference proceedings National

Title: Proceedings of the 7th Portuguese Conference on Pattern Recognition

Pages: 311-318

7th Portuguese Conference on Pattern Recognition

Aveiro, 23 - MARÇO - 1995

Other information

Resumo (PT): Higher-Order Statistics are a powerful tool to analyse and classify signals, particularly effective in the presence of non linearities or non gaussianities. The introduction of such methods in engineering and in pattern recognition problems requires the introduction of efficient ways to estimate the higher-order statistical parameters. It has been shown that, due to the symmetry properties of the higher order statistics, the cumulants and moments of 3rd and 4th order, bi- and tri-spectra can be significantly simplified. This paper introduces the properties of symmetry of these parameters and their implications in the simplification of the corresponding estimation. Special attention will be given to tri-spectra aiming to clarify and complement the present literature.

Language: Portuguese

Type (Professor's evaluation): Scientific

No. of pages: 8