Nonlinear Radar-Based Classification System for Electronic Devices in Cluttered Close-Range Scenarios

This article proposes a system for detecting and classifying realistic electronic devices in the presence of linear clutter objects based on a frequency-modulated continuous-wave (FMCW) nonlinear radar. For classification, the proposed system utilizes statistical features of the time-segmented baseband signal envelope of the second harmonic responses from targets, which contain nonlinear coefficients of targets, and a support vector machine (SVM) is used as the classifier. The system is validated using an experimental apparatus in an anechoic chamber, and a minimum allowable signal-to-noise ratio (SNR) is set to ensure accurate target classification in the presence of a linear clutter object. The transmit frequency band of the FMCW nonlinear radar used for the experiment is 3.0–3.2 GHz, and the receive frequency band is 6–6.4 GHz, which is the second harmonic of the transmit band. The experimental results show that the proposed system can detect and classify targets with a detection rate of 85.5% and classification accuracy of 89.2%. The proposed system has the potential to provide an effective solution for detecting and classifying unauthorized electronic devices in various scenarios.