Detection of Cyclostationary Electromagnetic Emissions Using Degree of Cyclostationarity

Abstract

Detection and localization of physical radiating sources allows to allocate hotspots with substantial emitting power on the surface of the printed circuit board (PCB) of the electronic device. Electromagnetic emissions caused by data transferring signals can be mathematically described by cyclostationary stochastic processes. The optimal detection of cyclostationary stochastic process with known two-dimensional autocorrelation function (ACF) assumes two-dimensional crosscorrelation between the shifted product of the stochastic process and relevant ACF. The proposed detection algorithm is based on the degree of cyclostationarity (DCS), defining by comparison of evaluated ACFs obtained from the measured realizations of stochastic process. Experimental verification of the proposed algorithm was implemented by near-field scanning of two spatially distributed sources with different cyclic frequencies on the surface of the PCB.