Spectral Analysis for System Identification from Input/Output Binary Measurements

Hicham Oualla (1) , Mathieu Pouliquen (1) , M. Frikel (1) , Said Safi (2) , Tristan Bonargent (1)

1 LAC - Laboratoire d'automatique de Caen
2 USMS - Université Sultan Moulay Slimane

This short paper focus on an aspect of system identification from input/output binary measurements: the estimation of the frequency response of the system. Currently, no algorithm is available in literature for such an estimation. The novelty of this paper is to propose an algorithm. It is based first on the estimation of the auto-correlation functions of the input and of the output using binary measurements. In a second time Blackman-Tukey approach is used for the estimation of the spectrum of the input and of the output. A numerical example is provided to demonstrate the effectiveness of the proposed algorithm.

Keywords: Estimation; Frequency response; Signal to noise ratio; Erbium; Gaussian distribution; Microsoft Windows; Spectral analysis; Correlation theory; Estimation theory; Spectral Analysis for System Identification; Input/Output Binary

Domains: Engineering Sciences [physics]; Signal and Image processing; Engineering Sciences [physics]; Automatic

Hicham Oualla, Mathieu Pouliquen, M. Frikel, Said Safi, Tristan Bonargent. Spectral Analysis for System Identification from Input/Output Binary Measurements. European Control Conference (ECC), May 2020, Saint-Pétersburg, Russia. pp.731-734, ⟨10.23919/ECC51009.2020.9143871⟩. ⟨hal-02989099⟩