Unintentional modulation microstructure enlargement

Radio frequency fingerprinting (RFF) is a technology that identifies the specific emitter of received electromagnetic signal by external measurement minuscule hardware-level, device-specific imperfections. The RFF-related information mainly in form unintentional modulation (UIM), which subtle enough to be effectively imperceptible and submerged intentional (IM). It necessary minimize influence IM expand slight differences between emitters for successful RFF. This paper proposes UIM microstructure enlargement (UMME) method based on feature-level adaptive decomposition (ASD), accompanied autocorrelation cross-correlation analysis. common part evaluated analyzing newly-defined benchmark feature. Three different indexes are used quantify similarity, distance, dependency RFF features from devices. Experiments conducted real-world signals transmitted 20 same type radar working mode. visual image qualitatively shows magnification feature differences; indicators quantitatively describe changes features. Compared with original feature, recognition results Gaussian mixture model (GMM) classifier further validate effectiveness proposed algorithm.