The design of suboptimal inputs for identifying parameters in linear time-varying system

This paper presents a method of suboptimal inputs design (SID) for identifying parameters in linear time-varying system (LTVS) based on the idea of orthogonal approximation of functions. The SID can be reduced to find an eigenvector of the "response matrix" corresponding to its maximum eigenvalue. This method leads to a tractable computation algorithm for the LTVS and its convergence has also been proven. Computational examples are presented to illustrate the utility of this method.