A Simulation Study of an Optimized Impedance Spectroscopy Approach for Gas Sensors

This paper presents a simulation study of an optimized and computationally low-cost method for estimating the Electrical Impedance Spectra (EIS) of semiconductor gas sensors and in particular for Metal-Oxide (MOX) chemiresistive sensors. The approach is applied directly to the sensor without using a voltage divider and is based on a well-known signal processing principle: a Linear Time-Invariant (LTI) system's Impulse Response (IR) is estimated by stimulating the system with a Maximum Length Sequence (MLS) and thus performing the circular cross-correlation between input and output signals. Finally, the system's frequency response, i.e. the impedance spectrum, is obtained through the Fast Fourier Transform (FFT) of the estimated impulse response. The technique is demonstrated in simulation environment using a time-invariant passive network that simulates a MOX sensor. Simulation results and performance analysis are discussed showing design trade-offs.