Distinguishing Electronic Devices Using Harmonic Radar Based on a Linear Model

Abstract

A linear model using harmonic radar to distinguish electronic devices is proposed in this article. Nonlinear characteristics of the electronic devices are captured by using power varying signals as incident waves. Three harmonics of the received powers are analyzed in harmonic space. As a major contribution of this study, power series model is employed to calculate the input-output relationship of the electronic devices. As a first in this area, we construct a linear model that relates the measurements to the vectors of parameters characterizing the nonlinear behaviors of the Electronic Circuits Under Test (ECUT). Each nonlinear circuit has a distinct response to a single-tone time-varying signal with varying power. Subsequently, a unique unknown deterministic vector of parameters can be estimated from this linear model for each device. We estimate the unique vectors of parameters using a Maximum Likelihood Estimator (MLE) in the presence of Complex White Gaussian Noise (CWGN). We show that the statistical features of the normalized estimated vectors of parameters can be used to distinguish various nonlinear electronic devices.