IAM   02674
INSTITUTO ARGENTINO DE MATEMATICA ALBERTO CALDERON
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Lowpass FIR Filter Affine Biased Identification
Autor/es:
MAGDALENA BOUZA; FERNANDO GAMA; BRUNO CERNUSCHI FRÍAS
Lugar:
Buenos Aires
Reunión:
Congreso; 24 Congreso Argentino de Control Automático AADECA 2014. Asociación Argentina de Control Automático (AADECA); 2014
Institución organizadora:
AADECA
Resumen:
The problem of lowpass FIR filters identification is addressed in thispaper. Discrete-time, linear time-invariant, causal and stable FIR filters designedusing windowing methods provide a well-known structure that can be exploited inorder to improve the performance of the identification method. In this respect, affineestimators are used to include this a-priori information in a deterministic setting,yielding better results, especially for the adverse case of high noise variance on themeasurements. Simulation results show that the proposed technique indeed yields aconsiderable improvement in terms of the mean squared error.