Offset removing in the domain of signal shapes

Hervé Rix (1)

1 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe SIGNAL

The recognition of a given shape in a positive signal using the Distribution Function Method (DFM) assumes an affine transform on the abscissa and only a multiplicative coefficient on the ordinates, without any offset. The aim of the paper is to extent DFM when shape equality includes an offset. In fact, this problem is a particular case of signal shape recognition in a sum of two signals whose shapes are known. The first application in mind is the beat to beat extraction of a given electrocardiographic (ECG) wave (e.g. T-wave) added to a random offset. The method for signal shape recognition will be first presented in view of this application. Then a simulation study is done, pointing out the influence of noise on estimations.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Hervé Rix. Offset removing in the domain of signal shapes. Colloque National Recherche en Imagerie et Technologies pour la Santé, Apr 2011, Rennes, France. pp. CDROM. ⟨hal-00589089⟩