Original structure for Walsh–Hadamard transform on sliding window

The Nth-order Walsh–Hadamard transform on a sliding window is described here by means of FIF filters. Factoring the transfer functions of the filters allows for a synthesis of the transform as a tree of comb filters. The trees built in ascending or descending order of the delay value allow the compu...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Electronics letters Jg. 51; H. 23; S. 1850 - 1852
Hauptverfasser:	Tisserand, E, Berviller, Y
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	The Institution of Engineering and Technology 05.11.2015 IET
Schlagworte:	Algorithms ascending order Circuits and systems coefficient computation comb filters Delay delay value descending order digital arithmetic Electronics Engineering Sciences Field programmable gate arrays Hadamard order Hadamard transforms Micro and nanotechnologies Microelectronics parallel field programmable gate array implementation Projection RIF filters sequency order Signal and Image processing Sliding sliding window Synthesis transfer functions Transforms Trees Walsh‐Hadamard transform Hadamard transforms descending order field programmable gate arrays Walsh-Hadamard transform parallel field programmable gate array implementation coefficient computation sliding window delay value Hadamard order sequency order digital arithmetic RIF filters comb filters ascending order transfer functions
ISSN:	0013-5194, 1350-911X, 1350-911X
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	The Nth-order Walsh–Hadamard transform on a sliding window is described here by means of FIF filters. Factoring the transfer functions of the filters allows for a synthesis of the transform as a tree of comb filters. The trees built in ascending or descending order of the delay value allow the computation of the coefficients, respectively, in the sequency order and in the Hadamard order. The proposed algorithm requires 2N−2 additions for each projection. The resources used for a parallel field programmable gate array implementation are presented.
Bibliographie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0013-5194 1350-911X 1350-911X
DOI:	10.1049/el.2015.1246