Chipless RFID sensor for corrosion characterization based on frequency selective surface and feature fusion

Abstract

Chipless RFID sensors attract attention to structural health monitoring (SHM) because of its advantages of being low-cost, wireless, passive, and having multiple resonances for sensing. Its application for corrosion sensing, however, receives little attention and faces challenges in terms of sensitivity and reliability. This paper proposes a chipless RFID sensor for corrosion characterization based on frequency selective surface (FSS) and feature fusion. An FSS pattern on a substrate is designed to generate three resonances within 2–6 GHz. The ability of the FSS to characterize corrosion thickness was simulated and validated in the experiments. The experimental results using dedicated corrosion undercoating samples show that the FSS based chipless RFID sensor can be used to characterize corrosion, where the three resonance frequency features provide sensitivity and consistent monotonic relations to the corrosion progression. Furthermore, feature fusion using simple sum and confidence weighted averaging (CWA) can enhance the sensitivity and reliability of the sensor. With the low-profile and printability of the sensor, this work paves the way for smart coatings for corrosion sensing and monitoring on metallic structures.