Second-order stationary statistical models for inverse frequency processes

Birsen Yazici, Purdue University

Abstract

In this thesis, we have proposed two new classes of nonstationary stochastic processes for 1/f phenomena, developed efficient and convenient spectral representations, and explore some potential applications. The first class is a family of self-similar processes with the correlation structure of the form $E\lbrack X(t)X(\lambda t)\rbrack=t\sp{2H}\lambda\sp HR(\lambda)$ where $t,\lambda>0, {-\infty}0, {-\infty}

Degree

Ph.D.

Advisors

Kashyap, Purdue University.

Subject Area

Electrical engineering

COinS