Detection of Rogue Devices using Unintended Near and Far-field Emanations with Spectral and Temporal Signatures

In this work, using data collected from a desktop computer in an anechoic chamber, we have characterized both near and far-field emissions over a wide frequency band and demonstrated that it matches closely with the monopole antenna in Fresnel and Fraunhofer regions and can be used to detect the start-up of a rogue device in a security-critical room. Detailed spectral analysis shows that emission is strong at lower frequencies. Additionally, high-frequency leakage can also be detected from a 200 cm distance. Frequency shifting of the peaks is observed over time and linked to the program execution. Finally, detection of the onset of a rogue device in a static facility has been demonstrated for different SNRs.