A General and Accurate Impedance Demodulation Method Based on Improved DFT for Electrical Impedance Spectrometer

Electrical impedance spectroscopy (EIS) is a noninvasive, inexpensive, and rapid detection technique that is widely used for industrial troubleshooting and medical disease diagnosis. Accurate impedance demodulation is essential to obtain reliable EIS information. This article proposes a general and accurate impedance demodulator based on the improved discrete Fourier transform (DFT), which overcomes the inherent drawbacks of the traditional DFT-based demodulator operating under noninteger period sampling condition. First, the sampled sequence is truncated by the user-selected window function. Second, the discrete spectral line with the largest amplitude is found and its spectral value is calculated by the DFT operation. Finally, a general formula for calculating impedance that is independent of sampling conditions is derived. The effects of the proposed method on the impedance demodulation accuracy with different sampling lengths, window functions, analog-to-digital converter quantization bits, and signal-to-noise ratios are investigated by simulations. Moreover, several passive two-resistors and one-capacitor circuits and fresh ex-vivo biological tissues, including carrot, pumpkin, and beef, are measured using a custom-built EIS meter based on the NI-PXI (PCI eXtensions for Instrumentation) platform. Simulation and experimental results demonstrate that the proposed method can not only achieve accurate impedance demodulation at arbitrary measurement frequencies but also has excellent noise immunity. Importantly, this method has the potential to be applied directly in numerous industrial and medical applications.