A new time-domain method for frequency measurement of sinusoidal signals in critical noise conditions

• Proposed method enhances the traditional approach based on zero-crossing events.
• Frequency measurement based on histogram of time intervals of noise and semiperiods.
• Performance in numerical tests very close to those of the most reliable methods.
• Statistical characterization carried out by means of design of experiments techniques.

Frequency measurement of sinusoidal signals corrupted by noise is dealt with. A new time-domain, digital signal processing method, based on an enhanced version of the zero-crossing technique, is in particular proposed. To obtain the desired frequency value, the method exploits the histogram of time intervals between consecutive zero-crossing events, and works with success in the presence of zero-mean additive noise.Several tests conducted on simulated and actual sinusoidal signals highlight the good performance of the method also in critical noise conditions, which make high-end measurement instruments like digital counters poorly fail. Moreover, the obtained results concur with those granted by competitive digital signal processing proposals operating in the frequency domain. With respect to these latter, the main advantage of the method relies on the limited computational resources required, which suggests its implementation even in low-performance processing devices for the realization of cost-effective measurement instruments.