ABSTRACT
ABSTRACT: Impedance spectroscopy (IS) is suitable for different applications like fuel cells, batteries or food analysis. Another field of interest with promising results are IS meas urements in semiconductor (SC) gas sensor applications. The high costs o f measurement systems impede the use of IS in field tests, which are essential for developing sensor sys tems efficiently. In this paper we present an IS hardware concept which is based on FPGAtechnology for both sensor stimulation and data acquisition combined with adapted feature extraction methods for signal interpretation. Based on signal theory we are using a MLS signal, which we identified as an optimal signal for sensor stimulation and identification, respectively. This signal with a bandwidth up to 100M Hz is generated by an FPGA and then reshaped and amplified in order to obtain a high quality MLS signal with variable ampli tude. The sensor response is acquired by a high speed ADC in combination with the FPGA. Furthermore, the FPGA performs an FFT to obtain the impedance spectrum. Based on measurements with a commercial IS analyzer we transfer state-of-the-art feature extraction to sc-gas sensor applications. Relevant features are extracted by fitting impedance curves using equivalent circuit models. In order to develop a field test system that combines IS and temperature cycling of the sensor, we enhance and adapt the feature extraction algorithms. Information extracted from the curve shapes are significant features for gas discrimination.
