A recursive system identification method based on binary measurements

Abstract-- An online approach to parameter estimation problems based on binary observations is presented in this paper. This recursive identification method relies on a least-mean squares approach which makes it possible to estimate the coefficients of a finite-impulse response system knowing only the system input and the sign of the system output. The impulse response is identified up to a positive multiplicative constant. The role of the regulative coefficient is investigated thanks to simulated data. The proposed method is compared with another online approach: it is shown that the proposed method is competitive with the other one in terms of estimation quality and of calculation complexity.

• Type: Journal
• Year: 2010
• Where: CDC
• Authors: Kian Jafari, Jérôme Juillard, Éric Colinet
• Keywords: Binary Observations, CDC 2010, Control Systems, Estimation Problems, Online Approach