Instrumental variable least squares blind identification for SIMO systems
Feng Ding · Systems engineering and electronics · 2009
As an important identification method,the instrumental variable method can give the unbiased parameter estimation for systems with colored noises and unknown noise models.The key is how to choose the instrumental variables to generate the instrumental matrices.In the traditional identification approaches with known input signals,the instrumental variables/matrices are formed by using the inputs,but for blind system identifcation,difficulty arises in that the system inputs are unavailable.For single-input,multi-input systems,this paper studies the corresponding blind identification methods using the instrumental variable technique.The basic idea is to identify two combined subsystems simultaneously using the outputs of the third subsystem as the instrumental variables/matrices and presents the instrunmental variable least squares(IVLS) blind identification algorithm and its recursive form.The convergence of the algorithm is also analyzed.A simulation example is included.