Model set validation and update for time-varying SISO systems

This paper presents a method of the model set validation (VD) and update (UD) for time-varying SISO systems by using time-domain data. As a representation of such systems, we use a discrete-time linear difference equation with time-varying parametric uncertainties measured by the weighted /spl infin/-norm. It is shown that the validation can be simply conducted by comparisons of distances in a parameter space, and that the update is done by solving a convex optimization problem in general, and by giving an explicit solution in a special case. Based on these results, we propose an off-line and online algorithm of the model set VD and UD, respectively.