Nonparametric estimation of the best linear time-invariant approximation of a linear time-varying system

Nonparametric frequency response function (FRF) estimates are a standard tool to get quickly insight into the dynamic behavior of complex systems. However, in quite some applications the time-invariance assumption is not exactly met. Therefore, it is important to detect and quantify the deviation from the ideal time-invariant framework that is inherent to the concept of an FRF. In this paper we present a new (direct) method for estimating the best linear time-invariant approximation of a linear time-variant system, and compare it with an existing indirect approach.