CSRR Based Metamaterial Inspired Sensor for Liquid Concentration Detection Using Machine Learning

Abstract

A sensor to accurately predict chemical concentrations has been proposed in this research work. Inspired by Metamaterials, the sensor is composed of Complementary Split-Ring Resonators (CSRRs) and utilizes the Machine Learning technique to accurately predict the concentrations. The sensor is designed to maximize the interaction of the Material Under Test (MUT) with the sensitive regions of the CSRRs. The usage of costly and complex fluidic channels and sample containers is avoided by using filter paper for the liquid MUT placement in between the resonators. The proposed sensor is small (2.3 cm × 2.3 cm), simple, employs a low-cost fabrication technique, and offers an alternate sensing mechanism that requires a minimal amount of the MUT. The multiple resonances exhibited by the proposed sensor add to the reliability and accuracy of the sensor.