Robust and accurate real-time estimation of sensors signal parameters by a DSP approach

The paper presents a spectral method for robust and accurate real-time frequency estimation when a square wave affected by impulsive noise is considered. The method, tailored for signals coming from typical resonant sensors (frequency from 2 kHz to 22 kHz), has been implemented and tested using a floating-point digital signal processor (DSP) and a 16-bit analog-to-digital converter (ADC). It has been observed that there is a greater immunity to glitches with respect to the classic solution based on Constant Elapsed Time, while resolution (10-6) and measuring time (10 ms), are close to the ones of traditional high-cost laboratory instrumentation