Ya.A. Parts, Yu.K. Evdokimov
In this paper we discuss the problem of definition of Cramer-Rao lower bound of variance of multi-frequency damped signal parameters estimation from a number of discrete-time observations in additive white Gaussian noise.
We present the common algorithm for estimation of lower bound value by numerical calculations for any number of signal model parameters. We derive and investigate analytical expressions for some important cases of reduced signal model. We consider the questions of error value calculation, a choice of optimum parameters of realization, an estimation of mutual influence of components of a signal. We give also some numeric examples in application to the task of parameters estimation of the multi-frequency piesoresonant sensors.